Practical guides for x-ray photoelectron

spectroscopy: Quantitative XPS

Cite as: J. Vac. Sci. Technol. A 38, 041201 (2020); https://doi.org/10.1116/1.5141395
Submitted: 05 December 2019 • Accepted: 13 February 2020 • Published Online: 08 July 2020

Alexander G. Shard

COLLECTIONS

Paper published as part of the special topic on Special Topic Collection: Reproducibility Challenges and Solutions

This paper was selected as Featured

This paper was selected as Scilight

ARTICLES YOU MAY BE INTERESTED IN

Nature of the use of adventitious carbon as a binding energy standard

Journal of Vacuum Science & Technology A 13, 1239 (1995); https://doi.org/10.1116/1.579868

Silicon (100)/SiO2 by XPS

Surface Science Spectra 20, 36 (2013); https://doi.org/10.1116/11.20121101

Introduction to lateral resolution and analysis area measurements in XPS

Journal of Vacuum Science & Technology A 38, 053206 (2020); https://
doi.org/10.1116/6.0000398

J. Vac. Sci. Technol. A 38, 041201 (2020); https://doi.org/10.1116/1.5141395 38, 041201

© 2020 Author(s).
 TUTORIAL avs.scitation.org/journal/jva

Practical guides for x-ray photoelectron

spectroscopy: Quantitative XPS
Cite as: J. Vac. Sci. Technol. A 38, 041201 (2020); doi: 10.1116/1.5141395
Submitted: 5 December 2019 · Accepted: 13 February 2020 · View Online Export Citation CrossMark
Published Online: 8 July 2020

Alexander G. Sharda)

AFFILIATIONS
National Physical Laboratory, Hampton Road, Teddington TW11 0LW, United Kingdom

Note: This paper is part of the Special Topic Collection on Reproducibility Challenges and Solutions.
a)
Electronic mail: [email protected]

ABSTRACT
X-ray photoelectron spectroscopy (XPS) is widely used to identify chemical species at a surface through the observation of peak positions
and peak shapes. It is less widely recognized that intensities in XPS spectra can also be used to obtain information on the chemical composi-
tion of the surface of the sample and the depth distribution of chemical species. Transforming XPS data into meaningful information on
the concentration and distribution of chemical species is the topic of this article. In principle, the process is straightforward, but there are a
number of pitfalls that must be avoided to ensure that the information is representative and as accurate as possible. This paper sets out the
things that should be considered to obtain reliable, meaningful, and useful information from quantitative XPS. This includes the necessity
for reference data, instrument performance checks, and a consistent and methodical method for the separation of inelastic background from
peaks. The paper contains relevant and simple equations along with guidance on their use, validity, and assumptions.

Published under license by AVS. https://doi.org/10.1116/1.5141395

I. INTRODUCTION that include XPS as one of many “characterization tools” rather

It is the normal expectation that, over time, improvements than papers that employ XPS as the main analytical method. This
are made and things get better. There are often ebbs and flows in implies that a large number of relatively inexperienced researchers
this process and from some perspectives it may appear that, in use XPS to measure samples. These users may assume that the
certain periods, there is regression rather than progression. As instrument manufacturer has written the appropriate software,
x-ray photoelectron spectroscopy (XPS) transforms from being a included all the calibration procedures, and has the appropriate ref-
specialist subject into a mainstream analytical tool, a period of erence data to make their analysis meaningful. Sadly, this is often
relearning is likely to occur. The fact that XPS has achieved wide- not the case. This problem, combined with the difficulty of access-
spread adoption is a testament to the community of experts who, ing coherent, simple, and accessible guidance, results in a signifi-
in the final decades of the 20th century, solved most of the associ- cant proportion of errors in XPS data analysis. One misconception
ated challenges and issues. Through that work, XPS data were that I have heard expressed by those outside the surface analysis
shown to be highly repeatable and reproducible, the instrumenta- community is that XPS cannot be used in a quantitative manner.
tion became more efficient and automated, simple algorithms to With perhaps slightly more justification, I have heard XPS
interpret the data were shown to be sufficiently accurate for most described by casual users as “semiquantitative.” Both points of view
purposes and in some cases, such as the measurement of very relate to the manner in which some researchers use XPS and are
thin silicon oxide layers, XPS was found to be the most accurate fundamentally wrong. Like all other measurement methods, XPS is
measurement method.1 quantitative if the instrument is calibrated and reference materials
Since the turn of the century, XPS instrument sales have con- or data are used. This article is part of a series of practical guides
tinued to grow and the number of papers that include XPS data is and covers in more detail one of the points highlighted in an intro-
doubling every ten years. This is a faster rate than the increase in ductory article, “First steps in planning, conducting and reporting
the number of scientific publications, which generally double every XPS measurements.”3 The assumption is that, having planned and
15 years.2 The majority of this growth can be ascribed to articles considered the experiment, you have decided that a quantitative

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-1
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

result regarding thickness, concentration, or surface coverage not change during XPS analysis before considering that the instru-
should be extracted from the XPS data. ment performance has changed.
In writing this article, my intention is to compile what I think
are the important considerations in performing quantitative analy- 1. Variability (critical)
sis with XPS. Quantitative XPS is the process of comparing two or
There are many sources of variability, all of which should be
more intensities in XPS spectra to determine the amount of mate-
assessed. Usually, the most notable changes of a few percent in
rial at the surface of a sample. These intensities may be from the
intensity are due to variations in x-ray power, particularly, during
same peak or different peaks in the same spectrum or the same set
warmup after the instrument has just been switched on. Find out
of peaks in different spectra. The amount of material may be
how long this takes and let the instrument equilibrate before
expressed as atomic concentration, coverage, area fraction, or thick-
taking measurements.
ness. It is all quantitative XPS and involves the following process:
It is a good practice to assess the performance of the instru-
acquire data, extract peak areas, and employ an equation or algo-
ment on a weekly or monthly basis. This is generally performed
rithm to arrive at a result.
on a reference material, such as freshly sputter-cleaned silver or
If one reads the detailed literature, some of the issues that I
gold. Details on this form of regular monitoring of instrument
omit in this article will be highlighted more strongly than those I
performance are provided in another article in this series8 and
have emphasized. The reader may have the impression that I have
in ISO 16129:2018 “SCA-XPS-Procedures for assessing the
glossed over a number of significant problems. I make no claim to
day-to-day performance of an X-ray photoelectron spectrome-
infallibility and hope that, if the reader knows better than myself
ter.” The simplest assessment is to acquire a survey spectrum,
about such issues, they are already beyond the stage where this
divide this by a reference spectrum taken using the same condi-
article can be of help to them.
tions, and sample at an earlier point in time and identify issues
Before progressing, it is worth commenting that all measure-
from any deviations in the intensity ratio.9 Such practices ensure
ments have an associated error. The important question is “how
that the data from your instrument are at least consistent over
much error?” Therefore, I have tried to highlight the most signifi-
time and provide you with a value for the repeatability of your
cant (>30%) sources of error as critical for quantitative XPS;
data. It is important to remember that every operational mode of
those that should be considered to achieve approximately 10%
an XPS instrument, such as pass energy, slits, lens mode, aper-
error as important; and finally mention a number that are, in the
tures, and so on, produces a different, energy-dependent inten-
context of most works, trivial. In this vein, I have simplified
sity response. Therefore, each of the modes that you use for
many equations so that they contain, as far as possible, only
quantitative analysis should be assessed for repeatability.
things that an average analyst might be expected to know.
Another aspect of variability is the comparability of results
Further detail and a more thorough treatment can be found in
generated by your XPS instrument to those generated by other XPS
textbooks on the subject.4,5
instruments. Such assessments are intermittently carried out
I suggest that the reader begins at the summary, where all
through inter-laboratory studies. Previous inter-laboratory studies
the major points raised in this article are listed, and then proceed
have shown that there can be significant variability between even
to any of the sections that they are interested in or would like
nominally identical XPS instruments in different laboratories.10–13
more understanding. For specific types of materials, more infor-
These differences can be mitigated by commonly applied calibra-
mation is available in other guides in this series, for example,
tion schemes,14–18 but it is of prime importance to remember that
polymers6 and catalysts.7
uncalibrated data from your instrument should not be compared
directly to those from other instruments.
II. BEFORE YOU START
If it is your intention to use XPS to measure something, then 2. Analyzer (important)
there are several things that you should have at hand. First, of If one is simply identifying peak positions in XPS, there are
course, it is essential to have an operational XPS instrument, but it many instrument imperfections that can be ignored. For this “simple
is also necessary to know something about the instrument and its identification” purpose, the performance metrics are energy resolu-
performance. Having the correct pieces of information ensures that tion and energy scale calibration. For quantitative XPS, these metrics
the data, once you have it, are meaningful. are less important as long as the peaks being analyzed can be identi-
fied and distinguished from other features.
The performance metric for quantitative XPS is detector
A. Know your instrument linearity.19–21 This should be assessed during instrument commis-
Modern XPS instruments are reliable and they produce the sioning, servicing, and at approximately yearly intervals if no
same data from the same sample, typically to within a few percent regular servicing is performed. Different detector types have differ-
variability. This is an excellent starting point for quantitative XPS, ent characteristics, but most modern instruments have excellent lin-
but it should not be taken for granted. All instruments drift in per- earity up to a saturation level. Beyond this limit, many strange
formance over time and any maintenance work may also alter per- things can happen that are often difficult to unravel. For accurate
formance. If you have any doubt that the XPS signal varies, check work, it is important to avoid this regime, which will typically start
that the sample gives the same spectrum on two different occasions somewhere between 100 kcps and 1 Mcps. Modern instruments are
during the analysis. It is important to confirm that the sample does quite capable of achieving such count rates in routine operation,

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-2
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

particularly, with calibration materials such as silver and gold. this regard because they often contain a large number of elements,
Reducing the count rate is simply a matter of reducing the emission are electrically conductive, and can be cleaned by ion bombard-
from the x-ray anode. Do not alter instrument settings (e.g., irises ment. The use of sensitivity factors is considered in detail in
or apertures) to reduce the count rate unless you have reference another article in this series.28
spectra or a calibration for the new settings. It is essential to note that changing any of the following can
Consideration should also be given to dark noise in the detec- change the transmission function of the analyzer: aperture or iris
tor and scattering in the spectrometer.21,22 These may affect data by settings; lens settings; and analyzer pass energy. The x-ray illumina-
producing a broad background but are usually of minor signifi- tion area also affects the instrument response because transmission
cance except in the case of high-resolution, low-pass energy experi- efficiency varies away from the focus of the lens. It is also worth
ments. Dark noise can be assessed simply by running a spectrum noting that the sample itself may affect instrument response due to
without the x-ray source on, or with both the sample and sample electric or magnetic fields or topographic features that change the
holder well away from the analysis position. Scattering can be iden- x-ray illumination area.
tified from low-pass energy survey spectra and will appear as an
anomalous background in the XPS spectrum, a shoulder on peaks
or, in monochromated instruments, significant counts at energies 3. Geometry (trivial, but occasionally important)
higher than the x-ray energy. One instrument that I was responsible The angle between incoming x-rays and detected photoelec-
for produced a distinct peak at the high kinetic energy side of each trons at the sample should be known. This is a parameter of your
peak immediately after installation. This feature resulted from elec- instrument that cannot be changed without spanners and a great
trons scattering from a movable plate near the detector that was deal of work. This is not the same as the electron emission angle,
failing to close fully. The problem, after identification, was swiftly or electron take-off angle, which is the geometry of the experiment
fixed by the manufacturer. In general, these effects should be and can be easily changed by tilting the sample.
resolved by the manufacturer, but it is important that the user is For almost all laboratory-based instruments with x-ray sources,
able to identify such problems before generating data. the geometry is not important for most quantitative work, but you
Finally, and of considerable importance, a calibration of the should be able to report what it is. The geometry affects the photo-
intensity scale should be performed. Most spectrometers transmit electron intensity that reaches the analyzer due to the angular distri-
electrons from the sample to the detector with an efficiency that bution of emission from the atom.29 In contrast to photoelectron
changes with the kinetic energy of the electrons, the mode that the emission, Auger electron emission from an isolated atom is isotropic,
analyzer is operated in and the analyzer pass energy or retarding and, therefore, the average intensity is the same in any direction.
ratio of the lens. Each of the instrument modes has an energy- This distribution is indicated in Fig. 1(a), where the incoming x-ray
dependent response, which is commonly called the “transmission is in blue, the atom is a small silver sphere and the analyzer direction
function.” The instrument response has to be applied to the raw indicated by a black line and cone to indicate the analyzer acceptance
data from the instrument to enable a quantitative comparison angle (here about 15°). The semi-transparent sphere represents the
between different modes of the same instrument. It also permits a uniform angular emission and the red overlap represents the
comparison of your data to data from other instruments which detected intensity. A typical magic angle (∼55°) geometry between
have been calibrated in the same way. the x-ray source and analyzer is depicted in Fig. 1(a).
Most manufacturers will either perform this calibration For photoemission using unpolarized x-rays, the electron
during service visits or provide a routine for the user to calibrate intensity is distributed typically as the closed “doughnut” or red
the instrument. Other methods are also available, and it is worth blood cell shape in Fig. 1(b), the red volume is approximately iden-
pointing out that there is a true XPS spectrum (the number of tical to that in Fig. 1(a) at the magic angle geometry. The most
electrons emitted at a particular angle per photon eV steradian) intense emission is at a right angle to the incoming x-ray and
from any sample, which can be obtained from a properly cali- many early instruments, including the instrument on which my
brated XPS instrument.10 Interpretation of XPS results is facili- Ph.D. work was carried out, had a 90° angle between the x-ray
tated by having the true spectrum, or at least a spectrum source and analyzer as depicted in Fig. 1(c). Other instruments
proportional to the true spectrum, because such data can be have 45° angles, with a smaller direct intensity into the analyzer,
directly compared with the theory.23–25 but a reduction in shadowing effects for topographic samples.
In conjunction with intensity scale calibration, the use of rela- In the dipole approximation, which is useful for XPS with alu-
tive sensitivity factors (RSFs) is common in XPS analysis. It is minum and magnesium x-ray sources, the degree of asymmetry in
important to select the RSFs that are valid for the calibration proce- photoemission is characterized by the parameter β, which has a value
dure employed on the instrument. For example, the sensitivity between 0 (isotropic emission) and 2 (emission from s orbitals).
factors of Wagner et al.26 are appropriate to an instrument with a Orbitals which have angular momentum, i.e., p, d, f subshells have β
transmission, which declines approximately inversely with electron values that range from approximately 0.6 to 1.8. Depending upon the
kinetic energy. Others, such as those from National Physical depth distribution of the element in the sample, this angular distribu-
Laboratory (NPL), are based on theory and applicable to true XPS tion is modified by elastic scattering of electrons and the appropriate,
spectra. It is a good idea to check that the combination of “trans- “effective” value of β is smaller than for an isolated atom.
mission function” and RSFs that you are using is consistent by So, why have I categorized this as mainly trivial? For the vast
measuring spectra from materials of known stoichiometry that have majority of commercial instrument geometries, the error is less
a clean surface. Ionic liquids16,18,27 are potentially very useful in than 10% even if angular anisotropy in photoemission is not taken

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-3
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

FIG. 1. Illustration of the angular distribution of electron emission from an atom. (a) Isotropic (spherical) emission, typical of Auger electrons. (b)–(f ) Example photoelec-
tron emission (β = 1.5) with (b) unpolarized x-rays at 55°, (c) unpolarized x-rays at 90°, (d) polarized x-rays at 55° with x-ray e-vector 45° out of plane, (e) polarized x-rays
at 55° with x-ray e-vector in plane, and (f ) polarized x-rays at 90° with x-ray e-vector in plane. The intensities are nearly equivalent in (a), (b), and (d). In (c), (e), and (f ),
they are larger by an approximate factor of 1.3, 1.8, and 2.5, respectively.

into account. The worst case occurs for 90° instruments when
errors could reach up to 20% or more and, for these instruments,
corrections for angular distributions in photoemission intensity
should be made.
Figure 2 illustrates the effect of changing the angle between
x-rays and analyzer from 90° (black line) to 45° (red dashed line).
Note that the Auger electron intensity is constant in the two geom-
etries and also that the inelastic backgrounds converge at kinetic
energies lower than the photoelectron peaks because elastic scatter-
ing randomizes the direction of electrons which have a long path
length in the sample.
The problem of electron angular distribution can, however,
become critical if a polarized x-ray source, such as a synchrotron, is
used. Although laboratory-based instruments with monochroma-
tors also have a degree of polarization from the Bragg reflection,
this is almost always a trivial effect. Synchrotrons are often
designed to generate near-perfect plane-polarized x-rays, which
causes some issues for quantitative XPS. Figure 1(d) shows the best
geometry for quantitative XPS in the dipole approximation, having
the magic angle between incoming x-rays and the analyzer and also
the plane of polarization 45° out of the x-ray/sample/analyzer
plane. Note that the cross section through the angular emission has
the same shape as in Fig. 1(b). In this case, the detected intensity
per photon will be similar to an isotropic distribution. However, I
have never encountered a synchrotron end station that has this
FIG. 2. Effect of x-ray to analyzer angle on XPS spectra. Monochromated Al geometry, typical geometries are: magic angle with in-plane x-ray
Kα XPS of clean copper at 45° (red dashed) and 90° (black). The absolute polarization, shown in Fig. 1(e), and the “maximum intensity in
intensities of photoelectron peaks change by ∼50% and the Auger electron
the peak” 90° geometry shown in Fig. 1(f ). Such geometries pose
peak intensities are unchanged.
interesting interpretive challenges if quantitative XPS is desired.

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-4
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

B. Information, reference data, and samples that you

will need
In this section, some of the underpinning detail is provided
on how XPS can be made quantitative. In the most general form of
XPS analysis, the peak intensities are extracted from spectra and
divided by sensitivity factors to provide numbers that correspond
to relative compositions. Such analysis places a large amount of
trust in the instrument, its calibration, the sensitivity factors, and
the ideality of the sample.

1. Reference materials (critical)

This section is vital for careful quantitative analysis in which FIG. 3. Why we need reference materials. (a) Simulated spectrum of an
accuracy is required. Even if you only intend to carry out routine element with plasmon losses, which are both intrinsic and extrinsic. The inelas-
XPS work, it is worth looking through this section to understand tic background (red, lower solid curve) is shown under the total simulated spec-
the uncertainties that arise from assumptions used in more general trum (black, upper solid curve). (b) Simulated spectrum of the metal oxide,
and routine XPS analysis. which is modeled without a shake-up structure. (c) Simulation of an overlayer of
oxide on the metal, the practical analysis region is indicated by dashed lines,
A sample that requires XPS measurement will have more than which misses the intrinsic plasmon intensity.
one chemical phase in it. For the most accurate measurements, it is
best to have pure samples of each of the phases in the sample. Ideal
reference materials are flat, single phase, pure compounds of
known stoichiometry with a very clean surface. The essential point Figure 3(c) represents an overlayer of the oxide on the metal, where
is that the reference material composition is known by other means the relative intensities in the two main peaks could be used to measure
and that it will not be determined by XPS. If your samples are to the oxide thickness. Using the reference spectra, this is straightforward
be measured with better than 5% accuracy, then the reference mate- and the analysis can be carried out with little error. However, if theo-
rials should be pure phases of the individual mixed phases in the retical intensities were used, it would be difficult to account for the dif-
sample. Each of the reference materials should be measured using ferent amounts of intrinsic intensity outside the analysis region and
the same conditions as that for the sample measurement, ideally the resulting thickness would almost certainly be erroneous.
several times both before and after the test sample to establish mea- In practice, it is often hard to get hold of such clean, pure, and
surement precision and account for instrument drift. For each flat reference materials and therefore some thought is required to
phase, this then provides a “pure material” signal intensity, Ii,∞, acquire suitable materials. It is often necessary to find reference
that can be used to normalize the signal from the sample, Ii, to materials that are not identical to the phases of the sample under
provide an accurate measurement using equations presented later. study, but do contain chemically similar environments in known
You may ask why this is necessary, because XPS is a well under- concentrations. For example, to find the correct reference materials
stood technique and, therefore, we should be able to predict the inten- for organic overlayers on gold surfaces and nanoparticles, it was
sity arising from each material. The answer is that we can predict the not possible to source the exact organic materials. Instead, other
intensity, but an unknown fraction of that intensity goes into satellite pure organic materials were used and an assumption made that
peaks. In general, the majority (typically >80%, but sometimes less they have similar effective attenuation lengths (EALs) and atomic
than 50%) of the intensity will be in a main, easily visible, peak.28 densities.30,31 It was found that the experimental ratio of reference
However, a fraction of the intensity will be in satellite features, which intensities was 40% different to that predicted theoretically, almost
usually appear at a lower kinetic energy than the main peak and are certainly due to consistent errors in determining peak areas and
due to additional, intrinsic energy losses that occur during photoemis- excluding some of the intrinsic structure.
sion. These are called intrinsic plasmon losses, shake-up, or shake-off The two examples of oxide on metal, with ∼25% error in a
features and arise from additional electronic excitations that can occur peak intensity, and organic on gold, with ∼40% error in a peak
simultaneously with the photoemission event. In a few cases, such as intensity, illustrate the point of this section. The standard XPS
for the Cu2+ 2p peaks and Ce4+ 3d peaks, these loss features are sharp, practice of reporting elemental concentrations without reference
easily identified, and can be included in the quantification. In other material data is likely to be inaccurate. In the worst case, if there
cases, they are broad and hard to separate from the inelastic back- are different phases in the sample, then the relative error could be
ground, which may also include similar lumps and bumps. as high as a factor of 1.5. However, this lack of accuracy does not
Figure 3(a) shows an example, simulated spectrum representative detract from the fact that XPS is remarkably precise and relative
of the 2p region of pure aluminum or silicon where the features at changes in composition of less than 1% are often easy to detect.
lower kinetic energy contain contributions from both the inelastic For most practical purposes, this is sufficient.
background and the intrinsic peak. In the simulation, about 25% of
the total intrinsic intensity is in the plasmon feature. These contribu-
tions are almost impossible to separate from the inelastic background 2. Sensitivity factors (important)
in a practical analysis. Figure 3(b) is a simulation of the metal oxide The majority of quantitative XPS measurements express an
spectrum which is, naïvely, modeled without shake-up features. analysis result as an atomic fraction of each element or chemical

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-5
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

state. For this purpose, it is essential to have a set of sensitivity necessary to take a full survey spectrum of the sample and report
factors, Si, for each type of photoelectron in the sample. These all of the elements detected. There should be no need to say any
essentially represent the intensity of the peak relative to another, ref- more than this, but there may be a temptation among the inexperi-
erence, elemental peak. If your instrument is not calibrated in any enced to save time by not taking a full survey spectrum or to disre-
way, then it is necessary to find these yourself. Wagner et al.26 did gard certain elements, which are not considered important.
this for an Al Kα XPS instrument by measuring the intensities of a
set of compounds of fluorides of known stoichiometry and, there- 4. Information about the sample (important)
fore, the natural reference peak was F 1s. They found that the C 1s
The more information you have about the sample, the better. It
sensitivity factor was SC1s = 0.25, implying that for polytetrafluoro-
is always helpful to have some idea which elements you are looking
ethylene (PTFE) (CF2), the F 1s peak is eight times more intense
for and at roughly what concentrations. XPS has variable sensitivity:
than the C 1s. A similar approach was taken by Edgell et al.32 when
generally very good for heavy elements and not so good for light ele-
they determined experimental sensitivity factors for an Ag Lα x-ray
ments. It is also good to check whether the experiment is even feasi-
source. As noted previously, such sensitivity factors are ideal for the
ble or identify if it will take a long time due to poor sensitivity.39
instrument and operating mode that was used to measure them, but
There are often peak overlaps that make the identification of certain
only applicable to instruments and operating modes that have the
elements difficult, for example, aluminum in copper for most stan-
same energy-dependent transmission, or the same intensity scale
dard x-ray sources which cannot access the Al 1s orbital.
calibration procedure. For an Al Kα instrument with constant trans-
If the sample is topographic, for example, is rough or has a
mission at all energies, the F 1s peak for CF2 is actually only six
cylindrical or spherical form, then this will affect the measurement.
times as intense as the C 1s peak.18 The ∼25% difference between
Often, the topography will merely reduce the intensity of peaks due
the sensitivity factors found in these references is indicative of the
to shadowing effects. However, there will also be a wider variety of
error incurred by using the wrong sensitivity factors. More detail
electron emission angles relative to the local surface normal. This
can be found in ISO 18118:2015 “SCA-XPS-Guide to the use of
will alter the calculation method for film thickness or overlayer cov-
experimentally determined relative sensitivity factors for the quanti-
erage measurements.40–43 For such samples, microscopic data such
tative analysis of homogeneous materials.”
as atomic force microscopy or electron microscopy are helpful.
If you have any doubt whatsoever, then analyze a few samples
The sample is best considered as part of the instrument and
with known stoichiometry and a clean surface such as: PTFE,
some samples may change the performance of the XPS spectrome-
freshly spin-cast or scraped poly(methyl methacrylate), a salt such
ter in a significant way. Magnetic materials may significantly alter
as a fluoride, or an ionic liquid.16,18 If you do not get the answer
the trajectory of electrons and many instruments use magnetic
you are expecting, then something is wrong.
lenses that can change the magnetization of the sample itself.
For true XPS spectra, it is possible to estimate sensitivity
Considerable expertise is required to deal with these types of mate-
factors from theoretical parameters. In magic angle instruments,
rials and quantitative analysis should not be attempted from XPS
this can be as simple as multiplying the theoretical photoionization
data of strongly magnetic materials without understanding the pit-
cross section by the IMFP, that is, Si is proportional to σi λi,M and
falls. Highly topographic, nonconducting samples can also affect
the only question is which representative material “M” to use for
XPS intensities by altering the trajectory of emitted electrons when
the IMFP and which element to use as the reference element. This
they become charged under an electron flood gun. It is difficult to
type of calculation is quite adequate for most purposes,18 especially
predict the effect that this will have on the instrument transmission,
when it is considered that the relative variation in theoretical cross
and in these cases, quantitative analysis should also be approached
sections may be in the region of 10% depending upon which
with caution.
theory is used.33–35 Far more detailed analysis has been carried
If the sample is a single crystal, you need to be careful about
out,24,25,36,37 but the typical relative scatter between experimental
the orientation of the crystal to the emission angle of the electrons.
and theoretical sensitivity factors is ∼10%.38 For any sort of routine
Most quantitative XPS analyses assume that the sample is either
work without appropriate reference materials, this should be con-
amorphous or polycrystalline. The angular distribution of photo-
sidered as the best possible accuracy of standard XPS measure-
electrons and Auger electrons are influenced by the local arrange-
ments, while remembering that compositional changes smaller
ment of atoms, which is a significant effect for single crystals44 and
than 1 at. % can be identified routinely.
this is known as photoelectron diffraction.45,46 These effects can be
Finally, in some sources of theoretical sensitivity factor data,
used as a powerful form of structural analysis but, in the context of
the separate spin–orbit doublets are assigned a sensitivity factor,
routine quantitative XPS analysis, they can be annoying. These dif-
but these can sometimes be impossible to separate in a spectrum. If
fraction effects can be reduced by both selecting the sample angle
this is the case, the correct approach is to sum the two sensitivity
to avoid collecting electrons emitted along the low-index axes of
factors, for example, SSi2p, the sensitivity factor for the silicon 2p
the crystal and also by increasing the collection angle of the ana-
peak will be the sum of the sensitivity factors for the silicon 2p3/2
lyzer to average out diffraction effects.47
and 2p1/2 orbitals.
III. USING THE DATA
3. Full survey spectrum (important) After checking that you have all the necessary data and infor-
In quantitative XPS, all the elements present, except hydrogen mation, it is time to start the process of quantitative analysis. The
and helium should be considered and reported. Therefore, it is ability to assign peaks correctly is an important step and for

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-6
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

quantitative analysis, this becomes critical in the case that there are under the oxide layer, they have a small intensity even though the
peak overlaps. Also remember that, if you have one peak from an metal contributes significantly to the inelastic background. The
element, then all the other peaks from that element should also be oxide overlayer contributes two broad peaks, but less intensity to
observed in the survey spectra. If an element is present, but the the inelastic background. The blue dashed line is the division
main peak overlaps another peak from a different element, then the between peak and background from the model and it is unreason-
two recourses are to (1) find another peak from the same element able to expect any standard background shape to match it exactly
that does not overlap or (b) attempt a peak fit; an example of this because there is no way of inputting the layer structure. By choos-
method is provided later. Peak fitting different chemical states of ing end points close to where the model background starts and
the same element is also a form of analysis that may be quantita- ends and a variety of background shapes, the gray shaded area rep-
tive. This is not covered in detail in this article because it is dealt resents the range of estimated background positions. In this case,
with in another article in this series.48 both standard Tougaard and Shirley backgrounds provide areas
within 10% of the model and a linear background within the gray
A. Selecting regions and backgrounds area has a 15% discrepancy. Thus, the background shape can be
important but is not usually critical.
A good quantitative analysis hinges upon making a sensible
The dashed red line with a big cross through it illustrates the
choice for which peaks to use and how they should be measured.
effect of a badly chosen end point. In this case, a linear background
There are a number of issues here, but, in general, it is best to use
is shown because in this case, it is most sensitive to end point selec-
the sharpest peaks. These tend to be the “leading” peaks with the
tion with an approximate 100% error in peak area for the line in
highest orbital angular momentum in each shell, i.e., 1s, 2p, 3d,
the figure. It is worth noting that, for these end point choices, a
and 4f. Sharp peaks are easier to find in a spectrum and it is
Shirley background (not shown) is hardly better than a linear back-
slightly less challenging to define a suitable background for them.
ground and even a Tougaard background using normal parameters
and assuming depth homogeneity has a 50% error. The link
1. Choosing a suitable background (critical) between the peak intensity and the background intensity is not
The separation of an XPS peak from its associated background straightforward because of the layer structure and accurate back-
is a process that is difficult to get right. If accuracy in the raw peak ground subtraction requires a more detailed model.
area is required, then one requires either a reference material, as Figure 4(b) illustrates a simple error made by inexperienced
discussed previously, or a great deal of detailed and time consum- XPS users. Here, there is a weak peak with significant noise and a
ing work. For most practical work, the commonly adopted linear background is used. The two red dashed lines use single
approach is to select a background that is fit-for-purpose and this
essentially restricts the choice to linear, Shirley49 or Tougaard.50 A
linear background usually works well for overlayers, or if the other
two are not working for various reasons. Shirley is useful if there is
a “step” in the background after a peak and it is commonly used
for peak fitting. It makes the assumption that the change in back-
ground intensity is proportional to the peak intensity above the
background. Tougaard’s background has a physical basis, calculat-
ing the background from an inelastic scattering cross section.51
Although the choice of the cross section can be uncertain,
Tougaard’s background should be preferred for quantitative analy-
sis, particularly for peak areas from survey spectra, because it pro-
vides results that are often similar to accurate experimental
results.52–54 Tougaard’s background requires a large energy range
in the background at lower kinetic energy to the peak, typically
more than 30 eV, and this should be considered before collecting
data. Additionally, there can be problems from overlapping peaks
within this large energy window.
However, the really critical issue in determining peak areas is
not the shape of the background, but the choice of the end points
on either side of the peak that define the background. FIG. 4. Choosing a background. (a) Simulated XPS spectrum typical of the 2p
region of a 3d metal with an oxide overlayer. The blue dashed line is the back-
Figure 4 illustrates some of the problems that may be encoun-
ground from the simulation and the gray area is the range of background positions
tered in defining a background. The spectrum in Fig. 4(a) is gener- found using reasonable selections of algorithms and end points. The red dashed
ated as a simplistic model of the 2p region of a transition metal line with a cross is an unreasonable linear background. (B) Linear background on
such as iron, which has an oxide overlayer. The reason for using a a noisy peak; dashed red lines use one point in the data and the blue line uses an
model is so that it is clear which region corresponds to the peak average position over a large number of background data points. (C) Problems
and which corresponds to the background. The two metal peaks, encountered analyzing peaks on a sloping background. Naïve interpretations of the
Tougaard (red dashed) and Shirley (red solid curve) backgrounds are unphysical,
2p1/2 and 2p3/2, have a higher intensity in the pure material, a the blue line is linear and the only reasonable choice here.
lower binding energy, and are sharp. However, because the metal is

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-7
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

points in the data to define the end points of the background. 2. Peak fitting (important)
These are clearly poor choices compared to the blue line, which The purpose of a peak fit is to separate overlapping peaks in
uses an average of many data points to define the end points in the the same region of the spectrum in a physically meaningful way and
background before and after the peak. The lesson to remember is in order to extract useful information. This can be relatively straight-
that the line describing the background should go through the forward if the relative peak positions are known from the literature
middle of the noise in the data before and after the peak. This is and the peaks are sharp and well-defined. For organic materials,
very easy to assess visually. It is also quantitatively critical; the two this is often the case and reference works are available.57 For other
red dashed lines provide areas of either zero or twice that provided materials, care should be taken, especially if there are extensive and
by the blue line, i.e., 100% error. Guidance is provided by NIST intense shake-up satellites.58 In some cases, peak fitting is the only
(Ref. 55) on the means of achieving the lowest statistical uncer- means of performing an analysis. Assessing the accuracy of a peak
tainty in XPS peak areas. There should be at least as many data fit is not at all easy and I will not attempt to do it here. However, it
points on each side of the peak to define the background end should be noted that uncertainty calculations from the reduced
points as there are data points across the peak itself. Thus, it is chi-squared will only tell you the level at which the model agrees
essential to collect at least twice as much background data as peak with the data; it does not tell you whether the model itself is
data if you want to minimize statistical uncertainty in the peak correct. Most peak fits fail the “physically meaningful” criterion for
area. It is also possible to “actively” fit the background and peak the following reasons: the peak assignments are flaky; the back-
simultaneously, which may help if you have inadvertently not col- ground is poorly modeled; shake-up structure is not considered; or
lected enough background data.56 The uncertainty associated with the depth distribution of components is not taken into account.
this approach is not clear and it is affected by the choice of both Naturally, it is important to have sufficient energy resolution to dis-
the peak shape and the background shape. Of course, this does not tinguish the peaks of interest and this often means using a small
ensure your choice of background is correct, but it is a warning pass energy. It is possible to compare relative intensities of peaks
against trying to make experiments faster by chopping the back- within a small kinetic energy range, but the raw intensities should
ground regions out of the scan. not be directly compared to higher pass energy data, or different
While on the topic of correct background choices, Fig. 4(c) is energy regions, without transmission function correction.
included to emphasize that software will let you do things that even a Figure 5 illustrates spectra from a material containing: lantha-
novice should be able to spot as an error. Problems can arise for num, zirconium, oxygen, lithium, and variable amounts of gallium.
peaks on a sloping background, particularly, one that increases in The concentration of gallium was of interest but this was a minor-
intensity with kinetic energy. This is actually quite common for ity component and the only sharp features suitable for analysis
surface species on a bulk material in calibrated XPS data, but is not were the Ga 2p peaks, which overlap the La 3p3/2 peak. This issue
so common for uncalibrated data from older instruments. Some was identified before XPS analysis and a sample made without any
implementations of the Tougaard and Shirley backgrounds assume gallium. The gallium-free sample spectrum is shown in Fig. 5(a)
that there is no slope in the background and this leads to a flat line and a spectrum from one of the gallium-containing sample in
for the Tougaard, the red dashed line, and a negative peak area. Since Fig. 5(b). The reasons for the peak shapes here are not important,
the “flatline Tougaard” is so obviously wrong, I have not seen it used but may largely be ascribed to damage after ion bombardment. To
in publications or presentations. However, I have seen the “upside- obtain the gallium peak areas, the best approach was to first find a
down Shirley” shown by the red solid line. In the case shown in description of the La 3p3/2 peak shape as shown in Fig. 5(a). This
Fig. 4(c), it gives a similar area to the blue linear background because shape was then applied to the data in Fig. 5(b) along with Ga 2p
of the end point choices, but in other cases it can give very inaccurate peaks, in this case the constraints included: the known separation
peak areas and should never be used in peak fitting. between the spin–orbit doublets; the known 2:1 intensity ratio

FIG. 5. Using a reference spectrum to

cope with overlaps. (a) La 3p3/2 peak
shape determined from a sample
without gallium in. (B) Fitting of the Ga
2p peaks in a sample containing
gallium. Sample courtesy of Federico
Pesci and Sarah Fearn, Imperial
College, UK.

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-8
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

between j = 3/2 and j = 1/2 pairs and a constraint to keep the remembered that there is an underlying assumption. The assump-
widths of each spin–orbit couple the same (NB: this constraint is tion is that the sample is homogeneous and single phase within the
not a strict necessity, but is always a useful starting point). Two XPS sampling depth. Very few samples actually meet this criterion
pairs of Ga 2p peaks were found to be required. The fit is driven by and it is a matter of faith that both the person reporting the result
the Ga 2p3/2 peaks, which are relatively clear of the structure at and the person reading the report understand the assumption. If
higher binding energy. The fact that the whole data are matched by there is any doubt, I recommend making it clear using the phrase
the fit without any additional peaks provides confidence that the “equivalent homogeneous composition.” The calculation [Eq. (1) in
Ga 2p area has been measured with reasonable accuracy. Table I] is very simple. For each element, select a peak and divide
the area of that peak, Ii, by the sensitivity factor, Si, to obtain a nor-
3. Data mixing and matching (important) malized peak area, Ii/Si. The equivalent homogenous atomic frac-
tion, Xi, of each element is simply that elements normalized peak
The direct comparison of data taken using different modes of area divided by the sum of all normalized peak areas. To get
an XPS instrument should be carried out with care. In some cases, atomic percent (at. %), simply multiply Xi by 100%.
it is necessary. For example, a quantitative analysis of atomic con-
centrations is best carried out at high pass energy due to higher
counts and the lower influence of effects like scattering in the spec- 2. Relative area
trometer. At the same time, an element may be in different states If the sample is suspected to have different phases within the
and will also require a low-pass energy, high-resolution spectrum volume analyzed by XPS, then the equivalent homogeneous compo-
to separate these. The low-pass energy spectrum can be used to sition can only be used as an indication of the amount of each
find the fraction of elements in each state and this information material present. However, if the distribution of material is known,
combined with the high pass energy quantitative analysis. If there then more accurate information is available. A simple, but rare, case
are valid transmission function corrections available for the is a mixture of two phases (P and Q) at the surface, which occupy a
low-pass energy mode, then quantitative analysis can also be constant fractional area of the surface within the XPS information
carried out using those spectra. Since a high pass energy survey depth. The normalized relative intensity Ap,q [Eq. (2) in Table I]
spectrum is required in any case, the data may as well be employed represents the relative area of phase P to phase Q. If these are the
to support a quantitative analysis. If it disagrees with the low-pass only two phases, then the fractional area of phase Q is (1 + Ap,q)−1.
energy results, then you need to check whether your sample is The most tricky problem is finding the ratio of the pure phase
damaging or whether you have a problem with your instrument. intensities, Iq,∞:Ip,∞. This is best done with pure, flat reference
Make sure that all the peak areas have the same units, which materials as described earlier. If this is not possible, but there are
are typically written as cps.eV or counts.eV. If they do not have the clean mixed materials available with similar surface conditions,
same units, then nonsense will result. I also strongly advise you to then the pure material intensities can be estimated by assuming
make sure you understand what any software packages you use do that the signal is proportional to the fractional area of the surface.
to your data and check that the peak areas they report are what you For a binary mixture, this is illustrated in Fig. 6(a), by plotting the
think they are. For example, you should understand when and how intensity of a peak, q, from one component, Q, against a peak, p,
a transmission function is applied. Sometimes, it is applied to the from the other component, P. It is then possible to extrapolate to
whole data set and sometimes it is applied only to the peak areas the axes and estimate the pure phase intensities or use the slope of
that have been extracted from the raw data. You should ensure that a linear fit to obtain the ratio of pure material intensities. The error
it is not applied twice to your data. It is worth working through the can be estimated from the scatter of the points around a linear fit.
calculations yourself for one or two sets of data and checking that A wide range of compositions is required to reduce uncertainty.
the software gives the same result. Please note that a consistent variation in contamination, topogra-
phy, or instrument performance will cause the estimated intensities
B. Using peak areas to be wrong and the plot itself will not identify such issues.
Now we have peak areas, we have to turn them into something Therefore, these possibilities should be checked by looking for con-
understandable. This is always done by comparing one peak area to taminants in the survey spectra, measuring topography, measuring
another and the simplest example is a peak fit of the C 1s region. the samples in the same area at least twice in different sequences
The area of each peak, compared to the total C 1s area, can be and using different areas of each sample to assess variability.
interpreted as the fraction of carbon in the chemical environment If no other information is available, the pure material intensity
represented by that peak, weighted by the depth distribution of the ratio can be estimated from first principles using photoionization
various species and the depth sensitivity of XPS. The comparison cross sections, inelastic mean free paths (IMFPs), and atomic densi-
of peak intensities from different elements or to measure film ties, if these are known or can be estimated for the material under
thickness is somewhat more involved. investigation. For reasons given earlier, these estimates should be
associated with a high (>20%) uncertainty.
Although the use of the normalized relative intensity Ap,q as a
1. Equivalent homogeneous atomic fraction measure of relative area is not common, I have introduced it here
The most commonly used method of reporting XPS data is as because it is used in other XPS measurements, such as the coverage
atomic fractions and usually as an atomic percent (at. %). This con- of one material on another [Eq. (3) in Table I]. It is much more fre-
vention is perfectly fine and understandable as long as it is quently used as an input for the measurement of overlayer thickness.

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-9
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

TABLE I. Summary of some equations used in quantitative XPS. each material can be used in a similar manner to that described for
relative area reference intensities as shown in Fig. 6(b).
Quantity, If the electrons from peak “p” have a different kinetic energy
Expression assumption to the electrons from peak “q,” then Eq. (4) is no longer valid.
(1) Atomic Also, the plot in Fig. 6(b) should no longer be a straight line, but
Ip /Sp this may be hard to spot in the presence of noise or if there is a
X¼P fraction
j Ij /Sj Homogeneous limited range of thicknesses in the samples. Dashed black lines are
included to indicate the expected effect of unequal sampling depths
(2) Relative area of for the electrons. Estimating the ratio of effective attenuation
Ip Iq,1 lengths is quite easily done because the electron kinetic energy is
Ap,q ¼ P to Q
Ip,1 Iq Homogeneous known [Eq. (A2) in the Appendix].
in depth Assumption (1) concerning exponential attenuation is a good
(3) Coverage of P one; however, it is important to use EALs rather than IMFPs. The
λp,P former accounts for the fact that electrons can change the direction
ΦP ¼ Ap,q cos θ on Q
aP Low coverage in a sample.
Assumption (2) concerning flat samples requires checking by
(4) Thickness of P microscopies, preferable AFM because this provides quantitative
t ¼ ln(Ap;q þ 1)Lp,P cos θ
on Q height and slope information. Typically, samples can look quite
Flat sample rough in AFM due to an exaggerated height scale, but the lateral
and Ep ≈ Eq distance over a height change often means the local surface normal
(5) Thickness of P is close to the “average” surface normal of the sample. For XPS, the
p,q ln(Ap,q )Lq,P Lp,P þ 2Ap,q Lq,P Lp,P
A2:2 0:95 0:05 0:42 0:58
t ¼ cos θ on Q slope is important but it should not be a major concern unless the
p,q þ 1:9
A2:2 Flat sample range of local emission angles from the sample exceeds ∼10° or so.
If the sample consists of, for example, spheres or fibers, then the
equations given here are not valid and corrections are required.41,42
Core–shell nanoparticles also require special consideration.40,43,59
Assumption (3) regarding the uniformity of the overlayer is
3. Overlayer thickness difficult to test without the aid of microscopy. For flat samples,
The most common assumptions in measuring film thickness changing the take-off angle and hence the information depth
are (1) that the intensity of electrons decline exponentially with dis- could identify some forms of nonuniformity.60 Similarly, photo-
tance travelled through a material, (2) that the substrate is flat, (3) electron peaks from the same element but with different EALs
that the overlayer has uniform thickness, and (4) that the electrons could be used.61 If your instrument has two, quite different x-ray
all have the same effective attenuation length in the overlayer mate- energies, then nonuniformity may be assessed by changing the
rial. With these assumptions, the thickness of the overlayer, P, on x-ray source and analyzing the same area.18 A more elegant
the substrate, Q, is given by Eq. (4) in Table I. The equation uses approach is to perform a detailed analysis of the inelastic back-
the normalized intensity ratio, Ap,q, and once again the main ground, which can reveal nonuniformity.62,63 All these methods
problem is finding the pure material reference intensities. If refer- are far from routine, require considerable understanding, and can
ence samples are not available, but a range of samples with different generally only identify the grossest forms of nonuniformity (such
film thicknesses are available, then a plot of peak intensities from as holes or islands). The effect of unidentified overlayer

FIG. 6. Estimating pure material inten-

sities from a set of samples. (a) Two
phases P and Q, homogeneous in
depth. Black solid line represents ideal-
ity and data points are with added
noise. Red dotted line is a linear fit to
the data. (b) Two phases with P on top
of Q; the black solid line is the expec-
tation with equal effective attenuation
lengths and the dashed black lines
with factors 3/4 and 4/3 difference in
effective attenuation lengths for the
overlayer and substrate electrons.

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-10
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

nonuniformity on an XPS thickness measurement is that the TABLE II. Topics covered in this article.
average thickness of the layer will be underestimated and this is
true in all cases, from flat films62 to nanoparticles.64 Section Main points
Assumption (4) is applicable when the electron kinetic ener- Instrument Assess typical repeatability and variability
gies being compared are within ∼5% of each other. The important Assess drift on a regular basis
effective attenuation length is that for electrons travelling through Check the linear range of your detector
the overlayer material, P. Because the kinetic energies of the elec- Calibrate both the energy and intensity scale
trons p and q are similar, then the EAL for the overlayer electrons, Know your instrument geometry
Lp,P, is approximately the same as that for the substrate electrons, Information Find appropriate reference materials
Lq,P. If the EALs are different, then it is clear that another approach Check that your sensitivity factors are useful
is required. The methods to cope with this are either by iteratively Always take a survey spectrum to identify
changing t in the relevant equation [Eq. (A3) in the Appendix] contaminants
to match the experimental Ap,q; using a graphical method;65 or use If necessary, measure the topography of the sample
an accurate empirical equation such as that given in Eq. (5). Data Collect more background data than peak data
Equation (5) in Table I is mathematically identical to a previous, Choose background shapes and positions with care
validated direct equation for flat surfaces with uniform overlayers.40 Peak fitting should be approached with caution
Understand what your software does to the data
IV. REPORTING RESULTS Carefully explain your calculation method
When reporting the results of an XPS analysis, it is important to Provide enough information for others to repeat the
provide sufficient details so that the analysis can be repeated by others. calculation
More details can be found in ISO 13424:2013 “SCA-XPS-Reporting of Reporting Report everything that may affect your result
results of thin-film analysis,” ISO 15470:2017 “SCA-XPS-Description
of selected instrumental performance parameters,” ISO 19830:2015 e. The software used to analyze data.
“SCA-XPS-Minimum reporting requirements for peak fitting in X-ray f. Area of peaks used in quantification and whether these are
photoelectron spectroscopy,” and ISO 20903:2019 “SCA-XPS- raw or calibrated.
Methods used to determine peak intensities and information g. Full details of any peak fitting: position, width, shape,
required when reporting results.” In an academic publication, constraints.
there is no barrier to provide all the necessary information in (5) Results
electronic supplementary information. Ideally, the following a. Equations used to analyze the data, with references.
should be specified as a minimum: b. Estimate of the uncertainty in the results.
(1) Instrument details This may seem a very exhaustive and exhausting list, but many of
a. Make and model of the XPS instrument. these details will be identical for a large number of analyses carried
b. Lens mode and pass energy or retarding ratio. out in your laboratory. Therefore, if a system is set up to record
c. Angular range of electrons collected. and report these details for each set of samples, there is some initial
d. Dimensions of analysis area. work but then the additional burden for each sample is not so
e. Type of anode or x-ray source. great. On the other hand, if you do not know what some of these
f. Geometry of the instrument. are, you should find out before reporting your XPS results.
g. Linear range of detector in cps.
h. Method of energy and intensity calibration.
(2) Information used V. SUMMARY
a. References to sources of RSFs, EALs, IMFPs, and so on. The messages from each of the sections in this article are sum-
b. A list of the values used, especially if these are not in a marized in Table II.
readily accessible reference.
(3) Samples ACKNOWLEDGMENTS
a. Emission angle of electrons.
The author acknowledges the “Metrology for Advanced
b. X-ray intensity: anode potential and current, power or flux.
Coatings and Formulated Products” theme of the UK National
c. Surface roughness or topography, if known.
Measurement System, funded by the Department of Business,
d. Number of areas analyzed and number of repeats.
Energy and Industrial Strategy (BEIS). Support, information, and
e. Any details of degradation or change during analysis.
advice were provided by Don Baer (PNNL), David Cant (NPL),
f. Reference materials and samples.
Steve Spencer (NPL), Andrew Pollard (NPL), and Dick Brundle
(4) Data
(C. R. Brundle and Associates).
a. Survey spectrum, with all detected elements identified.
b. Regions analyzed.
c. Width of regions and data point spacing. NOMENCLATURE
d. Type of background and method of end point ai = size of an atom of element i calculated using Mi = ρiNAa3i
determination. Ai,j = normalized intensity ratio of photoelectron peaks i and j

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-11
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

cps = counts per second 1. Inelastic mean free path

EAL = effective attenuation length
These are dealt with in detail within another article in this
Ei = kinetic energy of photoelectron peak i
series.66 Inelastic mean free paths (IMFPs) can be calculated using
Ii = background-subtracted peak area of photoelectron peak i
the TPP-2M formula67 with ∼10% accuracy. It requires the inputs
Ii,∞ = background-subtracted peak area of photoelectron peak
of electron kinetic energy, E, number of valence electrons per atom
i from a pure, homogeneous, flat reference material
or molecule, Nv, density of material in g cm−3, ρ, atomic number or
IMFP = inelastic mean free path
molecular weight of material, M, and the bandgap energy in eV, Eg.
ISO = International Organization for Standardization
A more recent expression,68 S1, given in Eq. (A1) has a similar
Li,M = effective attenuation length (EAL) of photoelectron
accuracy to TPP-2M and is easier to apply in the case of materials
peak i in material M
where some of the details are unknown. In this case, Z is the
Mi = relative atomic mass of element i
number averaged atomic number (Z = 4 for organic materials) and
NA = Avogadro constant
a is the atomic size in nm, which is typically 0.25 nm. The same
NIST = National Institute of Standards and Technology, USA
paper provides some even more general equations that can be used
NPL = National Physical Laboratory
in the case that both the composition and the IMFP require consis-
PTFE = polytetrafluoroethylene
tency during an automated analysis.
RSF = relative sensitivity factor
SCA = surface chemical analysis
Si = relative sensitivity factor of photoelectron peak i 2. Effective attenuation length
t = thickness of a uniform overlayer For practical purposes, the effective attenuation length (EAL)
X = atomic fraction of specified element, often expressed is more useful than the IMFP. There are a number of EALs gener-
as atomic percent ated for different purposes,66 but generally the ones that are used
XPS = x-ray photoelectron spectroscopy are those suitable for measuring overlayer thickness. They account
Zi = atomic number of element i for elastic scattering of electrons: the fact that electrons do not
always travel in straight lines.69,70 There are a number of relation-
Greek letters ships available that involve the use of a parameter called the single
scattering albedo, for which information is available for some mate-
β = angular asymmetry parameter for photoelectron
rials. The single scattering albedo depends upon the elemental
emission
composition of the material and the electron kinetic energy and,
λi,M = inelastic mean free path (IMFP) of photoelectron peak i
therefore, does not lend itself to simple analysis without extensive
in material M
databases. The EALs that result from such calculations are summa-
Φi = fractional surface coverage of atom i
rized in a database available from NIST.71 A relatively simple equa-
ρi = density of element i
tion is given in Eq. (A2) which has the same inputs as Eq. (A1)
σi = photoionization cross section of photoelectron peak i
and a slightly reduced accuracy.72
θ = electron emission angle relative to the surface normal

3. Substrate-overlayer intensities
APPENDIX
Equation (8) predicts the normalized intensity ratio for a
Some useful relationships for quantitative XPS are given below
substrate-overlayer system using the straight line approximation.
and in Table III.
Substrate Q generates electrons, q, with EAL Lq,P in overlayer mate-
rial P, which generates electrons, p, with EAL Lp,P in itself.
TABLE III. Some equations for electron inelastic mean free paths and effective
attenuation lengths. REFERENCES
1
Expression Comment M. P. Seah et al., Surf. Interface Anal. 41, 430 (2009).
2
P. Larsen and M. Von Ins, Scientometrics 84, 575 (2010).
3
(A1) S1, λ in nm D. R. Baer et al., J. Vac. Sci. Technol. A 37, 031401 (2019).
(4 þ 0:44Z 0:5 þ 0:104E0:872 )a1:7 4
λ¼ D. Briggs and M. P. Seah, Practical Surface Analysis: Auger and X-Ray
Z 0:3 (1  0:02Eg ) Photoelectron Spectroscopy (Wiley, Chichester, 1996).
5
J. F. Watts and J. Wolstenholme, An Introduction to Surface Analysis by XPS
and AES (Wiley, Chichester, 2003).
(A2) S3, L in nm 6
(5:8 þ 0:0041Z 1:7 þ 0:088E0:93 )a1:82 C. D. Easton, C. Kinnear, S. L. McArthur, and T. R. Gengenbach, J. Vac. Sci.
L¼ Technol. A 38, 023207 (2020).
Z 0:38 (1  0:02Eg ) 7
P. R. Davies and D. J. Morgan, J. Vac. Sci. Technol. A 38, 033204 (2020).
8
J. Wolstenholme, “A procedure which allows the performance and calibration
(A3) Overlayer-substrate of an XPS instrument to be checked rapidly and frequently,” J. Vac. Sci. Technol.
1  exp(t/Lp,P cos θ)
Ap,q ¼ A (to be published).
exp(t/Lq,P cos θ) 9
J. Wolstenholme, Surf. Interface Anal. 45, 1071 (2013).
10
M. P. Seah and G. C. Smith, Vacuum 41, 1601 (1990).
11
A. G. Shard et al., J. Phys. Chem. B 119, 10784 (2015).

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-12
Published under license by AVS.
 TUTORIAL avs.scitation.org/journal/jva

12 40
N. A. Belsey et al., J. Phys. Chem. C 120, 24070 (2016). A. G. Shard, J. Phys. Chem. C 116, 16806 (2012).
13 41
M. Seah, M. Jones, and M. Anthony, Surf. Interface Anal. 6, 242 (1984). A. G. Shard, J. Wang, and S. J. Spencer, Surf. Interface Anal. 41, 541 (2009).
14 42
M. P. Seah, J. Electron Spectr. Relat. Phenom. 71, 191 (1995). R. C. Chatelier, H. A. St John, T. R. Gengenbach, P. Kingshott, and
15
M. P. Seah and S. J. Spencer, J. Electron Spectr. Relat. Phenom. 151, 178 (2006). H. J. Griesser, Surf. Interface Anal. 25, 741 (1997).
16 43
M. Holzweber, A. Lippitz, R. Hesse, R. Denecke, W. S. Werner, and C. J. Powell, W. S. Werner, H. Kalbe, A. G. Shard, and D. G. Castner, J. Phys.
W. E. Unger, J. Electron Spectr. Relat. Phenom. 233, 51 (2019). Chem. C 122, 4073 (2018).
17 44
A. G. Shard and S. J. Spencer, Surf. Interface Anal. 51, 618 (2019). S. Evans and M. D. Scott, Surf. Interface Anal. 3, 269 (1981).
18 45
A. G. Shard, J. D. Counsell, D. J. Cant, E. F. Smith, P. Navabpour, X. Zhang, C. S. Fadley, Phys. Scr. 1987, 39 (1987).
46
and C. J. Blomfield, Surf. Interface Anal. 51, 763 (2019). D. Woodruff and A. Bradshaw, Rep. Prog. Phys. 57, 1029 (1994).
19 47
M. P. Seah, Surf. Interface Anal. 36, 1645 (2004). H. Bishop, Surf. Interface Anal. 17, 197 (1991).
20 48
M. P. Seah, I. S. Gilmore, and S. J. Spencer, J. Electron Spectr. Relat. Phenom. G. H. Major, N. Fairley, P. M. Sherwood, M. R. Linford, J. Terry,
104, 73 (1999). V. Fernandez, and K. Artyushkova, “Practical guide for curve fitting in X-ray
21
R. Wicks and N. Ingle, Rev. Sci. Instrum. 80, 053108 (2009). photoelectron spectroscopy,” J. Vac. Sci. Technol. A (submitted).
22 49
M. Seah, Surf. Interface Anal. 20, 865 (1993). D. A. Shirley, Phys. Rev. B 5, 4709 (1972).
23 50
M. P. Seah, J. Electron Spectr. Relat. Phenom. 100, 55 (1999). S. Tougaard, Surf. Sci. 216, 343 (1989).
24 51
M. P. Seah and I. S. Gilmore, Phys. Rev. B 75, 149901 (2007). S. Tougaard, Surf. Interface Anal. 25, 137 (1997).
25 52
M. P. Seah and I. S. Gilmore, Phys. Rev. B 73, 174113 (2006). M. P. Seah, Surf. Interface Anal. 24, 830 (1996).
26 53
C. Wagner, L. Davis, M. Zeller, J. Taylor, R. Raymond, and L. Gale, Surf. M. P. Seah, Surf. Sci. 420, 285 (1999).
54
Interface Anal. 3, 211 (1981). M. P. Seah, I. S. Gilmore, and S. J. Spencer, Surf. Sci. 461, 1 (2000).
27 55
E. F. Smith, F. J. Rutten, I. J. Villar-Garcia, D. Briggs, and P. Licence, S. B. Hill, N. S. Faradzhev, and C. J. Powell, Surf. Interface Anal. 49, 1187 (2017).
56
Langmuir 22, 9386 (2006). A. Herrera-Gomez, M. Bravo-Sanchez, O. Ceballos-Sanchez, and
28
C. R. Brundle and B. V. Crist, “XPS: A perspective on quantitation accuracy M. Vazquez-Lepe, Surf. Interface Anal. 46, 897 (2014).
57
for composition analysis of homogeneous materials,” J. Vac. Sci. Technol. A (to G. Beamson and D. Briggs, High Resolution XPS of Organic Polymers: The
be published). Scienta ESCA300 Database (Wiley, Chichester, 1992).
29 58
R. F. Reilman, A. Msezane, and S. T. Manson, J. Electron Spectr. Relat. E. Paparazzo, J. Phys. Condens. Matter 30, 343003 (2018).
59
Phenom. 8, 389 (1976). D. J. H. Cant, Y.-C. Wang, D. G. Castner, and A. G. Shard, Surf. Interface
30
S. Ray, R. T. Steven, F. M. Green, F. Hook, B. Taskinen, V. P. Hytonen, and Anal. 48, 274 (2016).
60
A. G. Shard, Langmuir 31, 1921 (2015). K. Piyakis, D.-Q. Yang, and E. Sacher, Surf. Sci. 536, 139 (2003).
31 61
N. A. Belsey, A. G. Shard, and C. Minelli, Biointerphases 10, 019012 (2015). G. Vereecke and P. Rouxhet, Surf. Interface Anal. 27, 761 (1999).
32 62
M. Edgell, R. Paynter, and J. Castle, J. Electron Spectr. Relat. Phenom. 37, 241 A. G. Shard and S. J. Spencer, Surf. Interface Anal. 49, 1256 (2017).
63
(1985). A. Cohen Simonsen, J. P. Pøhler, C. Jeynes, and S. Tougaard, Surf. Interface
33
J. H. Scofield, Report No. UCRL51326, Lawrence Livermore Laboratory Anal. 27, 52 (1999).
64
Report, CA, USA (1973). Y.-C. Wang, M. H. Engelhard, D. R. Baer, and D. G. Castner, Anal. Chem. 88,
34
L. Sabbatucci and F. Salvat, Radiat. Phys. Chem. 121, 122 (2016). 3917 (2016).
35 65
M. Trzhaskovskaya and V. Yarzhemsky, At. Data Nucl. Data Tables 119, 99 P. J. Cumpson, Surf. Interface Anal. 29, 403 (2000).
66
(2018). C. J. Powell, J. Vac. Sci. Technol. A 38, 023209 (2020).
36 67
M. P. Seah, I. S. Gilmore, and S. J. Spencer, J. Electron Spectr. Relat. Phenom. S. Tanuma, C. J. Powell, and D. R. Penn, Surf. Interface Anal. 21, 165 (1994).
68
120, 93 (2001). M. P. Seah, Surf. Interface Anal. 44, 497 (2012).
37 69
M. P. Seah, I. S. Gilmore, and S. J. Spencer, Surf. Interface Anal. 31, 778 A. Jablonski, Surf. Interface Anal. 14, 659 (1989).
70
(2001). A. Jablonski, J. Phys. D Appl. Phys. 48, 075301 (2015).
38 71
C. Battistoni, G. Mattogno, and E. Paparazzo, Surf. Interface Anal. 7, 117 C. J. Powell, see https://www.nist.gov/srd/nist-standard-reference-database-82
(1985). (2009).
39 72
A. G. Shard, Surf. Interface Anal. 46, 175 (2014). M. P. Seah, Surf. Interface Anal. 44, 1353 (2012).

J. Vac. Sci. Technol. A 38(4) Jul/Aug 2020; doi: 10.1116/1.5141395 38, 041201-13
Published under license by AVS.