Transcript CH915: Elemental Analysis

Session 3
Optical Spectroscopy:
Introduction/Fundamentals
Atomic and molecular spectroscopies
Instrumentation
Overview




Physical basis of absorption and emission
 Atomic spectra
 Molecular spectra
Instrumentation: components of optical
systems for spectrometers
Common techniques in atomic
spectroscopy: AAS and ICP-OES
Calibration
Useful websites for spectroscopy




http://www.shsu.edu/~chm_tgc/sounds/fl
ashfiles/ICPwCCD.swf
http://www.thespectroscopynet.com/Inde
x.html?/
http://teaching.shu.ac.uk/hwb/chemistry/t
utorials/molspec/
http://www.chemguide.co.uk/analysis/uvvi
siblemenu.html
See also individual citations on slides
3
Electromagnetic radiation
Spectroscopy = interactions between light & matter
E = hn = hcl
n = frequency; l = wavelength
http://www.spectroscopynow.com/coi/cda/detail.cda?id=18411&type=
EducationFeature&chId=7&page=1
This primer also contains a wavelength-energy converter
4
Fundamentals

Absorption and emission of light by compounds is
generally associated with transitions of electrons
between different energy levels
E2
DE2
E2
excited states
DE2
DE = hn = hc/l
E1
E1
DE1
E0
ground state
Emission: Sample (in an excited state)
produces light/looses energy
DE1
E0
Absorption: sample takes up energy
Consumes light of appropriate wavelength
Atomic spectra: line spectra provide specificity: each element has its
own pattern, as each element has its own electronic configuration
http://physics.nist.gov/PhysRefData/ASD/lines_form.html
5
Fundamentals

The population of different states is
given by the Boltzmann equation:
N1 g1
 e
N0 g0
 ΔE
kT
N0: number of atoms in ground state
N1: number of atoms in excited state
g1/g0 : weighting factors
Note: Equation contains temperature:
Excitation can be achieved by providing thermal
energy
6
Atomic emission:
Flame spectroscopy
Observation
Caused by...
Persistent golden-yellow
flame
Sodium
Violet (lilac) flame
Potassium, cesium
carmine-red flame
Lithium
Brick-red flame
Calcium
Crimson flame
Strontium
Yellowish-green flame
barium, molybdenum
Green flame
Borates, copper,
thallium
Blue flame (wire slowly
corroded)
Lead, arsenic,
antimony, bismuth,
copper
Qualitative method
Lithium
Cesium
Sodium
7
A simple spectroscope



Spectroscope: Device for
qualitative assessment of a
sample
E.g. used in flame analysis
E.g. used in gemmology
8
Atomic Spectroscopies - Synopsis
Techniques for determining the elemental composition of an
analyte by its electromagnetic or mass spectrum
Optical
spectroscopies
Fluorescence
Spectroscopy
AES
Others
See
table
Mass spectrometries
AAS
ICP-MS
SIMS
Others
(L. 6)
ICP-OES
Flame AAS
GFAAS
9
Atomic spectroscopies
Technique
Atomisation/Excitation
Sample etc
Arc/spark
e
Electric arc/spark
Solid sample on carbon
electrode
Laser microprobe
e
Laser
Solid sample on support
Glow discharge
e
Glow discharge
lamp
Solid sample disc
ICP-OES
e
Electromagnetic
induction
Liquid sample, sprayed
into gas plasma
Flame photometry
(atomic emission)
e
Flame
Liquid sample, sprayed
into flame
AAS
a
UV/Vis light
Liquid sample, sprayed
into flame or furnace
Atomic fluorescence
fe
UV/Vis light
Liquid sample, sprayed
into flame or furnace
X-ray fluorescence
fe
X-radiation
Solid or liquid
ICP-MS
-
n/a
Liquid sample, sprayed
into gas plasma
10
Atomic spectroscopies

Common principles:



Sample introduction:
Nebulisation, Evaporation
Atomisation (and excitation or
ionisation) by flame, furnace,
or plasma
Spectrometer components:



Light source (can be sample
itself - Only AA requires
external light source
Optical system (or mass
spectrometer)
Detector
11
Atomic spectra vs molecular spectra:

Lines
Bands
(nm)
Typical atomic spectrum
e.g. acquired by AAS
Two typical molecular spectra
Acquired by UV-Vis spectroscopy
Y axes: intensity of absorbed light. Under ideal conditions proportional
to analyte concentration (I  c; Beer’s law).
12
Origin of bands in molecular spectra
Molecules have chemical bonds
 Electrons are in molecular orbitals
 Absorption of light causes electron transitions
between HOMO and LUMO
 Molecules undergo bond rotations and
vibrations: different energy sub-states
occupied at RT and accessible through
absorption: many transitions possible:
A band is the sum of many lines

LUMO
Vibrational substates
rotational substates
HOMO
13
Quantitative analysis by molecular
absorption: Colorimetry

Because absorption spectroscopy is widely
applicable, sensitive (10-5-10-7 M), selective, accurate
(0.1-3% typically), and easy:



95% of quantitative
analyses in field of health
performed with UV/Vis tests
Hemoglobin in blood
First step in analysis: establish working conditions



Select l
Selection, cleaning and handling of cells
Calibration: determine relationship between absorbance and
14
concentration
Instrument components
AAS Spectrometer
Light
Source
Monochromator
Sample
Detector
Readout/Data
system
ICP-OES Spectrometer
Sample =
light
source
Monochromator
Detector
Readout/Data
system
UV-Vis Spectrometer:
Light
Source
Monochromator
Sample
Detector
Readout/Data
system
15
UV-Vis spectrophotometer (dual beam)
Monochromator
Slit
Diffraction
grating
Mirror
Slit
Light sources
Filter
Reference
HalfMirror
Detector
Mirror
Sample
http://www.spectroscopynow.com/coi/cda/detail.cda?i
d=18412&type=EducationFeature&chId=7&page=1
16
Example for a dual beam spectrometer
17
Single beam
18
UV-Vis spectroscopy
practicalities: Referencing


Matrix (solvent, buffer etc) might also
have absorbance: Must be taken care of
In dual beam:



Simultaneous measurement of reference cell
eliminates absorbance of background
Recording of baseline recommended
Single beam:


Requires measurement of reference spectrum,
can be subtracted from sample spectrum
Preferentially in same cuvette
19
Light sources
Wavelength(nm)
100
200
VAC
400
UV
700
Visible
2000
Near IR
4000
7000
10,000
IR
20,000
40,000
Far IR
Spectral region
Continuum
Ar lamp
Xe lamp
D2 lamp
Tungsten lamp
Nernst glower (ZrO2 + Y2O3)
Line
Nichrome wire
Globar (SiC)
Hollow cathode lamps
Lasers
20
Example of a continuum source:
Output from Tungsten lamp
Widely applied in UV-Vis spectrometers
21
Hollow cathode lamp
Used in AAS
 Filled with Ne or Ar at a pressure of 130-700 Pa (1-5
Torr).
 When high voltage is applied between anode and
cathode, filler gas becomes ionised
 Positive ions accelerated toward cathode
 Strike cathode with enough energy to "sputter" metal
atoms from the cathode
to yield cloud with
excited atoms
• Atoms emit line spectra

22
Example: Output from iron hollow
cathode lamp



Small portion of spectrum
from Fe hollow cathode lamp
Shows sharp lines
characteristic of gaseous
atoms
Linewidths are artificially
broadened by
monochromator (bandwidth
= 0.08 nm)
23
Wavelength selectors: dispersive
elements and filters
Wavelength(nm)
100
200
VAC
UV
400
700
Visible
2000
4000
Near IR
7000
10,000
IR
20,000
40,000
Far IR
Spectral region
Fluorite prism
Fused silica or quartz prism
Continuous
Glass prism
NaCl prism
KBr Prism
3000 lines/nm
Gratings
50 lines/nm
Interference filters
Discontinuous
Interference wedge
Glass filters
24
Monochromators

Consist of






Entrance slit
Collimating lens or mirror
Dispersion element (prism or grating)
Focusing lens or mirror
Exit slit
Czerny-Turner grating monochromator:
Mirrors
Common in UV-Vis spectrometers
25
Dispersers

Separate polychromatic light into its
components


Prism
Diffraction grating: patterned surface which
diffracts light
Prisms
Blazed diffraction
grating
Holographic
grating
26
Echellette grating:



Extra pathlength travelled by wave 2 must
be multiple of l for positive interference:
nl = d(sin i + sin r)
for UV 1000-2000 lines/mm: d = 0.5-1 mm
echelle: French for ladder
27
Bandwidth of a monochromator



Spectral bandwidth: range
of wavelengths exiting the
monochromator
Related to dispersion and
slit widths
Defines resolution of
spectra: 2 features can only
be distinguished if effective
bandwidth is less than half
the difference between the l
of features
28
Effect of slit width on peak heights
29
Components of optical system in
an ICP-OES spectrometer





spherical and cylindrical lenses
flat and spherical mirrors
parallel planes
optical path under vacuum or
controlled nitrogen atmosphere
(necessary for wavelengths <200
nm; air absorbs far UV light)
Disperser(s)
30
Old models: Sequential type
Can only measure one wavelength at a given time: Slow
31
Newer: Simultaneous type
CCD detector: 2D detector
Echelle cross disperser
(polychromator):
Consists of Echelle
grating and prisms/
echellette: separates
lights in 2 dimensions
This combination allows
high-speed measurement,
providing information on
all 72 measurable
elements within 1 to 2
minutes
32
Detectors
Wavelength(nm)
100
200
VAC
400
UV
700
Visible
2000
4000
Near IR
7000
10,000
IR
20,000
40,000
Far IR
Spectral region
Photographic plate
Photomultiplier
Photon
detectors
Phototube
Photocell
Silicon diode
Charge-coupled device (170-1000)
Thermal
detectors
Photoconductor
Thermocouple
Golay pneumatic cell
Pyroelectric cell
33
Photomultiplier: detects one
wavelength at a time






Based on photoelectric
effect
Photocathode and
series of dynodes in an
evacuated glass
enclosure
Photons strike cathode and electrons are emitted
Electrons are accelerated towards a series of dynodes
by increasing voltages
Additional electrons are generated at each dynode
Amplified signal is finally collected and measured at
anode
34
Photodiode arrays: measure several
wavelengths at once


linear array of discrete photodiodes on an integrated
circuit (IC) chip
Photodiode: Consists of 2 semiconductors (n-type and
p-type)



Light promotes electrons into conducting band: generates
electron-hole pair
“Concentration” of these electron-hole pairs directly
proportional to incident light
a voltage bias is present and the concentration of lightinduced electron-hole pairs determines the current through
35
semiconductor
Detection in simultaneous ICP-OES:
CCD:
Charge-coupled device
• Also integrated-circuit chip
• Contains an array of capacitors that store charge
when light creates electron-hole pairs
• Accumulated charge is read out at given time
interval
• Each wavelength is detected at a different spot
• Much more sensitive than photodiode array
detectors
http://www.chemistry.adelaide.edu.au/external/soc-rel/content/ccd.htm
36
Lecture 4
AAS and ICP-OES
Sample preparation
Interferences
Calibration
37
Crucial steps in atomic spectroscopies
and other methods
Laser ablation etc.
Solid/liquid
sample
Nebulisation
Solution
Sample
preparation
M+
X-
Desolvation
M+
Atoms in gas
phase
Ionisation
MX(g)
Vaporisation
M(g) + X(g)
Sputtering, etc.
Molecules in
gas phase
Atomisation=
Dissociation
Excitation
Ions
 ICP-MS and
other MS methods
Adapted from www.spectroscopynow.com (Gary Hieftje)
Excited
Atoms
 AAS and AES,
X-ray methods
38
Sample Introduction: liquid samples



Often the largest source of noise
Sample is carried into flame or plasma as aerosol,
vapour or fine powder
Liquid samples introduced using nebuliser
39
Sample preparation for analysis in
solution: Digestion




Digestion in conc. HNO3 and mixtures
thereof (e.g. aqua regia)
Br2 or H2O2 can be added to conc. acids
to give a more oxidising medium and
increase solubility
Certain materials require digestion in
conc. HF
Common to use microwave digestion
40
Microwave digestion
Rotor
Supplied with dedicated vessels (e.g. PTFE)
Closed vessel digestion minimises sample contamination
Faster, more reproducible, and safer than conventional
41
methods
Sample preparation and sample
handling for trace analysis



As always – sample preparation is key
Ultra-trace: Contaminations introduced during
sample processing can seriously limit performance
characteristics
Points to consider:





Purity of reagents
Chemical inertness of reaction vessels and any other material
samples come into contact with
Working environment
Preparation of standards and blanks crucial
Also measure a “process blank”:

Important for determination of LOD and LOQ
42
Common Units in trace analysis





ppm, ppb, ppt, ppq…..: parts per million etc.
ppm: mg/kg; often also used as mg/L
ppb: mg/kg
ppt: ng/kg
ppq: pg/kg
43
Atomic absorption
spectroscopy
44
Atomic Absorption Spectroscopy



Flame AAS has been the most widely used of all atomic
methods due to its simplicity, effectiveness and low cost
First introduced in 1955, commercially available since
1959
Qualitative and quantitative analysis of >70 elements
 Quantitative: Can detect ppm, ppb or even less
 Rapid, convenient, selective, inexpensive
H
Li
He
Be
B
C
N
O
F
Ne
Na Mg
Al
Si
P
S
Cl
Ar
K
Ca
Sc
Ti
V
Rb
Sr
Y
Zr
Cs
Ba
La
Hf
Fr
Ra
Ac
Cr Mn Fe
Co
Ni
Cu Zn
Ga Ge
As
Se
Br
Kr
Nb Mb Tc
Ru Rh
Pd
Ag Cd
In
Sn
Sb
Te
I
Xe
Ta
Os
Pt
Au Hg
Tl
Pb
Bi
Po
At
Rn
W
Re
Ir
45
Flame AA Spectrometer
Hollow cathode lamps with
characteristic emissions
Burner
Flame fuelled by (e.g.)
acetylene and air
Nebuliser and
Spray chamber
Hollow cathode lamps available for over 70 elements
Can get lamps containing > 1 element for determination
of multiple species
46
Schematic
I0
Light Source
E.g. Hollow
cathode lamp
Analyte solution
It
Monochromator
Atomiser
Detector
Amplifier
Fuel (e.g. acetylene)
Air
Nebuliser, spray
chamber, and
burner
47
Flame atomisation:
Laminar flow burner - components






Nebuliser: converts sample solution into aerosol
Spray chamber: Aerosol mixed with fuel, oxidant and burned in
5-10 cm flame
Fuel: Acetylene or nitrous oxide
Oxidant: Air or oxygen
Burner head:
Laminar flow: quiet
flame and long pathlength
But: poor sensitivity
(not very efficient
method, most of
sample lost)
from: Skoog
48
Structure of a flame

Relative size of
regions varies with
fuel, oxidant and
their ratio
49
Electrothermal atomisation: GFAAS

Provides enhanced sensitivity



entire sample atomised in very
short time
atoms in optical path for a second
or more
(flame 10-4 s)
Device: Graphite furnace
50
Sensitivity and detection limits in
AAS


Sensitivity: number of ppm of an element to give 1%
absorption.
Limit of detection: dependent upon signal:noise ratio:
S/N   Light intensity reaching detector
S/N=3.2
51
Interferences in AAS


Broadening of a spectral line, which can occur due to
a number of factors (Physical)
Spectral: emission line of another element or
compound, or general background radiation from the
flame, solvent, or analytical sample




Background correction can be applied
Chemical: Formation of compounds that do not
dissociate in the flame
Ionisation of the analyte can reduce the signal
Matrix interferences due to differences between
surface tension and viscosity of test solutions and
standards
Another caveat: Non-linear response common in AAS
52
Physical interferences:
Atomic line widths/ line shapes



Very important in atomic spectroscopy
Narrow lines increase precision, decrease
spectral interferences
Lines are broadened
by several mechanisms:



Natural broadening
Doppler effect
Pressure broadening
Figure taken from
http://www.cem.msu.edu/~cem333/
Week03.pdf
53
Natural linewidths



Width of an atomic spectral line is
determined by the lifetime of the excited
state
Consequence of the Heisenberg
uncertainty principle
For example, lifetime of 10-8 seconds (10
ns) yields peak widths of 10-5 nm
54
Doppler Effect
Photon detector



Due to rapid motion of atoms in gas phase
Atom moving toward the detector absorbs / emits radiation
of shorter l than atom moving perpendicular to detector.
Atom moving away from the detector absorbs / emits
radiation of longer l: detector perceives fewer oscillations
55
Pressure broadening

Results from collisions of absorbing/emitting species





With analyte atoms or combustion products of fuel
Deactivates the excited state – shorter lifetime - broader
spectral lines
Increases with concentration and temperature
E.g. in flame, Na absorbance lines broadened up to 10-3 nm.
Doppler and pressure effects broaden atomic lines by
1-2 orders of magnitude as compared with their
natural linewidths
56
Background correction in AAS


particularly important in GFAAS
Use beam chopper to distinguish the signal due to flame
from desired atomic line at the same wavelength (old
method)
Lamp and flame emission
reach detector
Only flame emission
reaches detector
Resulting signal
57
Background correction in AAS

High energy Deuterium background corrector
Detector
Hollow
cathode
lamp
Lamps are
pulsed out of
phase with
each other
Beam
combiner
Sample
Deuterium lamp
58
Minimising the effect of
Matrix Interferences



The term "matrix" refers to the sum of all compositional
characteristics of a solution, including its acid
composition
Calibration standards
and samples must be
matrix-matched in
terms of composition,
total dissolved solids,
and acid concentration
of the solution
Also advisable for
ICP-OES and -MS
Effect on K concentration on measured Sr
59
Specialised applications in AAS:
Flameless cold vapour methods


Mercury: has sufficient vapour pressure at RT
Hydride generation technique for determination
of As, Sb, Bi, Se, Te, Ge, Pb, and Sn




Generation of volatile metal hydrides (As, Sb, Bi, Se, Te,
Ge, Pb, and Sn)
Reduction by NaBH4 to form volatile hydride (e.g. SnH4)
Hydrides carried into light path by argon gas
Decomposed into elemental vapour by injection into
(electrothermally) heated silica cell
60
Calibration – some practical
aspects
61
Principles





Recap: Measured quantity must change with
analyte concentration in systematic and defined
way
Can be determined by calibration, using defined
standards
Stock solutions of standards can either be
prepared or purchased
Working solutions are best prepared by weighing
the amounts of stock solution and matrix (rather
than using volumetric ware)
NEVER extrapolate: concentration of sample must
be in same range as standards
62
Calibration in AAS
In theory, Beer’s law applies
for dilute solutions
 In practice, deviation from
linearity is usual
Small dynamic range
Possible to use non-linear
curve fitting for calibration
 Reasons: Self-absorption:


excited atoms emit light that
can also be absorbed instead of
that of source:  on average,
less light per number of atoms
is absorbed
Linear range
63
Alternative to matrix-matching:
Method of standard additions


Extensively used in absorption
spectroscopy, accounts for matrix effects
Several aliquots of sample
Sample (1): diluted to volume directly
 Samples (2,3,4,5…): known amounts of analyte
added before dilution to volume


BUT: Only makes sense if the added
standard closely matches the analyte
present in the samples chemically and
physically

 if simple, dissolved ions are analysed
64
Method of standard additions

If linear relationship exists between measured quantity and
concentration (must be verified experimentally) then:
kVx c x kVs cs
AT 

VT
VT






Vx, Cx: volume and concentration of analyte
Vs: variable volume of added standard
Cs: concentration of added standard
VT: total volume of volumetric flask
k: proportionality constant (= єl)
Ax, AT: absorbances of standard alone and sample + standard addition,
respectively.
65
Method of standard additions
Graphical
evaluation
slope = m = (єlcs) / VT
intercept = b = (єlVxcx) / VT
Limitations
• The calibration graph must be substantially linear since accurate
regression cannot be obtained from non-linear calibration points.
• Caution: The fact that the measured part of the graph is linear does not
always mean that linear extrapolation will produce the correct results
• It is also essential to obtain an accurate baseline from a suitable reagent
blank
66
Most simple version of standard
addition: spiking


Spiking means deliberately adding analyte to
an unknown sample
Involves:





preparation of sample and measurement of
absorbance
Addition of standard with known concentration,
measurement of absorbance
From difference in absorbance, calculate e
From reading of sample alone, calculate amount
of analyte
(use Beer’s law for calculations)
67
Other uses for spiking




Add spike at beginning of sample
preparation
Process sample with and without spike
Difference should correspond to amount
spiked
Deviation allows to calculate recovery
factor
68
Atomic emission
spectroscopy
69
Atomic emission spectroscopy


Historically, many techniques based on
emission have been used (See Table on p. 4)
Flame and electrothermal methods now
widely superseded by Inductively-Coupled
Plasma (ICP) method


Developed in the 1970s
Higher energy sources than flame or
electrothermal methods
70
ICP-AES/OES
Inductively coupled plasma-atomic emission spectroscopy
(or optical emission spectroscopy)

Offer several advantages over flame/electrothermal:






Lower inter-element interference (higher temperatures)
With a single set of conditions signals for dozens of
elements can be recorded simultaneously
Lower LOD for elements resistant to decomposition
Permit determination of non-metals (Cl, Br, I, S)
Can analyse concentration ranges over several decades (vs 1
or 2 decades for other methods)
Disadvantages:


More complicated and expensive to run
Require higher degree of operator skill
71
Modern ICP-OES spectrometer




Over 70 elements (in principle simultaneously)
Including non-metals such as sulfur, phosphorus,
and halogens (not possible with AAS)
ppm to ppb range
Principle: Argon plasma generates excited atoms
and ions; these emit characteristic radiation
72
ICP-AES Instrumentation
73
Components for sample injection
and the ICP torch
Up to 7000°C
www.cleanwatertesting.com/news_
NR149.htm
www.midwestrefineries.com
/refiningandassaying.htm
74
Meinhard nebuliser
Caution: The capillary is easy to block and difficult to unblock
75
ICP torch
water cooled induction
coil powered by RF
generator (2 kW power
at 27 MHz)
concentric quartz tubes
11-17 L/min
d=2.5 cm
76
Torch Ignition Sequence
Ionisation of
Argon initiated by
spark from Tesla
coil
Start gas flow
Switch on RF power
After leaving injector, sample
moves at high velocity
Punches hole in centre of
plasma
Plasma generated
77
Atomisation / Ionisation

In plasma, sample moves through several zones



Preheating zone (PHZ): temp = 8000 K:
Desolvation/evaporation
Initial radiation zone (IRZ): 6500-7500 K: Vaporisation,
Atomisation
Normal analytical zone (NAZ): 6000-6500 K: Ionisation
78
Advantages of plasma

Prior to observation, atoms spend ~ 2 sec at
4000-8000 K (about 2-3 times that of hottest
combustion flame)



Chemically inert environment for atomisation


Atomisation and ionisation is more complete
Fewer chemical interferences
Prevents side-product (e.g. oxide) formation
Temperature cross-section is uniform (no cool
spots)


Prevents self-absorption
Get linear calibration curves over several orders of
magnitude
79
Radial and axial observation
Axial
Radial. Can achieve higher
sensitivity
Combined viewing expands dynamic range
http://las.perkinelmer.com/content/relatedmaterials/brochures/bro
80
_atomicspectroscopytechniqueguide.pdf
Applications

ICP-OES used for quantitative analysis of:

Soil, sediment, rocks, minerals, air








Geochemistry
Mineralogy
Agriculture
Forestry
Fornensics
Environmental sciences
Food industry
Elements not accessible using AAS

Sulfur, Boron, Phosphorus, Titanium, and Zirconium
81
Homework for revision

Read
http://las.perkinelmer.com/content/
relatedmaterials/brochures/bro_ato
micspectroscopytechniqueguide.pdf
82
Lab Experiment 3

Analyse a Chromium complex for [Cr] in three ways:




UV (absorbance & extinction coefficient)
Titration (moles Cr and charge)
AAS (Cr standard curve and unknown concentration)
AAS data analysis

Fit standards to quadratic equation


A=a[Cr]2 + b[Cr] + c
Use a, b, and c to calculate unknown concentration
83