Elemental Analysis - Atomic Spectroscopy
A) Introduction
Based on the breakdown of a sample into atoms, followed by the measurement of the
atom’s absorption or emission of light.
i. deals with absorbance fluorescence or emission (luminescence) of atoms or
elemental ions rather then molecules
- atomization: process of converting sample to gaseous atoms or
elementary ions
ii. Provides information on elemental composition of sample or compound
- UV/Vis, IR, Raman gives molecular functional group information, but no
elemental information.
iii. Basic process the same as in UV/Vis, fluorescence etc. for molecules
h
E1
Eo
Absorbance
Fluorescence
iv. Differences for Molecular Spectroscopy
- no vibration levels  much sharper absorbance, fluorescence, emission
bands
- position of bands are well-defined and characteristic of a given element
- qualitative analysis is easy in atomic spectroscopy (not easy in molecular
spectroscopy)
Examples:
carbon
oxygen
nitrogen
B) Energy Level Diagrams
energy level diagram for the outer electrons of an element describes atomic spectroscopy
process.
i. every element has a unique set of atomic orbitals
ii. p, d, f split by spin-orbit coupling
iii. Spin (s) and orbital (l) motion create magnetic fields that perturb each other (couple)
- parallel  higher energy; antiparallel  lower energy
• Similar pattern between atoms but
different spacing
• Spectrum of ion different to atom
• Separations measured in
electronvolts (eV)
1eV = 1.602x10-19 J
= 96.484 kJ mol-1
Na
Mg+
• As number of electrons increases,
number of levels increases
emission spectra more complex
Li 30 lines
Cs 645 lines
Cr 2277 lines
Note slight differences in energy due to
magnetic fields caused by spin
C) Desire narrow lines for accurate identification
Broadened by
i. uncertainty principle
Uncertainty principal:
\
Dt . DE ≥ h
Dt . D ≥ 1
Dt – minimum time for measurement
D – minimal detectable frequency difference
Peak line-width is defined as width in wavelength at half the signal intensity
C) Desire narrow lines for accurate identification
Broadened by
ii. Doppler effect
Doppler effect
- emitted or absorbed wavelength changes as a result of
atom movement relative to detector
- wavelength decrease if motion toward receiver
- wavelength increases if motion away from receiver
Usage in measurement of velocity of galaxies, age of universe and big bang theory
C) Desire narrow lines for accurate identification
Broadened by
iii. Pressure broadening
Pressure broadening:
Collisions with atoms/molecules transfers small quantities of vibrational energy
(heat) - ill-defined ground state energy
Effect worse at high pressures:
• For high pressure Xe lamps (>10,000 torr) turns lines into continua!
D) Effect of Temperature on Atomic Spectra
- temperature changes number of atoms in ground and excited states
- need good temperature control
Boltzmann equation
N1 and No – are the number of atoms in excited and ground states
k – Boltzmann constant (1.28x10-23 J/K)
T – temperature
DE – energy difference between ground and excited states
P1 and Po – number of states having equal energy at each quantum level
Na atoms at 2500 K, only 0.02 % atoms in first excited state!
Less important in absorption measurements - 99.98 % atoms in ground state!
E) Sample Atomization – expose sample to flame or high-temperature
i.
Need to break sample into atoms to observe atomic spectra
ii.
Basic steps:
a) nebulization – solution sample, get into fine droplets by spraying thru thin nozzle or
passing over vibrating crystal.
b) desolvation - heat droplets to evaporate off solvent just leaving analyte and other
matrix compounds
c) volatilization – convert solid analyte/matrix particles into gas phase
d) dissociation – break-up molecules in gas phase into atoms.
e) ionization – cause the atoms to become charged
f) excitation – with light, heat, etc. for spectra measurement.
E) Sample Atomization – expose sample to flame or high-temperature
iii.
Types of Nebulizers and Atomizers
F) Atomic Absorption Spectroscopy (AAS)
– commonly used for elemental analysis
– expose sample to flame or high-temperature
– characteristics of flame impact use of atomic absorption spectroscopy
Flame AAS:
• simplest atomization of gas/solution/solid
• laminar flow burner - stable "sheet" of flame
• flame atomization best for reproducibility (precision) (<1%)
• relatively insensitive - incomplete volatilization, short time in flame
i.
Different mixes and flow rates give different temperature profile in flame
- gives different degrees of excitation of compounds in path of light source
ii. Types of Flame/Flame Structure – selection of correct flame region is important for
optimal performance
a) primary combustion zone – blue inner cone (blue due to emission from C2, CH &
other radicals)
- not in thermal equilibrium and not used
b) interconal region
- region of highest temperature (rich in free atoms)
- often used in spectroscopy
- can be narrower in some flames (hydrocarbon) tall in others (acetylene)
c) outer cone
Temperature varies across flame –
- cooler region
need to focus on correct part of flame
- rich in O2 (due to surrounding air)
- gives metal oxide formation
Primary region
for spectroscopy
Not in thermal equilibrium
and not used for
spectroscopy
Flame profile: depends on type of fuel
and oxidant and mixture ration
Most sensitive part of flame for AAS varies with analyte
Consequences:
- Sensitivity varies with element
- must maximize burner position
- makes multi-element detection difficult
iii. Basic instrument design (Flame atomizer)
Single beam
Double beam
a) atomizer
1) Laminar Flow Burner
- adjust fuel/oxidant mixture for optimum excitation of desired
compounds
- usually 1:1 fuel/oxidant mix but some metals forming oxides use
increase fuel mix
- different mixes give different temperatures.
Laminar – non-turbulent streamline flow
i.
ii.
sample, oxidant and fuel are mixed
only finest solution droplets reach
burner
iii. most of sample collects in waste
iv. provides quite flame and a long path
length
2) Electrothermal (L’vov or Graphite furnace)
- place sample drop on platform inside tube
- heat tube by applying current, resistance to current creates heat
- heat volatilizes sample, atomizers, etc. inside tube
- pass light through to measure absorbance
Po
P
Place sample
droplet on platform
3) Comparison of atomizers
a) Electrothermal (L’vov or Graphite furnace) :
advantages:
- all sample used
- longer time of sample in light beam
 lower limit of detection (LOD)
 can use less sample (0.5 – 10)
disadvantage:
- slow (can be several minutes per element or sample)
- not as precise as flame (5-10% vs. 1%)
- low dynamic range (< 102, range of detectable signal intensity)
\ use only when there is a need for better limit of detection or
have less sample than Laminar flow can use
b) Laminar Flow Burner
advantages:
- good b (5-10 cm)
- good reproducibility
disadvantages:
- not sample efficient (90-99% sample loss before flame)
- small amount of time that sample is in light path (~10-4 s)
- needs lots of sample
b) Light source
- need light source with a narrow bandwidth for light output
- AA lines are remarkably narrow (0.002 to 0.005 nm)
- separate light source and filter is used for each element
1) problem with using typical UV/Vis continuous light source
- have right l, but also lots of others (non-monochromatic light)
- hard to see decrease in signal when atoms absorb in a small bandwidth
- only small decrease in total signal area
- with large amount of elements  bad sensitivity
2) Solution is to use light source that has line emission in range of interest
- laser – but hard to match with element line of interest
- hollow cathode lamp (HCL) is common choice
Hollow Cathode Lamp
Coated with element
to be analyzed
Process: use element to detect element
1. ionizes inert gas to high potential (300V)
Ar  Ar+ + e2. Ar+ go to “-” cathode & hit surfaces
3. As Ar+ ions hit cathode, some of deposited element is excited and
dislodged into gas phase (sputtering)
4. excited element relaxes to ground state and emits characteristic radiation
- advantage: sharp lines specific for element of interest
- disadvantage: can be expensive, need to use different lamp for each element tested.
c) Source Modulation (spectral interference due to flame)
- problem with working with flame in AA is that light from flame and light source
both reach detector
- measure small signal from large background
- need to subtract out flames to get only light source signal (P/Po)
i. done by chopping signal:
Flame + P
Flame only
P
Flame + P
ii. or modulating P from lamp:
Flame only
time
d) Corrections For Spectral Interferences Due to Matrix
- molecular species may be present in flame
- problem if absorbance spectra overlap since molecular spectrum is much
broader with a greater net absorbance
- need way of subtracting these factors out
Methods for Correction
1) Two-line method
- monitor absorbance at two l close together
> one line from sample one from light source
> second l from impurity in HCL cathode, Ne or Ar gas in HCL, etc
- second l must not be absorbed by analyte
> absorbed by molecular species, since spectrum much broader
- A & e are ~ constant if two l close
- comparing Al1, Al2 allows correction for absorbance for molecular species
Al1 (atom&molecule) – Al2 (molecule) = A (atom)
Problem: Difficult to get useful second l with desired characteristics
2) Continuous source method
- alternatively place light from HCL or a continuous source D2 lamp thru flame
- HCL  absorbance of atoms + molecules
- D2  absorbance of molecules
advantage:
-available in most instruments
-easy to do
disadvantage:
-difficult to perfectly match lamps (can give + or – errors)
3) Zeeman Effect
- placing gaseous atoms in magnetic field causes non-random orientation of
atoms
- not apparent for molecules
- splitting of electronic energy levels occurs (~ 0.01 nm)
- sum of split absorbance lines  original line
Background
- only absorb light with same orientation
- can use Zeeman effect to remove background
> place flame polarized light through
sample in magnetic field get
absorbance (atom+molecule) or
absorbance (molecule) depending
on how light is polarized
z
z
*
*
*
Background+Absorbance
z * z
e) Chemical Interference - more common than spectral interference
1) Formation of Compounds of Low Volatility
- Anions + Cations  Salt
Ca2+ +SO42-  CaSO4 (s)
- Decreases the amount of analyte atomized  decreases the absorbance signal
- Avoid by:
> increase temperature of flame (increase atom production)
> add “releasing agents” – other items that bind to interfering ions
eg. For Ca2+ detection add Sr2+
Sr2+ + SO42-  SrSO4 (s)
increases Ca atoms and Ca absorbance
> add “protecting agents” – bind to analyte but are volatile
eg. For Ca2+ detection add EDTA4Ca2+ + EDTA4-  CaEDTA2-  Ca atoms
2) Formation of Oxides/Hydroxides
M + O  MO
non-volatile & intense molecular absorbance
A
M + 2OH  M(OH)2
- M is analyte
- Avoid by:
> increase temperature of flame (increase atom production)
> use less oxidant
3) Ionization
M  M+ + e- M is analyte
- Avoid by:
> lower temperature
> add ionization suppressor – creates high concentration of esuppresses
M+ by shifting equilibrium.
G) Atomic Emission Spectroscopy (AES) –
similar to AA with flame now being used for
atomization and excitation of the sample for
light production
1) Atomic Processes
heat
Degree of Excitation Depends on Boltzmann Distribution:
N1 and No – are the number of atoms in excited and ground states
k – Boltzmann constant (1.28x10-23 J/K)
T – temperature
DE – energy difference between ground and excited states
P1 and Po – number of states having equal energy at each quantum level
Increase Temperature  increase in N1/No (more excited atoms)
\ I (emission) 
N1, so signal increases with increase in temperature
Need good temperature control to get reproducible signal
eg. For Na, temperature difference of 10o 2500  2510
results in a 4% change in N1/No
Temperature Dependence Comparison between AA and AES:
- AA is relatively temperature independent. Need heat only to get
atoms, not atoms in excited state.
- AA looks at ~ 99.98% of atoms
- AES uses only small fraction (0.02%) of excited atoms
2) Comparison of AA and AES Applications
AES - emission from multiple species simultaneously
Comparison of Detection Limit
Flame Emission More
Sensitive
Sensitivity About the
Same
Flame Absorption
More Sensitive
Al, Ba, Ca, Eu, Ga, Ho,
In, K, La, Li, Lu, Na,
Nd, Pr,Rb, Re, Ru, Sm,
Sr, Tb, Tl, Tm, W, Yb
Cr, Cu, Dy, Er, Gd, Ge,
Mn, Mo, Nb, Pd, Rh,
Sc, Ta, Ti, V, Y, Zr
Ag, As, Au, B, Be, Bi,
Cd, Co, Fe, Hg, Ir, Mg,
Ni, Pb, Pt, Sb, Se, Si,
Sn, Te, Zn
Some better by AA others better by AES
3) Instrumentation
- Similar to AA, but no need for external light source (HCL) or chopper
> look at light from flame
> flame acts as sample cell & light source
Atomization Sources:
Source
Temperature (oC)
Flame
1700-3150
Plasma
4,000-6,000
Arc/Spark
4,000-5,000/40,00
Electrothermal usually not used – too slow and not as precise
a) Flame Source:
- used mostly for alkali metals
> easily excited even at low temperatures
- Na, K
- need internal standard (Cs usually) to correct for variations flame
Advantages
- cheap
Disadvantage
- not high enough temperature to extend to many other elements
b) Plasma (inductively coupled plasma - ICP)
- plasma – electrically conducting gaseous mixture (cations & electrons)
- temperature much higher than flame
- possibility of doing multiple element analysis
> 40-50 elements in 5 minutes
Advantages
- uniform response
- multi-element analysis, rapid
- precision & accuracy (0.3 – 3%)
- few inter-element interferences
- can use with gas, liquid or solids sample
Inductively Coupled Plasma (ICP) Emission Spectroscopy
- involves use of high temperature plasma for sample atomization/excitation
- higher fraction of atoms exist in the excited state, giving rise to an increase
in emission signal and allowing more types of atoms to be detected
Magnetic field
Ions forced to flow in closed
path, Resistance to flow
causes heating
Ar charges
by Tesla coil
(high voltages at high frequency)
Temperature Regions
in Plasma Torch
Overall Design for ICP Emission Spectrometer
Rowland circle:
- curvature corresponds to focal curve of
the concave grating.
- frequencies are separated by grating
and focused onto slits/photomultiplier
tubes positioned around the Rowland
circle
- slits are configures to transmit lines for
a specific element
Arc & Spark Emission Spectroscopy
- involves use of electrical discharge to give high temperature environment
- higher fraction of atoms exist in the excited state, giving rise to an increase
in emission signal and allowing more types of atoms to be detected
- can be used for solids, liquids or gas phase samples
- types of discharge used:
DC arc: high sensitivity, poor precision
DC spark: intermediate sensitivity and precision
AC spark: low sensitivity, high precision
Because of difficulty in reproducing the arc/spark conditions, all elements of interest
are measured simultaneously by use of appropriate detection scheme.
Concave grating disperse frequencies,
photographic film records spectra
Arc created by electrodes separated by a few
mm, with an applied current of 1-30 A
Comparison of ICP and Arc/Spark Emission Spectroscopy
- Arc/Spark first instrument used widely for analysis
- all capable of multielement detection with appropriate instrument design
(e.g. 40-50 elements in 5 min for ICP
- ICP tends to have better precision and stability than spark or arc methods
- ICP have lower limits of detection than spark or arc methods
- ICP instruments are more expensive than spark or arc instruments
Example 11: For Na atoms and Mg+ ions, compare the ratios of the number of particles in
the 3p excited state to the number in the ground state in a natural gas-air flame (2100K)