Atomic (Flame) Emission Spectrometry
The potential of atomic absorption (AA) of radiant energy
for qualitative analysis was demonstrated many years ago by Fraunhoffer in his
analysis of the sun's elemental composition.
However, the modern era of AA began in 1955 when the Australian chemist
Walsh designed and built a convenient
instrument for measuring the wavelengths and radiant power absorbed by
neutral atoms in the vapor state. Walsh
also pointed out that the technique is advantageous over its complement, flame
photometry, in regard to both sensitivity and selectivity. For these reasons, you don't see much flame
photometry being done anymore, but you do see a lot of ICP (Inductively Couple
Plasma, which uses the same principle as flame photometry without the
flame). We don't have an ICP however,
so we'll stick to AA and flame photometry.
In atomic absorption
spectroscopy measurement is made of the radiation absorbed by the nonexcited
atoms in the vapor state. In emission
spectroscopy, measurement is made of energy emitted when atoms in the excited
state return to the ground state. Flame
emisssion spectroscopy (what we called flame photometry in the 1st paragraph)
is a special area of emission spectroscopy in which a flame is used to excite
the atoms.
Flame absorption and flame emission techniques usually
involve introduction of a sample solution in aerosol form into a flame. Solvent evaporation and vaporization of the
salt occur prior to dissociation of the salt into free gaseous atoms. At the temperature of an air/acetylene flame
(~2300oC) atoms of many elements exist largely in the ground
state. When a beam of radiant energy
that consists of the emission spectrum for the element that is to be determined
is passed through the flame, some of the ground state atoms absorb energy of
characteristic wavelengths (resonance lines) and are raised to a higher energy
state. The radiation not removed by
absorption is isolated by a monochrometer and detected by a
photomultiplier. For example, at 283.3
nm,
Pb ---> Pb*
by absorption of a photon
The amount of radiant energy
absorbed as a function of
concentration of an element in the flame is the basis of atomic absorption spectroscopy. The amount of
light absorbed is proportional to the elemental concentration, assuming Beer's
Law holds.
For a few elements, such as the alkali metals Na and K,
an air/acetylene flame is hot enough not only to produce ground state atoms,
but to raise some of the atoms to an excited electronic state. The radiant energy emitted when the atoms return to the ground state is proportional
to the concentration and is the basis of flame
emission spectroscopy. For example,
at 589 nm,
Na -------> Na*
(energy from flame)
Na* ------->
Na + hn (at 589 nm)
The emitted radiant energy
from flame emission is isolated by a monochromator and detected by a
photomultiplier. The measured signal
may or may not be linearly related to analyte concentration.
Within the flame, there are many more atoms in the ground
state than in the excited state. For
Zn, for instance, in a 2000K flame, there are 7.3 ´ 1015 atoms in
the ground state for every excited atom.
(You can do these calculations if or when you take 2nd semester
P-Chem.) Some 65 elements can be
determined with good precision and sensitivity by AA, in contrast to about 10
by conventional flame emission spectroscopy, since a flame provides sufficient
energy for only a few elements to have a signficant number of atoms in an
excited state. The alkali metals are
elements with unocuppied atomic orbitals of low enough energy to be
sufficiently populated by a flame.
Sodium and potassium ions play an important role in a
number of biological systems and their functions. Since these ions form few insoluble compounds and exhibit
essentially no acidic or basic properties, they cannot be determined readily by
conventional wet chemical techniques and are usually measured
instrumentally. The usual techniques
employed in determining these ions include atomic absorption spectroscopy
(AAS), atomic emission spectroscopy (AES) and ion selective electrodes (ISE). All of these methods require that the sample
exist as, or be converted to, an aqueous form.
Choosing between the individual methods is based primarily on the
sensitivity of the method required as well as the speed and convenience of the
method. Since the detection limit for
these ions is lowest by AES and AES is simple, this method is generally used
for their determination.
The spectrum obtained with the flame is due to the
following order of events. The solution
of the metal salt in question is sprayed into the flame and the solvent
evaporates leaving the finely powdered salt.
The salt is vaporized, atomized, and a valence electron is raised to a
higher energy state. The energy emitted
when this electron drops down into a vacant lower level is given off as radiant
energy of a wavelength determined by the Planck-Einstein relationship:
DE = hn = hc/l
Flame spectra are mainly
composed of ground state transitions, while higher energy sources give rise to
transitions between excited states. You
will be treated to this sort of thing in the laser laboratory in 2nd semester
P-Chem lab.
Flame photometry is largely an empirical method and is
sensitive to experimental conditions. The
signal intensity from a flame is dependent on the flame temperature, the rate
of flow of liquid into the flame, the pressure and rate of flow of fuel gases,
and any of many other variables which affect the character of the flame or
atomizing of the sample. Thus the
compounds in which the ion is found and the viscosity of the solution have a
great effect. A consequence of this situation is that reliable results can be
obtained only after painstaking attention to details, with repeated checks of
reproducibility and the effects of altering conditions.
Atomic emission analyses are most commonly and routinely
performed on solutions. Therefore your sample must be converted to liquid form
prior to analysis. This is most conveniently done using a microwave to digest the sample, leaving a solution
that can then be analyzed. There must
be a sufficient concentration of analyte for the spectrometer to detect. Prior to performing atomic emission
analysis, you will need to determine the minimum detection limit for the
element of interest.The minimum quantifiable limit (the lowest concentration of
analyte which can be quantitatively determined) is generally 3-5 times the
minimum detection limit.
The minimum detection limit is intimately related to the
concept of signal-to-noise ratio (S/N) and the standard deviation of a measurement.
Noise is conveniently measured as the standard deviation of numerous
measurements of the signal. It is possible to detect a signal when the S/N is
three or greater. Thus the minimum detection limit is when the S/N = 3. If you
measure the standard deviation of a number of measurements you can then
calculate the minimum detectable signal. The minimum detection limit is
obtained by experimentally determining the concentration of analyte which gives
the minimum detectable signal. From the minimum detection limit calculate the
minimum quantitiable limit.Check your calculations to see if the AA will detect
the analyte concentration you expect in your sample. If you cannot detect that
low of a concentration, decide if a preconcentration
step is in order.