Chapter 8

1
Chapter 8 9 Atomic Absorption Spectroscopy
2
Atomic Absorption Process

• A neutral atom in the gaseous state can absorb
  radiation and transfer an electron to an excited
  state.
• Simple electronic transitions possible with no
  vibrational and rotational energy levels
  possible. Bandwidth much narrower!
• Occur at discreet l
• Na(g) 3s 3p and 3p 5s as well as other
  transitions are possible at the correct photon
  energy a transition.

3
Atomic Absorption Transitions
4
THE FLAME AND EXCITED STATES

• 3 steps required before measurements are possible
  in an A.A. experiment 1. vaporization 2.
  reduction to the elemental state and then 3.
  exposure to radiation.
• The first two steps are accomplished by a flame.
• Effect of flame temperature Since flame is at
  high temperature might have an effect on fraction
  of atoms in excited state.
• Boltzmann's equation describes effect of flame
  temperature where
• N of atoms in a given state
• g statistical factor for a given level and
  measures the number of possible electrons in each
  level
• g 2J 1 where J Russel-Saunders coupling
  constant and is given by J L S or L ? S where
  L orbital angular momentum quantum (0,1,2,3
  for s, p, d, f respectively) and S spin ½.
• E.g. For the Na transition
• 3s½ 3p3/2 gu 2(LS) 1 2(1 ½) 1 4
  and
• go 2(0 ½) 1 2.
• 3s½ 3p½ go 2 and gu 2(1 - ½) 1 2.
• Overall population of both of these states since
  they are only separated by 5Å, let's use average
  of their wavelengths and add population for the
  two excited states
• g 4 2 6 and go 2 (as before) lave 5892Å.
  

5
The Flame and Excited States

• Assume Air-acetylene flame (2400C) Temperatures
  for different flames used in AA are listed in
  text. T (2400 273)K 2673K
• Substituting into the Boltzmann equation
• 3.23x10?4
• Very small fraction of the atoms in the flame are
  excited to this excited state.

6
Relative population of higher energy transitions

• 3p 5s transition is also possible and has l
  6161Å (E 3.22x10?12 erg.
• The fraction of 3p electrons excited to the 5 s
  orbital is calculated as before
• 5.34x10-5
• Fraction involved in this transition even
  smaller.
• Finally, we can estimate the fraction of
  electrons in the 5s state relative to the 3s
  state
• 5.34x10?5?3.23x10?4 1.72x10?8 QED
• Only very small proportion of the absorbing
  species is in the excited state from excitation
  by flame higher energy transitions much less
  likely than the lower energy transitions.

7
MEASURING ATOMIC ABSORPTION

• Recall Beer's Law (A log ebC ) is obeyed
  when line width
• small compared to absorption band.
• Atoms or molecules absorb radiation at discrete
  wavelengths.
• Broadband radiation contains photons of several
  wavelengths, some of which may be useful but many
  of which will not. This will make Po ( Pusable
  Puseless) larger and the absorbance smaller
  than would be expected with only the usable
  portion of the light available for absorption.
• Besides the Pusable can be composed of
  wavelengths with different absorptivities i.e.
  the sample does not absorb all radiation to the
  same degree.
• Non-linear behavior observed when l range of
  excitation source is greater than l range of
  absorber bandwidth of excitation source must be
  narrower than bandwidth of absorber.

8
Linewidth of Atomic Transitions

• Natural linewidth of an absorption spectrum is
  very small (10?4Å) but is broadened by
• Doppler broadening Random thermal motions of
  atoms relative to the detector
• Pressure broadening In the atomic absorption
  experiment the pressure is large enough that
  atoms can undergo some interatomic collisions
  which cause small changes in the ground state
  levels.
• Normal line width of excitation lines much
  greater than this line width
• Monochromator cannot be used to select l range
  in AA (bandwidth ? few tenths of a nm).

9
SOURCES

• Solution to line width problem Use atomic source
  of same material.
• e.g. For Na analysis Na vapor is used.
• Atoms are excited by electrical discharge the
  excited atoms emit a characteristic l. The
  bandwidth of the source ltlt sample linewidth since
  it is generated under conditions where there is
  less broadening.
• Hollow Cathode Tube Hollow cathode made of the
  material needed is vaporized and emits radiation
  of the characteristic wavelength.
• The ion current to the cathode controls photon
  intensity Increasing the voltage between the
  anode and cathode will control the current and
  thus total photon flux.
• Optimum current for each lamp (1-20ma).

10
FORMATION OF ATOMIC VAPOR

• Four methods used to vaporize sample from
  solution
• Ovens Sample placed in an oven after
  evaporating solvent, sample vaporized into
  irradiation area by rapidly increasing
  temperature.
• Electric arc or spark Sample subjected to high
  current or high potential A.C. spark.
• Ion bombardment Sample placed on cathode and
  bombarded with ions (Ar). Sputtering process
  dislodges them from cathode and directs them to
  irradiation region.
• Flame atomization Sample sprayed into flame
  where it undergoes atomization and irradiation.

11
FLAME ATOMIZERS

• Total consumption burner Separate channels bring
  sample, fuel, and oxidant to combustion area. All
  of the sample, that is carried into the burner,
  is burned
• Sensitivity is greater than in a burner where the
  sample is not completely burned.
• extra turbulence in the flame from variations in
  droplet size increase noise.

12
Premix (laminar flow) burner

• Sample, fuel, and oxidant mixed prior to entering
  flame.
• Turbulence drastically reduced by removing larger
  droplets.
• Mixing baffles insure only fine mist makes it
  through to burner.

13
ELECTROTHERMAL ATOMIZATION

• all of the sample used is atomized in furnace
  (electrothermal) atomizer.
• detection limit is 100-1000x lower than with
  aspiration techniques.
• only a few mL of solution is used.
• Basic Principle
• sample container resistively heated to vaporize
  the metal atoms.
• sample dried (evaporate solvent) at 110C
• ash sample called "burn off" (200-300C)
• atomization.(2000-3000C)
• comparison with flame atomization
• interaction with sample matrix and electrode
• poorer reproducibility
• detection limits of 10?10-10?12g (or 1ppb) are
  possible.

14
FUELS/OXIDANTS

• Low T flames easily reduced elements (Cu, Pb,
  Zn, Cd)
• High T flames difficult to reduce elements (e.g.
  alkaline earths).
• Fuels natural gas, propane, butane, H2, and
  acetylene
• Oxidants- Air and O2 (low temperature flames).
  N2O (high temperature flames).
• Flame characteristics
• Sample enters flame, is vaporized, reduced and
  eventually oxidized.
• Exact region of flame in which each of these
  occurs depends upon
• flow rate,
• drop size, and
• oxidizability of sample.
• Optimum position for flame for many metals.

15
Flame Profiles in AA
16
MEASUREMENT OF AA

• Ideally, the amount of light reaching the
  detector is given by Beers Law P Po10?ebC .
• several interferences can change this to
• P Po10?ebC Pemission ? Pbackground ?
  Pscattering.
• Pemission is due to analyte emission in the flame
  
• eliminated from the absorbance by modulation of
  the light source measures only AC levels
  emission DC level.
• Pbackground, Pscattering due to absorption by
  the flame or are induced by sample matrix and are
  independent of the analyte.
• Broad band in nature.
• Flame interferences nulled by comparing a blank
  with sample
• Sample matrix is a problem. Caused by, for
  example, high salt content (e.g. NaCl or KI).
  These have broad band absorption spectrum in
  flame since they are not reduced by it. Most
  common approach uses secondary continuum source
  (e.g. D2 lamp)
• Each lamp (D2 and HCT) modulated but are 180 out
  of phase with each other.
• Detection system measures difference between two
  absorbance signals AHCT Asample Abrdband
  while Acontinuum source Abrd band. will be
  absorbance of sample.

17
D2 Source Elimination of Background
18
MONOCHROMATOR

• Needed to choose one of several possible emission
  lines (lemitted) associated with HCT.
• Since they are usually reasonably well separated
  from the line of interest, it is straightforward
  to use a monochromator to eliminate this
  interference.

19
ANALYTICAL TECHNIQUES

• Beer's law, A kC, not always true making a
  calibration curve necessary.
• Standard addition method is used to minimize the
  effects from the matrix
• Anion- height of the absorbance peak is
  influenced by type and concentration of anion.
  It can reduce the number of atoms made. An
  unknown matrix is thus hard to correct for
• Cation The presence of a second cation sometimes
  causes stable compounds to form with the cation
  being analyzed. e.g. Al Mg produces low results
  for Mg due to the formation of an Al/Mg oxide.

20
Sample Problem

• The nickel content in river water was determined
  by AA analysis after 5.00 L was trapped by ion
  exchange. Rinsing the column with 25.0 mL of a
  salt solution released all of the nickel and the
  wash volume was adjusted to 75.00 mL 10.00 mL
  aliquots of this solution were analyzed by AA
  after adding a volume of 0.0700 ?g Ni/mL to each.
  A plot of the results are shown below.
  Determine the concentration of the Ni in the
  river water.

21
Return to Slide 10