CompleteEASE

6
Software Manual

J.A. Woollam
©2004–2020 J.A. Woollam Co., Inc. All Rights Reserved.

This manual was written with features available in CompleteEASE 6.54.

J. A. Woollam Co., Inc.

645 M Street, Suite 102
Lincoln, NE 68508
USA

Phone: 1.402.477.7501
Fax: 1.402.477.8214

www.jawoollam.com
[email protected]
[email protected]
[email protected]
[email protected]

Copying and Distribution

You are welcome to make backup copies of this manual for your own use and protection. However, you are not
permitted to make copies for others or for resale.

Trademarks
J.A. Woollam, CompleteEASE, M-2000, RC2, alpha-SE, WVASE, VASE, Gen-Osc, and Global Fit are registered trademarks
of J.A. Woollam. All other trademarks used within are the property of their respective owners.
 Contents

1. Introduction 1-1
1.1. Manual Overview ..................................................................................................... 1-1
1.2. Conventions Used in this Manual ............................................................................. 1-3
Tabs .......................................................................................................................... 1-4
Panels ....................................................................................................................... 1-4
Buttons ..................................................................................................................... 1-4
Commands ................................................................................................................ 1-4
Mouse-Selected Menus ............................................................................................ 1-5
Pipeline Descriptions ................................................................................................ 1-6
1.3. File Structure and File Types .................................................................................... 1-7
COMMON Location .................................................................................................. 1-8
File Types .................................................................................................................. 1-8
1.4. File Dialog ................................................................................................................. 1-9
Recent ....................................................................................................................... 1-9
Projects ................................................................................................................... 1-11
Library ..................................................................................................................... 1-13
Opening Files from Windows Explorer ................................................................... 1-14
1.5. Multiple Users ........................................................................................................ 1-14
1.6. Shortcut Guide........................................................................................................ 1-18
1.7. What is Spectroscopic Ellipsometry?...................................................................... 1-20

2. Data Acquisition 2-22

2.1. alpha-SE Systems .................................................................................................... 2-22
2.2. M-2000 and RC2 Systems ....................................................................................... 2-25
Recipes ................................................................................................................... 2-25
Acquisition Parameters .......................................................................................... 2-28
Scan Pattern ........................................................................................................... 2-32
Model ..................................................................................................................... 2-33
Running a Recipe and Viewing Results ................................................................... 2-34

3. Data Analysis 1 – Basic 3-37

3.1. General Data Analysis Theory................................................................................. 3-39
Layered Optical Model ........................................................................................... 3-39
Surface Roughness ................................................................................................. 3-41
Grading ................................................................................................................... 3-41
Anisotropy .............................................................................................................. 3-41
Goodness of Fit: Definition of the MSE ................................................................. 3-42
“Fitting” Data .......................................................................................................... 3-43
Evaluating the Fit Results ....................................................................................... 3-43
“Physical” Optical Constants .................................................................................. 3-45
3.2. Samples with Known Optical Constants: SiO2 on Si ............................................... 3-46
Fit Results and Parameter Error Bars ..................................................................... 3-48
Si with Thermal Oxide.mod .................................................................................... 3-50
Derived Parameters ................................................................................................ 3-51
Hiding, Renaming, and Bounding Fit Parameters ................................................... 3-51
Angle Offset ............................................................................................................ 3-53
Thickness Pre-Fit ..................................................................................................... 3-53
Visualizing Parameter Changes .............................................................................. 3-58

CompleteEASE Software Manual • iii

 3.3. Transparent Thin Films 1: PECVD SiO2 on Si .......................................................... 3-58
Si with Transparent Film.mod ................................................................................ 3-60
Global Fit ................................................................................................................ 3-64
Rename Layer & Fit Parameters ............................................................................. 3-67
3.4. Transparent Substrates: Bare Glass ....................................................................... 3-69
Substrate Backside Reflections............................................................................... 3-69
Graphing Multiple Data Sets .................................................................................. 3-70
Using the Graph Scratchpad ................................................................................... 3-71
Glass Substrate.mod ............................................................................................... 3-73
Graphing Depolarization......................................................................................... 3-75
Including Backside Reflections ............................................................................... 3-78
Fitting Depolarization ............................................................................................. 3-79
Glass Substrate (with backside reflection).mod ..................................................... 3-80
3.5. Transparent Films 2: SiC on Si ................................................................................ 3-85
Surface Roughness ................................................................................................. 3-86
Simple Grading ....................................................................................................... 3-87
Simple Anisotropy .................................................................................................. 3-90
Try Alternate Models .............................................................................................. 3-93
3.6. Log and Report a Series of Films............................................................................. 3-97
Copy to Clipboard ................................................................................................... 3-97
Fit Log ..................................................................................................................... 3-98
Comparing Results .................................................................................................. 3-99
Compare Optical Constants .................................................................................. 3-100
Reanalyzing Multiple Data Sets ............................................................................ 3-100
Generate Reports ................................................................................................. 3-104
Add Optical Constants to Report .......................................................................... 3-106
Copy Analysis Report to Clipboard ....................................................................... 3-107
Save Log File ......................................................................................................... 3-108
Snapshot ............................................................................................................... 3-108

4. Data Analysis 2 – Intermediate 4-109

4.1. Metal Substrates .................................................................................................. 4-111
Pseudo Optical Constants ..................................................................................... 4-111
WvlByWvl Layer .................................................................................................... 4-113
B-Spline Layer ....................................................................................................... 4-118
Save Layer Optical Constants ............................................................................... 4-122
Save Model ........................................................................................................... 4-124
4.2. Self-Assembled Monolayer on Au ........................................................................ 4-125
Self-Assembled Monolayers ................................................................................. 4-127
Fit Parameter Uniqueness .................................................................................... 4-129
4.3. Semi-Absorbing Thin Films (B-Spline) ................................................................... 4-132
Absorbing vs. Transparent Region ........................................................................ 4-133
‘Set Ranges’ .......................................................................................................... 4-133
Transparent Region Analysis ................................................................................ 4-136
Expansion to Absorbing Region ............................................................................ 4-138
B-Spline Layer ....................................................................................................... 4-139
Wvl. Range Expansion Fit ..................................................................................... 4-141
4.4. Semi-Absorbing Thin Films (Gen-Osc) .................................................................. 4-146
Gen-Osc Layer....................................................................................................... 4-146
Save Model ........................................................................................................... 4-158
4.5. Multiple Layer Film Stacks .................................................................................... 4-159
Superlattice .......................................................................................................... 4-166

iv • CompleteEASE Software Manual

 Thickness Pre-Fit ................................................................................................... 4-166
Global Fit .............................................................................................................. 4-167
Parameter Coupling .............................................................................................. 4-167
Layer Coupling ...................................................................................................... 4-168
4.6. Thin Absorbing Films ............................................................................................ 4-169
SE + Transmission ................................................................................................. 4-169
Interference Enhancement................................................................................... 4-177

5. Data Analysis 3 – Advanced 5-181

5.1. Map Data and Model Non-Idealities .................................................................... 5-183
Show Map Data .................................................................................................... 5-183
Selecting a Point from the Map ............................................................................ 5-184
Si with Transparent Film ....................................................................................... 5-185
Wavelength Units – eV ......................................................................................... 5-186
Fixing the Dispersion ............................................................................................ 5-187
Sellmeier Dispersion ............................................................................................. 5-190
Model Calculation................................................................................................. 5-192
Model Calculation – Thickness Non-Uniformity ................................................... 5-192
Model Calculation – Bandwidth (nm) ................................................................... 5-194
Include Depolarization Data ................................................................................. 5-196
Fit Scan Data ......................................................................................................... 5-198
5.2. Multi Sample Analysis........................................................................................... 5-202
Shortcuts for Mapping Data ................................................................................. 5-203
Blank Model.......................................................................................................... 5-203
Set Symbol Size ..................................................................................................... 5-205
Zooming on Map .................................................................................................. 5-205
Selecting Multiple Points on a Map ...................................................................... 5-206
Switching Graph Views ......................................................................................... 5-208
Fitting Multiple Data Sets ..................................................................................... 5-208
Graph All Data Sets ............................................................................................... 5-209
Selecting and Un-selecting Data Sets ................................................................... 5-210
Multi Sample Analysis (MSA) ................................................................................ 5-211
Clear Multi-Data Set Mode ................................................................................... 5-214
Multi Sample Analysis (revisited) ......................................................................... 5-220
5.3. Creating Optical Constants Library for Composition- or Temperature-Dependent
Materials ............................................................................................................................. 5-228
Opt. Const. Compare Model ................................................................................. 5-228
Save OC Library ..................................................................................................... 5-232
5.4. Anisotropic Films .................................................................................................. 5-234
Convert to Anisotropic ......................................................................................... 5-236
Biaxial Layer .......................................................................................................... 5-237

6. Data Analysis 4 – In-situ Data 6-241

6.1. In-situ Data Analysis ............................................................................................. 6-242
Shortcuts for In-situ Data ..................................................................................... 6-242
Navigating In-situ Data ......................................................................................... 6-243
Viewing Spectroscopic Data at a Specific Time .................................................... 6-243
Turning off the “Auto-Fit” function ...................................................................... 6-245
Building a Model ................................................................................................... 6-245
‘Fit Dynamic’ ......................................................................................................... 6-247
Selecting Multiple Timeslices ............................................................................... 6-253

CompleteEASE Software Manual • v

 Manipulating Multiple Data Sets .......................................................................... 6-254
Multi-Sample Analysis (for multiple timeslices) ................................................... 6-254
Fitting Optical Constants ...................................................................................... 6-257
6.2. Growth Rate and Optical Constants ..................................................................... 6-260
Virtual Interface .................................................................................................... 6-260
Growth Rate and Optical Constants (GROC Model) ............................................. 6-261
Multilayer Example ............................................................................................... 6-263

7. Reference – Graph 7-273

7.1. Basic Screen Layout .............................................................................................. 7-273
7.2. Graph .................................................................................................................... 7-274
‘Show Data’........................................................................................................... 7-274
Select Range ......................................................................................................... 7-274
Zoom All................................................................................................................ 7-275
7.3. Right-Click Menu (Graph) ..................................................................................... 7-275
Zoom All................................................................................................................ 7-276
Statistics ............................................................................................................... 7-276
Copy Graph to Clipboard ...................................................................................... 7-276
Copy Data to Clipboard ........................................................................................ 7-276
Copy Data to Graph Scratchpad ........................................................................... 7-276
View Graph Scratchpad ........................................................................................ 7-277
Show Legend ........................................................................................................ 7-277
Show Symbols ....................................................................................................... 7-277
Set Symbol Size ..................................................................................................... 7-278
Auto X-, Y-, or Z-Scale ........................................................................................... 7-278
Manual X-, Y-, or Z-Scale ....................................................................................... 7-278
Scale Z-Range by n-Sigma ..................................................................................... 7-278
Graph Points as a Line .......................................................................................... 7-278
Draw Surface Plot ................................................................................................. 7-279
Use Clipboard Aspect Ratio .................................................................................. 7-280
Cancel Menu ......................................................................................................... 7-280
7.4. +Data Menu .......................................................................................................... 7-280
Psi, Delta ............................................................................................................... 7-281
Re(rho), Im(rho) .................................................................................................... 7-282
N, C, S.................................................................................................................... 7-282
<Pseudo> Transforms (<e1>, <e2>, <n>,<k>) ....................................................... 7-282
Intensity ................................................................................................................ 7-284
Depolarization ...................................................................................................... 7-284
Graph vs. Angle..................................................................................................... 7-285
Double-Y Axis ........................................................................................................ 7-286
Difference Mode................................................................................................... 7-286
More Options ....................................................................................................... 7-286
Show GE Items ...................................................................................................... 7-287
Select Ratios to Graph .......................................................................................... 7-288
Show Mueller Items.............................................................................................. 7-289
Show MM Graph Utility ........................................................................................ 7-293
7.5. +Model Menu ....................................................................................................... 7-295
Layer Optical Constants ........................................................................................ 7-295
All Layer Optical Constants ................................................................................... 7-296
Optical Constants Profile ...................................................................................... 7-297
7.6. +Results Menu ...................................................................................................... 7-297
7.7. Special Graph Features ......................................................................................... 7-298

vi • CompleteEASE Software Manual

 Multiple Data Sets ................................................................................................ 7-298
Rotation Data ....................................................................................................... 7-300
Translation (Uniformity Map) Data ...................................................................... 7-302
Dynamic (In Situ) Data .......................................................................................... 7-305
7.8. Graph Scratchpad ................................................................................................. 7-307
Edit........................................................................................................................ 7-309
Delete ................................................................................................................... 7-310
Delete All .............................................................................................................. 7-310
Create Curve ......................................................................................................... 7-310
Add Trend ............................................................................................................. 7-311
Right-Click Menus ................................................................................................. 7-312
Graph Settings ...................................................................................................... 7-313
Open ..................................................................................................................... 7-314
Save ...................................................................................................................... 7-314

8. Reference – Measurement Tab 8-315

8.1. System Status Panel ............................................................................................. 8-315
8.2. Measurement Controls Panel (alpha-SE).............................................................. 8-315
Mode .................................................................................................................... 8-315
Sample Alignment ................................................................................................ 8-316
Focus Probes Installed .......................................................................................... 8-317
Model ................................................................................................................... 8-317
Save Data after Measurement ............................................................................. 8-318
‘Measure’.............................................................................................................. 8-318
8.3. Measurement Controls Panel (M-2000 & RC2) .................................................... 8-318
Recipe ................................................................................................................... 8-319
Save Data after Measurement ............................................................................. 8-321
‘Measure’.............................................................................................................. 8-321
Acquire Multiple Measurements .......................................................................... 8-321
Show Recipe Scheduler ........................................................................................ 8-322
8.4. Acquisition Parameters ........................................................................................ 8-329
Data Type.............................................................................................................. 8-329
Sample Alignment ................................................................................................ 8-331
Common Acquisition Parameters ......................................................................... 8-333
8.5. Scan Pattern ......................................................................................................... 8-333
Scan Region of Interest ......................................................................................... 8-335
Common Scan Patterns ........................................................................................ 8-336
8.6. Fit Results Panel ................................................................................................... 8-336
Right-Click Menu .................................................................................................. 8-336
‘View Previous Fit Results’ .................................................................................... 8-337

9. Reference – Analysis Tab 9-341

9.1. Data Panel ............................................................................................................ 9-341
‘Open’ ................................................................................................................... 9-341
‘Save’ .................................................................................................................... 9-343
‘Info’ ..................................................................................................................... 9-345
‘Set Ranges’ .......................................................................................................... 9-345
9.2. Fit Panel ................................................................................................................ 9-346
‘Generate’ ............................................................................................................. 9-346
‘Fit’ ........................................................................................................................ 9-346
‘Fit Scan Data’ ....................................................................................................... 9-346

CompleteEASE Software Manual • vii

 ‘Fit Dynamic’ ......................................................................................................... 9-347
‘Fit Rotator Data’ .................................................................................................. 9-347
Graphs of Fit Results ............................................................................................. 9-348
‘Reset’ ................................................................................................................... 9-349
9.3. Fit Panel Right-Click Menu .................................................................................... 9-349
Auto Fit ................................................................................................................. 9-350
View Fit Stats ........................................................................................................ 9-350
View Parameter Stats ........................................................................................... 9-350
Copy to Clipboard – Formatted ............................................................................ 9-351
Copy to Clipboard – Table .................................................................................... 9-351
Copy Parameters to Clipboard ............................................................................. 9-351
Copy Analysis Report to Clipboard ....................................................................... 9-351
Add to Fit Log........................................................................................................ 9-353
View Fit Log .......................................................................................................... 9-353
Advanced Graph Options ..................................................................................... 9-358
Graph OCs vs. Time / Graph OCs vs. Position ....................................................... 9-358
9.4. Model Panel.......................................................................................................... 9-359
‘Open’ ................................................................................................................... 9-360
‘Save’ .................................................................................................................... 9-360
‘Clear’.................................................................................................................... 9-361
‘Open Snapshot’ ................................................................................................... 9-361
‘Save Snapshot’ .................................................................................................... 9-361
Layer Commands: Add.......................................................................................... 9-361
Layer Commands: Delete ..................................................................................... 9-362
Layer Commands: Save......................................................................................... 9-362
Include Surface Roughness ................................................................................... 9-363
Pictorial Model Layer Stack .................................................................................. 9-363
Model Panel Right-Click Menu ............................................................................. 9-363
Alternate Model Panels (Model #2 & Model #3) ................................................. 9-364
9.5. Model Layer Stack ................................................................................................ 9-365
Editing Layers (Layer Right-Click Menu) ............................................................... 9-365
Editing Fit Parameters (Fit Parameter Dialog Box) ............................................... 9-374
9.6. +MODEL Options .................................................................................................. 9-377
Angle Offset .......................................................................................................... 9-378
Analyze All Rotation Angles .................................................................................. 9-378
Include Substrate Backside Correction ................................................................. 9-378
Model Calculation................................................................................................. 9-379
+Parameter Smearing ........................................................................................... 9-380
Patterning ............................................................................................................. 9-381
Ambient Index > 1 ................................................................................................ 9-381
Use Scattering Factor ........................................................................................... 9-382
Wvl Shift (nm) ....................................................................................................... 9-382
Return Path Ellipsometer ..................................................................................... 9-382
Delta Offset .......................................................................................................... 9-382
Psi Offset .............................................................................................................. 9-382
Source Rot. / Receiver Rot. ................................................................................... 9-382
+Multi-Model Patterning ...................................................................................... 9-383
+Multi-Sample Analysis (MSA) ............................................................................. 9-384
+Parameter Coupling ............................................................................................ 9-386
Use Previous Results............................................................................................. 9-387
9.7. +FIT Options ......................................................................................................... 9-389
Include Initial Fit Stage ......................................................................................... 9-389
Perform Thickness Pre-Fit .................................................................................... 9-390

viii • CompleteEASE Software Manual

 Use Global Fit ....................................................................................................... 9-391
Selected Data........................................................................................................ 9-394
Fit Weight ............................................................................................................. 9-395
Include Depolarization Data ................................................................................. 9-396
Include Intensity Data........................................................................................... 9-396
Transmission Data % Weight ................................................................................ 9-396
Define MSE using L*a*b* ..................................................................................... 9-397
Limit Wvl. for Fit ................................................................................................... 9-397
Limit Angles for Fit ................................................................................................ 9-397
Max. Acceptable MSE ........................................................................................... 9-398
Skip Data Points in Fit ........................................................................................... 9-398
Max. Fit Iterations ................................................................................................ 9-398
Auto Fit Parameter Reset ..................................................................................... 9-398
Include Derived Parameters ................................................................................. 9-399
+Translation Scan: First Point Only ....................................................................... 9-403
9.8. +OTHER Options ................................................................................................... 9-403
Wvl. Range Expansion Fit ..................................................................................... 9-404
Try Alternate Models ............................................................................................ 9-404
Fit Parameter Uniqueness .................................................................................... 9-406
Fit Parameter Error Estimation ............................................................................. 9-407
Simulate Data ....................................................................................................... 9-411
Add Optical Constants to Report .......................................................................... 9-412
# of Processor Cores to Use .................................................................................. 9-412
Color Calculation .................................................................................................. 9-412
Configure Options................................................................................................. 9-414
Turn Off All Fit Parameters ................................................................................... 9-414
9.9. Additional Models ................................................................................................ 9-414
Basic Library Models ............................................................................................. 9-415
Growth Rate and Opt. Const. (GROC) Model ....................................................... 9-415
Opt. Const. Compare (OCC) Model....................................................................... 9-416
Pore Size Model .................................................................................................... 9-416
Retardance Model ................................................................................................ 9-416

10. Reference – Layer Types 10-418

10.1. Tabulated ............................................................................................................ 10-420
10.2. Cauchy ................................................................................................................ 10-420
10.3. B-Spline ............................................................................................................... 10-422
Resolution (eV) ................................................................................................... 10-425
Draw Node Graph ............................................................................................... 10-425
Fit Optical Constants .......................................................................................... 10-426
Use KK Mode ...................................................................................................... 10-426
+Kramers-Kronig ................................................................................................. 10-426
+Nodes................................................................................................................ 10-427
+Advanced .......................................................................................................... 10-430
10.4. Gen-Osc .............................................................................................................. 10-431
Show Dialog ........................................................................................................ 10-432
Real Part (+e1 Components)............................................................................... 10-433
Imaginary Part (+e2 Components) ..................................................................... 10-434
10.5. EMA (Effective Medium Approximation) ........................................................... 10-449
10.6. Intermix .............................................................................................................. 10-451
10.7. Void .................................................................................................................... 10-451
10.8. Additive .............................................................................................................. 10-452

CompleteEASE Software Manual • ix

 10.9. Biaxial ................................................................................................................. 10-453
10.10. Biaxial 2 .............................................................................................................. 10-454
10.11. Cauchy_Wvl ........................................................................................................ 10-455
10.12. Cauchy_Extended ............................................................................................... 10-455
10.13. Coupled .............................................................................................................. 10-456
10.14. EMA-Coupled ..................................................................................................... 10-456
10.15. Full Tensor .......................................................................................................... 10-457
10.16. GenOsc 3D .......................................................................................................... 10-458
10.17. Glass Catalog ...................................................................................................... 10-459
10.18. Graded ................................................................................................................ 10-460
# of Slices ............................................................................................................ 10-462
Grade Type ......................................................................................................... 10-462
10.19. KK Material ......................................................................................................... 10-469
10.20. Porous EMA ........................................................................................................ 10-470
10.21. Sellmeier ............................................................................................................. 10-470
10.22. Temperature or Composition Library ................................................................. 10-471
10.23. User Defined ....................................................................................................... 10-471
10.24. Virtual Substrate................................................................................................. 10-472
10.25. WvlByWvl ........................................................................................................... 10-473
10.26. Zero .................................................................................................................... 10-473

11. Reference – In situ Tab 11-474

11.1. Displaying the In situ Tab.................................................................................... 11-474
11.2. In situ Tab Features ............................................................................................ 11-475
Status Box ........................................................................................................... 11-476
Hardware Control Section .................................................................................. 11-476
Models ................................................................................................................ 11-479
View .................................................................................................................... 11-481

12. Reference – Hardware Tab 12-484

12.1. Configuration Files .............................................................................................. 12-484
hardware.cnf ...................................................................................................... 12-484
CompleteEASE.cnf .............................................................................................. 12-484
CompleteEASEhard.cnf ....................................................................................... 12-484
Privileges.cnf....................................................................................................... 12-484
rce-cal.cnf, rce-comp.cnf, rce-dc.cnf, rce-refl.cnf .............................................. 12-484
12.2. Hardware Tab – alpha-SE ................................................................................... 12-485
Hardware ............................................................................................................ 12-485
Signal .................................................................................................................. 12-485
Calibration .......................................................................................................... 12-487
Show Logs ........................................................................................................... 12-487
Camera ............................................................................................................... 12-488
Routine Test Measurement ................................................................................ 12-489
12.3. Hardware Tab – M-2000 and RC2 ...................................................................... 12-490
Hardware Status ................................................................................................. 12-490
Controls .............................................................................................................. 12-491
General ............................................................................................................... 12-491
Calibration .......................................................................................................... 12-494
System ................................................................................................................ 12-495
Misc. ................................................................................................................... 12-497
System Information ............................................................................................ 12-499

x • CompleteEASE Software Manual

 13. Reference – Options Tab 13-504
13.1. Display Units ....................................................................................................... 13-504
Wavelength Units ............................................................................................... 13-504
Thickness Units ................................................................................................... 13-504
Optical Constants ............................................................................................... 13-505
‘Use Defaults’ ..................................................................................................... 13-505
13.2. Miscellaneous ..................................................................................................... 13-505
‘Reset Window Size’ ........................................................................................... 13-505
‘Manage Users’ ................................................................................................... 13-505
‘Show Manual (F1)’ ............................................................................................. 13-506
‘View CompleteEASE Shortcuts’ ......................................................................... 13-506
‘Deactivate’ ......................................................................................................... 13-507
13.3. Configuration Controls ....................................................................................... 13-508
‘Edit Configuration’............................................................................................. 13-508
General ............................................................................................................... 13-509
Memory Management ....................................................................................... 13-512
Display Units ....................................................................................................... 13-513
Remote Communications ................................................................................... 13-514
Window Setup .................................................................................................... 13-516
Graph Clipboard Parms ...................................................................................... 13-517
Hardware Simulation Mode ............................................................................... 13-517
In Situ Tab Parameters ....................................................................................... 13-518
Data Export Parameters ..................................................................................... 13-519
Shutter Controls ................................................................................................. 13-520
‘Import Configuration’ / ‘Save Configuration’ .................................................... 13-521
13.4. About CompleteEASE ......................................................................................... 13-521
‘Create Debug File’ ............................................................................................. 13-522
‘View Error Log’ .................................................................................................. 13-522

14. Appendices 14-523

14.1. Installing CompleteEASE ..................................................................................... 14-523
Before Installation .............................................................................................. 14-523
Upgrading to CompleteEASE 6 ........................................................................... 14-523
Installing CompleteEASE (Windows) .................................................................. 14-523
Installing CompleteEASE (Mac) .......................................................................... 14-525
Operating Hardware ........................................................................................... 14-527
14.2. Activating CompleteEASE ................................................................................... 14-528
Types of CompleteEASE Licenses ....................................................................... 14-528
Online Activation ................................................................................................ 14-528
Offline Activation ................................................................................................ 14-529
Deactivating CompleteEASE ............................................................................... 14-533
Activation Troubleshooting ................................................................................ 14-534
14.3. General Troubleshooting .................................................................................... 14-537
Reporting Problems ............................................................................................ 14-537
Creating a Debug file .......................................................................................... 14-537
14.4. Increase CompleteEASE Memory Usage ............................................................ 14-537
14.5. Network Card Configuration .............................................................................. 14-538
14.6. CompleteEASE Data Transfer (to/from WVASE)................................................. 14-541
Importing CompleteEASE Data into WVASE ....................................................... 14-541
Exporting WVASE Data to CompleteEASE .......................................................... 14-543
14.7. Comparing CompleteEASE and WVASE Results ................................................. 14-544

CompleteEASE Software Manual • xi

 Fit Weighting ...................................................................................................... 14-544
Depolarization Data ............................................................................................ 14-544
Intensity Data ..................................................................................................... 14-545
Surface Roughness ............................................................................................. 14-545
14.8. CompleteEASE Remote Commands.................................................................... 14-545
Configuring Communication Settings ................................................................. 14-545
In-Situ Acquisition Modes................................................................................... 14-546
Remote Commands ............................................................................................ 14-547
Asynchronous Real-Time Reporting ................................................................... 14-554
Parameter Value Query ...................................................................................... 14-554
Comm Test Program ........................................................................................... 14-555

15. Index 15-1

xii • CompleteEASE Software Manual

1. Introduction

Thank you for purchasing a J.A. Woollam Spectroscopic Ellipsometer (SE) system.
We hope the information contained in this manual will help you develop a better
understanding and appreciation for data analysis using CompleteEASE.

1.1. Manual Overview

This manual is organized into the following chapters:

Chapter 1: Introduction
Chapter 1 will introduce you to CompleteEASE. It contains conventions used in this
manual, information on file structure and file types, how to configure
CompleteEASE for multiple users, a list of shortcuts, and a brief introduction to
spectroscopic ellipsometry.

Chapter 2: Data Acquisition

Chapter 2 provides a quick overview of data acquisition. As the details can vary
depending on your instrument and configuration, this chapter is meant only for
quick reference. See your hardware manual for more details.

Chapter 3: Data Analysis 1 – Basic

Chapter 3 begins a series of chapters devoted to data analysis procedures. Each
chapter steps through examples provided with this manual. Chapter 3 describes
the basic SE data analysis algorithms used by CompleteEASE. All users can benefit
from understanding how the most basic models are constructed. Examples in this
chapter include:
• Oxides on Silicon
• Transparent Thin Films
• Glass Substrate Considerations
• Surface Roughness, Index Grading, and Simple Anisotropy
• Logging and Reporting Results from a Series of Films

CompleteEASE Software Manual Introduction • 1-1

 Chapter 4: Data Analysis 2 – Intermediate
Chapter 4 continues to build on principles of data analysis, focusing on absorbing
materials. The examples are intended to teach approaches for absorbing materials
and a few advanced features within CompleteEASE. Examples in this chapter
include:
• Metal Substrates
• Organic Monolayers on Gold
• Semi-Absorbing Thin Films (B-Spline Approach)
• Semi-Absorbing Thin Films (Gen-Osc Approach)
• Multiple Layer Film Stacks
• Thin Absorbing Films

Chapter 5: Data Analysis 3 – Advanced

Chapter 5 continues the description of data analysis for advanced applications.
Examples in this chapter include:
• Working with Uniformity Maps
• Multi-Sample Analysis
• Sample Non-Idealities
• Creating Alloy or Temperature Files
• Anisotropic Thin Films

Chapter 6: Data Analysis 4 – In-situ Data

Chapter 6 details how to work with time-dependent data that is measured during
deposition or etch of a film. This chapter is also useful when working with liquid or
heat stages, where real-time data can be obtained for varying measurement
conditions.
• Working with in-situ data
• Growth Rate and Optical Constants Model

Chapters 7 – 13: Feature References

These chapters are targeted toward advanced users that need an explanation of
specific topics. The chapters are organized as follows:
Chapter 7: Graph
Chapter 8: Measurement Tab
Chapter 9: Analysis Tab
Chapter 10: Layer Types
Chapter 11: In-situ Tab
Chapter 12: Hardware Tab
Chapter 13: Options Tab

1-2 • Introduction CompleteEASE Software Manual

 Appendices: Additional Information
The Appendices include additional information for your reference.
• Appendix 14.1 Installing CompleteEASE
• Appendix 14.2 Activating CompleteEASE
• Appendix 14.3 General Troubleshooting
• Appendix 14.4 Increase CompleteEASE Memory Usage
• Appendix 14.5 Network Card Configuration
• Appendix 14.6 CompleteEASE Data Transfer (to/from WVASE)
• Appendix 14.7 Comparing CompleteEASE and WVASE Results
• Appendix 14.8 CompleteEASE Remote Commands

1.2. Conventions Used in this Manual

CompleteEASE is written in the Java programming language with a tab interface to
conveniently organize software features. Figure 1-1 shows the home screen of
CompleteEASE for an RC2. Your system may have different features and therefore
different measurement controls, but all versions of the software have tabs across
the top with panels and the graph window.

Figure 1-1. Example of the Measurement tab view in CompleteEASE.

To help you navigate this manual, the following conventions will be followed to
describe features in CompleteEASE.

CompleteEASE Software Manual Introduction • 1-3

 Tabs
When CompleteEASE is used with an alpha-SE system, there are typically four tabs,
as shown across the top of the software screen: Measurement, Analysis,
Hardware, and Options. An additional In-situ tab is available for some systems that
may be applied to real-time measurements. Throughout this manual, tabs will be
written in bold, italic text.

Panels
Within each tab, there are screen areas referred to as panels. From the
Measurement tab shown in Figure 1-1, the panels include System Status,
Measurement Controls, and Fit Results. Throughout this manual, panels will be
written in bold, blue text.

Buttons
Figure 1-2 shows an example Hardware tab for CompleteEASE. The figure below
has six panels: Hardware Status, Controls, Alignment, Other, Routine Test
Measurement, and System Information. Within panels are various buttons.
Buttons will be designated with single quote marks, such as ‘Align Sample’ and
‘Display Signal'.

Figure 1-2. Example Hardware tab.

Commands
Figure 1-3 shows an example Analysis tab. Within the Analysis tab, there are Data,
Fit, and Model panels. The Model panel includes many choices – some of which are
shown in bold, underlined, red letters. Within this manual, they will also be shown
in bold, underlined, red letters such as Add, Delete, and Save. Model details are
further divided into expandable sections, shown in bold letters preceded by a red +
or - symbol. The symbol preceding the text is interactive and shows whether the
section is collapsed (+) or expanded (-). These sections will be listed using bold text
with a red + symbol: +MODEL Options, +FIT Options, and +OTHER Options.

1-4 • Introduction CompleteEASE Software Manual

 Figure 1-3. Analysis Tab.

Mouse-Selected Menus
CompleteEASE also utilizes both mouse buttons to access additional menus and
features. Throughout this manual, if you need to select the right or left mouse
button, you may see the following symbols: (R) or (L), respectively. Figure 1-4
shows the menu that appears when the right mouse button (R) is clicked within
the Fit results box.

CompleteEASE Software Manual Introduction • 1-5

 Figure 1-4. Right-clicking (R) in the fit results box (white area) opens a menu with options
related to the fit results.

Pipeline Descriptions
The descriptions used in above sections can be “pipelined” to form a concise
description of software features or menus. In this case, the description will take on
a form like:
Tab > Panel > ‘Button’
For example, the ‘Open’ button within the Model panel under the Analysis tab is
conveniently described as:
Analysis > Model > ‘Open’
As another example, the command to right-click the mouse within the Fit Panel
from the Analysis tab and then choose the “Add to Fit Log” menu choice would
read as follows:
Analysis > Fit > (R) > ‘Add to Fit Log’

1-6 • Introduction CompleteEASE Software Manual

 1.3. File Structure and File Types
CompleteEASE is typically installed in C:\CompleteEASE. There are several default
subdirectories within this folder, as shown in Figure 1-5.

Figure 1-5. CompleteEASE directory.

Table 1-1 provides more details about each of these subdirectories and their
function within CompleteEASE.

Table 1-1. CompleteEASE directory information.

DIRECTORY FUNCTION
cnf Hardware configuration files
DAT COMMON data files, including a sub-folder of Examples that
are included with CompleteEASE. Data files saved in this
directory can be viewed and opened by all users.
HTML Files and graphs used by CompleteEASE to construct reports
Clipboard within the HTML clipboard.
jar Java application files
MAT COMMON material files to describe optical constants of
materials. This folder is divided into Advanced, Basic,
Dielectric, Metal, and Semiconductor folders. Files within
this directory can be viewed and opened by all users but are
read-only.
MOD COMMON model files to describe a sample and the
associated analysis strategy. COMMON snapshots are also
saved here. This folder is divided into Advance, Basic, and
Calibration Wafers.
Recipe COMMON recipe files describe a complete measurement
and model strategy. Recipes are not used with alpha-SE
systems.

CompleteEASE Software Manual Introduction • 1-7

 You can create your own directories to store data files, models, materials,
snapshots, and recipes, but any subdirectories created in the CompleteEASE folder
will be treated as “read-only”. This is to maintain the default files and libraries
included with CompleteEASE.

COMMON Location
As indicated in Table 1-1, CompleteEASE uses these folders as a COMMON location
for data, material, model, and recipe files which are accessible to all users. The
DAT, MAT, and MOD folders are also accessible for all recipes. For CompleteEASE
to locate material or model files associated with a recipe, it must be in the same
folder as the recipe or in the respective COMMON folder.
The location of these COMMON folders is as follows:
C:\CompleteEASE\DAT
C:\CompleteEASE\MAT
C:\CompleteEASE\MOD
C:\CompleteEASE\recipe

File Types
There are many different file types used within CompleteEASE. Table 1-2 provides a
list of different file types and their corresponding file extensions.

Table 1-2. List of File Types used by CompleteEASE and their corresponding extensions.

FILE TYPE EXTENSION

Experimental Data .SE
In-situ Experimental Data (data vs. time) .iSE
Model .mod
Material File (optical constants) .mat
Snapshot (data, model, and fit results) .SEsnap
Data Acquisition Parameters .parms
Mapping Pattern .scan
Recipe (points to Data Acquisition Parameters, .recipe
Mapping Pattern, and Model)
Measurement Schedule .schedule
Graph Scratchpad .scratch
Fit Log or Fit Results .fit

1-8 • Introduction CompleteEASE Software Manual

 1.4. File Dialog
CompleteEASE has a custom File Dialog to help you organize and locate your files
and folders. Key features are listed below:
• Tab for Recent Folders and Recently Opened Files
• Tab for Projects to list user-relevant folder locations
• Tab for Library files provided by J.A. Woollam
• Ability to drag and drop files between folders
• Ability to clear the listed folders without deleting the actual folders from
your computer – keeps file dialog clean and relevant
• Maintains different folder and file lists for each user (based on login).

Figure 1-6. File Dialog for models.

The tab you select from this dialog box (Recent, Projects, Library) will remain
“sticky” for each file type. For example, if the Library tab was last selected in the
Open Model dialog, that tab will remain selected the next time the Open Model
dialog is shown.
In addition, the selected project will also remain “sticky” for each file type. For
example, if the “research” folder shown above is selected in the Open Model dialog,
then it will be selected when this dialog box is shown next. This allows you to return
quickly to the same location again and again when working on the same project.
Each tab is described in more detail in the sections below.

Recent
Each of the file dialog boxes used for opening or saving files contains a “Recent” tab
which enables you to quickly find files and folders you have recently worked in.
CompleteEASE auto-populates the list of recent folders. This list excludes
COMMON files, as these are always available from the Projects or Library tabs. You
can add folders to this list using drag-and-drop from Windows Explorer.
First in the recent list is a blue folder for “Recently Opened Files”. This is
convenient when working with files across multiple directories.

CompleteEASE Software Manual Introduction • 1-9

 Select any folder in the Recent tab to show all applicable files.

Figure 1-7. Recent tab.

Add to Projects List

You can right-click any folder in the Recent tab and select “Add to Projects list” to
add the selected folder to the Projects tab. This may be preferred if the folder will
be regularly accessed or to create or view sub-folders within this location. The
Recent tab does not allow folders to be expanded.

Figure 1-8. Right-click on a Folder in the “Recent” list to Add that folder to the Projects list or
to Clear the Recent Folders list.

Clear Recent Folders List

You can right-click anywhere in the Recent tab and select “Clear Recent Folders
List”. This will not remove or affect the blue Recently Opened Files folder.

Browse File System

If the desired file or folder is not listed under the Recent, Projects, or Library tabs,
use the ‘Browse File System’ button within the Recent tab to browse the entire
Windows file system. When you select a file from this window, the folder location

1-10 • Introduction CompleteEASE Software Manual

 will be added to the Recent list and can be added to the Projects tab using the right-
click menu described above.

Figure 1-9. Browse File System dialog box.

Projects
The Projects tab is not auto-populated by CompleteEASE. This tab can be used to
create a list of folders relevant to you, which is custom to each user (based on
login). Folders can be located anywhere on the computer or remote network drive
and will be shown together in a single list in this tab. The COMMON folder is always
included here and is accessible to all users.
Folders can be added to the Projects tab by:
• Adding from Recent tab (see Add to Projects List section above)
• Drag-and-drop from Windows Explorer
• ‘Add Folder Link’ button (see Add Folder Link section below)
Folders in the Projects tab can be expanded to show subfolders up to three levels
deep.

Figure 1-10. Projects tab.

CompleteEASE Software Manual Introduction • 1-11

 Remove From List
Folders can be removed from Projects list by right-clicking the top-level folder and
selecting “Remove From List”. This will not delete the folder from your computer; it
will only remove the folder link from your Project list in CompleteEASE. The
COMMON folder cannot be removed from Projects list.

Figure 1-11. Right-click on a Folder in the “Projects” list to remove that folder from the list or
to create subfolders within that Project folder.

Create New Sub Folder

Subfolders can be created within Project folders by right-clicking a 1st or 2nd level
folder and selecting “Create New Sub Folder”. This will create a new subfolder on
your computer wherever the folder is located. You cannot create subfolders in the
COMMON folder or in a 3rd level folder from the Projects tab.

Add Folder Link

If you would like to add a folder to the Projects tab, click on ‘Add Folder Link’. This
will open the system file dialog window and allow you to browse and find the
desired folder. When found, click ‘Select Folder’ to add it to the Projects tab for
easier saving/loading of files within this folder.

Figure 1-12. “Add Folder Link” opens global file dialog to search for a Folder to add to the
Projects list.

1-12 • Introduction CompleteEASE Software Manual

 Library
The Library tab contains the read-only folders located within the COMMON folder
location, as described in the previous section. For example, the three folders in
Figure 1-13 (Advanced, Basic, and Calibration Wafers) are all located within the
C:\CompleteEASE\MOD directory and contain COMMON models provided with
CompleteEASE. The Library tab is available for models, material files, and data.

Figure 1-13. Library tab.

The Library tab is disabled when saving files, as it is not intended for you to add
additional files into these read-only locations. However, additional folders can be
created in Windows Explorer and populated with special files that you may choose
to archive and wish to prevent other users from changing (as they will be read-only
files and folders). For example, five additional folders shown in Figure 1-14 were
added into the C:\CompletEASE\MOD location using Windows Explorer prior to
launching CompleteEASE. These folders are read-only, and no files can be added
into these folders from within CompleteEASE. Files can still be added or deleted
from Windows Explorer.

CompleteEASE Software Manual Introduction • 1-13

 Figure 1-14. Open Model Library tab with extra folders added.

Opening Files from Windows Explorer

Data (.SE, .iSE) and snapshot (.SEsnap) files can be opened directly from Windows
Explorer by double-clicking on the file. This will add the folder location for that file
to the Recent tab. It will also add the file into the Recently Opened Files list.

1.5. Multiple Users

To have your own Project lists within CompleteEASE, the software must be
configured for multiple users.
To do so, click on Options > Configuration Controls > ‘Edit Configuration’.

Figure 1-15. Edit Configuration.

1-14 • Introduction CompleteEASE Software Manual

 Figure 1-16. CompleteEASE Configuration window.

From the CompleteEASE Configuration window, expand the General section, scroll
to “User Log-In Mode = ” . Change User Log-In Mode to “Must Log In”, as shown in
Figure 1-17.

Figure 1-17. User Log-In Mode option.

Next, scroll to “Users Have Own Folders List = ” and check the box to turn this
parameter ON.

CompleteEASE Software Manual Introduction • 1-15

 Figure 1-18. Users Have Own Folders List option.

Optionally, you can decide whether to have all CompleteEASE folders visible to all
users by changing the “Always Add CE Folder” to True (Parameter ON) or False
(Parameter OFF). With this ON, all folders within the CompleteEASE directory on
your computer will show up for all users.

Figure 1-19. Always Add CE Folders option.

NOTE: CompleteEASE must be restarted to implement changes.

After configuring CompleteEASE for multiple users, you will want to set up each
User with their own account. Click Options > Miscellaneous > ‘Manage Users’ and
the User Management box will appear.

1-16 • Introduction CompleteEASE Software Manual

 Figure 1-20. Manage Users.

Figure 1-21. User Management window.

From the User Management window, click ‘Add New User’ and type in the name
and password for each user you wish to add.

NOTE: If the password area is left blank, no password will be required to log-in.

NOTE: CompleteEASE must be restarted to implement changes.

When CompleteEASE is restarted, it will open by asking for User Name and
Password. User names will be available from a drop-down list.

CompleteEASE Software Manual Introduction • 1-17

 Figure 1-22. User Log-In window.

Any folders that are added by an individual user will now be shown for their
account only. Folders within the CompleteEASE directory will be shown for all users
if “Always Show CE Folders” is configured ON. CompleteEASE COMMON folders will
be shown for all users.

1.6. Shortcut Guide

There are two types of shortcuts within CompleteEASE. First, like most software, if
a letter is shown that is underlined, the shortcut for that command is to hold down
the ALT button and press that letter on the keyboard. Additional shortcuts that
have a global scope incorporate the CTRL, SHIFT, and ALT buttons. These additional
shortcuts are shown in Table 1-3 and within CompleteEASE from:
Options > Miscellaneous > ‘View CompleteEASE Shortcuts’.

Table 1-3. Shortcut reference.

SHORTCUT FUNCTION
CTRL+M Go to Measurement Tab
CTRL+A Go to Analysis Tab
CTRL+H Go to Hardware Tab
CTRL+O Go to Options Tab
CTRL+U Go to In-situ Tab
CTRL+P Graph the Psi data curves
CTRL+D Graph the Delta data curves
CTRL+N Graph the “N” data curves (N,C,S format)
CTRL+C Graph the “C” data curves (N,C,S format)
CTRL+S Graph the “S” data curves (N,C,S format)
CTRL+1 Graph the <e1> data curves
CTRL+2 Graph the <e2> data curves
CTRL+I Graph the Intensity data
CTRL+Z Graph the Depolarization data

1-18 • Introduction CompleteEASE Software Manual

 CTRL+T Toggle between Graph Types
CTRL+G Show Graph ‘Data’ Menu
CTRL+L Add to Fit Log
CTRL+ALT+L View Fit Log
CTRL+R Copy Analysis Report to Clipboard
CTRL+Y Toggle between single and double Y axis
CTRL+ALT+O Toggle Default Optical Constant units between “e1 &
e2” and “n & k”
CTRL+ALT+W Toggle Default wavelength units between “nm” and
“eV”
CTRL+ALT+S Show Graph Statistics
ALT++ Add Layer to Model
CTRL+ALT++ Add Layer to Top of Model Stack
CTRL+ALT+SHIFT+U Add the In-situ Tab
SHIFT+Mouse Roller When positioned over model parameter, it will
increase or decrease this parameter.
CTRL+SHIFT+Mouse Same as SHIFT-Mouse roller, but with smaller
Roller parameter increments.

If you are working with Uniformity Mapping data, the shortcuts in Table 1-4 can
help manipulate the data and graphs.

Table 1-4. Shortcut reference for mapping data.

SHORTCUT FUNCTION
CTRL+Click on Point Select/Un-select Point
CTRL+ALT+Click on Point Delete Point
CTRL+ALT+SHIFT+Click on Point Show camera image from point (when
available)
CTRL+space bar Switch view from single point to map

Finally, when working with in-situ data or multi-sample analysis, there are
additional shortcuts to help quickly switch between different data sets and models,
as shown in Table 1-5.

Table 1-5. Shortcut reference for in-situ or multi-sample analysis.

SHORTCUT FUNCTION
CTRL+(F1, F2, F3...) Switch between models 1, 2, 3...
ALT+(1,2,3...) Switch between data sets 1, 2, 3...

CompleteEASE Software Manual Introduction • 1-19

 1.7. What is Spectroscopic Ellipsometry?
Ellipsometry is a non-destructive optical technique in which the sample to be
characterized is illuminated with a beam of polarized light, as shown in Figure 1-23.
Ellipsometry measures the change in polarization state of the measurement beam
induced by reflection from (or transmission through) the sample. The change in
polarization state is commonly presented in the form of the ellipsometric
parameters Psi () and Delta () defined in Equation 1-1:

Equation 1-1

𝑟̃p
tan Ψ ⋅ eiΔ = ρ̃ =
𝑟̃s

In this equation, rho (𝜌̃) is defined as the ratio of the complex reflection coefficient
for p-polarized light (𝑟̃p , polarized parallel to the plan of incidence) divided by the
complex reflection coefficient for s-polarized light (𝑟̃s , polarized perpendicular to the
plane of incidence). ρ̃ is a complex number, and the ellipsometric parameters
simply report this value in polar form: tan() is the magnitude of the reflectivity
ratio, and  is the phase. Ellipsometry offers three main advantages over simple
intensity-based reflection or transmission measurements:
1. Precision As the change in polarization state is defined by a ratio,
ellipsometry is not sensitive to changes in the absolute intensity of
the measurement beam (i.e., the sample itself is the ‘reference’ for
the measurement). Thus, it can remain accurate without knowing
“absolute” intensity values.
2. Sensitivity The phase information contained in the ellipsometric 
parameter provides enhanced sensitivity to ultra-thin films, even
down to the sub-nanometer thickness level.
3. Information Ellipsometry measures two values (Psi and Delta) at
each wavelength, doubling the information content compared to
an intensity reflection or transmission measurement.
The high accuracy, precision, and sensitivity of the ellipsometric measurement
make it highly suited for demanding thin-film metrology applications.
1. linearly polarized light ...
E p-plane

s-plane 3. elliptically polarized light !

p-plane
E
s-plane

plane of incidence

2. reflect off sample ...

Figure 1-23. Interaction of polarized light with a sample.

1-20 • Introduction CompleteEASE Software Manual

 In Spectroscopic Ellipsometry (SE), Psi and Delta values are acquired as a function of
wavelength. This greatly increases the information content in the data set, enabling
the simultaneous determination of multiple sample properties. However, to extract
sample parameters such as film thickness and optical constants from the measured
SE data set, an optical model must be built and compared to the experimental data.
CompleteEASE provides a graphical user interface for building models and
displaying measured and model calculated data. CompleteEASE also provides a
simple interface to operate the ellipsometer hardware, making acquisition of
accurate SE data fast and easy.

For more information on ellipsometry theory, see the following resources:

1. H. Tompkins and J. Hilfiker, Spectroscopic Ellipsometry: Practical

Application to Thin Film Characterization, Momentum Press, New York,
2016.
2. H. Tompkins and E. Irene, eds. Handbook of Ellipsometry, William Andrew
Publishing, New York, 2005.
3. H. Fujiwara, Spectroscopic Ellipsometry Principles and Applications John
Wiley & Sons, West Sussex, England 2007.
4. J.N. Hilfiker and J.A. Woollam, Ellipsometry, in Encyclopedia of Modern
Optics, edited by Robert D. Guenther, Duncan G. Steel and Leopold Bayvel,
Elsevier, Oxford, 2004.
5. J. A. Woollam et al., “Overview of Variable Angle Spectroscopic
Ellipsometry (VASE), Part I: Basic Theory and Typical Applications”, SPIE
Proc. CR72 (1999) 3.
6. B. Johs et al., “Overview of Variable Angle Spectroscopic Ellipsometry
(VASE), Part II: Advanced Applications”, SPIE Proc. CR72 (1999) 29.
7. H.G. Tompkins and W.A. McGahan, Spectroscopic Ellipsometry and
Reflectometry, John Wiley & Sons, 1999.
8. R.M.A. Azzam, and N.M. Bashara, Ellipsometry and Polarized Light, North
Holland Press, Amsterdam 1977, Second edition 1987.

CompleteEASE Software Manual Introduction • 1-21

2. Data Acquisition

The first step in any ellipsometry experiment is acquisition of data from the sample
with the ellipsometer. Details for data acquisition depend on the ellipsometer type
and configuration. This chapter provides a brief overview of data acquisition with
CompleteEASE. More details are provided in Chapter 8, which lists all features of
the Measurement Tab, and in the corresponding hardware manual for your system,
which can be found by pressing the F2 key on the instrument computer while
CompleteEASE is running.

NOTE: Instructions in this chapter are representative of various systems. For more
complete instructions, including details on system alignment and calibration, please
consult your hardware manual.

2.1. alpha-SE Systems

An example Measurement tab for an alpha-SE is shown in Figure 2-1.

Figure 2-1. Measurement tab for alpha-SE systems. The camera accessory is optional and
may not be available on your system.

The Measurement Controls panel contains all options for measurement. For an
alpha-SE, you must specify measurement mode, type of sample alignment,
measurement angles, model for analysis (optional), and “Save Data after
Measurement”, if desired. “Mode”, “Sample Alignment”, and “Model” are selected
from a drop-down menu, as shown in Figure 2-2.

2-22 • Data Acquisition CompleteEASE Software Manual

 Figure 2-2. Options for measurement Mode, Sample Alignment, and Model. Note your listed
model choices will differ depending on which models are in your common directory or opened
recently.

For most measurements, “Standard” Mode and “Standard” Sample Alignment are
adequate. A Model can be developed and applied later if not available at time of
measurement.
Once the Measurement Controls options are set, click ‘Measure’ to begin.
CompleteEASE will prompt you to change angle if needed. The alpha-SE will
automatically align the z-height according to the selected Sample Alignment option
and then acquire data according to the selected Mode. Finally, if a Model was
selected, the data will be analyzed to determine the resulting film properties.

Figure 2-3. Clicking ‘Measure’ will start the data acquisition procedure. The System Status
will be updated to show the current hardware operation.

If the “Save Data after Measurement” checkbox was selected, a dialog box will
appear, as shown in Figure 2-4. Navigate to the desired location, enter a File Name
and Comment (optional), and click ‘Save’. See Section 1.4 for details regarding the
CompleteEASE File Dialog.

NOTE: CompleteEASE data files are encrypted and saved with “.SE” or “.iSE”
extensions. These files are only accessible by CompleteEASE.

CompleteEASE Software Manual Data Acquisition • 2-23

 Figure 2-4. Save Data dialog box.

If a model was selected prior to measurement, the data will automatically be

analyzed after measurement according to the selected model. Resulting fit
parameters will be stored with the data set and displayed in the Fit Results panel.
The model fit to the measured data is also displayed in the graph. For a good fit to
the data, the black dashed “Model” curves should lie essentially on top of the
colored Psi and Delta measured data curves. If this is not the case, a different
model or changes to the existing model may be required.
As an example, the result for a thin oxide on silicon is shown in Figure 2-5. Model
parameters are reported in the Fit Results panel, and a graph of Psi and Delta
versus wavelength is displayed at the bottom.
If no model was selected before the measurement, the Fit Results panel will display
“No Results Available”. The data can be analyzed later from the Analysis tab.

Figure 2-5. Successful measurement of thin oxide film using an alpha-SE system.

2-24 • Data Acquisition CompleteEASE Software Manual

 2.2. M-2000 and RC2 Systems
There are many options and accessories for M-2000 and RC2 systems, leading to
various configurations and measurement options. This chapter will discuss basic
options for typical systems, but you should consult your hardware manual for
specific system and configuration details.
Figure 2-6 shows a representative Measurement tab for an M-2000 with a mapping
stage and camera. In the Measurement Controls panel, there is a drop-down menu
for “Recipe”. Recipes contain all information about what to measure, where to
measure, and how to analyze. Recipes can be created and saved for any
measurement or sample type.

Figure 2-6. Measurement tab for an M-2000 system with automated angle of incidence,
automated sample alignment, automated sample translation, focusing optics, and an
integrated camera.

Recipes
Each Recipe contains three basic components of data acquisition:
1) Acquisition Parameters
2) Scan Pattern (if applicable)
3) Model (optional)

CompleteEASE Software Manual Data Acquisition • 2-25

 The Recipe drop-down menu contains the following options:
• Prompt for Acquisition Parameters
• Prompt for Recipe Components
• List of COMMON recipes (saved in C:\CompleteEASE\recipe)
• ‘Choose From File Dialog’
• ‘Create/Edit Recipe’

These options are shown in Figure 2-7 and more details are provided in the
following sections.

Figure 2-7. Recipe Options.

<Prompt for Acquisition Parameters>

To measure data at a single location for later analysis, the quickest selection will be
<Prompt for Acquisition Parameters>. This option will allow you to specify the
acquisition parameters but not a scan pattern or model. Acquisition parameters
will be selected from a dialog box like Figure 2-8. Options shown in this window are
detailed later in this chapter.

2-26 • Data Acquisition CompleteEASE Software Manual

 Figure 2-8. Acquisition Parameters Setup dialog box.

<Prompt for Recipe Components>

Choose this option to quickly utilize previously saved acquisition parameters, scan
pattern (if applicable), and model (optional). A dialog box like Figure 2-9 will appear
and each component can be selected from a drop-down list. Click ‘Ok’ to start the
measurement. More details regarding each component will be provided later in
this chapter.

Figure 2-9. Choose Recipe Components dialog box.

‘Choose From File Dialog’

This button allows you to choose an existing recipe previously saved by navigating
to the recipe file.

CompleteEASE Software Manual Data Acquisition • 2-27

 ‘Create/Edit Recipe’
This button allows you to create or edit a recipe and save it for later use. You can
select previously saved acquisition parameters, scan patterns, and models from the
drop-down lists. You can edit/create new acquisition parameters and scan patterns
from the ‘Edit/Create’ buttons to the right these components. More details
regarding each component will be provided later in this chapter.

Figure 2-10. Create/Edit Recipe dialog box.

Acquisition Parameters
Acquisition parameters describe how the data will be collected at each location. A
typical acquisition parameters setup dialog box is shown in Figure 2-11. The
options available will depend on your system configuration. Only basic details are
provided in this chapter. Please see Chapter 8 and consult your hardware manual
for more details.

2-28 • Data Acquisition CompleteEASE Software Manual

 Figure 2-11. Acquisition Parameters Setup dialog box for M-2000 and RC2 systems. Choices
listed above can vary from system to system.

Data Type
“Data Type” options are shown in Figure 2-12. For most applications, “Standard”
will be appropriate.

Figure 2-12. Data Type options.

Acq. Time
This box sets the averaging time. For most samples, 1 to 5 seconds will yield good
data. Low reflectivity samples may require longer averaging than high reflectivity
samples.

Angle Scan
This selection sets the angle(s) of incidence to be used during measurement. For
most applications, two to three angles between 50° and 75° are recommended.

NOTE: The following settings will work well for many applications: Data Type =
Standard, Acq. Time = 3, Angle Scan = 55 to 75 by 10.

CompleteEASE Software Manual Data Acquisition • 2-29

 Alignment Options depend on the type of system and configuration. In general,
there are two basic steps to sample alignment: sample tilt alignment and sample
height alignment. Typical selections for each are described next.

Sample Tilt Alignment

Sample tilt alignment refers to the tip and tilt of the sample relative to the incoming
light beam such that the reflected beam is aligned on the receiver unit at the
correct angle of incidence. For systems with automated sample tilt, sample tilt
alignment can be set to Skip, Manual, or Automated, as shown in Figure 2-13.

Figure 2-13. Options for Sample Tilt Alignment.

Sample tilt alignment is often skipped when using focusing probes, as a focused
beam is less sensitive to sample tilt. It can also be skipped for applications where
common samples are measured that remain flat. For example, if a system is
continually used to measure 200mm silicon wafers, the tilt of each wafer will
probably be very similar from sample-to-sample.
For manual or automatic sample tilt alignment, the system will study the reflection
of a light beam off the sample that is incident on a four-quadrant detector or
camera. An example of the tilt alignment screen for a system with automatic four-
quadrant detector tilt alignment is shown in Figure 2-14.

Figure 2-14. Sample Tilt Alignment window.

2-30 • Data Acquisition CompleteEASE Software Manual

 Sample Height Alignment
Sample height alignment refers to adjustment of the ellipsometer beam relative to
the surface of a sample to ensure correct positioning and adjustment for samples of
different thicknesses. This adjustment becomes increasingly important for systems
with a small or focused spot. It is also important to perform height alignment when
many samples of different thicknesses will be measured on the same ellipsometer.
Typical options for sample height alignment are skip, manual, automatic-quick, and
automatic-robust. Other options may exist depending on your hardware. See
Figure 2-15 for a complete list of intensity- and camera-based options for sample
height alignment.

Figure 2-15. Options for Sample Height Alignment.

For automatic-quick and automatic-robust (most commonly used), the reflected

beam is detected by the receiver unit and the intensity is recorded versus different
z-height positions. This is demonstrated in Figure 2-16. The position of highest
intensity is assumed to be the correct z-height for alignment of the sample.

Figure 2-16. Sample Height Alignment window.

CompleteEASE Software Manual Data Acquisition • 2-31

 Scan Pattern
Scan patterns are available for systems with automated sample translation and are
used to describe where the sample should be measured across a sample. To create
a new scan pattern, click the ‘Edit/Create’ button to the right of this component in
the Choose Recipe Components dialog box, as shown in Figure 2-17.

Figure 2-17. Choose ‘Edit/Create’ Scan Pattern from the Choose Recipe Components dialog
box.

The Scan Patter Editor, shown in Figure 2-18, allows you to describe both circular
and rectangular samples. There are many options to automatically fill-in the points
to measure, including ‘Cartesian Grid Fill’, ‘Polar Grid Fill’, and ‘Line Fill’. Most of
these selections are intuitive and best learned by testing each button a few times.

Figure 2-18. Scan Pattern Editor.

2-32 • Data Acquisition CompleteEASE Software Manual

 Model
A model is a description of the sample structure and includes instructions for
CompleteEASE to determine film thickness, optical constants, or other material
properties with designated fit parameters and strategies. The model selected to
analyze data during a recipe should be appropriate for the sample and purpose of
the measurement. Models will be described in significant detail in the upcoming
chapters.
CompleteEASE has a several prebuilt models included with the installation of the
software. These can be found in the Library tab and are divided into “Advanced”,
“Basic”, and “Calibration Wafers” folders, as shown in Figure 2-19. The “Basic”
folder contains several options for transparent or absorbing films on silicon or glass
substrates, which can also be used as starting models for similar structures on other
substrate types.

Figure 2-19. From Analysis>Model>‘Open Model’, choose the Library tab, “Basic” folder on
the left to access several prebuilt models for glass or silicon substrates.

CompleteEASE Software Manual Data Acquisition • 2-33

 Running a Recipe and Viewing Results
After selecting or building a recipe, click ‘Measure’. CompleteEASE will collect data
according to the acquisition parameters selected at each of the locations described
in the scan pattern (if applicable). After each measurement, the data will be
analyzed using the selected model (if selected). An example of this process is
shown in Figure 2-20. System Status will show the current hardware operation. Fit
Results will be updated with the model fit results from each point. Scan Map will
show the status of a pattern map (if applicable) where green points are completed
locations, the blue point is the current location, and red points are locations yet to
be measured. The graph will update with each location.

Figure 2-20. CompleteEASE during a Recipe measurement with multiple points.

For maps of multiple points, the graph can be toggled to display parameters for all
completed locations by clicking Measurement > Fit Results > ‘Graph Parms’.
Parameters can be selected from a drop-down list above the graph, as shown in
Figure 2-21.
When the recipe is finished, a table of results will be displayed in the Fit Results
panel, as shown in Figure 2-22.

2-34 • Data Acquisition CompleteEASE Software Manual

 Figure 2-21. Viewing parameter results of a recipe in progress.

Figure 2-22. Viewing results after a recipe is completed.

If “Save Data after Measurement” was selected prior to starting the recipe, results
of the analysis are saved with the measurement data and can be accessed at any
time by clicking Measurement > Fit Results > ‘View Prev. Results’. Choose the file
location and file to display a tabulated list of results. Click ‘Graph Data’ from this
window to select parameters to graph. See Figure 2-23 and Figure 2-24.

CompleteEASE Software Manual Data Acquisition • 2-35

 Figure 2-23. Viewing previous results.

Figure 2-24. Graphing previous results.

2-36 • Data Acquisition CompleteEASE Software Manual

3. Data Analysis 1 – Basic

The best way to learn software features and data analysis is through a series of real-
world examples where you can follow along. To this end, we have developed
examples that spread over multiple chapters to teach common data analysis
procedures in CompleteEASE. Following this tutorial is instructive for beginner and
expert users. This chapter begins with basic data analysis examples. Here, we
concentrate on thin films that are transparent. Chapter 4 continues with
intermediate concepts, concentrating on absorbing materials and the extra
difficulties involved with their characterization. Chapter 5 introduces a few
advanced topics such as multi-sample analysis and anisotropic materials. Finally,
Chapter 6 introduces in-situ data analysis methods.
The section examples from this chapter are listed below, along with the primary
CompleteEASE features that are discussed.

Section 3.1 General Data Analysis Theory

Section 3.2 Samples with Known Optical Constants: SiO2 on Si

FEATURES IN THIS EXAMPLE

• Opening model and data files • Fitting data
• Parameter error bars • Hiding or renaming fit parameters
• Visualizing model changes • Thickness pre-fit
• Reporting “derived” parameters • Defining fit parameters
(turning on and off)

Section 3.3 Transparent Thin Films 1: PECVD SiO2 on Si

FEATURES IN THIS EXAMPLE

• Cauchy dispersion equation • Displaying optical constants
• Global fit procedures • Expanding layers
• Renaming layers

Section 3.4 Transparent Substrates: Bare Glass

FEATURES IN THIS EXAMPLE

• Graphing multiple data sets • Using the graph scratchpad
• Including backside reflections • Depolarization data

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-37

 Section 3.5 Transparent Films 2: SiC on Si

FEATURES IN THIS EXAMPLE

• Surface roughness • Simple grading
• Simple anisotropy • Try alternate models

Section 3.6 Log and Report a Series of Films

FEATURES IN THIS EXAMPLE

• Fit log • Copy analysis report to clipboard
• Open and save snapshots • Re-analyzing data from log
• Copying to clipboard • Comparing optical constants in fit
(formatted or table) log

3-38 • Data Analysis 1 – Basic CompleteEASE Software Manual

 3.1. General Data Analysis Theory
Data analysis is a very important part of spectroscopic ellipsometry (SE): without
data analysis, SE measures only the ellipsometric parameters Psi and Delta versus
wavelength. To determine sample properties of interest, such as layer thicknesses
and optical constants, a model-based analysis of the SE data must typically be
performed. The only exception to this is in characterizing the optical constants of
bulk samples, which is described on the next page. The model-based analysis
approach is summarized by the flowchart shown in Figure 3-1. The basic steps of
this approach are:
1. SE data is measured on the sample.
2. A layered optical model is built which represents the nominal structure of the
sample. This model is used to “generate” SE data.
3. Model fit parameters are defined and then automatically adjusted by
CompleteEASE to improve the agreement between the measured and model-
generated SE data. This is known as “fitting” the data.
4. The results of the fit are evaluated. If the results are not acceptable, the optical
model and/or defined fit parameters are modified, and the data is fit again.

Figure 3-1. SE Data Analysis flowchart.

The above steps will be described in more detail in subsequent sections. Keep in
mind that while the basic SE data analysis approach is straightforward, “real-world”
samples can often be difficult to analyze. CompleteEASE is designed to simplify
common SE data analysis procedures which work for many types of samples, but
there is no substitute for experience when dealing with complex samples. Please
contact your J.A. Woollam representative if you require assistance.

Layered Optical Model

The SE data analysis process for most samples begins by building a layered optical
model which corresponds to the nominal sample structure as shown in Figure 3-2.
Each layer is parameterized by a thickness (t1, t2, etc.) and optical constants. Optical
constants describe how light interacts with and propagates through the layer.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-39

 Using the optical model and standard textbook thin film equations (Snell’s law,
Fresnel equations, etc.), CompleteEASE can calculate “generated” or “simulated” SE
data. If the model is a good representation of the sample, the model-generated SE
data will be in good agreement with the SE data measured on the sample, as is
illustrated in Figure 3-3.

t2 n,k (film 2)

t1 n,k (film 1)

n,k (substrate)
Figure 3-2. Schematic representation of a layered optical model with two films,
parameterized by thicknesses t1 and t2, and optical constants n & k.

(a)

(b)
Figure 3-3. Layered optical model in CompleteEASE with model-generated data: (a) the
model-generated data (black dashed lines) are not in agreement with the measured SE data
(colored curves); (b) when the top layer thickness is adjusted appropriately, the model-
generated data lies directly on the measured data.

3-40 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Surface Roughness
The layered optical model as shown in Figure 3-2 indicates a perfectly smooth
surface, but in reality, sample surfaces often have some roughness. Optically,
roughness is like a lower-index surface film, as the roughness is a mixture of base
material and air (n=1, k=0). CompleteEASE models this “layer” using the
Bruggeman Effective Medium Approximation (EMA). Surface roughness in
CompleteEASE is a mixed layer assuming 50% void content and 50% base material.
A schematic of a sample with surface roughness is shown in Figure 3-4. This
approach is only an approximation to the actual sample but works extremely well
for modeling SE data when the size of the surface roughness is much less than the
wavelength of light used to measure the sample. For most SE systems, this implies
that the surface roughness features must be less than ~40nm.

(a) (b)
Figure 3-4. Schematic of: (a) physical surface roughness; (b) optical model of surface
roughness.

Grading
Another common non-ideality is that the optical properties of real samples are not
constant through layer depth. This may be caused by process variations during the
film deposition. CompleteEASE models these variations, called gradients, by
dividing the layer into smaller sublayers, with each sublayer having slightly different
optical properties, as shown in Figure 3-5.

(a) (b)
Figure 3-5. Schematic of: (a) continuous variation in the optical properties of a graded film,
(b) CompleteEASE approximation of a graded film with discrete layers.

Anisotropy
Some materials have different optical properties depending on polarization and
propagation direction of the light beam. This is called anisotropy, which is a broad
topic that is not covered fully in this section. Anisotropy can be uniaxial (where nx =
ny ≠ nz) or biaxial (where nx ≠ ny ≠ nz).

nx
ny nz

Figure 3-6. Anisotropic materials have different values for nx, ny, and/or nz.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-41

 Goodness of Fit: Definition of the MSE
Quantifying how well the optical model agrees with the measured data is an
important part of the SE data analysis process. In Figure 3-3, you can easily see that
model (b) fits the data better than model (a). However, to automate the analysis
process, this assessment is quantified by the “MSE” value defined in Equation 3-1:

Equation 3-1

n  N 
2 2 2
1  − N Gi   CEi − CGi   S Ei − SGi 
MSENCS =   Ei
3n − m i =1  .001
 +   +   
   .001   .001  

where “n” is the number of wavelengths, “m” is the number of fit parameters,
N=cos(2), C=sin(2)cos(), and S=sin(2)sin().
“MSE” is an acronym for Mean Squared Error. The definition given in Equation 3-1
is more accurately termed “Root Mean Squared Error” (as the error summation is
divided by the number of measurement points minus the number of fit parameters,
and the square root is taken). However, for historical reasons, we still use the term
“MSE” to quantify the goodness of fit.
Basically, the MSE sums, over all the measurement wavelengths, the differences
between the measured data (parameters subscripted with “E”) and model-
generated data (parameters subscripted with “G”). In CompleteEASE, the
difference between the measured and model-generated data is calculated in terms
of the “N”, “C”, and “S” parameters, which are derived from the ellipsometric Psi
() and Delta () parameters. The N, C, & S parameters have the following
properties which make them better suited for the MSE definition (compared to
using  and ):
1. N, C, & S are always bounded between -1.0 and 1.0.
2. The rotating compensator ellipsometer configuration measures data with
approximately the same precision and accuracy in N, C, & S on any sample.
The N, C, & S parameters for the current data set in CompleteEASE can be graphed
by selecting N, C, or S from the “+Data” menu above the graph, as shown in Figure
3-7.

Figure 3-7. Select N, C, or S option from the +Data menu to display corresponding data.

3-42 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Lower MSE values indicate better agreement between the measured and model-
generated SE data. As the typical precision and accuracy of the measured SE data in
terms of N, C, & S is ~0.001, a multiplicative factor of “1000” is included in the MSE
definition. This implies that an ideal model fit should have an MSE of ~1. While this
is achieved on certain types of samples (single thin films on Si substrates, for
example), the best model fits for more complex samples (with thick and/or multiple
layers) may exhibit much larger MSEs (>10) and still be considered acceptable. It is
also possible to get an MSE smaller than 1, but this implies the match to data is
better than accounted for with random measurement error. In other words, the
results likely include some of the “noise” present from data set.
In general, the MSE should be understood as a relative quantity specifying the
match between the model-generated and experimental data for a specific type of
sample. A smaller MSE implies a better match, but there is no target value that
must be achieved.

“Fitting” Data
To obtain the best agreement between the model-generated and experimentally
measured data (or in other words, to achieve the lowest MSE), parameters of the
optical model must be adjusted, and you must define which model parameters
should be adjusted. These adjustable parameters are also called fit parameters,
and they typically consist of layer thicknesses and parameters which define the
optical properties of the layers. CompleteEASE uses a standard, iterative, non-
linear regression algorithm (the Levenberg-Marquardt method) to automatically
minimize the MSE by adjusting the fit parameters. The Levenberg-Marquardt
method does an excellent job of quickly converging to the best (lowest) MSE value
if: 1) the optical model accurately represents the sample and 2) the initial starting
values for the fit parameters are reasonably close to their correct (best fit) values.
To help ensure that the best fit model is found, CompleteEASE includes the
following powerful features:
• “Try Alternate Models”: this command automatically fits the data using
the ideal layered optical model, then with surface roughness, index
grading, and simple anisotropy. The results of the various model fits are
summarized in tabulated format, and a suggestion is given for which model
is most appropriate for the sample.
• “Thickness Pre-Fit” and “Global Fit”: these options perform a large
number of trial fits to the data set, using a wide range of initial starting
values for selected fit parameters to find the best fit model.
These features will be demonstrated in data analysis examples presented later in
this chapter.

Evaluating the Fit Results

Evaluating the fit results is arguably the most important (and unfortunately the
most overlooked) aspect of SE data analysis. For a model fit to be acceptable, the
following requirements must be simultaneously met:
1. The model-generated data must fit the measured (experimental) data.
The simplest evaluation is visual: the model generated curves should
essentially lie on top of the experimental curves, reproducing all structure
and features (except noise). MSE should also be considered, but it is

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-43

 difficult to specify a target MSE value for an unknown sample. A good fit
for thin-film samples with relatively little data structure will typically have
MSE around 1-5. It is common to see higher MSE (10-20 or higher) for
thicker films, multilayers, graded-index materials, anisotropic films, and
other sample complexities. However, this can still be considered a good
fit. For any sample, the goal is not to achieve a specific MSE value but to
minimize the MSE while still meeting criteria #2 and #3 below.
2. The model should be unique.
It is always possible to add more layers, complexity, or fit parameters to
the model, but unless the MSE is significantly reduced, the additions may
not be justified. You should generally choose the simplest model that
adequately fits the data. As a guideline, the MSE should improve by ~20%
or better to justify added model complexity. Parameter confidence limits,
correlation statistics, and uniqueness tests are available in CompleteEASE
to help you evaluate the parameter sensitivity and uniqueness.
3. The model and fit parameters must be physically plausible.
Evaluating this requirement depends on a fundamental understanding of
the materials and sample structure. Some materials are known to have a
certain index range, absorptions in the UV, roughness, anisotropy, index
gradients, etc. A knowledge base from similar samples is helpful in
evaluating the physical plausibility. For example, many metal dielectrics
(Ta2O5, Nb2O5, etc.) have a graded index, whereas long-chain polymer films
can exhibit optical anisotropy. Transparent conductive oxides have strong
infrared absorption due to free carriers, and this sample property is often
graded through the film depth. Unphysical results should be disregarded.
Some examples of unphysical results include negative thickness values,
negative k values, and optical constants which do not maintain the physical
relationship between n and k (Kramers-Kronig consistency).
Complementary techniques such as stylus profilometry, AFM, XTEM, SIMS
or Auger depth profiling can also be used to corroborate the physicality of
the model fit results.
If the model fit does not meet the above criteria, the results are not trustworthy.
More information about the sample, a different modeling strategy, or changes to
the experimental data acquisition may be helpful or necessary to achieve confident
results.

3-44 • Data Analysis 1 – Basic CompleteEASE Software Manual

 “Physical” Optical Constants
One important condition for any model is that the resulting optical constants must
be physically plausible and maintain the physical relationship between n and k
(Kramers-Kronig consistency). This can be difficult to evaluate without experience.
The following graphs are a few examples of “physical” and “unphysical” optical
constants to introduce this concept.

A. Index (n) must increase toward shorter wavelengths when the material is
transparent (k=0).

Figure 3-8. Index must increase toward shorter wavelengths if k=0. These optical
constants are physical as they obey this rule.

Figure 3-9. Index must increase toward shorter wavelengths if k=0. These optical
constants are not physical because n is decreasing below 600nm without any
absorption (k=0).

B. Index (n) will decrease toward shorter wavelengths if material becomes

absorbing (k increasing).

Figure 3-10. Index will turn over and decrease toward shorter wavelengths when
material becomes absorbing (k increases toward shorter wavelengths). These
optical constants are physical as they obey this rule.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-45

 C. Index (n) will return to the normal behavior of increasing toward shorter
wavelengths as the amount of absorption begins to decrease (k
decreasing).

Figure 3-11. The graphed optical constants are physically plausible, as the index is
increasing for wavelengths above 400nm (where k=0), decreasing when absorption
becomes larger (near 350nm), and then returns to increasing toward shorter
wavelengths when the absorption decreases (below 320nm).

D. Extinction coefficient (k) cannot be negative.

Figure 3-12. The graphed optical constants are unphysical because the extinction
coefficient, k, is negative at the longer wavelengths.

Evaluating the fit results is an important part of SE data analysis. As examples are
presented in the following sections, comments related to evaluation of fit results
will be highlighted yellow.

3.2. Samples with Known Optical Constants: SiO2

on Si
FEATURES IN THIS EXAMPLE
• Opening model and data files • Fitting data
• Parameter error bars • Hiding or renaming fit parameters
• Visualizing model changes • Thickness pre-fit
• Reporting “derived” parameters • Defining fit parameters
(turning on and off)

3-46 • Data Analysis 1 – Basic CompleteEASE Software Manual

 The simplest types of samples to analyze are those in which the optical constants of
each material are known. For this case, only layer thickness needs to be
determined in the optical model fit. Unfortunately, most samples do not fall into
this category, as the “optical constants” for most materials are rarely constant.
Instead, they often depend on the material deposition and processing conditions.
The analysis examples in the following sections demonstrate how to determine
both layer thickness and optical constants from SE data. However, in this section
we show the classic ellipsometry example which assumes known optical constants
for the materials: thermally grown silicon dioxide films on silicon wafer substrates
(SiO2 on Si).
For high quality thermally grown oxide films on semiconductor-grade Si wafers, the
optical constants of both the Si wafer substrate and the SiO2 film are reliably
constant. CompleteEASE has a prebuilt model for this type of sample. To open this
model:
1. Start CompleteEASE (if the program is not open).
2. Select Analysis > Model > ‘Open’.
3. Click on the Library tab of the File Location panel, then the Basic folder.
4. Select “Si with Thermal Oxide.mod” in the Files panel and click ‘Open’.

Figure 3-13. Location of “Si with Thermal Oxide” prebuilt model.

Next, open an example data file that we previously acquired on an ellipsometer

system:
1. Select Analysis > Data > ‘Open’.
2. Click on the Library tab of the File Location panel, then the Examples
folder.
3. Select the “25nm SiO2 on Si.SE” file and click ‘Open’.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-47

 Figure 3-14. Location of “25nm SiO2 on Si” example data set.

Shortcut: After opening the specified model and data files, the CompleteEASE screen should
appear as shown in Figure 3-15. Note the experimentally measured SE data (the
CTRL+Y to turn double-Y
ellipsometric Psi and Delta parameters versus wavelength) appear in the graph
axis on or off as needed
panel as red and green curves, and the optical model contains a substrate
(SI_JAW) and two layers (INTR_JAW and SIO2_JAW).

Figure 3-15. CompleteEASE screen after opening the “Si with Thermal Oxide” model and the
“25nm SiO2 on Si” example data.

Fit Results and Parameter Error Bars

To analyze the data, click Analysis > Fit > ‘Fit’. The analysis will proceed very quickly
(less than one second on most computers), and the results will be displayed in the
Fit panel (Figure 3-16). Of course, the main parameter of interest is the total film
thickness (24.50nm). The value after the “±” is the estimated error bar on each fit
parameter.

3-48 • Data Analysis 1 – Basic CompleteEASE Software Manual

 The error bar values reported by CompleteEASE come from the 90% confidence
intervals that are calculated by the fit algorithm. Since the confidence interval
calculation makes several statistical assumptions that are not, in general, rigorously
satisfied, the error bar values reported by CompleteEASE should not be literally
interpreted. In practice, the error bars provide an estimation of the measurement
reproducibility and uniqueness of the specific parameter during the regression
analysis. Large error bars (relative to the magnitude of the fit parameter) are a
useful indicator that the model is not sensitive to the parameter (and the
parameter should likely be removed from the fit). As a rule of thumb, parameters
are considered uniquely determined if their error bars are smaller than 10% of the
parameter value. Note that after the fit, the black dashed “Model” generated data
curves are essentially on top of the colored experimental data curves.

Figure 3-16. SE data fit and results for SiO2 on Si example.

For more practice, use the “Si with Thermal Oxide” model to determine the total
film thickness for each SiO2 on Si example file: “60nm SiO2 on Si”, “500nm SiO2 on
Si”, and “1000nm SiO2 on Si ”. Open each data file (Analysis > Data > ‘Open’) and
click ‘Fit’; the results should agree with the values shown in Table 3-1. There are a
few important trends to note in this table:
1. The MSE increases with thicker films (from ~1 to >10), though all data fits
are visually good (i.e. black dashed model curves lie on red/green
experimental curves).
2. The reported “Total Thickness” values do not agree exactly with
thicknesses given in the file names (this is acceptable, as the file names are
only nominal film thickness values).

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-49

 Table 3-1. Fit results from “SiO2 on Si” example files.

Example MSE Angle Offset Thickness (nm)

25nm SiO2 on Si 0.688 0.00065837 24.50
60nm SiO2 on Si 1.693 0.0163 67.24
500nm SiO2 on Si 7.420 -0.0619 493.32
1000nm SiO2 on Si 20.357 -0.181 1027.65

The J.A. Woollam “Si with Thermal Oxide” optical model supplied with
CompleteEASE has several unique features which will be described in the following
sections.

Si with Thermal Oxide.mod

First, the oxide film is modeled as two layers: INTR_JAW, which is a higher-index
interface layer, and SIO2_JAW, which contains the optical constants for the
thermally-grown SiO2 film (which are different from the optical constants of bulk
SiO2 or “fused silica”). The optical constants for these layers and for the SI_JAW
substrate were published by Herzinger et al.1 The interface layer adds model
complexity, but it significantly improves the data fit when the model is applied over
a wide oxide thickness range.
To simplify the model and provide more unique fit results, the interface layer
(“Thickness #1”) is held constant at 1 nm. This value is typical for thinner oxide
films, and there is typically insufficient sensitivity to the exact value of this
interfacial layer. For example, open the “500nm SiO2 on Si” example and refit the
data with the “Si with Thermal Oxide” model. Now, add the interface layer
thickness as a fit parameter by right-clicking the mouse on the “Thickness #1”
Turn fit parameters on value (“1.00 nm”). The value is now shown in a bold font followed by the word
or off by right-clicking “(fit)”, indicating that this value is now defined as a fit parameter. (To turn off a fit
with the mouse on the parameter, simply right-click again on the parameter value.) Click on the ‘Fit’
parameter value. button, and the results shown in Figure 3-17 should appear. Note the fit value for
Thickness #1 is 1.63 nm, and the MSE is effectively the same as before. Unless the
explicit goal of the sample analysis is to characterize the interface thickness, it is
recommended to keep Thickness #1 fixed at 1 nm nominal value.

1
C.M. Herzinger, B. Johs, W.A. McGahan, J.A. Woollam, and W. Paulson, “Ellipsometric determination of optical constants for silicon
and thermally grown silicon dioxide via a multi-sample, multi-wavelength, multi-angle investigation”, J. Appl. Phys. 83 (1998) 3323.

3-50 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-17. Results from “Si with Thermal Oxide” model fit to “500nm SiO2 on Si” data with
interface layer (“Thickness #1”) added as a fit parameter.

Derived Parameters
The “Total Thickness” in the “Si with Thermal Oxide” model is not an actual fit
parameter, as the value is derived by summing all layer thicknesses in the model.
Derived parameters can be added to the results by expanding Model > +FIT Options
and turning the “Include Derived Parameters” option ON (Figure 3-18). In addition
to “Total Thickness”, additional derived parameters include n and k values for a
specified layer and wavelength, optical thickness, total optical thickness, and many
other possibilities. These will be addressed further in a later example.

Figure 3-18. “Include Derived Parameter” section of the model.

If you wish to delete a Derived Parameter, simply right-click on the number (shown
in blue) next to the parameter you wish to delete.

NOTE: The “Total Thickness” parameter error bar is simply the error bar of the top-
most layer’s thickness error bar (if the top-most layer’s thickness is turned on as a
fit parameter; if not, it takes the highest layer whose thicknesses is a fit parameter).

Hiding, Renaming, and Bounding Fit Parameters

“Thickness #2” is a fit parameter in the analysis, but this parameter doesn’t show up
in the fit results. “Thickness #2” was hidden to prevent confusion with the “Total
Thickness” that is reported. To hide a fit parameter so that it is not displayed in the

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-51

 Results, left-click on the fit parameter value. For example, if you left-click on
“491.63 nm”, the dialog box shown in Figure 3-19(a) will appear. “Minimum” and
“Maximum” values for the parameter can be specified to constrain the parameter
range during the regression analysis. If the “Show Adv. Parameters” box is checked,
the Parameter Name, Error Bars, and Specification sections are visible. If the “Hide”
box is checked, the fit parameter value will not be displayed as a fit result.

Figure 3-19. Fit parameter dialogs, showing how to (a) “Hide” or (b) “Rename” fit parameters. Low
and High Specs for the parameter are also set in (b).

Edit “Thickness #2” as shown in the Figure 3-19(b) dialog box:

• Check the “Rename” box, type in a different “Name:” for the parameter
(e.g., “Top SiO2”).
• Uncheck the “Hide” box.
• Type in values for “Low Spec.” (300) and “High Spec.” (400).
Click ‘Ok’ to close the dialog box and fit the data again (with the “500nm SiO2 on Si”
data file loaded). Fit results will appear as shown in Figure 3-20. Since the “Top
SiO2” value is outside the parameter Low and High Specs, an “(Out Of Spec.)”
indication appears with the result.

3-52 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-20. Fit results with parameter renamed “Top SiO2” and parameter result “Out Of
Spec”.

Angle Offset
In the “Si with Thermal Oxide” model, “Angle Offset” is also defined as a fit
parameter located within +MODEL Options, as shown in Figure 3-21. The “Angle
Offset” is added to the nominal angle of incidence in the optical model calculation.
For most analyses (and especially when fitting for optical constants) it is best to
leave the Angle Offset fixed at 0. However, since Si and SiO2 optical constants are
so well known and there is only one additional fit parameter in the model
(Thickness #2), fitting the “Angle Offset” can slightly improve the SiO2 thickness
result by compensating for small sample alignment errors. The best sensitivity to
angle of incidence occurs for thin layers (<100nm), and we generally only
recommend fitting for this value when measuring the calibration wafer supplied
with the ellipsometer.

Figure 3-21. Angle Offset parameter.

Thickness Pre-Fit
One further concept will be demonstrated using the “Si with Thermal Oxide”
model. For this example, reload both the “Si with Thermal Oxide” model and the
Thickness Pre-Fit quickly
“1000nm SiO2 on Si” example data file. From the Model > +FIT Options section,
finds good starting
turn the “Perform Thickness Pre-Fit” option to “OFF” by clicking on the underlined
values for the thickest
blue value.
layer in the sample
stack. Click Fit > ‘Generate’. The screen should appear as shown in Figure 3-22. Note
that the data generated by the model (black dashed curves) using the default
thickness (100nm) has roughly one oscillation while the experimental data has
many oscillations.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-53

 Figure 3-22. Data generated from “Si with Thermal Oxide” model using default parameter
values. The “Perform Thickness Pre-Fit” option is turned “OFF”.

Now click Fit > ‘Fit’. The result is shown in Figure 3-23. CompleteEASE fails to find a
good match to the data. The MSE is very high and model curves do not match
experimental curves. The results are meaningless. This illustrates what happens if
you start the fit with parameter values that are not reasonably close to the true or
best fit values.

Figure 3-23. Fit result with starting thickness = 100nm and Thickness Pre-Fit OFF.

3-54 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Evaluation of fit result: Model curves do not match experimental curves. MSE high.
Parameter results not physically plausible.

Now, set “Thickness #2” to “1000” and “Angle Offset” to “0” (by clicking on the
respective parameter values). Click Fit > ‘Generate’. The screen should appear as in
Figure 3-24. While the model and experimental data curves do not line up exactly
(slight shift), at least the basic structure and features in the data sets appear similar.

Figure 3-24. “1000nm SiO2 on Si” example with model data generated using a closer
(1000nm) starting thickness value.

Now click Fit > ‘Fit’. The result is shown in Figure 3-25. Because the starting value
was sufficiently close to the real value, CompleteEASE found a good match to the
data. MSE is reasonable and model curves visually match experimental curves. The
results are now meaningful.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-55

 Figure 3-25. Fit result with starting thickness = 1000nm and Thickness Pre-Fit OFF.

Evaluation of fit result: Model curves match experimental curves. Reasonable MSE.
Thickness result is believable based on nominal value.

A good starting value for “Thickness #2” is critical to achieve the best data fit.
Without the Thickness Pre-Fit, CompleteEASE tries to adjust the fit parameter
(thickness) to lower the MSE. However, the MSE profile is not a single minimum at
the correct thickness. It also contains many other minima, which are called local
minima. These local minima can trap the normal fit process. Each of these local
minima are at a much larger MSE value, as shown in Figure 3-26, so it is easy to
distinguish the correct answer from a thickness result that does NOT match the
experimental curves. However, it is tedious to guess starting film thickness values
for all samples.

NOTE: Films thicker than 1-2 microns are more susceptible to the model getting
trapped in local minima.

3-56 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-26. MSE profile for “1000nm SiO2 on Si” example data using the “Si with Thermal
Oxide” model. The only believable answer occurs when the SiO2 thickness is near 1000nm.
However, a normal fit will only find this answer if the starting thickness value is near the
1000nm. To overcome this problem, our “Thickness Pre-Fit” automatically searches for a
good starting thickness near the final answer.

The “Thickness Pre-Fit” option automatically provides a good starting value for the
thickest layer in the model, using a proprietary algorithm. To demonstrate this
feature, turn the “Perform Thickness Pre-Fit” option back to “ON”. Input a “bad”
starting value for the “Thickness #2” parameter, for example “50”. Click ‘Generate’
to verify that the model starting parameters are not close to the best fit. Now
when you click the ‘Fit’ button, a good data fit appears on the screen, even though
the initial thickness value (50nm) was not close to the final best fit value (1029.65
nm). For most analysis, it is useful to have the “Perform Thickness Pre-Fit” option
turned “ON”, especially if accurate starting thickness values are not known.
To better understand the Thickness Pre-Fit, consider that the data will oscillate
versus wavelength as the film becomes thicker due to interference between
reflected light from the surface and bottom of a thin film. As thickness increases,
more oscillations will occur versus wavelength. This is not independent of the
refractive index, but the underlying principle is that a larger “optical thickness” will
have more oscillations. The Thickness Pre-Fit works by quickly interpreting the data
oscillations to “estimate” the actual thickness, as illustrated in Figure 3-27. Often,
this fast calculation is adequate to get a good starting point for the normal fit
procedures.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-57

 Figure 3-27. Thickness Pre-Fit uses a proprietary algorithm to very quickly estimate the
thickness of a film based on oscillations in the data.

Visualizing Parameter Changes

It is useful to understand how changes in the model affect the generated data.
For example, an increase in thickness will shift the interference oscillations toward
The value of a fit
longer wavelengths. CompleteEASE allows convenient simulation as a model
parameter can be
parameter changes.
changed by holding the
moouse pointer over the 1. Position the mouse over the model parameter you wish to adjust.
parameter value, holding
2. Hold down the “Shift” key as you scroll the mouse wheel up and down.
the shift key, and scrolling
The model parameter is increased or decreased by a set increment and
the mouse wheel.
data are regenerated with the new model values.
3. For smaller parameter increments, simultaneously hold “Ctrl+Shift” as
you scroll the mouse wheel.

3.3. Transparent Thin Films 1: PECVD SiO2 on Si

FEATURES IN THIS EXAMPLE
• Cauchy dispersion equation • Displaying optical constants
• Global fit procedures • Expanding layers
• Renaming layers

The next set of examples will demonstrate how to determine the film thickness and
index of refraction “n” for transparent films deposited on substrates with known
optical constants. While there are certainly cases in which it is not known in
advance if the film is transparent (absorbing films are discussed in the following
chapter), there are also many applications where the film can be assumed
transparent, at least in the visible spectral range: optical coating materials (SiO 2,
TiO2, Ta2O5, MgF2 etc.), some nitrides (Si3N4, AlN), many organic films (photoresists,
PMMA, spin-on polymers, etc.), and so on. In addition, the presence of periodic

3-58 • Data Analysis 1 – Basic CompleteEASE Software Manual

 data oscillations (as exemplified by Figure 3-28) suggests interference between light
reflected at the top surface and at the interface between film and substrate.

Figure 3-28. Periodic data oscillations suggestive of a transparent film.

Figure 3-29. Interference within a transparent thin film.

The index of refraction dispersion for transparent films in the visible spectral range
is often modeled by the Cauchy dispersion equation (Equation 3-2). The A
parameter relates to the approximate amplitude for the material index, while the B
and C parameters provide the magnitude of curvature of the index versus
wavelength. Figure 3-30 shows index dispersion curves for three common
transparent materials. In a transparent region, the Kramers-Kronig relations imply
that index must curve upward toward shorter wavelengths, a behavior which is well
described by the Cauchy dispersion equation (Figure 3-31).

Equation 3-2

B C
n ( ) = A + +
 2
4

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-59

 Figure 3-30. Index of refraction “n” for common transparent materials.

Figure 3-31. Simulated index curves for Cauchy parameters displayed in graph legend.

Si with Transparent Film.mod

The “Si with Transparent Open the “PECVD SiO2 on Si” example data file from:
Film” model can be used
Analysis > Model > ’Open’ > Library tab > Basic folder.
for transparent films on
other substrates. Just Try to fit this data using the “Si with Thermal Oxide” model as described in Section
replace the substrate 3.2. Results from this model are shown in Figure 3-32. Notice the discrepancies
layer with the appropriate between the model-generated and experimental curves as well as the relatively
material file for your high MSE (~79). Also notice the high Angle Offset (0.658° ̶ well-aligned
substrate. ellipsometers will have angle accuracy better than a few hundredths of a degree).
Considering these factors, we can conclude that the model is not acceptable for
this data set.

3-60 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-32. “PECVD SiO2 on Si” data fit using “Si with Thermal Oxide” model – unacceptable
fit quality.

Evaluation of fit result: Model curves do not match experimental curves as well as
we expect for a simple structure. MSE is high. Angle Offset is too high and unlikely
real.

PECVD SiO2 is optically like thermal SiO2 but with slightly different refractive index.
Thus, a different optical model with more flexibility regarding the film refractive
index is required. To determine the index and thickness of the PECVD SiO2, open
the “Si with Transparent Film” model from the same location (see Figure 3-33).
After opening, click ‘Fit’. The result is significantly improved with better visual
agreement between model and experimental curves and lower MSE (~22) (see
Figure 3-34).

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-61

 Figure 3-33. Si with Transparent Film model location.

Figure 3-34. “PECVD SiO2 on Si” data fit using “Si with Transparent Film” model – acceptable
fit quality.

Evaluation of fit result: Model curves match experimental curves. MSE is

significantly improved. Layer 1 optical properties should be evaluated.

3-62 • Data Analysis 1 – Basic CompleteEASE Software Manual

 The reported fit parameters using this model are the PECVD SiO2 film thickness
(Thickness #1) and Cauchy parameters A, B, and C (which calculate the PECVD SiO 2
film’s index of refraction). The film’s index of refraction at 632.8 nm is also
reported as a derived parameter.
Recall from Section 3.1 that the most important aspect of SE data analysis is
evaluating the result. After fitting the optical properties of a layer, you should view
them to determine if the results are physically plausible.
Right-click the mouse To display the film’s index of refraction versus wavelength, right-click on the
over a layer name or material name (in this case, “Cauchy Film”) and choose the option “Graph Layer
expand the +Model Optical Constants” (Figure 3-35). You can also expand the +Model menu above the
menu above graph to graph and select from various options (Figure 3-36). The optical constants will be
display optical constants displayed in the graph as shown in Figure 3-37.
(n, k).

Figure 3-35. To view a layer’s optical constants, right-click the layer name and choose to
“Graph Layer Optical Constants”.

Figure 3-36. You can also expand the +Model menu above graph to display optical constants.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-63

 Figure 3-37. Optical constants displayed versus wavelength. Also, the Cauchy layer is
expanded, displaying the fitted A, B, and C values in the model layer.

Evaluation of fit result (continued): Layer 1 optical properties are physically

plausible. Index increases into the UV and is near expected value for this material.

To display and edit the Cauchy parameters, click the red “+” symbol to the left of
“Layer #1” to expand layer details. With the layer expanded, the Cauchy fit
parameters are now accessible, as shown in Figure 3-37.

NOTE: The “+Urbach Absorption Parameters” subsection within the Cauchy layer
can be used to add an exponential “Urbach” absorption tail to this otherwise
transparent layer. This is not required for this example. The Cauchy layer is
described in further detail in Section 10.2.

Global Fit
Like the “Si with Thermal Oxide” model, the “Si with Transparent Film” model uses
the Thickness Pre-Fit option to provide a good starting value for the layer thickness.
To provide starting values for the film index, a “Global Fit” is also performed. In a
Global Fit, a user-defined range of values and number of guesses within this range
for a specified fit parameter are used as starting values for a normal fit. The value
which yields the lowest MSE becomes the starting value for the “final” fit.

3-64 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Expand +FIT Options to find “Use Global Fit”. With this option “ON”, it can also be
expanded to display the Global Fit settings (Figure 3-38).

Figure 3-38. Settings for the “Si with Transparent Film” model; note the “Use Global Fit” and
“Include Derived Parameters” sections.

In the “Si with Transparent Film” model, the Cauchy A parameter (which is the
approximate index at long wavelengths) is defined as a Global Fit parameter. Min.
and max. values are set to 1.2 and 4.5, respectively, and the # of Guesses is set to
45. This means that CompleteEASE performs 45 trial fits with different starting
values for A, distributed in equal increments between 1.2 and 4.5. At each trial fit,
the Thickness Pre-Fit is also performed. The lowest MSE from these trial fits
determines the starting values for thickness and the Cauchy A parameter. A final fit
is then performed with these starting values to arrive at a “final” result (subject to
evaluation as described in Section 3.1). Min., Max., and # Guesses can be adjusted
if needed.
If more than one global fit parameter is defined, a grid of test values is created with
the total number of guesses being the product of the individual guesses for each fit
parameter. Depending on the total number of guesses, the Global Fit procedure
can be computation intensive. If the total number of guesses is small, you may not
notice a significant difference in the time it takes to arrive at a “final” result. If the
total number of guesses is large, the process may take several seconds or longer,
and a status box with a ‘Stop’ button will be displayed as shown in Figure 3-39.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-65

 Figure 3-39. Status box and ‘Stop’ button displayed during a Global Fit.

The Global Fit can be further customized (expand “+Customize Global Fit”
subsection). To minimize computational time, only a subset of the experimental
data points is used during trial fits (“# of Data Points”). The # of Iterations used in
each trial fit can also be limited. Default values for these options are 20 and 5,
respectively, which are adequate for most samples but can be increased if
necessary. Larger values may improve the robustness of the Global Fit but will
increase analysis time. More details regarding Global Fit customization can be
found in Section 9.7.

Figure 3-40. Customize Global Fit subsection.

3-66 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Rename Layer & Fit Parameters
The Fit panel with results after fitting the “PECVD SiO2 on Si” example data with the
“Si with Transparent Film” model is shown in Figure 3-41. Notice the parameter
names are not specific. This may be fine for single-layer structures or if the results
will not be utilized further, but it is often helpful to rename layers and fit
parameters according to the material or sample details.

Figure 3-41. Fit results for “PECVD SiO2 on Si” example data with “Si with Transparent Film”
model.

You can right-click on any layer and choose “Rename Layer & Fit Parameters”, as
shown in Figure 3-42.

Figure 3-42. Right-click on a layer and choose “Rename Layer and Fit Parameters”.

The “Rename Layer Options” dialog box will appear, as shown in Figure 3-43. You
can enter a new name and use the checkboxes to determine which elements of the
layer will be renamed:
• “Rename Layer” will change the displayed layer name in the model and
other references to the layer name.
• “Rename Layer Parameters” will append the layer name to the layer
parameters.
• “Rename Thickness” will specify thickness according to the layer name
rather than the layer number.
Click ‘Ok’ to apply changes and refit the data to update results displayed in the Fit
panel (Figure 3-44).

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-67

 Figure 3-43. “Rename Layer Options” dialog box.

Figure 3-44. Renamed layer, layer parameters, and thickness.

3-68 • Data Analysis 1 – Basic CompleteEASE Software Manual

 3.4. Transparent Substrates: Bare Glass
FEATURES IN THIS EXAMPLE
• Graphing multiple data sets • Using the graph scratchpad
• Including backside reflections • Depolarization data

In the previous section, we introduced the Cauchy equation to describe the index of
refraction for transparent films. The Cauchy equation can also be applied to
transparent substrates. This is the most common approach for glass or plastic
substrates. In this section, we will use the prebuilt “Glass Substrate” model which
utilizes the Cauchy layer to determine the index for glass substrates.

Substrate Backside Reflections

With transparent substrates, it is possible for light to reflect from the back surface
of the substrate and reach the detector. Detection of backside reflections is
generally undesirable, as they can combine incoherently with frontside reflections.
Backside reflections can shift the ellipsometric data and produce errors in the
analysis if not accounted for properly.
Fortunately, backside reflections can often be easily avoided. There are multiple
ways to avoid backside reflections from transparent substrates and this topic is
reviewed by Synowicki2. Methods to avoid collecting backside reflections include:
• Spatial separation of the backside beam via 1) thick substrate, 2) small
measurement beam size, or 3) wedged substrate.
• Scattering and/or absorption of the backside reflection via 1) roughening
the backside to scatter light or 2) applying index-matching material to the
backside (matte-finish/cloudy office tape typically works well for glass).

Figure 3-45. Backside reflections of transparent substrates can be avoided by roughening the
back surface or using index-matching tape to scatter the reflected light.

If the backside reflections cannot be avoided, then they must be included in the
software calculation. This will be demonstrated with the following example.
Depolarization data can be used to check for evidence of backside reflections in the
measured data. If the depolarization is essentially zero, then no backside
reflections reached the detector. This is the preferred condition. If backside

2 R.A. Synowicki, “Suppression of Backside Reflections from Transparent Substrates”, Phys. Stat. Sol. (c) 5, No. 5 (2008) 1085-1088.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-69

 reflections reach the detector, CompleteEASE can incorporate incoherent
contributions in the model to account for them.
Let’s consider data from a glass substrate, with and without backside reflections
detected, to better understand this situation.

Graphing Multiple Data Sets

There are three files for glass substrates located in the Library tab, Examples folder:
1. glass substrate_rough.SE – glass slide with backside mechanically
roughened
2. glass substrate_smooth.SE – glass slide with backside untreated
3. glass substrate_taped.SE – glass slide with matte-finish office tape applied
to backside
In this example, we will open the first data set and append the remaining two data
sets to graph all three simultaneously.
First, open “glass substrate_rough” by clicking Analysis > Data > ‘Open’.

Right-click on Next, append “glass substrate_smooth” and “glass substrate_taped” by right-

Data>‘Open’ to append clicking the same button to access the append menu (Analysis > Data > (R)
additional data sets. ‘Open’). See Figure 3-46.

Figure 3-46. Right-click on ‘Open’ button to access the “Append Data” option.

NOTE: While you can only OPEN one file at a time, you can APPEND multiple files
simultaneously.

Figure 3-47. Multiple data set graph menus.

Shortcut: When multiple data sets are open, new menus appear above the graph, as shown
in Figure 3-47. Check the “Graph All Data Sets” box (and turn OFF double Y axis, if
CTRL+Y to turn double Y
applicable) to compare Psi for each data set. In Figure 3-48, “glass
axis ON or OFF
substrate_smooth” is data set #2 with dark blue curves. This data set is from a
glass slide with normal, polished backside, and backside reflections are expected
and present in the data. Notice the shift around 300 nm compared to data sets #1
and #3, where backside reflections have been eliminated or suppressed via
roughening and tape, respectively. Below 300 nm, the glass is absorbing and light
doesn’t reach the backside or reflections from the back surface are absorbed

3-70 • Data Analysis 1 – Basic CompleteEASE Software Manual

 before reaching the top surface and traveling to the detector. This shift in the data
will lead to errors in results if backside reflections are not considered in the model.

Figure 3-48. Check “Graph All Data Sets” to view all data sets simultaneously.

Using the Graph Scratchpad

While the “Graph All Data Sets” option provides a quick comparison of the three
data sets, it is not customizable. For example, we can’t change the colors or select
a subset of the total curves. For this purpose, we turn to the Graph Scratchpad.
Right-click within the graph area and choose to “Copy Data to Graph Scratchpad”,
as shown in Figure 3-49. This will copy all curves currently shown on the graph and
place them in an internal “pad” for further manipulation. Next, choose to “View
Graph Scratchpad” from the same right-click menu. This opens the CompleteEASE
Graph Scratchpad, as shown in Figure 3-50. Play with the Graph Scratchpad by
selecting which curves to compare and by changing their colors and styles. Finally,
it is beneficial to change their curve names to correspond to the data set that they
came from. Figure 3-51 shows a comparison of the taped and rough glass data –
which are nearly identical. This shows that taping the backside can be as effective
as roughening the backside of a transparent substrate for removing backside
reflections.
Next, we will model each data set to show how this backside reflection can produce
error in refractive index results and how to correct for this backside reflection.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-71

 Figure 3-49. Right-click within the graph area to access the Graph Scratchpad.

Figure 3-50. The Graph Scratchpad allows customization of graphs.

3-72 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-51. The Graph Scratchpad is used to compare data from glass substrates where the
backside is roughened or taped to suppress backside reflections.

NOTE: Before proceeding, clear the contents in the Graph Scratchpad by viewing
the Scratchpad and clicking ‘Delete All’.

Glass Substrate.mod
In a later example, we will learn how to fit multiple data sets simultaneously.
However, for this example, we will focus on fitting each individual data set. Start by
opening the “glass substrate_rough” data file. Next, open the “Glass Substrate”
model from the Basic folder (Figure 3-52). Click ‘Fit’ and you should get the result
as shown in Figure 3-53. The data appear noisy, but this is primarily due to the
graph scale.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-73

 Figure 3-52. Glass Substrate.mod location.

Figure 3-53. Result for “glass substrate_rough” data with “Glass Substrate” model.

Display the optical constants of Cauchy Substrate by right-clicking this name in the
model or by selecting from the +Model menu above the graph.

3-74 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-54. Index of refraction from Cauchy model fit to bare glass substrate.

Evaluation of fit result: Model curves match experimental curves. MSE is

reasonable. Cauchy Substrate optical constants are physically plausible. Index
increases into the UV and is near expected value for this material.

Graphing Depolarization
Ellipsometry measurements use polarized light. If the sample is isotropic and ideal,
this measurement can be described by two values: Psi and Delta. However, if a
sample is partially depolarizing, two parameters are insufficient to fully describe the
change in polarization. The patented Rotating Compensator technology used in the
alpha-SE, M-2000, and RC2 systems permits measurements of three reported
quantities: , , and %depolarization. The “%depolarization” should remain zero
for ideal sample conditions.

NOTE: Some ellipsometers require DC measurement to acquire accurate

depolarization data. Please consult your hardware manual (or contact your J.A.
Woollam representative) to best measure the depolarization.

Shortcut: To view depolarization data, expand the +Data menu above the graph and select
“Depolarization”. See Figure 3-55.
CTRL+Z to graph
Depolarization

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-75

 Figure 3-55. Expand +Data menu to graph depolarization.

Notice the depolarization in Figure 3-56 is near zero. (Some noise is expected and
acceptable.) This indicates that backside reflections are not detected.

Figure 3-56. Depolarization data from glass slide with backside roughened.

Shortcut: Next, open the “glass substrate_smooth” data file. Graph the depolarization
data. %Depolarization now deviates from zero increasingly with angle. This
CTRL+P to graph Psi
indicates that backside reflections are detected.

3-76 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-57. Depolarization data from glass slide with backside smooth.

Now, graph Psi again and click ‘Generate’. See Figure 3-58. The model no longer
matches the experimental measurement even though this data is from the same
glass substrate (only now with backside reflections present in the data). The
backside reflections do more than introduce depolarization – they also shift the
ellipsometry values.

Figure 3-58. Data generated with “Glass Substrate.mod” do not match data from glass slide
with backside reflections.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-77

 Fit the data with the same model; you should get results as shown in Figure 3-59.

Figure 3-59. Fit results for “glass substrate_smooth” data without backside reflections
included in the model.

Evaluation of fit result: Model curves do not match experimental curves. MSE is
high. “Glass Substrate.mod” does not work when backside reflections are present
in the data.

Including Backside Reflections

If the backside reflections can’t be avoided, the “incoherent” summation of light
beams can be handled mathematically. To include backside effects in the model,
expand the +MODEL Options section of the Model. Turn “Include Substrate
Backside Corrections” to ON, as shown in Figure 3-60. The correction depends on
the amount of light collected from the backside, which can differ for different
substrates. Make “# Back Reflections” a fit parameter by right-clicking on the value
and set the starting value to a low number, such as 0 or 1.

3-78 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-60. “Include Substrate Backside Correction” = ON with “# Back Reflections” activated
as a fit parameter to match the value to the experimental data.

Fitting Depolarization
There can be correlation between the substrate index and the amount of backside
light collected when modeling ellipsometry data. However, depolarization data is
only influenced by the amount of backside light collected. The default setting is to
ignore the depolarization data and only fit Psi and Delta. In fact, the depolarization
fitting selection is a hidden feature of the model. To show this feature, scroll to the
bottom of the Model panel and click Configure Options. This will open the dialog
box shown in Figure 3-61. Check the box next to “Include Depolarization Data” in
the Fit Options section and then click ‘Ok’.

Figure 3-61. Click Configure Options to open the Show Model Options dialog box. Under Fit
Options section, check “Include Depolarization Data”.

Now that this option has been added to the model, it can be turned on. Expand the
+FIT Options section in the model and turn the “Include Depolarization Data” to
ON, as shown in Figure 3-62.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-79

 Figure 3-62. “Include Depolarization Data” is included in the +FIT Options section and is
turned ON.

Glass Substrate (with backside reflection).mod

CompleteEASE has a prebuilt model with the previous two features (include
backside reflections and fit depolarization data) already activated. Like the other
models we have introduced, it is in the Library tab, Basic folder.

Figure 3-63. Location of “Glass Substrate (with backside reflection)” model.

3-80 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-64. “Glass Substrate (with backside reflection)” model.

NOTE: For this example, we will limit the wavelength range to 375 nm and longer
to avoid absorption in the UV.

Open the “glass substrate_smooth” example data set if needed. Click the ‘Set
Ranges’ button in the Data panel to open the Select Data Range dialog box. Set the
min. wavelength to 375.

Figure 3-65. Select Data Range dialog box.

Using the “Glass Substrate (with backside reflection)” model, click Fit > ‘Fit’. Results
are shown in Figure 3-66.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-81

 Figure 3-66. Results from “glass substrate_smooth” data set with “Glass Substrate (with
backside reflection)” model.

Evaluation of fit result: Model curves match experimental curves, indicating that
the backside correction is effective. MSE is reasonable. However, roughness is
negative, and index has the beginnings of a downturn at the shortest wavelengths.
The model could use a few modifications.

Before we finalize the result, there are a couple of modifications we can make to
this model. It is prudent to reset the fit to restore the model to original starting
values. To do so, click Fit > ‘Reset’.
• First, as the previous result yielded a negative roughness value, let’s turn
this OFF and see if a similar fit quality can be obtained.

3-82 • Data Analysis 1 – Basic CompleteEASE Software Manual

 • Second, the slight index downturn at the lowest wavelengths of the
previous result is due to a negative C parameter in the Cauchy equation.
Often in the VIS-NIR, a material’s dispersion can be adequately described
using only the A and B parameters. Let’s set the C value to 0, turn this
parameter OFF, and see if a similar fit quality can be obtained.
With these changes, the model should now look like Figure 3-67.

Figure 3-67. “Glass Substrate (with backside reflection)” model with modifications – surface
roughness OFF and C parameter set to 0 and deactivated as a fit parameter.

Now fit the data again and re-evaluate the result (Figure 3-68).

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-83

 Figure 3-68. Results from “glass substrate_smooth” data set with modified “Glass Substrate
(with backside reflection)” model.

Evaluation of fit result: Model curves still match experimental curves. MSE is only
slightly higher. Index only increases with decreasing wavelength. Higher MSE is
acceptable with improved physicality of model.

NOTE: Prebuilt models are included with CompleteEASE for convenience but may
need to be modified for specific sample types. Always evaluate your results!

3-84 • Data Analysis 1 – Basic CompleteEASE Software Manual

 3.5. Transparent Films 2: SiC on Si
FEATURES IN THIS EXAMPLE
• Surface roughness • Simple grading
• Simple anisotropy • Try alternate models

For this example, load the “SiC on Si” example data and the “Si with Transparent
Film” model. Click the ‘Fit’ button. Results are shown in Figure 3-69 and Figure
3-70.

Figure 3-69. Results from “SiC on Si” data set with “Si with Transparent Film” model.

Figure 3-70. Resulting SiC optical constants using “Si with Transparent Film” model.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-85

 Evaluation of fit result: Model curves mostly match experimental curves. Unsure if
MSE is okay. Index result for SiC layer looks reasonable. The question is: can MSE
be reduced?

Let’s consider ways to possibly improve the model and lower the MSE. The
following three topics (Surface Roughness, Simple Grading, and Simple Anisotropy)
were introduced in Section 3.1. The sections below contain more details on how
these complexities are included in a model and how they may apply to this specific
example.

Surface Roughness
Surface roughness lowers the optical density and behaves optically as a lower-index
surface film. CompleteEASE utilizes an Effective Medium Approximation with 50%
void (n=1, k=0) and 50% base material to describe this condition. The option to
include Surface Roughness in the model is included at the top of the Model panel
(below Layer Commands) and is easily toggled ON/OFF.

Figure 3-71. Model with Surface Roughness OFF.

Figure 3-72. Model with Surface Roughness ON.

Ellipsometry is increasingly sensitive to surface roughness at low wavelengths (due

to scale of wavelength vs roughness thickness) and on absorbing films and high-
index films (due to increased optical contrast between film and air). Most real-
world surfaces are not perfectly smooth, so whenever you suspect a fit can be
improved, surface roughness is an easy, common addition to try in your model,
especially if measuring low wavelengths or absorbing or high-index films.
Try including surface roughness into the current example and refit the data. The
results should appear as shown in Figure 3-73.

3-86 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-73. Results from “SiC on Si” data set with “Si with Transparent Film” model with
Surface Roughness added.

Evaluation of fit result: MSE is reduced from ~22 to ~20. Is this enough? Can MSE
be reduced further? There is still some visual misfit between model and
experimental curves.

Simple Grading
If the index varies with depth through the film, the data may not be well-modeled
by a standard single layer. CompleteEASE can approximate the index variation by
dividing the layer into smaller sublayers, as shown in Figure 3-5.
Grading often appears in the data as compressed or expanded interference
oscillations, as we see for this example. Consider the 60° data for Psi only (Figure
3-74). Model curves are compressed compared to experimental curves.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-87

 Figure 3-74. “SiC on Si” example data – Psi only at 60̈°. Model curves are compressed
compared to experimental curves, which can be indicative of index grading.

Reload the “Si with Transparent Film” model and click ‘Fit’. Now, try adding a
simple index gradient by right-clicking on the name of Layer #1 (“Cauchy Film”) and
choosing “Grade Layer” (see Figure 3-75). This will add a simple gradient where
index is varied with depth of the film and % Inhomogeneity is defined as a fit
parameter. Click ‘Fit’. See Figure 3-76 for results.

Figure 3-75. Right-click on the layer name and choose “Grade Layer”.

3-88 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-76. “SiC on Si” example data with “Si with Transparent Film” model plus simple index
grading.

Right-click on “Graded Layer” and select “Graph Sub-Layer Optical Constants (n)”, as
shown in Figure 3-77, to display the index at each sublayer (“slice”) of the model.

Figure 3-77. Right-click on "Graded Layer" to show graphing options.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-89

 Evaluation of fit result: MSE is reduced from ~22 to ~5. Thickness and index are still
physically plausible. Adding grading to the model provides significant improvement
while still maintaining physical meaning.

NOTE: The optical effect of surface roughness is like a film with lower index at the
top, as both have the effect of lowering the optical density toward the surface. In
this case, the film may be modeled equally well by surface roughness or by a
negative index gradient. This is more likely to occur for thinner films where the size
of roughness is comparable to the film thickness or if the gradient is nonlinear. If
both roughness and grading yield similar MSE values, you can decide which is best
based on your understanding of the material or information you have from other
metrology.

Simple Anisotropy
Anisotropy is usually observed in the experimental data as a tilting of the curves, as
demonstrated by Figure 3-78. Note that while the maxima of the interference
oscillations are lined up at the same wavelengths, the minima position is shifted
relative to the isotropic model.

Figure 3-78. Anisotropy can tilt the experimental data.

The simplest form of anisotropy is uniaxial where nx = ny ≠ nz and nx, ny are in the
sample plane. To try this simple case in your model, right-click the layer name and
select “Convert to Anisotropic” (Figure 3-79). This will replace your layer with an
anisotropic layer where the previous optical constants are preloaded into the X and
Z orientations of a uniaxial model (Figure 3-80).

3-90 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-79. Right-click a layer and select "Convert to Anisotropic" to add anisotropy to your
model.

Figure 3-80. "Convert to Anisotropic" replaces the previous layer with an anisotropic layer
where the previous optical constants are preloaded into the X and Z orientations of a uniaxial
model.

As shown in Figure 3-80, two sets of optical constants are now active in the fit: Ex
and Ez. We can further simplify this model by using “Difference Mode”. Turn
“Difference Mode” ON to fit Ex only; Ez is simply shifted relative to Ex using the
“Index Differences” parameters.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-91

 Figure 3-81. Anisotropic model with Difference Mode = ON.

The index difference is described by Equation 3-3:

Equation 3-3
∆𝐼𝑅
∆𝑛(𝑒𝑉) = ∆𝐴(1 + ∆𝐵 ∙ 𝐸 2 + ∆𝐶 ∙ 𝐸 4 + ∆𝐷 ∙ 𝐸 6 + ) (𝐸 in eV)
𝐸2

The “A” parameter provides the base index difference and may be the only
parameter needed to adequately describe simple cases.

NOTE: “Difference Mode” often works well for uniaxial transparent films with the
differing axis oriented in the Z direction. More complex cases will require more
complex model and nonstandard experimental data.

Try simple anisotropy with the current “SiC on Si” example. Results are shown in
Figure 3-82.

3-92 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-82. "SiC on Si" example data with simple anisotropy added to "Si with Transparent
Film" model.

Evaluation of fit result: MSE is unchanged. Therefore, there is no justification for

including anisotropy in the model.

Try Alternate Models

The process of trying optical models with and without surface roughness, simple
grading, and simple anisotropy is so common in SE data analysis that CompleteEASE
has a command to automate the procedure: Try Alternate Models. The command
can be found by expanding +OTHER Options, as shown in Figure 3-83.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-93

 Figure 3-83. Expand +OTHER Options to find the Try Alternate Models command.

A good starting model should be first established, which approximates the film as
ideal. When Try Alternate Models is executed, CompleteEASE will automatically try
fitting the data using the ideal optical model, the same model with surface
roughness, then with simple grading, with both roughness and grading, and finally,
with simple anisotropy. When finished, a summary of the results will appear in a
new window, as shown in Figure 3-84. The recommended model will be highlighted
in green. The recommended model is determined solely by the MSE Improvement
Threshold, which is located next to “Try Alternate Models” and must be set prior to
executing. The default value for this threshold is 25%, which means that the added
complexity in the model must provide at least 25% improvement in MSE to be
recommended over a simpler model.

NOTE: Roughness often improves the fit for higher-index or absorbing films.
Grading is common with many metal-dielectrics. Anisotropy is common with
polymers.

With “SiC on Si” example data and “Si with Transparent Film” model open, use the
Try Alternate Models command. The recommended model (see Figure 3-84)
includes both surface roughness and index grading. The MSE is improved at least
25% over the ideal model, the model with surface roughness only, and the model
with index grading only, which suggests that both roughness and grading are
present in the sample.

3-94 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-84. Results of the Try Alternate Models command for “SiC on Si” example data and
“Si with Transparent Film” model.

NOTE: The Try Alternate Models command is a tool to automate the process of
testing each model but does not evaluate the results of each model beyond MSE.
This command does not eliminate the need to evaluate the results.

Before accepting the recommended model, evaluate the parameter results

displayed in the summary table. In this case, the “Roughness & Grading” results for
thickness, roughness, and % Inhomogeneity are physically reasonable.
Click on the ‘Show Graphs’ button in this window to scroll through the Psi/Delta fits
and difference plots for each model. These can be helpful in evaluating the results
from each model. An example graph is shown in Figure 3-85.

Figure 3-85. Example difference plot from summary of Try Alternate Models results.

Lastly, you can click the ‘<<’ and ‘>>’ buttons to change the selected model, then
‘Apply Chosen Model’ to update the Model panel with your selection. With the
“Roughness & Grading” column selected, click ‘Apply Chosen Model’ to update your
model for this example. CompleteEASE will then look like Figure 3-86.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-95

 Figure 3-86. “SiC on Si” example data with surface roughness and index grading.

Finally, upon updating the model, you can further evaluate the result by viewing
optical constants. Right-click on “Graded Layer” and select “Graph Sub-Layer
Optical Constants (n)”, as shown in Figure 3-77. The final optical constants for this
example are shown in Figure 3-87.

Figure 3-87. Optical constants for “SiC on Si” example with surface roughness and index
grading included in the model.

Evaluation of fit result: MSE is significantly improved by including both surface

roughness and index grading in the model. Thickness, roughness, % Inhomogeneity,
and index shape are physically plausible. Both surface roughness and index grading
are needed to find the best physically meaningful fit to the data.

3-96 • Data Analysis 1 – Basic CompleteEASE Software Manual

 3.6. Log and Report a Series of Films
FEATURES IN THIS EXAMPLE
• Fit log • Copy analysis report to clipboard
• Open and save snapshots • Re-analyzing data from log
• Copying to clipboard • Comparing optical constants in fit
(formatted or table) log

Ellipsometry is often used to measure a series of thin films to compare and report.
CompleteEASE enables simple logging, comparing, and reporting of results. This
example will step through a series of SiNx thin films on Si, using the analysis steps
taught in Section 3.3. However, the emphasis will be on using CompleteEASE to log
all fit results, compare optical constants for the various films, and finally create a
report of the results.

To begin this exercise, open the “sinx on si-1” data file and the “Si with Transparent
Film” model in the Analysis tab of CompleteEASE. Click ‘Fit’ and the results shown
in Figure 3-88 appear.

Figure 3-88. Fit to “sinx on si-1” example data with “Si with Transparent Film” model.

Copy to Clipboard
The Fit and Model panels in the Analysis tab of CompleteEASE offer right-click (R)
menu choices. Examples of these right-click menus are shown in Figure 3-89. The

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-97

 graph also features a right-click menu, as shown in Figure 3-90. From these menus,
models, fit results, and graphs can be copied to the clipboard and then pasted in
various programs such as Word, Excel, or PowerPoint. Fit results can be copied to
the clipboard using Formatted or Table format, where the Formatted is best for
pasting into a word processor and the Table is preferred for pasting into a
spreadsheet. Models can be copied in their entirety (includes all visible commands
and options) or with layers only.

Figure 3-89. Right-click menu choices from the Fit and Model panels.

Figure 3-90. Right-click menu choices from the graph.

Fit Log
Shortcut: CompleteEASE includes a “Fit Log” for easy comparison of multiple results. To
record a fit result to the Fit Log, right-click (R) on the Fit panel and select “Add
CTRL+L to Add to Fit Log
to Fit Log” after the fit is finished. To view the Fit Log, use the same menu and
CTRL+Alt+L to View Fit select “View Fit Log”.
Log
Use the Fit Log to compare results for all five “sinx on si” example data sets. Open
each file and fit using the “Si with Transparent Film” model. After each is finished,
add the result to the Fit Log. You will be prompted to enter a descriptive name for
each entry, or you can use the timestamp generated by CompleteEASE (Figure
3-91).

3-98 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-91. Use the timestamp or enter a descriptive name for Fit Log entry.

After all five data files have been fit and added to the Fit Log, view the Fit Log. It
will appear as a new window, as shown in Figure 3-92. Click on any of the individual
log entries to see the results from that entry. You can use ‘Load Log Entry’ to
restore the selected entry into the Analysis tab of CompleteEASE.

Figure 3-92. Fit Log showing results from all five “sinx on si” data files.

Comparing Results
The Fit Log allows easy comparison of multiple results. Use SHIFT+click or
SHIFT+up/down keys to select multiple entries from the Fit Log. With multiple
entries selected, click ‘Compare’. This will create a table of results for each sample,
as in Figure 3-93. Use the checkboxes to reverse columns and rows or to add
statistics. This table can also be easily copied to the clipboard and pasted into
another program.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-99

 Figure 3-93. Comparison Table showing results from all five SiNx films.

Compare Optical Constants

From the Comparison window, click ‘Compare Optical Constants’ and again a new
window appears (Figure 3-94). Use the Display Options and Layer Selection Options
to configure the comparison graph. You can also use the checkboxes in the white
panel to make individual selections. Graphs can be copied to the clipboard by right-
clicking in the graph area.

Figure 3-94. Compare Optical Constants Dialog Box.

Reanalyzing Multiple Data Sets

CompleteEASE can automate the process of analyzing multiple data sets using the
same or multiple models. Before proceeding, open the Fit Log and click ‘Clear Log’
to remove all previous entries.

3-100 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Next, from the Fit Log, click ‘Reanalyze’. This will open a new window for you to set
up batch analysis of multiple data sets and/or models (Figure 3-95).

Figure 3-95. Reanalyze Data dialog for batch analysis setup.

NOTE: The “Re-save Data” checkbox allows new fit results to be saved with the
data file, which can then be reviewed later.

In batch analysis, “rows” are used to separate groups of data or analysis models.
Each row will contain a “file group”, model, and log name (optional). To start, click
‘Add Row’. This will open a dialog for you to first select data files for the file group
(Figure 3-96).

Figure 3-96. Choose Files window.

Click ‘Add’ to browse to the desired file location. For this example, select the five
“sinx on si” example data files from the Library tab (Figure 3-97). When finished,
click ‘Ok’.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-101

 Figure 3-97. Select Files dialog.

CompleteEASE will ask for a name for the group (Figure 3-98). When finished, click
‘Ok’.

Figure 3-98. Enter name for file group.

The new file group will now appear in the “Choose Files” window with individual
files displayed on the right (Figure 3-99).

Figure 3-99. Choose Files dialog with new file group listed.

This is the only file group needed for this example, so with the new file group
selected, click ‘Ok’ to add it into a row of the batch analysis setup. The Reanalyze
Data dialog will now appear like Figure 3-100.

3-102 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-100. Batch analysis setup with one row added.

To add a model for this batch analysis, double-click where indicated and select a
prebuilt or previously saved model. For this example, select “Si with Transparent
Film” from the Library tab and click ‘Open’ (Figure 3-101). The model will be added
to the row in the batch analysis setup.

Figure 3-101. Open Model dialog.

Lastly, you can double-click under “Log Name” and enter a name for the batch
analysis (Figure 3-102). The log name will be appended to the file name for each
entry and will be displayed within the Fit Log.

Figure 3-102. Batch analysis setup.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-103

 When the File Group, Model, and Log Name (optional) are finalized, click
‘Reanalyze’. CompleteEASE will perform a fit for each row in the batch analysis,
applying the designated model to each data file within the file group for that row. If
multiple rows are set up, CompleteEASE will work through the rows in sequence.
When finished, the Fit Log will be open with all entries listed.

Figure 3-103. Fit Log after batch analysis. The log name is appended to the file name for
each Fit Log entry.

After the batch analysis, entries can be compared as described in the preceding
sections.

Generate Reports
To generate reports from the Fit Log, select one or more entries and click ‘Generate
Reports’ (Figure 3-104). CompleteEASE will ask if it’s okay to clear the current
model and data set. Click ‘Ok’ to continue (Figure 3-105). The report will be saved
as an RTF file; navigate to the desired save location and enter a name for the report
(Figure 3-106).

3-104 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Figure 3-104. Select one or more Fit Log entries and click ‘Generate Reports’ to create an RTF
report for each entry.

Figure 3-105. CompleteEASE will ask if it is okay to clear the current model and data set
before generating reports.

Figure 3-106. Browse to desired save location and enter name for report.

When finished, navigate to the designated save location and open the RTF file to
view. The report will look like Figure 3-107.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-105

 Figure 3-107. Example report generated by CompleteEASE.

Add Optical Constants to Report

Optical constants are not automatically added to reports. If you would like to
include optical constants in CompleteEASE-generated reports, you must configure
the model to include optical constants for a specified layer. This option can be
found under Analysis > Model > +OTHER Options. Turn “Add Opt. Const. to
Report” ON and specify the layer of the optical constants to be included. The
substrate is layer #0.

NOTE: The model must be SAVED with “Add Opt. Const. to Report” = ON if it is to
be used in batch analysis or generating reports from the Fit Log.

Figure 3-108. “Add Opt. Const. to Report”.

3-106 • Data Analysis 1 – Basic CompleteEASE Software Manual

 Copy Analysis Report to Clipboard
Reports can also be created for a single result. After data analysis, right-click in
the Fit panel and choose “Copy Analysis Report to Clipboard”, as shown in Figure
Shortcut:
3-109. This will create a report in the clipboard which can be pasted into any
CTRL+R to Copy Analysis program (Figure 3-110). For this example, the optical constants were included in
Report to Clipboard the report by configuring the model as described in the previous section.

Figure 3-109. Right-click in the Fit panel to “Copy Analysis Report to Clipboard”.

Figure 3-110. Example of Analysis Report, including Optical Constants graph.

CompleteEASE Software Manual Data Analysis 1 – Basic • 3-107

 Save Log File
Fit Logs can be saved using the ‘Save Log’ button contained within the Fit Log. This
is convenient if results will be compared later. However, the Fit Log does not save
data, it only points to the file locations. If data files are moved, or become
inaccessible due to permissions, the Fit Log will no longer be able to open the
corresponding log entries.

Snapshot
A more robust way to save individual results for sharing or long-term storage is by
saving a “Snapshot”. A snapshot will compress your data file, model, and fit results
into a single file with the “.SEsnap” extension. To save a snapshot, simply click the
‘Save Snapshot’ button located in the upper right of the Model panel. Similarly, to
open a snapshot, simply click the ‘Open Snapshot’ button.

Figure 3-111. Open or save snapshots using the buttons in the upper-right corner of the
Model panel.

NOTE: Snapshots are preferred when contacting J.A. Woollam or your local
representative for analysis help.

3-108 • Data Analysis 1 – Basic CompleteEASE Software Manual

4. Data Analysis 2 – Intermediate

This chapter continues the series of examples with intermediate concepts

concentrating on absorbing materials. Sections from this chapter are listed below,
along with the primary CompleteEASE features that are discussed.

Section 4.1 Metal Substrates

FEATURES IN THIS EXAMPLE

• Saving optical constants • Saving model
• Building a model • B-Spline layer (basics)
• WvlbyWvl layer • Graphing pseudo transforms

Section 4.2 Self-Assembled Monolayer on Au

FEATURES IN THIS EXAMPLE

• Parameter uniqueness • Fixing B-Spline optical constants

Section 4.3 Semi-Absorbing Thin Films (B-Spline)

FEATURES IN THIS EXAMPLE

• B-Spline layer • Wvl. range expansion fit
• Starting material • Selecting a graph/fit range
• Saving a model • Zoom all

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-109

 Section 4.4 Semi-Absorbing Thin Films (Gen-Osc)

FEATURES IN THIS EXAMPLE

• Gen-Osc layer • Parameterize layer
• Tauc-Lorentz oscillator • Cody-Lorentz oscillator

Section 4.5 Multiple Layer Film Stacks

FEATURES IN THIS EXAMPLE

• Superlattice • Coupling

Section 4.6 Thin Absorbing Films

FEATURES IN THIS EXAMPLE

• Appending data • Fitting SE and transmission data
simultaneously
• Parameter uniqueness • Graph scratchpad
• Interference enhancement

4-110 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 4.1. Metal Substrates
FEATURES IN THIS EXAMPLE
• Saving optical constants • Saving model
• Building a model • B-Spline layer (basics)
• WvlbyWvl layer • Graphing pseudo transforms

Substrate optical constants are an important consideration when modeling thin

films. Some materials, such as crystalline silicon, are well understood and remain
constant. Other materials, such as glass and metals, are more variable. In these
cases, it is best to measure a bare substrate to determine the substrate optical
constants. Glass substrates were discussed in Section 3.4. In this section, we will
discuss metal substrates. First, let’s consider pseudo optical constants.

NOTE: Any material coated with an opaque metal film will be optically equivalent
to a bulk metal substrate. If light is absorbed before it can reach the backside of the
metal, reflect, and exit out of the material into the detector, the film is opaque. For
most metals, this will occur around 100nm or less. The example in this section is a
glass or silicon substrate that is coated with an opaque metal film.

Pseudo Optical Constants

If only a single reflection is detected by the ellipsometer (for example,
measurement of a bulk sample without any oxide, surface roughness, or any other
films or overlayers), then it is possible to directly determine the optical constants of
the sample from the SE data.
Equation 4-1 can be used to transform the ellipsometric parameters Psi and Delta
into the material optical constants “n” and “k” (or equivalently, the complex
dielectric function values “1” and “2”).

Equation 4-1

 1−   
2

 =  1 + i  2 = n~ = ( n +i k )
2 2
= sin(  ) 2  1 + tan(  ) 2    
  1 +   

In Equation 4-1,  is the angle of incidence and  is defined in Equation 1-1.

However, the direct transformation from SE data to material optical properties
given by Equation 4-1 is quantitatively valid only if all the assumptions are met; that
is, the sample does not have any oxide, surface roughness, or other films or
overlayers. Since all these assumptions are rarely met (or at least it is very difficult
to independently verify that they are met), the term “pseudo optical constants” is
used to describe optical constants which are derived from this analytic
transformation. The “<>” brackets are used to denote pseudo optical constants.
The transformation in Equation 4-1 can be applied to any experimental  data. To
avoid confusion between pseudo optical constants and actual material optical
constants when films are present, CompleteEASE hides the pseudo optical
constants by default. To view the pseudo optical constants, you must turn this

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-111

 option on. Click Options > Configuration Controls > ‘Edit Configuration’ to open the
CompleteEASE Configuration dialog shown in Figure 4-1.

Figure 4-1. CompleteEASE Configuration dialog.

Check the “Show Advanced Config. Options” box and expand the General section.
Find “Display Pseudo Transforms” in the list and check the “Parameter ON”
checkbox. Click “Close” when finished. You will need to restart CompleteEASE for
this change to take effect.
After restart, the pseudo optical constants <n> & <k> (or the pseudo dielectric
functions <e1> & <e2>) of the current data set can be displayed by selecting the
“<Pseudo> Transforms” option from the +Data menu above the graph, as shown in
Figure 4-2. Viewing the data in terms of pseudo optical constants can be useful for
bulk samples.

Figure 4-2. Select <Pseudo> Transforms from the +Data menu above the graph to display
pseudo optical constants or the pseudo dielectric function.

NOTE: The pseudo optical constants are NOT equivalent to the true optical
constants if multiple reflections are present in the data, which occurs when surface
layers are present on the sample.

Open the “Ni Substrate” example data file and view the pseudo optical constants
Shortcut:
shown in Figure 4-3. Notice that <n> and <k> are the same for each angle. Because
CTRL+Y to turn double Y the Ni layer is opaque, only a single reflection occurs from each surface. In this
axis ON or OFF case, the pseudo optical constants are equivalent to the true optical constants. The
data must still be modeled to create a material file that contains the optical

4-112 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 constants for subsequent analysis of any coatings applied to this substrate. We will
demonstrate this concept using two different layer types: the WvlbyWvl layer and
the B-Spline layer.

Figure 4-3. Pseudo optical constants for the “Ni substrate” data.

WvlByWvl Layer
Open the “Ni Substrate” example data file if needed. Clear the Model panel by
clicking Analysis > Model > ‘Clear’. With a blank starting model, click “none” to
load a material file for substrate. From the Library tab, Advanced folder, choose the
“WvlByWvl.mat” file and click ‘Open’ (see Figure 4-4).

Figure 4-4. The WvlByWvl.mat file is located in the Advanced folder of the Library tab.

After loading the WvlByWvl layer into the model, expand the layer by clicking the +
sign (see Figure 4-5).

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-113

 Figure 4-5. Model with WvlByWvl layer as the substrate.

The default starting values for the WvlByWvl layer are n = 1.5 and k = 0 (at all
wavelengths). These values are not representative of a metal as metals are always
absorbing (k > 0). Good starting values are important for the fitting algorithm to
find the correct result. It is best to use starting optical constants close to the
expected values for the material, so you should change the initial values for n and k.
In this case, you can use the pseudo optical constants as a guide for the initial
values. For this example, you can try n = 3 and k = 5 (<n> ranges from 1 to 5; <k>
ranges from 2 to 8).
A more robust way to set the starting values is to load a starting material file into
the WvlByWvl layer. Click on (Starting Mat =) none and choose Ni.mat from the
Metal folder of the Library tab (see Figure 4-6). This will enter the Ni values from
Palik’s Handbook of Optical Constants of Solids as the starting values at each
wavelength within the WvlByWvl layer.

NOTE: Metal optical constants can vary significantly from source to source and are
even discontinuous within Palik from use of multiple references. The
CompleteEASE library has many material files for metals, but they are often best
utilized as starting material files rather than as truly representative of your metal
material.

4-114 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-6. Use Ni.mat as the starting material file for the WvlByWvl layer.

Right-click on (Starting Mat = ) Ni to graph the starting optical constants (see Figure
4-7), and you should see the values as shown in Figure 4-8.

Figure 4-7. Right-click on “Ni” within the WvlByWvl layer and select “Graph Layer Optical
Constants” to view the starting optical constants.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-115

 Figure 4-8. Optical constants of Ni from Palik’s Handbook of Optical Constants of Solids.

Click ‘Generate’ to see how closely the library Ni values match the experimental
data. Figure 4-9 shows Psi only. The model-generated data is not identical to the
experimental data because the optical constants from Palik are not identical to the
measured sample.

Figure 4-9. Generated data from “Ni.mat” compared to “Ni substrate” example data.

Now, click ‘Fit’. The WvlByWvl layer allows n and k to vary independently at each
wavelength to best match the experimental data. This produces a near-perfect fit
to the data, as shown in Figure 4-10. Right-click on the WvlByWvl layer to graph the
optical constants (see Figure 4-11). These optical constants can now be saved as a
reference material file for this Ni material. Instructions for saving optical constants
are found in the following sections. Before proceeding, copy the optical constants
graph to the Graph Scratchpad (see Figure 4-12).

NOTE: In the WvlByWvl layer, each wavelength is fit independently, without

consideration of neighboring wavelengths.

4-116 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-10. Fit to “Ni substrate” example data using the WvlByWvl layer with Ni.mat
starting values.

Figure 4-11. Optical constants of the WvlByWvl layer after fitting the “Ni substrate” example
data.

Evaluation of fit result: Visually, there is good agreement between the model-
generated data and experimental data, and the MSE is low. The shape of optical
constants is typical for metals with absorption increasing into the IR. Noise in the
optical constants is a direct result of the WvlByWvl layer fitting each wavelength
independently, picking up noise from the data.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-117

 Figure 4-12. Right-click in the graph to copy the optical constants to the Graph Scratchpad.

B-Spline Layer
While the WvlByWvl layer works well for this simple substrate example, it does not
use neighboring wavelengths to help determine the optical constants. Any noise in
data will produce noise in the resulting optical constants. To avoid this problem,
the B-Spline layer can be used. Full details of the B-Spline layer will be discussed in
a later example. In this simple example, we will build a model to fit any metal
substrate and save this model for future use.
Start by clearing any existing model from the Model panel. (Click Analysis > Model
> ‘Clear’). Click on (Substrate = ) none and load the B-Spline layer from Library tab,
Basic folder (Figure 4-13).

Figure 4-13. The B-Spline layer is in the Library tab, Basic folder.

After loading the B-Spline layer into the model, expand the layer by clicking the +
sign (see Figure 4-14). The B-Spline layer has several subsections which can be
further expanded. Expand the +Nodes section.
For this example, set the “Resolution” to 0.1eV. Click on (Starting Mat =) none
(within the +Nodes section) and load the same Ni.mat file used in the previous
section (see Figure 4-6).

4-118 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-14. Model with B-Spline layer as the substrate with Resolution set to 0.1eV and
Ni.mat loaded as the starting material.

The B-Spline layer specifies the optical constants versus wavelength using a series
of control points (“nodes”) which are equally spaced in photon energy (eV). The
approximate spacing of the control points is controlled by the Resolution. When a
starting material is specified, the nodes are immediately matched to the starting
material file. The starting material file does not have to be the same as the
measured material, but chances of the fitting algorithm finding the correct solution
are improved by choosing a starting material with similar values.

NOTE: For many metal substrates, any metal starting values are usually close
enough.

Click ‘Generate’ to see how closely the starting B-Spline values match the
experimental data. Figure 4-15 shows Psi only.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-119

 Figure 4-15. Generated data from B-Spline with Resolution = 0.1eV and Starting Mat = Ni
compared to “Ni substrate” example data.

Now, click ‘Fit’. The algorithm quickly converges to the result shown in Figure 4-16.

Figure 4-16. Fit to “Ni substrate” example data using B-Spline layer.

To view the optical constants, right-click on “B-Spline” and choose “Graph Layer
Optical Constants” (see Figure 4-17 and Figure 4-18). To view the B-Spline nodes,
click on Draw Node Graph within the B-Spline layer (see Figure 4-19).

4-120 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-17. Right-click on B-Spline to graph layer optical constants.

Figure 4-18. Resulting optical constants from B-Spline fit to “Ni substrate” example data.

Figure 4-19. Nodes at 0.1eV resolution fit to the “Ni substrate” example data.

Evaluation of fit result: Again, there is good visual agreement between the model-
generated data and experimental data, and the MSE is low. The shape of optical
constants is typical for metals with absorption increasing into the IR.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-121

 Before proceeding, copy the optical constants graph to the Graph Scratchpad.
Right-click in the graph
Then, open the Graph Scratchpad and compare the results of the B-Spline and
and select “View Graph
WvlByWvl layers. Notice the B-Spline results are smooth and continuous, unlike
Scratchpad” to open
the WvlByWvl result, but otherwise they are identical (Figure 4-20).
the Graph Scratchpad

Figure 4-20. Results from WvlByWvl and B-Spline layers for “Ni substrate” example data
compared in the Graph Scratchpad (curve labels and colors altered for easier comparison).

Save Layer Optical Constants

When finished with the previous exercise, click the Layer Command Save to save
the Ni optical constants (Figure 4-21). This opens the Save Layer Optical Constants
dialog box as shown in Figure 4-22. First, click ‘Tabulated’ in the Save Type panel.
Next, click on the substrate layer itself. After entering the file name and comment
(optional), the tabulated n and k values at each measured wavelength will be saved
for future use.

Figure 4-21. Use layer command to save layer optical constants.

NOTE: The Layer Command Save is different from the ‘Save’ button in the Model
panel. The Layer Command Save saves the layer optical constants as a material
(.mat) file, while the ‘Save’ button in the Model panel saves a model (.mod) file
with the layered model structure and any +MODEL, +FIT, or +OTHER Options.

4-122 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-22. Save Layer Optical Constants dialog box.

Alternatively, the layer optical constants can also be saved by right-clicking the layer
name (in this case “B-Spline”) and selecting “Save Layer Optical Constants” (see
Figure 4-23).

Figure 4-23. Right-click on a layer name to choose “Save Layer Optical Constants” from a
drop-down menu.

CompleteEASE will ask how you want to save the optical constants. For B-Spline,
you can choose to save as a Fixed B-Spline, an Editable B-Spline, or Tabulated list of
n and k values.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-123

 Figure 4-24. Choose to save the optical constants as a ‘Fixed B-Spline’, ‘Editable B-Spline’, or
as a tabulated list of optical constants at each wavelength.

Save Model
After optimizing the analysis for a sample, the model can be saved for future use.
Let’s use this Ni substrate as an example. If additional Ni substrates will be
measured in the future, a customized model would help reduce analysis time. For
this example, let’s modify two additional settings:
• Replace the Starting Mat. in the B-Spline layer with the metal optical
constants saved in this example (previous step).
• Expand +FIT Options and turn Include Derived Parameters ON. Add
derived parameters for both ‘n’ and ‘k’ of the substrate (layer #0). This will
report n and k at the specified wavelength (default = 632.8nm) in the Fit
Results.
The model should now appear like Figure 4-25.

Figure 4-25. Ni substrate model saved for future use.

4-124 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 When finished, click the ‘Save’ button at the top of Model panel (Figure 4-26) to
save the final model as “metal substrate”. We will use this model in the next
example.

Figure 4-26. Use save button to save model.

NOTE: The Layer Command Save is different from the ‘Save’ button in the Model
panel. The Layer Command Save saves the layer optical constants as a material
(.mat) file, while the ‘Save’ button in the Model panel saves a model (.mod) file
with the layered model structure and any +MODEL, +FIT, or +OTHER Options.

4.2. Self-Assembled Monolayer on Au

FEATURES IN THIS EXAMPLE
• Parameter uniqueness • Fixing b-spline optical constants

NOTE: This example uses the “metal substrate” model created in Section 4.1. You
will need to complete Section 4.1 before proceeding to this example.
When considering a film on a metal substrate, it is strongly recommended to first
measure an uncoated substrate to determine the substrate optical properties. For
this example, data were acquired from a bare Au substrate and from an Au
substrate with self-assembled monolayer film.
First, open the “Au substrate” example data file and fit using the “metal substrate”
model created in the previous section. The starting values should be close enough
to simply click ‘Fit’. Results should appear as shown in Figure 4-27 and Figure 4-28.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-125

 Figure 4-27. Results of "metal substrate" model on "Au substrate" example data.

Figure 4-28. Resulting optical constants of Au substrate.

Evaluation of fit result: There is good visual agreement between the model-
generated data and experimental data, and the MSE is low. The shape of optical
constants is typical for metals with absorption increasing into the IR, and the n and
k values are similar to library values for Au.

NOTE: Before fitting any overlayers, it is important to now fix the optical constants
from the bare substrate analysis. Turn off the fitting of the substrate optical
constants before proceeding.
To fix the B-Spline optical constants, expand the B-Spline layer and turn off “Fit
Optical Constants” (Figure 4-29). This will prevent the substrate optical constants
from changing when fitting for subsequent films.

4-126 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-29. Turn "Fit Optical Constants" to OFF before fitting any added layers.

Self-Assembled Monolayers
Self-Assembled Monolayers (SAMs) are very thin organic layers that consist of a
single monolayer. They are generally less than 10nm thick and are often closer to
1-2nm. Ellipsometers are very sensitive to layers of this thickness due to the phase
information available from the measured Delta parameter. However, the index of
refraction from such thin layers is not typically attainable. Instead, the nominal
index of refraction must be assumed in order to fit the thickness. As layers become
thicker than 10-20 nm, sensitivity to index of refraction increases. A common
approach for films less than 10-20 nm thick is to measure a thicker film of similar
material. The index is determined from the thicker film and then used for
determining thickness of the thinner films.

NOTE: Assuming optical constants from published values or from thicker films may
not always be true, so measurement of SAMs or other very thin layers is often
qualitative and best used for comparison between multiple samples.
Open the “Organic on Au” example data file and click ‘Generate’. The graph should
appear as in Figure 4-30. The model-generated curves represent a bare Au surface,
as they come from the model which currently contains only the substrate. Notice a
shift in Delta at all wavelengths. This indicates the presence of a thin surface layer.
A larger shift indicates a thicker layer on the Au surface.

Figure 4-30. “Organic on Au” example data compared to model-generated data from bare
Au. The Delta shift indicates a surface layer.

Layers can also be Now, add a layer to the model by clicking the Layer Command Add. Position the
added by using blue bar above the substrate (Au represented by B-Spline in this model) and click to
ALT+plus sign. Use add the layer (Figure 4-31).
arrow keys to position
blue bar, then Enter.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-127

 Figure 4-31. Use the mouse to position the blue bar above the substrate and then click once
to add a layer to the model in the selected position.

Next, choose Cauchy.mat from the Library tab, Basic folder and click ‘Open’. The
default values for the Cauchy layer will be close to the values needed for most
organic films. We will assume these values are representative of our SAM.

Figure 4-32. Cauchy.mat is in the Library tab, Basic folder.

With the Cauchy layer added to the model, activate the thickness of layer 1 as a fit
parameter by right-clicking on the value. The model should now look like Figure
4-33. Make sure that “Fit Optical Constants” within the B-Spline layer is off, as
these values were determined from the bare substrate. Click ‘Fit’ and you should
obtain a thickness of 7.46nm (Figure 4-34).

4-128 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-33. SAM on Au model with SAM represented by the Cauchy layer and Au
represented by the B-Spline layer.

Figure 4-34. Thickness result for “Organic on Au” example data.

NOTE: Thickness units can be changed in Options > Display Units.

Evaluation of fit result: There is good visual agreement between the model-
generated data and experimental data, and the MSE is low. The thickness is typical
for SAMs.

Fit Parameter Uniqueness

We will use this result to demonstrate a feature in CompleteEASE that tests the
uniqueness of a result. In the Model panel, expand +OTHER Options. Here you will
find the Fit Parameter Uniqueness command. Click on this command, and a new
dialog box will appear (Figure 4-35). Up to two parameters can be tested
simultaneously, but for this example, we only have one. To test the uniqueness of a
given parameter, you can enter minimum and maximum values and number of tries
or “fit points” within the range. CompleteEASE will fit the data at each fit point and

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-129

 report the MSE for each fit. If the MSE is the same for a wide range of values, we
can conclude that the value is not unique. However, if the MSE is well-minimized at
the original fit value, we can conclude that the value is unique within the given
model.
For this example, enter a minimum value of 2.5nm, a maximum value of 15nm, and
100 fit points, as shown in Figure 4-35. Click ‘Ok’ to start the uniqueness test.
When finished, a profile of the MSE vs. each fit point will be displayed in the graph.
The result for this example is shown in Figure 4-36.

Figure 4-35. Parameter Uniqueness Fit dialog box.

Figure 4-36. Uniqueness test result of the thickness parameter for a self-assembled
monolayer on Au with assumed optical constants.

Right-click on the graph area and copy the MSE profile to the Graph Scratchpad.
We will compare this MSE curve with our next test.

NOTE: You may want to clear the Graph Scratchpad (if results from previous
examples are still present) before copying the uniqueness test result. To do this,
open the Graph Scratchpad and click ‘Delete All’. Then go back and add the
uniqueness test result.

4-130 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Now, let’s try fitting for thickness and optical properties of the SAM layer to see
what happens. Expand the Cauchy layer and right-click the A, B, and C parameters
to activate them as fit parameters. The model will now appear as in Figure 4-37.

Figure 4-37. Model with Cauchy parameters A, B, and C activated as fit parameters.

Repeat the uniqueness test for Thickness #1 with the same test conditions as
before. The result is shown in Figure 4-38. Add this result to the Graph Scratchpad
and then view the Graph Scratchpad to better compare the two results.

Figure 4-38. Uniqueness test result of the thickness parameter for a self-assembled
monolayer on Au with fitted optical constants.

Change the color of one of the curves by clicking on the color box and selecting a
new color from the subsequent window. Give each curve a unique name by
double-clicking and typing in the Curve Name column. The comparison is shown in
Figure 4-39.
The V-shape of the uniqueness test result for thickness only indicates good
sensitivity to the thickness parameter within this model. The shallow curvature of
the uniqueness test result for thickness and Cauchy parameters indicates poor
sensitivity to both thickness and refractive index. This is typical for thin films less
than ~10 nm. When working with very thin films (< 10 nm), it is recommended to
fit for thickness only, while setting refractive index at reasonable values. The
default Cauchy parameters are recommended for many organic or transparent
materials.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-131

 Figure 4-39. MSE profiles from both tests compared in the Graph Scratchpad.

4.3. Semi-Absorbing Thin Films (B-Spline)

FEATURES IN THIS EXAMPLE
• B-spline layer • Wvl. range expansion fit
• Starting material • Selecting a graph/fit range
• Saving a model • Zoom all

Absorbing films are more difficult to analyze than transparent films because the
general structure of an absorbing film’s optical constants versus wavelength is not
always known before the measurement. There are two layer types in
CompleteEASE most commonly used for absorbing films: 1) the B-Spline layer,
which allows arbitrary flexibility in n and k versus wavelength, and 2) the Gen-Osc
layer, which uses a summation of standard oscillator line shapes. B-Spline will be
used in this example; Gen-Osc will be used in Section 4.4.
The B-Spline layer is perfect for materials that are transparent in part of the
measured spectral range and absorbing in part of the measured spectral range.
Examples of such materials for UV-VIS-NIR are:
• a-Si, a-Ge, a-SiGe
• ZnS, ZnSe
• SiC
• TiO2

4-132 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 • Most organics, such as polyimide, PEDOT, P2VP, Alq3, P3HT, photoresists,
and many more
• Diamond-like carbon (DLC)

Absorbing vs. Transparent Region

Open the “TiO2 on Si” example data file. Notice that Psi and Delta oscillate (Figure
4-40) as we saw in Chapter 3 for transparent film examples. These oscillations are
caused by interference between light reflecting from the top of the film and light
reflecting from the bottom of the film. The number of oscillations is correlated to
film thickness: thicker films produce more oscillations.
Notice the interference oscillations decrease in amplitude toward shorter
wavelengths. This is a strong indication of absorption in the ultraviolet spectrum.
The oscillations are dampened, as less light can penetrate the film and return to the
surface to interfere with the top-reflected light. This concept is illustrated in Figure
4-41. To model the TiO2 film on Si, we will need a material file that can describe
both transparent and absorbing regions.

Figure 4-40. Data from “TiO2 on Si” example data. The interference oscillations in data
suggest the film is transparent at longer wavelengths, while the dampening of these
oscillations at short wavelengths indicates UV absorption.

Figure 4-41. Interference oscillations occur when the light can travel through a thin film and
return from the bottom interface (red arrows). If the light is absorbed, the light only comes
from the top-surface (blue arrows) and will not produce interference oscillations.

‘Set Ranges’
The typical approach for modeling semi-absorbing thin films is to first fit the
transparent region only to determine film thickness and index in this region. In this

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-133

 manner, we can build from concepts demonstrated in Chapter 3. To start, we must
identify and select the transparent region of the data.
If you position the mouse where the oscillations begin to dampen, CompleteEASE
will display the graph coordinates of the mouse position. As shown in Figure 4-42,
the oscillations in the TiO2 on Si data set start to dampen around 330nm. To fit the
transparent range only, we want to select only wavelengths longer than this value.
In the Data panel, click on ‘Set Ranges’ (Figure 4-43) to open a dialog where you can
specify the min. and max. wavelengths (Figure 4-44).

NOTE: For the purpose of selecting the transparent range of a semi-absorbing film,
the min. and max. wavelengths do not need to be exact. Typically, for UV-
absorbing materials, you can enter a conservative value in the min. wavelength box
and leave the max. wavelength at the default value (which is the maximum
measured wavelength).
To ensure only the transparent region is selected, enter “400” as the min.
wavelength and click ‘Ok’ to close the dialog box. The graph will update to display
only the selected range (Figure 4-45).

Figure 4-42. Interference oscillations start to dampen around 330nm.

Figure 4-43. Data panel buttons, including the ‘Set Ranges’ button.

4-134 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-44. Select Data Range dialog box.

Figure 4-45. “TiO2 on Si” example data after selecting wavelengths 400nm and longer.

Alternatively, you can also simply click and drag the mouse within the graph to
select the transparent region. First, you will need to reselect the entire spectral
range. To do this, you can:
1) Re-open the example data file.
2) Reset the range from Data > ‘Set Ranges’ (enter 0 or any value below the
minimum measured wavelength for minimum wavelength).
3) Right-click in the graph and select “Zoom All”, click the “Zoom All” button in the
upper-right area of the graph, or double-click anywhere on the graph to
recapture the full measured range (see Figure 4-46).

Figure 4-46. Right-click in the graph and select "Zoom All" to recapture full range of
measured data.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-135

 Now, to select the transparent region only, position your mouse near the
wavelength where the oscillations begin to dampen. Again, the position does not
need to be exact but should err to the transparent side. Click and drag your mouse
over the graph. As you drag, the selected region will be highlighted gray (see Figure
4-47). When you reach the end of the range, release the mouse and the graph will
update to show only the selected data range.

Figure 4-47. Another way to select a range of data is to click and drag mouse over the graph.

Transparent Region Analysis

With only the transparent region selected, you can follow procedures learned in
Chapter 3 for a transparent film on Si. For this example and others like it, you can
simply load the “Si with Transparent Film” model and click ‘Fit’. Results are shown
in Figure 4-48 and Figure 4-49.

NOTE: Remember to evaluate the result before proceeding. If you do not have a
good result for the transparent region, you may have difficulty expanding the
analysis into the absorbing region.

4-136 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-48. Results of fitting transparent region (400-1700nm) of “TiO2 on Si” example data
set with “Si with Transparent Film” model.

Figure 4-49. Optical constants of transparent region.

Evaluation of fit result: The model-generated data and experimental data generally
agree, and the resulting optical constants are physically plausible with index around
the nominal expected value for TiO2 and only increasing into the UV. The MSE is 18
and there seems to be room for improvement in the visual match between the
curves. As learned in Chapter 3, we can use the Try Alternate Models command to
test for common complexities.

Expand +OTHER Options within the Model panel and click Try Alternate Models.
Results are shown in Figure 4-50. CompleteEASE suggests including surface
roughness in the model to reduce MSE to ~8. Click ‘Apply Chosen Model’ to accept
this recommendation and apply the change to the model.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-137

 Figure 4-50. Results of Try Alternate Models command for transparent range (400-1700nm)
of “TiO2 on Si” example data set with “Si with Transparent Film” model.

NOTE: If you are only interested in film thickness and/or the index value within this
range, the analysis is complete. It is not necessary to analyze the absorbing region
just because it was measured. The following steps demonstrate expansion into the
absorbing region, assuming that the optical constants or other properties in the
absorbing region are goals of the measurement.

Expansion to Absorbing Region

Now that we have a good fit to the transparent region of the data, we know the film
thickness and optical constants near the onset of the absorbing region. These
serve as good starting values for expansion into the absorbing region. First, we
must replace the Cauchy layer with a layer capable of modeling absorption (k > 0).
As suggested earlier in this chapter, we will use a B-Spline layer. Rather than simply
replacing the Cauchy layer with a B-Spline (which would cause us to lose the optical
constants already determined for the transparent region), we can convert the
existing Cauchy into a B-Spline by right-clicking on the layer and selecting
“Parameterize Layer” (Figure 4-51). Choose “B-Spline” from the side menu.

Figure 4-51. Right-click “Cauchy Film” and select “Parameterize Layer”. For this example, we
will choose B-Spline parameterization.

A new window will appear which contains the existing model (Cauchy) values and
“New Layer” (B-Spline) values (see Figure 4-52). The B-Spline will automatically
mimic the Cauchy values. We will learn more about this window in Section 4.4. For
now, simply click ‘Replace Layer’ to finalize conversion of the Cauchy layer into a B-
Spline layer. The model should now appear as shown in Figure 4-53.

4-138 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-52. Parameterize Layer window.

Figure 4-53. Model updated from Cauchy to B-Spline. Note the fit results and model curves
are left-over from the previous fit.

B-Spline Layer
Expand the B-Spline layer by clicking the “+” to view layer details. Recall from
Section 4.1 that the B-Spline layer specifies optical constants versus wavelength
using a series of control points which are equally spaced in photon energy (eV). The

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-139

 approximate spacing of the control points is controlled by the resolution in units of
eV, and the resulting number of control points within the current spectral range is
listed after this value. Basis (“B-”) spline curves (which are smooth and continuous
in the 0th, 1st, and 2nd derivatives, by definition) are used to interpolate the optical
constants between the control points.

Figure 4-54. Expanded B-Spline layer.

Click on Draw Node Graph to display the node positions as shown in Figure 4-55.
The B-Spline layer always works in terms of the complex dielectric function  = 1 +
i2 = (n + ik)2, which is often written as “e1 & e2”. One set of control points (and
associated spline curve) is used to specify the dispersion in “e1”, and a second set
of control points is used to describe “e2”. Using a smaller resolution value adds
more control points, resulting in more flexibility to accommodate sharper features
in the dielectric function spectra. However, setting the resolution too low can
result in unwanted noise or unphysical artifacts in the dielectric function spectra.
Further, the more control points you add, the more time CompleteEASE will take to
calculate fits.

Figure 4-55. B-Spline dielectric function “e1 & e2” showing the location of the “control
points” (red and green dots), which are equally spaced in eV.

The “Fit Opt. Const. = ” feature allows you to turn on and off the fitting of control
points from within the B-Spline, as was demonstrated in Section 4.2.
In the B-Spline layer, “Use KK mode” is an advanced calculation that maintains
Kramers-Kronig consistency between the resulting e1 and e2 spline curves. When
this feature is ON, the optical constants will be constrained to a physical shape.
This feature will also reduce the total number of fit parameters, as only the e2
spectrum is described with a series of control points while the e1 spectrum is

4-140 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 calculated from the e2 spectrum via the Kramers-Kronig transformation. Additional
features of the B-Spline layer are detailed in Chapter 10.

Wvl. Range Expansion Fit

To expand the current B-Spline layer, which currently only describes the
transparent range, we will use the Wvl. Range Expansion Fit command. This
command can be found in the Model panel under +OTHER Options, as shown in
Figure 4-56.

Figure 4-56. +OTHER Options in the model, including the Wvl. Range Expansion Fit
command.

The Wvl. Range Expansion Fit command starts with the currently selected
wavelength range, applies the current model, then gradually increases the range to
shorter (and longer, if applicable) wavelengths until the entire data set is included.
A fit is performed at each new increment of wavelengths. The default increment
value is 0.5 eV, which works well for broad absorption features, but smaller
increments may be needed for materials with narrow absorption features. You can
consider setting the increment value to match the B-Spline node resolution such
that the fitting process adds one B-Spline node at a time.
For this example, the default increment of 0.5 eV is fine. Click on the Wvl. Range
Expansion Fit command, and CompleteEASE will prompt you to choose the
wavelength expansion range. The default values will be the min. and max. of the
available data, which is what we want for this example. Click ‘Ok’ to start.

Figure 4-57. Expansion Wvl. Range dialog box.

CompleteEASE will add data in 0.5 eV increments, fit the newly expanded range,
and repeat until the full range specified for the Expansion Wvl. Range (Figure 4-57)
is included. The results of this Wvl. Range Expansion fit are shown in Figure 4-58
and Figure 4-59. As always, the results should be evaluated.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-141

 Figure 4-58. Results of Wvl. Range Expansion Fit for “TiO2 on Si” example data.

Figure 4-59. Optical constants after Wvl. Range Expansion Fit.

Evaluation of fit result: There is good visual agreement between the model-
generated data and experimental data, and the MSE is low. Index increases toward
the UV in the transparent region and decreases in the absorbing region. Are the
results Kramers-Kronig consistent? Does k drop below 0 near 450nm?

Upon evaluation of the optical constants, it appears that the results maintain the
physical relationship between n and k, but it can be difficult to determine by eye.
Turn on the KK Mode feature within the B-Spline layer to ensure that this physical
relationship is enforced.

4-142 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-60. "Use KK Mode" will enforce the physical Kramers-Kronig relationship between n
and k.

Click and drag the mouse over the optical constants graph to zoom in on a small
region near 500nm. Notice that k is slightly negative in this range. We can improve
the physicality of the results by constraining the B-Spline layer such that k (e2) is
always positive. Expand the +Nodes section within the B-Spline layer and turn
“Force E2 Positive” to ON. Even with this change, you may notice some small
fluctuations in k around 0 in the transparent region (see Figure 4-61).

Figure 4-61. TiO2 results with “Use KK Mode” and “Force E2 Positive” features on.

These fluctuations are not likely real and can be ignored, or we can further improve
the physicality of the results by assuming a transparent region. This feature is also
found within the +Nodes section within the B-Spline layer and allows you to define
the transparent region with a band gap (Assume Band Gap = ON) or by setting the
min. and max. wavelengths of the transparent region (Assume Band Gap = OFF).
For this example, turn “Assume Transparent Region” to ON, “Assume Band Gap” to
OFF, and set the transparent region from 400nm to 10000nm (infinity
approximation for this data range; effectively selects data 400nm and longer).
The final model and results for the “TiO2 on Si” example data set are shown in
Figure 4-62, Figure 4-63, and Figure 4-64.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-143

 Figure 4-62. Final model for "TiO2 on Si" example data.

Figure 4-63. Final results for "TiO2 on Si" example data.

Figure 4-64. Final optical constants for the TiO2 layer in “TiO2 on Si” example data.

4-144 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Evaluation of fit result: The thickness and surface roughness are about the same as
the unconstrained B-Spline result. Optical constants are more physically
meaningful. MSE is slightly higher, but we are still getting a good visual match to
the experimental data. Higher MSE is acceptable for improved physicality in the
model.

NOTE: CompleteEASE has a feature to automatically execute the previous steps. To

convert a transparent layer to B-Spline, followed by a Wvl. Range Expansion Fit, go
back to the Transparent Region Analysis on page 4-136. From the transparent
range result, right-click on the Cauchy layer and select “Convert to Transparent B-
Spline” (Figure 4-65). Follow the prompts shown in Figure 4-66, Figure 4-67, and
Figure 4-68. The result will be very similar to Figure 4-63 and Figure 4-64, with Use
KK Mode = ON, Force E2 Positive = ON, and Assume Transparent Region = ON.

Figure 4-65. “Convert To Transparent B-Spline” option.

Figure 4-66. Enter the desired resolution in eV.

Figure 4-67. Click ‘Yes’ to perform wavelength expansion fit.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-145

 Figure 4-68. Click ‘Ok’ to expand over the full measured range.

NOTE: The results of this example will be used for the next section. Save a
snapshot of this result before closing CompleteEASE. See page 3-108 for details on
saving snapshots.

4.4. Semi-Absorbing Thin Films (Gen-Osc)

FEATURES IN THIS EXAMPLE
• Gen-Osc layer • Parameterize layer
• Tauc-Lorentz oscillator • Cody-Lorentz oscillator

In the previous example, we introduced the analysis of semi-absorbing thin films

using the B-Spline layer. Another method is to use an oscillator model.
Gen-Osc.mat is a generalized oscillator layer in CompleteEASE containing various
oscillator types which can be used to model absorptions. In this example, we
continue the example from the previous section, only this time we will use a Tauc-
Lorentz oscillator to describe the TiO2 film.

NOTE: Section 4.3 must be completed before proceeding with this example. If
needed, open the snapshot saved at the end of Section 4.3. See page 3-108 for
details on opening snapshots.

Gen-Osc Layer
The Gen-Osc layer has a few advantages over the B-Spline layer. 1) Oscillators
provide a physically meaningful model as they represent the dipole response of the
material. 2) Oscillators can be used to determine comparable, physical parameters
such as band gap. 3) The Gen-Osc layer always enforces Kramers-Kronig
consistency.
A disadvantage is that it can be difficult to know where to place oscillators without
a reference. Therefore, the most common approach with the Gen-Osc layer is to
start with a reference set of optical constants and place oscillators accordingly. The
reference optical constants can be from the CompleteEASE library, from a previous
or published result, or from an initial WvlByWvl or B-Spline fit.
In this example, we will start with the final B-Spline result from Section 4.3. With
this result open in CompleteEASE, right-click on the B-Spline layer and select
“Parameterize Layer”. Choose “GenOsc” from the side menu, as shown in Figure
4-69.

4-146 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-69. Right-click a layer to parameterize to GenOsc.

A new window will appear which contains the existing model (B-Spline) values and
“New Layer” (Gen-Osc) values (see Figure 4-70). Unlike B-Spline parameterization,
which automatically mimics the reference values, oscillators must be manually
placed with a Gen-Osc parameterization.

Figure 4-70. “Parameterize Layer” window for Gen-Osc layer.

The goal is to match the line shape of the reference values with one or more
oscillators. The Gen-Osc layer allows a combination of various oscillators which are
detailed in Chapter 10. For this example, we will use a Tauc-Lorentz oscillator to
describe the single, asymmetrical absorption feature in this material, which is
typical for many amorphous materials.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-147

 The basic steps for matching optical constants with a new oscillator model are as
follows:
1. Match e2 by placing, adjusting, then fitting the oscillator parameters.
2. Match e1 by adjusting and then fitting the e1 offset or pole values.
3. Final adjustment to all parameters to fit both e1 and e2.

Each of these three steps will be demonstrated for this example, where we match
the previous B-Spline results with a Tauc-Lorentz oscillator.

Step 1: Match Imaginary Part Only

Most oscillators are Kramers-Kronig consistent. This provides an internal coupling
between the real (e1, n) and imaginary (e2, k) optical properties. As we adjust the
oscillator parameters to match the shape of e2 from our B-Spline result, we will
automatically adjust the e1 shape via the KK transformation. Step 2 will allow us to
further adjust only e1, which is why it is important to start with e2 matching.
To focus on e2 only, select “Imaginary Part only” from the Display panel of the
Parameterize Layer window (see Figure 4-71).

Figure 4-71. From the Display panel, select View/Fit: “Imaginary Part Only”.

The window will simplify to show only imaginary curves and components (see
Figure 4-72).

4-148 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-72. Parameterize Layer window after selecting to display “Imaginary Part Only”.

To add an oscillator, click the Oscillator Menu Add command within the Layer
Parameters panel. CompleteEASE will prompt you to choose the oscillator type you
wish to add. For this example, choose Tauc-Lorentz (see Figure 4-73). You can read
more about this specific oscillator type in Section 10.4. Click ‘Ok’.

Figure 4-73. Choose Tauc-Lorentz oscillator type.

CompleteEASE will place a Tauc-Lorentz oscillator into the layer with nominal values
for each of the oscillator parameters (see Figure 4-74).

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-149

 Figure 4-74. Parameterize Layer window after adding a Tauc-Lorentz oscillator.

Next, adjust the oscillator parameters such that the Tauc-Lorentz oscillator shape
resembles the reference curve. You can click on each of the blue, underlined
oscillator values within the Layer Parameters panel to enter new values, but
CompleteEASE provides control boxes in the graph for easier, visual manipulation.
Click and drag on any of the control boxes in the graph and watch how the oscillator
changes. Manipulate the band gap parameter control box such that it is located
near 3.2eV. Manipulate the amplitude and center energy control box such that the
amplitude and center energy of the Tauc-Lorentz oscillator approximately match
the amplitude and center energy of the reference curve. Lastly, manipulate the
broadening such that the broadening of the Tauc-Lorentz oscillator approximately
matches the broadening of the reference curve. When finished, your window
should look something like Figure 4-75.

NOTE: Left-click on the oscillator number in the Layer Parameters panel to select
an oscillator and view its control boxes. Right-click on the oscillator number to
delete the oscillator. Use ‘Expanded Axis’ button to view control boxes that are out
of range of the reference data.

4-150 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-75. Parameterize Layer window with Tauc-Lorentz oscillator adjusted to
approximate the reference curve.

With the oscillator shape approximately matching the reference values,

CompleteEASE can finalize the adjustment by fitting the oscillator to the reference
values. Right-click on each of the Tauc-Lorentz parameters or select Fit Menu: All
to activate all four parameters as fit parameters. Next, locate the Fit panel within
this Parameterize Layer window and click ‘Fit’.

NOTE: The Fit panel within the Parameterize Layer window is NOT the same as the
Fit panel in the Analysis tab. In the Parameterize Layer window, we are NOT fitting
the experimental data. We are fitting the oscillator parameters to reference optical
constants. We will fit the experimental data in a later step.

CompleteEASE varies the oscillator parameters to match the imaginary curve shape
of the reference (B-Spline) values (see Figure 4-76).

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-151

 Figure 4-76. Results of fitting the Tauc-Lorentz oscillator to reference values.

Step 2: Match Real Part Only

When finished matching the imaginary part with an oscillator, change the Display to
“Real Part Only”.

Figure 4-77. From the Display panel, select View/Fit: "Real Part Only".

The oscillator list will be hidden from view, as oscillators affect e2 only.
CompleteEASE automatically calculates e1 from e2 through the Kramers-Kronig
relationship (see Figure 4-78).

NOTE: The Kramers-Kronig equations integrate from 0 to infinity, so absorptions

outside of the measured spectral range can still affect e1. An offset or tilting of e1
is often necessary to account for absorptions outside of the measured spectral
range.

4-152 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-78. Parameterize Layer window displaying “Real Part Only”.

The e1 curve can be adjusted through a constant offset, UV pole, and/or IR pole.
Poles are unbroadened oscillators placed outside of the measured spectral range to
mimic the effect of absorption outside of the measured spectral range. A UV pole
will tilt the e1 curve upwards at short wavelengths; an IR pole will tilt the e1 curve
downwards at long wavelengths.
Experiment with the values of Einf (constant offset), UV Pole Amp., and IR Pole
Amp. and view the effect on e1. Position your mouse over one parameter at a time
and use SHIFT+CTRL+mouse scroll wheel to view changes. Notice the Einf
parameter raises or lowers the overall curve. The UV Pole Amp. tilts e1 up at short
wavelengths; the IR Pole Amp. tilts e1 down at long wavelengths. The IR pole
position cannot be changed. The UV pole position (UV Pole En.) default value is
11eV but can be higher or lower (farther or closer to measured spectral range) as
needed.
Reset the parameter values to their default values (Einf = 1.000, UV Pole Amp. = 0,
UV Pole En. = 11.000, and IR Pole Amp. = 0). Right-click on the Einf parameter value
to activate Einf as a fit parameter. Then click the ‘Fit’ button.

NOTE: The Fit panel within the Parameterize Layer window is NOT the same as the
Fit panel in the Analysis tab. In the Parameterize Layer window, we are NOT fitting
the experimental data. We are fitting the oscillator parameters to reference optical
constants. We will fit the experimental data in a later step.
See Figure 4-79. Notice the e1 curve needs to tilt up at low wavelengths. Try
adding the UV Pole Amp. and En. as fit parameters. Click ‘Fit’ again. Now the e1
curve closely resembles the reference curve at all wavelengths (see Figure 4-80).

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-153

 Figure 4-79. e1 curve adjusted with Einf parameter only.

Figure 4-80. e1 curve adjusted with Einf and UV pole.

4-154 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Step 3: Match All
If steps 1 and 2 were successful, the Gen-Osc will now match both real and
imaginary values of the reference. Change the Display panel to view/fit “All”
(Figure 4-81). There can be correlation between values used to match the e2 curve
and values used to match the e1 curve. For example, if an oscillator is centered
outside the measured spectral range, the data may not have sensitivity to the exact
position, and its effect may correlate with an e1 pole. You can perform a final fit to
both curves by clicking the ‘Fit’ button once more to balance any misfitting and
reduce any unneeded parameters.

Figure 4-81. From the Display panel, select View/Fit: “All”.

Figure 4-82. New Gen-Osc layer with e2 and e1 curves approximate to reference values.

NOTE: If the reference or initial values were not KK consistent, you will not be able
to get an exact match to the model curves. This is okay as the desired final
outcome is KK consistent optical properties.

Replace Layer
When you are finished optimizing the e2 and e1 curves to match the reference
curves, click ‘Replace Layer’ to apply the new oscillator layer in your model. Click
‘Generate’ to ensure that your model-generated data still resemble the
experimental data (see Figure 4-83).

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-155

 Figure 4-83. Model panel with new Gen-Osc layer expanded. Model-generated data still
resemble experimental data.

The new oscillator layer can now be used to fit data. Click ‘Fit’ to see how well the
Tauc-Lorentz oscillator can describe the experimental data (see Figure 4-84).

Figure 4-84. Result from one Tauc-Lorentz oscillator and UV pole to fit “TiO2 on Si” example
data.

4-156 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Evaluation of fit result: At first glance, the fit quality looks okay, but the MSE is
almost double the B-Spline result. Zoom in on UV region to see misfit (see Figure
4-85).

Figure 4-85. Zoomed in view of experimental data to show misfit in the UV region.

Let’s look at the Gen-Osc parameterization again. Expand the Gen-Osc layer and
click Show Dialog to reopen the Gen-Osc dialog. Try adding a small Gaussian
oscillator in the UV to better match the reference values; click the oscillator
command Add or hold the SHIFT key while pointing and clicking in the graph (see
Figure 4-86).

NOTE: CompleteEASE will add the most recently used oscillator type. You can click
on the blue, underlined type name to change from Tauc-Lorentz to Gaussian.
Repeat steps 1 through 3 and click ‘Close’ when finished. Fit the data again with the
two-oscillator layer. The final results are presented in Figure 4-87 and Figure 4-88.

Figure 4-86. Try adding a small Gaussian oscillator in the UV.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-157

 Figure 4-87. Final result for Gen-Osc modeling of “TiO2 on Si” example data.

Figure 4-88. Final optical constants for the TiO2 layer.

Evaluation of fit result: MSE is now in agreement with previous B-Spline result, with
better fit to the experimental data in the UV.

Save Model
If you would like to save a model for future use on similar samples, click the ‘Save’
button at the top of the Model panel (see Figure 4-89).

Figure 4-89. Use the ‘Save’ button to save model.

4-158 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 NOTE: The Layer Command Save is different than the ‘Save’ button in the Model
panel. The Layer Command Save saves the layer optical constants, while the ‘Save’
button in the Model panel saves the layered model structure and any +MODEL,
+FIT, or +OTHER Options.

4.5. Multiple Layer Film Stacks

FEATURES IN THIS EXAMPLE
• Superlattice • Coupling

Many applications include multilayer structures which are typically more

complicated than single-layer structures. Additional layers increase the number of
unknowns (thickness, n, and k for each layer) while the data content remains
generally the same (Psi and Delta at each measured wavelength). Whenever
possible, it is recommended to determine optical constants of each material first as
an individual layer (Figure 4-90) or as a progression of layers (Figure 4-91).

Figure 4-90. Determine optical properties one at a time through individual layer samples.

Figure 4-91. Determine optical properties one at a time through progression of layers.

Sometimes these methods are not possible or not successful, for example, if the
material properties in the stacked condition differ from the material properties in
the isolated condition. In this case, the typical approach is to build the starting
model with as much information as possible, using library or other published values
as starting values for each layer. In this example, we will consider a simple case of
SiN over SiO2 on Si.
Open “multilayer” example data file. The file comment indicates that the sample
structure is 90nm SiN over 15nm SiO2 on silicon. We can use this information to
build a starting model. First, clear any existing model by clicking the ‘Clear’ button
in the Model panel.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-159

 Figure 4-92. Click 'Clear' to clear the Model panel and replace with a blank starting model.

Next, build the expected sample structure using the following library files for each
material. Click on (Substrate = ) none to add “Si_JAW.mat” as the model substrate
(see Figure 4-93).

Figure 4-93. Si_JAW.mat is located in the Library tab, Semiconductor folder.

Then click Add, position the blue bar above the substrate, and click (Figure 4-94).
Add “SiO2_JAW.mat” as the first layer (Figure 4-95) and enter 15nm as the layer
thickness. The model should now look like Figure 4-96.

Figure 4-94. Position blue bar above substrate and click to add layer.

4-160 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-95. SiO2_JAW.mat is located in the Library tab, Semiconductor folder.

Figure 4-96. Model with first layer added.

Click Add again, position blue bar above layer 1, and click (see Figure 4-97). Add
“SiNx (CodyLor).mat” as the second layer (see Figure 4-98) and enter 90nm as the
layer thickness. The model should now look like Figure 4-99.

Figure 4-97. Position blue bar above the first layer and click to add a second layer.

Figure 4-98. SiNx (CodyLor).mat is located in the Library tab, Dielectric folder.

Figure 4-99. Model with both layers added.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-161

 With the model structure in place, click ‘Generate’ to see if the model-generated
data is close to the experimental data (see Figure 4-100).

Figure 4-100. Model-generated data compared to experimental data.

It might be tempting to turn on fit parameters for all layer thicknesses and all layer
optical properties, but doing so will usually result in high correlations and
unphysical results. It’s possible that only a few parameters need to be adjusted to
fit the measured data.
First, try fitting for thicknesses only. Activate thickness #1 and #2 as fit parameters,
then click ‘Fit’. The result is shown in Figure 4-101.

Figure 4-101. Result after fitting layer thicknesses only.

4-162 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Evaluation of fit result: MSE is very high (189) with visual mismatch between the
model-generated and experimental data. We can likely improve the fit quality by
allowing one or more of the layer optical properties to vary.

Fitting the layer thicknesses was not enough to match the experimental data. Now,
we can consider letting one or more of the layer optical properties to vary. Thermal
SiO2 optical properties are generally very stable, and library values normally work
well for this material. SiNx is more variable, depending on process conditions and
stoichiometry.
Expand the SiNx layer, and you can see that this is a prebuilt Gen-Osc layer with one
Cody-Lorentz oscillator, as the file name indicates. Activate Amp1, Br1, Eo1, Eg1,
and Ep1 as fit parameters. Your model should now look like Figure 4-102. Click ‘Fit’
again and results should appear as in Figure 4-103.

Figure 4-102. Expanded SiN layer with Amp1, Br1, Eo1, Eg1, and Ep1 activated as fit
parameters.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-163

 Figure 4-103. Result after fitting thicknesses and SiN layer optical properties.

Evaluation of fit result: MSE is significantly improved to ~12 with visual agreement
between the model-generated and experimental data. Can the model be improved
further?

The above result is good and may be adequate for most applications. SiN is a higher
index material with absorption in the UV. Optical contrast to air (n=1, k=0) and
measurement at short wavelengths typically increases sensitivity to surface
roughness. Try adding surface roughness and refit. The result is shown in Figure
4-104.

4-164 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-104. Result after adding surface roughness.

Evaluation of fit result: MSE is significantly improved again from ~12 to ~6. There
seems to be adequate sensitivity to surface roughness for this sample and
measured spectral range.

The best fit to the data based on MSE includes surface roughness, but what effect
does including surface roughness have on the results? Thickness #2 is reduced by
~2nm and Thickness #1 is increased by ~2nm. The SiN optical properties with and
without roughness in the model are shown in Figure 4-105. The optical constants
are very similar with only small changes in the UV and insignificant changes at long
wavelengths. If this level of change is significant in your research, it will be
important to include roughness in the model.

Figure 4-105. SiN results with and without roughness.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-165

 For more complicated structures, CompleteEASE has several features to minimize
the number of fit parameters and utilize known information about the sample.

Superlattice
If a bilayer structure is repeated, you can create a superlattice structure in the
model. Right-click the first layer of the repeated structure and choose “Start
Superlattice”. Then right-click the last layer of the repeated structure and choose
“End Superlattice” (see Figure 4-106).

Figure 4-106. Right-click layers to start and end superlattice.

With the superlattice structure enabled, the model now adds “SL Count” to the top
superlattice layer (Figure 4-107). Click on this value to change the number of
repeated superlattices (Figure 4-108).

Figure 4-107. "SL Cnt" = number of repeated superlattices.

Figure 4-108. Dialog box for superlattice count.

To clear the superlattice, right-click on the starting and ending superlattice layers
and select “Clear Superlattice”.

Thickness Pre-Fit
Thickness Pre-Fit can be helpful in fine tuning an unknown thickness of a layer
within a multilayer stack. See Section 3.2 for introductory details.

4-166 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 NOTE: Thickness Pre-Fit always operates on the thickest layer in the model.

Global Fit
A global fit for one or more parameters can be helpful in determining thickness or
index of an unknown layer. See Section 3.3 for introductory details.

Parameter Coupling
Individual parameters can be coupled to each other using a ratio, offset, or custom
equation. To access this function in the model, click Configure Options. Turn on
“Parameter Coupling” as shown in Figure 4-109. Click ‘Ok’ when finished, and this
option now appears under +MODEL Options section.

Figure 4-109. Show Model Options dialog box.

Figure 4-110. Parameter Coupling now available under +MODEL Options.

Click Add to open the Create/Edit Coupled Parameter dialog box (Figure 4-111).
Choose the Coupling Mode, Coupling Parameters, and enter the ratio, offset, or
equation. The equation used for coupling will be displayed at the bottom of this
dialog box. Click ‘Ok’ when finished to apply coupling to the model (Figure 4-112).
This feature is often useful for coupling layer thicknesses in a multilayer structure.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-167

 Figure 4-111. Create/Edit Coupled Parameter dialog box.

Figure 4-112. Exmaple model with Thickness #2 coupled to Thickness #1.

NOTE: Parameters must have unique names to properly couple. Rename

parameters if necessary. See Section 3.2 for instructions on renaming parameters.

Layer Coupling
CompleteEASE can also couple layer optical properties to other layers in the model.
Simply add or replace any layer with “Coupled.mat” (Figure 4-113). Then expand
the layer to designate which layer to couple (Figure 4-114).

4-168 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-113. Coupled.mat is located in the Library tab, Advanced folder.

Figure 4-114. Example model with Layer #2 coupled to Layer #1.

4.6. Thin Absorbing Films

FEATURES IN THIS EXAMPLE
• Appending data • Fitting SE and transmission data
simultaneously
• Parameter uniqueness • Graph scratchpad
• Interference enhancement

Earlier sections in this chapter demonstrate how to analyze opaque films (4.1) and
semi-absorbing films with a transparent region (4.3 and 4.4). What about films
which absorb at all wavelengths but are not yet opaque? This section will
demonstrate methods for analyzing thin absorbing films where k > 0 at all
measured wavelengths.
The challenge with thin absorbing films is that there are more unknown sample
properties (thickness plus n and k at each wavelength) than measured values (Psi
and Delta at each wavelength). This results in correlations between thickness and
optical properties and no unique answer.
Measuring multiple angles alone will not break these correlations. This section will
cover two methods for analyzing thin absorbing films: SE + Transmission and
Interference Enhancement.

SE + Transmission
This approach requires a transparent substrate. By combining SE data with
transmission intensity data, the amount of information contained in the data is now

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-169

 enough to determine the number of unknowns, and a unique result for thickness
and optical constants of the thin absorbing layer can be found.
Open the “Cr on glass_SE” example data file. This data was collected from a thin Cr
film on glass, with the backside taped to suppress backside reflections. Clear the
Model panel by clicking ‘Clear’, and build a starting model with the following
details:
• Add “7059 Glass (Cauchy).mat” as the model substrate (located in the
Library tab, Dielectric folder).
• Add “B-Spline.mat” as Layer #1 (located in the Library tab, Basic folder).
• Expand the B-Spline layer +Nodes section and enter “Cr.mat” as the
Starting Mat. (located in the Library tab, Metal folder).
• Turn on Thickness #1 as a fit parameter.
• Turn on Thickness Pre-Fit.

The model should now look like Figure 4-115:

Figure 4-115. Starting model for this example.

NOTE: Depending on the measured material and spectral range, optimization of

the B-Spline layer node resolution or other features may be required. For this
example, no further adjustments are required.

Click ‘Fit’ and the result in Figure 4-116 should appear.

4-170 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-116. Result from Cr on glass – SE data only.

Evaluation of fit result: MSE is low and there is good agreement between model-
generated and experimental data. Thickness is 3.19nm. Is this result unique? Click
‘Fit’ again and you will get a new thickness result with similar visual fit quality and
similar MSE. This is due to correlations between thickness and optical properties.

Parameter Uniqueness
Perform a uniqueness test for the Cr thickness (see page 4-129 for introductory
details). Expand +OTHER Options in the Model panel and find Fit Parameter
Uniqueness. Click on this command, enter minimum value of 1, maximum value of
30, and 30 fit points (see Figure 4-117). Click ‘Ok’. The uniqueness test result is
shown in Figure 4-118. Notice the MSE does minimize around 20nm, but the MSE
scale over the entire tested range is small, which indicates low sensitivity to the
exact value of this parameter. Copy the results of this test to the Graph Scratchpad
before proceeding.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-171

 Figure 4-117. Parameter Uniqueness Fit dialog.

Figure 4-118. Uniqueness test result for Cr thickness from 1 to 30 nm with SE data only.

Append Data
Now add transmission intensity data to the analysis. Right-click Analysis > Data >
‘Open’ button and select “Append Data” (see Figure 4-119). Browse to the Library
tab, Examples folder, select “Cr on glass_T”, and click ‘Append’ (see Figure 4-120).
This file contains intensity transmission data acquired from the same sample.

Figure 4-119. Right-click on the 'Open' button to append data.

4-172 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-120. "Cr on glass_T.SE" file location.

Notice the additional graph menus that appear above the graph when multiple data
sets are open. Use the drop-down menus to view each data set (see Figure 4-121).

Figure 4-121. Additional graph menus for multiple data sets.

With data set #2 (transmission intensity) selected, click ‘Generate’. Notice the
model-generated data do not agree with the experimental data in the UV-VIS (for
Layer #1 thickness ~8nm) (see Figure 4-122).

Figure 4-122. Model-generated and experimental transmission intensity data.

Click ‘Fit’ and the result in Figure 4-123 should appear.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-173

 Figure 4-123. Result from Cr on glass – SE+T data.

Evaluation of fit result: MSE is low and there is good agreement between model-
generated and experimental data for both data sets. Thickness is 20.86nm. Is this
result unique? Click ‘Fit’ again and you should get the same answer. This is
because the additional information contained in the transmission data breaks the
correlation between Cr thickness and optical properties.

Perform another uniqueness test using the same conditions as before. Notice the
MSE minimum is better defined with a larger scale for MSE (Figure 4-124). Copy the
result to the Graph Scratchpad for easier comparison (Figure 4-125).

Figure 4-124. Uniqueness test result for Cr thickness from 1 to 30 nm with SE+T data.

4-174 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-125. Comparison of uniqueness test results for SE data only and SE+T data.

NOTE: This method works best if you first analyze SE+T from a bare substrate.
Then, use the results from SE+T analysis of the bare substrate in your model when
analyzing SE+T for the coated substrate.

This example utilized a highly transparent substrate with negligible contribution to

the transmission intensity data from the substrate. This is not true for all substrate
types and is dependent on spectral range. Best practice of SE+T approach may
include two additional steps: specifying substrate thickness and increasing
transmission data fit weight.

Substrate Thickness
Standard modeling of SE data does not require accurate substrate thickness.
However, if backside reflections are present in the data or if data are acquired in
transmission mode (such as intensity transmission data utilized in the SE+T
approach for thin absorbing films), you may need to enter an accurate substrate
thickness. To do so, you must first turn on Substrate Backside Correction, located
under +MODEL Options. With this option enabled, you can now edit the Substrate
Thickness (see Figure 4-126). You may also need to consider the “# Back
Reflections” value. This value ONLY applies to the SE data. If backside reflections
were suppressed during measurement of the SE data (see Section 3.4 for review on
this topic), set this value to 0.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-175

 Figure 4-126. Turn on Include Substrate Backside Correction and set # Back Reflections to 0
for SE+T analysis where the SE data has backside reflections suppressed.

Transmission Data Weighting

Another factor you may need to consider for SE+T analysis is the transmission data
weighting. Sometimes it may be necessary to increase this weighting to find an
accurate fit to the transmission data or a balance in fit between the SE and T data.
To change this weighting, click on the Configure Options command at the bottom
of the Model panel. This will open the Show Model Options window. Scroll down
to Fit Options and select “Transmission Data Weighting” (Figure 4-127). Click ‘Ok’
when finished.

Figure 4-127. Show Model Options window.

A new option will now appear under +FIT Options section of the Model panel
(Figure 4-128).

4-176 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-128. “Transmission Data % Weight” now available under +FIT Options.

Interference Enhancement
Another approach to thin absorbing films is Interference Enhancement. This
approach requires that the thin absorbing film is deposited over a dielectric layer on
a substrate (thermal oxide on silicon works well) and that multiple angles are
measured. Interference enhancement increases information in the data by varying
the pathlength for different incident angles. This approach works best if the
dielectric layer is around 500nm or more. The advantage of this method over SE+T
is that it only requires one data set.
Open “NiFe on SiO2 on Si” from the Examples folder. Clear the Model panel by
clicking ‘Clear’ and build a starting model with the following details:
• Add “Si_JAW.mat” as the model substrate.
• Add “SiO2_JAW.mat” as Layer #1.
• Add “B-Spline.mat” as Layer #2.
• Expand the B-Spline layer +Nodes section and enter “NiFe.mat” as the
starting mat.
• Turn on Thickness #1 and Thickness #2 as fit parameters.
• Enter 500nm as the starting thickness value for Layer #1, as this is known
or measured prior to depositing the NiFe layer.
• Turn on Global Fit for Thickness #2 from 1-1000nm with 100 guesses.

NOTE: Use Global Fit instead of Thickness Pre-Fit for this example as Thickness Pre-
Fit only operates on the thickest layer in the model. In this case, we know the
thickness of the thickest layer (Layer #1) and want CompleteEASE to find the
thickness of Layer #2.

The model should now look like Figure 4-129:

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-177

 Figure 4-129. Starting model for this example.

NOTE: Depending on the measured material and spectral range, optimization of

the B-Spline layer node resolution or other features may be required. For this
example, no further adjustments are required.

Click ‘Fit’ and the result in Figure 4-130 should appear.

4-178 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

 Figure 4-130. Result for NiFe on SiO2 on Si.

Evaluation of fit result: MSE is low and there is good agreement between model-
generated and experimental data. Optical constants for NiFe layer (not shown
here) agree with expected values. Thickness #2 is 32.62nm and Thickness #1 is
488.04nm. Are these values unique?

Perform a uniqueness test for each layer thickness (see page 4-129 for introductory
details). For this example, perform a 2D uniqueness test using the settings shown in
Figure 4-131.

Figure 4-131. Suggested settings for 2D uniqueness test for this example.

CompleteEASE Software Manual Data Analysis 2 – Intermediate • 4-179

 CompleteEASE will generate a 2D plot based on MSE results of the 2D uniqueness
test (see Figure 4-132). Results indicate good sensitivity to both thicknesses.

Figure 4-132. 2D uniqueness test result.

NOTE: Good sensitivity to the NiFe thickness is a direct result of the interference
enhancement technique and due to the underlying SiO2 layer. The same NiFe film
deposited directly on Si would have strong correlations between thickness and
optical properties and, as a result, a worse parameter uniqueness curve.

4-180 • Data Analysis 2 – Intermediate CompleteEASE Software Manual

5. Data Analysis 3 – Advanced

This chapter includes further examples to demonstrate advanced and specialized

topics. Advanced topics include multi-sample analysis, creation of composition- and
temperature-dependent material files, and working with anisotropic or non-ideal
models. A few of the examples use data from mapped samples, which help
demonstrate how to work with data collected at multiple points on a sample.

NOTE: Not all ellipsometers have mapping capability. If your system does not have
a translation stage, some details specific to analysis of map data will not be
applicable. Other concepts are still more generally applicable, so it is still useful to
work through the exercises.
The sections from this chapter are listed below, along with the primary
CompleteEASE features that are discussed.

Section 5.1 Map Data and Model Non-Idealities

FEATURES IN THIS EXAMPLE

• Showing map data • Fitting scan data
• Selecting single points from map • Viewing map results
• Sellmeier layer • Model calculations
(non-idealities)
• Thickness non-uniformity • Bandwidth (nm)
• Setting ranges

Section 5.2 Multi Sample Analysis

FEATURES IN THIS EXAMPLE

• Selecting multiple points from a map • Selecting/un-selecting data sets
• Switching view (CTRL+space bar) • Zooming in/out on map
• Setting symbol size • Clearing multi-data set mode
• Graphing all data sets • Adding fit parameters
• Multi sample analysis • Turning off all fit parameters

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-181

 Section 5.3 Creating Optical Constants Library for Composition- or
Temperature-Dependent Materials

FEATURES IN THIS EXAMPLE

• Creating composition- or • OC library mode
temperature-dependent material
files
• Opt. Const. Compare.mod • Building a library
• Drawing optical constants
comparison graph

Section 5.4 Anisotropic Films

FEATURES IN THIS EXAMPLE

• Converting to anisotropic • Difference mode
• Graphing anisotropic optical • Biaxial layer
constants
• Graphing anisotropic differences

5-182 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 5.1. Map Data and Model Non-Idealities
FEATURES IN THIS EXAMPLE
• Showing map data • Fitting scan data
• Selecting single points from map • Viewing map results
• Sellmeier layer • Model calculations
(non-idealities)
• Thickness non-uniformity • Bandwidth (nm)
• Setting ranges

In this example, a silicon wafer with thick SiO2 coating was measured at many
locations across the wafer. Our goal is to determine the SiO2 thickness and
refractive index, along with the uniformity of these properties across the wafer.

Show Map Data

Open the “SiO2_Map” example data file. This map was created from a radial scan
of the wafer. You should see a 2D plot of Psi data across the center of a circular
sample. The Psi data in this view are plotted using a single wavelength. The
wavelength can be specified using Data > ‘Set Ranges’ (see Figure 5-1).

Figure 5-1. Select Data Range dialog. Use "Graph Wvl." box to specify wavelength for map
graphs.

Right-click in the graph and make sure that “Show Symbols” is checked. This will
mark the individual measurement locations, as shown in Figure 5-2.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-183

 Figure 5-2. Psi values for “SiO2_Map” data file displayed at 756.8nm.

As with previous examples, the data must be modeled to determine properties of

interest. The best way to construct a model is to start at a single location. Notice
the “Show Map Data” checkbox above the graph. Uncheck this box to display
spectroscopic data at a single location.

Selecting a Point from the Map

Notice the X and Y coordinates in the graph title. You can use the “Position” slider
bar to change the selected location (see Figure 5-3).

Figure 5-3. Spectroscopic data from a single point of the map. This view was accessed by
left-clicking the mouse on a map point.

You can also select a single location from the map view by simply clicking on any
point. Practice selecting different locations by using the “Position” slider bar and by
clicking on individual points within the map. (Remember to check “Show Map

5-184 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Data” to access map view.) Before proceeding, choose the center location (X=0,
Y=0) and view the spectroscopic data at that point.

Si with Transparent Film

The film is SiO2 but not necessarily thermal oxide. Therefore, we will try the “Si
with Transparent Film” model that was first described in Section 3.3 for an
unknown transparent layer. Go ahead and open this model (Model > ‘Open’ >
Library tab > Basic folder) and click Fit > ‘Fit’. With a single location selected, the
‘Fit’ button applies the model to the selected location only (see Figure 5-4).

Figure 5-4. Fit result for "Si with Transparent Film" model at single location of "SiO2_Map"
example data.

Evaluation of fit result: MSE is high (>200) but visual fit appears okay. Check optical
constants.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-185

 Unphysical result

Figure 5-5. Resulting index curve for "Si with Transparent Film" model at center location of
“SiO2_Map” example data.

Evaluation of fit result: Optical constants are not physically plausible as the index
should only increase towards UV wavelengths when k=0.

Thicker films have more interference oscillations, and it can be difficult to visually
confirm the fit quality at short wavelengths when displayed in units of nm.

Wavelength Units – eV
Shortcut: Use Interference oscillations are evenly spaced in units of photon energy (eV) rather than
CTRL+ALT+W to toggle wavelength (nm). Therefore, when working with thicker films, it is often convenient
wavelength units. to view the data in units of eV rather than nm. You can make this change from the
Toggle options depend Options tab. Go to Options > Display Units and select “eV” from the drop-down
on loaded data. menu next to “Wavelength Units” (see Figure 5-6).

Figure 5-6. Display Units panel in Options tab.

The units of nm and eV are inversely related. Therefore, shorter nm values

correspond to larger eV values. When viewing versus eV, the shortest UV
wavelengths are on the right side of the graph (see Figure 5-7).

5-186 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-7. Data shown versus eV instead of nm. Short UV wavelengths are now on the right
(high photon energies).

With the data displayed versus eV, the interference oscillations are evenly spaced,
and it is easier to see that the model is not fitting the data well in the UV. So, what
happened? We know this is a SiO2 film and should be transparent with n≈1.45 over
this spectral range.

Fixing the Dispersion

From Figure 5-7, we can see that the Cauchy parameters appear to fit the data well
until about 4.5eV. Above this range, the dispersion model is not correct and cannot
match the interference oscillations. To fix this, we could try to redo the fit with a
global search for the B and C parameters, but this could result in long computation
times. We can also range-select the region that fits well and slowly add more
wavelength range. The second approach will be demonstrated here.
First, turn off the Global Fit and Thickness Pre-Fit, as they are no longer needed.
Next, select data up to ~4.5eV using Data > ‘Set Ranges’ (see Figure 5-8).

Figure 5-8. Select Data Range up to 4.5eV.

Click ‘Fit’. The Cauchy parameters will adjust to better match this range. The result
is shown in Figure 5-9.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-187

 Figure 5-9. Result for data up to 4.5eV only.

Extend the data range to 5.0eV and refit. Keep adding additional data in 0.5eV
increments and fitting the data until all data are included (up to ~6.5eV). The result
is shown in Figure 5-10 and Figure 5-11.

NOTE: After fitting the data up to 4.5eV, you can perform the additional steps
automatically with the Wvl. Range Expansion Fit command found in +OTHER
Options. Click this command to add and fit small increments of data, up to the full
acquired range.

5-188 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-10. Final fit to all wavelengths using Cauchy dispersion.

Figure 5-11. Final Cauchy optical constants in eV.

The MSE is significantly improved (~40) compared to the previous MSE (>200).

Evaluation of fit result: The MSE, visual fit in the UV, and resulting optical constants
are all significantly improved. Index result is physically plausible with index only
increasing toward UV, but MSE is still ~40. Can it be further improved?

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-189

 Sellmeier Dispersion
The Cauchy equation works well to describe the index of refraction in a transparent
region, and its simplicity and convenience make it the most widely utilized model
for transparent materials, but it does have some limitations. When the spectral
range extends beyond the visible into the near-infrared (NIR), the refractive index
of many transparent materials trends downward. The Cauchy equation, which
flattens at long wavelengths, cannot fit the NIR index correctly.
A downward trend into the NIR is better described by a summation of Sellmeier
oscillators, which are zero-broadened Lorentz oscillators. To model a transparent
region, typically two Sellmeier oscillators are used: one in the UV and one in the IR.
Both Sellmeier oscillators function outside the analyzed range and effectively
account for absorptions outside the analyzed range.
Replace the Cauchy layer in your model by clicking the “Cauchy Film” layer name.
From the Library tab, Dielectric folder, choose “SiO2 (Sellmeier)” and click ‘Open’
(see Figure 5-12).

Figure 5-12. Location of SiO2 (Sellmeier).mat.

Figure 5-13. Model with “SiO2 (Sellmeier)” instead of "Cauchy Film".

The “SiO2 (Sellmeier)” material file is comprised of a two-term Sellmeier equation

with parameters values pre-loaded to represent SiO2 (see Figure 5-13). Click
‘Generate’ and the model-generated curves should be close to the experimental
data, due to the pre-loaded parameter values, as in Figure 5-14.

5-190 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-14. Generated data with the Sellmeier model for SiO2 (before fitting).

Activate all four Sellmeier parameters as fit parameters and click ‘Fit’. Results are
shown in Figure 5-15.

Figure 5-15. Result using Sellmeier model.

Evaluation of fit result: Sellmeier result better matches the SiO2 index dispersion
into the NIR. The MSE is now < 30. Can it be further improved?

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-191

 Model Calculation
In Figure 5-15, Psi and Delta are fit reasonably well. However, there is still some
visual discrepancy in the interference peaks at low wavelengths. It’s time to
consider depolarization data. Depolarization is caused by non-idealities in the
sample or measurement.
Shortcut: Use CTRL+Z Expand the +Data menu above the graph and select “Depolarization”. The increasing
to view depolarization depolarization toward the UV shown in Figure 5-16 is common for thick layers. The
data. cause of this depolarization is due to non-ideality in the measurement. The most
common non-idealities for thick layers are thickness non-uniformity within the
measurement beam or instrument bandwidth. Both can be included in the model.

Figure 5-16. Depolarization data for the thick SiO2 film.

Model Calculation – Thickness Non-Uniformity

Expand the +MODEL Options section in the Model panel and change the “Model
Calculation” from “Ideal” to “Include Thickness Non-uniformity”, as shown in Figure
5-17. This will calculate the effect on the data if there was a thickness variation
within the measurement spot.

Figure 5-17. Model Calculation options.

5-192 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-18. Model Calculation changed to "Include Thickness Non-uniformity".

Now, position your mouse on the % Thickness Non-uniformity value (0.00) (see
Figure 5-18). Hold the SHIFT key down and roll the mouse wheel until the model-
generated depolarization data approximately matches the measured depolarization
data, somewhere around 2.4% (see Figure 5-19).

Figure 5-19. Depolarization data with 2.4% Thickness Non-uniformity in model calculation.

Activate % Thickness Non-uniformity as a fit parameter and click ‘Fit’ again. Results
are shown in Figure 5-20.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-193

 Figure 5-20. Depolarization after adding % Thickness Non-uniformity as a fit parameter.

Evaluation of fit result: The MSE is now < 20. Can it be further improved?

Model Calculation – Bandwidth (nm)

Next, try changing the “Model Calculation” from “Include Thickness Non-
uniformity” to “Include Bandwidth (nm)” (see Figure 5-21)

Figure 5-21. Model Calculation options.

5-194 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Again, position your mouse on the Bandwidth (nm) value (0.00) (see Figure 5-22).
Hold the SHIFT key down and roll the mouse wheel until the model-generated
depolarization data approximately matches the measured depolarization data,
around 4.000 (see Figure 5-23), which is physically true for the M-2000 used for
these measurements.

Figure 5-22. Model Calculation changed to "Include Bandwidth (nm)".

Figure 5-23. Depolarization data with Bandwidth (nm) = 4.000.

Activate Bandwidth as a fit parameter and click ‘Fit’ again. Results are shown in
Figure 5-24.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-195

 Figure 5-24. Depolarization after adding Bandwidth (nm) as a fit parameter.

Evaluation of fit result: The MSE is now < 18, and depolarization data is matched
better than with Thickness Non-uniformity.

Include Depolarization Data

Thus far, we changed the Model Calculation to better match the depolarization
data, but the depolarization data were not actually included in the MSE calculation.
Thus, we have shown that the non-idealities will also better match the shapes of Psi
and Delta (as the MSE was reduced). Now, we include the depolarization data in
the MSE calculation. To do this, click the Configure Options command at the
bottom of the Model (see Figure 5-25).

Figure 5-25. Configure Options command.

5-196 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Scroll to the Fit Options section, then check “Include Depolarization Data”, as
shown in Figure 5-26. Click ‘Ok’ to add this as an option within the +FIT Options
section.

Figure 5-26. Click Configure Options to select hidden option “Include Depolarization Data”.

Now that the option has been made available in the model, we can turn this feature
ON. Expand the +FIT Options section and turn “Include Depolarization Data” to ON
(see Figure 5-27).

Figure 5-27. Model after adding “Include Depolarization Data” to +FIT Options and turning it
ON.

With “Include Depolarization Data” ON, click ‘Fit’ again. The MSE is further reduced
while the bandwidth stays close to its previous value (see Figure 5-28).

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-197

 Figure 5-28. Final result for center location of SiO2_Map data.

Fit Scan Data

Now that we have optimized the model for the center point from the map, we are
ready to fit data at all locations. Recall that the initial model utilized Thickness Pre-
Fit and a Global Fit. If the wafer is relatively uniform across the sample, with only
small variations expected or observed in the raw data, then the results from the
initial location are likely good starting values for all other locations. Therefore, we
can turn off the Thickness Pre-Fit and Global Fit for faster results.

NOTE: When there is significant variation across a sample, it may be necessary to

use Thickness Pre-Fit and/or Global Fit to find the best fit at all locations. In such
cases, the time required to fit all locations will be longer.

5-198 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-29. Final model for SiO2_Map example.

Using the final model as shown in Figure 5-29, click Fit > ‘Fit Scan Data’. This will
apply the model to all locations on the map. A status bar will display the progress
as CompleteEASE calculates the fit at each point (see Figure 5-30).

Figure 5-30. A status bar displayed while CompleteEASE fits each point.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-199

 When the fit is complete, you can use the “Position” slider bar above the graph to
view the fit quality at each location (see Figure 5-31).

Figure 5-31. User slider bar above graph to view fit quality at each location.

Notice the parameters listed in the Fit panel are now blue and underlined. Click on
any of these parameters to view the mapped result, such as Thickness #1, as shown
in Figure 5-32.

5-200 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-32. Final result at all locations for “SiO2_Map” example data. Thickness #1 is
displayed.

Map graphs can be copied to clipboard and pasted as shown in the figures below.

Figure 5-33. Maps of MSE and [email protected] for "SiO2_Map" example data.

Evaluation of fit result: MSE is similar at all locations, indicating a fit quality similar
to the center location which was evaluated independently.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-201

 5.2. Multi Sample Analysis
FEATURES IN THIS EXAMPLE
• Selecting multiple points from a map • Selecting/un-selecting data sets
• Switching view (CTRL+space bar) • Zooming in/out on map
• Setting symbol size • Clearing multi-data set mode
• Graphing all data sets • Adding fit parameters
• Multi sample analysis • Turning off all fit parameters

Once again, we will be working with map data from multiple locations across a
sample. This sample is more complex than the previous example. In addition to an
SiO2 coating across the wafer, there are 5 patterned locations on the substrate with
an extra layer of Cr over the SiO2 film. The Cr layers were deposited to nominally
10, 15, 20, 25, and 40nm thick. These layers are thin enough to allow light to
penetrate through to the underlying SiO2 film. They were part of a study of
different ellipsometric methods to measure thin absorbing layers.3 Figure 5-34
shows the location of each of the Cr regions along with their nominal thicknesses in
nanometers.

25

20 40 10

15

Figure 5-34. Overview of sample. The blue region represents a 100nm SiO2 layer. An
additional Cr layer is present at the five orange locations, with the nominal thicknesses in nm
shown for each.

3
J.N. Hilfiker et al. “Survey of methods to characterize thin absorbing films with Spectroscopic Ellipsometry”, Thin Solid
Films, 516 (2008) pp. 7979-7989.

5-202 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Shortcuts for Mapping Data
Before proceeding with this example, it is helpful to review the shortcuts when
dealing with mapping data. The table below shows the important features we will
use for this example.

NOTE: You can also right-click on any location to Unselect, Unselect All, and Delete.
Table 5-1. Shortcut reference for mapping data.

SHORTCUT FUNCTION
CTRL+Click on Point Select/Un-select Point
CTRL+ALT+Click on Point Delete Point
CTRL+ALT+SHIFT+Click on Point Show camera image from point (when
available)
CTRL+space bar Switch view from single point to Map

Blank Model
Open the “Cr on SiO2 MAP” data file. Data from a translation scan will be displayed
as shown in Figure 5-35. Notice the low Psi values (red/yellow) at the 5 locations
coated with Cr. What model should we use to fit the data? In previous examples,
we have started with pre-built models available in the CompleteEASE library. We
can also start from a blank model that has no values or settings pre-determined.

Figure 5-35. Translation scan graph for “Cr on SiO2 MAP”.

Click Model > ‘Clear’ to clear any previous models. Then, click on the substrate
layer name (“none”) (see Figure 5-36). Choose “Si_JAW.mat” from the Library tab,
Semiconductor folder (see Figure 5-37). Next, click the layer command Add to add
a layer above the substrate. Choose Cauchy.mat from the Library tab, Basic folder
to model the SiO2 layer. Expand the Cauchy layer and right-click on the A, B, and C
values to activate as fit parameters. Make Thickness #1 a fit parameter as well.
Expand the +FIT Options section and turn Thickness Pre-Fit ON. This model should
be adequate to fit any of the areas that do not have a Cr layer (see Figure 5-39).

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-203

 Figure 5-36. Starting from “Blank” model.

Figure 5-37. Location of Si_JAW.mat.

Figure 5-38. Location of Cauchy.mat.

Figure 5-39. Model for regions without Cr coating.

5-204 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Set Symbol Size
Before continuing, let’s increase the symbol size to better visualize the locations of
the measured points. Right-click on the graph panel and make sure “Show
Symbols” is checked. From the same right-click menu, choose “Set Symbol Size” as
shown in Figure 5-40. Change the symbol size to 3. The graph shown in Figure 5-41
has symbol size set to 3.

Figure 5-40. From right-click menu on graph, select the “Set Symbol Size” option.

Figure 5-41. Figure with Symbol Size set to 3.

Zooming on Map
Sometimes it is helpful to zoom in on a region of mapped data, for example, to
better isolate a specific point. To zoom in, click and drag your mouse across a
region of the sample, as shown in Figure 5-42. The graph will update to display only
the selected region, as shown in Figure 5-43. To zoom out to all points, simply
right-click in the graph area and choose “Zoom All” or click “Zoom All” from the
upper-right corner of the graph area.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-205

 Figure 5-42. Click and drag mouse over a graph area to zoom in to that region.

Figure 5-43. Zoomed area of the map.

Selecting Multiple Points on a Map

With our model ready to fit the SiO2 locations, go ahead and click on any of the
points away from the Cr patterns (blue, purple, and dark green locations). If Auto
Fit is enabled (default), the data from will be opened and fit automatically, as
shown in Figure 5-44. If Auto Fit is not enabled, you may need to click the ‘Fit’
button. (Auto Fit will be discussed more later in this example.)

5-206 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-44. Cauchy fit to a single point with only SiO2 film.

NOTE: Remember that you can change the scale of the X-axis using CTRL+ALT+W if
needed to match your graphs to this manual.

Check “Show Map Data” to return to the map view. Let’s select data from multiple
points – all of which only contain the SiO2 film – to be graphed and fit at the same
time. To select multiple points, hold the CTRL button down and click on the points
you want to select or use the right-click menu associated with each point. A grey
border will be shown around each data point that is selected, as shown in Figure
5-45.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-207

 Figure 5-45. Four points in the upper-left quadrant are selected. Notice the gray borders
surrounding the symbols at the selected locations.

Switching Graph Views

To toggle the graph from map view to spectroscopic view, you can either check/un-
check the “Show Map Data” checkbox, or simply hold down the CTRL button and
press the space bar.
From the spectroscopic view, notice the additional menu options for multiple data
sets. Because we selected multiple locations in the map view, these selected data
points are numbered sequentially and available for multi-data set viewing and
analysis.

Figure 5-46. When multiple data points are selected in the map view, they are also available
simultaneously in the spectroscopic view.

Fitting Multiple Data Sets

With multiple data sets selected, click ‘Fit’. The MSE increases from ~5 (see Figure
5-44) to ~22 (see Figure 5-47). This is because the model is trying to fit all four
locations with the same values.

5-208 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-47. Fit result with four locations selected.

Graph All Data Sets

Most likely, there is some variation between the four selected points. This is easier
to visualize by graphing all data sets at the same time. Before proceeding, simplify
the graph by turning off the Double Y axis (expand +Data menu or press CTRL+Y)
and selecting a 65° angle of incidence only (from ‘Set Ranges’ dialog, Selected
Angles panel, as shown in Figure 5-48).

Figure 5-48. Choose ‘Set Ranges’ and select only the 65° data from the Selected Angles
panel.

Now, check the “Graph All Data Sets” box above the graph to display Psi curves
from all four locations (see Figure 5-49).

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-209

 Figure 5-49. Graph of all four Psi curves from multiple data point selection.

Selecting and Un-selecting Data Sets

Expand the +Data Sets menu above the graph (see Figure 5-50). Here you will find
options for selecting and un-selecting data sets. A good test of this menu is to
select an individual data set. To do so, expand the +Data Sets menu and choose
“Select Current Data Set ONLY”.

Figure 5-50. Expand +Data Sets menu above graph and choose “Select Current Data Set
ONLY”.

This will exclude all other data points from a fit (see Figure 5-51).

Figure 5-51. Drop-down list of points shows which are not selected and therefore not active
in the fit.

Click ‘Fit’ and the result will be for the selected data set only (see Figure 5-52).

5-210 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-52. Result for single data set selected.

Before proceeding, select each data set individually and click ‘Fit’ to find the
thickness at each of the four selected locations. When finished, expand the +Data
Sets menu and choose “Select ALL Data Sets” before moving on to the next section.

NOTE: The “Graph All Data Sets” checkbox only applies to selected data. Initially,
all data sets are selected by default. If you limit the selected data sets, the un-
selected data sets will not be graphed. Expand the +Data Sets menu and choose
“Select ALL Data Sets” as needed to graph or fit all data sets.

Multi Sample Analysis (MSA)

So far, we have fit the four selected points as a combined group and each
individually. In some cases, it is helpful or necessary to fit multiple points with
some common parameters and some varied parameters. In this example, we
assume the optical properties are the same at all points, but thickness may vary
across the sample. We will demonstrate this first with the SiO2 locations and then
apply to locations with the additional Cr coating where the thickness variation is
more significant.
To apply this method, we need to turn on “Multi Sample Analysis” (MSA). Click the
Configure Options command at the bottom of the model. Under “Model Options”,
check “Multi Sample Analysis”. Then click ‘Ok’ (see Figure 5-53).

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-211

 Figure 5-53. Check the “Multi Sample Analysis” option from Model Options section.

Now, when you expand the +MODEL Options section, you will find a +Multi Sample
Analysis option available (see Figure 5-54). Expand +Multi Sample Analysis and
you will see a column containing all selected data sets. In this example, we have
four (as selected previously, see Figure 5-45).

Figure 5-54. +MODEL Options configured for Multi Sample Analysis.

Click Add Fit Parameter and select a fit parameter to vary between data sets from
the drop-down list. For this demonstration, choose “Thickness # 1”, which is the
SiO2 film thickness (see Figure 5-55).

5-212 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-55. Choose a fit parameter to vary between data sets.

The thickness for each data set will appear as an individual parameter in a new
column next to the numbered data sets (see Figure 5-56). Here, we can vary and fit
the thicknesses individually while the optical properties (as defined by the Cauchy
layer in this example) remain common for all data sets.

Figure 5-56. MSA section with Thickness #1 added as an MSA parameter.

With the Thickness #1 parameter added in the MSA section, click ‘Fit’. The result is
shown in Figure 5-57. Notice that each data set has its own thickness value and
model-generated curve in the graph.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-213

 Figure 5-57. Results for Multi Sample Analysis of four locations of SiO2 on Si. Optical
properties are the same for all data sets while SiO2 thickness varies.

Clear Multi-Data Set Mode

Now that we have determined the optical properties of the SiO2 layer, we are ready
to turn our attention to the Cr coated areas. Before we proceed, we want to fix the
fit parameters for the SiO2 film – including the thickness and Cauchy values – so
they do not change from the newly determined values. The fastest way to do this is
to click Turn Off All Fit Parameters at the bottom of the Model panel.
To clear the current multi-data set mode (four locations of SiO2 only) and return to
the original map data file, click +Data Sets and choose “Clear Multi-Data Set Mode”
(Figure 5-58).

Figure 5-58. +Data Sets menu with "Clear Multi-Data Set Mode" selected.

Use the “Show Map Data” checkbox (or CTRL+space bar) to view map data. Before
we select the Cr coated areas, we want to disable automated fitting. To do so,
right-click in the Fit panel and uncheck “Auto Fit” (see Figure 5-59).

5-214 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-59. Right-click in the Fit panel and turn off the “Auto Fit” checkbox.

NOTE: Turning off Auto Fit ensures that unintentional clicks do not result in model
changes.
Now, click the center point in the graph. For the Cr analysis, we need multiple
angles of incidence. Recall that we had selected only one angle in a previous step.
Click on ‘Set Ranges’, then ‘Select All Angles’, and ‘Ok’ to restore the additional
angles measured (see Figure 5-60). The data will appear as shown in Figure 5-61.

Figure 5-60. Click 'Select All Angles' to restore all measured angles.

Figure 5-61. Psi data from center location (X=0, Y=0) at all angles.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-215

 Shortcut: Use To fit this data, we need to add another layer to our model to represent the Cr layer.
CTRL+Add to Click the layer command Add, position the blue bar above “Layer #1 = Cauchy”, and
automatically add a click (see Figure 5-62).
layer to the top of the
model.

Figure 5-62. Add layer above “Layer #1 = Cauchy”.

Recall from Chapter 4 that metals can be described by a B-Spline layer. Choose the
B-Spline.mat as the new layer (see Figure 5-63).

Figure 5-63. B-Spline.mat is in the Library tab, Basic folder.

With Layer #2 added in the model, expand the B-Spline layer +Nodes section and
change the Starting Mat. From “none” to Cr.mat (see Figure 5-64). The B-Spline
nodes will update to match Cr optical constants as published in Palik’s Handbook of
Optical Constants of Solids.

Figure 5-64. Cr.mat is in the Library tab, Metal folder.

5-216 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-65. Model with second layer added and B-Spline Starting Mat. = Cr.

Hold the SHIFT key and roll the mouse wheel over the Thickness #2 value until the
model-generated data approximately matches the experimental data. You will find
that no thickness value matches the data exactly. This is because the optical
properties of the Cr layer on this sample are not quite the same as the optical
properties contained in Cr.mat. This is common for metals, which will differ
optically with different deposition methods and processing conditions. Set the
thickness to the nominal expected value of 40nm and click ‘Generate’ (see Figure
5-66).

Figure 5-66. Model-generated data with Thickness #2 set to the nominal expected value of
40nm.

Let’s use the expected thickness of 40nm to bring the optical constants closer to the
sample values. Leave the Cr thickness fixed at 40nm (do not activate as a fit
parameter) and click ‘Fit’. Results are shown in Figure 5-67.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-217

 Figure 5-67. Result for center location with Cr thickness fixed at 40nm.

Evaluation of fit result: The MSE is ~4, but there is a visual mismatch between the
model-generated data and experimental data around 360nm. This is an
interference enhancement from the underlying SiO2 film. We will only match this
region well with the correct Cr thickness.

Now that the optical constants of Cr in the model are closer to the measured data,
we can let the thickness adjust to a more accurate value. Add the Cr thickness as a
fit parameter and click ‘Fit’ again. The final result is shown in Figure 5-68.

5-218 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-68. Final fit to center location.

Evaluation of fit result: The MSE is ~2 and visual fit quality is good across the full
spectrum. B-Spline optical constants trend as expected for Cr, absorbing
everywhere and increasing into the NIR (see Figure 5-69). Thickness appears
unique over a wide range (see Figure 5-70).

Figure 5-69. B-Spline optical properties compared to Palik’s optical properties for Cr.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-219

 Figure 5-70. Uniqness test for Thickness #2 from 20-80nm.

Multi Sample Analysis (revisited)

Now that we have developed a model to match the center point, let’s try this model
for all points that are coated with Cr. Use the “Show Map Data” checkbox or
CTRL+space bar to return to map view. CTRL+click on each Cr-coated location,
using Figure 5-34 and the Psi color variations as guides. The selected locations will
add a gray border around the map symbol (see Figure 5-71).

Figure 5-71. Map view with five Cr-coated locations selected.

Next, switch to the Spectroscopic View (“Show Map Data” checkbox or CTRL+space
bar). Turn off the double Y axis and set ranges to only 65 degrees. This will allow a
more convenient comparison of the data sets, as shown in Figure 5-72. From this
data alone, we can estimate the points with thickest-to-thinnest Cr layers based on
the shifts in data. The thickest point was at center (in purple). As the data shifts,
the Cr film gets thinner.

5-220 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 NOTE: The order of points will differ depending on the order they were selected.
Thus, your colors may not match the same order as Figure 5-72.

Figure 5-72. Spectroscopic view comparing Psi curves from each of the five selected locations
(65° data only).

Now, let’s prepare the Multi Sample Analysis. Expand the +MODEL Options, then
+Multi Sample Analysis section. If any parameters are leftover from the previous
analysis, click Delete All Parms. Next, click Add Fit Parameter, and choose Thickness
#2. If done correctly, the MSA section will now appear as shown in Figure 5-73.

Figure 5-73. Multi Sample Analysis section with Thickness #2 added.

NOTE: The thickness values are based on the most recent fit result. If your value is
not ~47nm, manually change each thickness in the MSA section before proceeding.
Alternatively, you can right-click the “Thickness #2” column header and select “Set
Parm Values” (see Figure 5-74). Enter 47nm in the subsequent dialog.

Figure 5-74. Right-click the “Thickness #2” column header and select “Set Parm Values”.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-221

 Before we begin fitting, we need to adjust the starting thicknesses of each data set.
For this purpose, uncheck “Graph All Data Sets” and display data set #1. Then, in
Model > +MODEL Options > +Multi Sample Analysis, SHIFT+scroll your mouse
wheel over the Thickness #2 parameter for data set #1 until the model-generated
curve approximately matches the experimental curve. You can use
CTRL+SHIFT+scroll for smaller increments.

NOTE: Make sure you are adjusting the thickness parameter that corresponds to
the selected data set displayed in the graph.
After you have the thickness for data set #1 adjusted, graph data set #2 and adjust
the corresponding thickness parameter. Repeat until the thicknesses for data sets
#1-5 are completed (see Figure 5-75).

Figure 5-75. All five thickness values in the MSA section of the model have been adjusted to
approximately match the experimental data for each data set. Data Set #1 is displayed in the
graph.

After the thicknesses for each data set have good starting values, check “Graph All
Data Sets” again. Your graph should look similar to Figure 5-76. If the model-
generated curves are approximate to the experimental curves, you are ready to fit
the data. Click ‘Fit’. All five data points will be fit simultaneously with a single B-
Spline layer to describe the common Cr optical constants and individual thicknesses
for the Cr layers. The fit result should appear as shown in Figure 5-77.

5-222 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-76. Model-generated data and experimental data for all five data sets with
thicknesses for each data set adjusted for better starting values.

Figure 5-77. Fit result for all five Cr locations with Multi Sample Analysis.

Evaluation of fit result: The MSE is ~18 and visual fit quality is okay for most of the
data but the model-generated data visually differs from the experimental data
below 300nm and for data set #3.

Metal optical constants can vary with thickness due to microstructure. Data set #3
is the thickest Cr location. Let’s try excluding this location from the multi sample
analysis and see if a better fit to the remaining data sets is achieved.
To do exclude the thickest Cr location, choose the data set with (0,0) coordinates
from the drop-down list above the graph. Expand the +Data Sets menu and choose
“Un-Select Current Data Set” (see Figure 5-78). The curve for the center point
should disappear – leaving only four curves and their corresponding models. Click
‘Fit’ again and the MSE drops from ~18 to <10 (see Figure 5-79).

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-223

 Figure 5-78. With the (0,0) location selected, expand the +Data Sets menu and choose "Un-
Select Current Data Set".

NOTE: It is important to choose “Un-Select Current Data Set” rather than “Delete
Current Data Set”. If “Delete Current Data Set” is chosen, the mutli sample analysis
renumbers the remaining data sets, which would require re-establishing the
starting thickness value for each data set.

Figure 5-79. Result excluding the center (thickest) data set.

Evaluation of fit result: Excluding the thickest Cr location does in fact yield a better
MSE (~10) and fit quality is improved for the remaining data sets below 300nm.

Metals also typically have roughness and/or oxidation at the surface. To test this,
add surface roughness to your model (toggle “Include Surface Roughness = ON”)
and add it as another fit parameter in the +Multi Sample Analysis list (click “Add Fit
Parameter”, then choose “Roughness”). This will allow each point to have different
amounts of roughness. Click ‘Fit’(see Figure 5-80).

5-224 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-80. Fit to four thin Cr films with roughness added to the Multi Sample Analysis.
Notice the resulting roughness increases as the Cr film thickness increases.

Evaluation of fit result: MSE drops from ~10 to ~6. Surface roughness increases
with the thickness of the film.

With surface roughness included in the model, try adding the thickest Cr point back
into the analysis. See if the model with surface roughness better matches the
center location and if the trend of roughness increasing with thickness is consistent.
To add the center location back to the multi sample analysis, expand the +Data Sets
menu and choose “Select ALL Data Sets” (see Figure 5-81). The graph will again
display all five data sets, and five values for Thickness #2 and Roughness will be
tabulated in the +Multi Sample Analysis section of the model.

Figure 5-81. +Data Sets menu with "Select ALL Data Sets" selected.

With all five data sets in the analysis, click ‘Fit’ again (see Figure 5-82).

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-225

 Figure 5-82. Fit results for all five points, showing increasing roughness with increasing Cr
layer thickness.

Evaluation of fit result: The MSE increases slightly from ~6 (for the same model
applied to the four thinnest data sets) to ~8 but is significantly improved over the
previous result without roughness applied to all five data sets (MSE ~18). .

At this point, research could continue by testing the samples with other techniques
– such as Atomic Force Microscopy – to see if the roughness result agrees for this
sample.
As a final exercise, add all angles back to the analysis. Click ‘Set Ranges’ > ‘Select All
Angles’ > ‘Ok’. Notice that when multiple multi-angle data sets are selected, each
data set is graphed with the same color. For example, all curves for data set #1 are
magenta and all curves for data set #2 are blue. Before fitting again, turn on the
SiO2 thickness as a fit parameter and add Thickness #1 to the list of Multi Sample
Analysis parameters. Click ‘Fit’. The final result is shown in Figure 5-83.

5-226 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-83. Final fit to all angles for five locations with Cr, SiO2, and Roughness thicknesses
included in the Multi Sample Analysis.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-227

 5.3. Creating Optical Constants Library for
Composition- or Temperature-Dependent
Materials
FEATURES IN THIS EXAMPLE
• Creating composition- or • OC library mode
temperature-dependent material
files
• Opt. Const. Compare.mod • Building a library
• Drawing optical constants
comparison graph

In this example, we will demonstrate how to build a composition- or temperature-

dependent optical constant library. This procedure can be used to create special
material files for any series of samples that show a systematic shift in optical
properties related to composition, temperature, or other material property or
process condition. During the process, B-Splines are used to represent the
constituent optical constant spectra for each known composition or temperature.
The library is then built using an interpolation algorithm based on the critical-point
shifting algorithm of Snyder et al.4 In the critical-point shifting algorithm, the
dielectric function at an arbitrary composition or temperature is evaluated from a
weighted sum of the nearest reference dielectric function spectra after the
reference spectra are wavelength-shifted to line up the critical-point features.
CompleteEASE simplifies this process by automatically determining the wavelength-
shifting polynomials (that describe the path of critical points versus
composition/temperature) via a nonlinear regression analysis.
The full procedure for creating a composition- or temperature-dependent material
file is to obtain optical constants from a series of samples with varying, known
composition or temperature. After successfully modeling the experimental data to
determine accurate optical constants for a specific composition/temperature, save
the tabulated optical constants for each material file. In this example, we work with
a series of Si1-xGex material files that were obtained by fitting a series of layers with
different germanium compositions (x). The actual composition was measured by
other means – in this case x-ray diffraction (XRD). Thus, we now have the optical
constants (n, k) for a series of known Si1-xGex compositions.

NOTE: Before starting this example, change the Wavelength Units to “eV” and the
Optical Constant Units to “e1 & e2” from Options > Display Units.

Opt. Const. Compare Model

To start this example, open the “Opt. Const. Compare” model from the Library tab,
Advanced folder (see Figure 5-84).

4
P.G. Snyder, J.A. Woollam, S.A. Alterovitz, and B. Johs, “Modeling AlxGa1-xAs Optical Constants as Functions of
Composition”, J. Appl. Phys. 68 11 (1990) 5925.

5-228 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-84. Location of "Opt. Const. Compare.mod".

The model is shown in Figure 5-85. Click ADD MATERIAL seven times to add seven
layers to the model. Each of these layers will be used to hold the optical constants
for a specific composition of Si1-xGex. There are 10 Si1-xGex material files in the
Examples folder that were measured from a series of thin films with known
composition (as determined from XRD). In addition, we can use crystalline Silicon
and Germanium as the endpoints for this material file. The compositions related to
each material file are provided in Table 5-2.

Figure 5-85. “Opt. Const. Compare” model.

Table 5-2. Compositions for each Si1-xGex material file used in this example.

MATERIAL FILE NAME GERMANIUM COMPOSITION % (X)

Si_jaw.mat 0% (x=0.000)
SiGe-1.mat 5.2% (x=0.052)

SiGe-2.mat 9.8% (x=0.088)

SiGe-3.mat 15.5% (x=0.155)
SiGe-4.mat 20.4% (x=0.204)
SiGe-5.mat 26.3% (x=0.263)

SiGe-6.mat 31.9% (x=0.319)

SiGe-7.mat 37.1% (x=0.371)
SiGe-8.mat 42.3% (x=0.423)

SiGe-9.mat 49.3% (x=0.493)

SiGe-10.mat 53.9% (x=0.539)
Ge.mat 100% (x=1.000)

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-229

 For this example, click on “none” for each material to populate the seven layers
with Silicon, SiGe-2, SiGe-4, SiGe-6, SiGe-8, SiGe-10, and Germanium (see Figure
5-86). The SiGe files are found in the Library tab, Examples folder. Si_JAW.mat and
Ge.mat are found in the Library tab, Semiconductor folder.

Figure 5-86. Model populated with specified material files.

Turn “OC Library Mode” ON to set the compositions for each layer. Change the “x”
value for each material to the corresponding composition from Table 5-2 for each
layer (see Figure 5-87). The temperatures can remain at 25° Celsius for all files.

NOTE: If we were making a temperature-dependent file, we would adjust the

corresponding temperatures to match each material file.

Figure 5-87. Model populated with known compositions.

Before proceeding to build the OC Library from these reference files, set the Wvl.
Range to match the material files. For the SiGe material files, the optical constants
were measured from 248nm to 1240nm (1eV to 5eV). Also, the “# of Wvls” may

5-230 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 need to be increased to allow the B-Spline to more accurately match the reference
files over this wide spectral range. The original material files contain about 540
individual wavelengths each. Thus, we can set the “# of Wvls” to 540.
When ready, click Build Library, and a graph will appear as shown in Figure 5-88. In
this graph, the colored curves are the reference optical constants at the specified
compositions, and the black dashed curves are interpolated spectra at reference
and midpoint compositions. Thus, each colored graph should be matched with a
corresponding dashed black curve. The additional dashed black curves should
demonstrate whether the critical-point shifting algorithm has worked.

Figure 5-88. Results after building OC Library for SiGe.

You can simplify the graph by changing the “Graph Type” in the Draw Graph
command at the bottom of the Model panel. The default view is e1 & e2. Click on
this value to view other options, including e1 only or e2 only (see Figure 5-89).

Figure 5-89. Graphing selections after building an OC Library.

Viewing the Shifted Spectra and Wvl Shift Function can help assess whether the
composition library is successful. For advanced users, the Library Resolution and
Spectra Shifting Parameters can be modified, as shown in Figure 5-90.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-231

 Figure 5-90. Advanced features to customize the OC Library algorithm, including spectral
shifting parameters and resolution in eV.

You can zoom in on each of the graphs to further assess whether the library
algorithm was successful. Figure 5-91 shows a zoomed view of the e2 peak for the
SiGe films. This shows that the peaks are shifted to higher energy with increasing
silicon composition.

Figure 5-91. Zoomed view of the e2 peak for the SiGe compositions, showing a nice shift in
critical point toward higher energy for increasing silicon composition.

Save OC Library
After successful calculation of a composition- or temperature-dependent optical
constant library, click the Save OC Library command at bottom of the Model panel
to save your library as a single material file. For this example, save it in the Projects
tab, Common folder as “SiGe comp.mat” (see Figure 5-92).

Figure 5-92. Save SiGe comp.mat in the Projects tab, Common folder.

The newly created “SiGe comp.mat” can now be used to fit SiGe data files. To
demonstrate what this file looks like and how it operates, load “SiGe comp.mat” as
the Ref. Material in the “Opt. Const. Compare” model. Then, expand the layer to
view the composition value (see Figure 5-93).

5-232 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-93. Opt. Const. Compare model with the new “SiGe comp" file loaded as the Ref.
Material.

To adjust the composition value of the “SiGe comp” layer, hold down SHIFT+roll
mouse wheel. You will be able to visualize the optical constants shifting as the
composition is varied. An example is shown in Figure 5-94, where the composition
value of the “SiGe comp” layer has adjusted to x=0.75 (black curve).

Figure 5-94. “SiGe comp” (Ref. Material) composition set to 0.75. Notice it is well-spaced
between x=1 (Ge) and x=0.539 (sige-10).

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-233

 5.4. Anisotropic Films
FEATURES IN THIS EXAMPLE
• Converting to anisotropic • Difference mode
• Graphing anisotropic optical • Biaxial layer
constants
• Graphing anisotropic differences

In this example, we will allow a layer to have anisotropic optical constants, where
the index normal to the sample surface is different than the index in the sample
surface plane. This is common for many polymer films that are oriented along a
specific direction.
To start, open the “Anisotropic Polymer on Si” data file from the Examples folder.
Psi values for this data set are shown in Figure 5-95. The oscillations show that this
film is transparent in this spectral range. Open the “Si with Transparent Film”
model and click ‘Fit’. The results of this fit should appear as seen in Figure 5-96.

Figure 5-95. Psi values from “Anisotropic Polymer on Si”.

5-234 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-96. Result using “Si with Transparent Film” model to fit “Anisotropic Polymer on Si”
example data set.

Evaluation of fit result: MSE is high (Out Of Spec.) and visual fit to the data is poor.

Upon closer inspection, the data at 75° shows a common “signature” of anisotropy
in the film. The interference oscillations from anisotropic films will often “tilt” such
that the curves are higher on one side of the peak and lower on the opposite side.
This is demonstrated for a 900nm anisotropic film on Si at 75° in Figure 5-97. The
red curve represents an isotropic film. The green curves are higher on the long
wavelength side of a peak and lower on the short wavelength side. These represent
anisotropic films with Nx < Nz. The blue curves are lower on the long wavelength
side of a peak and higher on the short wavelength side. These represent
anisotropic films with Nx > Nz.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-235

 Figure 5-97. Simulation for transparent films showing the effect of anisotropy with Nx > Nz
and Nx < Nz as compared to isotropic data.

Use the Try Alternate Models feature described in Section 3.5 to test for surface
roughness, grading, and simple anisotropy. Results are shown in Figure 5-98.

Figure 5-98. Results after using Try Alternate Models command.

MSE is significantly reduced (from >100 to <10) by using an anisotropic model.

From here we could click ‘Apply Chosen Model’, but sections below will describe
how to construct an anisotropic model in more detail. So instead, click the “x” in
the upper-right corner to exit the results and manually add anisotropy to the
model.

Convert to Anisotropic
Right-click the layer name (in this example, “Cauchy Film”) to access the layer
options menu. From this list, choose to “Convert to Anisotropic” (see Figure 5-99).
The Cauchy layer will now become embedded as part of a new layer named
“Biaxial”. Biaxial is a term for anisotropy where Nx, Ny, and Nz are all different.
However, the layer name “Biaxial” is used here to describe any anisotropic film.
Within the Biaxial layer, you can choose either Uniaxial or Biaxial anisotropy. For
this example, we will assume that the film is only uniaxially anisotropic.

5-236 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 NOTE: Biaxial anisotropy would require more advanced measurements.

Figure 5-99. Right-click the layer name and select “Convert to Anisotropic”.

Biaxial Layer
The biaxial layer can be used in two different ways, and both will work equally well
for this example. The choice is whether to have Difference Mode OFF or ON (see
Figure 5-100 and Figure 5-101).

Figure 5-100. Biaxial layer with Uniaxial anisotropy and Difference Mode OFF.

Figure 5-101. Biaxial layer with Uniaxial anisotropy and Difference Mode ON.

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-237

 With the Difference Mode OFF, each direction is described by an individual material
file. In our case, each direction is described by a Cauchy layer, as that is the layer
we started with. The constituent materials could just as easily be B-Spline, Gen-
Osc, or a combination of these or other material files.
With the Difference Mode ON, the xy-direction is defined by a material file while
the z- direction is determined by adding the x-direction optical constants with
“difference” values calculated from an extended Cauchy-type dispersion described
as dZ. The index difference equation is shown in Equation 5-1, where wvl is in µm.

Equation 5-1

𝐵 𝐶 𝐷
𝑛𝑧 − 𝑛𝑥𝑦 = 𝐴 ∗ (1 + 2
+ 4
+ − 𝐼𝑅 ∗ 𝑤𝑣𝑙 2 )
𝑤𝑣𝑙 𝑤𝑣𝑙 𝑤𝑣𝑙 6

This method can be extremely useful when the fit is primarily sensitive to the index
difference and less sensitive to the absolute value of the index.

NOTE: Difference Mode is only recommended for spectral ranges where the
material is transparent. If absorptions are present, it can produce incorrect results.

For this example, turn the Difference Mode ON. Use SHIFT+roll the mouse wheel
over the dZ_A parameter. If you roll this parameter to positive numbers, the
simulated curve will look worse than the isotropic fit (tilt in wrong direction).
However, if you roll the dZ_A parameter to negative values, it will start to match
better.
Turn on dZ_A and dZ_B as fit parameters. Click ‘Fit’. The result is shown in Figure
5-102.

5-238 • Data Analysis 3 – Advanced CompleteEASE Software Manual

 Figure 5-102. Anisotropic fit using Difference Mode ON.

Evaluation of fit result: MSE is significantly improved and visual fit to the data is
good. What are the optical constants of the film?

Right-click Biaxial and choose “Graph Layer Optical Constants” (see Figure 5-103).
This will show ordinary (Nx = Ny) and extraordinary (Nz) optical constants (see
Figure 5-104). You can also choose “Graph Layer Optical Constants (anisotropic
differences)” to display the difference between ordinary (Nx = Ny) and
extraordinary (Nz). The graph will display “o-e” as shown in Figure 5-105.

Figure 5-103. Right-click the layer name and choose "Graph Layer Optical Constants".

CompleteEASE Software Manual Data Analysis 3 – Advanced • 5-239

 Figure 5-104. "Biaxial" layer optical constants.

Figure 5-105. Anisotropic differences in n and k.

You can also view the optical constants of only a single direction by right-clicking on
the individual material file describing that direction (see Figure 5-106). For access
to all directions, you must fit with Difference Mode OFF.

Figure 5-106. When Difference Mode is OFF, you can right-click on the various material files
describing the directions in the Biaxial layer to view them separately.

5-240 • Data Analysis 3 – Advanced CompleteEASE Software Manual

6. Data Analysis 4 – In-situ Data

This chapter includes further examples pertaining to in-situ data analysis. The most
common analysis strategies for dynamic measurements use “Multi-Timeslice”
analysis, which is discussed in the first example. We also demonstrate the use of a
Growth Rate and Optical Constants model (GROC) that is very easy to apply when
the film is an absorbing film with constant growth rate and optical constants.
The sections from this chapter are listed below, along with the primary
CompleteEASE features that are discussed.

Section 6.1 In-situ Data Analysis

FEATURES IN THIS EXAMPLE

• Viewing individual timeslices • Dynamic fit
• Selecting multiple timeslices from a • Multiple timeslice analysis
dynamic scan

Section 6.2 Growth Rate and Optical Constants

FEATURES IN THIS EXAMPLE

• GROC model • Virtual interface

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-241

 6.1. In-situ Data Analysis
FEATURES IN THIS EXAMPLE
• Viewing individual timeslices • Dynamic fit
• Selecting multiple timeslices from a • Multiple timeslice analysis
dynamic scan

In-situ data is dynamic ellipsometric data measured over time. Typically, some thin
film process is occurring during this time. The model may need to account for
changes in film thickness, optical constants, composition, or even temperature
versus time. The navigation, manipulation, and analysis of in-situ data have many
parallels with the corresponding processes for uniformity maps and multi-sample
analysis.
First, navigating and selecting timeslices from in-situ data will be discussed. Second,
the data will be analyzed via a multi-sample-analysis approach. Finally, the data will
be analyzed via a time-dependent data-analysis procedure. Previous chapters
should be reviewed in detail before proceeding with this example.

Shortcuts for In-situ Data

The table below shows shortcuts we will use for this example.

Table 6-1. Shortcut reference for dynamic data.

SHORTCUT FUNCTION
Click on dynamic graph Opens spectroscopic data from that timeslice.
CTRL+click dynamic graph Selects spectroscopic data from that timeslice.
Repeat for multiple timeslices.
CTRL+space bar Switch view from dynamic to spectroscopic
data.

6-242 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 Navigating In-situ Data
Open “a-Si Grow Sim.iSE” data file from Library tab, Examples folder. The time-
dependent data is shown in the graph in Figure 6-1. Notice that five wavelengths
are graphed. These curves are not the only wavelengths measured but represent
the shape versus time for the spectral range selected.

Figure 6-1. Graph shows in-situ data versus time.

Viewing Spectroscopic Data at a Specific Time

Click in the graph to select data for a specific timeslice. A line is drawn at that
timeslice, as shown in Figure 6-2. Now, you can press CTRL+space bar or
check/uncheck the “Show Dynamic Data” box above the graph to toggle between
the spectroscopic graph for that timeslice and the dynamic graph. Data for a
timeslice selected near 3 minutes is shown in Figure 6-3.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-243

 Figure 6-2. Click on the graph near x = 3 minutes to select this time. The selected timeslice is
shown as a vertical bar.

Figure 6-3. To view the spectroscopic data at the selected timeslice, use CTRL+space bar or
uncheck the “Show Dynamic Data” checkbox above the graph.

6-244 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 Turning off the “Auto-Fit” function
Before we proceed, you will want to turn off the “Auto-Fit” function so that the
data are not automatically fit every time we select a timeslice or open a new model.
To do this, right-click in the Fit panel and uncheck “Auto Fit”, as shown in Figure
6-4.

Figure 6-4. Right-click in the Fit panel and turn off the “Auto-Fit” checkbox.

Building a Model
Now, let’s build a model to fit this timeslice. We’ll start with the assumption that
the sample coating has known optical constants.
First, load “BK7 Glass.mat” from the Library tab, Dielectric folder, into the model
substrate. Next, add a layer over the substrate and load “a-Si parameterized.mat”
from the Library tab, Semiconductor folder. Activate the a-Si thickness as a fit
parameter. When finished, the model should look like Figure 6-5.

Figure 6-5. Starting model for this example.

Click ‘Fit’ to fit only this selected timeslice. The result is shown in Figure 6-6.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-245

 Figure 6-6. Fit result for a timeslice near 3 minutes using library optical constants for a-Si and
only fitting for the a-Si thickness.

Evaluation of fit result: Visual fit to the data could be improved.

Use the Try Alternate Models command to see if any common complexities
significantly improve the model. Results are shown in Figure 6-7. MSE drops from
near 80 to under 2 with surface roughness. Click ‘Apply Chosen Model’ to update
the model to include surface roughness. The improved fit with surface roughness is
shown in Figure 6-8.

Figure 6-7. Results from the Try Alternate Models command.

6-246 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 NOTE: Amorphous silicon typically has a naturally forming oxide at the surface.
Optically, this oxide is similar to surface roughness. Both are lower-index surface
layers. Because they are optically similar, these effects are typically
indistinguishable. You can try adding a silicon oxide layer to the surface instead of
roughness and compare the results. In this case, surface roughness provides a
better fit to the data.

Figure 6-8. Fit result for time near 3 minutes after including the surface roughness.

Evaluation of fit result: MSE and visual fit to the data are significantly improved.
Model is physically meaningful as a-Si typically has surface oxide, roughness, or
both.

‘Fit Dynamic’
Now, we can fit these two parameters versus time to see how they change as the
data vary across the measured time span. Simply click the ‘Fit Dynamic’ button in
the Fit panel, and CompleteEASE will start at the first timeslice and fit through the
last timeslice, using the previous timeslice results as starting values for the next
timeslice. A status bar will appear while CompleteEASE is working as shown in
Figure 6-9. When finished, the spectroscopic view of the last result should appear
as shown in Figure 6-10.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-247

 Figure 6-9. Status bar during Dynamic Fit.

Figure 6-10. Spectroscopic view of final timeslice fit results.

6-248 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 If you only consider the final timeslice, you might consider the analysis complete.
However, check ‘Show Dynamic Data’ or click MSE or Roughness to view these
values versus time, and it is clear that the model did not work well below the ~3-
minute timeslice (see Figure 6-11, Figure 6-12, and Figure 6-13 below).

Figure 6-11. Dynamic view of the fit results for all data from 0-10 minutes.

Figure 6-12. MSE vs. Time.

Figure 6-13. Roughness vs. Time.

Evaluation of fit result: Model failed below ~3-minute timeslice.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-249

 The fit failed at the earlier timeslices because the starting values in the model
(determined from the ~3-minute timeslice) were not sufficiently close to the real
sample values at the beginning of the measurement.
To remedy this problem, start with values from an earlier timeslice. Select a
timeslice near 0.3 minutes. You can select the timeslice as before in the dynamic
view (see Figure 6-14 for placement) or use the slider bar above the graph in the
spectroscopic view (see Figure 6-15).

Figure 6-14. Timeslice selected near 0.3 minutes.

Figure 6-15. Use slider bar above graph in spectoscopic view to select timeslice.

With a timeslice near 0.3 minutes selected, roll the thickness value until the model-
generated curves approximately match the experimental curves. Then click ‘Fit’.
The result is shown in Figure 6-16.

6-250 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 Figure 6-16. Result for 0.3-minute timeslice.

NOTE: It may be easiest to turn off surface roughness, adjust the a-Si thickness, and
repeat the Try Alternate Models command.

After a good fit to the 0.3-minute timeslice is achieved, click ‘Fit Dynamic’ again.
Results are shown in Figure 6-17 through Figure 6-20.

Figure 6-17. Dynamic fit result for all timeslices.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-251

 Figure 6-18. MSE vs. Time.

Figure 6-19. Roughness vs. Time.

Figure 6-20. Thickness #1 vs. Time.

Evaluation of fit result: The model fits well for all timeslices. MSE is low for all
timeslices. Roughness is mostly constant for all timeslices, and a-Si growth is
constant during process.

This time, with better starting values for the beginning of the measurement, the
model worked well for all timeslices.

6-252 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 NOTE: Sometimes it is best to start the fit from the endpoint rather than the
starting time. To fit the data in reverse-time order, right-click the ‘Fit Dynamic’
button and select “Fit Backwards in Time”. In this case, the model starting values
should represent the end of the measurement rather than the beginning.

Selecting Multiple Timeslices

The previous example assumes known optical constants for the a-Si. Often, optical
constants are not known. One advantage of in-situ data is that it provides
information about the sample at different stages of growth. Multiple timeslices can
be fit simultaneously as a form of multi-sample analysis.
To analyze multiple timeslices simultaneously, we must first select multiple
timeslices. To do so, view the dynamic data, hold the CTRL key, and click at multiple
points on the graph. With each CTRL+click, a black, numbered, vertical bar will
appear in the graph. Each numbered line is a selected timeslice (see Figure 6-21).

Figure 6-21. CTRL+click allows multiple timeslices to be selected.

With multiple timeslices selected, use CTRL+space bar or uncheck “Show Dynamic
Data” to display the spectroscopic view. The selected timeslices can now be viewed
individually using the drop-down menu or simultaneously by checking “Graph All
Data Sets” (see Figure 6-22). This process is similar to viewing multiple selected
locations with mapping data (see Chapter 5).

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-253

 Figure 6-22. Multiple timeslices graphed simultaneously. Use CTRL+Y to toggle between
single- and double-Y axis.

NOTE: To quickly select multiple consecutive timeslices, use CTRL+click and drag
over the graph area you wish to select. CompleteEASE will ask how many sets you
wish to add to the multi-data set analysis, and it will select evenly distributed
timeslices within the specified range.

Manipulating Multiple Data Sets

When working with multiple data sets, there are additional options in the
+Data Sets menu above the graph. This expanded menu is shown in Figure 6-23.

Figure 6-23. Expanded +Data Sets menu.

Use these options to select or un-select data sets from the multi-data set analysis or
to clear multi-data set mode and return to the full dynamic data.

NOTE: ‘Fit Dynamic’ cannot be used when multiple timeslices are selected.

Multi-Sample Analysis (for multiple timeslices)

Now, let’s set up the multi-sample analysis. Click Model > Configure Options.
Check “Multi Sample Analysis” and click ‘Ok’ (see Figure 6-24).

6-254 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 Figure 6-24. Choose “Multi Sample Analysis” from the Available Options within the Show
Model Options” dialog.

Now, when the +Model Options section is expanded, there is a +Multi Sample
Analysis subsection that can be further expanded. Expand the MSA section and
click Add Fit Parameter. Choose Thickness #1. When finished, your model should
appear as shown in Figure 6-25.

Figure 6-25. Model with Thickness #1 as MSA parameter.

Next, adjust the starting thickness value for each timeslice. To do so, select each
timeslice individually and scroll each thickness until the model-generated curve
approximately matches the experimental curve for each data set. When finished,

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-255

 check “Graph All Data Sets” and your model and graph should appear as shown in
Figure 6-26.

Figure 6-26. Starting points for all timeslices have been adjusted by rolling the mouse wheel
over each individual thickness from within the multi-sample analysis section of the model.

Click ‘Fit’ and all thicknesses (for each timeslice) along with a single roughness value
that is used to fit all data sets are varied. This result is shown in Figure 6-27. If the
roughness was also changing with time, its thickness could be added as a time-
dependent variable within the multi-sample analysis section.

6-256 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 Figure 6-27. Fit results when a-Si thicknesses were allowed to vary at each of the five
timeslices, along with a single roughness value for all timeslices.

Fitting Optical Constants

There are several ways to proceed with fitting the optical constants of the a-Si layer.
One general method is shown here.
Click on a-Si parameterized and replace it with B-Spline.mat (located in Library tab,
Basic folder). Expand the +Nodes section of the B-Spline layer and click on Starting
Mat. = none. Choose a-Si.mat from the Library tab, Semiconductor folder, and the
B-Spline nodes will automatically adjust to try to match the a-Si.mat values. At this
point, your screen should look something like Figure 6-28.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-257

 Figure 6-28. Layer #1 as a B-Spline layer with Starting Mat. = a-Si.

Click ‘Generate’ and notice that the model-generated curves do not quite match the
experimental curves (see Figure 6-29).

Figure 6-29. Generated data after setting up the model.

Now, click ‘Fit’ and CompleteEASE will adjust the a-Si optical constants via the
B-Spline layer to fit all five data sets simultaneously while allowing the thickness at
each timeslice to fit independently. The result is shown in Figure 6-30, with the
B-Spline optical constants shown in Figure 6-31.

6-258 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 Figure 6-30. Fit result with a-Si optical constants allowed to vary.

Figure 6-31. a-Si optical constants.

At this point, the B-Spline/a-Si optical constants could be saved and used for fitting
the full dynamic data.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-259

 6.2. Growth Rate and Optical Constants
FEATURES IN THIS EXAMPLE
• GROC Model • Virtual interface

In this example, we will consider a multilayer stack of tantalum metal and

amorphous silicon thins films where the process conditions are varied for each
layer. During deposition of each layer, the growth rate is constant, and the optical
constants should not vary. Our goal is to determine the process condition from
these six different films.
Because of the ideal sample conditions, we can apply a “Growth Rate and Optical
Constants” (GROC) model to the dynamic data analysis. This model can only be
applied to cases where i) the growth rate is constant, ii) the optical constants do not
vary during growth, and iii) a virtual interface approximation can be applied for the
underlying structure (typically true for semiconductors and metals, but not
dielectrics).

Virtual Interface
A virtual interface can simplify dynamic data analysis by allowing us to ignore some
of the sample structure and only consider the current thin-film growth. The virtual
interface is placed near the surface of the film that is being studied. For example, in
Figure 6-32 below, if we are determining the optical constants of Layer #3, we
would place the virtual interface inside Layer #3 and ignore underlying materials.

Actual Sample Simplified Model

Layer 4 Layer 4

Layer 3
Virtual
Layer 3
Interface
Layer 2

Layer 1
Pseudo-Substrate
Substrate

Figure 6-32. A virtual interface allows the underlying sample structure to be approximated by
a single interface. The optical constants at this interface are called the “pseudo-substrate”.

There are different ways to calculate the virtual interface, but we will limit our
discussions to the common pseudo-substrate approximation (CPA). This
assumption replaces everything below the virtual interface by an effective, semi-
infinite substrate that has optical constants equal to the pseudo dielectric values for
the stack below that level. This approximation works amazingly well for
semiconductors and other high-index materials but does not work for low-index
transparent materials, like glass.

6-260 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 Growth Rate and Optical Constants (GROC Model)
To start this example, open the “Growth Rate and Opt. Const.” model, which is in
the Library tab, Advanced folder (see Figure 6-33).

Figure 6-33. Location of GROC model.

The GROC model is shown in Figure 6-34. Notice it does not contain a layered
structure, unlike models in previous examples. It is designed to simplify the
simultaneous extraction of growth rate and optical constants from dynamic in-situ
data. The GROC model utilizes a ‘virtual interface’, which means that the user does
not have to specify an optical model for the underlying layers/materials. The GROC
model assumes a constant growth rate, constant surface roughness, and that the
optical constants do not vary with film thickness over the specified time.

Figure 6-34. Example of the GROC model.

The common pseudo-substrate approximation (CPA) used within the GROC model is
not generally valid for all materials but works well for growth of semiconductors on
semiconductors or the deposition of absorbing materials (such as metals, after
about 5 nm of growth).

NOTE: The GROC model assumptions are not valid during the nucleation phase of
growth.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-261

 Details of the GROC model are described below:

Rate
Rate specifies the constant growth rate of the layer. The GROC model opens with
this parameter active as a fit parameter. Like any other fit parameter, the starting
value should be close to the actual value.

Fit Opt. Const.

When ‘ON’:
• Layer optical constants are fit using a piecewise-continuous polynomial
representation for the optical constants. The spacing of the polynomial
control points are specified by the “Spectral Resolution in eV” setting.
Draw Opt. Const. will graph the extracted optical constants while
Save Opt. Const. will save the extracted optical constants in a tabulated
material file.
• Starting values for the polynomial match the “Layer” entry, if present. This
is helpful for setting the optical constants to good starting values.
When ‘OFF’:
• The optical constants are specified by the “Layer” entry. If the Layer entry
is empty or no fit parameters are turned on within the layer, then the
optical constants are fixed and only growth rate is varied. If parameters
are turned on within the “Layer” entry, then the optical constants are fit
using this layer and its dispersion equation fit results.

NOTE: Optical Constants can be determined by turning Fit Opt. Const. ON or by

turning on fit parameters within the “Layer” entry.

Layer
The layer is used to determine starting values for optical constants when “Fit Opt.
Const. = ON” and specifies the optical constants of the layer when “Fit Opt. Const. =
OFF”. Any fit parameters defined in the layer will be included in the GROC analysis
of the data when the polynomial fitting is turned off (“Fit Opt. Const. = OFF”).

Angle Offset
NOTE: Angle Offset is only applied when ‘Fit Opt. Const. = OFF’ to avoid 100%
correlation.
Specifies an offset to the nominal angle of incidence specified in the experimental
data file. Fitting the angle offset may be useful when known reference optical
constants are specified in the “Layer” field and are not allowed to vary.

Roughness
Specifies and/or fits the surface roughness of the growing film, which is assumed to
be constant during the selected time range. It is not always possible to uniquely
extract the optical constants, growth rate, and surface roughness; an adequate
thickness of grown material (>> ¼ wave) must be encompassed by the selected time
range. Otherwise, the roughness should not be included in the fit.

6-262 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 VI Mode
Specifies how the GROC model implements the Virtual Interface. The “CPA” and
“CPA – Fit VI” modes assume the common pseudo-substrate approximation, which
is typically valid for the growth of semiconductors on semiconductors or absorbing
materials such as metals.
• “CPA”: The ellipsometric Psi and Delta parameters at the first selected
time are simply inverted into pseudo optical constants, which are then
used to calculate the ellipsometric trajectory for the subsequent film
growth.
• “CPA-Fit VI”: This defines the virtual-interface pseudo optical constants as
fit parameters, which minimizes the impact of random or systematic errors
in the Psi and Delta data used for the virtual-interface inversion.

Max. Times
To minimize analysis time, the GROC model limits the timeslices during analysis to
the number specified. Selected timeslices are evenly distributed across the selected
time range.

Seed Thickness
To improve the robustness of the fit, the GROC model performs an initial fit with an
initial subset of the selected time range before fitting all of the data. The ‘Seed
Thickness’ specifies how much time should be included in this initial fit (the initial
growth rate is used to calculate the time, given the ‘Seed Thickness’). For most fits,
the default value is acceptable, but in some cases, smaller or (more likely) larger
values could be tried if the fit does not converge properly.

Surface Tracking Fit Mode

When OFF (which is the default), the dynamic data in the selected time range is
used in the GROC analysis. If ON, the dynamic data immediately preceding and
including (the # of points is specified by the “Max. Times” setting) the selected
timeslice is used in the analysis. Since the selected timeslice is automatically moved
through the data during a dynamic fit, this mode enables a GROC analysis versus
time, which could be useful for extracting time-dependent growth rate and/or
surface roughness changes.

Multilayer Example
Open the “Dynamic Ta-Si Multilayer” data file from the Library tab, Examples folder.
This file is a dynamic in-situ measurement, so it has the “.iSE” extension. If you
cannot locate the file, check the drop-down box to ensure the dialog is showing .iSE
files (see Figure 6-35).

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-263

 Figure 6-35. Location of “Dynamic Ta-Si Multilayer” example data file. Use drop-down box to
make sure the dialog is showing .iSE files.

Upon opening, the graph will display Psi versus Time (see Figure 6-36).

Figure 6-36. Dynamic Ta-Si Multilayer dynamic data.

The data were collected during growth of six consecutive films, as detailed in Table
6-2. For the first minute, the substrate was monitored without film growth. Then,
six films were deposited, alternating between Ta and a-Si. For each layer, the
sputter gun current was varied to change the growth rate. Our goal in this example
is to determine the growth rate and optical constants for each of the six layers to
characterize the process conditions for this sputter chamber.

Table 6-2. Process conditions for all six layers of the dynamic Ta-Si multilayer.

Layer Time Start Time End Film Gun Current

1 1 6 Ta 0.2
2 6 11 a-Si 0.3
3 11 14 Ta 0.4
4 14 17 a-Si 0.5
5 17 24 Ta 0.1
6 24 31 a-Si 0.15

6-264 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 To start, open or re-open the “Growth Rate and Opt. Const.” model from the
Library tab, Advanced folder.
Next, select the time range to analyze by referring to the Time Start and Time End
columns in Table 6-2. To select the first Ta layer, from the Data panel, click ‘Set
Ranges’. Under Time Range, enter 1 to 6 minutes (see Figure 6-37).

Figure 6-37. Click the ‘Set Ranges’ button to select the time range for the first Ta film (1 to 6
minutes).

Set the starting values for the virtual-interface optical constants by loading a
starting material file into the “Layer”. Click on none and choose Ta.mat from the
Library tab, Metal folder (see Figure 6-38). The Model panel and Graph should now
look like Figure 6-39. Next, click ‘Fit’ and you should get the results shown in Figure
6-40.

Figure 6-38. Location of Ta.mat.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-265

 Figure 6-39. Before fitting, the GROC model is loaded with Ta.mat as the starting layer and
data from 1-6 minutes is selected.

Figure 6-40. Results of GROC analysis over time range from 1-6 minutes for Ta film.

6-266 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 Evaluation of fit result: MSE is ~18. Visual discrepancy observed at 251.1 nm from
1-3 minutes.

Recall that the GROC model requires the optical constants and growth rate to be
constant over the selected time range. The Ta layer deposition may not have
started at exactly 1 minute and the first few data points of growth may show
nucleation effects where the metal optical constants are not consistent with the
thicker layer. Try selecting data from 2-6 minutes to ignore the early stages of film
growth. Click ‘Fit’ again. Results are shown in Figure 6-41.

Figure 6-41. GROC analysis for Ta film growth from 2-6 minutes, ignoring the early stages of
film growth where the film optical constants may vary during nucleation.

Evaluation of fit result: MSE drops from ~18 to ~1.5. Result is significantly
improved by selecting 2-6 minutes instead of 1-6 minutes.

Record these results for later comparison.

• Use CTRL+L to record the growth rate into the fit log. Use a descriptive
name such as “Layer 1 (Ta)”.
• Click the Draw Opt. Const. command from the Model panel to display the
extracted optical constants vs. nm. Then, right-click in the graph and select
“Copy Data to Graph Scratchpad”.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-267

 Repeat the GROC analysis procedure for the other two periods of Ta growth (11-14
minutes and 17-24 minutes). Remember to avoid the nucleation region at the
beginning of growth.

NOTE: Use the ‘Show Data’ button above the graph if needed when selecting new
time ranges. The ‘Show Data’ button always switches the graph view back to
showing the experimental data.

NOTE: When you start a new region, the GROC model will calculate a new virtual
interface, but the growth rate will be left over from the previous fit. You can adjust
the growth rate to a better starting value if needed by using the mouse wheel to
scroll the parameter until the model-generated data approximately match the
experimental data.

Results for the second and third Ta films are shown below (see Figure 6-42 and
Figure 6-43).

Figure 6-42. GROC analysis for Ta growth from 12-14 minutes.

6-268 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 Figure 6-43. GROC fit for Ta growth from 18-24 minutes.

A comparison of all three Ta films using the Fit Log and Graph Scratchpad is shown
below (see Figure 6-44, Figure 6-45, and Figure 6-46).

Figure 6-44. Comparison of MSE and growth rate for Ta from Fit Log.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-269

 Figure 6-45. Graph Scratchpad comparing n for all three Ta films.

Figure 6-46. Graph Scratchpad comparing k for all three Ta films.

6-270 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

 Conduct GROC analysis for the a-Si layers. Details are not shown here, but results
are shown below (see Figure 6-47, Figure 6-48, and Figure 6-49).

Figure 6-47. Comparison of MSE and growth rate for a-Si from Fit Log.

Figure 6-48. Comparison of n for a-Si from Graph Scratchpad.

Figure 6-49. Comparison of k for a-Si from Graph Scratchpad.

CompleteEASE Software Manual Data Analysis 4 – In-situ Data • 6-271

 Growth Rate versus Gun Current for both Ta and a-Si are shown below (see Figure
6-50).

3.5
Ta
3 a-Si 2.57

Growth Rate (Å/s)

Ta Linear Fit
2.5
a-Si Linear Fit 2.01
2

1.5 1.27
1.12
1
0.53 0.46
0.5

0
0 0.1 0.2 0.3 0.4 0.5 0.6
Gun Current

Figure 6-50. Growth Rate for Ta and a-Si films versus sputter gun current.

6-272 • Data Analysis 4 – In-situ Data CompleteEASE Software Manual

7. Reference – Graph

In this chapter, the features and capabilities of CompleteEASE will be described

using a “reference” format. That is, the features will be listed in the order they
appear on screen but not necessarily in the order one would use to perform typical
procedures. The previous chapters on Basic System Operation and Data Analysis
should be studied for step-by-step tutorials.

7.1. Basic Screen Layout

The basic layout of CompleteEASE consists of tabbed panes across the top of the
window with a graph always visible in the bottom section (Figure 7-1). The window
can be resized, maximized, or minimized using standard Windows keystrokes or
mouse operations. The divider bar between the tab panes and graph can be moved
to adjust the relative size of the graph. The CompleteEASE screen layout is saved
when the program exits and restored upon restart.

Figure 7-1. CompleteEASE with the Hardware tab selected. Screen may vary depending on
ellipsometer model.

CompleteEASE Software Manual Reference – Graph • 7-273

 7.2. Graph
The graph is an important part of the CompleteEASE user interface. Spectroscopic
data, model fit results, and optical constants are all plotted on the graph. The
features described in this section can be used regardless of the type of information
displayed in the graph window.
If the mouse is paused in the graph for a few seconds, a “tool tip” box will appear
which reports the X and Y values corresponding to the mouse position. The mouse
can be moved to other positions and the X and Y values will be updated.
The legend can be repositioned by clicking in the legend area and dragging the
legend box to the desired location.

‘Show Data’
If optical constants (or something else besides data) are being graphed, you can
press this button to go back to the data graph.

Select Range
To “zoom in” on a specific spectral range, click and drag over the desired spectral
range (Figure 7-2). The graph will then redraw, plotting only the selected spectral
range.

Figure 7-2. To select a specific spectral range, hold down the left mouse button and drag the
mouse over the desired range.

The same is true for uniformity maps, as shown in Figure 7-3. Dragging the mouse
over an area (left) will select this region to graph (right).

Figure 7-3. Click and drag the mouse over area of the map (left) to select that area of the
map (right).

7-274 • Reference – Graph CompleteEASE Software Manual

 Zoom All
To plot the entire spectral range (or entire uniformity map area), simply click “Zoom
All” in the top right corner of the graph (Figure 7-4). You can also right-click in the
graph area and select “Zoom All” or, when viewing spectroscopic data, double-click
in the middle of the graph.

NOTE: “Zoom All” does not appear in the top right corner of the graph if “Use
Clipboard Aspect Ratio” is active.

Figure 7-4. “Zoom All” by clicking in the top right corner of the graph or by right-clicking and
selecting “Zoom All”.

7.3. Right-Click Menu (Graph)

Right-click in the graph area to access the menus shown in Figure 7-5.

Figure 7-5. Right-click menus for a spectroscopic scan (left) and a map (right).

CompleteEASE Software Manual Reference – Graph • 7-275

 Zoom All
Graphs all data – spectroscopic, dynamic, or map data. See “Zoom All” in previous
section for more details.

Statistics
The graphed data undergo a variety of statistical calculations. For example, a
straight-through measurement of air can be used to verify measurement accuracy.
The statistics of this measurement can then be calculated, as in Figure 7-6.

Figure 7-6. Graph Statistics, taken from a straight-through measurement.

Copy Graph to Clipboard

Graph is copied to the clipboard for use in other programs.

Copy Data to Clipboard

Graphed data curve values are copied to the clipboard for use in a spreadsheet.
Information will include a title line that matches the graph title followed by columns
of information with a heading for each column. The first column is the x-axis
information – typically wavelength – followed by a column of information for each
curve on the graph.

Copy Data to Graph Scratchpad

This places the data curves from the graph into the Graph Scratchpad where they
can be viewed and compared to other curves or customized.

Profile Along X and/or Y Axis

For uniformity maps, the 2D plot is not copied to the Graph Scratchpad. Instead,
you are given a choice to copy the data points from along the X axis, Y axis, or both,
as shown in Figure 7-7.

Figure 7-7. Choose from X Axis, Y Axis, or X and Y Axes when copying map data to the Graph
Scratchpad. Note the reference point is shown in parentheses.

7-276 • Reference – Graph CompleteEASE Software Manual

 View Graph Scratchpad
The Graph Scratchpad is a convenient place internal to CompleteEASE to compare
curves. The curves can be any type of data that is graphed including raw data,
model-generated curves, optical constants, or even an MSE profile from a
parameter uniqueness test. An example Graph Scratchpad is shown in Figure 7-8.
See Section 7.8 for more details.

Figure 7-8. The Graph Scratchpad can be used to compare curves and/or customize graphs.

Show Legend
Show or hide the graph legend box.

Show Symbols
Turn on/off graph symbols. Figure 7-9 shows an example data graph with symbols
on.

Figure 7-9. Graph with symbols turned on.

CompleteEASE Software Manual Reference – Graph • 7-277

 Set Symbol Size
This menu option is available when viewing map data. Change the size of symbols
representing map data points. The default size is 1, but this can be increased or
decreased depending on the view required.

NOTE: If symbol size is set to zero, CompleteEASE will attempt to auto-size the
symbols to best fit the graphed data.

Auto X-, Y-, or Z-Scale

X-, Y-, or Z-axis is automatically selected to show all data.

Manual X-, Y-, or Z-Scale

User can select lower and upper X-, Y-, or Z-axis limits. All data are still included
when fitting, but only points within this range are graphed.

Scale Z-Range by n-Sigma

User can quickly select the lower and upper Z-axis limits for mapped data based on
a multiple of the standard deviation. This can be useful when there are a few “bad”
points on an otherwise uniform sample. When selected, a dialog box appears
requesting the number of standard deviations to use when calculating the range, as
shown in Figure 7-10. All data are still included when fitting, but only points within
this range are graphed.

Figure 7-10. Enter the number of standard deviations to use for calculating the Z-range.

Graph Points as a Line

Allows user to show map data unwrapped in a line graph. For example, a map of
SiO2 thickness over a wafer is graphed as a line in Figure 7-11.

7-278 • Reference – Graph CompleteEASE Software Manual

 Figure 7-11. Map data graphed as a line.

Draw Surface Plot

Allows user to show map data as a surface plot (see Figure 7-12). When checked,
an additional menu item becomes available in the right-click menu to edit the
surface plot settings (see Figure 7-13). Users can click and rotate surface plots or
edit surface plot settings (see Figure 7-14)

Figure 7-12. Example surface plot.

CompleteEASE Software Manual Reference – Graph • 7-279

 Figure 7-13. Additional menu item for surface plots: Edit Surface Plot Settings.

Figure 7-14. Surface Plot Settings dialog box.

Use Clipboard Aspect Ratio

The standard graph will scale with the size of the graph pane. However, when the
graph is copied to the clipboard, it always uses the configured aspect ratio. Choose
this button to view the graph as it will appear when copied to the clipboard. The
clipboard aspect ratio can be modified under the “Graph Clipboard Parms” section
of Options > Configuration Controls > ‘Edit Configuration’.

Cancel Menu
Close the right-click graph menu.

7.4. +Data Menu

The +Data drop-down menu in the top left corner of the graph lists different graph
options, as shown in Figure 7-15.

NOTE: The “<Pseudo> Transforms”, as shown in Figure 7-15, are hidden by default
in CompleteEASE. See section below for how to display them.

7-280 • Reference – Graph CompleteEASE Software Manual

 Figure 7-15. +Data menu for standard ellipsometry data.

If the experimental data is from a Generalized Ellipsometry (GE) measurement or a

Mueller Matrix Ellipsometry measurement, more menu options are available (see
Figure 7-16).

Figure 7-16. +Data menu for GE and Mueller Matrix data.

Psi, Delta
When Psi or Delta are selected, the graphs plot the standard ellipsometry
parameters for amplitude ratio, Psi (), and phase difference, Delta (). These are
traditional representations of an ellipsometry measurement, based on Equation
7-1.

CompleteEASE Software Manual Reference – Graph • 7-281

 Equation 7-1
~
Rp
 = tan( )e = ~
i

Rs

Re(rho), Im(rho)
Real and Imaginary rho graphs refer to an alternate view of the ellipsometry
measurement, based on rho (), as given in Equation 7-1.

N, C, S
Another alternate form of viewing ellipsometry measurements is based on the
Mueller Matrix formalism. For a standard, isotropic sample with no depolarization,
the Mueller Matrix response would appear as shown in Equation 7-2.

Equation 7-2

 1 −N 0 0
− N 1 0 0 
M isotropic = 
 0 0 C S
 
 0 0 − S C

Where

Equation 7-3

N = cos(2) , C = sin( 2) cos() , and S = sin( 2) sin( )

Thus, N, C, and S are the non-zero elements of the isotropic Mueller Matrix and are
related to the Psi and Delta curves traditionally graphed for ellipsometry. A few
advantages to the NCS notation include 1) they are always defined and bounded
between -1 and 1 and 2) they are linearly related to the intensity harmonics
measured by any ellipsometer system. Thus, they are closer to the direct
measurables of the ellipsometer and there is approximately equal precision in N, C,
and S.

<Pseudo> Transforms (<e1>, <e2>, <n>,<k>)

These options show in the +Data menu only if first enabled in the CompleteEASE
configuration. Click on Options > Configuration Controls > ‘Edit Configuration’.
Check “Show Advanced Config. Options” checkbox and expand the General section
to find “Display Pseudo Transforms” (see Figure 7-17).

7-282 • Reference – Graph CompleteEASE Software Manual

 Figure 7-17. CompleteEASE Configuration window to change Display Pseudo Transforms =
True.

The <Pseudo> Transforms menu expands to include <e1>, <e2>, <n>, and <k>, as
shown in Figure 7-18. These are referred to as the pseudo optical functions to
differentiate from the actual material optical functions. They are a direct
calculation based on the measured ellipsometry data with the assumption of a
single reflection from the surface. This calculation is given in Equation 7-4. The
transformation can be performed for any data set, but the pseudo curves lose
meaning when the assumption of single reflection is not valid. In other words, they
only transform the data correctly for a substrate. The pseudo optical constants are
NOT equal to the material optical constants if there are any thin films or surface
layers present during measurement of the ellipsometry data.

Equation 7-4

~  1−   
2

 = 1 + i  2 = n~ = ( n~ + i k ) 2 = sin 2 ( ) 1 + tan 2 ( )

2
 
  1 +   

NOTE: Pseudo optical constants do not represent a layer's optical constants. To

view the layer optical constants, right-click on the layer and choose "Graph Layer
Optical Constants" or expand the +Model menu above the graph and choose from
various options.

CompleteEASE Software Manual Reference – Graph • 7-283

 Figure 7-18. <Pseudo> Transforms

Intensity
This option shows the reflected or transmitted intensity that is collected at same
time as SE data using CompleteEASE. The intensity represented here is an average
of the p- and s- polarization components. The accuracy of the intensity data
depends on the instrument calibration. Please consult your hardware manual to
improve intensity accuracy.
Because M-2000 and RC2 spectroscopic ellipsometers collect data from the AC
signal, the intensity value measured here is also from the AC signal. Similarly, the
model calculations only consider the AC components. Depolarization is measured
from the DC signal and will not be measured (or modeled) as part of the intensity.

NOTE: See p-s (Total) Intensity under +Data > More Options to view the effects of
depolarization on intensity (see Figure 7-20).

Depolarization
Percent depolarization is collected on ellipsometers that incorporate adjustable or
rotating compensators. This information can be used to determine sample and
measurement non-idealities such as thickness non-uniformity, spectrometer
bandwidth, angular spread, and backside reflections.
In the presence of depolarization, the Mueller Matrix for an isotropic sample is
written in Equation 7-5. In ellipsometry, all Mueller Matrix elements are
normalized to the first element, M11.

7-284 • Reference – Graph CompleteEASE Software Manual

 Equation 7-5

 1 − PN 0 0 
− P  N P 0 0 
M sample =  
 0 0 P C PS
 
 0 0 − PS P  C

With the Polarization (P) defined as:

Equation 7-6

P = N 2 + C2 + S2

This gives the depolarization that is reported as:

Equation 7-7

%𝐷𝑒𝑝𝑜𝑙𝑎𝑟𝑖𝑧𝑎𝑡𝑖𝑜𝑛 = (1 − 𝑃2 ) ⋅ 100%

Graph vs. Angle

Most data are graphed versus wavelength. When multiple angles of incidence have
been collected, it can be useful to graph one wavelength versus angle to visualize
where the pseudo-Brewster angle occurs (when Psi goes through minimum and
Delta goes through 90° or 270°). Figure 7-19 shows example data graphed versus
angle of incidence.

Figure 7-19. Example data graphed versus angle of incidence.

CompleteEASE Software Manual Reference – Graph • 7-285

 Double-Y Axis
View two different data curves simultaneously, with each having its own Y-Axis as
shown in Figure 7-16 where Psi and Delta are graphed on two different axes on two
different sides of the plot.

Difference Mode
Difference Mode plots the difference between experimental data and model-
generated data.

More Options
Additional graphing choices are listed under “More Options” (see Figure 7-20).

Figure 7-20. More Options from the +Data menu.

p- s- Intensity
The p- and s- Intensity values are measured from the AC signal. Thus, these values
represent Rp and Rs for reflection and Tp and Ts for transmission. Because they are
calculated strictly from the AC signal of the M-2000 or RC2 ellipsometers, these
values will not contain any effect from depolarized light. When CompleteEASE
models this data, it also calculates values corresponding to AC measurements.

p- s- Intensity (Total)
The p- and s- Intensity (Total) values are measured from both the AC and DC signal
to measure the total reflection or transmission in the p- and s- directions and
include effects of depolarized light. This would be a better comparison to most
spectrophotometric measurements (or even measurements from WVASE) as most
do not distinguish between p-, s- and the unpolarized light along p-, s-.

Error Bars
The error bars of Psi, Delta, Depolarization, Intensity, and Mueller Matrix elements
are calculated during measurement to estimate the random error associated with
each data point.

7-286 • Reference – Graph CompleteEASE Software Manual

 Show GE Items
Anisotropic samples are often characterized using Generalized Ellipsometry
measurements, which consider the entire Jones Matrix, as shown in Equation 7-8.

Equation 7-8

~
rpp ~
rsp 
J Anisotropic =  ~
 rps rss 
~

As ellipsometry measurements are a ratio of Jones matrix elements, there are three
measured values possible to access the entire Jones Matrix. These values are
referred to as AnE, Aps, and Asp (see Equation 7-9, Equation 7-10, and Equation
7-11).

Equation 7-9
~
r
AnE = tan(  )ei = ~pp
rss

Equation 7-10
~
r
i p s
Aps = tan( ps )e = ~ps
rpp

Equation 7-11
~
r
i sp
Asp = tan( sp )e = ~sp
rss

When Generalized Ellipsometry data are present, you can select “Show GE Items”
from the +Data menu (Figure 7-21). In addition to menu options detailed above,
you will also find “Select Ratios to Graph”.

CompleteEASE Software Manual Reference – Graph • 7-287

 Figure 7-21. Show GE Items.

Select Ratios to Graph

You can select which of the anisotropic ellipsometry ratios to graph from the dialog
box shown in Figure 7-22.

Figure 7-22. Select Generalized Ellipsometry ratio(s) to graph.

7-288 • Reference – Graph CompleteEASE Software Manual

 Show Mueller Items
When Mueller Matrix data are present, you can select “Show Mueller Items” from
the +Data menu (Figure 7-23).

Figure 7-23. Mueller Matrix Graphing Options

The Mueller Matrix description of a sample is shown in Equation 7-12.

Equation 7-12

 m11 m12 m13 m14 

m m22 m23 m24 
 21
m31 m32 m33 m34 
 
m41 m42 m43 m44 

Mueller Matrix Diag, Off Diag

The diagonal and off-diagonal choices select specific elements from the Mueller
Matrix of Equation 7-12 to graph. The diagonal elements will include any of the
measured values from m11, m22, m33, and m44. The off-diagonal elements will
include all remaining elements that have been measured.

NOTE: The matrix is often normalized to m11. In this case, m11 is not graphed
when “Mueller Matrix Diag” is selected.

Mueller Matrix NCS

When choosing Mueller Matrix NCS, the standard Mueller Matrix elements that are
non-zero for an isotropic sample are graphed, which include all the following that
were measured: m21, m12, m33, m34, m43 and m44. These elements relate to NCS of
Equation 7-2 via the following:

CompleteEASE Software Manual Reference – Graph • 7-289

 Equation 7-13

 1 −N 0 0  1 m12 0 0 
− N 1 0 0  m21 1 0 0 
 = 
 0 0 C S  0 0 m33 m34 
   
 0 0 −S C  0 0 m43 m44 

Select Mueller Matrix Elements

Opens a dialog box to allow you to select any number of the various measured
Mueller Matrix elements for graphing, as shown in Figure 7-24. This dialog box then
allows you to quickly select or clear all elements, select the on-diagonal blocks or
select the off-diagonal blocks.

Figure 7-24. Select the Mueller Matrix elements you wish to graph by checking the
appropriate boxes.

The Mueller Matrix elements available for graphing will coincide with the measured
values. The last row/column of the Mueller matrix requires probing and detection
of the handedness of the polarization. This requires a compensator before or after
the sample. The RC2 has rotating compensators before and after the sample, so it
can detect all 16 elements of the Mueller Matrix, though generally represented as
ratios to the M11 element. Other ellipsometer types may have compensators
before or after the sample, which allows them to only collect the first three rows or
first three columns. Rotating analyzer and rotating polarizer ellipsometers (without
compensator before or after sample) can’t detect the fourth row and the fourth
column. These various configurations are shown in Equations Equation 7-14,
Equation 7-15, Equation 7-16, and Equation 7-17.

7-290 • Reference – Graph CompleteEASE Software Manual

 Equation 7-14

• • • 
• • • 
M RAE,RPE = 
• • • 
 
   

Equation 7-15
• • • •
• • • •
𝑀𝑅𝐶𝐸(𝐶𝑜𝑚𝑝𝑒𝑛𝑠𝑎𝑡𝑜𝑟 𝑏𝑒𝑓𝑜𝑟𝑒 𝑠𝑎𝑚𝑝𝑙𝑒) =[ ]
• • • •
× × × ×

Equation 7-16
• • • ×
• • • ×
𝑀𝑅𝐶𝐸(𝐶𝑜𝑚𝑝𝑒𝑛𝑠𝑎𝑡𝑜𝑟 𝑎𝑓𝑡𝑒𝑟 𝑠𝑎𝑚𝑝𝑙𝑒) =[ ]
• • • ×
• • • ×

Equation 7-17

• • • •
• • • •
M RC 2 = 
• • • •
 
• • • •

MM-Jones Quality
This graph helps determine whether the Mueller Matrix can be converted into a
corresponding Jones Matrix. The Mueller Matrix can handle both isotropic and
anisotropic samples with and without depolarization. The Generalized Jones matrix
can also handle both isotropic and anisotropic samples but does not handle
depolarization. Thus, the MM-Jones Quality factor will only be close to zero when
the conversion is successful, signifying that the depolarization is not a significant
issue.
The basic definition of the MM-Jones Quality factor is simply the RMS error
between the measured MM parameters and the "closest" (in a least squared sense)
Jones matrix. There are exact formulas for converting a Jones matrix (4 complex
values) into a Muller Matrix (16 real values). However, there is no exact conversion
from MM to Jones Matrix, so CompleteEASE performs a nonlinear regression fit to
find the closest Jones matrix to the MM data and report the RMS error. In
CompleteEASE, the RMS error is then multiplied by 1000, such that a MM-Jones
Quality of "1" means that the RMS difference between the closest Jones matrix and
the MM data is 0.001, which is the target value for MM data accuracy.

CompleteEASE Software Manual Reference – Graph • 7-291

 Depolarization Index
The depolarization index is a calculation of the depolarization for any sample
(isotropic or anisotropic) that requires measurement of the full MM. It ranges
between 0 (for completely depolarizing samples) and 1 (for non-polarizing samples).

MM Derived Parameter
This option will multiply the specified Stokes vector by the measured Mueller
Matrix to determine the "derived" outcome, as shown in Figure 7-25. The user gets
a choice of source polarization state and polarization state detector, as listed in
Figure 7-26. In addition, there are a large variety of derived parameters, as listed in
Figure 7-27.

Figure 7-25. Mueller Matrix Derived Parameter dialog box.

Figure 7-26. Choice of Source Polarization States.

7-292 • Reference – Graph CompleteEASE Software Manual

 Figure 7-27. Choice of Derived Parameters.

Show MM Graph Utility

This allows the user to view the Mueller Matrix elements graphically, as shown in
Figure 7-28 for a spectroscopic MM scan (X-Y Type) and in Figure 7-29 for a
rotational MM scan (Contour Type).

Figure 7-28. MM Graph Utility.

CompleteEASE Software Manual Reference – Graph • 7-293

 Figure 7-29. MM Graph Utility for rotational MM scan.

In addition to both X-Y and Contour graphing, the user can use the Graph Type
panel select the Data Type from choices of 1) Experimental Only, 2) Exp. and Gen.,
3) Difference: Exp. and Gen., or 4) Generated Only. The standard X-Y Axis will be
Wvl vs. Angle for spectroscopic single-position measurements. However, in the
case of automated rotation stages that record the orientation of the sample for
each measurement, there are also choices of Wavelength (Wvl) vs. Rotation Angle
or Angle of Incidence (AOI) vs. Rotation Angle. In these two cases, the remaining
parameter (AOI or Wvl) can be adjusted to different positions with the sliders.
From the Display panel, click ‘Graph Settings’ to access additional features, as
shown in Figure 7-30. The 15 normalized MM elements range from -1 to +1. Check
“Auto-Scale Y Axis” to view these elements over an appropriate range to view the
full scale of data for each element. Check “Always On Top” to maintain the MM
Graph Utility over other CompleteEASE windows and other software programs. If
“Un-Normalized Mueller Matrix” data was measured, check “Show M11 Graph” to
view M11 which is the reflected or transmitted intensity for an unpolarized
incoming light beam.

NOTE: ALL MM measurements are normalized to M11 (even if the user selects “Un-
Normalized Mueller Matrix” at the time of measurement). The difference is that for
“unnormalized” data, CompleteEASE also measures the intensity information such
that M11 can also be graphed. When “Show M11 Graph” is checked, the scale of
M11 will be 0 to +1, even when the other graphs are between -1 and +1. An
additional benefit of measuring the “unnormalized” Mueller Matrix is that
CompleteEASE can calculate the like- and cross-polarized intensities (pp, ss, ps, sp).

7-294 • Reference – Graph CompleteEASE Software Manual

 Figure 7-30. Graph Settings within the Mueller Matrix Graph Utility

7.5. +Model Menu

The +Model drop-down menu near the top left corner of the graph contains options
for graphing various quantities relating to the currently open model. Most often,
when using the standard multilayer model, the +Model menu will list different
options for graphing the optical constants of the model layers, as shown in Figure
7-31. Unlike the +Data menu, the +Model menu is more fluid, and its contents will
change based on the currently open model and the values of fit parameters. Since
the most widely used model is the standard multilayer model, only those +Model
options will be explored in this section. Other models (such as the Retardance
Model, Opt. Const. Compare Model, and Pore Size Model) will have other options in
this menu (see Section 9.9).

Figure 7-31. +Model menu for the standard multilayer model showing the different options
for graphing optical constants.

Layer Optical Constants

The +Model menu is the place to go to graph the optical constants of the materials
present in the model. Here, you will find a list of each layer in the model (e.g.
“Substrate = SI_JAW”, “Layer #1 = INTR_JAW”, and “Layer #2 = SIO2_JAW” in Figure
7-31 above) which can be clicked on to show a graph of that layer’s optical
constants. This is equivalent to right-clicking on the layer name (e.g. SIO2_JAW) in
the Model panel and selecting “Graph Layer Optical Constants”. The optical
constants will be graphed as either n and k or ε1 and ε2 depending on how
CompleteEASE is configured in the Options tab to display optical constants.

CompleteEASE Software Manual Reference – Graph • 7-295

 All Layer Optical Constants
Choosing “All Layer Optical Constants” or “All Layer Optical Constants (No
Substrate)” will plot the optical constants from all layers in the model including the
substrate or not including the substrate, respectively, on the same graph.
Depending on how CompleteEASE is configured in the Options tab to display optical
constants, you will either be given the choice between “n”, “k”, or both “n & k” or
“e1”, “e2”, or both “e1 & e2”. An example of plotting the optical constants for all
layers in the “Si with Thermal Oxide” model is shown for “All Layer Optical
Constants” in Figure 7-32 and for “All Layer Optical Constants (No Substrate)” in
Figure 7-33.

Figure 7-32. Plot of “All Layer Optical Constants” with both n and k for the “Si with Thermal
Oxide” model.

Figure 7-33. Plot of “All Layer Optical Constants (No Substrate)” with both n and k for the “Si
with Thermal Oxide” model.

7-296 • Reference – Graph CompleteEASE Software Manual

 Optical Constants Profile
Choosing “Optical Constants Profile” or “Optical Constants Profile (No Substrate)”
will plot the depth profile of the optical constants for all layers in the model
including the substrate or not including the substrate, respectively. The optical-
constants depth profile is a plot of the optical constants of the model layers at a
chosen wavelength versus thickness, which is measured as the distance from the
substrate. Depending on how CompleteEASE is configured in the Options tab to
display optical constants, the depth profile will be graphed as either n and k or ε1
and ε2. Plotting the optical-constants depth profile is often useful when using a
Graded layer (see Section 10.18). An example of plotting the optical-constants
depth profile for a model with a Graded layer is shown for “Optical Constants
Profile” with the substrate in Figure 7-34. An optical-constants depth profile is
shown from the substrate (left) to the surface of the layer stack (right).

Figure 7-34. Depth profile for a two-layer stack, where the bottom layer is graded.

7.6. +Results Menu

The +Results drop-down menu near the top left corner of the CompleteEASE graph
lists the results of model parameters obtained from fit results over translation
(uniformity map), dynamic (in-situ), or rotation data. The +Results menu is only
active once a model has been fit to all data in the data set (i.e. after ‘Fit Scan Data’,
‘Fit Dynamic Data’, or ‘Fit Rotation Data’ has been pressed). An example of the
+Results menu is shown in Figure 7-35. As can be seen below, the +Results menu
only contains the names of fit parameters and derived parameters from the model
as well as the MSE. The +Results menu can graph these fit parameter results for all
the positions or timeslices in the data set, producing uniformity-map or dynamic
graphs respectively. The graphs available in the +Results menu are equivalent to
those produced when clicking on the blue, underlined text of a fit parameter (e.g.
Thickness #1) in the Fit panel. Thus, the options listed in the +Results are not static;
rather, they change depending on which fit parameters or derived parameters are
present in the model and have been fit most recently.

CompleteEASE Software Manual Reference – Graph • 7-297

 Figure 7-35. CompleteEASE screen showing the +Results menu expanded for fit results with
translation data.

7.7. Special Graph Features

Multiple data sets, rotation data, map data, and dynamic data activate special
graph features that are normally not present. These features are described here.

Multiple Data Sets

When multiple data sets are present (via appending files) or selected (from rotation
data, map data, or dynamic data), a +Data Sets menu appears above the graph. The
multiple data sets are numbered and listed in a drop-down menu above the graph.
You can use this drop-down menu to select which data set to view, but all will be
included during a fit (see Figure 7-36). To view all data sets simultaneously, check
“Graph All Data Sets” (see Figure 7-37).

7-298 • Reference – Graph CompleteEASE Software Manual

 Figure 7-36. Use drop-down menu to view individual data sets. One data set will be
displayed, but all will be included in the fit.

Figure 7-37. Check “Graph All Data Sets” to view all data sets simultaneously.

When multiple data sets are present, the +Data Sets menu above the graph allows
selection options for viewing and fitting (see Figure 7-38).

Figure 7-38. +Data Sets menu.

CompleteEASE Software Manual Reference – Graph • 7-299

 • “Select Current Data Set” adds currently listed data set to the group for
viewing all simultaneously and fitting.
• “Un-Select Current Data Set” removes currently listed data set from the
group for viewing and fitting, but it remains in the list in case you would
like to reselect it later.
• “Select Current Data Set ONLY” will un-select all other data sets and select
only the current data sets. None of the other data sets are removed from
the list, in case you would like to reselect one of them later.
• “Select ALL Data Sets” will select all data sets from the list to be active for
viewing and fitting.
• “Delete Current Data Set” removes the selected data set from the Multiple
Data Sets list.
• “Clear Multi-Data Set Mode” will remove the Multiple Data Sets list and
open the currently selected data set only.

Rotation Data
Rotation data is collected versus azimuthal angle, often using an automated
rotation stage accessory. Rotation measurements are often helpful with
anisotropic samples.
In rotation data, each measured data set corresponds to a rotation azimuth. When
rotation data is opened, the graph will initially display spectroscopic data from the
first-measured azimuth (see Figure 7-39). Use the position slider bar above the
graph to change the rotation angle for viewing and fitting.

Figure 7-39. With “Show Rotation Data” unchecked, the graph will display data from a single
rotation angle.

The “Show Rotation Data” checkbox can be used to display data from a single
wavelength versus rotation angle (see Figure 7-40). The wavelength displayed can
be changed from the ‘Set Ranges’ button in the Data panel.

7-300 • Reference – Graph CompleteEASE Software Manual

 Figure 7-40. “Show Rotation Data” will display data from a single wavelength versus rotator
angle.

With “Show Rotation Data” selected, you can select multiple rotation angles to set
up a multi-data set list. Use CTRL+click to select multiple rotation angles in the
graph (see Figure 7-41) or CTRL+click and drag to first select a range of rotation
angles and then enter number of data sets to evenly select within that range (see
Figure 7-42). Uncheck “Show Rotation Data” to display spectroscopic data for all
selected rotations and for multi-data set options (see Figure 7-43).

Figure 7-41. Use CTRL+click to select multiple rotations.

CompleteEASE Software Manual Reference – Graph • 7-301

 Figure 7-42. Use CTRL+click and drag to select evenly spaced rotations over a range.

Figure 7-43. Spectroscopic view of selected rotations and multi-data set mode options.

Translation (Uniformity Map) Data

Map data is collected versus lateral position on the sample, often using an
automated mapping stage accessory. Map measurements are helpful for evaluating
sample uniformity.
In map data, each measured data set corresponds to a position on the sample.
When map data is first opened, the graph will initially display a 2D plot of the data
at a single wavelength versus measurement location, with a “Show Map Data”
checkbox selected (see Figure 7-44). Use the ‘Set Ranges’ button from the Data
panel to change the displayed wavelength.

7-302 • Reference – Graph CompleteEASE Software Manual

 Figure 7-44. “Show Map Data” displays the data from a single wavelength versus
measurement location.

If “Show Map Data” is unchecked, spectroscopic data from a single location will be
displayed (see Figure 7-45). A slider bar will also appear that allows you to change
the selected position.

Figure 7-45. With “Show Map Data” unchecked, the graph will display the spectroscopic
measurements from a single location. Use the slider bar above the graph to change locations
across the wafer.

With “Show Map Data” selected, you can select multiple measurement locations to
set up a multi-data set list. Use CTRL+click to select multiple measurement
locations on the map (see Figure 7-46) or CTRL+click and drag to select a region of
measurement locations (see Figure 7-47). Each selected measurement location will
show a gray border around the location symbol in the graph. Uncheck “Show Map
Data” to display spectroscopic data for all selected measurement locations and for
multi-data set options (see Figure 7-48).

CompleteEASE Software Manual Reference – Graph • 7-303

 Figure 7-46. Use CTRL+click to select multiple measurement locations.

Figure 7-47. Use CTRL+click and drag to select all measurement locations in a region.

7-304 • Reference – Graph CompleteEASE Software Manual

 Figure 7-48. Spectroscopic view of selected locations and multi-data set mode options.

Dynamic (In Situ) Data

Dynamic data is collected versus time, often using a special cell or stage accessory
or with the ellipsometer mounted to a deposition chamber. Dynamic data are
helpful for evaluating film growth and monitoring processes.
In dynamic data, each measured data set corresponds to a time after the start of
the measurement. When dynamic data is first opened, the graph will initially
display a limted number of wavelengths versus time, with a “Show Dynamic Data”
checkbox selected (see Figure 7-49).

Figure 7-49. Dynamic Data displaying five wavelengths versus time.

If “Show Dynamic Data” is unchecked, spectroscopic data from a single timeslice

will be displayed (see Figure 7-50). A slider bar will also appear that allows you to
change the selected timeslice.

CompleteEASE Software Manual Reference – Graph • 7-305

 Figure 7-50. With “Show Dynamic Data” unchecked, the graph will display spectroscopic data
from a single timeslice.

With “Show Dynamic Data” selected, you can select multiple timeslices to set up a
multi-data set list. Use CTRL+click to select multiple timeslices in the graph (see
Figure 7-51) or CTRL+click and drag to select a range of evenly spaced timeslices
(see Figure 7-52). Uncheck “Show Dynamic Data” to display spectroscopic data for
all selected timeslices and for multi-data set options (see Figure 7-53).

Figure 7-51. Use CTRL+click to select multiple timeslices.

7-306 • Reference – Graph CompleteEASE Software Manual

 Figure 7-52. Use CTRL+click and drag to select evenly spaced timeslices over a range.

Figure 7-53. Spectroscopic view of selected timeslices and multi-data set mode options.

7.8. Graph Scratchpad

The Graph Scratchpad can be used to customize graphs, compare curves, and add
trendlines. To add curves to the Graph Scratchpad, right-click in the graph and
select “Copy Data to Graph Scratchpad”. To view the Graph Scratchpad, right-click
in the graph and select “View Graph Scratchpad” (see Figure 7-54).

CompleteEASE Software Manual Reference – Graph • 7-307

 Figure 7-54. Right-click in the graph and select "Copy Data to Graph Scratchpad" or “View
Graph Scratchpad”.

An example of the Graph Scratchpad is shown in Figure 7-55. Curves in the Graph
Scratchpad are tabulated on the top panel and labeled C1, C2, C3, etc. Use
checkboxes in the “Graph” column to display/hide curves. Use checkboxes in the
“2nd Y” column to move any curve to a secondary Y axis. Click in the “Color”
column to change curve color. Click in the “Style” column to choose from a drop-
down list of curve styles (available styles include solid (0), dashed (1), circles (2),
thin (3), or thick (4)). Double-click in the “Curve Name” column to customize the
curve name and legend display.

Figure 7-55. The Graph Scratchpad can be used to customize, compare, and add trendlines to
curves.

7-308 • Reference – Graph CompleteEASE Software Manual

 Edit
Select any curve in the table and click ‘Edit’ to make all changes for a single curve
(see Figure 7-56). You can also click and drag to select multiple curves, then ‘Edit’
to make changes for all selected curves (see Figure 7-57). When multiple curves are
selected, an additional option appears in the Edit Selected Curve(s) dialog:
“Automatically Set Colors”. When this option is checked, CompleteEASE will
automatically assign a unique color to each curve (see Figure 7-58).

Figure 7-56. Edit Selected Curves(s) dialog for a single curve.

Figure 7-57. Edit Selected Curve(s) for multiple curves.

CompleteEASE Software Manual Reference – Graph • 7-309

 Figure 7-58. “Automatically Set Colors” option applied – CompleteEASE gives each curve a
unique color.

Delete
Click ‘Delete’ to remove selected curve from the Graph Scratchpad.

Delete All
Click ‘Delete All’ to remove all curves / clear the Graph Scratchpad.

Create Curve
You can create curves in the Graph Scratchpad by clicking ‘Create Curve’. This will
open the Edit Equation dialog, allowing you to enter an equation using standard
mathematical operations (see Figure 7-59). Existing curves are available as
parameters by their curve name and added to an equation by clicking ‘Add
Parameter’ (see Figure 7-60).

7-310 • Reference – Graph CompleteEASE Software Manual

 Figure 7-59. 'Create Curve' button allows you to edit your own equation with reference to the
curves already displayed.

Figure 7-60. 'Add Parameter' allows choice of the other curves or values available.

Add Trend
Click ‘Add Trend’ to fit a Linear, Quadratic, or Quartic curve to a selection of the
data (see Figure 7-61). You can extend this curve beyond the selected range by a
selectable number of “x units” and you can add legend information regarding either
the equation or the Y=0 intercept.

Figure 7-61. Click ‘Add Trend’ to open the Trend Editor dialog.

CompleteEASE Software Manual Reference – Graph • 7-311

 Figure 7-62. Graph Scratchpad with linear trend (T1) applied to a k curve (C2).

NOTE: Trends will be indicated in the Graph Scratchpad by T1, T2, etc. Curves will
be indicated by C1, C2, etc. Trends and Curves offer different right-click menu
options.

Right-Click Menus
The graph within the Graph Scratchpad contains generally the same right-click
options as described in Section 7.3, including “Copy Graph to Clipboard” for pasting
into other programs.
You can also right-click one or more curves to Add Offset, Graph Difference
(requires two curves), Graph Ratio, Graph Average, and Graph Standard Deviation
(requires three or more curves). See Figure 7-63 and Figure 7-64.

Figure 7-63. Right-click multiple curves for additional options.

7-312 • Reference – Graph CompleteEASE Software Manual

 Figure 7-64. Graph Scratchpad with various options added.

When two linear trends are selected, you can right-click and select “Calculate
Intersection” (see Figure 7-65).

Figure 7-65. Right-Click to "Calculate Intersection" between two trendlines.

Graph Settings
Click ‘Graph Settings’ to change titles, change font size, show or hide grid lines,
change the line width independent of the line style, and/or use log scale for X or Y
axis (as long as there is no negative or zero data) (see Figure 7-66).

CompleteEASE Software Manual Reference – Graph • 7-313

 Figure 7-66. 'Graph Settings' button allows access to titles, font sizes, grid lines, line width,
and log scales.

NOTE: Right-click in any title box in the Graph Settings dialog or in the main table of
the Graph Scratchpad to select from common symbols (see Figure 7-67).

Figure 7-67. Symbol Chooser dialog.

Open
Opens a previously saved Graph Scratchpad file with *.scratch extension.

Save
Saves the current Graph Scratchpad with *.scratch extension.

7-314 • Reference – Graph CompleteEASE Software Manual

8. Reference – Measurement Tab

An example Measurement tab is shown in Figure 8-1. This tab consists minimally of
three panels: System Status, Measurement Controls, and Fit Results. The view of
this tab and measurement choices available will vary depending on the hardware
configuration. A brief description of all commands within this tab is provided.

Figure 8-1. Measurement tab for alpha-SE systems. The camera accessory is optional and
may not be available on your system.

8.1. System Status Panel

This panel describes the current hardware status. Before the ellipsometer is ready
to measure, the message will state “Not Initialized”. If the ellipsometer is ready for
operation, “Waiting to Acquire Data” will be displayed.

8.2. Measurement Controls Panel (alpha-SE)

Mode
Specifies the type of measurement to collect on alpha-SE systems. Further details
on each mode are given in Table 8-1.

CompleteEASE Software Manual Reference – Measurement Tab • 8-315

 Sample Alignment
Specifies how the z-stage should be used to adjust the sample height before the
measurement. Table 8-1 provides details for each choice.

Table 8-1. Measurement Settings for an alpha-SE.

CATEGORY OPTION DESCRIPTION

Mode Standard Use for most measurements (~ 10 seconds)
Fast Quick measurements (~ 3 seconds)
Long Low reflectivity samples or high precision
measurements (~30 seconds)
Transmission Collects intensity transmission measurements.
M.M. Mueller Matrix measurement – for anisotropic and
depolarizing samples
M.M. Long High accuracy Mueller Matrix measurement
Sample None Stage location is below beam for straight-through
Alignment (90°) measurements.
Standard Use for standard reflected measurements. This
setting does a quick search for the light beam near
the previous alignment position.
Robust A thorough alignment of the light beam that
searches over the full translation range. Use for
glass substrates with multiple reflected beams.
Fixed Height Moves sample stage to height described in
hardware configuration file.
Prompt Allows user to enter the sample stage height for
Height measurement. It is often helpful to first perform a
z-stage scan from the Hardware tab.
Model Click “Choose from File Dialog.” All default models
are saved within various folders including
Common, Advanced, Basic, and Calibration Wafers.
Only models in the Common folder will appear in
the drop-down list.

For the “Standard” and “Long” modes, the motor home sensors are checked before
and after measurement. Also, they both implement zone-averaging, which collects
data with the input polarizer set to +45° and -45° and then reports the average.
This improves data accuracy.

The M.M. data modes allow measurement of the first three columns of the Mueller
matrix. The absolute intensity of m11 is measured, while all other measured
elements are normalized to m11.

NOTE: To improve accuracy of Transmission Intensity and Mueller Matrix

measurements, perform an “S-T Baseline” or “Off-Sample Baseline” prior to the
sample measurement (see Hardware tab).

8-316 • Reference – Measurement Tab CompleteEASE Software Manual

 Focus Probes Installed
Use this checkbox when focus probes are installed on the alpha-SE to correct for
lenses in the beam path. Focus probes are an optional accessory.

Model
Specifies which model will analyze the data. Models saved in the Common folder
are listed in the drop-down menu. You can also click ‘Choose From File Dialog’ to
open models from other locations. If “None” is selected, data will be acquired but
not analyzed.

NOTE: A model can also be applied later from the Analysis tab.

Figure 8-2. Model drop-down menu includes .mod files saved in the Common folder.

NOTE: Models can be added to the drop-down list by saving them in the
CompleteEASE\MOD folder or from Analysis>Model>‘Open’ by dragging .mod files
from any location to the Projects tab, Common folder (see Figure 8-3).

Figure 8-3. Click Analysis>Model>‘Open’ to drag-and-drop models into the Projects tab,
Common folder.

CompleteEASE Software Manual Reference – Measurement Tab • 8-317

 Save Data after Measurement
Select this checkbox to ensure the data are saved after each measurement. The
default setting for this option is ON, but it is “sticky” and will retain its new value
until CompleteEASE is turned off.

NOTE: Data can be saved manually after measurement from the Analysis tab.

‘Measure’
Begins a measurement using the settings specified in the “Mode” and “Sample
Alignment” fields. After data is acquired, the model is used to analyze the data.
Results are placed in the box in the upper right corner of the screen.

8.3. Measurement Controls Panel (M-2000 & RC2)

M-2000 and RC2 systems are available in many configurations. Consult your
hardware manual (by pressing F2 on the keyboard when CompleteEASE is open) for
specific details of your system and configuration. This chapter will discuss basic
options for typical systems. Figure 8-4 shows an example Measurement tab for an
M-2000 with mapping and camera accessories included.

Figure 8-4. Measurement tab for an M-2000 system with automated angle of incidence,
automated sample alignment, automated sample translation (mapping), focusing optics, and
an integrated camera.

8-318 • Reference – Measurement Tab CompleteEASE Software Manual

 Recipe
Recipes contain all details of a measurement - what to measure, where to measure,
and how to analyze. Recipes can be created for any specific measurement or
sample type. The three components of a Recipe are 1) Acquisition Parameters, 2)
Scan Pattern, and 3) Model.
The Recipe drop-down menu contains the following choices: Prompt for Acquisition
Parameters, Prompt for Recipe Components, or choose from list of all recipes in the
Common folder. You can also click ‘Choose From File Dialog’ or ‘Create/Edit Recipe’
(see Figure 8-5).

Figure 8-5. Recipe Options.

<Select a Recipe>
This listing is not an option but rather a note to the user that they need to select a
recipe or change this setting before they can proceed.

<Prompt for Acquisition Parameters>

This option will measure at the current location with no analysis. Choose this
option when you want to measure at a single point and save the data for later
analysis. Click ‘Measure’ with this option selected to view the Acquisition
Parameters Setup dialog (see Figure 8-6). Options available here are described in
Section 8.4.

CompleteEASE Software Manual Reference – Measurement Tab • 8-319

 Figure 8-6. Acquisition Parameters Setup dialog box.

<Prompt for Recipe Components>

This option allows the user to select the individual recipe components at time of
measurement. Click ‘Measure’ with this option selected to view the Choose Recipe
Components dialog (see Figure 8-7). Here, the user can use drop-down menus to
choose the Acquisition Parameters, a Scan Pattern, and a Model. If Acquisition
Parameters or Scan Pattern components have not already been created and saved,
new ones can be created from the ‘Edit/Create’ buttons.

Figure 8-7. Choosing the various components for a Recipe.

8-320 • Reference – Measurement Tab CompleteEASE Software Manual

 Common Recipes
All recipes that are saved in the Common folder (C:\CompleteEASE\recipe) will
appear in the drop-down list as shown in Figure 8-5.

‘Choose From File Dialog’

Click this button to browse to recipe files in locations other than the Common
folder.

‘Create/Edit Recipe’
Click this button to create a new recipe or edit an existing recipe.

Save Data after Measurement

Select this checkbox to ensure the data are saved after each measurement. The
default setting for this option is ON, but it is “sticky” and will retain its new value
until CompleteEASE is turned off.

NOTE: Data can be saved manually after measurement from the Analysis tab.

‘Measure’
The ‘Measure’ button begins a measurement according to the recipe or other
option selected from the drop-down list. There are also two right-click options for
the ‘Measure’ button, as shown in Figure 8-8.

Figure 8-8. Right-click on the ‘Measure’ button to choose “Acquire Multiple Measurements”
or “Show Recipe Scheduler”.

Acquire Multiple Measurements

Right-click on the ‘Measure’ button to “Acquire Multiple Measurements”.
CompleteEASE will then prompt for the number of times the measurement
procedure will be repeated (see Figure 8-9).

CompleteEASE Software Manual Reference – Measurement Tab • 8-321

 Figure 8-9. Enter number of measurement repeats.

Show Recipe Scheduler

The Recipe Scheduler is a utility for running multiple recipes automatically without
requiring attention or input from the user while the measurement is running. It is
primarily designed for cases where multiple samples are loaded onto the
translation stage at one time. The scheduler allows each sample to be measured
with its own Recipe. To activate the Recipe Scheduler utility, right-click ‘Measure’
and choose “Show Recipe Scheduler”. The Recipe scheduler supports the following
steps:
• Run Recipe – run a predefined recipe
• Delay – pause for the specified amount of time
• Set Temperature – set the temperature and wait a specified amount of
time
• Move Translator – move the translator to a specified position
• Acquire Dynamic Data – acquire dynamic data for a specified amount of
time
• Move Angle – move angle to specified position
• Run Temperature Profile – run saved temperature profile
• Move Z-stage – move z-stage to specified position
• Change Custom File Text – change the custom file name text prepended to
each data file (defined in 'Five Saving Settings' which is saved after
acquiring a recipe
• Display Message – pause the running of the schedule and display a custom-
defined message to the user
• Start Repeat Loop – start a loop where each step after this will be repeated
a specified number of times
• End Repeat Loop – end the steps included in the repeat loop
• Append and Fit Data – append all the data together that was acquired from
all recipes in the schedule and fit the appended data with the selected
model
• Hardware Check – perform a custom check on specific hardware-related
parameters from the recipe measurements in the schedule and determine
whether the recipe schedule should continue as is, repeat the most recent
recipe in the schedule, or stop running the schedule

8-322 • Reference – Measurement Tab CompleteEASE Software Manual

 Use the Recipe Scheduler (see Figure 8-10) to establish commands in list form. Only
commands that have the "Enabled" checkbox checked will be run in the schedule.

Figure 8-10. Recipe Scheduler dialog box.

Run Recipe
If the step type “Run Recipe” is added (see Figure 8-11), CompleteEASE will prompt
the user to select an existing recipe (see Figure 8-12). The Scheduler will then list
this new step as a row in the Recipe Scheduler (see Figure 8-13).

Figure 8-11. Run Recipe Step.

CompleteEASE Software Manual Reference – Measurement Tab • 8-323

 Figure 8-12. Edit Step to choose the Recipe.

Figure 8-13. Scheduler with single step entered.

After a recipe is added to the Recipe Scheduler, a translator offset can be specified
that offsets the recipe’s translator position by a specified amount. To add a
Translator Offset, double-click in the Translator Offset cell next to the recipe.
CompleteEASE will then prompt for the offset values (see Figure 8-14).

NOTE: For the offset to apply, the recipe must be set up such that the Do Not
Reposition Translator box is not checked.

Figure 8-14. Adding a Translator Offset.

Delay
The Delay step inserts a delay into the schedule. The delay time is specified in
minutes, as shown in Figure 8-15.

8-324 • Reference – Measurement Tab CompleteEASE Software Manual

 Figure 8-15. Adding a delay.

Set Temperature
The Set Temperature step is used to change the temperature of a supported
temperature controller and then wait a specified amount of time for the
temperature to stabilize, as shown in Figure 8-16.

Figure 8-16. Setting Temperature in the Scheduler.

Move Translator
The Move Translator step moves the translator to the specified position, as shown
in Figure 8-17.

Figure 8-17. Adding a Move Translator Step.

Acquire Dynamic Data

The Acquire Dynamic Data step allows the user to specify a model number (as
defined in the buttons of the In situ tab) and an amount of time to acquire data, as
shown in Figure 8-18.

Figure 8-18. Editing a step for dynamic data acquisition.

Move Angle
The Move Angle step moves the angle to specified position.

CompleteEASE Software Manual Reference – Measurement Tab • 8-325

 Figure 8-19. Move Angle in the scheduler.

Run Temperature Profile

The Run Temperature Profile step allows the user to choose a saved temperature
profile from file dialog to add to scheduler.

Figure 8-20. Run Temperature Profile in the scheduler.

Move Z-stage
The Move Z-stage step moves the z-stage to specified position.

Figure 8-21. Move Z-stage in the scheduler.

Change Custom File Text

The Change Custom File Text step allows the user to change the custom text that is
prepended to the names of the data files being saved by the recipes in the schedule
for the remainder of the schedule steps (or until another Change Custom File Text
step is encountered). See the 'File Saving Settings' section below for how to set the
custom prepended text initially.

Display Message
The Display Message step allows the user to have a custom, user-defined message
displayed on the screen in a dialog box. Doing so will pause the running of the
schedule until the user dismisses the message box.

Start Repeat Loop

The Start Repeat Loop step allows the user to define a loop of steps that will be
repeated for the specified number of times. This step defines where in the schedule
the repeat loop will start, meaning all steps between here and the corresponding
End Repeat Loop step after this step will be repeated.

8-326 • Reference – Measurement Tab CompleteEASE Software Manual

 End Repeat Loop
The End Repeat Loop step allows the user to define where a loop of steps will end.
This step is required for a repeat loop to run properly. Repeat loops can be nested.

Append and Fit Data

The Append and Fit Data step allows the user to specify a model that will be used to
analyze all data collected from all the recipes performed previous to this step in the
recipe scheduler. To do this, all data files collected from all prior recipes will be
appended together and then fit with the same model specified in this step. This
step also allows the results of the fit to be exported to a text file in a similar manner
to when the “Export Recipe Data as Text” configuration (in the “Data Export
Parameters” section of the configuration dialog box) is turned on for non-recipe-
schedule measurements.

Hardware Check
The Hardware Check step allows the user to define a check that can be performed
based on the values of parameters relating to the hardware (such as “Z Align”,
“SigInt”, “Tilt X”, “Tilt Y”, “Hardware OK”, etc.). See Figure 8-22for an example of
the dialog box for the Hardware Check step. The hardware check allows the user to
specify a custom equation for the values of a hardware parameter for each previous
recipe measurements. (In the dialog box, “[P1]”, “[P2]”, etc. stand for the values of
the specified parameter for each recipe measured in the schedule, in sequence.)
The custom equation will result in a “Test Value”, which will then be checked
against the “Test Value Range”. If the test value falls within the test value range,
then the hardware check will have passed and the recipe schedule will continue
without warning. If the test value falls outside of the test value range, then the
recipe schedule will pause and a message will be displayed to the user, asking if
they want to continue with the recipe schedule, repeat the previous recipe, or stop
the schedule altogether (see Figure 8-23). This message can optionally contain a
custom user-defined warning message about the significance of this hardware
check.

Figure 8-22. Hardware Check in the scheduler.

CompleteEASE Software Manual Reference – Measurement Tab • 8-327

 Figure 8-23. Hardware Check failure message.

File Saving Settings

The ‘File Saving Settings’ button sets the naming convention for the data files when
running a schedule. Figure 8-24 shows the File Saving Settings dialog. The default
option is “Automatically assign file name - <custom name><step #><time>”. This
option prompts the user for a text string when ‘Run’ is clicked and prepends this
text to the beginning of each automatically generated file name. The automatically
generated file name also consists of the step number and the current time. As
mentioned above, the prepended text can be changed throughout the schedule
using the ‘Change Custom File Text’ step.
The second option is like the first except that the user is not prompted for a text
string. The results are saved with a file name consisting of the recipe name, step
number, and time stamp.
The third option is to manually define the file names. When this option is chosen,
the user must go to each step in the scheduler that requires a file name and double
click on the table cell and choose a file name from the standard Save Data dialog.

Figure 8-24. Assigning file names in the scheduler.

Make Translator Offsets Relative To Initial Translator

Position
When checked, translator offsets will be relative to wherever the translator is when
the schedule is started. When unchecked, the translator offsets will be relative to
the center translator position (0,0).

8-328 • Reference – Measurement Tab CompleteEASE Software Manual

 Minimize This Window while Schedule is Running
When checked, this option will minimize the separate reipce scheduler window to
the taskbar so that it is not visible on the screen. However, if any message needs to
be displayed to the user, such as a custom, user-entered message or a hardware
check alert, then the window will become visible again.

8.4. Acquisition Parameters

The Acquisition Parameters describe how the data will be collected. An example
Acquisition Parameters Setup dialog is shown in Figure 8-25. The individual details
for these options are discussed in the corresponding hardware manual for your
instrument (which can be accessed by pressing F2 on the keyboard when
CompleteEASE is open).

Figure 8-25. Acquisition Parameters Setup for M-2000 and RC2 systems. Choices will vary
from system to system.

Data Type
The Data Type specifies the type of data to collect during the measurement, as
shown in Figure 8-26.

CompleteEASE Software Manual Reference – Measurement Tab • 8-329

 Figure 8-26. Options for Data Type.

Standard
Applies to most applications. This is simply a standard Spectroscopic Ellipsometry
measurement. The data collected will include Psi and Delta versus wavelength and
angle of incidence. In addition, the depolarization and reflected intensity will be
measured. However, the measurement accuracy of depolarization and reflected
intensity depends on the accuracy of recent DC calibrations and intensity baselines,
respectively. For more details, please consult your hardware manual. Most models
will ignore the depolarization and intensity measurements unless configured to
consider these extra data.

Transmission Intensity
Measure intensity of transmitted light through the sample.

Reflection Intensity
Measure intensity of reflected light from the sample.

Generalized Ellipsometry
Generalized ellipsometry measurements include the complete Jones matrix of a
sample, including the off-diagonal elements (which are assumed equal to zero for
Standard SE measurements). This type of measurement is useful for anisotropic
samples where the optical axis is not aligned parallel or perpendicular to the plane
of incidence. Generalized Ellipsometry measurements are not compatible with
measurements that exhibit depolarization, as it is based on a Jones-matrix
description of the interaction between measurement beam and sample.
Generalized Ellipsometry measurements consist of three ratios, so you will end up
with three Psi curves and three Delta curves for each measured angle of incidence.

Mueller Matrix
Measurement of the polarization change using Stokes vectors to describe the light
beam is done with the Mueller Matrix. The Mueller Matrix is a 4 x 4 matrix, but not
all elements can be measured with each ellipsometer. For example, the M-2000
measures 12 elements, the first three rows of the Mueller Matrix. The only
ellipsometer that can measure all 16 elements of the Mueller Matrix is the RC2. For
most samples, there is significant redundancy in the Mueller Matrix and no need to
measure all 16 elements. For isotropic samples (with or without depolarization),
the complete interaction can be described by measuring 4 elements (M11, M21 or
M12, M33 or M44, and M34 or M43). If the sample also contains anisotropy, the

8-330 • Reference – Measurement Tab CompleteEASE Software Manual

 number of necessary elements climbs to 7 or 8. Only for the most advanced
samples are more elements required for complete characterization. Thus, less than
1% of all ellipsometry applications require Mueller Matrix measurements.

Un-Normalized Mueller Matrix

ALL MM measurements are normalized to M11 (even if user selects “Un-
Normalized Mueller Matrix” at time of measurement). The difference is that for
“unnormalized” data, CompleteEASE also measures the intensity information such
that M11 can also be graphed. When “Show M11 Graph” is checked, the scale of
M11 will be 0 to +1, even when the other graphs are between -1 and +1. An
additional benefit of measuring the “unnormalized” Mueller Matrix is that
CompleteEASE can calculate the like- and cross-polarized intensities (pp, ss, ps, sp).

Sample Alignment
Sample alignment options depend on the type of system and configuration. In
general, there are two basic steps to sample alignment: tilt alignment and height
alignment. Common selections for each are described in this section.

Tilt Alignment
The sample surface should be perpendicular to the plane of incidence such that the
reflected beam is aligned on the receiver unit at the correct angle of incidence.
Tilting the sample surface to achieve this condition is required for most samples.
Tilt alignment options include Skip, Manual, or Automatic (see Figure 8-27).

Figure 8-27. Options for Sample Tilt Alignment.

Tilt alignment is sometimes skipped for flat, coplanar samples (if sample stage has
been previously aligned) or when using focusing probes (the measurement is less
sensitive to sample tilt with a small measurement beam).
When tilt alignment is set to manual or automatic, the ellipsometer uses a four-
quadrant detector to adjust the sample tilt until the reflected beam is centered
onto this detector and all quadrants have the same intensity. This method allows
very repeatable alignment of the beam from sample to sample. An example of the
tilt alignment step for an automatic system is shown in Figure 8-28.

CompleteEASE Software Manual Reference – Measurement Tab • 8-331

 Figure 8-28. Sample Tilt Alignment window.

Sample Height Alignment

Sample height alignment refers to the adjustment of ellipsometer beam relative to
the surface of a sample to ensure correct positioning for samples with different
thicknesses. This adjustment becomes increasingly important for systems with
smaller spot size (for example, if the ellipsometer has focus probes). It is also more
important to test the sample height alignment when many samples of different
thickness will be tested on the same ellipsometer. Typical choices for sample height
alignment are shown in Figure 8-29. “Skip” is typically only used if the same sample
was the last sample aligned. “Manual” allows the user to manually adjust the
height until maximum reflected intensity is detected. “Automatic-Quick” performs
a small search around the current height position for maximum reflected intensity,
and “Automatic-Robust” performs a search over the full z-range of the instrument
for the maximum reflected intensity (see Figure 8-30). The position of highest
intensity is assumed to be the correct height for alignment of the sample.

Figure 8-29. Options for Sample Height Alignment.

8-332 • Reference – Measurement Tab CompleteEASE Software Manual

 Figure 8-30. Sample Height Alignment window.

Common Acquisition Parameters

When finished setting the Acquisition Parameters, you can save a *.parms file. If
you want to use the same Acquisition Parameters for future measurements, the file
can be saved to the Projects tab, Common folder (C:/CompleteEASE/recipe) so it
will appear in the drop-down list when selecting your Recipe components.

8.5. Scan Pattern

The Scan Pattern is used to describe where each measurement should be
performed across a sample. This option is only applicable for ellipsometers with
automated sample translation. To make a new Scan Pattern, click ‘Edit/Create’ to
the right of the Scan Pattern selection of the Recipe Components dialog box (see
Figure 8-31).

CompleteEASE Software Manual Reference – Measurement Tab • 8-333

 Figure 8-31. Choose ‘Edit/Create’ Scan Pattern from the Recipe Components dialog box.

The Scan Patter Editor (see Figure 8-32) can be used to create custom patterns.
Choose from Circle or Rectangle for the Substrate Dimensions (or for the pattern
perimeter) and enter the corresponding diameter or dimensions. Next, add points
to the Point List. Points can be added one at a time (click ‘Add Point’) or
automatically filled using ‘Cartesian Grid Fill’, ‘Polar Grid Fill’, or ‘Line Fill’. Points
can also be created in a separate spreadsheet program and pasted into
CompleteEASE using the ‘Paste From Clipboard’ button. Test each option to gain
familiarity with these features. Click ‘Delete All’ to clear the Point List panel as
needed.

Figure 8-32. Scan Pattern Editor.

8-334 • Reference – Measurement Tab CompleteEASE Software Manual

 Scan Region of Interest
For Cartesian Grid Fill, you can “Scan Region of Interest”. Check this box and enter
perimeter coordinates for the region of interest (see Figure 8-33 and Figure 8-34).

Figure 8-33. Scan Region of Interest.

Figure 8-34. Updated Scan Pattern Editor after using Cartesian Grid Fill to Scan Region of
Interest.

The Scan Pattern also describes where the sample alignment procedure will occur.
For example, you can choose to align at the specified Alignment Position (often
center) only. You can also choose to fully align at each point or from other options
shown in Figure 8-35.

CompleteEASE Software Manual Reference – Measurement Tab • 8-335

 Figure 8-35. Alignment Position options.

Common Scan Patterns

When finished creating the Scan Pattern, you can save it into a *.scan file. If you
want to use the same Scan Pattern for future measurements, the file can be saved
to the Projects tab, Common folder (C:/CompleteEASE/recipe) so it appears in the
drop-down list when selecting your Recipe components.

8.6. Fit Results Panel

If a model was selected at the time of measurement (as part of a Recipe or selected
when prompted for Recipe components), results will be displayed in the Fit Results
panel (see Figure 8-36).

Figure 8-36. Fit Results panel in the Measurement Tab.

Right-Click Menu
Right-click in the Fit Results panel to copy results from the panel to clipboard.

Copy to Clipboard – Formatted

Copy results to the clipboard in a format ready to paste into a word processor, as
shown below.

8-336 • Reference – Measurement Tab CompleteEASE Software Manual

 MSE = 2.837
Thickness # 1 = 96.71 ± 0.018 nm
n of Cauchy Film @ 632.8 nm = 1.865

Copy to Clipboard - Table

Copy results to the clipboard in a format ready to paste into a table or spreadsheet,
as shown below.

PARAMETER VALUE ERROR BAR

MSE 2.837
Thickness # 1 (nm) 96.71 0.018
n of Cauchy Film @ 632.8 nm 1.865

‘View Previous Fit Results’

Review results from previous measurements. Click this button to open a new dialog
(see Figure 8-37). First choose the Location of the measurements you wish to
review. A list of measurements will then appear in the Files panel. Click on a file to
display results in the File Information panel. This table can be copied to the
clipboard with the ‘Copy Text to Clipboard’ and ‘Copy Table to Clipboard’ buttons.
If the selected data file is a map of multiple points across the sample, all points are
shown in the list, along with the statistics across the top: average, min, max,
standard deviation, %range, and %uniformity.

% Range
The percent range is defined as:
 MAX − MIN 
% RANGE = 100% 
 AVERAGE 

% Uniformity
The uniformity is defined as:
 STDEV 
%UNIFORMITY = 100% 
 AVERAGE 

CompleteEASE Software Manual Reference – Measurement Tab • 8-337

 Figure 8-37. View Previous Results dialog.

NOTE: You may need to scroll to view fit parameters (see Figure 8-38).

Figure 8-38. Scroll File Information panel to view all available information.

NOTE: Fit parameters will only be available if a model was designated at time of
measurement or if the data was resaved with fit results appended.

NOTE: Alignment information will be shown if the “Include Alignment Parameters

with Data” parameter is turned ON in the Hardware Configuration prior to
measurement (see Figure 8-39).

8-338 • Reference – Measurement Tab CompleteEASE Software Manual

 Figure 8-39. Include Alignment Parameters With Data.

For map data, the ‘Graph Data’ button can be selected to open a new window that
allows users to select the fit result they wish to view (see Figure 8-40).

Figure 8-40. Uniformity map of total thickness.

Previously measured data can be re-analyzed by selecting a file from the Files panel
and clicking ‘Load Data for Analysis’. This button opens the data file, but not any
model previously applied.

CompleteEASE Software Manual Reference – Measurement Tab • 8-339

 CTRL+click to select multiple results in the Files panel and the ‘Calc. Multi-Sample
Stats’ button will become active. Click and choose which values to use for statistics
(see Figure 8-41). Example multi-samples statistics are shown in Figure 8-42.

Figure 8-41. Choose values for statistics.

Figure 8-42. Multiple Sample Statistics, using values at (0.00, 0.00).

8-340 • Reference – Measurement Tab CompleteEASE Software Manual

9. Reference – Analysis Tab

The Analysis tab is shown in Figure 9-1. This tab consists of three panels: Data,
Model, and Fit.

Figure 9-1. CompleteEASE with the Analysis tab selected.

9.1. Data Panel

‘Open’
Opens a data file for analysis. Right-click menu is shown in Figure 9-2 below.

Figure 9-2. The right-click menu for ‘Open’ from the Data panel gives access to “Append
Data”, “Append Intensity Data From Text File”, and a list of up to 10 recently opened files.

CompleteEASE Software Manual Reference – Analysis Tab • 9-341

 Append Data
Appends another data file to the currently open data file. This will put
CompleteEASE in multi-data-set mode. See Table 9-1 below for the combinations of
data that can be appended.

Table 9-1. Types of CompleteEASE data that can be appended to other data.

DATA FILE TYPE EXT DATA FILE TYPES THAT CAN BE APPENDED
Standard Other standard data files and one-point translation
maps can be appended to standard .SE data. (One-
point translation maps will be converted to standard
.SE data files when appended.)
Translation No other data files can be appended to a translation
.SE
(Uniformity Map) data file.
Rotation Other rotation scans and standard data files can be
appended to rotation scans. (Standard data files will
be converted to one-point rotation scans when
appended.)
In-situ Only other in-situ data files can be appended to in-
(Dynamic Data) situ data files; however, appended in-situ data is
added, timewise, to the end of the current in-situ
.iSE
data (with an optional time delay), forming one
composite in-situ data set. This does not put
CompleteEASE in multi-data-set mode.

Append Intensity Data From Text File

Appends unpolarized transmission or reflection data from sources other than a J.A.
Woollam ellipsometer. This is only allowed after first loading or acquiring
ellipsometry data. A file dialog will appear allowing the user to select the desired
file. After appending, this data is treated just like transmission and reflection data
acquired by the ellipsometer. The format for the tab-delimited text file containing
the transmission or reflection data is as follows:

<Comment_Line>
<Data_Type> <Optional_AOI#1> <Optional_AOI#2> <…>
<Wavelength_Units>
<Wavelength> <Data_AOI#1> <Optional_Data_AOI#2> <…>
…

The first three lines in the file describe the content of the data. The first line is for a
comment. The second line describes the type of data in the file. The options are
“uT” for unpolarized transmission data and “uR” for unpolarized reflection data. By
default, the angle of incidence is assumed to be 0°, but additional items on the
second line are used to specify the angle(s). The third line specifies the wavelength
units for the file. The choices are “A”, “nm”, and “eV” for angstroms, nanometers,
and electron volts respectively. The rest of the lines specify the data in ascending
wavelength order. Example data files are shown below:

9-342 • Reference – Analysis Tab CompleteEASE Software Manual

 Normal Incidence Reflection Data File Example
Spectroscopic Reflection Data
uR
nm
600 0.337257
605 0.336592
610 0.335349
615 0.333637
620 0.333814
625 0.333993
630 0.331994
635 0.332227

Normal Incidence Transmission Data File Example

Spectroscopic Intensity Data
uT
nm
600 0.000414
605 0.000856
610 0.000245
615 0.000633
620 0.000226
625 0.000268
630 0.000047
635 0.000290

Multiple Angle Reflection Data File Example

Three Angle Reflection Data
uR 0 10 20
nm
250 0.521862 0.529956 0.551349
255 0.572673 0.578249 0.595404
260 0.617254 0.623597 0.637659
265 0.655014 0.658381 0.674699
270 0.681792 0.687090 0.698921
275 0.704360 0.704930 0.714147
280 0.702201 0.703172 0.714167
285 0.700212 0.703155 0.708637

Recent List
Up to ten recently opened data files are stored in a “Recent List”, accessed by right-
clicking on the ‘Open’ button. See Figure 9-2 for an example.

‘Save’
Saves the current data file. CompleteEASE data files are saved as encrypted files
that can only be opened with CompleteEASE. The file extensions are “.SE” for
standard measurements or “.iSE” for dynamic measurements. The right-click menu
with options for “Save Data Subset” and “Export to Text File” is shown in Figure 9-3.

Figure 9-3. The right-click menu for ‘Save’ from the Data panel gives access to “Save Data
Subset” and “Export to Text File”.

Save Data Subset

Saves a subset of the currently open data set. The data subset will consist of the
reduced data that are currently selected (wavelengths, angles, timeslices, multiple

CompleteEASE Software Manual Reference – Analysis Tab • 9-343

 points from a map, etc.). The options available depend on the type of data currently
open. Figure 9-4 shows the “Save Subset” dialog box that allows data at certain
wavelengths to be skipped when saving a data subset.

Figure 9-4. The “Save Subset” dialog box that allows the data subset to omit the data at
certain evenly-spaced wavelengths by specifying a wavelength resolution in nm or eV.

For a translation map, if no individual points are selected in multi-data-set mode,

saving a subset will save all of the data that is visible on the translation-map graph.
This means that a user can zoom into a subregion of a translation map and then use
the “Save Data Subset” option to save another translation map with only the
zoomed-in portion included. If some individual points on a translation map are
selected in multi-data-set mode, then the options for saving a subset will relate to
these selected points. Figure 9-5 shows the options for saving a data subset when
multiple specific translation map points are selected. “Save all selected points into a
multiple data set file” saves all selected map points into one multi-data-set file as
appended data sets. “Save each selected point into its own data file” saves an
individual standard data file for each map point selected.

Figure 9-5. The “Save Subset” dialog box with options for saving a data subset when multiple
translation map points are selected in multi-data-set mode.

For a dynamic data set, if no individual timeslices are selected in multi-data-set

mode, saving a subset will save all of the data that is visible on the dynamic graph.
This means that a user can zoom into a subregion of the dynamic data and then use
the “Save Data Subset” option to save another dynamic data file with only the
zoomed-in region included. If some individual timeslices are selected in multi-data-
set mode, then the options for saving a subset will relate to these selected
timeslices. Figure 9-6 shows the options for saving a data subset when multiple in-
situ timeslices are selected. “Save all selected points into a multiple data set file”
saves all selected timeslices into one multi-data-set file as appended data sets.
“Save each selected point into its own data file” saves an individual standard data
file for each timeslice selected. “Save all selected points as timeslices in an in-situ
file” saves an in-situ data file that only contains the timeslices selected. “Remove
selected timeslices and save remaining as an in-situ file” will save an in-situ data file
that retains only the timeslices that are not selected.

9-344 • Reference – Analysis Tab CompleteEASE Software Manual

 Figure 9-6. The “Save Subset” dialog box with options for saving a data subset when multiple
in-situ timeslices are selected in multi-data-set mode.

Export to Text File

Exports data sets to a specified text-file format. Multiple data sets can be exported
at once as data sets are selected from the file dialog and do not need to be
currently open in CompleteEASE to be exported. Data sets can be exported to a tab-
delimited “Table format” for use in a spreadsheet program, to the “WVASE .DAT”
format for use with WVASE®, or to the “NanoDiffract .dat” format for use with
NanoDiffract®, a software package provided by Onto Innovation™.

‘Info’
Shows information about the current data set opened in CompleteEASE. An
example of the “Exp. Data Information” dialog box is shown in Figure 9-7.

Figure 9-7. Information box showing details of data measurement.

‘Set Ranges’
Choose the wavelength range and selected angles for use with the current data set
during fit. This also allows the selection of selected translation point from
uniformity maps, as shown in Figure 9-8. When uniformity maps or rotation scans
are graphed, this dialog box also allows the choice of the selected point and the
graph angle and wavelength. When in-situ data are graphed, this dialog box also
allows selection of the time range and the selected timeslice.

CompleteEASE Software Manual Reference – Analysis Tab • 9-345

 Figure 9-8. Press ‘Set Ranges’ to select the wavelengths, angles, translation points and
graphing angles/wavelengths in the “Select Data Range” dialog box. The “Select Data Range”
dialog box shown here is for a translation map.

9.2. Fit Panel

‘Generate’
Generates data based on the current model and data set. (Generated data is what
the data would look like if the sample measured exactly matched the ideal model
description.) The shortcut keystroke to generate data is Alt+G.
Right-clicking on the ‘Generate’ button shows the “Clear Generated Data” option,
which removes the generated-data curves from the graph.

‘Fit’
Fits the model-generated data to the experimental data based on the current
model and data set open. Pressing ‘Fit’ starts a regression analysis in which model
fit parameters selected for fitting are varied to match the model-generated data to
the experimental data. A warning message will be shown if no fit parameters are
selected for fitting in the current model or for other common modeling issues that
can be encountered. For translation-map data, dynamic data, and rotation data,
only the currently graphed data set will be fit using the current model (unless
otherwise defined within the model options). When CompleteEASE is in multi-data-
set mode, all currently selected data sets will be matched using the current model.
The shortcut keystroke to fit data is Alt+F.

‘Fit Scan Data’

When a data set with a uniformity map is open, the third button in the Fit panel will
read ‘Fit Scan Data’, which fits each point from the uniformity map in sequence. If
any map points are selected in multi-data-set mode, ‘Fit Scan Data’ will be
unavailable. It is important to clear multi-data-set mode (by right-clicking any point

9-346 • Reference – Analysis Tab CompleteEASE Software Manual

 on the map and choosing “Unselect All” or using the “Clear Multi-Data Set Mode”
option in the +Data Sets drop-down menu above the graph after unchecking “Show
Map Data”) before ‘Fit Scan Data’ can be used.

‘Fit Dynamic’
When a dynamic (in-situ) data set is open, the third button in the Fit panel will read
‘Fit Dynamic’, which fits each timeslice from the dynamic data set starting with the
first timeslice in the selected time range and moving sequentially to the last
timeslice in the selected time range.
Right-clicking ‘Fit Dynamic’ gives options to “Append Fit Results”, “Fit Backwards in
Time”, or “Fit Backwards and Append”.

Append Fit Results

When fitting dynamic data, it can sometimes be helpful to fit different regions of
time separately. Thus, this option allows the use of different models for different
parts of the dynamic data set, for example, when growing multi-layer stacks. This
option can be used after the first region has already been fit, allowing a second
range to be fit and appended to the previous fit results. The dynamic fit results for
the currently selected range are analyzed using the current model and selected fit
parameters and then appended to the results of the previously analyzed parts of
the dynamic data set.

Fit Backwards in Time

This option allows the dynamic data to be fit backwards in time, starting from the
last timeslice in the selected time range and moving sequentially to the first
timeslice in the selected time range. Use this option for dynamic data where the
“end” result is better understood than at the beginning of the experiment. For
example, it can be used for etching studies, where the thickness at the last timeslice
is easier to determine.

Fit Backwards and Append

This feature combines the previous two options. It will perform the fit backward
starting at the last selected timeslice and moving to the first selected timeslice, but
then the results will be appended to previous dynamic-data fit results.

‘Fit Rotator Data’

When a rotation data set is open, the third button in the Fit panel will read ‘Fit
Rotator Data’, which fits each rotation position in the data set separately in
sequence. If any rotator positions are selected in multi-data-set mode, ‘Fit Rotator
Data’ will be unavailable. It is important to clear multi-data-set mode (by
unchecking “Show Rotation Data” and selecting “Clear Multi-Data Set Mode” from
the +Data Sets drop-down menu) before ‘Fit Rotator Data’ can be used. Oftentimes,
however, ‘Fit Rotator Data’ is rarely used; it is usually better and more
straightforward to use the option to “Analyze All Rotation Angles” found in the
+MODEL Options section of the model when rotation data is loaded. Refer to
Section 9.6 for a description of this feature and how it compares to the functionality
of the ‘Fit Rotator Data’ button.

CompleteEASE Software Manual Reference – Analysis Tab • 9-347

 Graphs of Fit Results
After ‘Fit Scan Data’, ‘Fit Dynamic’, or ‘Fit Rotator Data’ have been pressed, the
model will be fit to all the map points, timeslices, or rotation positions in the data
sets, respectively. This means that the fit parameters in the model may have
different values for different positions in the translation, dynamic, or rotation data.
In order to graph the individual fit parameters or the MSE over the whole set of
data (for each map point, timeslice, or rotation position), simply click on the blue,
underlined text of the fit parameter (e.g. Thickness #1) shown in the Fit panel. For
example, after fitting a uniformity-map data set, the fit parameters of the Sellmeier
layer and any derived parameters in the model are displayed in blue, underlined
text in the Fit panel, as shown in Figure 9-9. Click on any of these parameters to
view a graph of the fit results for the parameter over all the uniformity-map
positions, as shown in Figure 9-10. These same graphs can also be found in the
+Results menu (described in Section 7.6) that appears directly above the graph
area, which is activated whenever ‘Fit Scan Data’, ‘Fit Dynamic’, or ‘Fit Rotator Data’
have been pressed.

Figure 9-9. All fit parameters and derived parameters are displayed as blue, underlined text
after ‘Fit Scan Data’, ‘Fit Dynamic’, or ‘Fit Rotator Data’ are pressed. Click on these to show a
graph of the fit-parameter results over all map points, timeslices, or rotation positions.

Figure 9-10. An example of graphing a fit parameter, in this case “Thickness #1”, over all
points in a uniformity map.

9-348 • Reference – Analysis Tab CompleteEASE Software Manual

 ‘Reset’
Restores the values of all previous model parameters from before the most recent
fit. When the model is reset, generated-data curves on the graph are redrawn to
reflect the newly reset model’s state. An error message will be shown if the number
of fit parameters changed from the last time the ‘Fit’ button was pressed, for
example, by unselecting a fit parameter.
Right-clicking on ‘Reset’ will display the Reset List. This provides a list of reset
model-fit states that have been saved automatically by CompleteEASE before and
after each fit since the most recent data/model combination has been in use. An
example of the Reset List is shown in Figure 9-11. This functionality for the model is
somewhat equivalent to pressing the “undo” button repeatedly in word-processing
software.
Selecting a previous state on the Reset List can be done by double-clicking on the
previous state or by selecting the previous state and clicking ‘Ok’. When a previous
state is chosen from the Reset List, generated-data curves on the graph are redrawn
to reflect the newly reset model’s state.

Figure 9-11. The Reset List allows the user to reset to any recent model states, which are
automatically saved before and after each fit.

9.3. Fit Panel Right-Click Menu

Right-click the mouse within the Fit Panel text area to access the menu options as
shown in Figure 9-12. As seen there, the selections can vary depending on whether
one is fitting a single-point measurement, a uniformity map, or dynamic data.

Figure 9-12. Right-click options from the Fit panel after a fit to standard data and uniformity-
map data, respectively.

CompleteEASE Software Manual Reference – Analysis Tab • 9-349

 Auto Fit
When this checkbox is selected, a fit of the model to the data will happen
automatically when the measurement of new data is completed or when a new
timeslice in dynamic data or a map point in uniformity data is selected by clicking
on the graph. (Even with the “Auto Fit” checkbox checked, a model fit does not
happen automatically when the “Timeslice” or “Position” slider is moved to select a
new timeslice or map point.) When unchecked, the model fit only begins after the
user presses the ‘Fit’ button.

View Fit Stats

After each fit, there are statistical values available that include the 90% confidence
intervals on each fit parameter and a correlation matrix which shows the
correlation between all fit-parameter combinations, as shown in Figure 9-13.
Correlation values close to +1 or –1 show that the two parameters are most likely
achieving the same purpose and are not unique. Positive correlation indicates direct
variation (when one parameter increases, the other also increases), and negative
correlation indicates inverse variation (when one parameter increases, the other
decreases). High correlation values might indicate that there is not enough
sensitivity for both correlating parameters, perhaps due to unnecessary complexity
in the model. Correlations above 0.9 (in other words, 90%) can be highlighted in
yellow by checking the “Highlight Correlation Above 90%” checkbox.

Figure 9-13. “View Fit Stats” shows statistics such as the 90% confidence intervals in the “Fit
Parameters” section on the left side and the correlation matrix values after a fit on the right
side. High correlation values (above 90%) are highlighted in the correlation matrix.

View Parameter Stats

After fitting a uniformity map, this option allows you to view the statistics
calculated over all points in the map, along with the values from each location, as
shown in Figure 9-14.

9-350 • Reference – Analysis Tab CompleteEASE Software Manual

 Figure 9-14. “View Parameter Stats” for all points of a uniformity map.

Copy to Clipboard – Formatted

After fitting standard single-point data, the fit results of parameter values can be
copied to the clipboard in a text-formatted layout to be used by word processors or
presentations.

Copy to Clipboard – Table

After fitting standard single-point data, the fit results of parameter values can be
copied to the clipboard in a tabulated layout (tab-delimited) to be used by
spreadsheets.

Copy Parameters to Clipboard

After fitting points in a uniformity map, the fit results from all points can be copied
to the clipboard in a tabulated layout (tab-delimited) to be used by spreadsheets.

Copy Analysis Report to Clipboard

After fitting, the results can be easily placed into an automatically generated
analysis report. The report is generated in RTF, which stands for “Rich Text Format”.
This format is used because it is compatible with copying and pasting into common
word-processing software. The layout of the report will depend on the type of data
analysis being performed. For example, a single-point analysis will appear as shown
in Figure 9-15. The inclusion of the optical constants graph for the model in the
report is optional. This option, called “Add Optical Constants to Report”, can be
found in the +OTHER Options section of the Model panel. After analysis of a
uniformity map, the graphs from the map are copied as part of the report, rather
than the spectral fit quality, as shown in Figure 9-16. Finally, the graphs from each
spectral fit will all be copied if a multi-sample analysis has been performed, as
shown in Figure 9-17. The keyboard shortcut to create the Analysis Report is
CTRL+R.

CompleteEASE Software Manual Reference – Analysis Tab • 9-351

 Figure 9-15. Analysis Report from a single-point fit, with optical constants for the graded
layer added (from +OTHER Options in the Model panel).

Figure 9-16. Analysis Report from a uniformity map.

9-352 • Reference – Analysis Tab CompleteEASE Software Manual

 Figure 9-17. Analysis Report from a multi-sample analysis.

Add to Fit Log

Adds the current fit results to the Fit Log for later viewing. CTRL+L is the shortcut to
add fit results to the Fit Log.

View Fit Log

Displays the Fit Log, which lists any fit results recorded there by the user (by
selecting “Add to Fit Log”) since CompleteEASE was started, as shown in Figure
9-18. The Fit log is cleared when CompleteEASE is closed, but it can be saved and
opened at a later time by selecting the ‘Save Log’ and ‘Open Log’ buttons in the Fit
Log dialog box. The keyboard shortcut to view the Fit Log is CTRL+ALT+L.

Figure 9-18. Fit Log dialog box.

CompleteEASE Software Manual Reference – Analysis Tab • 9-353

 ‘Rename’
Gives a new name to the selected log entry.

‘Delete’
Deletes the selected log entry from the Fit Log.

‘Compare’
Compares statistics of multiple selected log entries, as shown in Figure 9-19. From
this comparison dialog box, you can “Add Statistics”, “Reverse Columns/Rows” and
even ‘Compare Optical Constants’, which brings up a new dialog box as shown in
Figure 9-20. (Section 3.6 contains an example using ‘Compare Optical Constants’.)

Figure 9-19. Fit Log comparison shows results from each fit log entry.

Figure 9-20. ‘Compare Optical Constants’ brings up a dialog box that shows each model and
allows you to choose the different layers to graph from each model.

‘Reanalyze’
Since some data files from CompleteEASE contain not only the experimental data
but also the results of the analysis of that data using a model (for example, when
using a measurement recipe including a model), the Reanalyze Data feature is used

9-354 • Reference – Analysis Tab CompleteEASE Software Manual

 to reanalyze experimental data using a different model than the one that was used
to create the results saved in the data files. (To see an example of using the
Reanalyze Data feature, refer to Section 3.6.)
The Reanalyze Data dialog box is shown in Figure 9-21. It uses a table for specifying
analysis steps, consisting of a File Group, a Model, and a Log Name. A File Group is
a list of files to be reanalyzed. It can consist of one or more files in the same
directory. The model is a model file that has been saved to the computer hard
drive. The Log Name is a text string that will be appended to the data file name
when adding the results of the reanalysis to the Fit Log.

Figure 9-21. Reanalyze Data dialog box.

To add an analysis step to the table, press ‘Add Row’. This brings up the Choose
Files dialog, as shown in Figure 9-22. By default, the only file group available is the
“Selected Log Entries” group which contains the files from any fit-log entries that
were selected before pressing the ‘Reanalyze’ button. To add a group of files to the
available file groups, press the ‘Add’ button. This brings up the standard file dialog
except that it allows multiple files to be selected. After selecting the desired files
and closing the dialog, you are given the chance to name the file group, as shown in
Figure 9-23. Note that if only one file is selected for a group, then the name is
automatically set to the name of the file. After giving the group a name, it now
shows up in the Choose Files dialog, as in Figure 9-24. The file groups that are
created will be available until CompleteEASE is closed.

Figure 9-22. Choose files to add to groups for reanalysis.

CompleteEASE Software Manual Reference – Analysis Tab • 9-355

 Figure 9-23. Enter a file group name.

Figure 9-24. The new group name shows up with the associated files.

To finish adding the row, choose a file group from the Choose Files dialog and press
‘Ok’. A row is now displayed in the reanalysis table with the file group specified, as
shown in Figure 9-25. A different group can be specified by double-clicking on the
cell you wish to change. This also applies to the Model and Log Name items. To
finish setting up the reanalysis, specify a model and log name by double clicking the
appropriate cell and entering these parameters, as demonstrated in Figure 9-26.

Figure 9-25. File Groups show ups as a row in the Reanalysis dialog box.

Figure 9-26. Click to add Models and optional Log Names for each row.

9-356 • Reference – Analysis Tab CompleteEASE Software Manual

 Once the analysis is set up, the ‘Reanalyze’ button can be pressed to start the
reanalysis. CompleteEASE will analyze each of the files and add the results to the
current fit log. An example of the fit log after a reanalysis is shown in Figure 9-27.

Figure 9-27. Fit Log after reanalysis, with each file from the file group added as a log entry.

‘Generate Reports’
Copies the results of the selected log entries into a report in RTF (“Rich Text
Format”), which is saved to a file. Generating reoprts will clear the currently open
model and data set. (For an example using ‘Generate Reports’, refer to Section 3.6.)

‘Load Log Entry’

You can select a log entry from the list and open this result for analysis in the main
window of CompleteEASE by pressing ‘Load Log Entry’. This will replace the
currently open data set, model, and fit results with the ones from the selected log
entry. The same can be accomplished by double-clicking on a log entry.

‘Clear Log’
Clears the fit log of all its entries.

‘Open Log’ / ‘Save Log’

Opens a saved fit log from your computer or saves the fit log currently open to your
computer. This is a convenient way to save results on your computer, but it
requires that the data files are located in a directory reachable by the computer on
which the fit log is saved. This is because the data files are not saved as part of the
fit log file itself; only a path to each data file (such as “C:\CompleteEASE\DAT\
SampleData.SE”) is saved to the fit log file. If the saved fit log file is transferred to
another computer or if the data files are moved from their current location, the link
between the saved fit log and its experimental data files will be broken. Thus, it is
usually best practice to keep the data files and the associated saved fit logs in the
same directory with each other.

‘Close’
Closes the Fit Log dialog box.

CompleteEASE Software Manual Reference – Analysis Tab • 9-357

 Advanced Graph Options
After fitting a uniformity map or dynamic data, there are “Advanced Graph
Options”, as shown in Figure 9-28, which allow the graphing of various
combinations of parameters on an X-Y graph. This allows graphical interpretation of
possible correlation, graphing parameters over external quantities (such as
temperature or partial pressure), and other relationships between parameters. An
example using dynamic temperature-based data is shown in Figure 9-29.

Figure 9-28. After fitting uniformity-map data or dynamic data (shown above), there are
“Advanced Graph Options”, which allow various combinations of parameters.

Figure 9-29. Using Advanced Graph Options, the thickness is plotted versus temperature from
a heat-cell measurement after a dynamic data analysis.

Graph OCs vs. Time / Graph OCs vs. Position

This choice allows the spectroscopic optical constants to be graphed versus time for
dynamic data or versus position for translation data. When selected, you will be
asked for the type of optical constant graph (as shown in Figure 9-30), the layer
number, and for dynamic data, the total number of curves to draw (as shown in
Figure 9-31). These curves will be equally spaced versus time across the selected
range of the data analysis. For translation data, all positions on the uniformity map
will be included. An example of dynamic data is shown for the index of refraction of
amorphous silicon growth in Figure 9-32.

9-358 • Reference – Analysis Tab CompleteEASE Software Manual

 Figure 9-30. Choose to graph optical constants such as n, k, ε1, or ε2 or combinations thereof.

Figure 9-31. Enter number of curves to draw, which will be equally spaced over the fit time or
position range.

Figure 9-32. Example of 10 index graphs from equally spaced sections of a dynamic fit.

9.4. Model Panel

The panel in the right half of the Analysis tab displays all the options and
specifications for the current data analysis model. An example model is shown in
Figure 9-33. Details of the open and save functions at the top of the model panel as
well as options within the basic multilayer model are given below.

CompleteEASE Software Manual Reference – Analysis Tab • 9-359

 Figure 9-33. Basic Model Panel in the Analysis Tab.

‘Open’
Clicking on the ‘Open’ button opens a new model selected from the file dialog box.
Right-clicking on ‘Open’ gives the option to merge models together and shows a list
of up to ten models that have been recently opened by CompleteEASE, as shown in
Figure 9-34.

Merge Models
Right-click on the ‘Open’ button to merge models together. This is usually used with
in-situ applications, where fit parameter values from one model need to be
transferred to another model that contains additional layers. (Note: To perform a
merge correctly, fit parameter and layer names must be identical in order for
parameter values to merge correctly.)

Figure 9-34. Right-click on model ‘Open’ button to either “Merge Models” or get a list of
recently opened models.

‘Save’
Clicking on the ‘Save’ button saves the current model and all model settings to a
.mod file. Right-clicking on ‘Save’ gives the option to “Save Changes” to the already
open model file. In CompleteEASE, clicking on ‘Save’ is analogous to what is
sometimes called “Save As…” in other programs, and right-clicking on ‘Save’ is
analogous to what is typically called simply “Save”.

Save Changes
Right-click on ‘Save’ to just save the changes made to the model file that is already
open instead of saving the entire model to a new file.

9-360 • Reference – Analysis Tab CompleteEASE Software Manual

 ‘Clear’
Removes the currently open model and reverts back to the standard blank
multilayer model. Note that clearing the model will lose all changes and model
settings in the current model unless they have been saved to a model file.

‘Open Snapshot’
Clicking on ‘Open Snapshot’ opens a snapshot file. A snapshot is a single file that
compresses all information from the CompleteEASE screen, including the data
measurement, model, and fit results. This is a convenient way to save and store all
information in order to restore CompleteEASE to a previous state at a later time.
Right-clicking on ‘Open Snapshot’ shows a list of up to ten recently opened
snapshots to open.

‘Save Snapshot’
Clicking on ‘Save Snapshot’ saves a snapshot file, which is a single file that
compresses all information from the CompleteEASE screen, including the data
measurement, model, and fit results. This is a convenient way to save and store all
information in order to restore CompleteEASE to a previous state at a later time.
Because it includes all the data, the snapshot can be overwhelming in size if created
from large uniformity-map data. Snapshot files have the “.SESnap” file extension.

Layer Commands: Add

Press this button to add a layer to the model stack. The “Add Layer To Model”
dialog box appears as in Figure 9-35. Select the position of the new layer by moving
the mouse to adjust the blue insertion bar and press the mouse button. (This can
also be done by using the arrow keys on the keyboard to move the blue insertion
bar and pressing ‘Enter’ on the keyboard to select.) If the layer type is set to
“Intermix” before pressing the blue insertion bar in place, the layer added will be an
EMA mixture between the layers below and above. The keyboard shortcut to add a
layer by viewing this dialog box is ALT+=.

Figure 9-35. “Add Layer To Model” dialog box.

NOTE: Holding the CTRL key down when clicking the Add button will automatically
add the new layer on top of the current model layer stack rather than showing the
“Add Layer To Model” dialog box and asking for the position.

CompleteEASE Software Manual Reference – Analysis Tab • 9-361

 NOTE: With “Include Substrate Backside Correction” turned on, it is possible to add
layers below the substrate.

Layer Commands: Delete

Press this button to delete a layer from the model. Choose the layer with the
mouse (making it blue) and click to delete as shown in Figure 9-36. (This can also be
done by using the arrow keys on the keyboard to select the layer and pressing
‘Enter’ on the keyboard to delete.)

Figure 9-36. “Delete Layer From Model” dialog box.

Layer Commands: Save

Press this button to save the optical constants for a layer. Before clicking on the
layer you wish to save in the “Save Layer Optical Constants” dialog box as shown in
Figure 9-37, choose the “Save Type”, either ‘Dispersion Parameters’ or ‘Tabulated’.
(Selecting the layer to save can also be done by using the arrow keys on the
keyboard to select the layer and pressing ‘Enter’ on the keyboard to save.)
Saving a layer as a tabulated file saves a .mat file that contains a table of values that
has wavelengths and associated optical-constant values. By default, a tabulated
.mat file is saved with wavelength in angstroms and optical constant values in ε1
and ε2. These settings can be changed in the CompleteEASE configuration with the
“Tabulated Mat File Save Mode” setting in the “General” section. Saving a layer as
dispersion parameters (also called a parameterized file) saves a .mat file that
contains only the values of the parameters in the layer’s dispersion equation.

NOTE: Layers from .mat files that are already tabulated cannot be saved as
“Dispersion Parameters” because they are not described by a dispersion equation
to begin with. Also, anisotropic layers (such as the Biaxial layer) cannot be saved as
tabulated, since they require multiple directions to describe their optical constants.

NOTE: Saving a layer’s optical constants can also be done by right-clicking on the
layer in the main model panel and clicking “Save Layer Optical Constants”; for
certain layers (such as the B-spline), there may be more options than just
“Dispersion Parameters” and “Tabulated”. See Section 9.5 for more details.

9-362 • Reference – Analysis Tab CompleteEASE Software Manual

 Figure 9-37. “Save Layer Optical Constants” dialog box.

Include Surface Roughness

Turn this on to fit a thin layer of roughness on the surface of the existing model. A
surface with roughness is modeled with a Bruggeman Effective Medium
Approximation mixing the top film with 50% void. The only fit parameter available
is the thickness of the rough surface. If the thickness moves to negative values, the
roughness layer is then equivalent to -50% void mixing, which increases the surface
region optical properties (rather than lowering them). The surface roughness layer
is a “stealing” layer, meaning that it takes half of its thickness from the layer below.
Thus, if a 100-nm Cauchy layer has 20 nm of surface roughness on top, the model is
actually calculated as 90 nm of the Cauchy layer with 20 nm of surface roughness.
(See Section 3.1 for further explanation of surface roughness and Section 3.5 for an
example using surface roughness.)

Pictorial Model Layer Stack

The model is shown as a layer-by-layer picture with details regarding thickness and
other parameters to describe the individual layers. Any layer with a “+” to the left
can be expanded to show more details about that layer. This is demonstrated in
Figure 9-38. More detail about the model layer stack is given in Section 9.5, and
each type of layer is described in Chapter 10.

(a) (b)
Figure 9-38. Model with layers collapsed (a) and with Cauchy layer expanded (b).

Model Panel Right-Click Menu

Right-click within the Model panel to access menu options shown in Figure 9-39.

Figure 9-39. Right-click menu from within the Model Panel.

CompleteEASE Software Manual Reference – Analysis Tab • 9-363

 Copy Model To Clipboard
This option copies the entire model to the clipboard, including the +MODEL
Options, +FIT Options, and +OTHER Options that are saved along with the layer
stack. To view the details from one of these sections on the clipboard image, such
as +MODEL Options, it needs to be expanded by pressing the “+” next to that
section before copying to the clipboard. An example of this “full” model is shown in
Figure 9-40.

Figure 9-40. “Full” model as copied to the clipboard.

Copy Model (Layers Only) To Clipboard

This option copies only the layer stack of a model to the clipboard. If you prefer to
see a layer collapsed or expanded, make sure to pre-select your view for each layer
present in the model layer stack before copying. This feature is commonly used to
copy the model structure into a word processor, slideshow presentation, or
publication manuscript. An example of only the model layer stack being copied to
the clipboard is shown in Figure 9-41.

Figure 9-41. “Layer-only” model as copied to the clipboard.

Alternate Model Panels (Model #2 & Model #3)

Alternate model panels are available, allowing CompleteEASE to hold multiple
models at once. Holding down the CTRL key and pressing the F1, F2, and F3 keys
can toggle through these other model panels. (Also, right-clicking on the Model
panel title will show a right-click menu to access them.) This feature is commonly

9-364 • Reference – Analysis Tab CompleteEASE Software Manual

 used when desiring to test multiple independent models with the same data set,
when using the Retardance model, or when performing porosimetry experiments
using the Pore Size model (Section 9.9).

9.5. Model Layer Stack

The model is shown as a layer-by-layer picture with details regarding thickness and
other parameters to describe the individual layers. Any layer with a “+” to the left
can be expanded to show more details about that layer. This is demonstrated in
Figure 9-42. This section gives information about the model layer stack, including
the options in the right-click menu for each layer as well as options for fit
parameters. The options described here are common to most types of layers.
Specific descriptions of each type of individual layer are given in Chapter 10.

(a) (b)
Figure 9-42. Model with layers collapsed (a) and with Cauchy layer expanded (b).

Editing Layers (Layer Right-Click Menu)

Each layer has a name shown in blue, underlined text (for example, Cauchy). Click
on the layer name to open a new material file in place of the current layer in that
position. Right-click on the layer name to see a menu of options relating to that
layer, which include graphing its optical constants, changing its name, saving its
optical constants to a file, or converting the layer to a different but related layer.
This menu is shown in Figure 9-43.

Figure 9-43. Right-click on a layer name to access this menu of options.

Graph Layer Optical Constants

Right-click on a layer to select this option which will display a graph of the layer’s
optical constants. This graph will be shown as either n and k values or ε1 and ε2
values depending on which configuration is selected in the Options tab. The optical
constants of a layer can also be graphed by selecting the layer name in the +Model
menu above the graph.

CompleteEASE Software Manual Reference – Analysis Tab • 9-365

 Graph Layer Absorption Coefficient
Right-click on a layer to select this option which will graph the absorption
coefficient (sometimes called the attenuation coefficient) for the layer. The
absorption coefficient (α) is derived from the layer’s optical constants and is
present in Beer’s Law (also known as the Beer-Lambert Law) in Equation 9-1, where
I is the intensity of light at a distance d from the layer’s surface and I0 is the initial
intensity at the layer surface:
Equation 9-1

𝐼 = 𝐼0 𝑒 −𝛼𝑑
The absorption coefficient α is then described by Equation 9-2, where k is the
extinction coefficient (imaginary part of the complex refractive index) and λ is the
wavelength:
Equation 9-2

4𝜋𝑘
𝛼=
𝜆
Clicking on this option will bring up a dialog box as shown in Figure 9-44, which
allows you to select whether you want to graph the absorption coefficient or the
logarithm of the absorption coefficient and which units (cm, mm, or μm) you want
to use for the reciprocal wavelength.

Figure 9-44. Graphing the absorption coefficient allows the selection of standard or
logarithm graphing and reciprocal wavelength units.

Graph Layer Optical Constants (anisotropic differences)

Right-click on an anisotropic layer (such as the Biaxial layer) to select this option,
which will bring up one of two dialog boxes, as shown in Figure 9-45. For layers with
uniaxial anisotropy, the differences between the E X and EZ components, EX – EZ, can
be graphed. For layers with biaxial anisotropy, the in-plane differences between the
EX and EY components, EX – EY, can be graphed along with the out-of-plane
differences option that can be selected. The retardance of the layer can also be
graphed from here. The retardance is calculated by Equation 9-3, where δ is the
retardance, c is a unit-conversion coefficient (see

Table 9-2 below for different values of this coefficient), t is the thickness in
nanometers, nX is the index of refraction in-plane (the ordinary direction), and nZ is
the index of refraction out-of-plane (the extraordinary direction):
Equation 9-3

𝛿 = 𝑐 ∙ 𝑡 ∙ |𝑛𝑋 − 𝑛𝑍 |

9-366 • Reference – Analysis Tab CompleteEASE Software Manual

 Table 9-2. Unit-conversion coefficient values for calculating retardance.

UNITS UNIT-CONVERSION COEFFICIENT

360°
Degrees 𝑐= where λ is the wavelength in nanometers
𝜆
2𝜋
Radians 𝑐= where λ is the wavelength in nanometers
𝜆

Nanometers 𝑐=1
1
Waves 𝑐= where λ is the wavelength in nanometers
𝜆

a) b)
Figure 9-45. “Graph Anisotropic Differences” dialog box for uniaxial anisotropy (a) and for
biaxial anisotropy (b), which allows the graphing of various differences between the different
anisotropic optical-constant directions and the retardance.

Graph Sub-Layer Optical Constants (e1) / (e2)

Right-click on a graded layer to select this option, which will graph the optical
constants of each individual slice of the graded layer for either ε1 or ε2. See Figure
9-46 for an example.

Figure 9-46. Graph of “sub-layer” ε2 dielectric functions, which displays ε2 for each slice of the
graded layer.

Rename Layer and Fit Parameters

Right-click on a layer to select this option which will bring up a dialog box, as shown
in Figure 9-47. There are three options for renaming the layer; you can select any
combination of these options. “Rename Layer” will simply rename the layer as
shown after the “Layer # X = ” text. “Rename Layer Parameters” will rename all the
layer fit parameters, except the thickness, by appending the layer name to the end
of each layer fit parameter name. (For example, the parameter “A” in a Cauchy
layer named “My Cauchy Layer” would become “A_My Cauchy Layer”.) This option
is useful for differentiating layers of the same type by giving parameters in different

CompleteEASE Software Manual Reference – Analysis Tab • 9-367

 layers unique names, which is required when using advanced model features such
as certain Derived Parameters, Custom User Equations, Parameter Coupling, and
Multi-Sample Analysis. However, note that renaming fit parameters will break
existing references to those fit parameters used in Derived Parameters, Custom
User Equations, Parameter Coupling, and Multi-Sample Analysis. Therefore, use
caution when renaming fit parameters, and remember that it is best to rename
layer and fit parameters before using these advanced features. The last option,
“Rename Thickness”, will rename the thickness fit parameter of the layer by
removing the number and adding the layer name to the beginning of the thickness
fit parameter name. (For example, a layer in position 1 called “My Cauchy Layer”
with thickness fit parameter called “Thickness # 1” would be renamed to “My
Cauchy Layer Thickness”.) The renaming of any layer fit parameter will also carry
over to the model fit results.

Figure 9-47. “Rename Layer Options” dialog box, which includes options for renaming the
layer in the model, renaming the layer fit parameters, and renaming the thickness that is
reported in the fit results.

Save Layer Optical Constants

Right-click on a layer to select this option which will save the optical constants for
that layer as a material (.mat) file. For most layers, you will be given the option to
save its material file as ‘Parameterized’ or ‘Tabulated’ (see Figure 9-48). Saving a
layer as a tabulated file saves a .mat file that contains a table of values that has
wavelengths and associated optical constant values. By default, a tabulated .mat
file is saved with wavelength in angstroms and optical constant values described by
ε1 and ε2. These settings can be changed from the Options tab by changing the
“Tabulated Mat File Save Mode” setting in the “General” section of the
CompleteEASE configuration. Saving a layer as a parameterized file saves a .mat file
that contains the values of the parameters in the layer’s dispersion equation.

Figure 9-48. “Save Layer Optical Constants” dialog box.

NOTE: Layers from .mat files that are already tabulated cannot be saved as
parameterized because they are not described by a parameterized dispersion
equation to begin with. (If you want to save a tabulated material file as a dispersion
equation, you will have to parameterize the layer, which can be done by right-
clicking on the layer and selecting “Parameterize”. See the next part of this section.)
Also, anisotropic layers (such as the Biaxial layer) cannot be saved as tabulated,
since they require multiple directions to describe their optical constants.

9-368 • Reference – Analysis Tab CompleteEASE Software Manual

 NOTE: Saving a layer’s optical constants can also be done by clicking Save in the
“Layer Commands” menu above the layer stack.

The B-Spline layer contains special options for saving its optical constants, as shown
in Figure 9-49. ‘Tabulated’ saves the B-Spline optical constants as a normal
tabulated .mat file as described above. An ‘Editable B-Spline’ is a parameterized
.mat file that allows all B-Spline node and parameter values to be fully editable
when it is reopened. A ‘Fixed B-Spline’ is a parameterized .mat file that does not
allow the B-spline parameter and node values to be edited by the user when it is
reopened. While its node positions are fixed once it is saved, the parameter and
node values of a ‘Fixed B-Spline’ can be varied by CompleteEASE during a fit when it
is reopened and used, unlike a tabulated file. Still, the positions of the nodes in a
“Fixed B-Spline” cannot be edited by a user after it is reopened.

Figure 9-49. “Save Layer Optical Constants” dialog box for a B-Spline layer.

Parameterize Layer
Right-click a layer to select this option to parameterize the optical constants of that
layer using a dispersion model, such a Gen-Osc, B-Spline, or Cauchy layer. This will
open the layer optical constants as reference values in the “Parameterize Layer”
dialog box, as shown in Figure 9-50. The three most common choices are the Gen-
Osc, B-Spline, and Cauchy models, which are easily selected from the Parameterize
Layer sub-menu. To parameterize to a different layer, such as a Sellmeier, simply
select “GenOsc” in the Parameterize Layer sub-menu. When the “Parameterize
Layer” dialog box opens, select ‘Open Material File’ and choose the layer you wish
to parameterize to (in this example, the Sellmeier). Refer to Sections 4.3 (B-Spline
example) and 4.4 (Gen-Osc example) for a detailed explanation and example of
using the Parameterize Layer functionality.

CompleteEASE Software Manual Reference – Analysis Tab • 9-369

 Figure 9-50. “Parameterize Layer” dialog box.

View Layer Comment

Right-click on a layer to select this option to view the layer comment, which often
contains information from the creator of the layer file about its source and other
measurement details (see Figure 9-51). For tabulated material files, the wavelength
range over which the material file is valid is also displayed. It is generally not best
practice to model experimental data that expands beyond the valid wavelength
range of the material files in the model layer stack. CompleteEASE will warn you if
this is the case and will attempt to extrapolate the optical constants beyond their
valid range, but this may not always be accurate, for example, when expanding a
transparent-film model into the absorbing regions.

Figure 9-51. Example of a layer comment for Cesium optical constants. The comment shows
the source of the optical constants (Palik’s Handbook of Optical Constants of Solids) and the
valid wavelength range.

Convert to EMA
Right-click on a layer to select this option which replaces the layer with an Effective
Medium Approximation (EMA), in which the original layer is now embedded as the
host material (Material 1). By default, the other material (Material 2) is set to
“Void” with “EMA %” set to 0 but selected as a fit parameter. Click on “Material 2”
to fill in the second material in the EMA. An example is shown in Figure 9-52.

9-370 • Reference – Analysis Tab CompleteEASE Software Manual

 Figure 9-52. Example of the EMA layer where the Cauchy layer was automatically converted
to the host material (Material 1) with VOID as the other material (Material 2).

Unwrap from EMA

After converting a layer to EMA, you can right-click on the EMA layer to unwrap the
host material (Material 1) back to a single layer without the EMA.

Convert to Transparent B-Spline

Right-click on a Cauchy layer to select this option, which converts the Cauchy layer
to an equivalent B-Spline layer. The “Convert to Transparent B-Spline” feature is a
quick and easy way to convert a Cauchy material describing the transparent region
of a semi-absorbing layer to an equivalent B-Spline material that describes the
optical constants over the whole spectrum. This feature is an automated process
that can replace much of the work involved for semi-absorbing samples like the
example from Section 4.3 (refer to the pink “Note” box at the end of the section).
To use the “Convert to Transparent B-Spline” feature, open a data file for a semi-
absorbing material and create a model with a Cauchy layer. Zoom in the wavelength
range until only the transparent region is selected and fit the Cauchy model to the
data. (For more information on the transparent-region analysis, refer to the
beginning of Section 4.3.) An example of a transparent region with corresponding
Cauchy-layer model is shown in Figure 9-53.

Figure 9-53. Example of a transparent region fit with a Cauchy model.

Once the transparent region is fit, right-click on the Cauchy layer and select
“Convert to Transparent B-Spline”. You will be asked to set the eV resolution for the
B-Spline layer, as shown in Figure 9-54. Typically, values of 0.2 eV or 0.3 eV work
best. Then, you will be asked to perform a “wavelength expansion fit” as shown in

CompleteEASE Software Manual Reference – Analysis Tab • 9-371

 Figure 9-55. Clicking ‘No’ will simply replace the Cauchy layer with an equivalent B-
Spline layer in the transparent region. Clicking ‘Yes’ will fit both the transparent and
absorbing regions with the B-spline layer. This is done via a wavelength-range
expansion fit, which will expand the analyzed wavelength range from the currently
selected transparent range to a user-defined range which may include absorption
regions by sequentially adding small wavelength ranges (with step size determined
by the resolution value) and analyzing the updated wavelength range with the B-
Spline model. After selecting yes, you will be prompted to enter the full wavelength
range in which to expand the B-Spline fit, as shown in Figure 9-56. The maximum
wavelength range values for the currently open data set will be automatically filled
in here, so it is generally okay to just click ‘Ok’. This will produce a fit result for the
entire spectral range of your data set, as shown in the example of Figure 9-57.

Figure 9-54. Prompt to enter the B-Spline resolution for “Convert to Transparent B-Spline”.

Figure 9-55. Prompt to perform a wavelength-range-expansion fit for “Convert to

Transparent B-Spline”.

Figure 9-56. Prompt to enter the wavelength range for the wavelength-range-expansion fit
with “Convert to Transparent B-Spline”. Generally, default values will work fine.

Figure 9-57. Example of a fit to both the transparent region and absorbing region of a
material using “Convert to Transparent B-Spline”.

9-372 • Reference – Analysis Tab CompleteEASE Software Manual

 Convert to Anisotropic
Right-click on an isotropic layer to select this option to replace the layer with an
anisotropic Biaxial layer, in which the original layer is now embedded as the starting
material for both the Ex and Ez directions. The default configuration is to have no
orientation present (determined by Euler angles) and anisotropy set to Uniaxial
(Ex = Ey  Ez). An example is shown in Figure 9-58. (Refer to Section 5.4 for more
about converting a film to anisotropic and to Section 10.9 for information about the
Biaxial layer.)

Figure 9-58. Example of an anisotropic Biaxial layer.

Convert to Isotropic
Right-click on an anisotropic layer (such as the Biaxial layer) to select this option
which replaces the anisotropic layer with an isotropic layer. The isotropic layer will
contain the layer parameters from the Ex direction of the anisotropic layer.

Grade Layer
Right-click on a layer to select this option to replace the layer with a Graded layer,
where the original layer is embedded as the average refractive index. A Graded
layer will incrementally vary the index of the film in a specified way throughout the
thickness of the layer. The “% Inhomogeneity” is set to 0% by default but can be
varied to allow the index to increase or decrease from the bottom to top of the
layer. An example of this is shown in Figure 9-59. (Refer to Section 10.18 for more
information on how to use the Graded layer.)

Figure 9-59. Example of Graded Layer.

Remove Grading
After grading a layer, you can right-click on Graded Layer and select “Remove
Grading” to unwrap the host film (from the “Material =” section) back to a standard
layer without grading.

CompleteEASE Software Manual Reference – Analysis Tab • 9-373

 Start Superlattice
Right-click on a layer to select this option in order to set the layer as the start of a
repeated superlattice structure. (For more information about using the superlattice
feature, refer to the “Superlattice” example in Section 4.5.)

End Superlattice
Right-click on a layer to select this option in order to set the layer as the end or top
of the repeated superlattice structure. When selected, an additional parameter will
be added to this layer to control the number of repeats (called “SL Count” for
“Superlattice Count”), as shown in Figure 9-60.

Figure 9-60. Model with superlattice start, end, and number of counts (repeats).

Clear Superlattice
Right-click on a superlattice layer to select this option which removes the layer from
the repeated superlattice structure. To clear the entire superlattice structure, you
must select “Clear Superlattice” on both the “SL Start” and “SL End” layers. (Note
that if you only “Clear Superlattice” on the “SL Start” layer, the “SL End” layer will
be automatically converted to the “SL Start” layer.)

Editing Fit Parameters (Fit Parameter Dialog Box)

In the Model panel, each fit parameter has a name shown in plain text (e.g.
“Thickness #1”) with its associated value (and sometimes its units) in blue,
underlined text (e.g. 100 nm) after an equal sign. If the fit parameter is active in the
fit (i.e. turned on as a fit parameter), then its value will be shown in bold, blue,
underlined text with the word “fit” next to it in parentheses (e.g. “100 nm (fit)”) as
shown in Figure 9-61.

Figure 9-61. Example of a fit parameter with parameter value activated for fitting.

Left-clicking on the fit parameter value (e.g. 100 nm) brings up a dialog box for
editing the value and other settings of the fit parameter (Figure 9-62). Right-clicking
on the fit parameter value toggles the fit parameter between being active in the fit
(allowing its value to be varied) and being inactive in the fit (holding its value
constant). Holding the SHIFT key down on the keyboard and rolling the mouse
wheel with the mouse pointer positioned over the fit parameter value will change
the parameter’s value in small increments and adjust the generated-data curves on
the graph accordingly to reflect the change in value. Holding down the CTRL and
SHIFT keys with the mouse pointer positioned over the fit parameter value and
rolling the mouse wheel will change the parameter’s value in even smaller
increments. This is generally referred to as “rolling the parameter value”, and it is a
useful way to easily adjust the value of the fit parameter and visualize the effect of
the parameter on the generated data at the same time.
Fit parameters have several settings that change how they appear in fit results or
how they behave during a fit. These options can all be found by left-clicking on the

9-374 • Reference – Analysis Tab CompleteEASE Software Manual

 blue, underlined fit parameter value (e.g. 100 nm) which will display the “Edit Fit
Parameter” dialog box as shown in Figure 9-62. These fit parameter settings are
discussed briefly below.

Figure 9-62. “Edit Fit Parameter” dialog box containing all settings for fit parameters.

Edit Value
The text field at the top of the “Edit Fit Parameter” dialog box contains the current
value of the fit parameter. To edit this value, simply type a new number in this field
and click “Ok” or press the “Enter” key on the keyboard. Keep in mind that the
number entered here cannot be outside the “Minimum” and “Maximum” values for
the fit parameter range.

Fit
The “Fit” checkbox indicates whether the fit parameter is turned on for fitting; in
other words, it controls whether the fit parameter is active in the fit, meaning its
value will vary when fitting, or whether the fit parameter is inactive in the fit,
meaning its value will be held constant during the fit. This option can also be
toggled without opening the “Edit Fit Parameter” dialog box by right-clicking on the
blue, underlined text of the parameter value (e.g. 100 nm) in the model panel.

Minimum / Maximum
The “Minimum” and “Maximum” values define the range in which the fit parameter
is allowed to vary when it is being fit (i.e. when the fit algorithm is varying its value).
The fit algorithm will never let a fit parameter wander outside of its specified
“Minimum” and “Maximum” values. In addition, CompleteEASE does not allow a fit
parameter to be rolled outside of its range with the mouse wheel, and entering a
value outside of the range for the fit parameter is not allowed as well. Therefore,
these fit parameter range values can be used to keep a parameter from having

CompleteEASE Software Manual Reference – Analysis Tab • 9-375

 values that are known to be incorrect (such as negative energy or negative
thickness values). Also, in cases where a certain fit parameter just doesn’t seem to
“lock on” to a correct value, check the fit parameter range as its minimum or
maximum values may be limiting the possible values it can have during the fit.

Show Advanced Parameter Settings

Checking “Show Advanced Parameter Settings” will display additional settings for fit
parameters that are less often modified from their default settings, such as the
parameter name, error bars, specification range, derivative increment, and
querying option.

Parameter Name
The “Parameter Name” area contains options to rename a fit parameter. To give a
fit parameter a different name than its default name, check the “Rename”
checkbox, and type the desired new name in the “Name” field. This is useful for
giving fit parameters unique names, which is required when using advanced model
features such as certain Derived Parameters, Custom User Equations, Parameter
Coupling, and Multi-Sample Analysis. This is mainly used when the model layer
stack contains two layers of the same type that have the same default names for
their fit parameters. This option can then help to differentiate fit parameters from
layers of the same type by making their names unique. However, note that
renaming fit parameters will break existing references to these fit parameters used
in Derived Parameters, Custom User Equations, Parameter Coupling, and Multi-
Sample Analysis. Therefore, use caution when renaming fit parameters, and
remember that it is best to give layers and fit parameters unique names before
using these advanced features. In order to rename all the fit parameters in a layer
in the same manner all at once, simply right-click on the layer name and choose
“Rename Layer and Fit Parameters”.
All fit parameters in the model are shown in the Fit Results (in the Fit panel and in
automatically generated analysis reports) by default. However, in order to hide a fit
parameter and prevent it from appearing in the fit results, check the “Hide”
checkbox in this section.

Parameter Error Bars

The “Maximum Error” specification controls how much error is allowed on a fit
parameter before it is considered “Out of Spec.”. This means that the “Out of Spec.”
designation will appear beside the fit parameter in the Fit Results in parentheses
(e.g. “(Out of Spec.)”) if its error bar exceeds the “Maximum Error” value. This value
is set intentionally high by default (10,000) because this feature should only be used
carefully in certain specific situations.

Parameter Specification
The “Low Spec.” and “High Spec.” values define the range in which a parameter is
considered to be within specification. When a fit parameter’s value ends up outside
of this range after a fit, it will be tagged with the “Out of Spec.” designation in the
Fit Results and any analysis reports (e.g. “(Out of Spec.)”). This feature is mostly
used when using CompleteEASE for quality control purposes, which allows the user
to be warned when a certain sample does not meet predetermined specifications.
The MSE for a fit can also have a maximum specification; this is set with the “Max.

9-376 • Reference – Analysis Tab CompleteEASE Software Manual

 Acceptable MSE” parameter in the +FIT Options area (see Section 9.7). Derived
parameters can also have high and low specifications, which are defined in the
“Include Derived Parameters” section for each parameter.

Derivative Increment
The “% Nominal Derivative Increment” controls how quickly a parameter will
change its value during a fit. It is specified as a percentage of the “nominal”, or
default, derivative increment for that parameter. Usually, this value should be left
at 100% as most CompleteEASE parameters are tuned to have derivative
increments that work well for optimal fitting. However, this feature is usually useful
when custom fit parameters are defined in user-defined equations. Since these
parameters are not created by CompleteEASE, this value may need to be adjusted
for them to vary correctly in a fit. When the Levenberg-Marquardt fitting algorithm
is varying a fit parameter, it will iteratively adjust its value slightly on each iteration
by the derivative increment value. The derivative increment is also used when the
fit parameter values are rolled up and down by the mouse wheel while holding
down the SHIFT key. In order to test whether the derivative increment is optimal for
a parameter, roll the mouse wheel over the parameter while holding down the
SHIFT key. The increment by which the parameter’s value changes with each roll is
the derivative increment. If large jumps are observed, then decrease the “%
Nominal Derivative Increment”. Likewise, if the parameter value doesn’t change
fast enough, then increase the “% Nominal Derivative Increment”.

Query Parameter
This section of the “Edit Fit Parameter” dialog box is only visible when
CompleteEASE is configured to ask a remote system (outside of CompleteEASE) for
what its parameter values should be. To turn on the ability for CompleteEASE to
query for parameter values from a remote source and see this section, go to the
Options tab and press ‘Edit Configuration’. In the “CompleteEASE Configuration”
dialog box, check “Show Advanced Configuration Options”, expand the “Remote
Communications” section under “Configuration Parameters”, and set “Allow
Parameter Query” to “True” (“Parameter ON”). Doing this will enable
CompleteEASE to ask a remote system for the fit parameter’s value on a port two
positions above the one specified with the “TCP/IP Port #” configuration setting. (By
default, this main port specified is 4444, meaning that in this case, the remote
parameter querying will happen on port 4446.) Once this is established, checking
the checkbox “Ask Remote Computer For Value” in the “Edit Fit Parameter” dialog
box will instruct CompleteEASE to ask for this parameter’s initial value from a
remote computer (on port 4446 by default) before performing a fit. Refer to
Appendix 14.8 for more information about remote communication features
available in CompleteEASE.

9.6. +MODEL Options

The full list of +MODEL Options are shown in Figure 9-63; however, many of these
options are not visible in the Model panel by default. To view many of these hidden
advanced options, you first need to select Configure Options at the bottom of the
Model panel and choose which advanced options you would like to use.

CompleteEASE Software Manual Reference – Analysis Tab • 9-377

 Figure 9-63. List of all +MODEL Options, including hidden advanced options.

Angle Offset
The “Angle Offset” applies a constant offset to the angle of incidence when
calculating the model, which can account for slight shifts or inaccuracies in the
angle-of-incidence data. By default, this parameter is a constant offset applied
equally to all angles of incidence in the data set. Generally, only one constant offset
for all angles of incidence is sufficient for most data sets requiring the “Angle
Offset”. However, by turning on “Multiple Angles of Incidence” in Configure
Options, this parameter can be fit individually for each angle of incidence in the
experimental data, as shown above in Figure 9-63.

Analyze All Rotation Angles

This option will only be shown if rotation data is loaded for analysis. Turning
“Analyze All Rotation Angles” to ON will simultaneously analyze all rotation angles
in the rotation data scan with the same model, forcing all fit parameter values to be
the same for all rotation angles except for the Euler angle “Phi” which, if present,
will be set to the value of the rotation angle at each rotation point. (This means that
the data generated by the model will be the same for all rotation angles in the data
set with the exception of the only parameter that changes with rotation angle, the
Euler angle “Phi”.) This option is different than pushing the ‘Fit Rotator Data’
button. When “Analyze All Rotation Angles” is turned OFF and the ‘Fit Rotator Data’
button is pressed, the same model structure will be fit to each rotation angle in the
data set, but the values of the fit parameters will be allowed to vary independently
for each rotation angle. Thus, there will be a different fit at each rotation angle, just
like there would be different fits for each timeslice when fitting dynamic data or for
each translation point in uniformity-map data. (This means that the data generated
by the model could be different for all rotation angles in the data, depending on the
variation in fit parameter values for each rotation angle.)

Include Substrate Backside Correction

“Include Substrate Backside Correction” calculates the incoherent contribution of
light reflecting from the backside of transparent substrates in the model. When

9-378 • Reference – Analysis Tab CompleteEASE Software Manual

 turned ON, the section will expand, as shown in Figure 9-64. Here, you can specify
if the SE data were collected in transmission through the sample or through the
reverse side of the substrate (opposite the coated side). There is also a selection
for the number of backside reflections and the percentage of first reflection that is
collected (from the top side); both of these are fit parameters.

Figure 9-64. When “Include Substrate Backside Correction” is ON, there are additional
options to consider transmission SE data, reverse direction measurements, and the number of
reflections from front and back.

For more explanation of the substrate backside correction and to see an example of
correcting for substrate backside reflections, refer to Section 3.4. Whenever
“Include Substrate Backside Correction” is turned ON, the substrate is given a
thickness value, as shown in Figure 9-65. This is often used with transmission data.
For an example of transmission data fitting using the substrate thickness, consult
Section 4.6. Also, when “Include Substrate Backside Correction” is turned ON,
CompleteEASE allows layers to be added below the substrate, also as shown in
Figure 9-65. To add layers below the substrate, use the Add command at the top of
the model and position the blue cursor below the substrate.

Figure 9-65. When “Include Substrate Backside Correction” is ON, the substrate is given a
thickness value, and CompleteEASE allows layers to be added below the substrate.

NOTE: When fitting transmission intensity data (not transmission SE data), even
when “Include Substrate Backside Correction” is OFF, CompleteEASE still uses the
value of “% 1st Reflection” and the value of the substrate thickness in the model
calculation. (“% 1st Reflection” is used to correct for the percentage of your
baseline intensity that could be lost at the detector due to scattering or other
reasons.) Therefore, if “% 1st Reflection” is not at its default value of 100% or if the
substrate thickness is set at a strange value, this will still affect the fit in an
undesired manner even though “Include Substrate Backside Correction” is OFF and
these fit parameters are not shown in the model. Thus, when fitting transmission
intensity data, make sure to set “% 1st Reflection” back to 100% and check the
value of the substrate thickness if you turn off the backside correction. (However, if
you leave “Include Substrate Backside Correction” ON when fitting transmission
intensity data, you’ll need to set “# Backside Reflections” to 0 to get the correct fit.)

Model Calculation
The “Model Calculation” section allows you to choose different model calculations
to handle nonideal sample situations. For most samples, the default setting of
“Ideal” is preferred. However, there are many choices related to spectrometer
bandwidth, film thickness nonuniformity, and beam angular spread to use (as
shown in Figure 9-66) when these nonideal situations are suspected. These
nonideal options are best used in conjunction with depolarization data fitting, as an
ideal sample/measurement will not produce any depolarization.

CompleteEASE Software Manual Reference – Analysis Tab • 9-379

 When a nonideal model calculation is selected, this section will expand to allow fit
parameters describing the amount of nonideality, as shown in Figure 9-67. With
advanced options selected, there is even the ability to describe two different
bandwidths, with a selectable switching wavelength. This is useful for systems that
have two separate spectrometers. Each nonideality is modeled by convolving
multiple data sets with different values. For example, a model with thickness non-
uniformity would be calculated by convolving the data from multiple different
generated data sets with the model thickness varied over a range of values. The
number of convolved data sets to approximate the nonideality can be controlled, as
seen in Figure 9-67, where the shape is set to Gaussian and “# of Pts” is equal to 9.
For examples of fitting data sets that include nonidealities such as “Thickness Non-
uniformity” and “Bandwidth”, refer to Section 5.1.

Figure 9-66. Different model calculation types.

Figure 9-67. Expansion of the Model Calculation section to include settings for Bandwidth,
Thickness Non-uniformity, and Angular Spread.

+Parameter Smearing
“Parameter Smearing” is a very advanced feature that allows other parameters to
be “smeared” to account for very specific nonidealities. Smearing a parameter acts
similarly to the nonidealities of “Bandwidth” (which is technically a smearing of
wavelength) and “Angular Spread” (which is a smearing of angle of incidence) as
described in “Model Calculation” above. “Parameter Smearing” can accommodate
smearing of any of the fit parameters. When turned on, this section expands as
shown in Figure 9-68. (A common use of this feature is when dealing with a
material, like a twisted liquid crystal, where you want to smear the amount of
“twist” present on the surface of a Biaxial grading.)

Figure 9-68. Parameter Smearing expands to allow smearing of any fit parameter.

9-380 • Reference – Analysis Tab CompleteEASE Software Manual

 Patterning
“Patterning” is another advanced feature that describes a situation where a sample
is only partially covered by a designated layer or the entire layer stack (relative to
the size of the measurement beam). This option assumes that the pattern
dimensions are much larger than the probing wavelength, and thus totally
incoherent overlap of light reflected from the coated and uncoated areas occurs on
the detector. This section expands when turned ON to calculate the model with
only partial coverage of any designated layer. (In other words, a pattern exists on
the surface of the sample when it can be described by a layer that does not cover
the full surface of the layer below, but only in select areas.) If all layers on the
substrate are patterned, you can choose “Layer # = 0” to pattern all of them, as
shown in Figure 9-69. The “% Patterned” value is the amount of film that is missing
from the measurement area. Thus, 0% would be a fully intact film and 100% would
be a non-existent film.

Figure 9-69. “Patterning” allows calculation with partial coverage of a single layer or multiple
layers over the layer below.

NOTE: The Patterning option is not suitable to describe periodic structures on the
surface of the sample with dimensions smaller than the wavelengths for which
coherent overlap of grating modes occurs on the detector.

NOTE: Patterning may be usurped if “Multi-Model Patterning” is also turned on.

Ambient Index > 1

This option is used to describe situations in which the probing light beam passes
through an ambient medium other than air (n = 1) before interacting with the
sample and where the light beam entered and exited said ambient medium at
normal incidence. This situation arises when working with Liquid Cells (where the
sample is immersed in an ambient of water or some other solution) or isosceles
prisms for which the angle of incidence is chosen to allow for normal incidence on
the liquid cell windows or prism faces. (This option is named “Ambient Index > 1”
because the default ambient for the model is “air” or “vacuum”, which has an index
of refraction of n = 1. Thus, any other ambient will have an index of refraction
greater than n = 1.) When “Ambient Index > 1” is turned ON, the section expands as
shown in Figure 9-70. A material file can be selected to describe the index of the
ambient material. If this material file is a dispersion equation (such as Cauchy), the
index of the ambient can be fit and its optical constants saved in this section. If the
ambient material happens to be different on the front (top) of the layer stack than
on the backside of the sample, “Use Separate Backside Ambient” can be turned ON
to specify a separate ambient for the backside.

Figure 9-70. Ambient Index > 1 allows another material file to describe ambient index.

CompleteEASE Software Manual Reference – Analysis Tab • 9-381

 Use Scattering Factor
If “Use Scattering Factor” is turned ON, the material file loaded (which needs to
represent a dispersion equation) will be used to describe the nonideality of
preferential scattering, where not all of the p-reflectance is captured vs. s-
reflectance due to really rough sample surfaces (like solar cells). This is mostly
meant to be used with T-Solar systems.

Wvl Shift (nm)

The “Wvl Shift” is used to simulate a wavelength shift if the ellipsometer
wavelength calibration for the already acquired data is questioned. In other words,
if it appears that the wavelengths in the measured data may not be “labeled”
correctly due to an offset in the wavelength calibration, this can be used to shift the
entire wavelength spectrum to fit the data. This option should only be used if the
optical properties of a test sample are well known. Ideally, “Wvl Shift” should only
be used when analyzing samples of multiple thicknesses. If the wavelength
calibration is in question, a wavelength and system calibration might be advisable.
Contact your J.A. Woollam representative for further instructions in this regard.

Return Path Ellipsometer

This option represents a special configuration of the M-2000 where the source and
detector are mounted side-by-side and the measurement beam reflects from the
sample, then from a mirror, and back off the sample surface on its way back to the
ellipsometer detector.

Delta Offset
The “Delta Offset” is used to correct for windows in the beam path, such as with a
Heat Cell or a prism. CompleteEASE can account for Delta offsets from a “Prism
(Cauchy)” or a “Window (1/)”. For most applications involving J.A. Woollam
attachments (such as the Heat Cell or Environment Cell), “Window Effects” are
automatically calibrated during a “System Check” and data is corrected for such
effects; thus, there is no need for this model option to be applied in these cases.

Psi Offset
The Psi offset can be used to model prisms in the beam path before and after the
sample. J.A. Woollam company instruments that use prisms for beam steering
already have a defined calibration method that accounts for these offsets, so this
feature of the model is rarely used.

Source Rot. / Receiver Rot.

These options are used to rotate the optical system on either the source or receiver
side from the standard plane of incidence. Enter a rotation angle in degrees to
rotate the respective part of the optical system. This feature is typically used in
special hardware configurations for in-situ monitoring on chambers, where there is
no room for the standard ellipsometer configuration.

9-382 • Reference – Analysis Tab CompleteEASE Software Manual

 +Multi-Model Patterning
When +Multi-Model Patterning is turned ON, it overrides the simplified patterning
feature discussed earlier in the +MODEL Options section. Multi-Model Patterning
allows you to mix calculations from multiple model structures where all model
structures are assumed to incoherently contribute to the total detector signal (i.e.
there is no interaction between beams reflected off of different sample areas). All
layers need to be added to the model in the order they will appear from the
different regions of the sample. For example, the “layered structure” shown in
Figure 9-71 needs to be constructed in order to model the hypothetical patterned
sample shown in Figure 9-72. Note that the “layered structure” picture shows all
layers overlapping all others regardless of the actual physical region they occupy.

Figure 9-71. Layered structure to be used in a patterned model calculation.

Surface oxide

a-Si Film

Cr Film

SiO2 Film
Cr Film

Silicon Substrate

Figure 9-72. Example of patterned sample, to be represented in CompleteEASE by Multi-

Model Patterning.

To set up the multi-model pattern for this sample, we will need three different
models – one for each of the areas that are different on the sample (within the
measurement beam spot). The first area is just a Cr film on Si (for this example, it
was assumed to be 25% of the measurement area). The second area is Cr on SiO2
on Si (assumed to be another 25% of the area). Finally, the complete stack of
surface oxide on a-Si on Cr on SiO2 on Si exists over the remaining 50% of the area.
To build this example using multi-model patterning, first choose “# of Models” to be
3, since there are three distinct areas of the film surface. Then, you can select the
various areal percentages. Next, add a fit parameter for each of the layers that will
vary from area to area – in this case, 3 of the 4 layers have less than 100% coverage.
The Cr film is coating the entire surface, and if we make the assumption that it is
the same thickness in all areas, then we do not need to add this parameter to the
list. However, it may be easier to add this parameter to the list for the sake of
easier visualization. The individual thicknesses in the different areas can be set,
fixed, or even fit. The final model for this sample is shown in Figure 9-73.

CompleteEASE Software Manual Reference – Analysis Tab • 9-383

 Figure 9-73. Model using Multi-Model Patterning to represent a sample with three distinct
areas.

By right-clicking on the fit parameter name in the Multi-Model Patterning table

(shown in Figure 9-74), you can access options that allow you to “Fit All”
parameters in that column, “Un-Fit All” parameters in that column, “Set Parm
Values” for all values in that column, and “Delete Multi-Sample Parameter” from
the Multi-Model Patterning table.

Figure 9-74. Options available in parameter right-click menu for Multi-Model Patterning.

+Multi-Sample Analysis (MSA)

It is possible to fit multiple data sets simultaneously using +Multi-Sample Analysis
(MSA). These multiple data sets can come from appended data files, multiple
selected points from a uniformity map, multiple rotations from automatic rotation
scans, or multiple timeslice selections from dynamic data. Multi-Sample Analysis
can be needed in each of these scenarios when some parameters in the model are
consistent over all data sets and other parameters vary from data set to data set.
For an example of using the Multi-Sample Analysis feature, refer to Section 5.2.
An example of Multi-Sample Analysis is shown in Figure 9-75, where three oxides on
Si are fit with a common Cauchy layer to describe the same index, but the
thicknesses for each data set are allowed to vary to adjust to the varying
thicknesses.

9-384 • Reference – Analysis Tab CompleteEASE Software Manual

 Figure 9-75. Example of Multi-Sample Analysis, where all three data sets are fit
simultaneously with common Cauchy optical constants and different film thickness.

By right-clicking on the fit parameter name in the Multi-Sample Analysis table

(shown in Figure 9-76), you can access options that allow you to turn on that
parameter as a fit parameter for all data sets (“Fit All”), turn off that parameter as a
fit parameter for all data sets (“Un-Fit All”), set the value of the parameter for all
data sets (“Set Parm Values”), set the range minimum and maximum for this
parameter value for all data sets (“Set Parm Ranges”), or delete this parameter
from the Multi-Sample Analysis (“Delete Multi-Sample Parameter”).

Figure 9-76. Options available in parameter right-click menu for Multi-Sample Analysis.

It is possible to set different fit weights for each data set in the MSA. Right-click on
the Data Set number and select “Set Fit Weight” (see Figure 9-77). This will open a
dialog box allowing you to enter a custom weight for that data set in the fit. By
default, all data sets are weighted equally (100%). You may wish to increase or
decrease the fit weighting of a particular set or sets if, for example, one set is
significantly noisier than others. The fit weight can also be used to balance a fit
result when a good fit can be achieved with one set but not others. (Be sure to
consider sample structure and model first before editing the fit weighting.) To
remove the custom fit weights from all data sets, you can right-click on Data Set
and select “Reset Fit Weights” (see Figure 9-77).

CompleteEASE Software Manual Reference – Analysis Tab • 9-385

 Figure 9-77. Right-click on Data Set number to change the fit weight for that data set and on
Data Set itself to reset the fit weights for all data sets.

NOTE: The names of the parameters in the model are very important when using
+Multi-Sample Analysis. If any parameters share the same name in the model,
CompleteEASE may not choose the correct parameter to use in the MSA. Thus,
before setting up MSA, make sure that all parameters to be used in the MSA have
unique names. If not, you can rename layer fit parameters by right-clicking on the
layer and choosing “Rename Layer and Fit Parameters”. If, after MSA has been set
up, a parameter involved in the MSA has its name changed, it will break the link
with the MSA, and the parameter with the new name will have to be re-added to
the MSA.

+Parameter Coupling
Parameter coupling allows users to enforce relationships between different fit
parameters in the model. There are three types of coupling: simple constant ratio,
simple constant offset, or custom equation. Figure 9-78 shows an example of
parameter coupling. The corresponding “Parameter Coupling” dialog is shown in
Figure 9-79. As an example, the thickness of the first and second data sets in a
multi-sample analysis, “Thickness #1 (1)” and “Thickness #1 (2)”, are set to be
double the thickness of the third data set in the MSA, “Thickness #1 (3)”. This ratio
will be enforced, and the thicknesses will vary together during the fit. Only active fit
parameters can be coupled to other fit parameters. The final coupling equation is
shown at the bottom of the dialog box to ensure that it is correct. To see an
example of a data fit using Parameter Coupling, see Section 4.5.

Figure 9-78. Example of Parameter Coupling, where thicknesses in an MSA are coupled by a
factor of 2 to another thickness in the MSA.

Figure 9-79. Example of Parameter Coupling, where thicknesses in an MSA are coupled by a
factor of 2 to another thickness in the MSA.

9-386 • Reference – Analysis Tab CompleteEASE Software Manual

 Parameter Coupling can also be done using a custom equation entered by the user.
An example of this is shown in Figure 9-80. For more information on how to use the
equation editor and its available operators and functions, refer to the “Custom
Oscillator” subsection of Section 10.4 and Table 10-3.

Figure 9-80. Example of Parameter Coupling with a custom equation, where a thickness in an
MSA is coupled by a factor of 2 to another thickness in the MSA.

NOTE: The names of the parameters in the model are very important when using
+Parameter Coupling. If any parameters share the same name in the model,
CompleteEASE may not couple either of those parameters correctly. Thus, before
setting up parameter coupling, make sure that all parameters to be coupled have
unique names from other parameters. If not, you can rename layer fit parameters
by right-clicking on the layer and choosing “Rename Layer and Fit Parameters”. If,
after a coupling relationship has been set up, a parameter involved has its name
changed, it will break the parameter coupling link to this parameter. In this case,
any coupling relationships involving this parameter will have to be re-added.

Use Previous Results

When CompleteEASE saves a data file during a measurement, it will also include the
results of any data analysis that was performed automatically during that
measurement (by using a recipe which includes a model) in the saved data file.
These analysis results from previous measurements can then be used for
consecutive sample measurements that may require information from the previous
measurements about the underlying sample structure.
For example, you may wish to map the uniformity of a layer on a substrate before it
undergoes some processing. The fit results from this first map can be used during
the analysis of the next measurement after processing. When working with multiple
layers, you may wish to analyze the thickness of the first layer as a uniformity map
prior to depositing the second layer and to use the results for the first underlying
layer when analyzing the second layer.
“Use Previous Results” requires that the same map locations are used for each run
so that measurement points can be directly compared. When using “Use Previous
Results”, the fit parameter values from the previous results will be used as the
values for the corresponding parameters in the current results. An example of the
options for “Use Previous Results” is shown in Figure 9-81.

CompleteEASE Software Manual Reference – Analysis Tab • 9-387

 Figure 9-81. “Use Previous Results” allows you to import results from a previous fit (stored
inside the data file).

If using a B-Spline layer to fit for the optical constants of the first (underlying)
measurement, make sure that two settings in the +Advanced section of the
B-Spline layer are turned ON. Turn “Show Parameters in Fit” ON so that all of the
B-Spline node parameters are saved into the results section of the data file. Also,
you may want to turn “Fix node bounds when all wavelengths selected” ON so that
the wavelength range for the B-Spline is consistent between both measurements.
Refer to the +Advanced part of Section 10.3 for an explanation of these options.
If a model is not loaded during data acquisition (i.e. as part of a recipe), then no
analysis results will be recorded in the measured data file. In this case, it will be
necessary to use the Fit Log’s reanalysis feature to get the model results saved into
the data file (using the “Resave Results to Data Files” checkbox) before this feature
can be used. Refer to Section 9.3 for a reference on the various Fit Log functions
used for the reanalysis of data. Also refer to Section 3.6 for an example of using the
Fit Log to reanalyze data.
The “Parse From Acquisition File Name” option is to be used instead of “Results
File” if the data file containing the results is not explicitly specified by “Results File”
but rather by a prefix of the current data file being analyzed. Turning this option ON
tells the model that the data file containing the results (“Results File”) should be
found by looking at the file name given to the current experimental data file. (Thus,
when using “Parse From Acquisition File Name”, the “Results File” option should be
set to “<None>”. This field can be cleared by clicking on the blue underlined file
path and then clicking “Cancel” on the file dialog without opening a file.) The
“Delimiter” option tells the model how to parse the name of the data file containing
the previous results from the current experimental data file name. Note that this
parsing feature will only occur during a recipe measurement.
For example, say a wafer is initially measured, its data file is saved as “wafer.SE”,
and its fit results are saved into the same experimental data file, “wafer.SE”. Then,
the wafer undergoes some sort of post-processing. After that, the post-processed
wafer is measured, its data file is saved as “wafer-post.SE”, and it uses a model with
“Use Previous Results” turned ON and “Parse From Acquisition File Name” turned
ON with “-” as the “Delimiter”. In this case, to obtain the fit results for the pre-
processed wafer in order to fit the post-processed wafer, CompleteEASE will
acquire the previous results by parsing the experimental data file name, “wafer-
post.SE”, using only the text before the specified delimiter, in this case “-”, so that
the file name “wafer.SE” will be parsed from the beginning of the file name “wafer-
post.SE”. CompleteEASE will then extract the previous results from “wafer.SE” in
order to analyze the data in “wafer-post.SE”. (Note that the pre-processed data file
containing the previous results must be in the same directory as the post-processed
data file.) If the delimiter is not found in the data file name, then an error message
will appear and the data will be analyzed only with the values in the current model.

9-388 • Reference – Analysis Tab CompleteEASE Software Manual

 9.7. +FIT Options
The full list of +FIT Options are shown in Figure 9-82; however, some of these
options are not visible in the Model panel by default. To view some of these hidden
advanced options, you first need to select Configure Options at the bottom of the
Model panel and choose which advanced options you would like to use.

Figure 9-82. List of all +FIT Options, including hidden advanced options.

Include Initial Fit Stage

This feature allows a model to define a “pre-fit” (a global fit that happens before
the main model fit occurs) where only a subset of the fit parameters is allowed to
vary before allowing all of the defined fit parameters to be fit. An example of
“Include Initial Fit Stage” could be finding the thickness value of one layer using only
nominal thicknesses of other layers before all layer thicknesses are fit, as shown in
Figure 9-83. Refer to the “Use Global Fit” section below for descriptions of what
some of the options under “Include Initial Fit Stage” do, as they are very similar to
the global fit options.

Figure 9-83. “Include Initial Fit Stage” allows you to define a subset of fit parameters that will
be included in a global fit that occurs before the main model fit with all of the fit parameters.

CompleteEASE Software Manual Reference – Analysis Tab • 9-389

 Perform Thickness Pre-Fit
When “Perform Thickness Pre-Fit” is turned ON, the model will use J.A. Woollam’s
patented algorithm to quickly estimate the thickness for the thickest layer in the
model before all of the fit parameters are allowed to fit. This algorithm is based on
calculation of zero-crossings when the data shows interference oscillations in the
presence of a film on a substrate due to the film thickness. The regression analysis
algorithm used for model fitting is sensitive to initial values and can get lost in local
minima (in the MSE space) if these values deviate too much from the final values.
“Perform Thickness Pre-Fit” can be extremely useful by ensuring that generally one
of the most consequential parameters, the thickness of the thickest layer in the
layer stack, is close enough to its actual value so that the fit has a much higher
probability of finding the correct solution. To see an example of this, refer to
“Thickness Pre-Fit” in Section 3.2. The thickness pre-fit options are shown in Figure
9-84; however, for most samples, just turning ON “Perform Thickness Pre-Fit” (using
default values) is sufficient. Only for rare special cases do these parameters need to
be changed from their default values.

Figure 9-84. “Perform Thickness Pre-Fit” options.

As stated above, the menu of options under “Perform Thickness Pre-Fit” should
rarely be changed from its default values. Only in rare cases should this be required.
Generally, if the thickness pre-fit fails to work properly, a global fit on the thickest
layer in the model should be used instead. Each extra thickness-pre-fit parameter is
briefly described below:
• Auto Wavelength Range: The thickness pre-fit algorithm operates by
looking for zero crossings in the C and S data. “Auto Wavelength Range”
indicates that CompleteEASE will automatically select the range over which
to look for zero crossings, generally looking for zero crossings in a subset of
the selected wavelength range. Turning “Auto Wavelength Range” to OFF
means it will use the full selected wavelength range when looking for zero
crossings. If thickness pre-fit fails, this is generally the first parameter to
change before attempting to use thickness pre-fit again.
• # of Compare Points: “# of Compare Points” affects the speed of the
thickness pre-fit algorithm. The algorithm compares generated and
experimental data to determine which thickness produces the best results,
but it does not use all the wavelengths in the spectral range. Rather, it only
selects 20 wavelengths from the full range by default, but this can be
changed with “# of Compare Points”.
• Subtract Average: Sometimes, the C- and S-curve oscillations are not
centered around zero and thus zero crossings may not be representative of
the thickness. To correct for this, “Subtract Average” can be turned ON to
subtract the mean value of the C and S curves (i.e. normalize them) when
searching for zero crossings.
• S or C Only: The thickness pre-fit algorithm looks at the data and
automatically decides whether to use S or C data when finding zero
crossings. When the thickness pre-fit algorithm generates “test” data

9-390 • Reference – Analysis Tab CompleteEASE Software Manual

 based on S or C to compare with experimental data, it compares the data
(at the compare points) for the N, C, and S data. Determining the pre-fit
initial thickness value depends on the best fit between this “test” data and
the experimental data. By default, the N, C, and S data are all used to
determine goodness of fit, but when “S or C Only” is turned ON, only the S
data (if the algorithm selects S to find zero crossings) or only the C data (if
the algorithm selects C to find zero crossings) will be used to determine the
goodness of fit and thus select the initial pre-fit thickness.
• Slope Threshold: This value indicates the minimum slope of the curve (i.e.
its derivative) that must occur at a zero crossing for it to be counted as a
“true” zero crossing. This is done to help lessen the chance that
measurement noise in the data around zero is mistaken for zero crossings.

Use Global Fit

Because the standard CompleteEASE fit algorithm searches for parameter values in
a region close to their initial values, the fit can sometimes produce a result that is
not the globally best solution, because it found only a locally good solution. The
Global Fit can help the fit algorithm to avoid getting trapped in local minima and
find the solution that is globally the best. The Global Fit works by searching an
entire grid of parameter-value combinations and finding which of these positions
will produce the lowest MSE (see Figure 9-85 for a graphic of this concept in the 2D
case). A normal fit is performed at each grid point where the global-fit grid values
are used as the starting values for the global-fit parameters and the fit-parameter
nominal values are used as starting values for all other fit parameters. Once the
parameter-value combination that can produce the lowest MSE value is identified
for the parameters in the global fit, those values (from the global-fit grid) are used
as the starting values for a normal fit to finalize the fit results for all fit parameters.
In general, the Global Fit should only be used if needed, as it greatly increases the
time involved in calculating fit results. The options available for the Global Fit are
shown in Figure 9-86. An example of using the Global Fit is described in Section 3.3.

Figure 9-85. Visual representation of what the Global Fit does with two fit parameters. It
scans an entire grid of parameter-value combinations and finds the one that will produce the
lowest MSE.

CompleteEASE Software Manual Reference – Analysis Tab • 9-391

 Figure 9-86. Options available when using the Global Fit.

Parameters
The “Parameters” section defines the fit parameters from the model that will be
used in the Global Fit. Parameters in the model must be turned on as fit parameters
before they can be used in the Global Fit.

Add
Select Add to add fit parameters from the model to the Global Fit.

Delete All
Select Delete All to delete all of the fit parameters present in the Global Fit. If “Use
Global Fit” is turned on but no parameters are specified, then no Global Fit will be
performed. To delete only one Global Fit parameter and not all of them, simply click
on the red x to the left of the Global Fit parameter you wish to delete.

Param. #1, #2…

An unlimited number of parameters can be included in the Global Fit. However, as
more Global Fit parameter are added, more calculations will need to be performed
which will result in a slower fit. Note that the total number of grid points in a Global
Fit is the product of the “# Guesses” for each global-fit parameter. Therefore, by
adding more parameters to a Global Fit, the total calculation time will grow in a
rather exponential manner and not simply linearly. An estimate of the total time
required to finish the Global Fit will be shown after the first few iterations of the
Global Fit are performed. If the Global Fit is expected to take too much time to
complete, it can be canceled and parameters can then be removed from the Global
Fit to reduce the calculation time before starting it again.

Min., Max., # Guesses

The “Min.”, “Max.”, and “# Guesses” options are vital to defining the “search” range
for the Global Fit. These parameters must specify the minimum value and maximum
value for the range over which the Global Fit will search for that parameter’s best
value. It is best to limit this range as much as possible to ensure fit efficiency. The
“# Guesses” option specifies how many values within the “search” range will be
tested. Again, this value should be kept to a lower number for quicker fit speed, but
be sure to provide enough guesses to search the whole range adequately. For

9-392 • Reference – Analysis Tab CompleteEASE Software Manual

 example, if searching for a layer thickness value with a minimum of 200 nm and a
maximum of 1000 nm with 5 guesses, the values tested for that layer thickness in
the Global Fit would be 200 nm, 400 nm, 600 nm, 800 nm, and 1000 nm.

+Customize Global Fit

This section contains optional settings that can change the way that the Global Fit
performs, most of which affect the speed at which the Global Fit operates.

# of Data Points
When performing a Global Fit, CompleteEASE does not use all of the measured data
to calculate the MSE values in order to increase speed and efficiency. Instead,
CompleteEASE selects a representative subset of data points from the data set
when performing the Global Fit. This option allows the user to specify the number
of data points CompleteEASE will use. A lower number means a faster Global Fit and
a higher number means a slower Global Fit.

# of Iterations
When performing a Global Fit, CompleteEASE does not perform all of the fit
iterations that it would normally perform for a regular fit while testing the global-fit
parameters. This is done to increase speed and efficiency because calculating the
full number of iterations for every parameter-value combination in the Global Fit
would be very time-consuming. Generally, a lower number of iterations is sufficient.
This value allows the user to specify how many iterations should be performed for
each parameter-value combination in the Global Fit grid. (A value of 0 entered here
will tell CompleteEASE to only calculate the MSE of the model using the nominal
values of each parameter, with no fit iterations performed.)

Random Search
By default, when CompleteEASE performs a Global Fit, it searches through the
parameter value combinations sequentially, similar to that shown in Figure 9-85.
However, the time taken to reach the results can be greatly reduced if
CompleteEASE does not have to exhaustively search all the parameter-value
combinations. If “Random Search” is turned ON, CompleteEASE will instead test
parameter-value combinations in random order until a result is found that is below
the desired “MSE Threshold”. When that result is found, the Global Fit terminates,
and a regular fit is performed to finalize fit parameter results and reach the
minimum MSE possible.

MSE Threshold / MSE Threshold Value

If the “MSE Threshold” is on, CompleteEASE will only perform the Global Fit until a
parameter-value combination is found that is below this threshold. When that
result is found, the Global Fit terminates, and a regular fit is performed to finalize fit
parameter results and reach the minimum MSE possible. Note that the “MSE
Threshold” is required to be ON to use “Random Global Fit”.

CompleteEASE Software Manual Reference – Analysis Tab • 9-393

 Flash Search / # of Fits
A “Flash Search” is a type of Global Fit that is designed to be faster than any other
type of Global Fit. In the “Flash Search”, CompleteEASE performs a Global Fit over
the entire parameter grid, but it only calculates the MSE of the model with the
nominal values of the fit parameters; it does not perform any fit iterations for any
parameter-value combinations. (Nominal values are simply the values that are
defined by the Global Fit “Min.”, “Max.” and “# Guesses” for parameters in the
Global Fit and the model-defined value for parameters not in the Global Fit.) Once
the “Flash Search” completes, it considers a subset of the parameter-value
combinations that contain the lowest MSE values of all the calculations that have
been performed. CompleteEASE then performs regular fits on only the parameter-
value combinations in that subset and finds the parameter-value combination that
has the lowest MSE after these regular fits are performed. (This is done because
MSE calculations on nominal values are faster than full iterative fits.) The “# of Fits”
setting specifies how many parameter-value combinations CompleteEASE will
include in the subset for regular, iterative fits.

Include Wvl. Range Expansion Fits

Turning this option ON specifies whether the Global Fit should also use the “Wvl.
Range Expansion Fit” feature located in +OTHER Options (see Section 9.8). The
"Increment (eV)" setting currently defined in the +OTHER Options section will be
applied for the increment used with the wavelength-range-expansion fits
performed here.

Limit Wvl. for Fit

When “Include Wvl. Range Expansion Fits” is turned ON, this option allows a limited
spectral range to be used for the expansion fits during a Global Fit to increase its
speed, since an expansion fit can greatly increase the time required to perform a fit.

Selected Data
This option specifies the type of data used to calculate the MSE during the fit.
Choices include Standard Ellipsometry (SE) data (Psi and Delta), Generalized
Ellipsometry (GE) data, Mueller Matrix data, and Intensity data only, as shown in
Figure 9-87. If any of these are selected, then the data set must have been acquired
with that type of data included. If Mueller Matrix data is selected, there are more
options available. You can choose whether or not the first element (m11) of the
Mueller matrix should be included in the fit. You can also specify which specific
Mueller-matrix elements should be included by turning “Use MM Elements Selected
for Graphing” to ON. Then, to select the elements, open the +Data menu above the
graph and choose “Select Mueller Matrix Elements” (refer to Section 7.4).

9-394 • Reference – Analysis Tab CompleteEASE Software Manual

 Figure 9-87. Choose the data type to use during fitting.

Fit Weight
This option is only available if Standard Ellipsometry data or Generalized
Ellipsometry data is specified in the “Selected Data” option. The Fit Weight setting
allows you to select the fit weighting, as shown in Figure 9-88. Fit weighting
determines which values affect the calculation of the MSE, which drives the fit
procedure as it is being minimized during the regression analysis. The quality of fit
for the model, as determined by the match between model-generated and
experimental data, is evaluated solely based on the data types selected in this
option, as the MSE is only calculated on types of data selected here. By default,
CompleteEASE uses N, C, and S data for fit weighting, as shown in Equation 9-4,
where n is the number of data points (which includes wavelengths × angles of
incidence × rotation angles/timeslices/map points if present, etc.), m is the number
of fit parameters, N = cos(2), C = sin(2)∙cos(), and S = sin(2)∙sin(). NE
represents the experimental data for N, and NG represents the generated data for
N. The same is true for both C and S. Changing this fit weighting will change the
data types which Equation 9-4 includes in its sum.
Equation 9-4

𝑛 2 2 2
1 𝑁𝐸 − 𝑁𝐺𝑖 𝐶𝐸 − 𝐶𝐺𝑖 𝑆𝐸 − 𝑆𝐺𝑖
𝑀𝑆𝐸𝑁𝐶𝑆 = √ ∑ [( 𝑖 ) +( 𝑖 ) +( 𝑖 ) ]
3𝑛 − 𝑚 . 001 . 001 . 001
𝑖=1

Figure 9-88. Options for fit weighting.

CompleteEASE Software Manual Reference – Analysis Tab • 9-395

 NOTE: Only the fit-weight option "Exp. Std. Dev.", which matches the experimental
and model-generated Psi and Delta data, uses the measurement error bars
recorded in the data file for fit weighting. This approach emphasizes parts of the
spectrum with low error bars. All other data types, when used as the basis for fit
weighting, use predefined, constant fit weights (error bars) for each wavelength
assuming that all data points in the spectrum contain meaningful information
(absolute MSE). In Equation 9-4, a weight of 0.001 is used for N, C, and S data,
which is the estimated error bar for typical Mueller-matrix data obtained over the
full spectral range of an M-2000 or RC2 ellipsometer.

Include Depolarization Data

Turn this setting ON when you want the depolarization data to be specifically
included as part of the fit weighting. (It will be included as an additional term in the
sum of Equation 9-4.) You can then select the percentage of weighting compared to
the other data types. Set the “% Weight” to 100% for equal weighting, lower for less
importance than other types of data, and higher for more importance than other
data types.

NOTE: Depolarization (e.g. caused by thickness non-uniformity) also effects the

standard ellipsometry data (Psi and Delta) and can be characterized sufficiently
without specifically turning this “Include Depolarization Data” option ON.

Include Intensity Data

Turn this setting ON when you want the intensity data to be included in the fit
weighting. (It will be included as an additional term in the sum of Equation 9-4.) You
can then select the percentage of weighting compared to other data types. Set the
“% Weight” to 100% for equal weighting, lower for less importance than other
types of data, and higher for more importance than other data types. Be very
careful when turning this option ON, as intensity data is always acquired and stored
along with the ellipsometry data during a measurement, but intensity data is only
meaningful if a baseline was acquired shortly before the ellipsometry
measurement. If it is not the case that a baseline was acquired, then the intensity
data will negatively affect the fit.

Transmission Data % Weight

When Transmission Intensity data are appended to the data set, they are included
in the fit regardless of the “Include Intensity Data” option above. This setting allows
more or less importance to be placed on the Transmission Intensity data by
increasing or reducing its weight percentage, respectively. As a rule of thumb,
based on how the MSE is calculated (see Equation 9-4), J.A. Woollam generally
recommends weighting the transmission intensity data at 200% for each angle of
incidence present in the data (e.g. five angles of incidence would result in 1000%
transmission data weighting).
This rule of thumb can be thought about intuitively in the following way: The MSE
calculation weights the N, C, and S data equally, as in Equation 9-4. When you
include transmission-data weighting, CompleteEASE adds an additional term to the
MSE sum, so now it has four data terms (N, C, S, and Transmission). Assuming you
have only one angle of incidence for the N, C, and S data, weighting the

9-396 • Reference – Analysis Tab CompleteEASE Software Manual

 transmission data equally to all the SE data (N, C, and S) would require a
transmission data weighting of 300%. A weighting of 100% means that
CompleteEASE weights the transmission data equal to any other "source" of SE
data, but since there are three "sources" of ellipsometric data (N, C, and S),
transmission data weighting at 100% represents only 1/4 of the overall weight in
the MSE calculation. Based on the fact that, while N, C, and S represent three
quantities, they are only related to two independent underlying quantities (Psi and
Delta), it has been observed that weighting transmission data equally at 300% tends
to weight it "too much", so relaxing that value to 200% seems to works well in
general.
Now, when you have multiple angles of incidence in your data set, you effectively
have additional "sources" of data (N, C, and S) because ellipsometric data depends
on the angle of incidence. However, that is not so with transmission intensity data,
since it is often only acquired at normal incidence. Thus, as you add more angles of
N, C, and S to the fit, the weight of your N, C, and S terms grows with respect to the
transmission intensity data, so if you want your transmission intensity data weight
to "keep up" with your N, C, and S data, you need to use 200% transmission-
intensity data weighting per angle of incidence in the data.

Define MSE using L*a*b*

In almost all cases, the MSE in CompleteEASE is defined as the mean-squared error
that results between experimental data and model-generated data. This setting,
however, allows the MSE that drives the fit results to be defined by the difference
between the model-calculated L*a*b* color coordinates and target values set by
the user. In this case, the “fit” of the model is considered with regard to specific,
user-defined target color coordinates, not the experimental data. This feature is
thus used for simulation purposes, for example, when searching for an optimal
design. This setting will find the model fit-parameter values that will most closely
result in certain desired color coordinates specified by the user. When this is turned
ON, the settings for the target values become visible. More options for color
calculation can be found in the Color Calculation dialog box found under +OTHER
Options (Section 9.8) when the model is configured to show color calculation
settings.

Limit Wvl. for Fit

Turning this option ON lets you reduce the wavelength range to a specifically
defined range to use during fitting. If you would rather limit the number of
wavelengths included in the fit by skipping every few wavelengths, use “Skip Data
Points in Fit”. With this option OFF, the wavelength range that is currently graphed
will be analyzed during the fit.

Limit Angles for Fit

Turning this option ON lets you select the specific angles of incidence to use during
fitting. You can use a comma-separated list to designate specific angles to include in
the “Included Angles” setting. With this setting OFF, the angles of incidence that are
currently selected for graphing are analyzed during the fit.

CompleteEASE Software Manual Reference – Analysis Tab • 9-397

 Max. Acceptable MSE
The value entered here will prompt CompleteEASE to tag the MSE with the “Out of
Spec.” designation if the MSE exceeds this value after the fit, as shown in Figure
9-89. To adjust the ranges for an individual fit parameter to be tagged with the “Out
of Spec.” designation, see the advanced fit parameter settings in Section 9.5.

Figure 9-89. Example of the MSE being “Out of Spec.”.

Skip Data Points in Fit

This option reduces the number of wavelengths used during fitting by skipping this
number of data points. In other words, only every n-th wavelength in the data will
be included in the fit calculation. If you would rather limit the number of
wavelengths included in the fit using a specifically defined wavelength range
instead of by skipping every few wavelengths, use “Limit Wvl. For Fit”.

Max. Fit Iterations

A fit in CompleteEASE is terminated once the MSE becomes smaller than 10-10, once
the change in MSE between fit iterations is smaller than 10-10, or once a predefined
number of fit iterations is reached (which, by default, is 50 iterations). If the
maximum number of fit iterations is reached, the fit will terminate and the current
in-process values for each fit parameter will be reported as the fit results.
This option allows you to change the number of maximum fit iterations performed.
Note that you can view the number of performed iterations after a fit has
terminated in the “Fit Statistics” window (see Section 9.3) in order to see if the
maximum number of fit iterations was reached. If a fit ended simply because it
reached that maximum number of fit iterations, it may indicate that the best match
between model-generated and experimental data was not found due to extended
time taken by the fit regression algorithm to reach a satisfactory result. In this case,
increasing the “Max. Fit Iterations” may help the model to find the best result or
checking the “% Nominal Derivative Increment” (see Section 9.5) for certain fit
parameters may also help, as this may indicate that certain fit parameters are
changing too slowly on each iteration to allow the fit to reach the best result in a
limited number of fit iterations.

Auto Fit Parameter Reset

Use this feature with caution, as turning it ON and not understanding what it is
doing can cause seemingly strange and erratic behavior when fitting data. This
option will reset all fit parameters to their initial values before fitting new data.
(Note that these initial values are the current model values at the moment that this

9-398 • Reference – Analysis Tab CompleteEASE Software Manual

 option is turned ON, not the current model values before pressing the ‘Fit’ button.)
Auto Fit Parameter Reset can be very useful with uniformity maps or dynamic data
in order to avoid fit parameters getting “lost” and not being able to recover after
measuring a single “bad” data point. Thus, this option ensures that the fit always
starts from a reasonable starting point. When this option is turned ON, all the
parameter values in the model are automatically recorded, and after this, no matter
how the values are changed, the model will reset the values back to the values as
they were when “Auto Fit Parameter Reset” was turned ON. Therefore, in order to
make changes to your default starting values, you must first turn Auto Fit
Parameter Reset to OFF, then change the values to the correct initial values you
want, and then turn it back to ON.

Include Derived Parameters

A derived parameter is a value that is not a fit parameter in the model, but its value
can be directly calculated from other fit parameters in the model (or sometimes
from the experimental data directly). After fitting data using a model with derived
parameters, the fit results section will show both the model fit parameters and the
derived parameters. The derived parameters are shown in gray text instead of
black, as shown in Figure 9-90. To add a derived parameter to a model, turn
“Include Derived Parameters” to ON. This allows you to add multiple derived
parameters. Figure 9-91 shows how a few derived parameters appear in the model.

Figure 9-90. Fit panel showing a derived parameter, the index of the SiO2 layer at 632 nm.

Figure 9-91. Examples of derived parameters in the “Include Derived Parameters” section.

Clicking on the blue, underlined text after “Type =” (e.g. n as in derived parameter
1: in Figure 9-91) allows you to change the type of derived parameter being
calculated using the dialog box shown in Figure 9-92. Many of the available types of
derived parameters have additional associated options for controlling how the
derived parameter is calculated and how it is displayed in the Fit Results. A brief
description of each type of derived parameter is shown in Table 9-3, and more
detailed descriptions of a few of the derived parameters are given below that.

CompleteEASE Software Manual Reference – Analysis Tab • 9-399

 Figure 9-92. Selecting the derived parameter type.

Table 9-3. Types of Derived Parameters in CompleteEASE.

DERIVED PARAMETER DESCRIPTION

Optical Constants The value of one of the optical constants for a specific
n, k, ε1, and ε2 layer in the model at a desired wavelength.
Anisotropic Differences The anisotropic differences of a uniaxial or biaxial layer in
the model. Refer to the “Graph Layer Optical Constants
(anisotropic differences)” part of Section 9.5 for more
explanation of these quantities.
Optical Thickness The optical thickness of a specified layer at a desired
wavelength. (Optical thickness is the product of the
layer’s physical thickness with its index of refraction. It
represents how “thick” the layer appears to an incident
light beam of the specified wavelength.)
Total Thickness The sum of all the thickness values of all the layers in the
model (excluding the substrate). This only includes
thickness values of films on top of the substrate. Films
below the substrate and the substrate itself are never
considered.
Total Optical Thickness The sum of all the optical thickness values for all layers
above the substrate in the model, multiplying each
layer’s physical thickness by the appropriate index of
refraction for each layer at a specified wavelength.
Psi and Delta values The value of Psi or Delta from the experimental data at a
desired wavelength.
Depolarization The depolarization from the experimental data at a
desired wavelength.
Pseudo Optical Constants The value of one of the pseudo-optical constants
<n>, <k>, <ε1>, and <ε2> calculated from the data at a desired wavelength. (For
more information on pseudo-optical constants, refer to
Section 4.1.)
Reflectance (Calculated) The reflectance or transmittance values, at a specified
Transmittance (Calculated) wavelength and angle of incidence, calculated from the
model once the data has been fit. These are not
measured values; they are simply what the model
implies.

9-400 • Reference – Analysis Tab CompleteEASE Software Manual

 Reflectance Extrema The maximum or minimum model-calculated reflectance
Transmittance Extrema or transmittance values over a specified wavelength
range at a desired angle of incidence and the position
within the specified range at which the maximum or
minimum occurs.
Constant A constant value that is meant to be used in equations
for the “User Equation” type of derived parameter. This
allows the constant in the equation to be named and
edited for future use.
User Equation An option allowing the user to enter an equation for a
custom derived parameter based on model fit parameter
values.
Color Coordinates The values of the color coordinates based on the
specular reflectivity of the final model
Sheet Resistance The sheet resistance of layers between the specified
starting and ending layer numbers. This is the sum of the
resistivities of each layer as if they were parallel
resistors.
Parameter Correlation The correlation between two fit parameters in the
model. Correlation for all fit parameters can be seen in
the “View Parameter Stats” menu on the Fit panel
(Section 9.3).

User Equation
The “User Equation” derived parameter allows the user to enter a custom equation,
based on the model fit parameters, that will be evaluated after the fit has finished
and displayed in the Fit Results on the Fit panel. If there is an error with the
equation, then the user equation derived parameter will not be displayed in the Fit
panel after a fit is completed.
To edit the equation for the derived parameter, press the Edit Equation command.
This brings up the “Edit Equation” dialog box as shown in Figure 9-93. The variables
used in an equation are fit-parameter names specified in enclosing brackets. An
example would be “[Thickness # 1]”. Any fit parameter name can be specified as
well as any other derived parameter names in the list of derived parameters. For
more information on how to use the equation editor and its available operators and
functions, refer to the “Custom Oscillator” subsection of Section 10.4 and Table
10-3. In addition to specifying the equation, this dialog allows the user to enter the
number of decimal places that should be displayed.

Figure 9-93. Edit Equation dialog box.

CompleteEASE Software Manual Reference – Analysis Tab • 9-401

 Color Coordinates
This derived parameter calculates the color coordinates based on the specular
reflectivity of the final model5. An example of color coordinates is shown in Figure
9-94, where the color coordinates are listed as both a derived parameter in the
Model panel and as results shown in the Fit panel. When “Color Coordinates” are
selected as a derived parameter, you have the additional choices of the type of
color coordinates (L*a*b*, Tristimulus, or RGB), the angle of incidence for
calculation of the reflectivity based on the final model, the standard observer
(which defines the underlying color sensitivity curves used in the calculation), and
the standard illumination light source (which defines the illumination spectrum
assumed for the calculation). For more options and settings involving color
calculation, see the Color Calculation dialog box found under +OTHER Options
(Section 9.8) when the model is configured to show color-calculation settings.

Figure 9-94. Color Coordinates shown as a derived parameter along with the fit results using
the L*a*b* type.

Sheet Resistance
This derived parameter, shown in Figure 9-95, calculates the sheet resistance of
layers between the specified starting and ending layer numbers. It sums the
resistivities as if they are parallel resistors, as shown in Figure 9-96. There are
“Offset” and “Multiplicative Corrections” available for matching ellipsometry-
measured values to alternative techniques if desired (e.g. a four-point probe). The
sheet resistance calculation rules are as follows:
• The calculation assumes all layers in the selected range “touch” the contacts.
The user sets the selected range (start and end layer numbers) appropriately.
• By default, all layers have an infinite resistivity.
• Gen-Osc layers with a Drude(RT) oscillator will use the “Resistivity” parameter
for the calculation. Gen-Osc layers with a Drude(NMu) oscillator will calculate
the resistivity from the “N” and “mu” parameters, assuming the current
effective mass is valid (see “Drude(RT) & Drude(NMu) Oscillators” in Section
10.4 for calculation details).
• Graded Gen-Osc layers using the parametric grade type will have the resistivity
of the slices summed up as if they are individual layers.
• EMA layers and graded layers not using the parametric grade type are
considered to have infinite resistivity even if they include a Gen-Osc layer with a
Drude oscillator.
• Intermix and Surface Roughness layers are ignored.

5
B. Johs, H. Arwin, T. Wagner, D. Appel, and D. Peros, “Accuracy of color determination from spectroscopic
ellipsometry measurements”, Thin Solid Films 519 (2011) 2711-2714.

9-402 • Reference – Analysis Tab CompleteEASE Software Manual

 Figure 9-95. Derived parameter showing Sheet Resistance options.

Figure 9-96. Calculation of sheet resistance from multiple layers in a graded model using the
parallel-resistor model.

+Translation Scan: First Point Only

This menu defines specific fit parameters that will be turned off or entire layers that
will be removed after the first point on a uniformity map (translation scan) is fit.
This option only applies when pushing the ‘Fit Scan Data’ button or when fitting the
data as it is being acquired.

9.8. +OTHER Options

The full list of +OTHER Options are shown in Figure 9-97; however, some of these
options are not visible in the Model panel by default. To view some of these hidden
advanced options, you first need to select Configure Options at the bottom of the
Model panel and choose which advanced options you would like to use.

CompleteEASE Software Manual Reference – Analysis Tab • 9-403

 Figure 9-97. List of all +OTHER Options, including hidden advanced options.

Wvl. Range Expansion Fit

The Wavelength Range Expansion Fit is used to expand a model fit from one region
of the spectrum into other regions of the spectrum. It is useful when a good fit and
thus good starting values have been obtained in another spectral region which can
then guide the fit in a new spectral region. When using a Wavelength Range
Expansion Fit, the model will be fit over the selected wavelength range first. Then,
the data range will be increased toward shorter and/or longer wavelengths in
increments defined with the “Increment (eV)” setting. As each new range of data is
added, the data are fit again. This method works well in combination with a B-
Spline layer and is demonstrated in the example of Section 4.3. A Wavelength
Range Expansion Fit can also be performed automatically by converting a Cauchy
layer to a B-Spline layer using the “Convert to Transparent B-Spline” option, as
explained in Section 9.5.

Try Alternate Models

Try Alternate Models will automatically fit the data with several common models
that include different options for different types of sample nonidealities to see
which matches the experimental data best. First, the data is fit using the ideal
layered optical model. Then, surface roughness is added as well as simple grading
and simple anisotropy. The results of the various model fits are summarized in a
table, and CompleteEASE will recommend which model should be used. The user
can then decide which model is most appropriate for their sample based on these
results and the recommendation by using the arrow buttons to choose the model
they want to apply and then clicking ‘Apply Chosen Model’. An example of the
report generated by Try Alternate Models is shown in Figure 9-98. To see examples
where Try Alternate Models is used, refer to Sections 3.5, 4.3, 5.4, and 6.1.
The way that CompleteEASE makes its recommendation in Try Alternate Models is
as follows: CompleteEASE will sequentially try each model in order of increasing
complexity and will see if the MSE value improves enough to justify the increase in
model complexity. First, CompleteEASE fits the ideal model to determine the
baseline MSE. Then, CompleteEASE adds surface roughness to the ideal model and
performs a fit. If the MSE of the model with roughness improves by at least the MSE
Improvement Threshold (MIT), then the model with roughness will be selected as
the new baseline model on which to compare all others. If not, the ideal model will
be retained as the baseline model for comparison. (By default, the MIT is 25%, so if
the MSE of the model with roughness is 25% better than the ideal model, i.e., its

9-404 • Reference – Analysis Tab CompleteEASE Software Manual

 MSE is 75% or less of the ideal model MSE, then CompleteEASE determines that the
model with roughness is better than the ideal model.)
Then, CompleteEASE removes the surface roughness but adds simple grading to the
thickest layer in the model, fits the data with that model, and records the MSE. If
the MSE is improved by at least the MIT over the previous model selected as a
baseline (whether ideal or with roughness), then the model with grading is selected
as the new baseline for comparison. Next, CompleteEASE keeps the simple grading
on the thickest layer in the model and then adds surface roughness. The data is fit
once again, and the resultant MSE is compared to the MSE of the previously best
determined model. If the previous baseline model included either roughness or
grading alone (one element of complexity), then for CompleteEASE to select the
model with roughness and grading (one additional element of complexity), only a
one-fold improvement of the MIT is required (by default, a 25% improvement).
However, if the previous baseline model was still the ideal model, then for
CompleteEASE to select the model with roughness and grading combined (two
additional elements of complexity), a two-fold improvement of the MIT is required
(by default, a 43.75% improvement, since “two-fold” is defined as
1–(1–MIT)∙(1–MIT) in CompleteEASE).
Finally, CompleteEASE removes the surface roughness and grading in the model and
then applies a simple, uniaxial anisotropic difference model to the data. The MSE in
this case is calculated and then compared to the previous baseline model for
comparison. For anisotropy to be selected by CompleteEASE as the baseline model,
a three-fold improvement of the MIT is required (by default, a 57.81% improvement
since “three-fold” is defined as 1–(1–MIT)∙(1–MIT)∙(1–MIT) in CompleteEASE). The
reason that such a high bar is required for anisotropy to be recommended is
because it can be easy for an anisotropic model to look like its fits the data well
when the sample is not actually anisotropic, so care should be taken to confirm that
your sample is actually anisotropic before switching to an anisotropic model.
By the end of the sequence of models, CompleteEASE recommends whichever
model was most recently chosen as the baseline for comparison, as that means that
it exhibited the most justified increase in complexity relative to its fit-quality
improvement as measured by the MSE.

Figure 9-98. Example of using Try Alternate Models.

CompleteEASE Software Manual Reference – Analysis Tab • 9-405

 Fit Parameter Uniqueness
This feature allows a user to test the uniqueness of fit parameters. In other words,
how does the quality of the model fit change as a parameter’s value is varied? If the
parameter value can be varied over a surrounding range and the quality of the
model fit is unaffected, that would indicate that the model is not sensitive to this
parameter, and thus there is no guarantee that the final fit-parameter value is
accurate. This determination can be made by performing a parameter uniqueness
test. To perform a parameter uniqueness test, one or two fit parameters are
selected to vary over a range of values. These ranges can be set in the “Parameter
Uniqueness Fit” dialog box, shown in Figure 9-99. At each value over the specified
ranges, the chosen fit parameters are fixed while all other fit parameters are varied
to find the best fit (lowest MSE). The MSE is recorded at all of these points and then
graphed over the values of the parameters in order to visualize the shape of the
MSE profile/contour versus the varied fit parameters. An example one-dimensional
MSE profile is shown in Figure 9-100, and an example two-dimensional MSE
contour is shown in Figure 9-101. For good examples of how and when to use Fit
Parameter Uniqueness, refer to Section 4.2 and Section 4.6.

Figure 9-99. Parameter Uniqueness Fit dialog box.

Figure 9-100. MSE profile from a one-dimensional parameter uniqueness test.

9-406 • Reference – Analysis Tab CompleteEASE Software Manual

 Figure 9-101. MSE contour map from a two-dimensional parameter uniqueness test.

Fit Parameter Error Estimation

This command tests the current model and data set for both random errors and
systematic errors. When pressed, a dialog box will appear as shown in Figure 9-102.

Figure 9-102. Fit Parameter Error Estimation dialog box.

Random Errors (Noise)

The “true” fit-parameter precision can be experimentally determined by calculating
the standard deviation of the fit parameters obtained by multiple measurements on
the same sample. However, multiple sample measurements can be very time-
consuming and impractical; thus, it is desirable to estimate the parameter precision
based on a single measurement. The “Random Errors (Noise)” section will test the
current data set and model in three possible ways, as described below, to estimate
the fit parameter precision. An example of these tests is shown in Figure 9-103 for
measurement of a native oxide on silicon. The test compares random error

CompleteEASE Software Manual Reference – Analysis Tab • 9-407

 estimations for various selections of angles of incidence. The “Actual Std. Dev.” are
calculated by measuring the same sample 50 times and checking the statistics of
the thickness “precision”. The “Fit Stats” are the 90%-confidence error bars that
are calculated after fitting a single data set. The “Error Bars MC” and “Bootstrap
MC” stand for the “Experimental Error Bars” and “Bootstrap” Monte-Carlo
simulations, as described below. For this sample, these estimates all have similar
magnitude, lending credibility to the value in estimating model precision based on a
single measurement.
Native Thick Standard Deviation, Fixed Index
0.05

All Angles
0.045
45° Only
0.04
55° Only
0.035 65° Only
Standard Deviation in Angstroms

0.03 75° Only

0.025

0.02

0.015

0.01

0.005

0
Actual Std. Dev. Fit Stats Error Bars MC Bootstrap MC

Figure 9-103. Random Error tests for a native oxide on silicon, comparing results for all
angles of incidence with results from various single-angle measurements.

Experimental Error Bars

The experimental data are fit multiple times but are manipulated before each fit
such that individual data from each wavelength are modified randomly based on
their measured error bars. The percentage listed here is associated with the
amount the experimental data are adjusted based on a percentage of the individual
error bars.

Bootstrap Method
The Bootstrap method6 also fits the experimental data multiple times to check for
“precision”. The subset of data points used by each trial fit are selected at random
from the total population of experimental data points. As the name suggests, each
time an individual data point is selected, it is replaced back into the general
population of experimental data points. This means that, during any single trial fit,
the trial sample can contain one, two, or more copies of any given data point – or
no copies at all, if it is never selected. The user determines the size of the trial
sample relative to the total available data points by entering a “% Replacement”.

Parameter Perturbation
The experimental data are fit multiple times. However, before each trial fit, the fit
parameters in the model are perturbed by a percentage of their current value to
check if the fit is robust enough to return to the same solution.

6
R. Rosa, “The inverse problem of ellipsometry: a bootstrap approach,” Inverse Problems 4 (1988) 887.

9-408 • Reference – Analysis Tab CompleteEASE Software Manual

 Systematic Errors
Random errors (precision) do not consider the effects of systematic errors on the
model fit results. In many cases, the systematic errors can be more significant.
Thus, CompleteEASE allows the accuracy to be estimated in the presence of
different possible systematic errors. These errors may include measurement errors
such as angle offset, wavelength shift, or data offsets or they may include modeling
errors such as substrate-optical-constant inaccuracies. Another method is also used
that adds the fit “errors” to the data to estimate the inaccuracy of the modeling.
The systematic errors are either added to the experimental data sets (constant
offsets in the ellipsometric data and fit error magnitudes) or to the optical model
(angle offset, wavelength shift, and offsets in substrate optical constants). The
changes in the fit parameters in the presence of the assumed systematic errors are
reported.

Angle Offset
Fit results are compared with and without an angle offset.

Wavelength Shift
Fit results are compared with and without a wavelength shift.

Psi Offset
Fit results from the standard data set are compared to a data set that is
manipulated by applying a constant offset to the Psi values.

Delta Offset
Fit results from the standard data set are compared to a data set that is
manipulated by applying a constant offset to the Delta values.

Transmission Offset
Fit results from the standard data set are compared to a data set that is
manipulated by applying a constant offset to the Transmission Intensity values.

Substrate ‘n’
The standard fit is compared to a fit where the model substrate index of refraction
(n) is manipulated by applying a constant offset.

Substrate ‘k’
The standard fit is compared to a fit where the model substrate extinction
coefficient (k) is manipulated by applying a constant offset.

Fit Errors Magnitude (+/-)

After a fit, there will be differences between the experimental data curves and the
model-generated curves. These can be viewed as a “difference graph”. These
represent systematic errors of an unknown kind (and combination) where the data

CompleteEASE Software Manual Reference – Analysis Tab • 9-409

 are not perfectly matched by the model. Without knowing the type of systematic
errors that produced these differences, CompleteEASE tries to estimate the effect
of this error on the fit parameter accuracy by adding or subtracting the same
“magnitude” of error at every wavelength to the existing data set and refitting the
manipulated data.
For example, Figure 9-104 graphs the difference between experimental data and
the model-generated curves for a thin organic film over a thick, oxide-coated silicon
wafer (P3HT). The “Fit Error Magnitude” takes the absolute value of this curve and
then adds or subtracts this from the existing experimental data set (giving the +/-
curves), as shown in Figure 9-105.
This test was applied to the P3HT film, which is anisotropic and absorbing in the
visible spectral region. This makes it difficult to gain sensitivity to the extraordinary
direction, which is out-of-plane (normal to sample surface). This is seen in the
results of Figure 9-106 for the ordinary direction and Figure 9-107 for the
extraordinary direction.

Fit Error Example (P3HT on 500nm SiO2/Si)

0.02

0.015
Difference: Exp Data and Model

0.01

0.005

-0.005

-0.01

-0.015
300 400 500 600 700 800 900
Wavelength in nanometers
Figure 9-104. Difference Graph between experimental data and the model-generated curves
for a P3HT film on SiO2-Si substrate.

Fit Error Magnitudes for Systematic Error Calculation

0.02

Fit Error Magnitude (+)

0.015
Fit Error Magnitude (-)
0.01

0.005
Magnitude

-0.005

-0.01

-0.015

-0.02
300 400 500 600 700 800 900
Wavelength in nanometers

Figure 9-105. “Fit Error Magnitudes (+/-)” calculated from the difference graph.

9-410 • Reference – Analysis Tab CompleteEASE Software Manual

 P3HT on 500nm SiO2/Si (ordinary optical constants)
6 3

e1_o, Initial Fit

5 e1_o, Fit Errors Mag. (+) 2.5
e1_o, Fit Errors Mag. (-)
e2_o, Initial Fit
4 e2_o, Fit Errors Mag. (+) 2
e2_o, Fit Errors Mag. (-)

2
1
3 1.5

2 1

1 0.5

0 0
0 200 400 600 800 1000 1200 1400 1600 1800
Wavelength in nanometers

Figure 9-106. Optical constants for the P3HT layer ordinary direction (in-sample-plane) where
“Fit Error Magnitudes” show the accuracy of the ordinary direction to be very good.

P3HT on 500nm SiO2/Si (extraordinary optical constants)

3.5 0.7

3 0.6

2.5 0.5

2 0.4

2
1

e1_e, Initial Fit

1.5 0.3
e1_e, Fit Errors Mag. (+)
e1_e, Fit Errors Mag. (-)
1 e2_e, Initial Fit 0.2
e2_e, Fit Errors Mag. (+)

0.5 e2_e, Fit Errors Mag. (-) 0.1

0 0
0 200 400 600 800 1000 1200 1400 1600 1800
Wavelength in nanometers

Figure 9-107. Optical constants and “Fit Error Magnitudes” for the extraordinary direction
(out-of-plane, normal to surface) of the P3HT film show less sensitivity to the values in this
direction.

Simulate Data
Experimental data can be simulated in CompleteEASE from the model using the
Simulate Data feature. This feature generates ellipsometric data from the current
model and uses it as though it were experimental data. This feature must be turned
on in Configure Options before it can be used.
To use this feature, click on Simulate Data to open the “Simulation Settings” dialog
box as shown in Figure 9-108. This dialog allows the user to specify the wavelength
range and angle range for the simulated data as well as choose the data type to be
simulated. Click “Ok” to create the simulated data set. (The checkbox called
“Perform Sensitivity Calculation” is a rarely used function that is designed to give
the user an idea of which angles and wavelength ranges give the most sensitivity to
a particular fit parameter. This option simulates data with the current model, then
increments a selected fit parameter by a specified amount, and then simulates the

CompleteEASE Software Manual Reference – Analysis Tab • 9-411

 data with the new fit-parameter value. After this, it calculates and graphs a mean-
squared comparison metric between these two simulated data sets and calls this
the “Absolute MSE”.)

NOTE: To simulate transmission results, “Include Substrate Backside Correction”

needs to be turned to ON and then “Transmission SE Data” also needs to be ON.

Figure 9-108. Simulation Settings dialog box.

Add Optical Constants to Report

Choose this option to automatically add graphs of the model optical constants to
the analysis report. An analysis report can be created and copied to the clipboard
by pressing CTRL+R or by right-clicking in the Fit panel text area and choosing “Copy
Analysis Report to Clipboard” (see Section 9.3).

# of Processor Cores to Use

CompleteEASE can increase calculation speed by using multiple processor cores (or
more accurately, threads). By default, when set to “0” all available cores will be
used when fitting. (It should be indicated on the screen how many cores are
available to be used.) Setting this number lower than the number of available cores
can limit the amount of processing power that CompleteEASE will use on a given
computer when fitting a model to experimental data, but it may allow the user to
perform other tasks with other software on the same computer while a longer task
(such as a fit to a whole uniformity map) is running in CompleteEASE.

Color Calculation
This feature is used to estimate the perceived color of a sample described by the
current model by calculating standard color coordinates assuming specular
reflection from the sample when illuminated by a standard illumination source of
predefined spectral intensity distribution under a defined angle of incidence and
considering the color sensitivity of the eye of a standard observer 7.
Press Color Calculation to show a dialog box representing the color coordinates of
the current model, as shown in Figure 9-109. The type of color coordinates (L*a*b*,
Tristimulus, or RGB), the angle of incidence for calculation of the reflectivity, the

7
B. Johs, H. Arwin, T. Wagner, D. Appel, and D. Peros, “Accuracy of color determination from spectroscopic
ellipsometry measurements”, Thin Solid Films 519 (2011) 2711-2714.

9-412 • Reference – Analysis Tab CompleteEASE Software Manual

 standard observer (which defines the underlying color sensitivity curves used in the
calculation), and the standard illumination light source, called the “Illuminant”,
(which defines the illumination spectrum assumed for the calculation) can all be
modified to see the resulting effect on the perceived color-coordinate values. The
results are reported in terms of the chosen color coordinates as well as the
respective color shading of the box behind the coordinates, giving users the ability
to visually compare the model-predicted color shown on the dialog box screen with
the apparent color of their actual sample.
The values of each color coordinate can be simulated when varying one or two
model fit parameters using the ‘Simulate’ button. The “Color Calc Parameter
Simulation” dialog box for this purpose is shown in Figure 9-110, and an example of
a color-coordinate simulation over the thickness of a Cauchy film on silicon is shown
in Figure 9-111. (Note that the coordinates that the simulation dialog allows will
match those selected in the drop-down menu of the main “Color Calculation”
dialog.)

NOTE: The Color Calculation feature displays the CIE L*a*b* color coordinates for
the model, using ASTM Standard E308-99.

Figure 9-109. Color Calculation dialog box shows the color coordinates and estimated
graphical “color” for specular reflection based on current model, specified Angle of Incidence
(AOI), standard observer, and illuminant light.

Figure 9-110. Color Calc Parameter Simulation dialog box lets the user simulate the values of
the color coordinates by varying the values of one or two parameters.

CompleteEASE Software Manual Reference – Analysis Tab • 9-413

 Figure 9-111. Graph of the results of the color-coordinate simulation.

Configure Options
Choose this command to show hidden advanced features in the +MODEL Options,
+FIT Options, and +OTHER Options sections of the model. The “Show Model
Options” dialog box is shown in Figure 9-112, which contains a brief description of
each of the available advanced options.

Figure 9-112. Show Model Options dialog box, which displays hidden options for the model.

Turn Off All Fit Parameters

Turns off all active fit parameters in the model, meaning that all values will be
treated as fixed values. After this option is clicked, the ‘Fit’ button can no longer be
pressed until at least one fit parameter is turned back on (activated for fitting).

9.9. Additional Models

All of the models contained in the “Basic” folder of the CompleteEASE model library
are simply different versions of the standard multilayer model (“Blank.mod”)
configured for different situations. CompleteEASE does have other models than the
standard multilayer model available in the “Advanced” folder of the model library.
These models are used for very specific applications and special cases that the
standard multilayer model does not account for. Brief descriptions of these models
and what they do are provided below; however, because of the specific and highly

9-414 • Reference – Analysis Tab CompleteEASE Software Manual

 specialized nature of these models, detailed instructions about their use are not
provided. If you are interested in obtaining more specific information on these
models, contact J.A. Woollam.

Basic Library Models

The basic models in the CompleteEASE library are all special versions of the
standard multilayer model (“Blank.mod”) that are used for specific types of
samples. Using these models can be a quick and easy way to analyze common types
of samples without having to spend time building a model, provided that your
sample type can be represented by one of the pre-built models. As can be seen
from the list in Figure 9-113, options include amorphous silicon on a glass substrate,
glass substrates with transmission data, glass substrates with an absorbing film, and
silicon substrates with absorbing films, transparent films, and thermal and native
oxides.

Figure 9-113. The “Basic” section of the CompleteEASE model library.

Growth Rate and Opt. Const. (GROC) Model

The Growth Rate and Opt. Const. (GROC) model is meant to model dynamic data
wherein a film is being grown on a substrate or layer stack. The GROC model can
only be applied to cases where all three of these conditions are met: the growth
rate is constant, the optical constants do not vary during growth, and a virtual
interface approximation can be applied for the underlying structure. (This is
typically true for semiconductors and metals but not dielectrics). The GROC model
is designed to simplify the simultaneous extraction of the film growth rate and
optical constants from dynamic (in-situ) data. Because the GROC model uses a
virtual interface, the user does not have to specify an optical model for any layers
or materials underlying the film being grown. For more in-depth information about
the GROC model and for an example of its use, refer to Section 6.2.

CompleteEASE Software Manual Reference – Analysis Tab • 9-415

 Opt. Const. Compare (OCC) Model
The Opt. Const. Compare (OCC) model has a variety of uses related to the variation
and comparison of optical constants over material composition or temperature. The
OCC model can be used to create composition- or temperature-dependent optical-
constant “libraries”, which are a special type of material file for materials whose
optical constants vary with composition or temperature. When constructing these
libraries, B-Splines are used to represent the constituent optical-constant spectra
for each known composition or temperature. The library is then built using an
interpolation algorithm based on the critical-point shifting algorithm8. In the critical-
point shifting algorithm, the dielectric function at an arbitrary composition or
temperature is evaluated from a weighted sum of the nearest reference dielectric
function spectra after the reference spectra are wavelength-shifted to line up the
critical-point features. CompleteEASE simplifies this process by automatically
determining the wavelength-shifting polynomials (that describe the path of critical
points versus composition/temperature) via a nonlinear regression analysis. For
more information and an example of the use of the OCC model, refer to Section 5.3.

Pore Size Model

The Pore Size model is a special model that can only be used with an Environment
Cell. The Pore Size model is used to obtain various quantities associated with
porosimetry experiments such as the solvent volume, refractive index, thickness as
a function of relative pressure, and the pore size distribution. Oftentimes, the Pore
Size model is used in conjunction with the Porous EMA layer (Section 10.20). In
order to use the Pore Size model, you must open it in a second model panel
(Model #2) with a standard multilayer model open in the first model panel (Model).
See Section 9.4 for an explanation of multiple model panels. For more information
on using the Pore Size model, refer to Chapter 5 of the J.A. Woollam Environment
Cell Manual, “Analysis of Porous Samples”.

Retardance Model
The Retardance model is designed to simplify the analysis of certain anisotropic
substrates which are typically classified as “linear retarders”. Common examples of
such anisotropic samples are birefringent polymeric sheets (e.g., PET, PEN,
Polyimide), waveplates, quartz, and sapphire substrates. This model requires
ellipsometry data measured in transmission mode. The first part of the model is
based on the Jones decomposition9. The Jones decomposition shows that the
product of any number of retarders and rotators can be equivalently described by
the product of a retarder and a rotator. The Retardance model applies the
decomposition method on generalized ellipsometry (GE) data through a 2×2 matrix
diagonalization. The output parameters are the linear retardance, fast axis, rotary
power, and any nonidealities of the data. This model is also capable of finding the
polarizing parameters through ellipsometric modeling rather than using direct
calculation. When the “In-Plane and Out-Of-Plane” option under “Model Analysis”
is selected, CompleteEASE will create a biaxial Cauchy model for the data. Before

8
P.G. Snyder, J.A. Woollam, S.A. Alterovitz, and B. Johs, “Modeling AlxGa1-xAs Optical Constants as Functions of
Composition”, J. Appl. Phys. 68 11 (1990) 5925.
9
R.C. Jones, “A New Calculus for the Treatment of Optical Systems I. Description and Discussion of the Calculus”, J. Opt.
Soc. Am. 31, p. 488-493 (1941).

9-416 • Reference – Analysis Tab CompleteEASE Software Manual

 the model is created, the user needs to choose the data type, substrate thickness,
and fit parameters associated with the anisotropic index. After the ‘Fit’ button is
pressed, CompleteEASE will generate the model which provides the best-fit biaxial
refractive indices (nx, ny, and nz) and the Euler angle “Phi”. This model will appear in
the second model panel (Model #2), which can be accessed by pressing CTRL+F2.
(See Section 9.4 for an explanation of multiple model panels.) For more details
about the Retardance model and directions for its use, contact J.A. Woollam.

CompleteEASE Software Manual Reference – Analysis Tab • 9-417

10. Reference – Layer Types

CompleteEASE provides many different ways to represent the optical constants of a

material versus wavelength. These representations range from simple tabulated
lists of the optical constants to complex dispersion models. All of these different
representations are stored as layers (.mat files) in CompleteEASE. The layers
currently available are listed in Table 10-1. This chapter describes the operation
and typical uses of the different layers in detail.

Table 10-1. Layer types in CompleteEASE. Most of these layers can be found in the
CompleteEASE material files library. The asterisk (*) indicates layers that are not present in
the CompleteEASE library but are obtained in other areas of CompleteEASE.

LAYER NAME LAYER DESCRIPTION

BASIC LAYERS
Tabulated A layer that contains tabulated values for the optical constants
(n and k values or of a material described by either n and k or ε1 and ε2. Most files
ε1 and ε2 values) in the CompleteEASE library with an element or compound
name (e.g. “Ag” or “BaTiO3”) with no other descriptor in the
name (such as “_g” or “(Cauchy)”) are tabulated.
Cauchy A simple dispersion equation for transparent materials using
three terms to describe the index of refraction (n) and optionally
an exponentially decaying function to independently describe
the extinction coefficient (k) in some simple cases (Urbach
absorption).
B-Spline A dispersion layer that uses basis-spline functions and a series of
“nodes” to describe both the ε1 and ε2 dielectric functions in a
flexible yet Kramers-Kronig-consistent manner.
Gen-Osc A dispersion layer that combines multiple oscillators of various
(General Oscillator) types to be used to represent the optical constants of a
material.
EMA A layer used for mixing the optical constants of two materials
(Effective Medium together to create a composite material using various methods
Approximation) such as the Bruggeman EMA or Maxwell-Garnett EMA.
Intermix* A simple EMA layer that mixes the optical constants from the
material below and material above in a 50%-50% ratio. Intermix
layers are found in the “Add Layer to Model” dialog box by
clicking Add in the model.
Void A simple layer that contains the optical constants for air
(vacuum), which are n = 1 and k = 0 for all wavelengths.
ADVANCED LAYERS
Additive A layer that sums the optical constants of multiple constituent
layers together.

10-418 • Reference – Layer Types CompleteEASE Software Manual

 Biaxial An anisotropic layer used to describe the anisotropic optical
constants of both uniaxial and biaxial materials. It contains the
traditional phi, theta, and psi Euler angles.
Biaxial 2 A version of the Biaxial layer that allows two different theta
Euler angles, theta X and theta Y, to compensate for small X and
Y tilts so that phi does not need to be rotated by 90°.
Cauchy_Extended A more advanced form of the Cauchy layer with extra terms,
such as a λ-6 term (“D”), an “IR” pole term, and a “Cody
Amplitude”.
Cauchy_Wvl A simple form of the Cauchy layer with only the real-part (index
of refraction) parameters, where the index can be specially
defined at a specific “design” wavelength instead of the usual
Cauchy “A” parameter.
Coupled A layer that inherits its optical constant values from a layer
defined elsewhere in the model layer stack. This is used when
two or more layers (perhaps in an alternating stack) are
expected to have exactly the same optical-constant values
during the fit.
EMA-Coupled A layer that combines the EMA and Coupled layers together in
order to form an EMA layer that mixes the optical constants of
two materials together, where one of the materials (Material
#1) is a Coupled layer which gets its optical constants from a
separate layer in the model.
Full Tensor An anisotropic layer that can represent the real and imaginary
parts of all nine elements of the dielectric-function tensor using
constituent material files and which can also rotate the
dielectric tensor using standard Euler angles.
GenOsc 3D An anisotropic layer that can combine oscillators of arbitrary
eigen-dielectric polarizability to create every physically possible
dielectric function tensor. It can be used to model monoclinic
and triclinic materials.
Glass Catalog A layer that allows easy representation of glass materials from
various manufacturers by using the standard three-term
Sellmeier equations given in glass catalogs.
Graded* A layer that allows the optical constants described by a material
file to incrementally vary across the thickness of the layer in a
specified way. Most layers can be converted to a Graded layer
by right-clicking and selecting “Grade Layer”.
KK Material A layer that allows the optical constants of either ε1 or ε2 to be
(Kramers-Kronig) described by a material file and then integrated using the
Kramers-Kronig transform to obtain the other dielectric-function
component.
Porous EMA An EMA layer that is used to represent a porous material whose
pores are being filled with a solvent vapor (such as water or
toluene) in porosimetry experiments.
Sellmeier A simple dispersion layer useful for modeling dielectric materials
that is slightly more advanced and more physically realizable
than a Cauchy layer.
Temperature* & A layer that contains optical constants of a material as a function
Composition* of temperature and/or composition. (Refer to Section 5.3.)
User Defined A layer that lets the user enter a custom equation for the optical
constants of a material.

CompleteEASE Software Manual Reference – Layer Types • 10-419

 Virtual Substrate A layer for dynamic (in-situ) measurements that allows the
underlying sample structure to be approximated by a single
interface. The optical constants at this interface are called the
“pseudo substrate”.
WvlByWvl A layer that allows optical constants to be individually
(Wavelength by determined on a wavelength-by-wavelength basis. This layer
Wavelength) does not enforce Kramers-Kronig consistency, can result in
spectral discontinuities, and inherently contains a maximum
number of fit parameters. Thus, it should be used with caution.
Generally, a B-Spline layer, which can be Kramers-Kronig
consistent and guarantees continuous results, should be used
instead.
Zero A layer with optical constants ε1 = 0 and ε2 = 0 for all
wavelengths that is mostly used with the Full Tensor layer to set
certain elements of the dielectric tensor to zero.

10.1. Tabulated
Most of the layers in the CompleteEASE material file library are standard tabulated
optical-constant lists. The optical constants for this type of layer are represented as
a table of optical-constant values versus wavelength. The table can contain the real
and imaginary parts of a material’s dielectric function (1 and 2) or the material’s
complex index of refraction (n and k). If you find optical constant tables in literature
and wish to use them in CompleteEASE, you can type the table into a text file in the
correct format (tab delimited), give it the “.mat” extension, and then open it as a
layer in a CompleteEASE model.
Figure 10-1 shows how typical tabulated layers are represented in the model.
Unlike the rest of the layer types described in this chapter, these have no user-
adjustable parameters to define the optical properties, making tabulated materials
the simplest layer type to use as only the thickness of the layer can be varied.

Figure 10-1. A model constructed of standard tabulated layers. Each set of optical constants
is described by an underlying table of optical-constant values versus wavelength.

10.2. Cauchy
The Cauchy layer is the most commonly used layer for determining the optical
constants of a transparent or partially transparent film (such as dielectrics and
semiconductors below the fundamental bandgap). Over part of the spectral range,
the optical constants of these materials can be represented by an index that varies
slowly as a function of wavelength and an exponential absorption tail. The index of
refraction of the Cauchy layer is represented by an inverse power series containing
only even-power terms, and the extinction coefficient is represented by a simple
exponential tail (called the Urbach absorption tail).
The Cauchy is a dispersion layer that describes the index of refraction using a three-
term equation, as shown in Equation 10-1, where λ is the wavelength in μm. (For
the Cauchy layer, the A, B, and C parameters always correspond to a wavelength in

10-420 • Reference – Layer Types CompleteEASE Software Manual

 μm regardless of the spectral light units that CompleteEASE is configured to use.
Since the Cauchy dispersion equation is fundamentally an energy equation, there is
usually a scaling factor of 1.240 eV∙μm that is used for conversion between
wavelength in μm and energy in eV. In the version of the Cauchy equation in
CompleteEASE shown in Equation 10-1, however, there is no scaling factor included,
and thus the scaling factor is assumed in the values of the A, B, and C parameters.)
Equation 10-1

𝐵 𝐶
𝑛(𝜆) = 𝐴 + 2
+ 4
𝜆 𝜆

In addition to the three terms for index of refraction (n), there are three extra
(optional) terms to describe an Urbach absorption tail, as shown in Equation 10-2,
where α is the “k amplitude”, the amplitude of the absorption tail; β is the
“Exponent”, which scales the shape of the absorption tail; γ is the “Band Edge”,
which is measured in μm in Equation 10-2; and λ is the wavelength in μm. (Once
again, Equation 10-2 assumes both a band edge and wavelength in μm; however,
CompleteEASE allows entry of the band edge in whichever light units are
configured, and it will automatically modify the scaling factor in Equation 10-2
accordingly.) The Urbach absorption is optional and is only used if the material
being measured is not completely transparent across the whole spectral range and
captures the onset of a stronger absorption feature outside the measured spectral
range. Note that the n and k values in the Cauchy layer are described independently
of one another entirely; in other words, no Kramers-Kronig consistency is enforced.
Equation 10-2

1.24 1.24
𝑘(𝜆) = 𝛼 exp (𝛽 ( − ))
𝜆 𝛾
Figure 10-2 shows the Cauchy layer as displayed in the CompleteEASE model, and
Figure 10-3 shows an example graph of Cauchy optical constants. The “A”, “B”, and
“C” parameters are variable fit parameters that determine the index (n) dispersion.
The Urbach parameters are found under the +Urbach Absorption Parameters
section. The “k Amplitude” and “Exponent” are fit parameters for determining the
shape of the extinction coefficient (k) dispersion. The “Band Edge” parameter can
be set manually, but it is not a fit parameter since it is entirely correlated with the
“k Amplitude” parameter. (In other words, the extinction coefficient is equal to “k
Amplitude” at the “Band Edge”, which corresponds to k(γ) in Equation 10-2.) For
examples of data analysis using a Cauchy layer, refer to Section 3.3, Section 3.5, and
the “Transparent Region Analysis” part of Section 4.3.

Figure 10-2. Cauchy dispersion layer in CompleteEASE with Urbach Absorption Parameters
section expanded.

CompleteEASE Software Manual Reference – Layer Types • 10-421

 Figure 10-3. Graph of the Cauchy-layer optical constants for the values shown in Figure 10-2.

10.3. B-Spline
The B-Spline layer describes the optical constants versus wavelength of a material
using a series of basis-spline functions, which rely on a series of control points,
called “nodes”, whose positions are fixed in photon energy (eV). The approximate
spacing of the nodes is controlled by the “Resolution (eV)” value, and the resulting
number of nodes within the current spectral range is listed after this value. The
basis-spline curves are used to interpolate between the control points10,11.
B-Spline curves are defined by a series of nodes, which have a fixed position and a
variable amplitude. Basis polynomial functions at each node are defined by the
recursive B-Spline formula12 in terms of the positions of the adjacent nodes.
Summing the basis functions together (weighted by the amplitude of each node)
produces the resulting curve. B-Spline curves have many desirable properties for
modeling dielectric functions:
• B-Spline curves and their derivatives (up to the spline order minus one) are
continuous. (The B-Spline layer uses third-order B-Splines.)
• The B-Spline node amplitudes affect only the “local” shape of the curve;
e.g., changing a node amplitude in the UV does not affect the curve in the
visible spectrum.
• B-Spline curves exhibit the “convex hull” property; e.g., if all the node
amplitudes are non-negative, then the B-Spline curve is also non-negative,
thus avoiding nonphysical, negative 2 values.
• Since the basis functions depend only on the node positions, the node
amplitudes which define the resulting curve are linearly independent,
which greatly increases computation efficiency.
• The Kramers-Kronig (KK) integral can be applied to the B-Spline recursion
formula to generate KK-consistent basis functions and thus KK-consistent
dielectric functions; i.e., the 1 curve is defined by the KK transform of the
2 curve, which is built with basis-spline functions.

10
B. Johs and J. S. Hale, “Dielectric Function Representation by B-Splines”, Phys. Status Solidi A 205 (2008).
11
J. Mohrmann, T. E. Tiwald, J. S. Hale, J. N. Hilfiker, and A. C. Martin, “Application of a B-spline Model Dielectric
Function to Infrared Spectroscopic Ellipsometry Data Analysis”, J. Vac. Sci. Technol. B 38 (2020).
12
W. Cheney and D. Kincaid, Numerical Mathematics and Computing, 5th ed. (2004), Chap. 9, pp. 419–436.

10-422 • Reference – Layer Types CompleteEASE Software Manual

 Figure 10-4 shows the B-Spline layer in the Model panel of CompleteEASE. The
“Resolution (eV)” field defines the nominal spacing of the nodes, which are equally
distributed by this resolution in eV over the selected spectral range of the data set.
The number of nodes and the range over which they are distributed are shown to
the right. Clicking the “Draw Node Graph” command plots the positions and
amplitudes of the B-Spline nodes, along with the calculated dielectric function of
the layer (Figure 10-5). The “Fit Optical Constants” setting turns ON or OFF all the
optical-constant fitting parameters for the layer (e.g., node amplitudes and
positions, pole amplitudes, etc.). If “Use KK Mode” is set to OFF, then separate and
unrelated B-Spline curves are calculated and fit for 1 and 2. When “Use KK Mode”
is set to ON, a B-Spline curve is used to define 2, and then 1 is calculated from the
KK transform of 2 (plus contributions from absorption outside the measured
spectral range). Figure 10-6 shows the fully expanded B-Spline layer with “KK
Mode” turned ON. Figure 10-7 shows the node positions when “KK Mode” is turned
ON. Notice that there are nodes to describe 2, along with a few nodes outside the
measured spectral range to help with the Kramers-Kronig integration (called “tie-
off” nodes).

NOTE: The B-Spline layer will automatically recalculate node positions and
interpolate node amplitudes if the selected spectral range of the experimental data
is adjusted, if the node resolution is changed, or if “KK Mode” is toggled ON or OFF.

Figure 10-4. B-Spline layer in CompleteEASE with “KK Mode” turned OFF.

Figure 10-5. B-Spline layer node graph with “KK Mode” turned OFF.

CompleteEASE Software Manual Reference – Layer Types • 10-423

 Figure 10-6. Fully expanded B-Spline layer in CompleteEASE with “KK Mode” turned ON.

Figure 10-7. B-Spline layer node graph with “KK Mode” turned ON.

NOTE: In the graph of Figure 10-7, the solid red and green lines for “e1” and “e2”
represent the reference material used to initialize the B-spline layer and might not
match the “Spline Model” curves (black dotted lines) after fitting experimental data.
This is expected behavior and is not something to be concerned about necessarily,
because the dielectric function implied by your fit results should be a more accurate
representation of the material you are fitting than any reference optical constants
that you used as a starting point.

To see examples of data analysis using the B-Spline layer, refer to Section 4.3. All
the available options for the B-Spline layer are shown in Figure 10-6 for a B-Spline
with “KK Mode” OFF and Figure 10-8 for a B-Spline with “KK Mode” ON. A brief
description of each option follows:

10-424 • Reference – Layer Types CompleteEASE Software Manual

 Figure 10-8. Fully expanded B-Spline layer with all available options shown, with “KK Mode”
turned ON.

Resolution (eV)
The resolution (also called the “node spacing”) defines the uniform spacing of the
B-Spline nodes across the spectrum in eV. The resolution must be small enough to
capture variation in the experimental data but large enough to avoid matching
thickness interferences or noise. (A warning will be shown if unusually small
resolution values are entered.) The resolution can be adjusted in specific, smaller
subregions of the spectrum by using the “Node Spacing Spectral Regions” option
described in the +Nodes section below.

Draw Node Graph

This shows a graph of the node values across the spectrum and the interpolated
B-Spline optical constants. This is useful to view whether the node resolution is
adequate to resolve the spectral features of a reference material. Example node
graphs are shown in Figure 10-5 and Figure 10-7 above.

CompleteEASE Software Manual Reference – Layer Types • 10-425

 Fit Optical Constants
Turn this ON to allow the B-Spline layer to be able to fit the material optical
constants. In other words, this option allows the node amplitude values (found
under the “Show Nodes” option in +Nodes) to vary to match the experimental data.
After fitting is completed, this can be turned OFF to hold the final optical constant
values fixed in order to adjust other parts of the model. This is particularly
important to use after determining the optical constants of a substrate with a
B-Spline before analyzing any thin-film measurements.

Use KK Mode
When “Use KK Mode” is set to OFF, separate, unrelated B-Spline curves are
calculated and fit for 1 and 2. When “Use KK Mode” is set to ON, a B-Spline curve
is used to define 2, and 1 is calculated from the KK transform of 2 (plus
contributions from absorption outside the measured spectral range). This option
adds the additional +Kramers-Kronig menu. KK Mode is being used when “Use KK
Mode” is turned on and when “(In Use)” appears to the right of the “Use KK Mode”
option, indicating that KK Mode is being used. If “(Not In Use)” appears to the right
of the “Use KK Mode” option, this means that there are not enough nodes to allow
the KK integration to be calculated, and therefore KK Mode is not being used, even
though “Use KK Mode” is set to ON. In order to activate KK Mode in this case,
decrease the “Resolution (eV)” value to add more nodes to the spectrum; KK Mode
requires at least four nodes to be used.

+Kramers-Kronig
This section is only visible when “Use KK Mode” is set to ON. It contains the variable
parameters that are only applicable when KK Mode is enabled, such as “Einf”, “IR
Amp”, “IR Br”, and extra nodes outside the measured range (called “tie offs”) to
accommodate the KK transformation and determine the correct 1 curve. The
options in the +Kramers-Kronig section are shown in Figure 10-9.

Figure 10-9. Kramers-Kronig section of the B-Spline layer.

E Inf
This represents the 1 value “at infinity”. When in KK mode, this accounts for the
effects of absorption far outside the measured spectral range by adding a constant
offset to the 1 curve.

10-426 • Reference – Layer Types CompleteEASE Software Manual

 IR Amp
This value is the amplitude of a zero-energy oscillator which accounts for the effects
of absorption at low energies outside of the measured spectral range, typically due
to free-carrier absorption (i.e., a Drude term). If the “IR Br” parameter is fixed at
zero, the “IR Amp” parameter will act like an IR pole.

IR Br
This value is the broadening of the zero-energy oscillator mentioned in “IR Amp”
above. By default, this value is fixed at zero in order to describe a pole which
accounts for strong IR absorption outside the measured spectral range. Allowing
this value to be nonzero turns the IR pole into a Drude absorption tail, which might
be useful for some metallic dielectric functions, but it can lead to unwanted
correlation with the tie-off nodes.

Use Default Tie Off Behavior

“Tie Offs” are the nodes which account for absorption outside the spectral range
when in KK Mode. Similar to a pole, the tie offs describe a single, strong feature that
accounts for all absorptions outside the measured spectral range. The tie offs do
not represent the actual absorption in that range but only the effect it has on the
real part of the dielectric function within the measured spectral range. In contrast
to a pole, the tie-off absorption can reach into to the measured spectral range to
account for strong absorption at the edge of the measured spectral range. The
“default” tie-off behavior (which specifies default tie-off node positions, with
amplitudes being fit for each node) is adequate for most applications. Turning “Use
Default Tie Off Behavior” to OFF allows for complete flexibility for the user to
choose the tie-off positions and amplitudes. This is typically only needed if there is a
strong absorption feature right outside the measured spectral region that is
dramatically influencing the optical constants. In this case, generally only “Tie Off
n+1” at max+E needs to have its energy position fit.

View Tie Off Positions

The positions and amplitudes of the tie offs can be set and defined as fit
parameters. The positions are defined relative to the ends of the spectral range
and relative to each other to guarantee monotonically increasing node positions.
Five node positions are defined above the spectral range and five below the
spectral range. For example, Tie Off n+1 is placed at “max+E”, which is the
maximum energy value of the spectral range plus the blue-underlined text value of
energy in eV; thus if the spectral range ends at 4.0 eV and Tie Off n+1 has a value of
0.5 eV, then it would be placed at 4.5 eV. The amplitudes of the first and last two tie
offs are fixed at zero, while the other six tie-off node amplitudes can be fit. (Each
tie-off amplitude is indicated by a “spline_e2” fit parameter located to the right of
its associated tie-off position.)

+Nodes
This section contains all the parameters that affect node amplitude values and node
position values. The options in the +Nodes section are shown in Figure 10-10.

CompleteEASE Software Manual Reference – Layer Types • 10-427

 Figure 10-10. Nodes section of the B-Spline layer.

Init. Values
Entering values for n or k in the “Init. Values” fields will reset all node values across
the entire spectrum to these constant values. Usually this option should only be
used when setting up the B-Spline layer before the first fit or with a Global Fit if
good starting values for n and k are not known.

Starting Mat.
Loading a material file in the “Starting Mat.” field will reset all the node values to
match this reference material file across the spectrum. This can be an effective way
to seed reasonable “starting” node values into the B-Spline layer. Generally, a
B-Spline layer will not converge to the correct solution without good starting values.
When using the “Parameterize Layer to B-Spline” feature (see Section 9.5), the
material file being parameterized is added as the “Starting Mat.” and can thus be
replaced here if needed.

Force E2 Positive
Setting this to ON prevents 2 nodes within the measured spectral range from going
below zero, which is unphysical behavior for the 2 dielectric function. The tie-off
nodes outside the spectral range are still allowed to go below zero to accommodate
effects outside the measured spectral range, so it is possible that 2 may still go
negative at the edge of the spectrum. While forcing 2 positive maintains “physical”
dielectric functions, sometimes allowing slightly negative 2 values improves the fit
convergence algorithms by allowing flexibility for noise in the data.

Assume Transparent Region

When a material is transparent in a certain region, its 2 dielectric function will also
be zero in that region. Thus, for known transparent regions, a bandgap or spectral
region can be specified for which 2 is fixed at zero, meaning that all nodes with
positions inside the specified transparent region will have an amplitude value of
zero and will not be fit with the rest of the node amplitudes. If “Assume Band Gap”
is ON, then the transparent region is specified by a “Band Gap”; specifying the
“Band Gap” means that all nodes at energies below the band gap will be fixed at
zero amplitude. If “Assume Band Gap” is OFF, then a transparent region can be
specified in whichever wavelength units CompleteEASE is configured to use.

10-428 • Reference – Layer Types CompleteEASE Software Manual

 Show Nodes
This option shows each node value and position in the B-Spline, so that you can see
where each individual node is located and what specific amplitude it has.
Sometimes, when changes are made to the B-Spline model that affect the node
values, “Show Nodes” may need to be turned OFF and then ON again to refresh the
view of these node values after making changes. Performing a fit, however, will
automatically refresh these values.

Spline_e2(photon energy) / Spline_e1(photon energy)

Each node that is shown is given as either the ε1 (e1) or ε2 (e2) value at a specific
photon energy. These nodes are positioned equally across the spectrum according
to the specified “Resolution (eV)” value except in special “Node Spacing Spectral
Ranges”. When KK Mode is OFF, there is a complete set of both ε1 and ε2 nodes.
When KK Mode is ON, only the ε2 nodes are listed, as the ε1 curve is calculated by
the Kramers-Kronig transformation.

Node Spacing Spectral Ranges

A single, uniform node spacing (resolution) across the entire spectrum may not be
optimal for all materials. For this reason, CompleteEASE allows special spectral
ranges to be added to the overall B-Spline model with smaller or larger node
spacing than the default “Resolution (eV)” specified above. GaP is a good example
of this, as shown in Figure 10-11, where the optical constants have no spectral
features below the bandgap (above 500 nm) but quickly change at shorter
wavelengths. It may be desirable to reduce the total number of nodes by having a
wide node spacing at longer wavelengths and narrow node spacing at shorter
wavelengths. Press Add to the right of “Node Spacing Spectral Ranges” and you will
be presented with three dialog boxes. The first will ask for the minimum
wavelength range, the second for the maximum wavelength range, and the third for
the custom node spacing (resolution) to be used within this range. For GaP, a node
resolution of 0.5 eV was set for the whole spectral range with “Resolution (eV)”
while a custom node spectral range from 190-500 nm of 0.1 eV was added to better
resolve the sharp features in that region, as shown in the model in Figure 10-12.
The final node graph is shown in Figure 10-13.

Figure 10-11. Optical Constants for GaP from ultraviolet to near infrared.

CompleteEASE Software Manual Reference – Layer Types • 10-429

 Figure 10-12. B-Spline for GaP with custom node spacing between 190-500 nm (0.1 eV
resolution).

Figure 10-13. GaP optical constants described by a B-Spline layer with 0.5-eV node spacing
above 500 nm (below 2.48 eV) and 0.1-eV spacing below 500 nm (above 2.48 eV).

+Advanced
A few additional options are hidden in the +Advanced section, as shown in Figure
10-14, to avoid unneeded complexity since these are only used for special purposes.

Figure 10-14. Advanced section of the B-Spline layer.

Show Parameters in Fit

This option allows all node positions and all node-amplitude fit parameters to be
displayed in the Fit Results. By default, node positions and amplitudes are hidden to
reduce the number of fit parameters displayed in the Fit Results, as it can be
overwhelming and unnecessary. However, when using “Use Previous Results” in the
model, it will be necessary to include all parameters in the Fit Results. (See Section
9.6 for information about how “Use Previous Results” is used.)

10-430 • Reference – Layer Types CompleteEASE Software Manual

 Pre-Fit when Changing Wavelengths
It is difficult to maintain the optical constants of a B-Spline when changing the
wavelengths in the spectral range because of the effect of tie-off positions outside
the selected spectral range on the optical constants. When the spectral range is
modified, the tie-off positions and amplitude values will adjust to the overall optical
constant shape. When this feature is turned ON, the B-Spline will save off its optical
constants before adjusting to the new spectral range that can be used to “pre-fit”
the new nodes to match. This is needed because of the automatic recalculation of
node positions when the spectral range is modified. This does not ensure the
optical constants described by the B-Spline nodes will not slightly adjust when
changing wavelengths, but it helps minimize the adjustment by pre-fitting to the
previous values. (This is generally applied when taking a B-Spline that was
constructed based on one subregion of the data-set spectral range and then using it
on either a larger or the full spectral range of the data set when exact node position
values need to be maintained.)

Fix Node Bounds When All Wavelengths Selected

This is meant to be used in conjunction with either “Query Remote System for
Optical Constants” (usually with M-2000 ellipsometers) or “Use Previous Results”.
Because of the automatic calculation of node positions based on the spectral range,
if data is being used from another ellipsometer with a slightly different spectral
range, exactly matching node positions are needed. This option will fix the bounds
of the spectral range for the purposes of node calculation so that node position
values match exactly for data sets from different ellipsometers. This keeps all the
node positions consistent even though the spectral ranges of the data sets from
each ellipsometer may vary slightly. (See Section 9.6 for information about how
“Use Previous Results” is used.)

Query Remote System for Optical Constants

This option is used with multiple ellipsometers to “feed” the optical constants from
a remote system (e.g., previous ellipsometers in a chain of ellipsometers) to the
B-Spline layer on the current ellipsometer when fitting layers that need this
updated information from fits performed on previous ellipsometers. This means the
B-Spline will query the previous ellipsometer (remote system) for starting values.

10.4. Gen-Osc
The Gen-Osc layer is a generalized oscillator material that allows the summation of
different oscillator line shapes. Figure 10-15 shows a Gen-Osc layer with two
oscillators of different types, along with the associated graph of the resulting optical
constants. For examples that involve using the Gen-Osc layer to analyze data, refer
to Section 4.4.

CompleteEASE Software Manual Reference – Layer Types • 10-431

 Figure 10-15. Gen-Osc dispersion layer with both a Cody-Lorentz and a Gaussian oscillator.

Show Dialog
Press Show Dialog (as shown in Figure 10-15) to bring up the Gen-Osc layer editor
to manipulate the Gen-Osc fit parameters while visualizing the changes to the
optical constants graphically. This “Edit Layer” dialog allows the oscillator
parameters to be matched to reference optical constants. This dialog box functions
in a similar manner as the “Parameterize Layer” dialog box and is shown in Figure
10-16.

Figure 10-16. “Edit Layer” dialog box to view and manipulate the Gen-Osc parameters.

10-432 • Reference – Layer Types CompleteEASE Software Manual

 The fit parameters of each oscillator in the “Edit Layer” dialog can be manipulated
directly on the graph by dragging the small gray “drag boxes” located at the
significant points of each oscillator. To display the “drag boxes” on the graph for an
oscillator, simply select it by left-clicking on the number to the right of its oscillator
type name (e.g. “2.” Next to “Type = Gaussian” in Figure 10-16 above).

NOTE: The oscillators in the Gen-Osc can be selected or deleted by pressing the
number to their left. Use care when clicking on the oscillator number, as each
mouse-button has a different function: Left-clicking on the oscillator selects the
oscillator and shows the gray “drag boxes”. Right-clicking opens a dialog asking
whether to delete the oscillator from the list.

Real Part (+e1 Components)

There are four parameters at the top of the Gen-Osc Layer Parameters panel that
are used to describe the real part of the dielectric function. These include the Einf,
UV Pole Amplitude, UV Pole Energy, and IR Pole Amplitude.

Einf
“Einf”, or “” as it is often referred to in scientific literature, is a purely real
constant added to 1. This represents the 1 value “at infinity”, which accounts for
the effects of absorption far outside the measured spectral range.

UV Pole Amp. & UV Pole En.

Poles, which the dielectric function will approach asymptotically, are equivalent to
Lorentz oscillators with zero broadening. Their position should remain outside the
“fit” spectral range, as they describe dispersion created by absorptions that occur
outside the measured spectral range. As such, they only affect the real part of the
dielectric function (1). The UV pole describes absorptions that occur at higher
energies outside the measured spectral range. A pole is defined mathematically by
Equation 10-3 where uv_pole(E) is 2 of the UV pole as a function of the energy (E) in
eV, En is the center energy of the pole (“UV Pole En.”) in eV, and An is the amplitude
of the pole (“UV Pole Amp.”) in eV2.
Equation 10-3

𝐴𝑛
𝜀uv_pole (𝐸) = 2
𝐸𝑛 − 𝐸 2

IR Pole Amp.
The IR Pole describes absorptions that occur at lower energies outside the
measured spectral range. The IR pole is always centered at 0 eV (which is
technically implemented as 10-8 eV for calculation purposes), thus the IR pole is
defined as in Equation 10-4 where ir_pole(E) is 2 of the IR pole as a function of the
energy (E) in eV and An is the amplitude of the pole (“IR Pole Amp.”) in eV2.
Equation 10-4

−𝐴𝑛
𝜀ir_pole (𝐸) =
𝐸2

CompleteEASE Software Manual Reference – Layer Types • 10-433

 Imaginary Part (+e2 Components)
Within the Gen-Osc layer, there are several commands for adding and manipulating
oscillators. These options, Add, Delete, Delete All, and Sort, are located in the
“Oscillator Menu”. There are also several commands for turning on or off the fit
parameters of all oscillators at once. These options, All, None, Amp., Br., and En.,
are located in the “Fit Menu.”

Oscillator Menu
The Oscillator Menu contains commands to manipulate the list of oscillators by
adding, deleting, and sorting them.

Add
This command adds a new oscillator to the list. If there are no oscillators present in
the list, then a dialog box will appear allowing you to select the oscillator that you
want to add as shown in Figure 10-17. If there are already oscillators present in the
list, then Add will just add another copy of the last oscillator type that was added.
In order to change the oscillator type, simply click on the blue underlined text of the
oscillator type name (e.g. Gaussian), and the dialog box to select the oscillator type
will appear as shown in Figure 10-17.

Figure 10-17. Choices for oscillator types within the Gen-Osc layer.

Delete
This command deletes the oscillator from the list which was last selected. To select
an oscillator, simply left-click on the number to the left of the oscillator type name.
To quickly delete a specific oscillator, you can also right-click on the number to the
left of the oscillator type name (e.g. 1:) to open a dialog box asking whether to
delete the oscillator, or you can hold the SHIFT key down and right-click on the
number to the left of the oscillator type to delete an oscillator without warning.

Delete All
This command opens a dialog box asking whether to delete all the oscillators
present in the list.

Sort
This command toggles the visibility of the “Sort Menu”, as seen in Figure 10-18,
which allows you to move each oscillator to a new position in the list or sort all

10-434 • Reference – Layer Types CompleteEASE Software Manual

 oscillators by a specific parameter value. Move Up and Move Down will move the
selected oscillator one position up or down in the list respectively. (To select an
oscillator, simply left-click on the number to the left of the oscillator type name.)
Sort by En. will sort the oscillator list by each oscillator’s center energy (its “En#”
parameter value), and Sort by Amp. Will sort the oscillator list by each oscillator’s
amplitude (its “Amp#” parameter value). Left-clicking on either sort command will
sort the oscillator list in ascending order whereas right-clicking on either sort
command will sort the oscillator list in descending order.

Figure 10-18. Choices for oscillator types within the Gen-Osc layer.

Changing the order of oscillators in the oscillator list by using the “Sort Menu”
should not change the resulting optical-constant values. Its only purpose is to allow
easier visualization of which oscillators are present in the Gen-Osc layer. Typically,
when constructing a Gen-Osc layer model, oscillators are added in arbitrary order,
and thus navigating a long, unsorted list of oscillators can become tedious.

NOTE: Since the fit parameters associated with an oscillator are named with the
number of the oscillator’s position in the list (e.g. “Amp2”, “En3”, “Br1”), sorting or
moving any oscillator will rename all of its associated fit parameters to reflect its
new position in the list. Changing oscillator fit parameter names will break
parameter references if they are used with Parameter Coupling, Derived
Parameters, Custom Equations, or Multi-Sample Analysis. Use caution when moving
or sorting oscillators in the oscillator list. If you happen to break the parameter
relationships by sorting oscillators, the Reset List (see Section 9.2) can be used to
restore previous parameter relationships if a fit had been performed before sorting
the oscillators.

Fit Menu
The Fit Menu contains commands to turn on or off some or all of the fit parameters
of all oscillators in the list for convenience.

All & None

The All command turns on all the fit parameters of all the oscillators in the list
(except parameters found in nonexpanded sub-menus, such as the Urbach
parameters of the Cody-Lorentz oscillator). The None command turns off all the fit
parameters of all oscillators in the list (including parameters found in sub-menus).

Amp., Br., & En.

These commands are toggle switches that turn on or off each respective parameter
for all oscillators in the list.

Lorentz Oscillator
This is a version of the classic Lorentz oscillator model described in Equation 10-5
where Lorentz(E) represents the complex dielectric function contribution of the
oscillator as a function of photon energy (E) in eV, En is the center energy of the

CompleteEASE Software Manual Reference – Layer Types • 10-435

 oscillator in eV, Brn is the broadening of the oscillator in eV,
An is the unitless amplitude of the oscillator, and i is the imaginary unit.
Equation 10-5

𝐴𝑛 𝐵𝑟𝑛 𝐸𝑛
𝜀Lorentz (𝐸) = 2
𝐸𝑛 − 𝐸 2 − i𝐵𝑟𝑛 𝐸
In this style of the Lorentz oscillator, An approximately equals 2 at its peak value,
and Brn is approximately the full width at half-maximum (FWHM). Note that the
maximum position shifts to lower energies compared to the center energy En as
broadening (Brn) increases.

Lorentz Im(Amp) Oscillator

Adding an imaginary-amplitude term to the Lorentz oscillator allows 2 to be
asymmetric around the center energy, En. Note, however, that 2 will become
negative when the imaginary term Ãn is nonzero. This means that another oscillator
must be added to force 2 ≥ 0 so that the total, summed Gen-Osc function is
actually physical. The Lorentz oscillator with imaginary-amplitude term is shown in
Equation 10-6 where Lorentz_iAmp(E) represents the complex dielectric function
contribution of the oscillator as a function of photon energy (E) in eV, En is the
center energy of the oscillator in eV, Brn is the broadening of the oscillator in eV, An
is the unitless amplitude of the oscillator, Ãn is the unitless imaginary amplitude of
the oscillator, and i is the imaginary unit.
Equation 10-6

(𝐴𝑛 + i𝐴̃𝑛 )𝐵𝑟𝑛 𝐸𝑛

𝜀Lorentz_iAmp (𝐸) =
𝐸𝑛 2 − 𝐸 2 − i𝐸𝐵𝑟𝑛
In this oscillator, An approximately equals 2 at its peak value, and Brn is
approximately the full width at half-maximum (FWHM).

Harmonic Oscillator
The Harmonic oscillator function is similar to that derived from perturbation theory
in quantum-mechanical treatments of single-electron transitions, and it is typically
written in the literature as a two-term sum13. The Harmonic oscillator is shown in
Equation 10-7 where Harmonic(E) represents the complex dielectric function
contribution of the oscillator as a function of photon energy (E) in eV, En is the
center energy of the oscillator in eV, Brn is the broadening of the oscillator in eV, An
is the unitless amplitude of the oscillator, and i is the imaginary unit.
Equation 10-7

𝐴𝑛 𝐵𝑟𝑛 1 1
𝜀Harmonic (𝐸) = ( + )
2 1 1
𝐸𝑛 − 𝐸 − i 𝐵𝑟𝑛 𝐸𝑛 + 𝐸 − i 𝐵𝑟𝑛
2 2
An approximately equals 2 at its peak value, and Brn is approximately the full width
at half-maximum (FWHM). When the broadening becomes much less than the
center energy (Brn << En), the Harmonic oscillator is effectively equivalent to the
Lorentz oscillator above.

13
For example, see S. Adachi, Optical properties of crystalline and amorphous semiconductors: materials and
fundamental principles,” (Kluwer Academic Publishers, Norwell, MA, USA, 1999), p. 64.

10-436 • Reference – Layer Types CompleteEASE Software Manual

 Harmonic Im(Amp) Oscillator
Adding an imaginary-amplitude term to the Harmonic oscillator above allows 2 to
be asymmetric around the center energy En. Note, however, that 2 becomes
negative when the imaginary term Ãn is nonzero. This means that another oscillator
must be added to force 2 ≥ 0 so that the total, summed Gen-Osc is actually
physical. The Harmonic oscillator with imaginary-amplitude term is shown in
Equation 10-8 where Harmonic_iAmp(E) represents the complex dielectric function
contribution of the oscillator as a function of photon energy (E) in eV, En is the
center energy of the oscillator in eV, Brn is the broadening of the oscillator in eV, An
is the unitless amplitude of the oscillator, Ãn is the unitless imaginary amplitude of
the oscillator, and i is the imaginary unit.
Equation 10-8

(𝐴𝑛 + i𝐴̃𝑛 )𝐵𝑟𝑛 1 1

𝜀Harmonic_iAmp (𝐸) = ( + )
2 1 1
𝐸𝑛 − 𝐸 − i 𝐵𝑟𝑛 𝐸𝑛 + 𝐸 − i 𝐵𝑟𝑛
2 2
An approximately equals 2 at its peak value, and Brn is approximately the full width
at half-maximum (FWHM).

Gaussian Oscillator
The Gaussian oscillator produces a Gaussian line shape in 2 with a Kramers-Kronig
consistent line shape for 1. The Gaussian oscillator is shown in Equation 10-9
where Gaussian(E) represents the complex dielectric function contribution of the
oscillator as a function of photon energy (E) in eV, En is the center energy of the
oscillator in eV, Brn is the broadening of the oscillator in eV, An is the unitless
amplitude of the oscillator, i is the imaginary unit,  is a convergence series that
produces a Kramers-Kronig-consistent line shape for the 1 dielectric function14,15,
Z + and Z – (equivalent to Gaussian “Z-values”) are defined in Equation 10-10, and σn
(equivalent to Gaussian “standard deviation”) is defined in Equation 10-11.
Equation 10-9

𝜀Gaussian (𝐸) = 𝐴𝑛 (Γ(𝑍 − ) + Γ(𝑍 + ) + i[exp(−(𝑍 − )2 ) − exp(−(𝑍 + )2 )])

Equation 10-10

𝐸 + 𝐸𝑛 𝐸 − 𝐸𝑛
𝑍+ = 𝑍− =
𝜎𝑛 𝜎𝑛
Equation 10-11

𝐵𝑟𝑛
𝜎𝑛 =
2√ln 2
An approximately equals 2 at its peak value (~Enn). By defining σn as in Equation
10-11, Brn approximately equals the full width at half maximum value (FWHM).

Drude(RT) & Drude(NMu) Oscillators

The classic Drude model describes free-carrier effects on the dielectric response. Its
form is equivalent to a Lorentz oscillator with center energy equal to zero. There

14
See section 3 of D. De Sousa Meneses, M. Malki, & P. Echegut, J. Non-Cryst. Solids 351 no.2 (2006) 769-776.
15
K.-E. Peiponen and E. M. Vartiainen, Phys. Rev. B 44 (15) (1991) 8301-8303.

CompleteEASE Software Manual Reference – Layer Types • 10-437

 are two versions of the Drude oscillator in the Gen-Osc layer: Drude(RT) and
Drude(NMu).
Drude(RT) has resistivity, , in Ω∙cm and mean scattering time,, in femtoseconds as
fit parameters. It is described by Equation 10-12, where Drude(RT)(E) represents the
complex dielectric function contribution of the oscillator as a function of photon
energy (E) in eV, ħ is Planck’s constant divided by 2π with units of eV∙s, ε0 is the
permittivity of free space (vacuum dielectric constant) in F∙cm-1, n is the resistivity
in Ω∙cm, n is the mean scattering time in seconds (but entered into CompleteEASE
as femtoseconds), and i is the imaginary unit.16
Equation 10-12

−ℏ2
𝜀Drude(RT) (𝐸) =
𝜀0 𝜌𝑛 (𝜏𝑛 ∙ 𝐸 2 + iℏ𝐸)
Drude(NMu) has the carrier concentration, N, in cm-3; the carrier mobility, , in
cm2V-1s-1; and the carrier effective mass, m*, which is unitless and is given as a
multiple of the free-electron rest mass, as fit parameters. These parameters are
related to the resistivity, , and mean scattering time,, by Equation 10-13 where
me is the free-electron rest mass in kilograms and q is the elementary charge in
Coulombs. (Note that m*, , and N are all completely correlated, so the user will
have to choose one to fix and then let the other two fit. The usual choice is to fix m*
to a known value and then let  and N vary in the fit.)
Equation 10-13

me 𝑚∗ 1
𝜌= 2
=
𝑁q 𝜏 q𝜇𝑁
Substituting Equation 10-13 into Equation 10-12 for  and  yields the Drude(NMu)
equation, shown in Equation 10-14, where Drude(NMu)(E) represents the complex
dielectric function contribution of the oscillator as a function of photon energy (E)
in eV, ħ is Planck’s constant divided by 2π with units of eV∙s, ε0 is the permittivity of
free space (vacuum dielectric constant) in F∙cm-1, me is the electron rest mass in
kilograms, q is the elementary charge in Coulombs, Nn is the carrier concentration in
cm-3;  is the carrier mobility in cm2V-1s-1, m* (called “mstar” in CompleteEASE) is
the unitless carrier effective mass, and i is the imaginary unit.
Equation 10-14

−ℏ2 q2 𝑁𝑛 𝜇𝑛
𝜀Drude(NMu) (𝐸) =
𝜀0 (𝜇𝑛 𝑚𝑒 𝑚∗ ∙ 𝐸 2 + iqℏ𝐸)
(Note that in all of the definitions for the Drude oscillators, CompleteEASE uses the
pre-2019 values for ħ, the elementary charge q, and the permittivity of free space
ε0. The values of these constants were changed with the 2019 SI Units Redefinition.)

16
T.E. Tiwald, D.W. Thompson, J.A. Woollam, W. Paulson, R. Hance, Thin Solid Films 313-314 661 (1998).

10-438 • Reference – Layer Types CompleteEASE Software Manual

 Tauc-Lorentz & Cody-Lorentz Oscillators
The Tauc-Lorentz17 and Cody-Lorentz18 dispersion functions both model the main
absorption of amorphous materials using a broad Lorentzian line shape with zero
absorption below a defined bandgap energy. The primary difference between the
two functions is how they model absorption at energies slightly larger than the
bandgap energy. In this region the Tauc-Lorentz model follows the Tauc-law
absorption formula, given in Equation 10-15, while the Cody-Lorentz follows the
Cody absorption formula, given in Equation 10-16, where 2(E) is a function of the
photon energy (E) in eV and Eg is the defined bandgap energy in eV:
Equation 10-15

(𝐸 − 𝐸𝑔)2
𝜀2 (𝐸) ∝
𝐸2
Equation 10-16

𝜀2 (𝐸) ∝ (𝐸 − 𝐸𝑔)2
The Cody-Lorentz oscillator also includes an Urbach absorption term with two extra
fit parameters to model absorption at energies below Eg.

Tauc-Lorentz Oscillator
The equation for the Tauc-Lorentz oscillator function is given in Equation 10-17,
which depends on Equation 10-18 and Equation 10-19. In these equations,
Tauc-Lorentz(E) represents the complex dielectric function contribution of the
oscillator as a function of photon energy (E) in eV, Eon is the center energy of the
oscillator in eV, Egn is the bandgap energy of the oscillator in eV, Brn is the
broadening of the oscillator in eV, An is the amplitude of the oscillator in eV, and i is
the imaginary unit. Equation 10-19 is the Kramers-Kronig transformation, where PV
indicates the Cauchy Principal Value of the integral. The bandgap energies (Egn) of
any Tauc-Lorentz oscillator can be coupled to Eg1, the Eg of the first Tauc-Lorentz
oscillator in the oscillator list, by turning “Common Eg” to ON.
Equation 10-17

𝜀Tauc−Lorentz (𝐸) = 𝜀𝑛1 (𝐸) + i𝜀𝑛2 (𝐸)

Equation 10-18

𝐴𝑛 𝐸𝑜𝑛 𝐵𝑟𝑛 (𝐸 − 𝐸𝑔𝑛 )2 1

∙ , 𝐸 > 𝐸𝑔𝑛
2 2
𝜀𝑛2 (𝐸) = {(𝐸 2 − 𝐸𝑜𝑛 ) + 𝐵𝑟𝑛 2 𝐸 2 𝐸
0, 𝐸 ≤ 𝐸𝑔𝑛
Equation 10-19
∞
2 𝜔 ∙ 𝜀𝑛2 (𝜔)
𝜀𝑛1 (𝐸) = 𝑃𝑉 [ ∫ 2 𝑑𝜔]
𝜋 𝜔 − 𝐸2
𝐸𝑔𝑛

17
G.E. Jellison, Jr. and F.A. Modine, “Parameterization of the optical functions of amorphous materials in the interband
region,” Appl. Phys. Lett. 69, 371 (1996), Erratum, Appl. Phys. Lett. 69, 2137 (1996).
18
A.S. Ferlauto, G.M Ferreira, J.M. Pearce, C.R. Wronski, R.W. Collins, Xunming Deng, and Gautam Ganguly, “Analytical
model for the optical functions of amorphous semiconductors from the near-infrared to ultraviolet: Applications in thin
film photovoltaics,” J. Appl. Phys. 92, 2424 (2002).

CompleteEASE Software Manual Reference – Layer Types • 10-439

 While the Kramers-Kronig transformation is presented here as an integral for
brevity (Equation 10-19), CompleteEASE uses the complete analytical solutions to
the Kramers-Kronig integral as presented in Jellison and Modine17. (See especially
the Erratum to that publication.) In order to maintain ε2 ≥ 0, CompleteEASE
internally constrains Egn and Brn such that Egn < Eon and Brn < 2Eon when fitting
data. However, these values are not limited externally; thus, the user can enter any
value. Therefore, a range of external values of Egn and Brn that the user could enter
will produce the same dielectric function according to Equation 10-18.

Cody-Lorentz Oscillator
The Cody-Lorentz oscillator is similar to the Tauc-Lorentz in that it defines a
bandgap energy, “Eg”, and a Lorentzian absorption peak (with parameters “Amp”,
“Eo”, and “Br”). However, for the Cody-Lorentz, Equation 10-16 applies in the
region just above the bandgap energy. The Cody-Lorentz also defines two
transition-energy fit parameters: “Ep” and “Et”. “Ep” is the offset from the bandgap
energy where the absorption transitions from Lorentzian to Cody behavior, and “Et”
is the offset from the bandgap energy where the absorption transitions from Cody
to Urbach behavior. The fit parameter “Eu” defines the exponential rate of the
Urbach absorption. See Figure 10-19 below for a graphical representation of these
fit parameters.
The equation for the Cody-Lorentz oscillator function is given in Equation 10-20,
which depends on Equation 10-21, Equation 10-22, Equation 10-23, and Equation
10-24. In these equations, Cody-Lorentz(E) represents the complex dielectric function
contribution of the oscillator as a function of photon energy (E) in eV, Eon is the
center energy of the oscillator in eV, Egn is the bandgap energy of the oscillator in
eV, Epn is the relative Lorentz-to-Cody transition energy of the oscillator in eV, Etn is
the relative Cody-to-Urbach transition energy of the oscillator in eV, Eun is the
exponential rate of the Urbach absorption of the oscillator in eV, Brn is the
broadening of the oscillator in eV, An is the amplitude of the oscillator in eV, and i is
the imaginary unit. G(E), defined in Equation 10-22, is the near-bandgap function
which defines the Cody behavior of the oscillator. L(E), defined in Equation 10-23, is
the Lorentz oscillator function for the Lorentz behavior of the oscillator. Equation
10-24 is the Kramers-Kronig transformation, where PV indicates the Cauchy
Principal Value of the integral. The bandgap energies (Egn) of any Cody-Lorentz
oscillator can be coupled to Eg1, the Eg of the first Cody-Lorentz oscillator in the
oscillator list, by turning “Common Eg” to ON.
Equation 10-20

𝜀Cody−Lorentz (𝐸) = 𝜀𝑛1 (𝐸) + i𝜀𝑛2 (𝐸)

Equation 10-21

𝐸𝑔𝑡𝑛 = 𝐸𝑔𝑛 + 𝐸𝑡𝑛

𝐸𝑔𝑡𝑛 ∙ 𝐺(𝐸𝑔𝑡𝑛 ) ∙ 𝐿(𝐸𝑔𝑡𝑛 ) 𝐸 − 𝐸𝑔𝑡𝑛
∙ exp , 0 < 𝐸 ≤ 𝐸𝑔𝑡𝑛
𝜀𝑛2 (𝐸) = { 𝐸 𝐸𝑢𝑛
𝐺(𝐸) ∙ 𝐿(𝐸), 𝐸 > 𝐸𝑔𝑡𝑛
Equation 10-22

(𝐸 − 𝐸𝑔𝑛 )2
𝐺(𝐸) =
(𝐸 − 𝐸𝑔𝑛 )2 + 𝐸𝑝𝑛 2

10-440 • Reference – Layer Types CompleteEASE Software Manual

 Equation 10-23

𝐴𝑛 𝐵𝑟𝑛 𝐸𝑜𝑛 𝐸
𝐿(𝐸) = 2
(𝐸 2 − 𝐸𝑜𝑛 2 ) + 𝐵𝑟𝑛 2 𝐸 2
Equation 10-24
∞
2 𝜔 ∙ 𝜀𝑛2 (𝜔)
𝜀𝑛1 (𝐸) = 𝑃𝑉 [∫ 2 𝑑𝜔]
𝜋 𝜔 − 𝐸2
0

While the Kramers-Kronig transformation is presented here as an integral for

brevity (Equation 10-24), CompleteEASE uses the complete analytical solutions to
the Kramers-Kronig integral as presented in Ferlauto18.

Eo

Eg+Et

Eg

Eg

Ep

Figure 10-19. Parameter definitions for the Cody-Lorentz oscillator in Gen-Osc.

In Equation 10-21, G(E) represents the Cody behavior (defined in Equation 10-22)
while L(E) is a Lorentz oscillator (defined in Equation 10-23). The sum Eg + Ep is the
energy where the oscillator transitions from a Cody absorption (Equation 10-16) to
a Lorentzian absorption. Note that G(E) → (E – Eg)2/Ep2 when E → Eg+, and
G(E) → 1 when E >> Ep.

The sum Eg + Et is the energy that marks transition between Urbach absorption and
Cody absorption. It is important to note that there are differences between the
Ferlauto18 and CompleteEASE definitions. Ferlauto defines Et as an absolute energy,
while CompleteEASE defines it as relative to the gap energy, as shown in Equation
10-25.
Equation 10-25

𝐸𝑡𝐹𝑒𝑟𝑙𝑎𝑢𝑡𝑜 = 𝐸𝑔𝐶𝑜𝑚𝑝𝑙𝑒𝑡𝑒𝐸𝐴𝑆𝐸 + 𝐸𝑡𝐶𝑜𝑚𝑝𝑙𝑒𝑡𝑒𝐸𝐴𝑆𝐸

The Urbach absorption is fully Kramers-Kronig consistent; that is, the exponential
Urbach absorption in 2(E) has a Kramers-Kronig-transformed counterpart in the
1(E) function. The coefficient before the exponential in Equation 10-21 guarantees
that the 2 function transitions smoothly at the Eg + Et transition point. For the
Urbach absorption, Eu defines its exponential rate; specifically, Eg + Et is the energy
at which the Urbach absorption equals e-1 of its maximum value. The coefficient
before the exponential in Equation 10-21 guarantees that the Urbach exponential
function exactly matches G(E)∙L(E) at Eg + Et.

CompleteEASE Software Manual Reference – Layer Types • 10-441

 Tanguy Oscillator
Tanguy19,20 developed this expression for an oscillator as an analytical expression of
Wannier excitons, including bound and unbound states. (The oscillator function
makes no distinction between bound and unbound states.) The Tanguy oscillator
equation is shown in Equation 10-26, which depends on Equation 10-27, Equation
10-28, and Equation 10-29. In these equations, Tanguy(E) represents the complex
dielectric function contribution of the oscillator as a function of photon energy (E)
in eV, Egn is the fundamental bound-state energy (FBSE) of the oscillator in eV, Rn is
the unscreened binding energy of the oscillator, Brn is the broadening of the
oscillator in eV, An is the amplitude of the oscillator in eV2, gn is the unitless
screening factor of the oscillator, and i is the imaginary unit. Equation 10-29 gives
the definitions of the digamma (ψ) and gamma (Γ) functions, which are functions of
a complex variable. The FBSE (Egn) of any Tanguy oscillator can be coupled to Eg1,
the Eg of the first Tanguy oscillator in the oscillator list, by turning “Common Eg” to
ON.
Equation 10-26

𝐵̃𝑛 (𝐸) = 𝐸 + i𝐵𝑟𝑛

𝐴𝑛 √𝑅𝑛
𝜀Tanguy (𝐸) = [𝑔̃ (𝜉 (𝐵̃𝑛 (𝐸))) + 𝑔̃ (𝜉 (−𝐵̃𝑛 (𝐸))) − 2𝑔̃(𝜉(0))]
(𝐵̃𝑛 (𝐸))2
Equation 10-27

𝑔𝑛 𝜉 1
𝑔̃(𝜉) = −2𝜓 ( ) − − 2𝜓(1 − 𝜉) −
𝜉 𝑔𝑛 𝜉
Equation 10-28

2
𝜉 (𝐵̃𝑛 (𝐸)) =
̃ (𝐸) 𝐸 ∙ 𝑔𝑛 − 𝐵̃𝑛 (𝐸) 4
√𝐸 ∙ 𝑔𝑛 − 𝐵𝑛 +√ +
𝑅𝑛 𝑅𝑛 𝑔𝑛
Equation 10-29
𝑑 ∞
𝜓(𝑧) = ln Γ(𝑧) Γ(𝑧) = ∫0 𝑥 𝑧−1 𝑒 −𝑥 𝑑𝑥, ℜ(𝑧) > 0
𝑑𝑧

The parameter gn is a screening factor that can vary between 0 and  (excluding
zero). As g → , it approaches an unscreened Coulomb potential, and as g → 0, it
approaches total screening. In CompleteEASE, since gn is bound between 0 and ,
the gn parameter is not fit for directly. Instead, the fit parameter called “lng#” fits
for the natural logarithm of gn; thus, Ing = ln(gn). Substituting the relationship
gn = exp(ln(gn)) = exp(Ing) into Equation 10-27 and Equation 10-28 above yields
equations directly using the CompleteEASE fit parameter “Ing#”.

Psemi Oscillators (M0, M1, M2, M3, & Tri)

The Psemi-M0, Psemi-M1, Psemi-M2, Psemi-M3, and Psemi-Tri oscillators are a
subset of the more general Herzinger-Johs Parameterized Semiconductor Oscillator

19
Christian Tanguy, “Optical Dispersion of Wannier Excitons,” Phys. Rev. Lett. 75, 4090 (1995).
Errata, Phys. Rev. Lett. 76, 716 (1996).
20
Christian Tanguy, “Analytical expression of the complex dielectric function for the Hulthén potential,” Phys. Rev. B.
60, 10660 (1999).

10-442 • Reference – Layer Types CompleteEASE Software Manual

 function originally introduced in the WVASE® “psemi.mat” layer21. The Psemi
oscillators combine a highly flexible functional shape with Kramers-Kronig-
consistent properties.
The 2 part of the Psemi function consists of four polynomial splines connected end-
to-end. Each spline connects smoothly with the adjacent spline, forming a single,
continuous function, as shown in Figure 10-20 and Figure 10-21. The function’s
shape is controlled by adjusting the center point, “Ec”; the end points, “WL” and
“WR”; and the mid-width control points, “AL” and “AR”. Thus, the overall shape is
controlled by the midpoints and location of the endpoints with respect to “Ec”.
The function is forced to equal zero near the endpoints, and a Gaussian broadening
parameter, “Br”, rounds off the endpoints and peak (Figure 10-20 and Figure
10-21). The flexibility of these functions can be seen in the figures, as combinations
of left and right control point amplitudes can produce a wide variety of
asymmetrical 2 absorptions.
The Psemi-M0 oscillator models the shape of an M0 critical point seen in direct gap
semiconductors such as GaAs and InP, especially in the bandgap energy region.
Figure 10-20 shows the effects on the 2 line-shape for three different combinations
of “AR” and “PR” values. Likewise, the Psemi-M1 oscillator models the shape of an
M1 critical point; the Psemi-M2 oscillator models the shape of an M2 critical point;
and the Psemi-M3 oscillator models the shape of an M3 critical point. The Psemi-Tri
oscillator shown in Figure 10-21 is a general-use Psemi function. The variable
parameters and fixed values for each Psemi oscillator can be seen in Table 10-2. The
energy (“E”), broadening (“Br”), and absorption-region width (“WR” and “WL”)
parameters have units of eV. The other Psemi parameters are unitless.

WR

Amp=10
Br=0.05eV
O2R=0 AR=0.8

AR=0.5

AR=0.3

PR=0.5

PR=0.6

Figure 10-20. Graph of 2 showing the effects on the line shape of the Psemi-M0 oscillator
for different AR and PR values.

21
The psemi.mat layer is described in detail in the WVASE® software manual.

CompleteEASE Software Manual Reference – Layer Types • 10-443

 n  oscillator #

Fit parameters (units):

Left Control Amp
Point AL
Right Control
Point AR
Ampn (no units),
Brn (eV),
AL=0.6 Ecn (eV),
AL=0.5 AR=0.5 WLn (eV)
WRn (eV)
AR=0.2 ALn (no units)
ARn (no units)
Ec
Left End Point Right End Point
Ec – WL Ec + WR

Figure 10-21. Two views of Psemi-Tri oscillator. Top: Graph of 2 showing the “Amp”, “Ec”,
endpoints and control-points. Bottom: Psemi-Tri for 3 different “Br” values.

Table 10-2. Variable and fixed parameter values for Psemi oscillator functions.

OSCILLATOR EN. AMP. BR. WL WR PL AL PR AR O2L O2R

Psemi-M0 Fit Fit Fit 0 Fit 0.5 0.5 Fit Fit 0 Fit
Psemi-M1 Fit Fit Fit Fit Fit 0.75 Fit 0.5 Fit 1 0
Psemi-M2 Fit Fit Fit Fit Fit 0.5 Fit 0.75 Fit 0 1
Psemi-M3 Fit Fit Fit Fit 0 Fit Fit 0.5 0.5 Fit 0
Psemi-Tri Fit Fit Fit Fit Fit 0.5 Fit 0.5 Fit 0 0

CPPB Oscillator (Critical Point Parabolic Band)

The CPPB (Critical Point Parabolic Band) oscillator was developed by Aspnes22 to
model the shape of the dielectric functions at semiconductor critical points (CPs).
The shape of each CPPB oscillator depends on five parameters: the amplitude
(“Amp”), phase projection factor (“Phase”), threshold energy (“En”), broadening
(“Br”), and exponent (“Mu”). The “Mu” parameter can have any one of three
values:
n = +½ for one-dimensional CPs
n = “0” (becomes ln(2En – 2E – iBr)) for two-dimensional CPs
n = -½ for three-dimensional CPs

22
D. E. Aspnes, “Modulation Spectroscopy/Electric Field Effects on the Dielectric Function of Semiconductors,” from
Handbook on Semiconductors, Vol. 2, edited by M. Balkanski (North Holland, 1980), pp. 125 – 127.

10-444 • Reference – Layer Types CompleteEASE Software Manual

 The equation for the CPPB oscillator is shown in Equation 10-30, where CPPB(E)
represents the complex dielectric function contribution of the oscillator as a
function of photon energy (E) in eV, En is the threshold energy of the oscillator in
eV, Brn is the broadening of the oscillator in eV, An is the unitless amplitude of the
oscillator, n is the unitless phase projection factor, n is the exponent, and i is the
imaginary unit.
Equation 10-30
𝜇𝑛
𝐵𝑟𝑛
𝐴𝑛 𝑒 i𝜙𝑛 ( ) , 𝜇𝑛 = ± 1⁄2
𝜀CPPB (𝐸) = { 2𝐸𝑛 − 2𝐸 − i𝐵𝑟𝑛
i𝜙𝑛
𝐴𝑛 𝑒 ln(2𝐸𝑛 − 2𝐸 − i𝐵𝑟2 ) , 𝜇𝑛 = 0

CPM0 & CPM1 Oscillators (Critical Point Model)

The CPM0 and CPM1 oscillators were developed by Adachi23,24 to describe the
dielectric functions at semiconductor critical points. See Schubert 25 for an example
of the application of these oscillators. The CPM0 oscillator is shown in Equation
10-32 and the CPM1 oscillator is shown in Equation 10-33. In these equations,
CPM0(E) and CPM1(E) represent the complex dielectric function contribution of the
oscillator as a function of photon energy (E) in eV, En is the center energy of the
oscillator in eV, Brn is the broadening of the oscillator in eV, An is the amplitude of
the oscillator which has units of eV1.5 for CPM0 and is unitless for CPM1, χn is
defined in Equation 10-31, and i is the imaginary unit.
Equation 10-31

𝐸 + i𝐵𝑟𝑛
𝜒𝑛 =
𝐸𝑛
Equation 10-32

𝐴𝑛
𝜀CPM0 (𝐸) = (2 − √1 + 𝜒𝑛 − √1 − 𝜒𝑛 )
𝐸𝑛1.5 𝜒𝑛 2
Equation 10-33

−𝐴𝑛
𝜀CPM1 (𝐸) = ln(1 − 𝜒𝑛 2 )
𝜒𝑛 2

TOLO Oscillator (TO & LO Phonon Resonance)

The TOLO oscillator is a factorized oscillator that describes resonances with
significant anharmonic coupling effects that are frequently observed in multiple-
phonon-mode materials26,27. In the literature, TOLO functions are generally
included as part of a product representation of the dielectric function. The Gen-Osc
representation, on the other hand, is always a summation of oscillators. As a result,
not all of the TOLO fit parameters in the Gen-Osc will equal the same values when
using the product version of the TOLO oscillator for the same material. The TOLO
oscillator as used in CompleteEASE is given by Equation 10-34 where TOLO(E)

23
S. Adachi, Phys. Rev. B 35, 7454 (1987).
24
S. Adachi, “Optical dispersion relations for GaP, GaAs, GaSb, InP, InAs, InSb, Al xGa1-xAs, and In1-xGaxAsyP1-y” J. of Appl.
Phys. 66, 6030 (1989).
25
M. Schubert, T. Hofmann, B. Rheinländer, I. Pietzonka, T. Sass, V. Gottschalch, and J. A. Woollam, “Near-band-gap
CuPt-order-induced birefringence in Al0.48Ga0.52InP2” Phys. Rev. B 60(24), 16618 (1999).
26
M. Schubert, T. E. Tiwald, and C. M. Herzinger, Phys. Rev. B 61 8187 (2000).
27
J. Humlíčcek, R. Henn, and M. Cardona, Phys. Rev B 61(21) 14554 (2000).

CompleteEASE Software Manual Reference – Layer Types • 10-445

 represents the complex dielectric function contribution of the oscillator as a
function of photon energy (E) in eV, ELO and ETO are the center energies of the LO
and TO modes of the oscillator in eV respectively, BLO and BTO are the broadenings
of the LO and TO modes of the oscillator in eV, respectively, An is the unitless
amplitude of the oscillator, and i is the imaginary unit.
Equation 10-34

(𝐸𝐿𝑂𝑛 )2 − 𝐸 2 − i(𝐵𝐿𝑂𝑛 )𝐸
𝜀TOLO (𝐸) = 𝐴𝑛
(𝐸𝑇𝑂𝑛 )2 − 𝐸 2 − i(𝐵𝑇𝑂𝑛 )𝐸
The broadening parameters, BLO and BTO, are associated with the LO phonon and TO
phonon resonance, respectively. If BTO = BLO, then the TOLO resonance assumes the
classic Lorentz line shape. According to Schubert26,  will remain physical (i.e., 2 ≥
0) only if ∑𝑘𝑛=1(𝐵𝐿𝑂𝑛 − 𝐵𝑇𝑂𝑛 ) > 0 for all k TOLO oscillators. On the other hand,
Humlíčcek27 states that ∑𝑘𝑛=1(𝐵𝐿𝑂𝑛 − 𝐵𝑇𝑂𝑛 ) = 0. Regardless of which is correct,
these conditions are not automatically enforced by the Gen-Osc layer and must be
enforced manually by the user.

Custom Oscillator
The custom oscillator allows the user to define a layer’s complex dielectric or
refractive-index function by entering a custom equation.

Edit Equation dialog box

Selecting Edit Equation (located under “Type = Custom” in the oscillator list) opens
the “Edit Equation” dialog box. The box is shown in
Figure 10-22. The user will need to select the following:

• Optical Constant type: e1 & e2 or n & k.

• Wavelength units for the [wvl] parameter in Å, nm, m, eV, or 1/cm.
The user can then type the real and imaginary parts of the desired function into the
appropriate boxes. All parameters and variables must be enclosed in brackets.
As an example, we will recreate the Lorentz oscillator from Equation 10-5. One way
to do this is to separate both of its real and imaginary parts, as shown in Equation
10-35.
Figure 10-22 shows the Re{} and Im{} parts for Lorentz entered into the Custom
Oscillator “Edit Equation” dialog box.
Equation 10-35

𝐴𝑛 𝐵𝑟𝑛 𝐸𝑛 (𝐸𝑛 2 − 𝐸 2 ) 𝐴𝑛 𝐵𝑟𝑛 2 𝐸𝑛 𝐸

𝜀Lorentz (𝐸) = 2 +i 2
(𝐸𝑛 2 − 𝐸 2 ) + 𝐵𝑟𝑛 2 𝐸 2 (𝐸𝑛 2 − 𝐸 2 ) + 𝐵𝑟𝑛 2 𝐸 2

10-446 • Reference – Layer Types CompleteEASE Software Manual

 Figure 10-22. Custom Oscillator “Edit Equation” dialog box for Equation 10-35.
After clicking “Ok”, all of the fit parameters defined between brackets (in this case
“[Amp]”, “[Br]”, and “[En]”) will be available in the model as fit parameters.
An easier way to enter a complex-valued function, however, is simply to enter the
entire complex equation into the “e1” field while leaving the “e2” field blank. (In
the example above, instead of entering Equation 10-35 for the Lorentz oscillator,
you would enter Equation 10-5 for the Lorentz oscillator into the “e1” field.)
CompleteEASE will then interpret the complex-valued equation in the “e1” field as
describing both the 1 and 2 parts of the dielectric function even though both parts
are entered as one complex-valued equation in the “e1” field.

Operator List
A number of built-in operators and functions are available for use in the Custom
Oscillator equation. They are listed in Table 10-3.

Table 10-3. List of built-in operators and functions available in the CompleteEASE equation
editor dialog box for building a custom oscillator.

OPERATOR NAME DESCRIPTION

BASIC
+ Addition
– Subtraction
* Multiplication
/ Division
^ Exponentiation (Powers)
exp(arg) Exponentiation with Base e
sqrt(arg) Square Root
log(arg) Logarithm with Base 10
ln(arg) Natural Logarithm (with Base e)
re(arg) Real Part
im(arg) Imaginary Part
abs(arg) Absolute Value
TRIGONOMETRIC
sin(arg) Sine
cos(arg) Cosine
tan(arg) Tangent

CompleteEASE Software Manual Reference – Layer Types • 10-447

 SPECIAL
parmLog(arg) Access any hardware parameter recorded during
a measurement. Examples: parmLog(AlignX),
parmLog(Temperature)
isPositive(test_val, eq_pos, eq_neg) If test_val is positive, then eq_pos will be used in
that position. If test_val is negative, then eq_neg
will be used in that position. Example with
parameter value [x]: isPositive([x], [x]+5, -[x]+4)
will return 8 if [x] = 3 and 7 if [x] = -3.
layerEps(layer_num, wvl_in_Å) Retrieves a specified layer’s complex dielectric
function value at a specific wavelength
specificed in angstroms. Example: layerEps(2,
5400) will return ε1 + iε2 for Layer #2 at 540 nm.
CONSTANTS
PI Pi: π = 3.14159…
(Uppercase “PI” required)
e Euler’s Number: e = 2.71828…
(Lowercase “e” required)
h Planck’s Constant: h = 6.6260689633E-34 J∙s
(Lowercase “h” required – Note: CompleteEASE uses the pre-2019
defintion of Planck’s Constant from before the 2019 SI Units Redefinition.)
hbar Reduced Planck’s Constant: ħ = h/(2π) = 1.05457162853E-34 J∙s
(Lowercase “hbar” required – Note: CompleteEASE uses the pre-2019
defintion of Planck’s Constant from before the 2019 SI Units Redefinition.)
i Imaginary Unit: i = sqrt(-1)
(Lowercase “i” required)

Edit Marker Parameters

Selecting Edit Marker Parameters (located under “Type = Custom” in the oscillator
list) opens the “Edit Marker Parameters” dialog box (Figure 10-23). For each
custom oscillator function, the user can select which of their custom-defined
variables correspond to the three common markers, “Amplitude”, “Broadening”,
and “Center Energy”.

Figure 10-23. Custom Oscillator “Edit Marker Parameters” dialog box for Lorentz equation.

Once the common markers are set, the user-defined parameters can then be varied
by simply dragging the marker boxes around the Graph (Figure 10-24) using a
mouse. Setting the common markers also includes the user-defined parameters in
the specific Amp., Br., and En. toggle switches in the “Fit Menu”.

10-448 • Reference – Layer Types CompleteEASE Software Manual

 Amplitude/Center Energy
marker

Broadening Broadening
marker marker

Center Energy
marker

Figure 10-24. Graph panel with markers identified for a user-defined oscillator.

Pole
Poles, which the 1 part of the dielectric function will approach asymptotically, are
equivalent to Lorentz oscillators with zero broadening. As such, they only affect the
real part of the dielectric function (1) and not the imaginary part (2). A pole is
defined mathematically by Equation 10-3, the same as for the UV pole. However,
poles added in the oscillator list can be placed anywhere the user desires; they are
not restricted to the UV region.

10.5. EMA (Effective Medium Approximation)

The EMA layer uses an Effective Medium Approximation to calculate the optical
constants of a mixed material based on the optical constants of constituent
materials (see Figure 10-25). Calculation is based on mixing the optical constants of
two or three constituent materials (defined in their own .mat files), the percentage
of each material in the mixture, a depolarization “screening” factor, and the type of
EMA calculation to perform.

Figure 10-25. Graphical depiction of an effective medium approximation.

There are many ways to implement an EMA in CompleteEASE. The EMA Layer
described here is the most general way to construct an EMA; however, some
simpler methods include surface roughness (Section 9.4), which is a Bruggeman
EMA with 50% of the top layer in the model and 50% void; the Intermix Layer
(Section 10.6), which is a Bruggeman EMA with 50% of the layer above and 50% of
the layer below; and the Graded Layer (Section 10.18), which can increase or
decrease the index of a material by mixing it with void. (For more advanced
applications, such as porosimetry measurements, see Section 10.20 for the Porous
EMA layer.)

CompleteEASE Software Manual Reference – Layer Types • 10-449

 The EMA layer in CompleteEASE provides three different EMA mixing methods for
the user to choose28,29,30. These include a simple Linear combination, Bruggeman
EMA, and Maxwell-Garnet EMA (as well as anisotropic versions of the Bruggeman
EMA). The Linear mixing simply interpolates between each constituent material’s
dielectric functions to get the optical constants of the composite material.
However, the Maxwell-Garnett and Bruggeman EMAs are more common and more
accurate approaches than the simple Linear EMA. The Maxwell-Garnett EMA
assumes that spherical inclusions of material #2 (and material #3 if defined) exist in
a host matrix of material #1. The Maxwell-Garnett EMA is given by Equation 10-36
where εEMA is the dielectric function of the EMA layer, εa is the dielectric function of
the host material, εb is the dielectric function of the constituent material, and fb is
the fraction of material b mixed into the host material a.
Equation 10-36
𝜀𝐸𝑀𝐴 − 𝜀𝑎 𝜀𝑏 − 𝜀𝑎
= 𝑓𝑏
𝜀𝐸𝑀𝐴 + 2𝜀𝑎 𝜀𝑏 + 2𝜀𝑎
If it is difficult to determine which material is the host material and which is not,
then the Bruggeman EMA can be used, which makes the EMA layer itself the host
material of both constituent materials. The Bruggeman EMA is given by Equation
10-37 where εEMA is the dielectric function of the EMA layer, εa is the dielectric
function of one constituent material, εb is the dielectric function of the other
constituent material, fa is the fraction of material a in the EMA mixture, and
fb = (1 - fa) is the fraction of material b in the EMA mixture.
Equation 10-37
𝜀𝑎 − 𝜀𝐸𝑀𝐴 𝜀𝑏 − 𝜀𝐸𝑀𝐴
𝑓𝑎 + 𝑓𝑏 =0
𝜀𝑎 + 2𝜀𝐸𝑀𝐴 𝜀𝑏 + 2𝜀𝐸𝑀𝐴
An EMA layer is shown in Figure 10-26. The Maxwell-Garnett and Bruggeman EMAs
are very useful for modeling surface and interfacial roughness as well as modeling
poly-crystalline materials by mixing together amorphous and crystalline optical
constants of the material. When using the EMA layer, it can often be useful to use
the Coupled layer (Section 10.13) as an EMA constituent material to connect the
EMA with other layers in the model or the EMA-Coupled layer (Section 10.14),
which is a special case of the EMA layer with coupling built in.

Figure 10-26. EMA model for mixed constituent layers.

28
D. E. Aspnes, “Optical properties of thin films,” Thin Solid Films, 89 (1982).
29
D. E. Aspnes, J. B. Theeten, and F. Hottier, “Investigation of effective-medium models of microscopic surface
roughness by spectroscopic ellipsometry,” Phys. Rev. B, 20 (1981).
30
H. Fujiwara, J. Koh, P. I. Rovira, and R. W. Collins, “Assessment of effective-medium theories in the analysis of
nucleation and microscopic surface roughness evolution for semiconductor thin-films,” Phys. Rev. B, 61 (2000).

10-450 • Reference – Layer Types CompleteEASE Software Manual

 10.6. Intermix
An Intermix layer represents a Bruggeman EMA with 50% of the material above the
Intermix layer and 50% of the material below the Intermix layer. An Intermix layer
can be added between two materials in the model by using Add in the “Layer
Commands”. Select Intermix as the Layer Type, and then click on the desired
position in the model (see Figure 10-27). Intermix thickness can be entered or
selected as a fit parameter.

Figure 10-27. Choose “Intermix” before adding a layer between two layers to create an
Intermix between materials.
An example of the Intermix layer is shown in Figure 10-28. Notice the Intermix is
not a numbered layer. Intermix thickness is an independed thickness quantity, with
half of its thickness extending into the layer below, and half of its thickness
extending into the layer above. The thickness of the numbered layers remains the
same. Defining the Intermix “layer” in this manner maintains the total optical
thickness of the film stack, independent of the intermix thickness. This helps to
orthogonalize the layer thickness and intermix thickness fit parameters, as the layer
thickness essentially specifies the optical thickness of the film (which, for example,
determines the period of any thin-film interference oscillations), and the intermix
thickness specifies how diffuse or non-abrupt the interface is (which, for example,
may affect the magnitude or damping of any thin-film interference oscillations).

Figure 10-28. Example of Intermix layer in a model.

10.7. Void
The Void layer is a special case of a tabulated layer where the optical constants are
fixed for all wavelengths at the nominal values for air (vacuum): n=1 and k=0. The
void layer is often used in combination with other layers, such as the EMA layer
(Section 10.5) and the Graded layer (Section 10.18) when using the “Standard” type
of grading.

CompleteEASE Software Manual Reference – Layer Types • 10-451

 10.8. Additive
The Additive layer sums optical constants together by adding together the ε1 and ε2
functions of the constituent materials. Click Add Material to add layers to the sum
of optical constants, and click Remove Material to remove layers from the sum. The
Additive layer is shown in Figure 10-29, and a simple example of adding a constant
offset is shown in Figure 10-30. In this example, the User Defined layer (Section
10.23) is used to add the constant offset, by defining ε1 to be 1. One common use of
the Additive layer is to add free carrier contributions (like the Drude oscillator does
in the Gen-Osc) to a tabulated material file of a semiconductor.

Figure 10-29. The Additive layer.

Figure 10-30. Additive layer example of adding a constant offset of e1=1 to SiO2 optical
constants.

The Additive layer can also be used to sum together the dielectric-function tensors
of anisotropic materials (such as the Biaxial and Full Tensor layers). When an
anisotropic material is loaded into the Additive layer, it will automatically detect the
anisotropy and will add the dielectric tensor elements together instead of just ε1
and ε2. When in anisotropic mode, Euler angle rotations within each constituent
anisotropic material (“internal Euler Angles”) are applied before summing up the
tensor elements. Then, a set of "External Euler Angles" allows the total, summed
dielectric-function tensor to be rotated to account for sample surface cuts and
orientations. An example of the Additive layer in anisotropic mode is shown in
Figure 10-31.

10-452 • Reference – Layer Types CompleteEASE Software Manual

 Figure 10-31. The Additive layer in anisotropic mode.

10.9. Biaxial
The Biaxial layer allows the user to model anisotropic materials for which an eigen
coordinate system exists in which the dielectric-function tensor is diagonal for all
wavelengths (orthorhombic and higher symmetry). This layer is used for both
uniaxial anisotropy (two independent diagonal elements) and biaxial anisotropy
(three independent diagonal elements). It has three Euler-angle terms to rotate the
orientation of principal axes relative to the ellipsometer measurement orientation.
There are two modes which can be used within the Biaxial layer. With “Difference
Mode” OFF, the optical constants along the principal directions (two for uniaxial,
three for biaxial) are independently described by separate material files. This is
shown in Figure 10-32. With “Difference Mode” ON, the optical constants along the
x-direction (“Ex”) are described by a material file while an extended Cauchy
“difference” is established for the index difference between the x-direction and the
other principal directions. This is shown in Figure 10-33. An isotropic layer can be
converted to a Biaxial layer by right-clicking on the layer name and choosing
“Convert to Anisotropic”. For examples of data analysis using the Biaxial layer, see
Section 5.4.

Figure 10-32. The anisotropic Biaxial layer with “Difference Mode = OFF”.

Figure 10-33. The anisotropic Biaxial layer with “Difference Mode = ON”.

CompleteEASE Software Manual Reference – Layer Types • 10-453

 NOTE: Biaxial layer optical constants cannot be saved as a tabulated material file
because multiple directions of optical constants are included. In order to save a
“tabulated” biaxial material, first right-click on each constituent material (“Ex”,
“Ey”, “Ez”) in the Biaxial layer and save each as a tabulated .mat file (see Section
9.4). Then, replace each parameterized constituent layer in the Biaxial layer with
the tabulated .mat file it was saved to. Now, the Biaxial should contain all tabulated
.mat files as its constituents. Finally, save the Biaxial layer as “Parameterized”, and
the Biaxial layer will be saved with each tabulated constituent material file
embedded in the parameterized Biaxial .mat file.

10.10. Biaxial 2
The Biaxial 2 layer is a special version of the Biaxial layer that contains different
Euler angle definitions in order to make the analysis of specific materials more
convenient. Biaxial 2 allows two different theta Euler angles, “Theta X” and “Theta
Y”, to compensate for small X and Y tilts so that the phi Euler angle (here called
“Phi2”) does not need to be rotated by 90° to describe the optical axis orientation.
The Biaxial 2 layer can be seen in Figure 10-34.
The standard Biaxial layer uses an Euler-angle definition that works well if the theta
angle is large. However, if the c-axis of a crystal is only slightly tilted away from the
normal, it always favors the tilt along the y-direction. A slight tilt along the y-
direction only needs a small theta value. Then, with the standard Biaxial layer, if a
slight tilt along the x-direction is needed, phi needs to be rotated by a full 90° and
then theta can be slightly tilted. To solve this, the Biaxial 2 layer contains the Euler
angles in terms of a slight x-tilt and a slight y-tilt. (Thus, there are two theta angles
and no psi angle). The behavior of Biaxial 2 is very similar to the standard Biaxial
layer, and you can create the same tensor symmetries with Biaxial 2 as with Biaxial,
but the Biaxial 2 layer is much simpler if you are dealing with small x-tilts and y-tilts
in the crystal, perhaps due to a mis-cut. The relationship between Biaxial Euler
angle definitions and the Biaxial 2 Euler angle definitions are given in Equation
10-38 where φ, θ, and ψ are the Euler angles of the Biaxial layer and φ 2, θX, and θY
are the Euler angles of the Biaxial 2 layer.
Equation 10-38

tan 𝜃𝑋
𝜙 = 𝜙2 − arctan
tan 𝜃𝑌

𝜃 = arctan (√tan2 𝜃𝑋 + tan2 𝜃𝑌 ) ∙ sign(tan 𝜃𝑌 )

tan 𝜃𝑋
𝜓 = arctan
tan 𝜃𝑌

Figure 10-34. The Biaxial 2 layer is a special version of the Biaxial layer for handling rotation
when there are small X and Y tilts.

10-454 • Reference – Layer Types CompleteEASE Software Manual

 10.11. Cauchy_Wvl
The Cauchy_Wvl layer is a special representation of the Cauchy layer. It provides
the same dispersion shape as the Cauchy but is described with the terms shown in
Figure 10-35. This allows the user to specify a known index of refraction at a
specific wavelength instead of the standard Cauchy “A” parameter, without concern
for the dispersion terms. The Cauchy_Wvl formula is given in Equation 10-39 where
n(λ) is the refractive index as a function of wavelength in m, nd is the defined
refractive index at the design wavelength (“n” parameter), λd is the design
wavelength converted to m, and B and C are the Cauchy parameters.
Equation 10-39

𝐵 𝐶
𝑛𝑑 + 2 + 𝜆4
𝑛(𝜆) = 𝜆
𝐵 𝐶
1+ +
𝑛 𝑑 𝜆𝑑 2 𝑛 𝑑 𝜆𝑑 4

Figure 10-35. The Cauchy_Wvl layer is the same dispersion shape as Cauchy, but it allows
entry of the refractive index (n) at a design wavelength in place of the “A” parameter of the
standard Cauchy.

10.12. Cauchy_Extended
The Cauchy_Extended layer is a special representation of the Cauchy layer. It
provides a similar shape to the standard Cauchy layer, but it contains the ability to
more precisely describe the refractive-index dispersion with an additional “D”
parameter, and it can decrease index quickly toward the NIR spectral region by
adding an “IR” term. An example layer is shown in Figure 10-36 with the resultant
optical constants shown in Figure 10-37. The refractive index is given in Equation
10-40 where λ is the wavelength in m. The extinction coefficient is given in
Equation 10-41 where λ is the wavelength in m, γ is the band edge energy in eV,
E(λ) is the energy of λ relative to the band edge in eV, η(λ) determines whether the
Cody Amplitude should be included (i.e., is the wavelength greater than the band
edge?), α is the Urbach amplitude, β is the Urbach exponent, and ρ is the Cody
amplitude.
Equation 10-40

𝐵 𝐶 𝐷
𝑛(𝜆) = 𝐴 + + + − 𝐼𝑅 ∙ 𝜆2
𝜆2 𝜆4 𝜆6
Equation 10-41

1.24 1, 𝐸(𝜆) > 0

𝐸(𝜆) = − 𝛾 𝜂(𝜆) = {
𝜆 0, 𝐸(𝜆) ≤ 0
𝑘(𝜆) = 𝛼 exp(𝛽 ∙ 𝐸(𝜆)) + 𝜂(𝜆) ∙ 𝜌 ∙ 𝐸 2 (𝜆)

CompleteEASE Software Manual Reference – Layer Types • 10-455

 Figure 10-36. Example of the Cauchy_Extended layer.

Figure 10-37. Optical constants from a Cauchy_Extended layer. Note the downwards bending
at the longest wavelengths and the absorption at the shortest wavelengths due to inclusion
of the "IR" parameter and Urbach/Cody absorption, respectively.

10.13. Coupled
The Coupled layer has the same optical constants as another layer specified
elsewhere in the model layer stack. As the optical constants of the layers in the
model change during a fit due to adjustment of fit parameters, the optical constants
of the Coupled layer will also change to match the layer to which it is coupled. An
example is shown in Figure 10-38. The Coupled layer is often used as a constituent
material within another layer, such as the EMA layer (see Section 10.5). The
Coupled layer is also often used in multilayer stacks where different layers share the
same optical properties; however, for a repeating structure that alternates between
two layers, using the superlattice feature is more convenient (see Section 4.5 for an
example of using the superlattice).

Figure 10-38. The Coupled layer gets optical constants from another layer in the model. In
this case, it is using the optical constants from Layer #1, the Cauchy layer.

10.14. EMA-Coupled
The EMA-Coupled layer is a combination of the EMA and Coupled layers. The EMA-
Coupled layer operates exactly like the EMA layer described in Section 10.5, except

10-456 • Reference – Layer Types CompleteEASE Software Manual

 the first constituent material is automatically set up to be coupled to another layer.
This allows “Material 1” to obtain its optical constants from a separate layer in the
model, designated by the layer number. An example of the EMA-Coupled layer is
shown in Figure 10-39.

Figure 10-39. Example of the EMA-Coupled layer, where “Material 1” receives its optical
constants from a separate layer in the model, referenced by the layer number.

10.15. Full Tensor

The Full Tensor layer is an anisotropic layer that can model the real and imaginary
parts of all nine elements of the dielectric tensor using constituent material files.
The Full Tensor layer allows the dielectric tensor to be rotated by the standard Euler
angles. The Full Tensor layer can model the most complex material systems and
measurements, even ones obtained with external constraints such as electric or
magnetic fields. The KK Material layer (Section 10.19) or the User Defined layer
(Section 10.23) can be used within the Full Tensor layer as flexible ways to describe
any mathematically possible situation. The Full Tensor layer is set up in terms of
diagonal contributions, fully symmetric off-diagonal contributions, and fully anti-
symmetric contributions for easy restriction of the tensor symmetry. Any general
tensor can be created by including all three of these contributions. The convention
used for the optical constants of the dielectric tensor is shown below in Equation
10-42. The Zero layer can be used to set any of the dielectric-tensor elements to
zero (1 = 0 and 2 = 0), or if no material file is loaded ("none"), both 1 and 2 are
assumed to be equal to zero. An example of the Full Tensor layer is shown in Figure
10-40.
Equation 10-42
𝜀𝑥 𝜀𝑠𝑥𝑦 − 𝜀𝑎𝑥𝑦 𝜀𝑠𝑥𝑧 − 𝜀𝑎𝑥𝑧
𝜀⃡Full Tensor 𝜀
= [ 𝑠𝑥𝑦 + 𝜀𝑎𝑥𝑦 𝜀𝑦 𝜀𝑠𝑦𝑧 − 𝜀𝑎𝑦𝑧 ]
𝜀𝑠𝑥𝑧 + 𝜀𝑎𝑥𝑧 𝜀𝑠𝑦𝑧 + 𝜀𝑎𝑦𝑧 𝜀𝑧
𝜀𝑥 0 0 0 𝜀𝑠𝑥𝑦 𝜀𝑠𝑥𝑧 0 −𝜀𝑎𝑥𝑦 −𝜀𝑎𝑥𝑧
= [ 0 𝜀𝑦 0 ] + [𝜀𝑠𝑥𝑦 0 𝜀𝑠𝑦𝑧 ] + [𝜀𝑎𝑥𝑦 0 −𝜀𝑎𝑦𝑧 ]
0 0 𝜀𝑧 𝜀𝑠𝑥𝑧 𝜀𝑠𝑦𝑧 0 𝜀𝑎𝑥𝑧 𝜀𝑎𝑦𝑧 0

CompleteEASE Software Manual Reference – Layer Types • 10-457

 Figure 10-40. The Full Tensor layer with all sections expanded. The B-Spline layer is used to
describe some optical constants, the User Defined layer is used for others, and the remaining
elements are set to zero using the Zero layer.

10.16. GenOsc 3D
The GenOsc 3D layer is an anisotropic layer that can combine oscillators of arbitrary
eigen-dielectric polarizability unit vectors to create every physically possible
dielectric function tensor, according to Equation 10-43. Within this model, each
oscillator has its own set of Euler angles, “Alpha#” and “Beta#”, which rotate the
eigen polarization direction within the x-y plane and away from the x-y plane,
respectively. The GenOsc 3D layer can be used to model monoclinic and triclinic
materials by using any oscillator model available in CompleteEASE (see Section 10.4
for descriptions of available oscillators). The GenOsc 3D layer is shown in Figure
10-41.
Equation 10-43
𝑁

𝜀⃡𝐺𝑒𝑛𝑂𝑠𝑐 3𝐷 (𝜆) = ∑ 𝜌𝑖 (𝜆)(𝑒̂𝑖 ⨂ 𝑒̂𝑖 )

𝑖=1

Figure 10-41. The GenOsc 3D Layer.

Clicking on Show Dialog will show the specialized “Edit Layer” dialog box for the
GenOsc 3D layer, as shown in Figure 10-42. Reference materials can be added as
any dielectric-function tensor model (e.g. a Full Tensor layer or Biaxial layer) by
clicking on the ‘Open Material File’ button. Individual tensor elements can also be

10-458 • Reference – Layer Types CompleteEASE Software Manual

 replaced with an isotropic material file by clicking on the respective hyperlink in the
Reference Material panel. Note that only the currently selected dielectric-function-
tensor elements are considered when the ‘Fit’ button is pressed, and it is up to the
user to make sure that the currently selected fit parameters have an effect in those
elements. Otherwise, the fit parameters will get lost. The selected tensor elements
can be saved to individual tabulated material files by clicking on the ‘Save tabulated
.mat Files’ button. An example of Ga2O3 optical constants that were modeled using
the GenOsc 3D layer are shown in Figure 10-43.

Figure 10-42. The specialized “Edit Layer” dialog box for the GenOsc 3D layer.

Figure 10-43. Optical constants of Ga2O3 modeled using the GenOsc 3D layer.

10.17. Glass Catalog

The Glass Catalog layer is used to easily represent many kinds of glass materials
found in glass catalogs that are usually described by three terms in the general
Sellmeier equation, shown in Equation 10-44, where n(λ) is the refractive index as a

CompleteEASE Software Manual Reference – Layer Types • 10-459

 function of wavelength, λ, which is measured in μm, the B parameters are unitless,
and the C parameters are in μm2. An example of the Glass Catalog layer for BK7
glass is shown in Figure 10-44.
Equation 10-44

𝐵1 𝜆2 𝐵2 𝜆2 𝐵3 𝜆2
𝑛(𝜆) = √1 + + +
𝜆2 − 𝐶1 𝜆2 − 𝐶2 𝜆2 − 𝐶3

Figure 10-44. The Glass Catalog layer for BK7 glass.

10.18. Graded
Most layer types can be graded to model the optical variation with depth through
the film. This is achieved by right-clicking a layer and choosing to “Grade Layer”. An
example of a Graded layer is shown in Figure 10-45. There are three types of
grading that can be used in this layer: Simple, Standard, and Parametric. Parametric
grading has several grading profiles which allow linear, exponential, one-segment,
two-segment, and custom grading. Grading is accomplished by dividing the thin film
into a series of ‘slices’ that will have varying optical constants to approximate the
index gradient profile, as demonstrated in Figure 10-46. The number of slices can be
designated. The primary fit parameter for simple grading is the “% Inhomogeneity”,
which will be a positive number when the index is larger toward the surface of the
film than it is toward the substrate interface and a negative number when the index
is smaller toward the surface.

Figure 10-45. Example of a Graded layer.

Figure 10-46. Grading by “slices” demonstrated for a sample of TiO2 on a glass substrate.

10-460 • Reference – Layer Types CompleteEASE Software Manual

 NOTE: Right-clicking on the substrate does not allow grading. This is because the
substrate, by definition, is considered to be effectively of “infinite” thickness; thus,
grading it is impossible.

If you right-click on the title Graded Layer, several options are given to either
remove the grading or graph the optical constants, as shown in Figure 10-47.
“Remove Grading” simply converts the Graded layer material to a simple layer with
uniform optical constants throughout. “Graph Layer Optical Constants” will show
either n and k or 1 and 2 at both the top and bottom of the film, as shown in
Figure 10-48. Choosing to “Graph Sub-Layer Optical Constants” will graph the
optical constants for each slice in the grading model, as shown in Figure 10-49.
In order to graph a profile of the graded optical constants versus film thickness for a
specific wavelength, you can select the “+ Model” menu above the graph and
choose either “Optical Constants Profile” or “Optical Constants Profile (No
Substrate)”. These options will produce graphs like in Figure 10-50. Note that these
options apply to the model as a whole, so they will graph the optical-constant
profiles of all layers in the model, not only the Graded layer.

Figure 10-47. Right-click menu of the Graded layer allowing the graphing of optical
constants.

Figure 10-48. Example graph of graded optical constant at the top and bottom of the layer by
choosing “Graph Layer Optical Constants”.

CompleteEASE Software Manual Reference – Layer Types • 10-461

 Figure 10-49. Example graph of graded optical constants for each slice of the layer in the
grading model by choosing “Graph Sub-Layer Optical Constants”.

Figure 10-50. Example graph of an optical-constant profile for a Graded layer.

# of Slices
For each type of grading, the profile is described by an equation between the top
and bottom of the film. Then, the number of slices, specified by “# of slices”, is
evenly distributed through the film and the value of each slice is taken from the
value at the center of each slice.

NOTE: The bottom and top “slice” do not have to be equal to the values of the
equation at the bottom and top of the film, as they get their value from the “slice”
center.

Grade Type
There are three different types of grading available, as shown in Figure 10-51.
Details for each are described below.

10-462 • Reference – Layer Types CompleteEASE Software Manual

 Figure 10-51. Grade type selections include Simple, Standard, and Parametric.

Simple Grading
Simple grading allows linear variation of the index of the film by fitting or adjusting
the “% Inhomogeneity” parameter. The index can either increase or decrease from
the bottom of film to the film surface, depending on whether “% Inhomogeneity” is
positive or negative, respectively. An example of simple grading is shown in Figure
10-52, where a Cauchy layer exhibits 4% inhomogeneity. The Cauchy layer remains
embedded in the Graded Layer, so the “average” index of the layer can still be
adjusted by the Cauchy parameters. The grading of the refractive index will occur in
both directions of the Cauchy – with a total range of index between the top and
bottom of the layer equaling the “% Inhomogeneity”.

NOTE: Simple grading only affects the index (n) of the film; it does not affect the
extinction coefficient (k).

Figure 10-52. Example of simple grading for a Cauchy Layer.

Standard Grading
Standard grading uses an EMA to mix the current material being graded with
another material. Usually, the Void material (Section 10.7) is used, and it is selected
by default when standard grading is chosen. Unlike simple grading, standard
grading affects both index of refraction (n) and extinction coefficient (k) of the film.
Standard grading only uses the EMA mixing in order to vary the optical constants
between the bottom and top of the layer. The EMA top and bottom percentages
are not necessarily a physical mixture of the material with void. Rather, the EMA is
a mathematical calculation to adjust the optical constants up or down to vary with
film thickness. In fact, the EMA percentages can go negative to specify the index
and extinction coefficient going “up” (further away from the other material) rather
than “down”.
Standard grading with the “Exponent” parameter fixed at a value of “1.0” is
sometimes referred to as “linear grading”, while allowing the exponent to vary to
other values is sometimes referred to as “nonlinear grading”. Specifying that
“Symmetric Profile” should be ON indicates that the exponential variation should
be symmetric about the center of the film, whereas turning “Symmetric Profile” to
OFF indicates that the exponential variation should proceed from the bottom to the
top of the film.

CompleteEASE Software Manual Reference – Layer Types • 10-463

 Figure 10-53 shows a Graded layer with standard grading that has an embedded
Cauchy material. The “Bottom %” and “Top %” can be set (or fixed) along with the
“Exponent” to describe the grading shape. Press Draw Profile to see how the EMA
percentages affect the way the second material is being distributed through the
various slices of the Graded layer. This option graphs only the optical-constant
profiles for the graded layer, not for the whole model’s layer stack. A wide variety
of grading profiles are possible, depending on whether the profile is symmetric or
not and the value of exponent, as shown in Figure 10-54 through Figure 10-58.

Figure 10-53. Example of nonlinear Standard grading.

Figure 10-54. Grade profile for a Graded layer with 5 slices and standard grading with
“Exponent = 1.0” (linear grading).

Figure 10-55. Grade Profile for a Graded layer with 21 slices and standard grading with
“Exponent = 0.2” and “Symmetric Profile” turned ON.

10-464 • Reference – Layer Types CompleteEASE Software Manual

 Figure 10-56. Grade Profile for a Graded layer with 21 slices and standard grading with
“Exponent = 2.5” and “Symmetric Profile” turned ON.

Figure 10-57. Grade Profile for a Graded layer with 21 slices and standard grading with
“Exponent = 0.2” and “Symmetric Profile” turned OFF.

Figure 10-58. Grade Profile for a Graded layer with 21 slices and standard grading with
“Exponent = 2.5” and “Symmetric Profile” turned OFF.

Parametric Grading
Parametric grading allows certain dispersion-model parameters to vary with depth
through the film, while the remaining parameters are constant with respect to film
depth. Thus, parametric grading only works with parameterized layers (not
tabulated layers) such as the Gen-Osc and Cauchy. An excellent example for the use
of parametric grading is modeling the free-carrier profile in transparent conductive

CompleteEASE Software Manual Reference – Layer Types • 10-465

 oxides such as indium tin oxide (ITO). The free carriers lead to infrared absorption,
which is modeled with a Drude oscillator. Thus, the variation in the film properties
is not well represented by EMA mixing with void. Instead, the amount of IR
absorption will change from bottom to top, requiring a variation in the Drude
oscillator amplitude. The electronic inter-band transition absorption in the UV
spectral range will generally stay constant through the film; thus, we want to grade
only the Drude component. Figure 10-59 is an example of an ITO layer with
parametric grading by linear variation of the Resistivity of the Drude(RT) oscillator.

Figure 10-59. Example of parametric grading with a linear variation of the Drude resistivity
term.

After the “Grade Type” is selected to be “Parametric”, there are five choices for the
grading equation, called the “Profile”, as shown in Figure 10-60. The “Linear” and
“Exponential” grade profiles allow the parameters to be varied through the film in a
linear or exponential fashion. However, these grade profiles can be difficult for
parameters that are close to zero. The “One Segment” grading profile allows the
specific parameter values to be specified for the bottom and top of the film with
linear grading in between. The “Two Segment” grading profile splits the film into
two segments, where the values of the parameter being graded can be specified at
the bottom, middle (midway split between segments), and top of the film, with
linear grading in between. The “Custom” grading profile allows the user to enter
their own equation to describe the parameter grading.

Figure 10-60. Choice of “Profile” (the grading equation) for parametric grading.

10-466 • Reference – Layer Types CompleteEASE Software Manual

 Linear Grade Profile
The linear grade profile allows a parameter in the graded layer’s dispersion
equation to be varied between the top and bottom of the film in a simple linear
fashion.

Exponential Grade Profile

The exponential grade profile allows a parameter in the graded layer’s dispersion
equation to be varied between the top and bottom of the film in an exponential
fashion. Figure 10-61 shows an example of an exponential grade profile for the
Drude resistivity parameter.

Figure 10-61. Example of an exponential grading profile for the Drude resistivity parameter.

One-Segment Grade Profile

The one-segment grade profile is very similar to the linear grade profile in that both
exhibit linear grading from bottom to top of the film. However, unlike the linear
grade profile, the one-segment grade profile allows the actual values of the
parameter being graded to be set directly at the top and the bottom of the film,
instead of just specifying the grade percentage.

Two-Segment Grade Profile

The two-segment grade profile allows variation of the parameters between the
bottom, top, and a middle point in the film with linear variation in between. In
addition, the position of the middle node can be varied. An example of two-
segment parametric grading is shown in Figure 10-62, where the bottom and
middle resistivity values are set to 0.1 while the top value is 0.2. The middle point
position is set to 60% through the film. This gives the grade profile shown in Figure
10-63.

NOTE: Even though the grade profile has two segments, the 21 slices specified in
the Graded layer are distributed equally through the entire layer.

CompleteEASE Software Manual Reference – Layer Types • 10-467

 Figure 10-62. Example of parametric grading with the two-segment grade profile. The
bottom and middle resistivity values are set to 0.1 while the top value is 0.2. The middle
position is set to 60% through the film.

Figure 10-63. Grade profile for resistivity associated with the two-segment profile in the
previous figure.

Custom Grade Profile

A custom grade profile allows the user to enter a custom equation to describe the
parametric grading. When entering a custom equation, the built-in variables for the
grade-profile equation to use are the position in the layer, called “[pos]”, which
varies from 0 to 1 and the value of the parameter being graded, called “[value]”.
Variable names must be enclosed in brackets. For more information on how to use
the equation editor and its available operators and functions, refer to the “Custom
Oscillator” subsection of Section 10.4 and Table 10-3.
Figure 10-64, Figure 10-65, and Figure 10-66 below show how the custom grade
profile can be set up to model a linear grade of the amplitude of a Cody-Lorentz
oscillator. The grade equation is evaluated at the center of each slice to determine
the parameter value for that slice. If a grade equation does not include the “[value]”
variable, then the parameter being graded should be turned off as a fit parameter
as it will be singular. If you forget to do this, the fit will proceed correctly, but the
graded parameter will have an error bar value of 9999. Also, when adding custom
grade equations for more than one parameter, any fit parameter names that are
the same between equations will be treated as the same fit parameter. This
eliminates the need to manually couple parameters after entering the equations.

10-468 • Reference – Layer Types CompleteEASE Software Manual

 Figure 10-64. Custom grade profile with parametric grading. Click <Enter an Equation> to
open the “Edit Equation” dialog box.

Figure 10-65. “Edit Equation” dialog box.

Figure 10-66. A custom grade profile has been applied. The user-defined parameter
“% Grade” is now available as a fit parameter.

10.19. KK Material
The KK Material layer is used to perform the Kramers-Kronig (KK) transform
integration. The optical constants of either the real part or the imaginary part of the
dielectric function are specified by another material file, and the KK Material layer
will then calculate the other part of the dielectric function using the KK transform,
as given in Equation 10-45 and Equation 10-46, where 1 and 2 are the real and
imaginary parts of the dielectric function, respectively; E is the energy in eV;  is
the offset of the real part, called “e1 Offset” in the KK Material layer (See the “Einf”
part of Section 10.4); and PV denotes the Cauchy principal value of the integral. An
example of the KK Material layer is seen in Figure 10-67. Often, the KK Material
layer is used in conjunction with the User Defined layer (Section 10.23), where the
user can specify one component of the dielectric function and then let the KK
Material layer calculate the other component. (In order to add flexibility for the
shape of the real part of the dielectric function, two poles are given that can be
specified. For more information on poles, see the “UV Pole”, “IR Pole”, and “Pole”
subsections of Section 10.4.)

CompleteEASE Software Manual Reference – Layer Types • 10-469

 Equation 10-45
∞
2 𝜔 ∙ 𝜀2 (𝜔)
𝜀1 (𝐸) = 𝜀∞ + 𝑃𝑉 [∫ 2 𝑑𝜔]
𝜋 𝜔 − 𝐸2
0

Equation 10-46
∞
2𝐸 𝜀1 (𝜔) − 1
𝜀2 (𝐸) = − 𝑃𝑉 [∫ 2 𝑑𝜔]
𝜋 𝜔 − 𝐸2
0

Figure 10-67. An example of the KK Material layer.

10.20. Porous EMA

The Porous EMA layer is a special case of the EMA layer that is used to represent a
porous material whose pores are being filled with a solvent vapor (such as water or
toluene) in porosimetry experiments. In this layer, the porous material, ambient,
and solvent material can be specified. The Porous EMA layer not only accounts for
the solvent filling fraction and porosity in your porosimetry experiment but also for
the percentage of accessible pores, potential anisotropy, and graded sample
properties. An example of the Porous EMA layer can be seen in Figure 10-68. In
most cases, the Porous EMA layer is used in conjunction with the Environment Cell
and the Pore Size model (Section 9.9). Refer to Chapter 5 of the J.A. Woollam
Environment Cell Manual, “Analysis of Porous Samples”.

Figure 10-68. An example of the Porous EMA layer.

10.21. Sellmeier
The Sellmeier layer is another dispersion layer that can be used to represent
dielectrics. It can be considered a slightly advanced version of the Cauchy layer. It is
used to represent the optical constants of transparent dielectric materials.
CompleteEASE uses a two-term Sellmeier model to account for UV and IR
absorption. The representation for the index of refraction is given in Equation 10-47
where n(λ) is the refractive index as a function of wavelength λ in μm,  is an index
offset (“Einf”), A is the amplitude (“Amp.”) of the absorption, E is the center energy

10-470 • Reference – Layer Types CompleteEASE Software Manual

 (“Center En.”) of the absorption, and P is the amplitude of a pole in the infrared (“IR
Pole Amp.”). An example Sellmeier layer is shown in Figure 10-69. To use the three-
term Sellmeier equation often associated with glass materials, use the Glass Catalog
layer (Section 10.17).
Equation 10-47

𝐴𝜆2
𝑛(𝜆) = √𝜀∞ + − 𝑃𝜆2
𝜆2 − 𝐸 2

Figure 10-69. The Sellmeier layer.

10.22. Temperature or Composition Library

Layers that are marked as “Comp Library” or “Comp & Temp Library” were created
to show variation in optical constants as the composition or temperature are
varied. Several of the material files provided with CompleteEASE contain
information about the optical constants of a material as a function of temperature
and/or alloy fraction (composition). One example of this type of material is
AlxGa1-xAs. The file “AlGaAs (Comp & Temp Library).mat” supplied with
CompleteEASE contains information on the optical constants of Al xGa1-xAs over x
values from 0 to 1 and temperature values from 577°C to 650°C. Using this type of
layer is very similar to using a standard tabulated layer for the material; the only
difference is that a Temperature or Composition Library material contains
temperature and/or alloy parameters that can be adjusted and used in a fit. Figure
10-70 shows how such a layer is represented in the model. Temperature and
Composition libraries can be constructed using the Opt. Const. Compare model.
Refer to Section 5.3 for an example of how to construct a Temperature or
Composition Library material file.

Figure 10-70. A Comp. & Temp. Library layer as displayed in the model. The composition and
temperature values are editable fit parameters.

10.23. User Defined

The User Defined layer allows the user to enter custom equations for both the real
and imaginary parts of the optical constants; however, the User Defined layer on its
own does not enforce Kramers-Kronig consistency between these parts. If Kramers-
Kronig consistency is desired, use the User Defined layer as the “Material” in the KK
Material layer (Section 10.19).
Click Edit Equation to open the “Edit Equation” dialog box as shown in Figure 10-72.
This dialog allows an equation to be entered in terms of n and k or 1 and 2.
Variable names must be enclosed in brackets. The units of the wavelength variable
[wvl] can be set to Å, nm, um, eV, or 1/cm. For more information on how to use the

CompleteEASE Software Manual Reference – Layer Types • 10-471

 equation editor and its available operators and functions, refer to the “Custom
Oscillator” subsection of Section 10.4 and Table 10-3. An example of how a Cauchy
dispersion layer can be set up using the User Defined layer is shown in Figure 10-71.
After clicking “Ok” in the “Edit Equation” dialog box, any variables defined within
brackets (“a”, “b”, and “c” in the example below) will be shown as fit parameters.

Figure 10-71. The User Defined layer with an example Cauchy equation entered.

Figure 10-72. “Edit Equation” dialog box with an example Cauchy equation entered.

10.24. Virtual Substrate

Sample structure below the layer of interest may not be known or may be difficult
to model. The Virtual Substrate layer can model the underlying structure as a single
“pseudo substrate” and simplifies the optical model of a multilayer sample by
allowing us to ignore the underlying sample structure and only consider the current
thin-film growth. (See Figure 10-73 for an illustration of this.)

Actual Simplified
Sample Model
Layer 4 Layer 4

Layer 3
Virtual Layer 3
Interface
Layer 2

Layer 1
Pseudo-Substrate
Substrate

Figure 10-73. Illustration of Virtual Interface.

The “pseudo substrate” optical constants are obtained from a direct inversion of Psi
and Delta from experimental data using the bulk-substrate assumption. This
approximation is nearly exact for materials with a high index of refraction and low
index contrast (e.g. growth of semiconductors on semiconductors).
The Virtual Substrate layer is always used with dynamic data from in-situ
experiments. To set the timeslice in the dynamic data at which the “pseudo
substrate” optical constants calculation should occur, hold down the ALT key and

10-472 • Reference – Layer Types CompleteEASE Software Manual

 click on a timeslice in dynamic data. (Use caution when doing this; the model will
allow any timeslice to be selected, even when the “pseudo substrate”
approximation is not valid). Alternatively, you can select a range of time for
averaging the “pseudo substrate” approximation using the “Start Time” and “End
Time” parameters found in the Virtual Substrate layer. By default, CompleteEASE
will ignore any timeslices with out-of-spec hardware parameters (i.e. insufficient
intensity recorded during measurement). Turn ON “Use All Times for VI” to override
this behavior. (To check whether there is an out-of-spec hardware parameter at any
timeslice, check for the parameter “Hardware OK” to be equal to zero (meaning the
hardware was NOT okay) by graphing it. Select “Hardware OK” in the drop-down
box next to “Graph Hardware Parm” above the graph when dynamic data is open.)

Figure 10-74. Virtual Substrate layer.

10.25. WvlByWvl
The WvlByWvl layer (short for Wavelength-by-Wavelength layer), as shown in
Figure 10-75, allows the optical constants to vary independently at each wavelength
(without regard to neighboring wavelengths measured). This layer has the utmost
flexibility but also the largest number of fit parameters and great propensity for
poor-quality results. Due to its use of only individual wavelengths, the WvlByWvl
layer will often reflect any measurement noise present in the data. As it does not
maintain Kramers-Kronig consistency or even smooth, continuous curves for the
optical constant dispersion, it should be used with extreme caution.
In the days before the widespread use of the B-Spline layer, the WvlByWvl layer
was the most common method for obtaining a test set of optical constants on
which to begin constructing a Gen-Osc model. Now, however, this process works
better when using a B-Spline layer instead of a WvlByWvl layer since the B-Spline
layer is guaranteed to produce smooth, continuous, and KK-consistent optical
constants. (Refer to Section 4.1 for a simple WvlByWvl layer example and refer to
Section 4.3 and Section 4.4 for examples of using the B-Spline to construct optical
constants for building a Gen-Osc model.)

Figure 10-75. The WvlByWvl layer.

10.26. Zero
The Zero layer is a layer with optical constants of ε1 = 0 and ε2 = 0 for all
wavelengths that is mostly used with the Full Tensor layer (Section 10.15) to set
certain elements of the dielectric tensor to zero.

CompleteEASE Software Manual Reference – Layer Types • 10-473

11. Reference – In situ Tab

The In situ tab is used for real-time dynamic ellipsometry measurements where the
sample properties might change with time. It provides several useful options for
real-time dynamic data acquisition and display. The In situ tab may be required for
M-2000 and RC2 systems mounted onto a chamber or for accessories such as liquid
cells or heat stages mounted onto M-2000, RC2, and alpha-SE systems.

11.1. Displaying the In situ Tab

The In situ tab should be visible among the other tabs. If it is not, press
CTRL+ALT+SHIFT+U to display the In situ tab for a single session.
To make the In situ tab always visible, use the following procedure.
1. From the Options tab, click ‘Edit Configuration’ (see Figure 11-1).

Figure 11-1. Options tab with ‘Edit Configuration’ button indicated.

2. Next, from the “CompleteEASE Configuration” dialog, expand the

“General” section, scroll to find “Always Show In situ Tab”, and turn
this Parameter ON (see Figure 11-2).

11-474 • Reference – In situ Tab CompleteEASE Software Manual

 Figure 11-2. “Always Show In situ Tab”.

3. When finished, click ‘Close’ and then ‘Yes’ to apply changes (see Figure
11-3).

Figure 11-3. Click ‘Yes’ to apply changes.

11.2. In situ Tab Features

The In situ tab is shown in Figure 11-4. This tab consists of a status box, hardware
control section, View panel, Models panel, and graph.

Figure 11-4. In situ tab.

CompleteEASE Software Manual Reference – In situ Tab • 11-475

 Status Box
The status box (upper-left corner of Figure 11-4) provides information about the
current status of the ellipsometer, such as “Initializing Hardware”, “Waiting to
Acquire Data”, etc.

Hardware Control Section

This section is comprised of four buttons and a checkbox immediately below the
status box (see Figure 11-5). From here, you can ‘Align Hardware’, perform a
‘System Check’, and ‘Display Signal’ from the spectrometer. Measurements are also
started here using ‘Start Acquisition’. Users can also enable Fast Acquisition Mode.

Figure 11-5. Hardware control section shown prior to start of data acquisition.

NOTE: The settings for Standard and Fast acquisition modes are specified in the
Hardware Configuration. To access this, click ‘Edit Hardware Configuration’ in the
Hardware tab.

Start Acquisition / End Acquisition

This starts the data acquisition process. After the acquisition is started, the ‘Start
Acquisition’ button changes to ‘End Acquisition’ which ends the data acquisition
and saves the data to data file (see Figure 11-6).

Pause Acquisition / Resume Acquisition

This pauses or resumes the data acquisition when standard acquisition mode is
selected (see Figure 11-6 and Figure 11-7).

11-476 • Reference – In situ Tab CompleteEASE Software Manual

 Figure 11-6. ‘End Acquisition’ and ‘Pause Acquisition’ become available after standard data
acquisition is started.

Figure 11-7. 'Resume Acquisition' becomes available after acquisition is paused.

Trigger Acquisition
When the acquisition mode is set to Trigger (via “Hardware Configuration” dialog
box, see Figure 11-8), a ‘Trigger’ button will also appear after starting acquisition
(see Figure 11-9). Selecting this button will trigger a single data acquisition event.

CompleteEASE Software Manual Reference – In situ Tab • 11-477

 Figure 11-8. Hardware Configuration dialog with “Acquisition Mode” set to “Trigger”.

Figure 11-9. ‘Trigger Acquisition’ button.

Align Hardware
This allows the user to align the ellipsometer source and receiver units or the
sample stage.

System Check
Starts a System Check. See your ellipsometer’s hardware manual for more details.

Display Signal
Displays the raw intensity signal output by the CCD array spectrometer.

Fast Acquisition Mode

Checking this box enables the fast acquisition time settings as specified in the
Hardware Configuration.

11-478 • Reference – In situ Tab CompleteEASE Software Manual

 Models
This panel contains models that can be used for in-situ analysis. Up to forty models
can be added to the list (four pages of ten models per page). Clicking one of the
model buttons before or during data acquisition loads the model and any data
subsequently acquired will be fit to that model.

Figure 11-10. Models panel with two models added.

Model Setup Dialog

To add, edit, or delete a model from the Models panel, right-click on an empty or
filled model slot. This will open the “Model Setup” dialog (see Figure 11-11).

Figure 11-11. “Model Setup” dialog.

CompleteEASE Software Manual Reference – In situ Tab • 11-479

 Figure 11-12. Use the drop-down menu to select models from the COMMON folder.

• Selecting a Model
Select the desired model from the “Model File” list.

NOTE: Only model files that have been saved in or copied to the COMMON folder
will appear in this list.

• Clearing a Model
Select “(none)” if you wish to clear the model slot.
• Parameter Corrections
With a model selected, all fit parameters from that model will appear in
the parameter list. Highlighting any fit parameter in the list allows the user
to include corrections (Offset, Multiply, and AlignX) to that fit parameter.
• “Do thickness control”
This checkbox directs CompleteEASE to send a signal to another program
or device to stop growth or etch when a target thickness is reached.
• “Write report”
This checkbox activates a report-writing function.
• “Merge with current model”
This checkbox allows the user to merge the selected model with the
current model in the Analysis tab. If the models match on all layers except
the top one, we can retain the fitted parameters from previous layers
when modeling additional layers.
For example, consider two models: 1) thermal oxide on silicon (Figure
11-13) and 2) metal on thermal oxide on silicon (Figure 11-14). The models
are identical except for the top layer.
A sample of thermal oxide on silicon is measured and analyzed with Model
1. The fitted thickness is 568.15 nm. Next, a metal is deposited over the
same sample and analyzed with Model 2 with “merged with current
model” checked. The previously fitted thermal oxide thickness is copied
into Model 2, but is no longer a fit parameter.

Figure 11-13. Model 1: thermal oxide on silicon. Note that the thickness of Layer
#2 is 568.15 nm and is an active fit parameter.

11-480 • Reference – In situ Tab CompleteEASE Software Manual

 Figure 11-14. Model 2: metal on thermal oxide on silicon. After merging, note that
Layer #2 is retains the previously fitted thickness (568.15 nm) but is no longer a fit
parameter.

View
This panel selects which information to display in the graph. Select from
“Parameters”, “Log”, “Fit”, or “Dynamic Data” (see Figure 11-15).

Figure 11-15. The View panel.

Parameters
This button graphs the value of the fit parameters and hardware alignment
information during data acquisition. Figure 11-16 shows the parameter graph
during a practice in-situ measurement of a thermal oxide on silicon wafer. It is
displaying the “Total Thickness”, “MSE”, and “Hardware OK” versus time. The
parameters displayed are selected from the four drop-down menus above the
graph. Parameters listed in the two left boxes are shown on the left y-axis and
parameters listed in the two right boxes are displayed on the right y-axis. Besides
fit parameters and MSE, the following hardware parameters can also be displayed:
SigInt, Intensity, AlignX, AlignY, Hardware OK, and Fit OK. Custom parameters
added through a remote system can also be displayed here if available.

Figure 11-16. View “Parameters”.

CompleteEASE Software Manual Reference – In situ Tab • 11-481

 Log
This shows the information in the in-situ log and allows the user to record
information during data acquisition. Recorded information includes the time when
models are loaded or unloaded and when the data acquisition is stopped. Users
can also add information to the log. To add a comment, type it into the textbox at
the top (Figure 11-17, green arrow) and click ‘Add’ (Figure 11-17, red arrow). The
comment is time-stamped according to when it is added. Comments can only be
added to the log during data acquisition. When data is saved, the parameters graph
and log data are saved with it and can be retrieved by opening the experimental
data file.

Figure 11-17. View “Log”.

Fit
This displays ellipsometric data vs. wavelength for the last recorded timeslice as
well as data generated from the model if applicable (see Figure 11-18).

Figure 11-18. View “Fit”.

11-482 • Reference – In situ Tab CompleteEASE Software Manual

 Dynamic Data
This displays one or more measured data values vs. time (Psi, Delta, N, C, S,
Intensity, etc.). The +Data menu allows user to select which measured values are
displayed. This is shown in Figure 11-19 for a section of dynamic data.

Figure 11-19. View “Dynamic Data”.

CompleteEASE Software Manual Reference – In situ Tab • 11-483

12. Reference – Hardware Tab

Controls available in the Hardware tab depend on the ellipsometer configuration.

As a result, this chapter is divided into two major sections: one section describing
the alpha-SE Hardware tab and another section describing the M-2000 and RC2
Hardware tab.

NOTE: More details of the Hardware tab can be found in the hardware manual for
your ellipsometer, which can be accessed by pressing F2 on the keyboard when
CompleteEASE is open. This chapter is meant only as a general reference.

12.1. Configuration Files

CompleteEASE determines the hardware configuration based on the “cnf” files that
are located in the C:\CompleteEASE\cnf folder. These files may include the
following:

hardware.cnf
Describes the hardware configuration and options available with the current
system. These settings should not be edited by the user unless instructed by a
Woollam representative.

CompleteEASE.cnf
Configuration file for user-settable software parameters. These parameters should
be accessed by choosing ‘Edit Configuration’ from the Options tab.

CompleteEASEhard.cnf
Configuration file for user-settable hardware parameters. These parameters should
be accessed by choosing ‘Edit Hardware Config.’ from the Hardware tab.

Privileges.cnf
Contains user privileges when password-protected user log-ins are created.

rce-cal.cnf, rce-comp.cnf, rce-dc.cnf, rce-refl.cnf

Describe the most recent calibration and intensity baseline values for the hardware
based on recent calibrations, system checks, etc.

12-484 • Reference – Hardware Tab CompleteEASE Software Manual

 12.2. Hardware Tab – alpha-SE
The Hardware tab for an alpha-SE is shown in Figure 12-1. This tab consists of
multiple panels: Hardware, Signal, Calibration, Show Logs, Camera (optional), and
Routine Test Measurement. A brief description of all commands within this tab is
provided.

Figure 12-1. CompleteEASE with the Hardware tab selected.

Hardware
Re-initialize
Re-initializes the ellipsometer hardware, which initiates communication over the
USB link, scans the spectrometer, moves the motors to their home positions, and
tests the motors. Details of any errors that occur during hardware initialization can
be found in the “Hardware” and/or “Error” logs.

Park Z-Stage
Moves the Z-stage to the “park” position such that the Z-stage shipping lock can be
engaged. This should be done before moving/shipping the alpha-SE unit to avoid
damage to the Z-stage motor.

Signal
Display
Displays signal intensity of the light collected by the spectrometer. Average,
maximum, UV, and IR intensity values are also reported in the status box, along
with the current spectrometer “Dark Count” and electronics “Temperature”. An
example is shown in Figure 12-2.

Max. Sig. Display

Same as Display, except that the polarizer and compensator optics are adjusted to
find the maximum signal intensity. After maximum signal is found, the intensity will
display similar to Figure 12-2.

CompleteEASE Software Manual Reference – Hardware Tab • 12-485

 Figure 12-2. Signal intensity of the light collected by the spectrometer.

Z-Stage Scan
Scans the Z-stage to generate an intensity vs. sample height profile. A sample
should be mounted and the source and receiver should be set to 65, 70, or 75
degrees before clicking this button. The resulting signal intensity profile should
appear symmetrical, similar to the graph shown in Figure 12-3. The black dashed
curve is a polynomial fit to the measured profile, and the reported “MSE=” value
quantifies the symmetry of the profile. If the MSE value is greater than 0.02,
contact your J.A. Woollam representative.

Figure 12-3. Results of a typical z-stage scan.

12-486 • Reference – Hardware Tab CompleteEASE Software Manual

 Calibration
S-T Baseline
This acquires baseline spectra for the “Straight-Through” configuration such that
accurate transmission intensity and Mueller-matrix data can be acquired. For best
data accuracy, always perform an S-T Baseline after moving the ellipsometer optics
or if it has been more than one hour since the previous baseline.
The S-T Baseline will collect data from the measurement beam as the polarizer and
compensator positions are rotated. This will provide details to correct for any
misalignment of the optical element azimuths relative to each other and more
importantly measure the baseline intensities at various optical element
combinations.

Off-Sample Baseline
This acquires data at the current angle of incidence over a range of optical element
(polarizer and compensator) positions to improve data accuracy, especially for
intensity and Mueller-matrix measurements. Because this baseline is angle-
dependent, the best intensity and Mueller-matrix data follow an Off-Sample
Baseline at the angle of incidence intended for measurement.
This option also calibrates and stores the angle of incidence which is used for
subsequent measurements. For best angle of incidence accuracy, perform an Off-
Sample Baseline after moving the source or receiver units or after mounting the
sample chuck.

Angle Offsets
Calculates the angle offsets. This is usually performed as part of a calibration
procedure.

Show Logs
Hardware / Error
These buttons display the Hardware and Error logs, which contain information
useful for diagnosing and debugging problems with the instrument. An example
Error Log is shown in Figure 12-4. Use the ‘Copy to Clipboard’ button to paste the
log information into another program such as Microsoft Word.
You can use the “Date Range” fields to limit the range of data shown in the log.
Enter starting and ending dates and click ‘Update Range’. To return to the full log,
click ‘Show Full Range’.
If the instrument is not working properly, your J.A. Woollam representative may
require these files to diagnose potential problems. From the Options tab, click
‘Create Debug File’. This will create a zip file containing both logs
(C:\CompleteEASE\CompleteEASE_Debug.zip). Email this zip file to your J.A.
Woollam representative.

CompleteEASE Software Manual Reference – Hardware Tab • 12-487

 Figure 12-4. Error Log.

NOTE: If the size of the ErrorLog.txt file is greater than 5 MB, it is compressed and
saved to C:\CompleteEASE\cnf\Archive and a new file is created.

Camera
The camera is an optional accessory for the alpha-SE that may or may not have
been purchased with your system.

Show Image Dialog

Click ‘Show Image Dialog’ to open the camera image (see Figure 12-5). Use the
checkboxes to “Show Grid” and “Show Beam Position”. Images can also be saved
using the ‘Save Image’ button.

NOTE: The measurement beam will not be visible on smooth, specular surfaces, but
the beam position can still be identified using the “Show Beam Position” checkbox,
if ‘Mark Beam Position’ has been completed.

12-488 • Reference – Hardware Tab CompleteEASE Software Manual

 Figure 12-5. ‘Show Image Dialog’.

Mark Beam Position

The measurement beam position can be marked such that users can identify where
the beam is located, even for smooth, specular samples. See your hardware
manual for full details.

Illumination On
Use this checkbox to illuminate the sample if needed for better camera image.

Routine Test Measurement

Measure
Click this button to start the routine test measurement. CompleteEASE assumes
that the 25 nm calibration wafer has been loaded and aligned (see Figure 12-6).

Figure 12-6. Routine Test Measurement.

CompleteEASE Software Manual Reference – Hardware Tab • 12-489

 Show Results
After completing at least two routine test measurements, the user will be able to
view the results by clicking the ‘Show Results’ button to graph history of the routine
test measurements (see Figure 12-7).

Figure 12-7. Example results from routine test measurements performed over ~1 month.

12.3. Hardware Tab – M-2000 and RC2

The Hardware tab for an M-2000 or RC2 system consists of three main panels:
Hardware Status, System Information, and Controls (see Figure 12-8). This section
describes these panels and the controls found on them.

NOTE: Full details of the Hardware tab can be found in your hardware manual.
This chapter is meant only as a general reference.

Figure 12-8. Example RC2 Hardware tab.

Hardware Status
Displays the current state of the hardware. The same information is also displayed
in the System Status panel of the Measurement tab.

12-490 • Reference – Hardware Tab CompleteEASE Software Manual

 Controls
Provides access to hardware functions other than those needed for acquiring data.
These hardware functions are separated into groups displayed as tabs. The tabs are
General, Calibration, System, and Misc.

General
Contains several panels and buttons and is nearly the same for all M-2000 and RC2
configurations (Figure 12-9).

Figure 12-9. General tab.

Alignment
Align Sample
Starts sample alignment procedure without acquiring data after alignment is
complete. For systems with automated alignment capabilities, a dialog is displayed
allowing the user to choose between manual alignment and automated (“full”)
alignment (see Figure 12-10). Choosing ‘No’ in the dialog will proceed with manual
alignment. An alignment screen with 4-quadrant detector position and Z-intensity
is shown in Figure 12-11.

Figure 12-10. The “Sample Alignment” dialog gives the user a choice of automated (“full”)
alignment or manual alignment.

CompleteEASE Software Manual Reference – Hardware Tab • 12-491

 Figure 12-11. Example alignment screen.

Display Signal
Displays the raw signal intensity (see Figure 12-12). This screen is useful for
diagnosing light intensity problems. The Average AC Intensity includes all channels
of the CCD spectrometer. The DUV AC Intensity is an average of the first 3
channels, while the UV AC Intensity is an average of the first 10% of CCD channels.
The IR AC Intensity is an average of the last 10% of CCD channels.

Figure 12-12. Display Signal Screen.

12-492 • Reference – Hardware Tab CompleteEASE Software Manual

 Other
System Check
Starts a System Check, which is the most common procedure for calibrating the
ellipsometer. See hardware manual for details on the System Check procedure.

Move To Load Pos. / Return to Center

‘Move To Load Pos.’ will position the stage, source, and receiver to make loading a
sample easier, if these components are automated. After moving to the load
position, this button will display ‘Return to Center’. ‘Return to Center’ will move
the stage to (0, 0) after sample load is complete.

Routine Test Measurement

Measure
Starts a routine test measurement. Using this feature requires that a recipe be
created and placed in the “\CompleteEASE\Recipe\Routine Test Measurement”
directory. The recipe name also needs to be listed in the “Routine Test
Measurements” section of the Hardware Configuration, as shown in Figure 12-13.
To make this feature useful, users will need to be trained to always use the same
sample and to put the sample in the same position on the stage each time they
perform the measurement.

Figure 12-13. “Routine Test Measurements” section of the Hardware Configuration.

Show Results
Shows all past results of the “Routine Test Measurement”. This is a convenient way
to track the performance of a system with results on the same sample. The values
that can be graphed include any fit parameters for the model of the designated test
sample (thickness, refractive index, etc.) and many hardware parameters such as
intensity and alignment values.

CompleteEASE Software Manual Reference – Hardware Tab • 12-493

 Calibration
This tab appears different depending on the type of ellipsometer. Figure 12-14
shows the Calibration tab for an M-2000 and Figure 12-15 shows the tab for an
RC2.

NOTE: Calibrations should only be performed when instructed by your J.A.

Woollam representative.

Initialize Hardware
Re-initializes the ellipsometer hardware, which initiates communication over the
USB or TCP/IP link, scans the spectrometer and moves motorized components to
their home positions. Details of any errors that occur during hardware initialization
can be found in the error log.

DC Offset (M-2000) / DC Offset Calibration (RC2)

Measures the ambient light and electronic noise contribution to the measured raw
intensity. This button is most often used before acquiring depolarization data to
improve measurement accuracy, as depolarization data is determined from the DC
component of the measured signal. A System Check automatically performs a DC
Offset as well.

Figure 12-14. Calibration tab for M-2000 ellipsometers.

Coarse Calibrate and Fine Calibrate (M-2000)

These calibrations can be performed in place of a System Check when it is not
possible to use the standard calibration wafer that is supplied with the instrument.
This situation occurs during in-situ measurements on a deposition chamber that
requires a specific sample size. These calibrations will not determine the in-plane
window effects.

NOTE: ‘Coarse Calibrate’ and/or ‘Fine Calibrate’ should only be used when
instructed by your J.A. Woollam representative.

Phase Sensor Info (M-2000)

This diagnostic function displays information used to determine if the phase sensor
is working and the compensator motor is spinning. When troubleshooting a
hardware problem, a J.A. Woollam representative may ask for this information.

12-494 • Reference – Hardware Tab CompleteEASE Software Manual

 Figure 12-15. Calibration tab for RC2 ellipsometers.

Standard Calibration (RC2)

A Standard Calibration is equivalent to a Coarse Calibration on the M-2000. The
System Check is the preferred method for calibrating the instrument when the
standard calibration wafer can be used.

NOTE: ‘Standard Calibration’ should only be used when instructed by your J.A.
Woollam representative.

Light Source Intensity (RC2)

Changes the current light intensity. Normally this is left in automatic mode.

System Status (RC2)

Displays a diagnostics dialog with information on the phase sensors and USB
communication. When troubleshooting a hardware problem, a J.A. Woollam
representative may ask for this information.

Lamp ON (RC2)
Turns the ellipsometer light source (lamp) on or off.

System
Controls in the System tab vary depending on the ellipsometer base type and
configuration or accessories available. Figure 12-16 shows an example System tab
for a horizontal-base system and Figure 12-17 shows an example System tab for a
vertical-base system. Brief descriptions of the different buttons on this panel can
be found below. See hardware manual for more details.

CompleteEASE Software Manual Reference – Hardware Tab • 12-495

 Figure 12-16. Example System tab for a horizontal-base system.

Angle, Z-Stage, Translator, Sample Tilt (Horizontal)

Allows user to change the positions of the motorized base components.

Install New Sample Stage (Horizontal)

Steps the user through the process of installing a sample stage attachment such as
a translator, heat stage, liquid cell, etc. On bases with automatic detection
capabilities, CompleteEASE will automatically detect which stage is being installed.
Bases without automatic detection will display a dialog with stage options for the
user to select.

Install/Remove Focus Probes (Horizontal)

Configures software and provides instructions for installing/removing focus probes.

Figure 12-17. Example System tab for a vertical-base system.

Move Angle (Vertical)

Moves angle to a specified position.

12-496 • Reference – Hardware Tab CompleteEASE Software Manual

 Set Angle (Vertical)
Vertical-base systems do not have a home sensor for the angle, so ‘Set Angle’
allows users to set or correct the current angle based on the goniometer readings.

Move Translator / Move Rotator (Vertical)

Moves translator or rotator to specified position.

Home Translator / Home Rotator (Vertical)

Homes translator or rotator to determine maximum positions and/or correct the
read position.

Configure System Components (Vertical)

Displays a dialog that allows the user to specify the attachments that are currently
installed (see Figure 12-18). These attachments can include a translator, focus
probes, camera, rotator, etc.

Figure 12-18. Installed Components dialog for vertical-base systems.

Activate In Situ Mode

Click this button and restart CompleteEASE to activate In situ mode, which holds all
base motors stationary and removes control buttons for base motors.

Misc.
Other functions that are dependent on the ellipsometer base and installed
accessories (see Figure 12-19). Some systems will not have a Misc. tab.

CompleteEASE Software Manual Reference – Hardware Tab • 12-497

 Figure 12-19. Example Misc. tab.

Display Sample Image (Systems with Camera)

Displays the camera image of the sample in a dialog. The dialog can stay open while
other operations are performed.

Z-Stage Scan (Systems with Auto-Z)

Scans the Z-Stage over a specified range and graphs the signal intensity versus Z-
Stage position. This is especially useful when working with transparent substrates
to determine if the backside beam is being collected.

Additional Camera Functions

Additional camera functions can include ‘Mark Measurement Beam Location’ and
‘Measure Camera Field Of View’, as shown in Figure 12-20. See hardware manual
for details.

Figure 12-20. Additional Camera Functions.

Additional Hardware Functions

May include other hardware functions specialized for the individual system. An
example with ‘Angle Offset Correction’ and ‘Check Probe Delta Offsets’ is shown in
Figure 12-21.

12-498 • Reference – Hardware Tab CompleteEASE Software Manual

 Figure 12-21. Example Additional Hardware Functions dialog.

Angle Offset Correction

Measures the calibration sample over a range of angles and then applies a
correction to the mechanical positions of the motors to produce a more accurate
angle of incidence for subsequent measurements.

Check Probe Delta Offsets

Moves the ellipsometer back into the Straight-Through position with focusing
probes attached and remeasures delta offsets.

System Information
This panel displays information about the ellipsometer including the system type,
EID number, base type, wavelength range, and current motor positions. This panel
also provides access to the hardware configuration dialog, hardware manual, and
hardware log.

Figure 12-22. Example System Information panel.

NOTE: J.A. Woollam representatives may ask for information from the System
Information panel when providing assistance.

‘Show Hardware Manual (F2)’

Opens hardware manual.

CompleteEASE Software Manual Reference – Hardware Tab • 12-499

 NOTE: The hardware manual must be present in the CompleteEASE directory for
the hardware manual to open. Typically, the hardware manual is only available on
computers connected to the ellipsometer.

‘View Hardware Log’

This button displays the hardware log which contains information on calibrations
and initializations.

‘Edit Hardware Configuration’

Opens the Hardware Configuration dialog which contains many settings for
configuring the hardware behavior (see Figure 12-23).

Figure 12-23. Hardware Configuration dialog.

Expand a category under “Configuration Parameters” and click on a parameter to

show its description and edit the value (see Figure 12-24).

Figure 12-24. Expand category to view parameter descriptions and edit values.

12-500 • Reference – Hardware Tab CompleteEASE Software Manual

 NOTE: A few select parameters are described in detail below. For more detailed
descriptions of these configuration settings, refer to the descriptions provided
within the “Hardware Configuration” dialog in CompleteEASE or in your hardware
manual.

+Window Effects
Window effects are used only when an attachment is employed that has windows
through which the ellipsometer beam must pass when measuring the sample.
Examples include liquid cells, heat stage, environment cell, etc.

Out‐of‐Plane Effects
Out-of-plane effects are measured when performing a calibration such as a System
Check if Window Effects are enabled.
“WinRet1” and “WinRet2” are the out-of-plane window retardance for the entrance
and exit windows respectively.
“WinDisp1” and “WinDisp2” are the dispersion terms for the out-of-plane window
retardance of the entrance window and “WinDisp2-1” and “WinDisp2-2” are the
dispersion terms for the exit window.
“RotDisp1” and “RotDisp2” are terms for the rotary power of the entrance window
and “RotDisp1-2” and “RotDisp2-2” are the terms for the rotary power of the exit
window.

NOTE: The effects of the out‐of‐plane window retardance on ellipsometric data are
canceled out (at least to the 1st order) when performing a zone-averaged (high
accuracy) measurement.

In‐Plane Effects
In-plane effects are measured during a System Check or a Delta Offset Calibration if
Window Effects are enabled.
“Delta Offset Mode” sets whether the effects are describing prism dispersion or
window (1/λ) dispersion.
“Delta Offset Err. Tolerance” specifies the maximum error allowed on a Delta Offset
parameter. If the error bar is above this tolerance then the calibration is considered
to have failed.
“Min Delta Offset1 Limit” specifies the smallest Delta Offset 1 parameter before the
higher order terms will be set to 0 (turned off).
“# of Higher Order Terms” specifies the number of Delta Offset terms to fit for
beyond Delta Offset1 during a System Check or Delta Offset calibration.
“Delta Offset(1-4)” parameters are the terms for describing the in-plane window
retardance. The in-plane window retardance as a function of wavelength for a
prism (Delta Offset Mode=1) is given in Equation 12-1 where δ is the retardance as
a function of the wavelength λ in angstroms and Δn is the nth delta offset.
Equation 12-1

5000 2 5000 4 𝜆 2
δprism (λ) = Δ1 + Δ2 ( ) + Δ3 ( ) + Δ4 ( )
𝜆 𝜆 5000

CompleteEASE Software Manual Reference – Hardware Tab • 12-501

 The in-plane window retardance as a function of wavelength for a window (Delta
Offset Mode=2) is given in Equation 12-2 where δ is the retardance as a function of
the wavelength λ in angstroms and Δn is the nth delta offset.
Equation 12-2

5000 5000 2 5000 4 𝜆 2

δwindow (λ) = (Δ1 + Δ2 ( ) + Δ3 ( ) + Δ4 ( ) )
𝜆 𝜆 𝜆 5000

Psi Offsets
Psi Offset parameters can be used to correct the Psi data when using prisms to
bend the beam and are only necessary when the ellipsometer beam enters the
prism(s) at angles away from normal incidence. Users must determine the correct
Psi offset values – they are not determined during calibration. The Psi correction is
given by Equation 12-3 where ψ is the Psi correction as a function of the
wavelength λ in angstroms and Ψn is the nth psi offset.
Equation 12-3

2000 2 2000 4 2000 6 2000 8

ψcorrection (λ) = Ψ0 + Ψ1 ( ) + Ψ2 ( ) + Ψ3 ( ) + Ψ4 ( )
𝜆 𝜆 𝜆 𝜆

+Alignment Parameters

Include Alignment Parameters With Data

Configures CompleteEASE to save alignment information along with the
ellipsometry data (see Figure 12-25). This can be useful in conjunction with
mapping measurements. For example, if high MSE is obtained at some location, the
alignment parameters at that location can be checked for consistency.

Figure 12-25. “Include Alignment Parameters With Data”.

To view the alignment parameters:

1. From the Measurement tab, Fit Results panel, click ‘View Prev. Results’,
which opens the “View Previous Results” dialog (see Figure 12-26).
2. Click ‘Choose File’ or ‘Choose folder’ to add file(s) of interest to the Files
panel.
3. Select one or more files in the Files panel.

12-502 • Reference – Hardware Tab CompleteEASE Software Manual

 Figure 12-26. View alignment parameters in “View Previous Results” dialog.

A table of acquisition parameters will appear in the File Information panel. Click
‘Graph Data’ to graph parameters versus position (see Figure 12-27).

Figure 12-27. Example graph of signal intensity (“SigInt”) vs. translator position.

If you select more than one file, you can also select ‘Calc Multi-Sample Stats’ to see
parameters from more than one data file in the table and graphs (see Figure 12-28).

Figure 12-28. Example 'Calc Multi-Sample Stats'.

CompleteEASE Software Manual Reference – Hardware Tab • 12-503

13. Reference – Options Tab

The Options tab is shown in Figure 13-1. This tab consists of four panels: Display
Units, Configuration Controls, About CompleteEASE, and Miscellaneous. A brief
description of all commands within this tab is given.

Figure 13-1. CompleteEASE Options tab.

13.1. Display Units

Sets the wavelength, thickness, and optical constants display units.

Wavelength Units
Choose from Å, nm, m, eV, and 1/cm for the wavelength units.
Wavelength is the distance between adjacent peaks of an electromagnetic wave,
commonly given in angstroms (Å), nanometers (nm), or microns (m).
Wavenumber refers to the number of waves in a length of one centimeter, with
units of reciprocal centimeters (1/cm).
If light is treated as a quantum of radiant energy (photon), its energy is given in
electron volts (eV) which is proportional to frequency. Choice of units depends
largely on the spectral range and application. Equation 13-1 shows the conversion
between wavelength and photon energy used in CompleteEASE.
Equation 13-1

1240
𝜆𝑛𝑚 =
𝐸𝑒𝑉

Thickness Units
Choose from Å, nm, m, and mm for the thickness units.

13-504 • Reference – Options Tab CompleteEASE Software Manual

 Optical Constants
Display and report optical constants as either complex refractive index (n & k) or
complex dielectric function (e1 & e2, which stand for 1 and 2). The two
descriptions for optical constants are related by Equation 13-2.
Equation 13-2

𝜀̃ = 𝑛̃2

NOTE: All settings are saved when the program is exited and remain selected when
the program is restarted.

‘Use Defaults’
Return all settings to their default values, which include wavelength in nm,
thickness in nm, and optical constants as “n & k”.

13.2. Miscellaneous
The Miscellaneous panel contains five buttons, as shown in Figure 13-2.

Figure 13-2. Miscellaneous panel from Options tab.

‘Reset Window Size’

Returns the CompleteEASE window to the default size for basic, small-screen, low-
resolution monitors.

‘Manage Users’
Opens the “User Management” window where a user list can be created, along with
access levels and passwords (see Figure 13-3). Access level is selected from SysOp,
Engineer, or Technician.
• SysOp – access to all software features.
• Engineer – customizable from a list of select privileges.
• Technician – access to Measurement tab only. Users limited to technician
level cannot create or edit recipes or analysis models.
The Engineer level can be customized by clicking ‘Define Engineer Privileges’. A list
of selectable privileges for this level is shown in Figure 13-4. User customization is
further defined in the Configuration Controls (see section 13.3).

CompleteEASE Software Manual Reference – Options Tab • 13-505

 Figure 13-3. “User Management” window.

Figure 13-4. “Edit User Privileges” window for Engineer access level.

‘Show Manual (F1)’

Click ‘Show Manual’ or press F1 to open a PDF copy of this software manual.

NOTE: From the instrument computer, you can press F2 to open a PDF copy of the
hardware manual. Hardware manuals are specific to ellipsometer hardware
configuration.

NOTE: The manual(s) are in the C:/CompleteEASE folder and can be accessed
directly from there as well.

‘View CompleteEASE Shortcuts’

Shows list of CompleteEASE shortcut keys (see Figure 13-5). See also Section 1.6
for a table of CompleteEASE shortcuts in this manual.

13-506 • Reference – Options Tab CompleteEASE Software Manual

 Figure 13-5. List of CompleteEASE shortcuts.

‘Deactivate’
Refer to Appendix 14.2 for information on Activation or Deactivation.

CompleteEASE Software Manual Reference – Options Tab • 13-507

 13.3. Configuration Controls
‘Edit Configuration’
Click to access the “CompleteEASE Configuration” window (see Figure 13-6) where
customizable options within CompleteEASE are defined. Expand each section to
display customizable features. Check “Show Advanced Configuration Options” to
show the advanced list (see Figure 13-7).

Figure 13-6. “CompleteEASE Configuration” window.

Figure 13-7. “CompleteEASE Configuration” window with “Show Advanced Configuration

Options” checked.

Most of the settings within the CompleteEASE Configuration are self-explanatory

and contain the default settings for the software. If you select an item, a brief
description will be displayed on the right side of the window. Each configuration
setting is listed in this chapter, with a few discussed in detail.

13-508 • Reference – Options Tab CompleteEASE Software Manual

 General
General purpose configuration parameters (see Figure 13-8).

Figure 13-8. General section.

Font Name
Name of CompleteEASE font family. Changes are made from the drop-down menu,
as shown in Figure 13-9.

Figure 13-9. Font Name drop-down menu.

Font Size
Point size of CompleteEASE font.

Graph Font Size

Point size of font used for text in graphs.

CompleteEASE Software Manual Reference – Options Tab • 13-509

 Reverse Order Of Graph Colors
Reverse the order of the colors used for graphing contour and surface graphs.

Font Scaling Factor

All font sizes are multiplied by this factor to increase or decrease the size of text.
This parameter can be useful for enlarging text on high DPI displays.

Look & Feel

Specifies the look & feel of the user interface elements, with choices from a drop-
down list, as shown in Figure 13-10

Figure 13-10. Look & Feel drop-down menu.

System Time Mode

Specifies the method used by CompleteEASE to determine time, with the following
choices:
0. Java time call
1. Windows GetTickCount
2. Windows timeGetTime
3. Windows Performance Counter
4. CPU Counter

NOTE: Changing the default setting for “System Time Mode” is not recommended.

Operating System File Dialog

Indicates that the operating system’s file dialog will be used instead of the Java one.

User Log-In Mode

Specifies how users will log into and access software functions. This feature is
important when setting up multiple users, as described in Section 13.2.

Auto Fit Enabled

Specifies if “Auto-Fit” is enabled when CompleteEASE is started. This is related to
the “Auto Fit” checkbox from the Analysis tab, Fit panel. See Section 9.3 for a
description of this behavior.

13-510 • Reference – Options Tab CompleteEASE Software Manual

 Always Add CE Folders
Indicates that folders under the base folder (usually C:\CompleteEASE) should
always be added to the user’s recent file location list when CompleteEASE is
started. This feature is important when setting up multiple users, as described in
Section 13.2.

User Have Own Folders List

When user log-in is required, CompleteEASE will keep each user’s recent file
location list separately. This feature is important when setting up multiple users, as
described in Section 13.2.

Clear Data on Logout

When user log-in is required, CompleteEASE will clear the data, model, and selected
recipe when a user logs out. This is important to prevent proprietary results from
remaining within CompleteEASE after a user has logged out.

Always Show In situ Tab

When this option is selected, the In situ tab will always be displayed. When this
option is not selected, the In situ tab will only be displayed when the installed
sample stage supports in-situ measurements (liquid cell, heat stage, environment
cell, etc.).

NOTE: Use CTRL+ALT+SHIFT+U to access the In situ tab during the current session
of CompleteEASE only.

Use Unicode
Specifies the character encoding for saving configuration files and models. Must be
set on 'ON' to use extended (non-Latin) character sets; setting to 'OFF' improves
backwards compatibility with versions of CompleteEASE before CompleteEASE 4.

Force Read of Tabulated Mat Files

Indicates that all tabulated material files should be read from the disk instead of
from the model file. This should almost always be set to FALSE.

NOTE: This option can cause unexpected and erratic behavior if not used properly.

# of Cores for Data Analysis

Specifies the # of processor cores for data analysis: ‘0’ means all the available cores
are used. This value can be overridden by the model options for specific models
(see Section 9.8).

Always Save Measurement Data

Disables the “Save Data after Measurement” checkbox and forces the user to
always save measurement data.

Default To Double Y Axis

Indicates whether the double Y axis should be displayed on the graph by default
when starting CompleteEASE.

CompleteEASE Software Manual Reference – Options Tab • 13-511

 Display Pseudo Transforms
Specifies whether the Pseudo Transforms of <n>, <k>, <e1>, and <e2> should be
shown in the +Data menu (see Section 7.4 for an explanation of Pseudo
Transforms).

Show Greek Letter Symbols

Indicates whether Psi, Delta, and Epsilon should be shown as their Greek letter
symbols instead of words.

Minimum Delta Value

Indicates the minimum value delta can take before it wraps around by 360 degrees.

Enforce Fit Parameter Range Limits

Specifies whether CompleteEASE will allow a user to set or mouse-roll a fit
parameter’s value outside of its defined range.

Reverse Mouse Wheel Direction

Specifies which direction rolling the mouse wheel should move parameters. If you
are using Windows, this should generally be “false”. If you are using a Mac, this
should generally be “true” if using a traditional mouse and “false” if using a
trackpad.

Tabulated Mat File Save Mode

Specifies in which units values for tabulated material files will be saved to the file.
“Default Mode” will save every file in Angstroms with e1 and e2 values. “Display
Units Mode” will save each file with whichever units and optical constants are
selected for display in the configurations on the Options tab (see Section 13.1).

Memory Management
Maximum number of wavelengths and timeslices for which to allocate space (see
Figure 13-11).

Figure 13-11. Memory Management section.

13-512 • Reference – Options Tab CompleteEASE Software Manual

 Max. Wavelengths
Controls the amount of memory required by CompleteEASE by setting a maximum
number of wavelengths.

Max. Time Slices

Controls the amount of memory required by CompleteEASE by setting a maximum
number of timeslices for dynamic data.

NOTE: This option also controls the amount of memory required for translation and
rotation points, multiple data sets, multiple angles of incidence, etc. For example, if
you have a uniformity map with 500 points and each one has four angles of
incidence, then this number needs to be at least 2000 (500 * 4).

Log Memory Usage

When this option is on, additional memory logging files will be created in the
CompleteEASE install directory. Those files can be used by your J.A. Woollam
representative for debugging memory usage problems.

Display Units
Changes the units that different parameters are displayed in (see Figure 13-12).

Figure 13-12. Display Units section.

Light Units
Units to display light wavelength or photon energy with choices of Å, nm, m, 1/cm,
or eV (see Section 13.1).

Time Units
Units for time with choices of seconds, minutes, and hours.

Rate Units
Units for rate of dynamic growth/etch with choices of Å/s, Å/min, Å/hr, nm/s,
nm/min, nm/hour, and m/hour.

CompleteEASE Software Manual Reference – Options Tab • 13-513

 Thickness Units
Units for thickness with choices of Å, nm, m, and mm.

Opt. Const. Units

Units for Optical Constants with choices of “e1 & e2” or “n & k” (see Section 13.1).

Temperature Units
Units for temperature with choices of °C or K.

Translator Units
Units for translator coordinates with choices of cm or mm.

% Range Calculation
This allows the choice of calculation for “% Range” in the map statistics tables with
choices of Range/Average or Range/(2*Average).

File Save Prompt

Automatically prompt to save data after each measurement.

Remote Communications
Parameters for 1) setting up how communication between a remote computer (i.e.
a computer controlling a process) and the ellipsometer is handled and 2) setting up
how parameter values are returned through the “get()” functions (see Figure
13-13). For more information about controlling remote communication, refer to
Appendix 14.8).

Figure 13-13. Remote Communications section.

Communication Type
Method of communication between a remote computer and CompleteEASE.
Choosing “None” indicates no communication will take place. Other options
include “TCP/IP” and “RS232”.

13-514 • Reference – Options Tab CompleteEASE Software Manual

 TCP/IP Port #
Port number that the CompleteEASE server listens on.

Use System Proxy Settings

This setting tells CompleteEASE to use the system proxy settings when
communicating over the remote port. This should normally be set to false.

RS232 Baud Rate

Baud rate for serial communications.

Serial Port
Serial port to use with serial communications.

Auto Report
Turning Auto Report on will result in a string containing the values of current model
fit parameters to be sent out on the Asynchronous TCP/IP port after data is
acquired and fit. This is only done if the communication type is set to TCP/IP. See
Appendix 14.8 for a more detailed explanation of this feature.

Auto Report Format

Specifies the format of data reported using Auto Report. “Status Text” will return
the parameters as listed in the Simple Mode status box. “All Parms” will return all
of the parameters including those that are hidden.

Allow Parameter Query

This will add an additional checkbox in the “Edit Fit Parameter” dialog. When this
checkbox is checked for a specific fit parameter the model will try to ask via TCP/IP
for the value of the parameter before the fit (the remote computer must be
connected to port "TCP/IP Port #" plus 2). See Appendix 14.8 for a more detailed
explanation of this feature.

Save Results to Database

Specifies that the fit results from a recipe measurement will be added to a
database. The URL for the database should be set in a file named “DB-URL.cnf” in
the cnf folder.

RunRecipe – Out of Spec

Specifies that when a RunRecipe() command is received, the return string should
include “:OOS” for each parameter that is out of spec or “:EEL” for each parameter
whose error bar exceeds the limit as defined in the model.

RunRecipe – Time Out (seconds)

Specifies the maximum amount of time to wait for a recipe to finish before sending
an error message to the remote computer. The purpose of this feature is to allow
CompleteEASE to inform the remote computer that a problem has occurred instead
of having the remote computer wait indefinitely. The default is 180 seconds, so if
the user’s longest recipe is longer than 3 minutes, this value should be changed to

CompleteEASE Software Manual Reference – Options Tab • 13-515

 make sure that a false error message is not sent. For example, if the user’s longest
recipe takes 3 minutes to complete, the timeout can be set to 240 seconds.

RunRecipe – Save By Date

Specifies that when a RunRecipe() command is received without a specific directory
to save to, CompleteEASE will automatically save data into timestamped folders
under the REMOTE_DATA project folder.

System Commands Time Out

Specifies the maximum amount of time to wait for a system command to finish
before sending an error message to the remote computer.

Window Setup
These display settings are automatically saved when CompleteEASE is closed (see
Figure 13-14), except for “Title Text” and “Start Up Tab” which remain constant to
specified values.

Figure 13-14. Window Setup section.

Title Text
Additional text that appears in the title bar of the CompleteEASE program.

Start Up Tab
The tab that is selected by default when CompleteEASE starts. If users are required
to log in, then the selected tab is the default tab when a technician-level user logs
in.

NOTE: “Ex Situ” is equivalent to the Measurement tab.

Width, Height, Top Left Corner X, Top Left Corner Y,

Divider Position, Maximized
Each of these options stores information about how the CompleteEASE main
window appeared when CompleteEASE was exited. They are then used to restore
the CompleteEASE window to this state when it is started again. These parameters
are included here for display purposes only. Changing these values will have no

13-516 • Reference – Options Tab CompleteEASE Software Manual

 effect since they are overwritten every time CompleteEASE is closed and can only
be applied when CompleteEASE starts.

Graph Clipboard Parms

Parameters for setting the size and resolution of the graph for saving graph images
and RTF files to the clipboard (see Figure 13-15).

Figure 13-15. Graph Clipboard Parms section.

Width in Pixels
Width of the graph on the clipboard in pixels.

Aspect Ratio
Specifies the ratio (Graph Width)/(Graph Height) when copied to the clipboard.

Resolution Factor
Oversampling factor for the graph and model; this value is multiplied by the graph
width and aspect ratio to determine the total number of pixels stored in the image.
It is also used when the model is copied to the clipboard to determine the size of
the image.

Auto Legend Position

Automatically places the legend so that it minimizes data coverage when copying a
graph to the clipboard.

RTF Left/Right Margin

The left and right page margins to use when creating RTF content for the clipboard.

RTF Top/Bottom Margin

The top and bottom page margins to use when creating RTF content for the
clipboard.

Hardware Simulation Mode

Simulates the real time acquisition of data without using the ellipsometer hardware
(see Figure 13-16).

CompleteEASE Software Manual Reference – Options Tab • 13-517

 Figure 13-16. Hardware Simulation Mode section.

Simulation Enabled
When checked, CompleteEASE uses simulation mode instead of the physical
ellipsometer hardware to acquire data.

Simulation Data File

Specifies the CompleteEASE data file to automatically load if Simulation Mode is
enabled.

In Situ Tab Parameters

Contains settings specific to the In situ tab (see Figure 13-17).

Figure 13-17. In Situ Tab Parameters section.

Show Error Bars in Parameter Graph

When this setting is turned on, the parameter graph will display 2 curves instead of
4, with the error bars for curves 1 and 2 displayed on the right Y axis.

System Check Button on In Situ panel

When set to true, a ‘System Check’ button will appear on the In situ tab. When set
to false, a ‘Check Substrate’ button will appear.

13-518 • Reference – Options Tab CompleteEASE Software Manual

 # of Model pages-Insitu
Number of model pages available in the In situ tab. Each page holds 10 models.
There is a minimum of 4 pages and a maximum of 8.

Rep. Converter Name

If the name of a program is listed here, the program will be run after a model report
(generated by the simple in situ mode) has been written to disk. The program will
be given the report file name as a command line parameter. The program needs to
be located in the CompleteEASE directory.

Show Received Cmds

Displays the text of Received Commands in the In situ tab status box.

Allow Second Model

When this setting is on, the user can assign a second model to a button that will be
fit to the data just as the first, but no report will be generated. The second model’s
parameters will show up in the parameter graph provided the fit parameters are
named differently than the first model’s fit parameters.

Single Model Mode

Specifies which model button will be activated when CompleteEASE is started.
Setting to ‘0’ disables this mode.

Show Fast Acq. Box

Includes the “Fast Acquisition” checkbox in the In situ tab.

Data Export Parameters

Contains settings for setting up data export features (Figure 13-18).

Figure 13-18. Data Export Parameters section.

Custom Export Application

This entry specifies the name of a custom, user-supplied application that will be run
after each measurement to export the results to an external program or database.
“Export measurement results” should be checked (in the Create/Edit Recipe dialog,

CompleteEASE Software Manual Reference – Options Tab • 13-519

 see Figure 13-19), and the application must be located in the CompleteEASE
directory.

Figure 13-19. "Create/Edit Recipe" dialog.

Append Timestamp to Data File

When checked, the current time will be appended to the name of data files
acquired through a recipe measurement.

Export Recipe Data as Text

Performing a recipe measurement when this parameter is set to true will result in
CompleteEASE saving an additional file with the data recorded in text format.

Export Maps As Individual Data Files

When this entry is on along with “Export Recipe Data as Text”, recipes with scan
patterns will export map files as multiple files each containing the data at a single
measurement location.

Include Hardware in 1Pt Export Format

The new export format only applies when exporting results from single-point recipe
measurements (no maps). When this parameter is set to true, the hardware-
related parameters will be added to the exported text file. When this parameter is
false (default), only the fit parameters are added.

Export Results File Format Version

Sets the data format for the export results file. Version 1.0 is the default and
version 2.0 adds a variable-size header that allows additional information such as
EID and user name to be added to the file.

Shutter Controls
Optional control parameters for setting up user control of shutters (see Figure
13-20).

13-520 • Reference – Options Tab CompleteEASE Software Manual

 Figure 13-20. Shutter Controls section.

Treat as Signal
Treat the shutters as signals instead of shutters. The difference is that in process
control conditions, a signal remains off until the target thickness is reached whereas
the shutter opens (turns on) when the growth starts and closes (turns off) when the
target thickness is reached.

# of Shutters
The number of shutters to be configured in CompleteEASE. The maximum number
is four.

‘Import Configuration’ / ‘Save Configuration’

To copy the configuration from one computer to another, you can ‘Save
Configuration’ and then ‘Import Configuration’. This will help to setup
CompleteEASE in the same way on multiple computers.

13.4. About CompleteEASE

This panel displays information about CompleteEASE as shown in Figure 13-21,
which contains the software version number, memory availability, copyright
information, and license key information.

CompleteEASE Software Manual Reference – Options Tab • 13-521

 Figure 13-21. About CompleteEASE panel.

‘Create Debug File’

Click to create a debug file (CompleteEASE_Debug.zip) saved to C:\CompleteEASE\.
Your J.A. Woollam representative may ask for this when troubleshooting.

NOTE: A list of J.A. Woollam representatives by country can be found on our

website (www.jawoollam.com).

‘View Error Log’

View the error log.

Note: CompleteEASE_Debug.zip will also contain an error log.

13-522 • Reference – Options Tab CompleteEASE Software Manual

14. Appendices

14.1. Installing CompleteEASE

Before Installation
Before installing CompleteEASE, verify that the computer used at least meets all the
minimum requirements listed in Table 14-1. Recommended specifications are also
provided for your information. In general, the better the specifications on your
computer, the faster and more efficient CompleteEASE will operate.
Table 14-1. Computer requirements for CompleteEASE.

COMPUTER COMPONENT REQUIREMENTS

Operating System Microsoft Windows 7, 8, 8.1, 10 (Acquisition & Analysis)
macOS X 10.9 – 10.15 (Analysis Only)
Memory (RAM) Minimum: 2 GB
Recommended: 4+ GB
Storage Space Minimum: 1 GB Available
Recommended: 2+ GB Available
Processor Minimum: 1 GHz Clock Speed, 1 Core
Recommended: 3+ GHz Clock Speed, 4+ Cores
Display Minimum: 1024 x 768 Pixel Resolution

Upgrading to CompleteEASE 6
CompleteEASE 6 is the latest version of the CompleteEASE software package,
containing many new features for data analysis. It is backward-compatible with
previous versions of CompleteEASE along with previous data, model, material, and
snapshot files. This means that it can open files that were created with earlier
versions of CompleteEASE. However, CompleteEASE 6 uses a new file type for its
data files but maintains the same “.SE” and “.iSE” data-file extensions. If you try to
open a data file created with CompleteEASE 6 on an earlier version of
CompleteEASE, you will get a checksum error when opening the file. Unfortunately,
there is no workaround for this. Analysis of data acquired with CompleteEASE 6 will
need to be performed on CompleteEASE 6.

Installing CompleteEASE (Windows)

Run the “CompleteEASE_Setup_6.xx.exe” application from the provided
installation USB drive. This program is located on the USB drive along with a folder
named “cnf files” that contains the configuration files to match one specific
ellipsometer system. The setup program will install the CompleteEASE program

CompleteEASE Software Manual Appendices • 14-523

 files, the Java Virtual Machine (JVM) files (which are required to run
CompleteEASE), and other related files (material optical constant files, model files,
example data files, etc.).

NOTE: The CompleteEASE_Setup program will not automatically install the

calibration and configuration files for your specific instrument to the “cnf”
directory. This is done at the factory, and if your files become corrupt, you will need
to copy all files from the “cnf files” folder on the USB drive that came with your
instrument into the corresponding “C:\CompleteEASE\cnf” folder after installation.
These files will return the instrument to the factory settings and replace any
calibrations that have been recently performed. Please contact J.A. Woollam for
further advice and directions if this is required in your situation.
Follow the on-screen prompts provided by the CompleteEASE_Setup program. The
default directory for installation is “C:\CompleteEASE\”. To minimize compatibility
problems, it is recommended to install CompleteEASE in this directory. This is
shown in Figure 14-1. The USB drive that came with your ellipsometer includes
configuration files for one specific ellipsometer system. If you are installing
CompleteEASE to run this ellipsometer system, choose “Full Installation” as shown
in Figure 14-2. This will install the drivers required for hardware operation. Even
when performing a “Full Installation”, it is still required to copy the configuration
files to the appropriate directory after installation, as discussed above.
CompleteEASE can be installed on multiple computers for data analysis purposes
(most systems include a five-seat license). If you are installing software on a
computer that will not operate an ellipsometer system, choose “Data Analysis
Only”, as shown in Figure 14-2.

NOTE: CompleteEASE is only compatible with the encoded experimental data files
measured using CompleteEASE. Thus, you can install CompleteEASE on multiple
computers, but it can only be used with data collected by CompleteEASE. Also, note
that data files acquired with CompleteEASE 5 and earlier versions can be opened on
any version of CompleteEASE, but data files acquired with CompleteEASE 6 can only
be opened with a version of CompleteEASE 6.

Figure 14-1. Default folder location for CompleteEASE installation.

14-524 • Appendices CompleteEASE Software Manual

 Figure 14-2. CompleteEASE Setup choices. CompleteEASE can be installed for use with an
ellipsometer system or for data analysis only.

Installing CompleteEASE (Mac)

CompleteEASE also has a Mac version that is available upon request. However, the
Mac version can only be used for data analysis; it cannot be used with hardware to
acquire data. To install the Mac version, copy the CompleteEASE installation disk-
image (.dmg) file to the desktop and double-click on it to mount and open it. (If the
window in Figure 14-3 does not appear, double-click on the “CompleteEASE” disk
image that is mounted on the desktop, which appears as in Figure 14-4.) Once the
disk-image window shown in Figure 14-3 opens, follow the on-screen directions by
dragging the CompleteEASE icon into the “Applications” folder. If there is already a
version of CompleteEASE installed on the computer, choose to replace the old
version when asked. Once CompleteEASE has been copied to the “Applications”
folder, close the CompleteEASE installation window shown in Figure 14-3 and eject
the mounted disk image on the desktop (Figure 14-4) by right-clicking (or control-
left-clicking) on the mounted disk image and choosing “Eject ‘CompleteEASE.’”

Figure 14-3. CompleteEASE disk-image setup screen for Mac.

CompleteEASE Software Manual Appendices • 14-525

 Figure 14-4. CompleteEASE mounted disk image.

To start CompleteEASE for the first time, open the “Finder” application and navigate
to the “Applications” folder, as shown in Figure 14-5. (When installing for the first
time, do not use “Launchpad” as it may block CompleteEASE for security reasons.)
Double-click on the CompleteEASE icon in the “Applications” folder to launch
CompleteEASE for the first time. (If you receive a security message that says that
CompleteEASE cannot be opened, as shown in Figure 14-6, then you will need to
right-click (or control-left-click) on the CompleteEASE icon and select “Open” from
the menu of options. Then, click the ‘Open’ button on the new security message
that appears as in Figure 14-7.)

Figure 14-5. CompleteEASE located in the “Applications” folder of the Mac Finder application.

Figure 14-6. Mac security warning about CompleteEASE.

Figure 14-7. Mac security warning about CompleteEASE that allows you to open it.

14-526 • Appendices CompleteEASE Software Manual

 After CompleteEASE has been launched for the first time, it will ask for
administrator credentials to install its security service, as shown in Figure 14-8.
Choose “Yes” and provide the credentials, as shown in Figure 14-9. This step is
necessary for CompleteEASE to run correctly, and it may take a couple minutes to
complete.

Figure 14-8. Installing the CompleteEASE security service.

Figure 14-9. Providing administrator credentials to install the CompleteEASE security service.

If you have not installed CompleteEASE before on this computer, CompleteEASE will
also ask you to create a special directory in which to store configuration files, model
and material library files, and example data files, as shown in Figure 14-10. Choose
‘Yes’ to continue. After this, CompleteEASE should then arrive at its activation
screen. This means that CompleteEASE has been installed successfully, and you can
now consult Appendix 14.2 for activation.

Figure 14-10. Creating the CompleteEASE user folder.

Operating Hardware
Hardware operation for data acquisition can only be done when running
CompleteEASE on the Microsoft Windows operating system. Hardware operation
also requires the configuration files for your instrument to be present in the
“C:\CompleteEASE\cnf” directory. These are provided with the installation USB
drive that accompanies each instrument. If you are upgrading to a newer version of
CompleteEASE on the computer that came with the instrument, the configuration
files should already be installed on that computer from the factory and do not
require any additional attention. When installing newer versions of CompleteEASE,
you can install the new version over the top of old version. The CompleteEASE
installer knows to preserve the configuration files. If you are installing

CompleteEASE Software Manual Appendices • 14-527

 CompleteEASE on a new computer to operate the hardware (or are repairing a
version that became corrupted), you must copy the contents of the
"C:\CompleteEASE\cnf " folder from the old computer (preferred if available) or
from the USB drive provided with the instrument to the "C:\CompleteEASE\cnf "
folder on the new computer’s hard drive.

NOTE: Do not copy configuration files from the USB drive that came with your
system if your system is already operating with CompleteEASE, as this will overwrite
your latest calibration and return it to factory settings, which may require the
recalibration of your instrument. Please contact J.A. Woollam for further details.

14.2. Activating CompleteEASE

Types of CompleteEASE Licenses
Perpetual License
The Perpetual License is the most common type of license being distributed with
J.A. Woollam ellipsometers. Each of these licenses will have a limited number of
seats. “Seat” is a term that means “one individual license” that can be used on one
individual computer. One activation key may contain multiple seats, meaning that
the same key can be used for multiple computers. For instance, an activation key
with a five-seat license can be installed on five computers and can be used by an
unlimited number of users as long as those users use one of the five computers.
Once CompleteEASE is activated on a computer, you should never need to
reactivate. CompleteEASE does not connect to the Internet after activation; in other
words, it will not contact or interact with the activation servers once it has already
been activated.

Timed License
The Timed License is a type of license distributed in special cases, such as at a
J.A. Woollam Short Course. A Timed License will expire after a specified period of
time once it is activated. After a Timed License is activated, the clock starts running
for how long it can be used. Timed Licenses cannot be deactivated and then
reactivated later; if you deactivate a Timed License, it is no longer usable. With a
Timed License, CompleteEASE will not contact the activation servers after it has
been activated. CompleteEASE keeps track of the time remaining on the Timed
License locally.

Online Activation
1. Verify you are logged in as a local administrator on your computer and are
connected to the Internet before installing CompleteEASE. Administrator
privileges are required to activate CompleteEASE. However, once
CompleteEASE is activated, administrator privileges should not be needed
to continue using CompleteEASE in most cases.
2. Start CompleteEASE from its shortcut (usually placed on the Desktop in
Windows or in the “Applications” folder in Mac). (If you are using Mac and
have not started CompleteEASE, you will now be prompted to install the
security service. Consult Appendix 14.1 for more details.)

14-528 • Appendices CompleteEASE Software Manual

 3. The CompleteEASE Activation window will appear as shown in Figure
14-11. Enter your product key with the dashes. (If you have the USB flash
drive that came with your ellipsometer, you should find a text file
containing your product key which you can then copy and paste into the
CompleteEASE Activation window.) The key is not case sensitive.

Figure 14-11. CompleteEASE Activation window.

4. Click the ‘Activate’ button. Please be patient as it may take a moment to

communicate with the activation server. You should then see a dialog box
confirming activation was successful as shown in Figure 14-12.

Figure 14-12. CompleteEASE Activation successful.

5. If you instead see an error message, then the activation was not successful.
If activation was not successful because CompleteEASE indicated that it
could not access the Internet, try the “Offline Activation” by clicking
‘Continue Activation Offline’. If the activation was not successful for other
reasons, consult “Activation Troubleshooting” below.

Offline Activation
1. Verify you are logged in as a local administrator on your computer before
installing CompleteEASE. Administrator privileges are required to activate
CompleteEASE. However, once CompleteEASE is activated, administrator
privileges should not be needed to continue using CompleteEASE in most
cases.
2. Start CompleteEASE from its shortcut (usually placed on the Desktop in
Windows or in the “Applications” folder in Mac). (If you are using Mac and
have not started CompleteEASE, you will now be prompted to install the
security service. Consult Appendix 14.1 for more details.)
3. The CompleteEASE Activation window will appear as shown in Figure
14-13. Click on the ‘Continue Activation Offline’ button.

CompleteEASE Software Manual Appendices • 14-529

 Figure 14-13. CompleteEASE Activation window.

4. You will then see the “Offline Activation” dialog box as shown in Figure
14-14. We will follow the steps listed in this dialog box. Click on the ‘Save
Computer C2V File’ button. The “.c2v” file is a unique identifier for your
specific computer. You will want to save this file to a USB drive or some
portable location so that you have access to the .c2v file while at a
computer with an Internet connection.

Figure 14-14. CompleteEASE offline activation dialog box.

14-530 • Appendices CompleteEASE Software Manual

 5. You then need to access the activation website. You may click the ‘Save
Link’ button, which saves the URL of the activation website that you will
use to upload the .c2v file. It is recommended to save the link in the same
location as the .c2v file so that you have access to it on the computer with
an internet connection.
6. Now, move to a computer with Internet access and open the “link” file that
was saved in the previous step or navigate to the following URL:
https://license.jawoollam.com/ems/customerLogin.html (This URL is case
sensitive). You should then see a screen similar to Figure 14-15.

Figure 14-15. Activation Website.

7. On the activation website, enter your product key with the dashes. (If you
have the USB flash drive that came with your ellipsometer, you should find
a text file containing your product key which you can copy and paste into
the CompleteEASE Activation website.) The key is not case sensitive. You
should now see a screen similar to Figure 14-16.

Figure 14-16. Select ‘Offline Activation’ on the activation website.

8. Click on the ‘Offline Activation’ button, as shown in Figure 14-16. You

should now see a screen similar to the one shown in Figure 14-17. First,
click on the ‘…’ button as indicated by the red “arrow #1”. You will be then
prompted to browse for the .c2v file you created earlier. Once you have
selected your .c2v file, click on the ‘Open’ button in the file dialog. Second,
click on the ‘Generate’ button as indicated by the red “arrow #2” in Figure
14-17. Be careful to only click on this button one time, or you may
accidently consume more than one of your available licenses.

CompleteEASE Software Manual Appendices • 14-531

 Figure 14-17. “Generate License” window on the activation website.

9. After the license is generated, you should now see a screen similar to the
one shown in Figure 14-18. Click on the orange ‘Download V2C File’ link as
indicated by the red arrow. You will then be prompted to save the .v2c file.
Again, it is recommended that you save this file to a portable location, such
as a USB drive, as you will need to move this file to the computer where
CompleteEASE is installed.

Figure 14-18. Click ‘Download V2C File’ to download the generated license.

10. You can now log out of the activation website and close your browser.
Transfer the .v2c file back to the computer where CompleteEASE is
installed.
11. After returning to the computer where CompleteEASE is installed, click on
the ‘Apply V2C License File’ button shown in Figure 14-19 and then browse

14-532 • Appendices CompleteEASE Software Manual

 to the .v2c file you created in the previous steps. Activation should
complete successfully at this point.

Figure 14-19. Choose ‘Apply V2C License File’ to complete offline activation.

12. If you instead see an error message, then the activation was not successful.
Consult “Activation Troubleshooting” below.

Deactivating CompleteEASE
Perpetual License Deactivation
It is possible to deactivate a perpetual license. The computer MUST be connected to
the Internet for deactivation. There is no option for deactivating offline. Once you
deactivate, you will be able to reactivate the same seat on another computer.
There is no limit on how many times you can activate/deactivate a license;
however, it is recommended that you use this feature sparingly as needed. It is not
meant for moving a CompleteEASE license between computers on a frequent basis.
To deactivate a perpetual license, navigate to the Options tab and select the
‘Deactivate’ button. If you launch CompleteEASE after your software has been
deactivated, you will once again be prompted for an activation key.

Timed License Deactivation

It is possible to deactivate a timed license; however, once you complete the
deactivation, you will not be allowed to reactivate it on any computer. There are
cases where you may want to deactivate a timed license; for example, if you were

CompleteEASE Software Manual Appendices • 14-533

 initially given a timed license but now you have a perpetual license, you will want to
deactivate the timed license before activating the perpetual license. If you launch
CompleteEASE after it has been deactivated, you will once again be prompted for
an activation key.

Activation Troubleshooting
Activation problems are commonly caused by not being logged in with appropriate
administrator privileges, internal firewalls, antivirus software, or network security
policies. If you are having issues with activation, please review the sections below. If
you are still experiencing activation issues after addressing these steps, feel free to
email [email protected] for support.

Online Activation Failed

The easiest way to solve most problems with the online activation is to continue
with the CompleteEASE Offline Activation. Please refer to the previous section for
detailed instructions. (If CompleteEASE says that is not connected to the Internet or
that it cannot access the Internet, even if you are connected to the Internet, it is
still probably easiest to try the Offline Activation. In this case, you could just
perform all offline activation steps on the same computer where CompleteEASE is
installed instead of using two different computers.)

Antivirus is Preventing Activation

You may want to consider temporarily disabling your antivirus software during the
activation of CompleteEASE. Most antivirus software will not interrupt the
activation process. If you are using a McAfee product, those are known to cause
issues with preventing online activation.

Internal Firewall is Preventing Activation

To test your internal firewall, proceed to the following URL:
https://license.jawoollam.com/ems/customerLogin.html
(Note: This URL is case sensitive.)
This server is used for both online and offline activations. Please contact your IT
personnel to make sure you have access to this domain during activation.

Proxy Servers and Network Policies

During online activation, CompleteEASE should inherit any proxy settings from your
network. If you believe there is an error because of proxy settings or a network
policy, it is recommended to continue with the offline activation instead.

No Administrator Privileges / Enforcement Type

Incompatible
When activating CompleteEASE, you may receive an error message that says that
you “do not have administrator privileges on this computer” or that says
“Enforcement type is incompatible”. This most likely means that you do not have
sufficient administrator privileges needed to place activation files in the appropriate
locations. Below are some diagnostic steps that can be attempted:

14-534 • Appendices CompleteEASE Software Manual

 1. You may want to consider temporarily disabling any antivirus software you
are running while trying to activate CompleteEASE and performing the
troubleshooting steps below. Most antivirus software will not interrupt the
activation process, but some antivirus software brands may cause
CompleteEASE to think it does not have appropriate administrator
privileges. Temporarily disabling antivirus software may resolve this issue
completely, but if not, it is still good to consider as a precaution before
following the steps below.
2. Even though you may think you have administrator privileges on this
computer, if you are seeing these errors, it is likely that you only have
some specific administrator privileges provided to you that are controlled
by a network group policy or other user-account management system. The
easiest way to attempt to mitigate these issues is to right-click on the
CompleteEASE icon and choose “Run as administrator” and then activate,
as shown in Figure 14-20. If it prompts you for credentials, it is best to find
a full account administrator (such as an IT specialist or systems
administrator) and have them log in to activate CompleteEASE.

Figure 14-20. Select ‘Run as Administrator” to activate CompleteEASE.

3. If CompleteEASE did not activate successfully using “Run as Administrator”,

there may still be issues with the level of user administrator privileges. You
can check specific folders to see if you have the proper access to these
folders in order for the activation to work correctly. You need to verify that
you have write access to both the “C:\Windows\system32\” folder and
also to the “C:\Program Files (x86)\Common Files\Aladdin Shared\HASP\”
folder. To do this, attempt to save a simple text file (e.g. created using
Notepad) in those directories and see if Windows allows you to write to
those directories. That should demonstrate that your administrator
privileges are sufficient for CompleteEASE to write to those folders. If you
are not allowed to write to these folders, then there is some issue with
administrator privileges that needs to be resolved before CompleteEASE
can be activated. You will need to find a user with the appropriate set of
elevated credentials allowing them to write to these folders.
4. If you have write access to the folders in the previous step and activation
still continues to fail despite starting CompleteEASE with “Run as
Administrator” or if you were blocked from writing to those folders but
have now resolved that issue, then you may need to reinstall
CompleteEASE before trying to activate CompleteEASE again. When
reinstalling CompleteEASE, make sure to right-click on the
CompleteEASE_Setup.exe file and select “Run as Administrator” to run the
installation as an administrator. (It may be necessary to reinstall because

CompleteEASE Software Manual Appendices • 14-535

 some of the drivers may not have been properly installed if the computer
did not have the appropriate administrator privileges to write to the
folders mentioned in the previous step.)
5. If activation problems still continue to persist, then other specific actions
may need to be taken. In order to assist us in helping you diagnose and
resolve these issues, you will need to provide us with a few additional files.
On the CompleteEASE activation screen, click on ‘Continue Activation
Offline’, then click ‘Save Computer C2V File’, and save it to a convenient
location (such as the Desktop). Then, open an Internet browser and
navigate to http://localhost:1947/. In the menu of options on the left, click
on "Diagnostics" and then find a small button off to the far right that says
"Create ID File." Click that to save the file to a convenient location. Also,
while there, click on the "About" option and note the version number of
the run-time installer. Please email these files and runtime version
information to [email protected].

No Internet Connection
To resolve errors related to the Internet connection, generally the fastest and
easiest approach is to simply try the offline activation. However, there are a few
additional diagnostic steps that can be attempted.
1. Make sure your computer is connected to the Internet by opening a web
browser, such as Chrome, Safari, Firefox, or Edge, and navigate to any
website, such as http://www.jawoollam.com/. If you cannot do this
successfully, then it is likely that your computer has no Internet connection
at all and that this is the cause of the activation issues. Either make
attempts to connect your computer to the Internet or follow the steps for
Offline Activation by clicking on the ‘Continue Activation Offline’ button
and using another computer for the steps that require an Internet
connection.
2. If you were able to access typical websites in a browser from the previous
step, that means your computer is connected to the Internet, but
something else is blocking CompleteEASE from accessing the Internet to
perform activation. This can be due to firewall settings, a strict network
group policy, proxy server settings, antivirus software, website access
restrictions, etc. CompleteEASE is configured to use your computer’s proxy
settings, but this doesn’t always guarantee that it can successfully reach
the Internet. Check to see if you can access the specific activation website
from a web browser by navigating to the following URL:
https://license.jawoollam.com/ems/customerLogin.html. If you are
greeted by the website shown in Figure 14-15, then your computer can
successfully access the activation website. If you do not see that website,
then your activation issues are probably caused by certain website
restrictions or proxy-server settings preventing your computer from
accessing the activation website. You will need to contact your IT support
to figure out how to allow access to the activation website
(https://license.jawoollam.com). If this is not possible, you will have to
follow the steps for Offline Activation.
3. If your computer is able to access the activation website in the previous
step from a web browser but CompleteEASE still says that it cannot access
the Internet, this means that something is preventing CompleteEASE from

14-536 • Appendices CompleteEASE Software Manual

 contacting this website. This may be firewall settings, an antivirus program,
or a network group policy. In this case, try to eliminate as many of these
potential factors as possible temporarily during activation (e.g. turn off
firewall, disable antivirus, etc.). If the activation still doesn’t work at this
point, rather than diagnosing complex Internet connection issues, it’s
probably easier to follow the Offline Activation process.

14.3. General Troubleshooting

If you’re having problems with troubleshooting activation or deactivation, please
see the “Activation Troubleshooting” part of Appendix 14.2.
Please contact your J.A. Woollam representative with any questions relating to
CompleteEASE or your ellipsometer system. To reduce the support time, please
create and email the debug file described in the following section before contacting
your representative.

NOTE: Your closest J.A. Woollam Representative can be found on our website by
visiting www.jawoollam.com/contact-us/worldwide-representatives/.

Reporting Problems
If you have a problem with CompleteEASE or your hardware, please send a debug
file to J.A. Woollam or your local representative. The debug file contains your
current configuration, log files, and recent calibration data. You can email this file to
us along with a description of your problem and contact information.

Creating a Debug file

1. Click on the Options tab.
2. Click on the ‘Create Debug File’ button. This will create and save the debug file
on your computer.
3. The file path for the debug file is “C:\CompleteEASE\CompleteEASE_Debug.zip”

NOTE: The Debug File packages all hardware and software configuration files along
with the current Error Log into a single .zip file.
Email this file to J.A. Woollam or your representative to aid in debugging problems
with your ellipsometer system.

14.4. Increase CompleteEASE Memory Usage

CompleteEASE has limited memory that it can use because all of its memory is
allocated when it starts up and cannot be changed unless CompleteEASE is
restarted. CompleteEASE will reserve up to 1 GB of memory (RAM) by default when
it starts, depending on how much is available on the computer. To increase memory
beyond 1 GB, first create a shortcut to the CompleteEASE program. (One way to do
this is by right-clicking on “CompleteEASE.exe” and selecting “Send To” and then
“Desktop (Create Shortcut)”). Once the shortcut is created, right-click on it and
open its “Properties” dialog box. Add “ -memXXXX” after the executable name in
the “Target” field, as shown in Figure 14-21. The XXXX should be replaced with the

CompleteEASE Software Manual Appendices • 14-537

 number of MB of memory desired. In Figure 14-21, “-mem3500” tells
CompleteEASE to obtain roughly 3.5 GB (3,500 MB) of memory when it starts. If
there is not enough memory to do this, then CompleteEASE will not start. On a
standard 32-bit Windows computer, we do not recommend setting this higher than
1500 MB. On a standard 64-bit Windows computer, we recommend no more than
½ of the available memory (RAM).

Figure 14-21. Add “ -memXXXX” to the “Target” field to increase maximum memory usage
for CompleteEASE.

14.5. Network Card Configuration

For some J.A. Woollam ellipsometers, such as the M-2000, the functions of the
ellipsometer are controlled through the network card (ethernet port) on the back of
the computer. Because of this, acquiring data with one of these ellipsometers using
CompleteEASE requires the network card in your computer to be configured
correctly. Below, you will find a step-by-step process for how to configure a
computer’s network card to take be able to run a J.A. Woollam ellipsometer and
take data with CompleteEASE.
1. It is recommended that the ellipsometer computer be independent of any
existing network. However, if it is necessary for the computer to be
connected to an existing network, two separate network cards are needed.
One network card is needed to communicate with the external network,
and the other network card is needed to communicate with the
ellipsometer. It is recommended to use the available motherboard-
installed network card (the one that came with the computer) for the
ellipsometer connection and install the second network card for any
external network connections.

14-538 • Appendices CompleteEASE Software Manual

 2. Before installing the new network card, go to the Start Menu and find the
Control Panel. (This can be found by simply typing “Control Panel” in the
Windows search bar or while the Start Menu is open. Note that, on
Windows 10 computers, the Control Panel is different than the Settings
application.) Under the “Network and Internet” category, select “View
network status and tasks”. (If your Control Panel is not organized by
categories, then select “Network and Sharing Center”.)

Figure 14-22. Select “View network status and tasks” in the Control Panel.
3. Select “Change adapter settings” from the panel on the left side of the
screen.

Figure 14-23. Select “Change adapter settings” on the left side.

4. Take note of the current network configuration. In this example, there is a
single network interface card installed that is currently connected. In the
following steps, we will add another interface to this list.

Figure 14-24. Only one network connection exists.

CompleteEASE Software Manual Appendices • 14-539

 5. Shut down the computer and install a new network card.
6. Upon restarting, Windows should automatically install the driver for the
new network card. If it does not, consult the software that was bundled
with the new card for manual driver installation instructions.
7. Now, repeat steps 1-3. You should see that there is now an additional
network card displayed. In this example, the new network card is
disconnected with the name “Ethernet 2”. Double-click the network card
icon for the network card which will be used to run the ellipsometer to
open up its “Properties” dialog box.

Figure 14-25. Now, two network connections exist.

8. Disable all connections for this network card except “Internet Protocol
Version 4 (TCP/IPv4)” by unchecking the boxes. Then select “Internet
Protocol Version 4 (TCP/IPv4)” and click the “Properties” button.

Figure 14-26. Select only “Internet Protocol Version 4” from the network card
properties dialog box.
9. The TCP/IPv4 properties dialog box is displayed. Select “Use the following
IP address” and enter “192.168.0.1” for the “IP address” and
“255.255.255.0” for the “Subnet mask” and then select “OK”.

14-540 • Appendices CompleteEASE Software Manual

 Figure 14-27. “Internet Protocol Version 4 Properties” dialog box with correct
settings to run a J.A. Woollam ellipsometer using CompleteEASE.
10. The newly installed network card now provides the instrument computer
with capabilities of joining a network while remaining connected to the
ellipsometer. It is recommended to use the higher performance Network
Card for the ellipsometer (usually the one attached to the motherboard
that came with the computer) for the fastest and most reliable connection.
11. Start CompleteEASE to initialize the hardware and confirm the new
installation was successful. (CompleteEASE should begin initializing the
ellipsometer hardware instead of just waiting with the “Connecting to…”
message.)

14.6. CompleteEASE Data Transfer (to/from WVASE)

The data files in CompleteEASE and WVASE are not directly compatible because
they have different file formats. The WVASE files are written in ASCII text while the
CompleteEASE files use an encrypted format. If you have a copy of WVASE and the
software key (activator USB) that accompanies it, you can import and export data
files between WVASE or CompleteEASE. The “Experimental Data” window in WVASE
is used to execute the data transfer of files already saved.

NOTE: Any data open in the Experimental Data window will be removed (deleted if
not saved) by performing the following steps.

Importing CompleteEASE Data into WVASE

1. From WVASE, select “CompleteEASE Data Transfer” from the File menu of
the Experimental Data window and then select “Import Data from
CompleteEASE”.

CompleteEASE Software Manual Appendices • 14-541

 Figure 14-28. Choose the “CompleteEASE Data Transfer” menu from the “File”
menu of the “Experimental Data” window and then select “Import Data from
CompleteEASE”.

2. Browse to select data files you wish to import from CompleteEASE.

Multiple files can be selected.

Figure 14-29. Select files you wish to Import.

3. After the CompleteEASE data files are selected, browse to select the
directory where you wish the new WVASE files to be saved. Imported files
will be saved in the WVASE format in the directory chosen.

Figure 14-30. Choose the location to save imported data files.

14-542 • Appendices CompleteEASE Software Manual

 4. After the files are imported, a message will appear, as in Figure 14-31.
Previous “.SE” and “.iSE” files are now saved in WVASE “.dat” format.

Figure 14-31. Message after all files imported correctly.

Exporting WVASE Data to CompleteEASE

1. From the Experimental Data window in WVASE, select the File menu and
choose “CompleteEASE Data Transfer”. Then, select the “Export Data to
CompleteEASE” option as shown in Figure 14-32.

Figure 14-32. Choose the “CompleteEASE Data Transfer” menu from the “File”
menu of the “Experimental Data” window and then select “Export Data to
CompleteEASE”.

2. Browse and select the WVASE data files that you want to export to
CompleteEASE, as in Figure 14-33. Multiple files can be selected.

Figure 14-33. Choose WVASE data files you wish to export.

CompleteEASE Software Manual Appendices • 14-543

 3. Browse to select the directory for data exported to CompleteEASE, as
shown in Figure 14-34. Exported files will be saved in a CompleteEASE
format in the directory chosen.

Figure 14-34. Browse for location to place the newly created CompleteEASE files.

4. After the files are exported, the message shown in Figure 14-35 will
appear. Previous “.dat” and “.bdt” files are now saved in CompleteEASE
“.SE” and “.iSE” formats.

Figure 14-35. Message when all data files are exported. If files were unable to
export, the total number of successful files is listed.

14.7. Comparing CompleteEASE and WVASE Results

Users that are familiar with using WVASE may experience a longer learning curve as
they discover the features of CompleteEASE. When comparing results between data
analysis performed within WVASE and CompleteEASE, please consider the following
differences:

Fit Weighting
Note the difference in Fit Weighting during data analysis in both software packages.
The WVASE default is “Experimental Standard Deviations” while CompleteEASE uses
“NCS” for its default fit weighting.

Depolarization Data
If the data contain depolarization, WVASE will automatically include this in the MSE
calculation and fit. CompleteEASE will not use the depolarization data during MSE
calculation and fit unless you turn ON “Include Depolarization Data” in the hidden
options of the +FIT Options menu in the model (see Section 9.7).

14-544 • Appendices CompleteEASE Software Manual

 Intensity Data
If data are collected with CompleteEASE, intensity data is also recorded (but not
used in the fit unless “Include Intensity Data” in the hidden options of the +FIT
Options menu in the model is turned ON – see Section 9.7). If this CompleteEASE
data is imported into WVASE, it will also import the intensity data, and WVASE will
then automatically fit this data. Therefore, the intensity data needs to be removed
from the WVASE experimental data (using Range Select dialog) in order to match
the CompleteEASE standard fit (that excludes intensity data by default) with the
WVASE fit (that includes intensity data by default).

Surface Roughness
In CompleteEASE, surface roughness is an independent thickness quantity with half
of its thickness extending down into the top layer and the other half of its thickness
extending up above the top layer. In the actual model calculation (which is created
in the background and not visible to users), an EMA layer is added above the top
layer while the top layer’s thickness is reduced by half of the surface roughness
thickness. Thus, if your top layer shows that it is 100-nm thick with 2 nm of surface
roughness, the actual model calculation will use a 99-nm thick layer with a 2-nm
EMA layer on top to represent the surface roughness. In WVASE, surface roughness
is simply an EMA layer over the top layer thickness; it does not “take away” from
the top layer. If your layer in WVASE shows 100 nm of thickness with 2 nm of
surface roughness, this is the actual model calculation.

14.8. CompleteEASE Remote Commands

CompleteEASE can communicate with an external computer, allowing some
operations to be automated. These operations include controlling data acquisition
for in-situ and ex-situ measurements as well as adding data from external sensors
during in-situ measurements so that the values can be displayed along with the
ellipsometry data. For this purpose, CompleteEASE contains an ASCII-based remote
communication protocol that allows external systems to send ASCII command
strings to CompleteEASE through a TCP/IP or RS-232 (serial) port to control various
aspects of the acquisition and analysis processes. In computer systems terminology,
CompleteEASE would be the “server” and the external system would be the
“client”. All ASCII commands that are sent to CompleteEASE by the external system
must be terminated with the carriage-return/line-feed sequence (the CR and LF
characters, which are ASCII characters 13 and 10, usually coded as “\r\n”). Likewise,
all responses from CompleteEASE will be terminated with these characters with the
exception of the “ReportParameters()” command, which returns a formatted text
string with potentially multiple CR/LF sequences in it, so it is terminated with “XXX”
and the CR/LF sequence on the last line. If an error occurs while processing a
command, CompleteEASE will return “Error:” followed by a description of the error.

Configuring Communication Settings

Remote communication settings are turned on in CompleteEASE through the
“CompleteEASE Configuration” dialog box accessible from the Options tab. To
display the dialog box, press the ‘Edit Configuration’ button on the Configuration
Controls panel. The “Show Advanced Configuration Options” checkbox will need to
be checked to gain access the “Remote Communications” section of the

CompleteEASE Software Manual Appendices • 14-545

 configuration parameters. Figure 14-36 below shows the dialog with the “Remote
Communications” section visible.

Figure 14-36. “Remote Communications” section of the CompleteEASE configuration.

The “Communication Type” parameter sets the communication method. The

options are “None” (the default), “TCP/IP”, and “RS-232”. If the parameter is set to
“TCP/IP”, then the “TCP/IP Port #” parameter determines the port that
CompleteEASE will communicate on. If the communication type is set to “RS-232”,
then the “Baud Rate” parameter sets the baud rate for the serial communication.
CompleteEASE assumes the other serial port settings are 8 data bits, 1 stop bit, and
no parity. The other parameters in the “Remote Communication” section are
related to TCP/IP communication when more than one connection is used.

In-Situ Acquisition Modes

CompleteEASE in-situ data acquisition can be run in several modes, from a simple
start-acquisition/end-acquisition mode to a variety of triggered modes. Some of the
commands in “Remote Commands” behave differently depending on which mode
you are operating in. The different modes available are described below. Each
acquisition mode has special configuration parameters that affect it (which are
found in the “Data Acquisition” section of the Hardware Configuration, found by
clicking the ‘Edit Hardware Configuration’ button in the Hardware tab). Basic
descriptions of these configurations can be found in the Hardware Configuration
dialog box. For more specific descriptions of all the options available, consult the in-
situ-specific Hardware Manual for your ellipsometer.

Standard Mode
Standard mode is the most common measurement mode. When the data
acquisition is started, the ellipsometer will acquire data until a pause command is
received or the acquisition is ended. Commands can be sent to set models and
retrieve hardware and fit parameters during acquisition.

Triggered Mode
Triggered mode is typically used when process monitoring and control is to be
automated. Commands are sent to CompleteEASE, CompleteEASE performs the
specified task, and then the acquisition pauses, waiting for the next command. For

14-546 • Appendices CompleteEASE Software Manual

 example, sending the command “Trigger()” tells CompleteEASE to acquire one data
point. Sending the command “SetModel(Model2,targetthick=200)” tells
CompleteEASE to make “Model 2” the active model and to acquire data until the
target thickness of 200Å is reached. When the target thickness is reached,
CompleteEASE sends a response and then pauses until the next command is
received. CompleteEASE can be told to acquire a specified number of points,
acquire data for a specified amount of time, or to acquire data until a specified
thickness is reached.

Intensity Trigger Mode

Intensity Trigger mode is used when only a single model is needed for monitoring or
control. In this mode, CompleteEASE starts acquiring data when the light intensity
reaches a specified threshold (indicating that a sample was placed in the beam
path) and stops acquiring data when the intensity drops back below the threshold.

Auto Trigger Mode

Auto Trigger mode is used when the sample moves out and then back into the
beam path during acquisition. It works just like Standard mode except that data is
only acquired when the intensity is above the “Minimum Intensity Threshold”
configuration parameter. Unlike Intensity Trigger mode, the data acquisition is not
ended when the intensity falls below the threshold; it is just paused until the
intensity is back above the threshold.

Aux. Port Trigger Mode

Aux. Port Trigger mode is used when data acquisition needs to be synchronized
with a specific signal. This mode works just like Auto Trigger mode except that data
is only acquired when the signal on Aux. Input Port #0 is high. (For more
information about using the Aux. Port for thickness monitoring and control, consult
the in-situ-specific Hardware Manual for your ellipsometer.)

One-Point Trigger Mode

One-Point Trigger mode is used for applications where only one data point needs to
be acquired when the sample becomes visible to the ellipsometer. An example of
this is an ALD deposition experiment where a shutter allows the ellipsometer to see
the sample after a number of deposition cycles. In this case, the ellipsometer will
acquire one data point each time the shutter is opened.

Remote Commands
In order to facilitate communication with external systems, CompleteEASE has a set
of commands that allow the remote system to communicate with CompleteEASE
and control how it operates. Some types of commands allow the external system to
query CompleteEASE for state information and parameter values (called
“Information Query Commands”) and others allow the external system to control
various aspects of the acquisition and analysis of data (called “Measurement
Control Commands”), such as setting models and target thicknesses, starting and
ending data acquisition, switching between acquisition modes, etc. Some

CompleteEASE Software Manual Appendices • 14-547

 commands relate to in-situ data acquisition and others relate to recipe-driven
ex-situ data acquisition. The commands for in-situ data acquisition are given in

Table 14-2 while those for ex-situ data acquisition are given in Table 14-3.
Additionally, there are some specific hardware commands that are only applicable
for certain ellipsometer systems; these are shown in Table 14-4. Unless a command
is supposed to return a specific value or string of text information, the command
will respond with “OK” if it has completed successfully or will return an error
message beginning with “Error” if it could not complete successfully.

Table 14-2. List of the remote commands available in CompleteEASE to be used by an

external system configured for in-situ measurement.

COMMAND DESCRIPTION
MEASUREMENT CONTROL
StartAcq(file_name) Starts in-situ data acquisition which will save the
data to the file specified in the file_name parameter.
StopAcq() Stops in-situ data acquisition and ends the
measurement process.
PauseAcq() Pauses in-situ data acquisition temporarily but does
not end the measurement process.
ResumeAcq() Resumes in-situ data acquisition after it has been
paused.
Trigger() Triggers an in-situ data measurement when in
Triggered Mode; in other words, performs a
measurement of a single data point.
StdAcq() Sets the acquisition time to “Standard Acquisition”,
which uses the normal acquisition time specified in
the Hardware Configuration. (This is equivalent to
unchecking the “Fast Acquisition Mode” checkbox
on the In situ tab.)
FastAcq() Sets the acquisition time to “Fast Acquisition”, which
uses the faster acquisition time specified in the
Hardware Configuration. (This is equivalent to
checking the “Fast Acquisition Mode” checkbox on
the In situ tab.)
DCOffset() Performs a DC offset calibration.
GetAlign() Performs a one-point alignment and returns the
signal intensity and the X and Y alignment values. An
error message is returned if the alignment cannot be
performed. (Ellipsometers without alignment
detectors only return signal intensity.)
GetAlign2() Performs the same function as “GetAlign()” but in
addition returns more specific intensity values, such
as the UV and NIR intensities separately.

14-548 • Appendices CompleteEASE Software Manual

 SetAcqMode(acq_mode) Sets the acquisition mode for in-situ measurements.
This changes the “Acquisition Mode” hardware
configuration parameter in the “Data Acquisition”
section. The argument acq_mode represents a string
with the name of the acquisition mode (e.g.
“Standard” or “Trigger”). These modes are described
in the section above entitled “In-Situ Acquisition
Modes”.
SetModel(model_name, Sets the current in-situ model to one that is on the
acq_setting) In situ tab. The model_name can be either the name
of the model or the word “MODEL” with its
associated position in the model list visible on the In
situ tab (e.g. “MODEL 3”). When in “Trigger” mode,
this command also starts data acquisition based on
the acq_setting parameter. The choices for
acq_setting are “targetthick=”, “pts=”, or “time=”.
(See the description below this table for more
information.)
SetTarget(float_value) Changes the target thickness for the measurement
currently being run. The target thickness is specified
by float_value as a floating-point value in angstroms.
The target thickness should first be set by using the
acq_setting argument of the
“SetModel(model_name, acq_setting)” command.
AddParm(parameter_name, Adds a custom parameter from the external system
value) to the in-situ data file that CompleteEASE acquires.
This custom parameter is added as a dynamic
parameter with the name parameter_name and is
given the specified value. The parameter is
timestamped with the time at which the previous
measurement point was acquired. Sending this
command multiple times with the same
parameter_name will dynamically record the values
of the parameter over time along with the in-situ
measurement data as provided by the external
system.
AddLog(text_string) Adds a text_string to the in-situ measurement log
that is timestamped in the log at the time the
command is sent.
INFORMATION QUERY: Hardware
Ready() Returns “OK” if the system is initialized, calibrated,
and ready to collect data. Returns “Error: Hardware
is not ready” if not.
Status() Returns the status string, which specifies what the
system is currently doing. When CompleteEASE is
ready for the next data-acquisition command, this
function returns “Waiting To Acquire Data”.
INFORMATION QUERY: Parameters
Get(parameter) Returns the most recent value of the parameter with
the parameter name specified by parameter.

CompleteEASE Software Manual Appendices • 14-549

 GetRT(parameter) Returns the “real-time” (RT) value of the parameter
with the parameter name specified by parameter.
The real-time value is the parameter’s most recent
value extrapolated to the time that this command
was sent.
SetRTParms(pts_to_fit) Sets the number of points used to fit the trendline to
determine the extrapolated “real-time” (RT) value
returned with the “GetRT()” command.
GetSmooth(parameter) Returns the “smoothed” value of the parameter
with the parameter name specified by parameter.
The smoothed value is the parameter’s value
averaged over the last x data points acquired for the
parameter.
SetSmoothParms(pts_to_avg) Sets the number of points used in calculating the
average of the most recent points to determine the
“smooth” value returned with the “GetSmooth()”
command.
GetInit(parameter) Returns the initial value of the parameter with the
parameter name specified by parameter. The initial
value is the value of the parameter at the start of an
in-situ measurement.
GetFinal(parameter) Returns the “final” value of the parameter with the
parameter name specified by parameter. The final
value is the parameter’s final value x number of
points away from the end of the in-situ
measurement and then averaging over y data points
at that position. (The x and y values are set with the
“SetFinalParms() command.)
SetFinalParms(pts_from_end, Sets both the number of points from the end of the
pts_to_avg) measurement and the number of points before that
to average over when determining the “final” value
of the parameter returned with the “GetFinal()”
command.
GetAll() Returns a list of all parameters that appear in the In
situ tab status panel.
ReportParameters() Returns a formatted string with the names and
values of the report parameters. There may be
several carriage-return/line-feed (CR/LF) character
combinations in this return string, so it is terminated
with “XXX+CR+LF” instead of the standard “CR+LF”.
During thickness control, the parameter report may
not be ready immediately after the end-point signal
is sent, so this command may need to be polled until
the parameter-report string is returned.
INFORMATION QUERY: Models
GetCurrentModel() Returns the name of the currently selected in-situ
model in the list visible on the In situ tab.
GetModels() Returns a list of all in-situ models available on the
list visible on the In situ tab.

14-550 • Appendices CompleteEASE Software Manual

 SetModel(model_name) Sets the current in-situ model to one that is on the
In situ tab. The model_name can be either the name
of the model or the word “MODEL” with its
associated position in the model list visible on the
In situ tab (e.g. “MODEL 3”). Sending “none” for -
model_name will clear the current model. (This
command can also be used to start data acquisition;
see the explanation below this table for more
information.)
GetCommonModels() Returns a list of the names of all models present in
the COMMON folder.
SetCommonModel(file_name) Opens a model from the COMMON folder in
CompleteEASE.
GetModelParms() Returns a list of the names of all fit parameters
present in the current model along with whether
they are controllable (e.g. with a target thickness)
and whether they are hidden from results reports.

SetModel(model_name, acq_setting) Command: The “SetModel()” command

behaves differently depending on how the “Acquisition Mode” parameter (in the
“Data Acquisition” section of the CompleteEASE hardware configuration) is
configured. If the acquisition mode is set to “Standard”, then the command is
“SetModel(model_name)” which takes only one argument, model_name, and sets
up the corresponding model. The model_name can be either the name of the model
or the word “MODEL” with its associated position in the model list visible on the
In situ tab (e.g. “MODEL 3”). Sending “none” for model_name will clear the current
model from the Analysis tab; this is equivalent to having no model selected in the
In situ tab. If the acquisition mode is set to “Trigger”, then the command is
“SetModel(model_name, acq_setting)” which can take an additional argument that
tells the ellipsometer how to acquire the data for this particular model. The second
argument can be “targetthick=#”, “time=#”, or “pts=#”, where the “targetthick”
value is given in angstroms, the “time” value is given in seconds, and the “pts”
(points) value is given as a number of data points to acquire. If the “targetthick”
option is used, CompleteEASE will return “OK” immediately after receiving the
command and will acquire data until the specified thickness has been reached.
CompleteEASE will notify the external system when this has occurred. If the “time”
or “pts” option is used, CompleteEASE will not return until after it has acquired the
data (based on the time or number of data points) and determined if all of the
parameters are within specification. If the measured parameters are within
specification, “OK” is returned; otherwise, an error message is returned.

Table 14-3. List of the remote commands available in CompleteEASE to be used by an

external system configured for ex-situ (recipe) measurement.

COMMAND DESCRIPTION
MEASUREMENT CONTROL
LoadRecipe(recipe_name) Loads the recipe indicated by recipe_name into
CompleteEASE. This recipe file must be located in
the COMMON Recipe folder (usually
C:\CompleteEASE\recipe).

CompleteEASE Software Manual Appendices • 14-551

 RunRecipe(recipe_name) Runs the recipe indicated by recipe_name. This
recipe file must be located in the COMMON Recipe
folder (usually C:\CompleteEASE\recipe). This
command will return a text string with the results
when finished. (Note that this cannot be canceled
after it is started nor can any other commands be
sent until it is finished.)
RunRecipe(recipe_name/file_ Runs the recipe indicated by recipe_name and saves
name) the data acquired with the file name file_name. The
recipe_name and file_name fields are separated
with a forward slash character “/”. The recipe file
must be located in the COMMON Recipe folder
(usually C:\CompleteEASE\recipe). This command
will return a text string with the results when
finished. (Note that this cannot be canceled after it
is started nor can any other commands be sent until
it is finished.)
RunRecipeAsync(recipe_name) Runs the recipe indicated by recipe_name
asynchronously. This recipe file must be located in
the COMMON Recipe folder (usually
C:\CompleteEASE\recipe). Running a recipe
asynchronously allows the recipe measurement to
be canceled. This command will return “OK” if it
starts the recipe measurement successfully, but it
will NOT return the results when finished. The
“Status()” command must be polled to know when
the recipe is finished.
RunRecipeAsync(recipe_name/ Runs the recipe indicated by recipe_name
file_name) asynchronously and saves the data acquired with
the file name file_name. The recipe_name and
file_name fields are separated with a forward slash
character “/”. This recipe file must be located in the
COMMON Recipe folder (usually
C:\CompleteEASE\recipe). Running a recipe
asynchronously allows the recipe measurement to
be canceled. This command will return “OK” if it
starts the recipe measurement successfully, but it
will NOT return the results when finished. The
“Status()” command must be polled to know when
the recipe is finished.
CancelRecipeAsync() Cancels the current recipe measurement only if the
current recipe measurement was started with the
“RunRecipeAsync()” command. Otherwise, the
recipe cannot be interrupted until it returns a result
indicating that it is finished.
INFORMATION QUERY
Status() Returns the status string, which specifies what the
system is currently doing. When CompleteEASE is
ready for the next data-acquisition command, this
function returns “Waiting To Acquire Data”.

14-552 • Appendices CompleteEASE Software Manual

 InitStatus() Returns more specific details about the hardware
initialization process than “Status()” does and will
give more specific error messages. After initialization
is complete, it returns either “Hardware Initialized”
or “Hardware Not Initialized”. If the latter, it will be
followed by “-Errors:” plus the errors encountered
during hardware initialization. (No errors present
mean no “hardware.cnf” file was found or a user
canceled initialization manually.)
ListRecipes() Returns a list of recipe files that are located in the
COMMON Recipe folder (usually
C:\CompleteEASE\recipe). The file names are
separated by a forward slash character “/”.
GetRecipeInfo(recipe_name) Returns the names of the acquisition parameters file
(.parms), the scan pattern file if present (.scan), and
the model file (.mod) associated with the recipe
indicated by recipe_name. The file names are
separated by a forward slash character “/”.

Table 14-4. List of the remote commands available in CompleteEASE to be used by an

external system for control of specific hardware elements, which are only available on specific
ellipsometer systems. If the command is not available on your ellipsometer, “Sensor Not
Available” or another type of error may be returned.

COMMAND DESCRIPTION
SPECIALIZED HARDWARE CONTROL
SystemLampOn() Turns the ellipsometer light source on. (It may take
up to 30 seconds for the lamp to turn on after
sending this command.)
SystemLampOff() Turns the ellipsometer light source off.
SystemLampIsOn() Returns “Yes” if current to the light source is
detected, indicating it is on. Returns “No” if no
current to the light source is detected, indicating it is
off. If the ellipsometer does not have a lamp current
sensor, then ‘Sensor Not Available’ is returned.
SystemLampShutterOpen() Opens the lamp shutter if it is available or moves the
filter wheel to its previous position before
“SystemLampShutterClosed()” was last sent.
SystemLampShutterClosed() Closes the lamp shutter if it is available or moves the
filter wheel to the blocked position. (Note the
misnomer with this command’s name. This
command does NOT detect if the shutter is closed; it
actually closes the shutter.)
SystemYellowFilterOn() Moves the filter wheel to the yellow filter position if
it is available.
SystemYellowFilterOff() Moves the filter wheel to its previous position
before “SystemYellowFilterOn()” was last sent.
SystemMoveToSampleLoad() Moves the ellipsometer stage to the sample load
position. (Sometimes, if the system is already at the
sample load position, this command may move the
stage back to the center position.)

CompleteEASE Software Manual Appendices • 14-553

 SystemGetCurrentAngle() Returns the current angle of incidence in which the
ellipsometer is positioned as a floating-point
number.

Asynchronous Real-Time Reporting

There are two ways that fit-parameter values can be retrieved in real-time after a
data point is measured and analyzed. The first method is to use one of the query
commands (i.e. “Get()”, “GetRT()”, etc.). This method is useful if only one or two
parameters are of interest. The second method is to use the asynchronous
reporting functionality of CompleteEASE, where CompleteEASE will automatically
send out a string with parameter names and values after a data point is measured
and analyzed. To turn on asynchronous reporting, the “Auto Report” configuration
setting must be turned on, which can be found in the “Remote Communications”
section (must have “Show Advanced Configuration Options” checked) of the
CompleteEASE Configuration found by clicking ‘Edit Configuration’ on the Options
tab (see Figure 14-36). When this is turned on, every time a data point is measured
and analyzed, CompleteEASE will send out the fit results asynchronously on a port
that is one above the port for sending commands. (By default, the port for sending
commands is Port 4444; thus, by default the asynchronous reporting port is Port
4445.)
The “Auto Report Format” configuration option (found in the same place as the
“Auto Report” option) allows the user to determine the format of the report.
Setting the format to “Status Text” will send the string of values that appears in the
In situ tab status box. Setting the format to “All Parms” will send all of the
parameters, including hidden parameters, each time a data point is analyzed. You
can use the “Comm Test” program described below to view these different formats
and choose the most appropriate one.

Parameter Value Query

CompleteEASE can be configured to ask an external system for the starting value of
a specified model fit parameter before performing a fit. To turn on the ability for
CompleteEASE to query for parameter values from a remote source, turn on the
“Allow Parameter Query” configuration option in the “Remote Communications”
section (must have “Show Advanced Configuration Options” checked) of the
CompleteEASE Configuration found by clicking ‘Edit Configuration’ on the Options
tab (see Figure 14-36). Turning this option on will enable CompleteEASE to ask a
remote system for a fit parameter’s value on a port two positions above the one
specified with the “TCP/IP Port #” configuration setting. (By default, the main port
specified is 4444, meaning that in the default case, the remote parameter querying
will happen on port 4446.) In order to set the fit parameters in the model for which
CompleteEASE will query starting values, click on the fit parameter to view the “Edit
Fit Parameter” dialog box, check the “Show Advanced Parameter Settings”
checkbox, and then check the “Ask Remote Computer For Value” checkbox under
“Query Parameter” (see Section 9.5). Once this is established, this parameter’s
initial value will be queried from a remote computer (on port 4446 by default)
before performing a fit.

14-554 • Appendices CompleteEASE Software Manual

 Comm Test Program
The CommTest.exe program is a stand-alone program provided with CompleteEASE
that lets a user test the remote commands by entering them manually and seeing
the output that CompleteEASE gives. (Thus, this program would act as a test
“client” in place of the external system.) This program is provided only as a
convenience for testing. It can run small scripts, but it is not necessarily meant as a
replacement for custom software being run by an external system as it has limited
functionality. The Comm Test program can communicate to CompleteEASE via
TCP/IP or RS-232, so both protocols can be tested. The communication protocol is
set up in the panel at the bottom of the Comm Test screen, as shown in Figure
14-37.

Figure 14-37. “Comm Test.exe” program provided with CompleteEASE to allow basic testing
of the remote command functionality.

To communicate using TCP/IP, make sure the “Serial?” checkbox is unchecked and
enter the IP address in the “IP Addr” box and the port number (4444 by default) in
the “IP Port” box. If CompleteEASE features requiring asynchronous communication
will be used, then make sure that the “Async IP Port” is set to one number higher
than the IP port number (4445 by default). Press the ‘Connect’ button. The “Status”
label at the bottom will change to say “Connected to…” if Comm Test was able to
connect to CompleteEASE. To communicate using RS-232, make sure that the
“Serial?” checkbox is checked, select the serial port using the drop-down menu, and
press the ‘Connect’ button. The “Status” label at the bottom will change to say
“Connected to…” if Comm Test was able to connect to the specified serial port.
Once a connection has been made, commands can be entered in the “Command”
field. Pressing the ‘Send’ button sends the command to CompleteEASE through the
configured connection. When a response is received from CompleteEASE, it is
displayed in the “Reply” box. Only one command can be sent at a time, and the next
command cannot be sent until a response to the previous command is received.
The “Poll” checkbox allows a command to be entered and then continuously polled
(i.e., sent repeatedly). The response to the poll feature can be found in the “Poll
Reply” box. (For instance, the poll feature could be used to continuously monitor

CompleteEASE Software Manual Appendices • 14-555

 the value of a certain parameter using the “Get()” command.) If an asynchronous
response message is received from CompleteEASE, it will be displayed in the
“Async. Msg” box.

14-556 • Appendices CompleteEASE Software Manual

15. Index

Add Opt. Const. to Report · 3-106

Add Optical Constants to Report · 9-412
# Add to Fit Log · 9-353
Add to Projects List · 1-10
# of Cores · 13-511 Add Trend · 7-311
# of Processor Cores to Use · 9-412 Additive Layer · 10-452
Advanced Graph Options · 9-358
a-Ge · 4-132
Align Sample · 12-491
% All Layer Optical Constants · 7-296
Allow Parameter Query · 13-515
%RANGE · 8-337 Alloy Files · 5-228
%UNIFORMITY · 8-337 Alloy layer · 10-471
alpha-SE · 2-22
Alq3 · 4-133
+ Alternate Model Panels · 9-364
Ambient Index > 1 · 9-381
+Data Menu · 7-280 Analysis Tab · 9-341
+Model Menu · 7-295 Analyze All Rotation Angles · 9-378
+Results Menu · 7-297 Angle Offset · 3-53, 9-378
Angle Offset Correction · 12-499
Angle Scan · 2-29
< Anisotropic Films · 5-234
Anisotropy · 3-41, 3-90
<e1> · 7-282 Append Data · 4-172, 9-342
<e2> · 7-282 Append Intensity Data From Text File · 9-342
<k> · 7-282 Append Timestamp to Data File · 13-520
<n> · 7-282 a-Si · 4-132
<Pseudo> Transforms · 7-282 a-SiGe · 4-132
Aspect Ratio · 13-517
Assume Transparent Region · 10-428
A Auto Fit · 9-350
Auto Report · 13-515
Auto X,Y,Z-Scale · 7-278
Absorbing Region · 4-138
Auto-Fit · 6-245
Absorbing Thin Films · 4-132
Average AC Intensity · 12-492
Acq. Time · 2-29
Acquire Dynamic · 8-325
Acquire Multiple Measurements · 8-321
Acquisition Parameters · 2-28, 8-329
B
Add Folder Link · 1-12
Add Layer · 9-361 Backside Reflections · 3-69, 3-78
Add Layer Below Substrate · 9-362 Bandwidth (nm) · 5-194

CompleteEASE Software Manual Index • 15-1

Biaxial · 10-453 CPPB · 10-444
Biaxial 2 · 10-454 Create Curve · 7-310
Biaxial Layer · 5-237 Create Debug File · 13-522
Blank Model · 5-203 Create New Sub Folder · 1-12
Bootstrap Method · 9-408 Create/Edit Recipe · 2-28, 8-321
Bounding Fit Parameters · 3-51 Critical Point Parabolic Band · 10-444
Browse File System · 1-10 Custom Export Application · 13-519
B-Spline · 4-132, 10-422 Custom Oscillator · 10-446, 10-449
B-Spline Layer · 4-118, 4-139
Build Library · 5-231
Buttons · 1-4 D
Data Analysis Theory · 3-39
C Data Collection · 2-22
Data Export · 13-519
Calibration · 12-494 Data Transfer (to/from WVASE32) · 14-541
Camera Functions · 12-498 Data Type · 2-29, 8-329
Cancel Menu · 7-280 DC Offset · 12-494
Cauchy · 3-59, 3-63, 10-420 Deactivating CompleteEASE · 14-533
Cauchy_Extended · 10-455 Debug file · 14-537
Cauchy_WVL · 10-455 Default TieOff Behavior · 10-427
Changing the Graph · 5-208 Delay · 8-324
Choose From File Dialog · 2-27 Delete Data Set · 7-300
Choose From File Dialog’ ∙ 8-321 Delete Layer · 9-362
Clear Model · 9-361 Delta · 7-281
Clear Multi-Data Set Mode · 5-214, 7-300 Delta Offset · 9-382
Clear Recent Folders List · 1-10 Depolarization · 3-75, 3-79, 7-284
Clipboard Aspect Ratio · 7-280 Depolarization Index · 7-292
Cody-Lorentz · 10-439, 10-440 Derived Parameters · 3-51, 9-399
Color Calculation · 9-412 Diamond-like carbon (DLC) · 4-133
Color Coordinates · 9-402 Difference Mode · 7-286
Common Acquisition Parameters · 8-333 Display Sample Image · 12-498
COMMON Location · 1-8 Display Signal · 12-492
Common Pseudo Substrate Approximation · 6-263 Display Units · 13-504
Common Pseudo-Substrate Approximation (CPA) · 6-260 Double-Y Axis · 7-286
Common Recipes · 8-321 Draw Node Graph · 4-140
Common Scan Patterns · 8-336 Draw Surface Plot · 7-279
Communication Type · 13-514 Drude(NMu) · 10-437
Compare Optical Constants · 3-100 Drude(RT) · 10-437
Comparing Results · 3-99 DUV AC Intensity · 12-492
CompleteEASE Activation Guide · 14-528 Dynamic Data · 7-305
CompleteEASE.cnf · 12-484
CompleteEASEhard.cnf · 12-484
Composition Library · 10-471 E
Configuration Controls · 13-508
Configure Options · 9-414 Edit Configuration · 13-508
Convert to Anisotropic · 5-236, 9-373 Edit Equation · 10-446
Convert to EMA · 9-370 Edit Hardware Config · 12-500
Convert to Isotropic · 9-373 Edit Marker Parameters · 10-448
Convert to Transparent B-Spline · 4-145 Editing Layers · 9-363, 9-365
Copy Analysis Report to Clipboard · 3-107 Effective Medium Approximation · 10-449
Copy Data to Clipboard · 7-276 Einf · 10-433
Copy Graph to Clipboard · 7-276 Ellipsometry · 1-20
Copy to Clipboard · 3-97, 9-351 EMA · See Effective Medium Approximation
Coupled · 10-456 EMA-Coupled · 10-456
Coupled Layer · 4-168 Error Bars · 7-286
CPA · 6-263 Error Log · 13-522
CPA-Fit VI · 6-263 Evaluating Fit Results · 3-43

15-2 • Index CompleteEASE Software Manual

Examples Gen-Osc · 4-146, 10-431
1000nm SiO2 on Si.SE · 3-49 GenOsc 3D · 10-458
25nm SiO2 on Si.SE · 3-47 Glass Catalog · 10-459
500nm SiO2 on Si.SE · 3-49 Glass Substrates · 3-69
60nm SiO2 on Si.SE · 3-49 Global Fit · 3-43, 3-64, 4-167
Anisotropic Polymer on Si.SE · 5-234 Global Fits · 9-391
a-Si Grow Sim.iSE · 6-243 Grade Layer · 9-373
Au substrate.SE · 4-125 Graded Layer · 10-460
Cr on glass_SE.SE · 4-170 Grading · 3-41, 3-87
Cr on glass_T.SE · 4-172 Graph · 7-273, 7-274
Cr on SiO2 MAP.SE · 5-203 Graph All Data Sets · 5-209
Dynamic Ta-Si Multilayer.iSE · 6-263 Graph Clipboard · 13-517
glass substrate_rough.SE · 3-70 Graph Layer Optical Constants · 9-365
glass substrate_smooth.SE · 3-70 Graph Points as a Line · 7-278
glass substrate_taped.SE · 3-70 Graph Scratchpad · 3-71, 7-276, 7-307
multilayer.SE · 4-159 Graph Settings · 7-313
Ni Substrate.SE · 4-112 Graph vs. Angle · 7-285
NiFe on SiO2 on Si.SE · 4-177 Graphing Multiple Data Sets · 3-70
Organic on Au.SE · 4-127 Growth Rate · 6-241, 6-260
PECVD SiO2 on Si.SE · 3-60
series of five SiNx thin films · 3-97
SiC on Si.SE · 3-85 H
SiO2_Map.SE · 5-183
TiO2 on Si.SE · 4-133 Hardware Simulation · 13-517
Expermental Error Bars · 9-408 Hardware Status · 12-490
Export Recipe Data as Text · 13-520 Hardware Tab · 12-484
Export to Text File · 9-345 hardware.cnf · 12-484
Exporting WVASE32 Data · 14-543 Harmonic · 10-436
Harmonic Im(Amp) · 10-437
Hiding Fit Parameters · 3-51
F
File Dialog · 1-9 I
File Structure · 1-7
File Types · 1-8 Im(rho) · 7-282
Fit · 9-346 Imaginary Part Only · 4-148
Fit Dynamic · 6-247, 9-347 Importing CompleteEASE Data · 14-541
Fit Errors Magnitude (+/-) · 9-409 In Situ Data · 6-241
Fit Log · 3-38, 3-97, 3-98 In situ tab · 11-474
FIT Options · 9-389 Align Hardware · 11-478
Fit Parameter Error Estimation · 9-407 Display Signal · 11-478
Fit Parameter Uniqueness · 4-129, 9-406 Pause (Resume) Acquisition · 11-476
Fit Results · 3-48 Start (End) Acquisition · 11-476
Fit Results Panel · 8-336 In situ Tab · 11-474
Fit Rotator Data · 9-347 Include Depolarization Data · 5-196, 9-396
Fit Scan Data · 5-198, 9-346 Include Hardware in 1Pt Export Format · 13-520
Fit Weight · 9-395 Include Initial Fit Stage · 9-389
Fitting Data · 3-43 Include Intensity Data · 9-396
Font · 13-509, 13-510 Include Substrate Backside Correction · 9-378
Full Tensor · 10-457 Include Surface Roughness · 9-363
Initialize Hardware · 12-494
Install/Remove Focus Probes · 12-496
G Installing CompleteEASE · 14-523
Intensity · 7-284
Gaussian · 10-437 Intensity (Total) · 7-286
Generalized Ellipsometry · 8-330 Interference Enhancement · 4-177
Generate · 9-346 Intermix · 10-451
Generate Report · 3-104 IR AC Intensity · 12-492

CompleteEASE Software Manual Index • 15-3

IR Pole · 10-433 Move Z-stage · 8-326
MSE · 3-42
Mueller Matrix · 8-330
K Mueller Matrix Diag · 7-289
Mueller Matrix NCS · 7-289
KK Material · 10-469 Mueller Matrix Off Diag · 7-289
Mueller Matrix Utility · 7-293
Multi Sample Analysis · 5-202
Multi-Data Sets · 6-254
L Multi-Layers · 4-159
Multi-Model Patterning · 9-383
L*a*b* · 9-397 Multiple Data Sets · 5-208, 7-298
Layer Coupling · 4-168 Multiple Users · 1-14
Layer Optical Constants · 7-295 Multi-Sample Analysis · 5-211, 5-220, 9-384
Layer Types · 10-418 Multi-Timeslice Analysis · 6-254
Library · 1-13
Library Models · 9-415
Look & Feel · 13-510
Lorentz · 10-435
N
Lorentz Im(Amp) · 10-436
N, C, & S · 3-42
N, C, S · 7-282
NiFe · 4-177
M Non-Idealities · 5-183
Non-Linear Grade · 10-463
M-2000 · 2-25, 8-318
Manage Users · 13-505
Manual · 13-506
Manual Conventions · 1-3
O
Manual X,Y,Z-Scale · 7-278
Map Data · 5-181, 5-183, 7-302 OC Library · 5-182, 5-228
Match All · 4-155 Offline Activation · 14-529
Max Time Slices · 13-513 Online Activation · 14-528
Max Wavelengths · 13-513 Open Data · 9-341
Maximum Memory Usage · 14-537, 14-538 Open Model · 9-360
Measure · 8-318, 8-321 Opening Files from Windows Explorer · 1-14
Measurement Tab · 8-315 Operator List · 10-447
Memory Management · 13-512 Opt. Const. Compare Model · 5-228
Metal Substrates · 4-111 Optical Constant type (Custom Osc.) · 10-446
MM Derived Parameter · 7-292 Optical Constants · 3-45, 13-505
MM-Jones Quality · 7-291 Optical Constants Profile · 7-297
Mode · 8-315 Options Tab · 13-504
Model · 2-33, 3-39
Model Calculation · 5-192, 9-379
Model Commands · 1-4 P
MODEL Options · 9-377
Models P2VP · 4-133
Glass Substrate (with backside reflection).mod · 3-80 P3HT · 4-133
Glass Substrate.mod · 3-73 Panels · 1-4
Growth Rate and Opt. Const.mod · 6-261, 9-415 Parameter Coupling · 4-167, 9-386
Opt. Const. Compare.mod · 9-416 Parameter Error Bars · 3-48
Pore Size.mod · 9-416 Parameter Perturbation · 9-408
Retardance.mod · 9-416 Parameter Smearing · 9-380
Si with Thermal Oxide.mod · 3-47, 3-50 Parameter Uniqueness · 4-129, 4-171, 9-406
Si with Transparent Film.mod · 3-60, 4-136, 5-185 Parameterize Layer · 4-138, 9-369
Models button Adding/Editing/Deleting · 11-479 Parametric Grade · 10-465
More Options · 7-286 Park Z-Stage · 12-485
Mouse-Selected Menus · 1-5 Patterning · 9-381
Move Angle · 8-325 PECVD SiO2 on Si · 3-58
Move Translator · 8-325 PEDOT · 4-133

15-4 • Index CompleteEASE Software Manual

Perform Thickness Pre-Fit · 9-390 Sample Tilt Alignment · 2-30
Perpetual License vs Timed License · 14-528 Save Data · 9-343
photoresists · 4-133 Save Data after Measurement · 8-318, 8-321
p-Intensity · 7-286 Save Data Subset · 9-343
p-Intensity (Total) · 7-286 Save Layer · 9-362
Pipeline Descriptions · 1-6 Save Layer Optical Constants · 9-368
Pole · 10-433 Save Log File · 3-108
polyimide · 4-133 Save Model · 4-124, 4-158, 9-360
Porous EMA · 10-470 Save OC Library · 5-232
Privileges.cnf · 12-484 Save Optical Constants · 4-122
Profile Along X and/or Y Axis · 7-276 Scale Z-Range by n-Sigma · 7-278
Projects · 1-11 Scan Pattern · 2-32, 8-333
Prompt for Acquisition Parameters · 2-26, 8-319 Scan Region of Interest · 8-335
Prompt for Recipe Components · 2-27, 8-320 Screen Layout · 7-273
Psemi-M0 · 10-442 SE + Transmission · 4-169
Psemi-Tri · 10-442 Seed Thickness · 6-263
Pseudo optical constants · 7-283 Select a Recipe · 8-319
Pseudo Optical Constants · 4-111 Select All Data Sets · 7-300
Psi · 7-281 Select Current Data Set ONLY · 7-300
Psi Offsets · 12-502 Select Data Set · 7-300
Select Mueller Matrix Elements · 7-290
Select Ratios to Graph · 7-288
R Selecting a Map Point · 5-184
Selecting Multiple Points on a Map · 5-206
Random Errors (Noise) · 9-407 Self-Assembled Monolayers · 4-109, 4-125, 4-127
Range-Select · 7-274 Sellmeier · 5-190, 10-470
RC2 · 2-25, 8-318 Semi-Absorbing Thin Films · 4-109, 4-110, 4-132, 4-146
Re(rho) · 7-282 Set Ranges · 4-133, 9-345
Real Part Only · 4-152 Set Symbol Size · 5-205, 7-278
Reanalyze · 9-354 Set Temperature · 8-325
Reanalyzing Multiple Data Sets · 3-100 Sheet Resistance · 9-402
Recent · 1-9 Shortcuts · 1-18, 5-203, 6-242, 13-506
Recipe · 2-25, 8-319 Show Data · 7-274
Recipe Scheduler · 8-322 Show GE Items · 7-287
Reflection Intensity · 8-330 Show Legend · 7-277
Re-initialize · 12-485 Show Map Data · 5-183
Remote Commands · 14-545 Show Mueller Items · 7-289
Remote Communications · 13-514 Show Symbols · 7-277
Remove From List · 1-12 Shutter Controls · 13-520
Remove Grading · 9-373 SiC · 4-132
Rename Layer & Fit Parameters · 3-67 SiC on Si · 3-85
Rename Layer and Fit Parameters · 9-367 Simple Grade · 10-463
Renaming Fit Parameters · 3-51 Simulate Data · 9-411
Report · 3-38, 3-97 s-Intensity · 7-286
Reset · 9-349 s-Intensity (Total) · 7-286
Reset Win. Size · 13-505, 13-507 SiNx · 3-97
Resolution Factor · 13-517 SiO2 on Si · 3-46
Return Path Ellipsometer · 9-382 Slice Value · 10-462
Rotation Data · 7-300 Snapshot · 3-108, 9-361
RS232 · 13-515 Source and Receive Rot. · 9-382
Run Temperature Profile · 8-326 Spectroscopic Ellipsometry · 1-20
Running a Recipe · 2-34 S-T Baseline · 12-487
Standard SE Measurement · 8-330
Statistics · 7-276
Substrate Thickness · 4-175
S Superlattice · 4-166, 9-374
Surface Roughness · 3-41, 3-86
Sample Alignment · 8-316, 8-331 Surface Tracking Fit Mode · 6-263
Sample Height Alignment · 2-31, 8-332

CompleteEASE Software Manual Index • 15-5

System Check · 12-493 Upgrading to CompleteEASE 6 · 14-523
System Status · 8-315 Use Previous Results · 9-387
Systematic Errors · 9-409 Use Scattering Factor · 9-382
User Defined Layer · 10-471
User Equation · 9-401
T User Log In · 1-17
UV AC Intensity · 12-492
Tabs · 1-4 UV Pole · 10-433
Tabulated n,k · 10-420
Tanguy · 10-442
Tauc-Lorentz · 10-439 V
Tauc-Lorentz oscillator · 4-147
TCP/IP · 13-515 VI Mode · 6-263
Thickness Non-Uniformity · 5-192 View Fit Log · 9-353
Thickness Pre-Fit · 3-43, 3-53, 4-166 View Fit Stats · 9-350
Thickness Units · 13-504 View Graph Scratchpad · 7-277
Thin Absorbing Films · 4-169 View panel
Tie Off Positions · 10-427 Dynamic Data · 11-483
Tilt Alignment · 8-331 Log · 11-482
Time Mode · 13-510 Parameters · 11-481
Timeslices · 6-243, 6-253 View Parameter Stats · 9-350
TiO2 · 4-132 View Previous Fit Results · 8-337
Transmission Data % Weight · 9-396 Viewing Results · 2-34
Transmission Data Weighting · 4-176 Virtual Interface · 6-260
Transmission Intensity · 8-330 Virtual Substrate Layer · 10-472
Transparent Films · 3-38, 3-85 Visualizing Parameter Changes · 3-58
Transparent Region · 4-136, 4-145 Void · 10-451
Transparent Substrates · 3-69
Transparent Thin Films · 3-37, 3-58
Troubleshooting · 14-537 W
Try Alternate Models · 3-43, 3-93, 4-137, 9-404, 9-405
Turn Off All Fit Parameters · 9-414 Wannier excitons (Tanguy oscillator) · 10-442
two segment grade equation · 10-467 Wavelength [wvl] (Custom Osc.) · 10-446
Wavelength Range Expansion Fit · 4-141, 9-404
Wavelength Units · 13-504
U Window Setup · 13-516
Wvl Shift (nm) · 9-382
Units WvlByWvl · 10-473
eV · 5-186 WvlbyWvl Layer · 4-113
Light · 13-513
Optical Constants · 13-514
Rate · 13-513 Z
Temperature · 13-514
Thickness · 13-514 ZnS · 4-132
Time · 13-513 ZnSe · 4-132
Translator · 13-514 Zoom All · 7-275, 7-276
Un-Normalized Mueller Matrix · 8-331 Zooming on Map · 5-205
Un-Select Data Set · 7-300 Z-Stage Scan · 12-486, 12-498
Unwrap from EMA · 9-371

15-6 • Index CompleteEASE Software Manual