Atomic absorption spectrometry can be used to determine elemental composition, not compounds. It requires a radiation source and high temperatures for atomization. There are two common atomization methods: flame and electrothermal. Flame atomization introduces samples into a flame where they undergo desolvation, volatilization, dissociation, and excitation/ionization. Electrothermal atomization atomizes samples in a graphite furnace at temperatures up to 3000°C, providing higher sensitivity than flame atomization. Radiation sources like hollow cathode lamps provide narrow, intense emission lines for selective analysis of elements. AA spectrometers can have single or double beam designs to correct for spectral and chemical interferences that can affect results.

1. 1.1 Atomic Absorption Spectrometry (AAS)
• determination of elements not compounds
• needs radiation source
• high temperature for atomization
Atomization
a. Flame
b. Electrothermal

2. 1.2 Flame atomizer for solutions
1. Desolvation: solvent evaporates to produce solid aerosol
2. Volatilization: form the gas molecules
3. Dissociation: produce atomic gas
4. {Ionization: ionize to form cations + electrons}
5. {Excitation: excited by heat of flame, emission}

3. Fig. 8-9 (p.225)
Samples are introduced into flames by a
Fig. 9-1 (p.231)
Processes occurring during atomization

4. Fig. 9-2 (p.231)
Regions in a flame
Fig. 9-3 (p.232)
Temperature (c) profile for a natural
gas-air flame

5. Flame structure
a. Primary combustion zone:
blue luminescence from emission of C2, CH
cool {thermal equilibrium not achieved)
initial decomposition, molecular fragments
b. Interzonal region:
hottest (several cm)
most free atoms, wildly used part
c. Secondary combustion zone:
cooler
conversion of atoms to molecular oxides {then disperse to the
surroundings}
Flame temperatures
Fuel Oxidant temperature (C)
Natural gas Air 1700 ~ 1900
H2 O2 2550 ~ 2700
Acetylene O2 3050 ~ 3000

6. Sensitive part of flame for AAS varies with analyte
 Sensitivity varies with element
Element rapidly oxides – near burner
Element poorly oxidizes – away from burner
 Optimize burner position for each element
 Difficult for multielement detection

7. Fig. 9-5 (p.233) A laminar-flow burner

8. Laminar flow burner
• Stable and quite flame
• Long path length for absorption
• Disadvantages: short residence time in the flame (0.1 ms)
low sensitivity (a large fraction of sample flows down the drain)
Flashback
Flame atomization
• Simplest atomization, needs preliminary sample treatment.
• Best for reproducibility (relative error <1%)
• Relatively intensive – incomplete volatilization, short time in beam

9. 1.3Electrothermal atomization (Method of choice when flame atomization fails)
• Analyis of solutions as well as solids
• Three stages: - dry at low temperature (120C, 20s)
- ash at higher temperature(500-1000C, 60s), removal of volatile
hydroxides, sulfates, carbonates
- atomize of remaining analyte at 2000-3000 C (ms~s)
• High sensitivity  less sample and longer residence time in optical path
(10-10 -10-13 g analyte, 0.5-10uL sample, 2x10-6 -1x10-5 ppm)
• Less reproducible (relative precision 5-10%)
• Slow (several minutes for each element)
• Narrow dynamic range
Two inert gas stream are provided
• External Ar gas prevents outside air from entering/incinerating tube
• Internal Ar gas circulate the gaseous analyte
Output signals from graphite furnace
• Drying
• Ashing (both from volatile absorbing species, smoke scattering)
• Atomize (used for analysis)

10. Fig. 9-6 (p.234) Graphite furnace
electrothermal atomizer
Fig. 9-7 (p.235) Typical output from
electrothermal atomizer

11. 2.1 Radiation source
• Each element has narrow absorption lines (0.002-0.005nm), very selective.
• For a linear calibration curve (Beer’s law), source bandwidth should be narrower than
the width of an absorption line.
- continuum radiation source requests a monochromator with eff < 10-4 nm, difficult!
• Solutions:
- LINE source at discrete wavelength,
resonance line, using 589.6 nm emission line of sodium as a source to probe Na in
analyte
- operate line source with bandwidth narrower than the absorption line width
minimize the Doppler broadening
lower temperature and pressure than atomizer

12. Hollow cathode lamp
• Electric discharge (300V) of Ar between tungsten anode and a cylindrical metal
cathode in a sealed glass tube filled with Ar (1-5 )
• Ar+ bombard cathode and sputter cathode atoms
• Fraction of sputtered atoms excited, then emit characteristic radiation
• Cathode made of metal of interest (Na, Ca, K, Fe,.. or mixture of several metals)
 give intense narrow line source of cathode material
Hollow cathode design:
Concentrate radiation in limited region;
Enhance the probability of redeposition on
cathode

13. Electrodeless discharge lamps
A few  of Ar and small quantity of metal of interest
Energized by an internal radio-frequency or microwave radiation
Discharged Ar+ excite the atoms of metal whose spectrum is sought
Higher intensities than hollow cathode lamp, but less relaiable

14. Fig. 9-10 (p.238) Absorption of a resonance line by atoms

15. 2.2 AA Spectrophotometers
- Single beam design
- Double beam design and lock-in
amplifier

16. 3.1 Spectral interference
- Absorption of interferant overlaps with that of analyte
- Absorption or scattering by fuel/oxidant or sample matrix
background should be corrected for
(reading assignment P241-244)
- Emission of radiation from flame at the same wavelength of AA
lock in amplifier, modulate the real atomic absorption at known frequency using a
lock-in amplifier,

17. 3.2 Chemical interference (more common)
1) Reactions of anions with analytes to form low volatile compound
releasing agent: cations that react preferentially with interferant
e.g.,Sr minimizes interference of phosphate with determination of Ca
protective agent: form stable but volatile compounds with analyte
e.g., EDTA-metal formation supresses the interference of Al, Si, phosphate, sulfate
in determination of Ca
2) Reverse atomization
MO  M + O
M(OH)2  M + 2OH
3) Ionization
M  M+ + e-
ionization suppressor: B  B+ + e-

18. 1. Quantitative determination of > 60 metals or metalloids
flame electrothermal
detection limit 0.001-0.002 pm 2x10-6 -1 x10-5 ppm
relative error 1-2% 5-10%
2. Less suitable for
weaker absorbers (forbidden transitions)
non-metals (absorb in VUV)
metal in low IP (alkali metals)