CEM 333

Instrumental Analysis
Simon J. Garrett
Room: CEM 234
Phone: 355 9715 ext 208
E-mail: [email protected]
Lectures: Tuesday, Thursday 9:00-9:50 am Room 136
Office Hours: Tuesdays 10:00-11:00 am
CEM 333 page 1.1
Course Objectives
Teach fundamentals of instrumental analysis
Lecture: Discuss theory and background for
(1) chemical/physical property measured
(2) origin of chemical/physical property
(3) instrument design and nature of response
(4) signal processing and relationship between readout
to property measured
Laboratory: Provides hands-on experience in
(1) relating lecture material to practical analysis
(2) design and operation of a real instrument
(3) measurements on range of instruments
(4) example analyses to illustrate value of technique
CEM 333 page 1.2
Introduction (Chapter 1)
Classification of Analytical Methods
Qualitative instrumental analysis is that measured property
indicates presence of analyte in matrix
Quantitative instrumental analysis is that magnitude of measured
property is proportional to concentration of analyte in matrix
Species of interest
All constituents
including analyte.
Matrix-analyte
=concomitants
Often need pretreatment - chemical extraction, distillation,
separation, precipitation
(A) Classical:
Qualitative - identification by color, indicators, boiling points,
odors
Quantitative - mass or volume (e.g. gravimetric, volumetric)
(B) Instrumental:
Qualitative - chromatography, electrophoresis and identification
by measuring physical property (e.g. spectroscopy, electrode
potential)
Quantitative - measuring property and determining relationship
to concentration (e.g. spectrophotometry, mass spectrometry)
Often, same instrumental method used for qualitative and
quantitative analysis
CEM 333 page 1.3
Types of Instrumental Methods:
Property Example Method
Radiation emission Emission spectroscopy - fluorescence,
phosphorescence, luminescence
Radiation absorption Absorption spectroscopy -
spectrophotometry, photometry, nuclear
magnetic resonance, electron spin
resonance
Radiation scattering Turbidity, Raman
Radiation refraction Refractometry, interferometry
Radiation diffraction X-ray, electron
Radiation rotation Polarimetry, circular dichroism
Electrical potential Potentiometry
Electrical charge Coulometry
Electrical current Voltammetry - amperometry, polarography
Electrical resistance Conductometry
Mass Gravimetry
Mass-to-charge ratio Mass spectrometry
Rate of reaction Stopped flow, flow injection analysis
Thermal Thermal gravimetry, calorimetry
Radioactivity Activation, isotope dilution
(Often combined with chromatographic or electrophoretic methods)
CEM 333 page 1.4
Stimulus Response
Analyte
(in matrix)
Energy Analytical
Data
encoded
information
Example:
Spectrophotometry
Instrument: spectrophotometer
Stimulus: monochromatic light energy
Analytical response: light absorption
Transducer: photocell
Data: electrical current
Data processor: current meter
Readout: meter scale
Data Domains: way of encoding analytical response in electrical or
non-electrical signals.
Interdomain conversions transform information from one domain to
another.
Light Intensity
Photocell
Current
Current Meter
Scale
Detector (general): device that indicates change in environment
Transducer (specific): device that converts non-electrical to
electrical data
Sensor (specific): device that converts chemical to electrical data
CEM 333 page 1.5
Non-Electrical Domains Electrical Domains
Physical (light intensity, color) Current
Chemical (pH) Voltage
Scale Position (length) Charge
Number (objects) Frequency
Pulse width
Phase
Count
Serial
Parallel
Time - vary with time (frequency, phase, pulse width)
Analog - continuously variable magnitude (current, voltage, charge)
Digital - discrete values (count, serial, parallel, number*)
CEM 333 page 1.6
Digital Binary Data
Advantages (1) easy to store (2) not susceptible to noise
2
2
1st bit 2nd bit 3rd bit 4th bit
Hi
Lo
Count
Serial
2
0
2
1
2
0
2
1
2
2
Parallel
Hi
Lo
Hi
Lo
Hi
Lo
Hi
Lo
3 separate
signals
2
0
=1, 2
1
=2, 2
2
=4...
CEM 333 page 1.7
Performance Characteristics: Figures of Merit
How to choose an analytical method? How good is measurement?
How reproducible? - Precision
How close to true value? - Accuracy/Bias
How small a difference can be measured? - Sensitivity
What range of amounts? - Dynamic Range
How much interference? - Selectivity
CEM 333 page 1.8
Precision - Indeterminate or random errors
Absolute standard deviation: s =
x
i
x ( )
2
i0
i= N

N1
Variance: s
2
Relative standard deviation: RSD =
s
x
Standard deviation of mean: s
m
=
s
N
Accuracy - Determinate errors (operator, method, instrumental)
Bias: bias = x x
true
Sensitivity
Calibration sensitivity:
S=
dSignal
dc
c + Signal
blank
= mc + Signal
blank
(larger slope of calibration curve m, more sensitive measurement)
Detection Limit
Signal must be bigger than random noise of blank
Minimum signal: Signal
min
= Av. Signal
blank
+ ks
blank
From statistics k=3 or more (at 95% confidence level)
CEM 333 page 1.9
Dynamic Range
At detection limit we can say confidently analyte is present but
cannot perform reliable quantitation
Level of quantitation (LOQ): k=10
Limit of linearity (LOL): when signal is no longer proportional to
concentration
Dynamic range:
LOL
LOQ
10
2
to > 10
6
Selectivity:
No analytical method is completely free from interference by
concomitants. Best method is more sensitive to analyte than
interfering species (interferent).
Matrix with species A&B: Signal = m
A
c
A
+ m
B
c
B
+ Signal
blank
Selectivity coefficient: k
B,A
=
m
B
m
A
k's vary between 0 (no selectivity) and large number (very
selective).
CEM 333 page 1.10
Calibration methods
Basis of quantitative analysis is magnitude of measured property is
proportional to concentration of analyte
Signal [x] or Signal = m[x]+ Signal
blank
[x] =
Signal Signal
blank
m
Calibration curves (working or analytical curves)
Instrument
Response
(Signal)
blank
Signal
DL
LOQ
LOL
Dynamic Range
[x]
Slope m
CEM 333 page 1.11
Example (if time):
Analyte Concentration (ppm*) Absorbance
0.0 (blank) 0.05
0.9 0.15
2.0 0.24
3.1 0.33
4.1 0.42
*ppm=1 g per L
Define Variance and Covariance:
S
xx
=
(x
i
x )
2

N 1
S
xy
=
(x
i
x )(y
i
y )
N 1
x = 2.02 y = 0.238
S
xx
=
2.02
2
+1.12
2
+ 0.02
2
+1.08
2
+ 2.08
2
( )
4
=
10.828
4
= 2.707
S
xy
=
2.02 0.188 ( ) + 1.12 0.088 ( ) + 0.02 0.002 ( )+...
4
=
0.9562
4
= 0.23905
Slope: m =
S
xy
Sx
=
0.23905
2.707
= 0.0883
Intercept:
b = y mx
= 0.238 (0.0883 2.02)
= 0.0596
Calibration expression is
Absorbance=0.0883[Analyte (ppm)]+0.0596
CEM 333 page 1.12