Transcript Atomic Spectroscopy - Professor Monzir Abdel

Atomic Spectroscopy
Introduction
1
Technique – Flame Test
2
An Introduction to Optical
Atomic Spectroscopy
In optical atomic spectrometry, compounds are first
converted to gaseous molecules followed by
conversion to gaseous atoms. This process is
called atomization and is a prerequisite for
performing atomic spectroscopy. Gaseous
atoms then absorb energy from a beam of
radiation or simply heat. Absorbance can be
measured or emission from excited atoms is
measured and is related to concentration of
analyte.
3
Atomic Energy Level Diagrams
As a start, we should be aware that only valence
electrons are responsible for atomic spectra
observed in a process of absorption or
emission of radiation in the UV-Vis region.
Valence electrons in their ground states are
assumed to have an energy equal to zero eV.
As an electron is excited to a higher energy
level, it will absorb energy exactly equal to the
energy difference between the two states. Let
us look at a portion of the sodium energy level
diagram where sodium got one electron in the
3s orbital:
4
5
6
The dark lines represent most probable transitions
and in an atomic spectrum they would appear
more intense than others. It should also be
indicated that two transitions, of very
comparable energies (589.0 and 589.6 nm),
from the 3s ground state to 3p excited state do
take place. This suggests splitting of the p orbital
into two levels that slightly differ in energy.
Explanation of this splitting may be presented as
a result of electron spin where the electron spin
is either in the direction of the orbital motion or
opposed to it.
7
Both spin and orbital motion create magnetic
fields that may interact in an attractive
manner (if motion is in opposite direction,
lower energy), or in a repulsive manner
when both spin and orbital motion are in
the same direction (higher energy). The
same occurs for both d and f orbitals but
the energy difference is so small to be
observed. A Mg+ ion would show very
similar atomic spectrum as Na since both
have one electron in the 3s orbital.
8
In cases where atoms of large numbers of
electrons are studied, atomic spectra
become too complicated and difficult to
interpret. This is mainly due to presence of
a large numbers of closely spaced energy
levels
It should also be indicated that transition
from ground state to excited state is not
arbitrary and unlimited. Transitions follow
certain selection rules that make a specific
transition allowed or forbidden.
9
Atomic Emission and Absorption
Spectra
At room temperature, essentially all atoms
are in the ground state. Excitation of
electrons in ground state atoms requires
an input of sufficient energy to transfer the
electron to one of the excited state through
an allowed transition. Excited electrons will
only spend a short time in the excited state
(shorter than a ms) where upon relaxation
an excited electron will emit a photon and
return to the ground state.
10
Each type of atoms would have certain
preferred or most probable transitions
(sodium has the 589.0 and the 589.6 nm).
Relaxation would result in very intense
lines for these preferred transitions where
these lines are called resonance lines.
Absorption of energy is most probable for
the resonance lines of each element. Thus
intense absorption lines for sodium will be
observed at 589.0 and 589.6 nm.
11
Atomic Fluorescence Spectra
When gaseous atoms at high temperatures are
irradiated with a monochromatic beam of
radiation of enough energy to cause electronic
excitation, emission takes place in all directions.
The emitted radiation from the first excited
electronic level, collected at 90o to the incident
beam, is called resonance fluorescence.
Photons of the same wavelength as the incident
beam are emitted in resonance fluorescence.
This topic will not be further explained in this text
as the merits of the technique are not very clear
compared to instrumental complexity involved
12
Atomic Line Width
It is taken for granted that an atomic line
should have infinitesimally small (or zero)
line width since transition between two
quantum states requires an exact amount
of energy. However, careful examination of
atomic lines reveals that they have finite
width. For example, try to look at the
situation where we expand the x-axis
(wavelength axis) of the following line:
13
14
The effective line width in terms of
wavelength units is equal to Dl1/2 and is
defined as the width of the line, in
wavelength units, measured at one half
maximum signal (P). The question which
needs a definite answer is what causes
the atomic line to become broad?
15
Reasons for Atomic Line
Broadening
There are four reasons for broadening observed in
atomic lines. These include:
1. The Uncertainty Principle
We have seen earlier that Heisenberg uncertainty
principle suggests that nature places limits on
the precision by which two interrelated physical
quantities can be measured. It is not easy, will
have some uncertainty, to calculate the energy
required for a transition when the lifetime of the
excited state is short.
16
The ground state lifetime is long but the
lifetime of the excited state is very short
which suggests that there is an uncertainty
in the calculation of the transition time. We
have seen earlier that when we are to
estimate the energy of a transition and
thus the wavelength (line width), it is
required that the two states where a
transition takes place should have infinite
lifetimes for the uncertainty in energy (or
wavelength) to be zero:
17
DE>h/Dt
Therefore, atomic lines should have some
broadening due to uncertainty in the
lifetime of the excited state. The
broadening resulting from the uncertainty
principle is referred to as natural line width
and is unavoidable.
18
2. Doppler Broadening
The wavelength of radiation emitted by a
fast moving atom toward a transducer
will be different from that emitted by a
fast atom moving away from a
transducer. More wave crests and thus
higher frequency will be measured for
atoms moving towards the transducer.
The same occurs for sound waves
19
Assume your ear is the transducer, when a
car blows its horn toward your ear each
successive wave crest is emitted from a
closer distance to your ear since the car is
moving towards you. Thus a high
frequency will be detected. On the other
hand, when the car passes you and blows
its horn, each wave crest is emitted at a
distance successively far away from you
and your ear will definitely sense a lower
frequency.
20
21
The line width (Dl) due to Doppler broadening can
be calculated from the relation:
Dl/lo = v/c
Where lo is the wavelength at maximum power
and is equal to (l1 + l2)/2, v is the velocity of the
moving atom and c is the speed of light. It is
noteworthy to indicate that an atom moving
perpendicular to the transducer will always have
a lo, i.e. will keep its original frequency and will
not add to line broadening by the Doppler effect.
22
In the case of absorption lines, you may visualize
the line broadening due to Doppler effect since
fast atoms moving towards the source will
experience more wave crests and thus will
absorb higher frequencies. On the other hand,
an atom moving away from the source will
experience less wave crests and will thus absorb
a lower frequency. The maximum Doppler shifts
are observed for atoms of highest velocities
moving in either direction toward or away from a
transducer (emission) or a source (absorption).
23
24
3. Pressure Broadening
Line broadening caused by collisions of
emitting or absorbing atoms with other
atoms, ions, or other species in the
gaseous matrix is called pressure or
collisional broadening. These collisions
result in small changes in ground state
energy levels and thus the energy required
for transition to excited states will be
different and dependent on the ground
state energy level distribution.
25
This will definitely result in important line
broadening. This phenomenon is most
astonishing for xenon where a xenon arc lamp at
a high pressure produces a continuum from 200
to 1100 nm instead of a line spectrum for atomic
xenon. A high pressure mercury lamp also
produces a continuum output. Both Doppler and
pressure contribution to line broadening in
atomic spectroscopy are far more important than
broadening due to uncertainty principle.
26
4. Magnetic Effects
Splitting of the degenerate energy levels does take place
for gaseous atoms in presence of a magnetic field. The
complicated magnetic fields exerted by electrons in the
matrix atoms and other species will affect the energy
levels of analyte atoms. The simplest situation is one
where an energy level will be split into three levels, one
of the same quantum energy and one of higher quantum
energy, while the third assumes a lower quantum energy
state. A continuum of magnetic fields exists due to
complex matrix components, and movement of species,
thus exist. Electronic transitions from the thus split levels
will result in line broadening
27
The Effect of Temperature on
Atomic Spectra
Atomic spectroscopic methods require the conversion
of atoms to the gaseous state. This requires the use
of high temperatures (in the range from 2000-6000
oC). Thee high temperature can be provided through
a flame, electrical heating, an arc or a plasma
source. It is essential that the temperature be of
enough value to convert atoms of the different
elements to gaseous atoms and, in some cases,
provide energy required for excitation. The
temperature of a source should remain constant
throughout the analysis especially in atomic
emission spectroscopy.
28
Quantitative assessment of the effect of
temperature on the number of atoms in the
excited state can be derived from Boltzmann
equation:
Where Nj is the number of atoms in excited state,
No is the number of atoms in the ground state, Pj
and Po are constants determined by the number
of states having equal energy at each quantum
level, Ej is the energy difference between excited
and ground states, K is the Boltzmann constant,
and T is the absolute temperature.
29
Boltzmann distribution
30
Atom
Wavelength
Cs
852.1 nm
Nj /N0 at
3000 K
7.24  10-3
Na
589.0 nm
5.88  10-4
Ca
422.7 nm
3.69  10-5
Zn
213.9 nm
5.58  10-10
To understand the application of this equation let
us consider the situation of sodium atoms in the
3s state (Po = 2) when excited to the 3p excited
state (Pj = 6) at two different temperatures 2500
and 2510K. Now let us apply the equation to
calculate the relative number of atoms in the
ground and excited states:
Usually we use the average of the emission lines
from the 3p to 3s where we have two lines at
589.0 and 589.6 nm which is:
31
32
Therefore, at higher temperatures, the number of
atoms in the excited state increases. Let us calculate
the percent increase in the number of atoms in the
excited state as a result of this increase in
temperature of only 10 oC:
33
34
Effect of Temperature on Atomic
Absorption and Emission
The question here is which technique would be affected
more as a result of fluctuations in temperature? The
answer to this important question is rather simple.
Atomic emission is the technique that will be severely
affected by fluctuations in temperature since signal is
dependent on the number of atoms in the excited state.
This number is significantly affected by fluctuations in
temperature as seen from the example above. However,
in the case of atomic absorption, the signal depends on
the number of atoms in ground state that will absorb
energy.
35
very high as related to the number of excited
atoms:
Nj/No = 1.72x10-4
or
172 excited atoms for each 106 atoms in ground
state
This suggests a very high population of the ground
state even at high temperatures. Therefore,
atomic absorption will not be affected to any
significant extent by fluctuations in temperature,
if compared to atomic emission spectroscopy.
36
However, there are some indirect effects of
temperature on atomic absorption
spectroscopy. These effects can be
summarized as:
1. Better sensitivities are obtained at higher
temperatures since higher temperatures can
increase the number of vaporized atoms at any
time.
2. Higher temperatures will increase the velocities
of gaseous atoms, thus causing line
broadening as a result of the Doppler and
collisional effects.
3. High temperatures increase the number of
ionized analyte and thus decrease the number
of atoms available for absorption.
37
Band and Continuum Spectra
Associated with Atomic Spectra
When the atomization temperature is insufficient to
cause atomization of all species in the sample
matrix, the existent molecular entities, at the
temperature of the analysis, impose very
important problems on the results of atomic
absorption and emission spectroscopy. The
background band spectrum should be removed
for reasonable determination of analytes.
Otherwise, the sensitivity of the instrument will
be significantly decreased.
38
39
As the signal for the blank is considered
zero and thus the instrument is made to
read zero, when the analyte is to be
determined, it got to have an absorbance
greater than the highest point on the
continuum and the instrument will assume
that the absorbance related to analyte is
just the value exceeding the background
blank value. This will severely limit the
sensitivity of the technique.
40
Putting this conclusion in other words we
may say that if the analyte signal is less
than the background blank, the instrument
will read it as zero. Therefore, it is very
important to correct for the background or
simply eliminate it through use of very high
temperatures that will practically atomize
all species in the matrix. We will come to
background correction methods in the next
chapter.
41
Atomization Methods
It is essential, as we have seen from
previous discussion, that all sample
components (including analytes,
additives, etc.) should be atomized. The
atoms in the gaseous state absorb or
emit radiation and can thus be
determined. Many ionization methods are
available which will be detailed in the next
two chapters. Generally, atomization
methods can be summarized below:
42
43
Sample Introduction Methods
The method of choice for a specific sample
will mainly depend on whether the sample
is in solution or solid form. The method for
sample introduction in atomic
spectroscopy affects the precision,
accuracy and detection limit of the
analytical procedure.
44
Introduction of Solution Samples
1. Pneumatic Nebulizers
Samples in solution are usually easily introduced
into the atomizer by a simple nebulization,
aspiration, process. Nebulization converts the
solution into an aerosol of very fine droplets
using a jet of compressed gas. The flow of gas
carries the aerosol droplets to the atomization
chamber or region. Several versions of
nebulizers are available and few are shown in
the figure below:
45
46
47
48
49
50
2. Ultrasonic Nebulizers
In this case samples are pumped onto the surface
of a piezoelectric crystal that vibrates in the kHz
to MHz range. Such vibrations convert samples
into homogeneous aerosols that can be driven
into atomizers. Ultrasonic nebulization is
preferred over pneumatic nebulization since
finer droplets and more homogeneous aerosols
are usually achieved. However, most
instruments use pneumatic nebulization.
51
3. Electrothermal Vaporization
An accurately measured quantity of sample (few
mL) is introduced into an electrically heated
cylindrical chamber through which an inert gas
flows. Usually, the cylinder is made of pyrolytic
carbon but tungsten cylinders are now available.
The signal produced by instruments which use
electrothermal vaporization (ETV) is a discrete
signal for each sample injection. Electrothemal
vaporizers are called discrete atomizers to
differentiate them from nebulizers which are
called continuous atomizers
52
4. Hydride Generation
Techniques
Samples that contain arsenic, antimony, tin,
selenium, bismuth, and lead can be
vaporized by converting them to volatile
hydrides by addition of sodium
borohydride. Volatile hydrides are then
swept into the atomizer by a stream of an
inert gas.
53
Introduction of Solid Samples
A variety of techniques were used to introduce solid samples into
atomizers. These include:
1. Direct Sample Insertion
Samples are first powdered and placed in a boat-like holder (from
graphite or tantalum) which is placed in a flame or an electrothermal
atomizer.
2. If the sample is conductive and is of a shape that can be directly
used as an electrode (like a piece of metal or coin), that would be
the choice for sample introduction in arc and spark techniques.
Otherwise, powdered solid samples are mixed with fine graphite and
made into a paste. Upon drying, this solid composite can be used as
an electrode. The discharge caused by arcs and sparks interacts
with the surface of the solid sample creating a plume of very fine
particulates and atoms that are swept into the atomizer by a flow of
an inert gas. This process of sample introduction is called ablation
54
3.
Laser Ablation
Sufficient energy from a focused intense laser will
interact with the surface of samples (in a similar
manner like arcs and sparks) resulting in
ablation. The formed plume of vapor and fine
particulates are swept into the atomizer by the
flow of an inert gas. Laser ablation is becoming
increasingly used since it is applicable to
conductive and nonconductive samples
55
56
57
4. The Glow Discharge
Technique
A low pressure envelope (1 to 10 torr argon)
with two electrodes with the conductive solid
sample is the cathode, as in the figure below.
The technique is used for sample
introduction and atomization as well. The
electrodes are kept at a 250 to 1000 V DC.
This high potential is sufficient to cause
ionization of argon which will be accelerated
to the cathode where the sample is
introduced. Collision of the fast moving
energetic argon ions with the sample
(cathode) causes atomization by a process
called sputtering.
58
59
60
Atomic Absorption
Spectroscopy
We will cover two main techniques of
atomic absorption spectroscopy (AAS),
depending of the type atomizer. Two
atomization techniques are usually
used in AAS:
61
1. Flame Atomization
Flames are regarded as continuous atomizers
since samples are continuously introduced
and a constant or continuous signal is
obtained. Samples in solution form are
nebulized by one of the described
nebulization techniques discussed
previously. The most common nebulization
technique is the pneumatic nebulization.
Nebulized solutions are carried into a flame
where atomization takes place.
62
Several processes occur during
atomization including:
a. Nebulized samples are sprayed into a
flame as a spray of very fine droplets
b. Droplets will lose their solvent content
due to very high flame temperatures in
a process called desolvation and will
thus be converted into a solid aerosol.
c. The solid aerosol is volatilized to form
gaseous molecules
63
d. Gaseous molecules will then be
atomized and neutral atoms are
obtained which can be excited by
absorption of enough energy. If energy
is not enough for atomization, gaseous
molecules will not be atomized and we
may see molecular absorption or
emission
e. Atoms in the gaseous state can absorb
energy and are excited. If energy is too
much, we may observe ionization.
64
65
66
The different processes occurring in flames
are complicated and are not closely
controlled and predicted. Therefore, it can
be fairly stated that the atomization
process in flames may be one of the
important parameters limiting the precision
of the method. It is therefore justified that
we have a closer look at flames and their
characteristics and the different variables
contributing to their performance.
67
Types of Flames
Flames can be classified into several types
depending on fuel/oxidant used. For
example, the following table summarizes
the features of most familiar flames.
Therefore, it can be clearly seen that
significant variations in flame temperatures
can be obtained by changing the
composition of fuel and oxidant.
68
On the other hand, flames are only stable at
certain flow rates and thus the flow rate of the
gas is very important where at low flow rates
(less than the maximum burning velocity) the
flame propagates into the burner body causing
flashback and, in some cases, an explosion. As
the flow rate is increased, the flame starts to rise
above the burner body. Best flames are obtained
when the flow rate of the gas is equal to the
maximum burning velocity. At this equity ratio the
flame is most stable. At higher ratios, flames will
reach a point where they will no longer form and
blow off the burner.
69
Flame Structure
Three well characterized regions can be identified in a
conventional flame. A lower region, close to the
burner tip, with blue luminescence. This region is
called the primary combustion zone which is
characterized by existence of some non atomized
species and presence of fuel species (C2 and CH,
etc.) that emit in the blue region of the
electromagnetic spectrum. The second well defined
region is called the interzonal region just above the
primary combustion zone. The interzonal region is
rich in free atoms and is the region of choice for
performing atomic spectroscopy.
70
It also contains the regions of highest
temperatures. The third region in the
flame is the outer region which is called
the secondary combustion region. It is
characterized by reformation of
molecules as the temperature at the
edges is much lower than the core.
These regions can be schematically
represented by the following
schematic:
71
72
Flame Absorbance profiles
Since the temperature of a flame depends on the
position from its tip, it is necessary to concentrate
our work on one spot in a flame and preferably
adjust the height of the flame to get best signal. In
fact, not all elements require a specific height above
burner tip but rather each element has its own
requirements which largely reflect some of its
properties. For example, one can use higher
distances from the tip so that higher temperatures
are achieved to analyze for silver. This is possible
since silver will not be easily oxidized.
73
Flame Absorption Profiles
We have seen that there are different temperature
profiles in a flame and temperature changes as
the distance from the burner tip is change
74
However, best results for the analysis of
chromium occur at lower heights (fuel rich
flames) since at higher heights oxygen from
atmosphere will force chromium to convert
to the oxide which will not be atomized at
flame temperatures. A third situation can be
observed for magnesium where increasing
the height above tip will increase the signal
due to increased atomization at higher
temperatures. However, at higher distances
the oxide starts to form leading to a decrease
in signal.
75
76
Flame Atomizers (Continuous
Atomizers)
Flame Atomizers (Continuous Atomizers)
There are several types of flame atomizers available.
The simplest is a turbulent flow burner that is very
similar to conventional Bunsen burner. This type of
burner suffers from fluctuations in temperature
since there is no good mechanism for homogeneous
mixing of fuel and oxidant. The drop size of
nebulized sample is also inhomogeneous which
adds to fluctuations in signal. The path length of
radiation through the flame is small which suggests
a lower sensitivity of the technique.
77
Turbulent flow burners are also susceptible to
flashback. These drawbacks were overcome
using the most widely used laminar flow
burner where quite flames and long path
length are obtained. Flashback is avoided
and very homogeneous mixing between fuel,
oxidant, and droplets take place. Larger
droplets are excluded and directed to a
waste container. A schematic representation
of the burner is shown below:
78
79
80
81
82
Fuel and Oxidant Regulators
The adjustment of the fuel to oxidant
ratio and flow rate is undoubtedly very
crucial. Although stoichiometric ratios
are usually required, optimization is
necessary in order to get highest
signal. However, in the determination of
metals that form stable oxides, a flame
with excess fuel is preferred in order to
decrease oxide formation.
83
Performance Characteristics of
Flame Atomizers
Reproducibility of flame methods are usually
superior to other atomization techniques.
However, the residence time of an atom in a
flame is in the order of 10-4 s which is very
short. This is reflected in a lower sensitivity
of flame methods as compared to other
methods. Also, conventional flames with
reasonable burning velocities can produce
relatively low temperatures which make them
susceptible to interference from molecular
species.
84
2. Electrothermal Atomization
These have better sensitivities than flame
methods. The increased sensitivity can be
explained on the basis that a longer atom
residence time is achieved (can be more than
1 s) as well as atomization of the whole
sample in a very short time. As the name
implies, a few mL of the sample are injected
into the atomization chamber (a cylinder of
graphite coated with a film of pyrolytic
carbon) where the following processes take
place:
85
a. Evaporation: the solvent associated with the
sample is evaporated in a low temperature
(~120 oC) slow process (seconds)
b. Ashing: sample is ashed to burn organics
associated with the sample at moderate
temperatures (~600 oC, seconds)
c. Atomization: The current is rapidly increased
after ashing so that a temperature in the
range from 2000-3000 oC is obtained in less
than1 second.
86
Electrothermal Atomizers
(Discrete Atomizers)
The heart of the atomizer, beside efficient
heating elements and electronics, is a
cylindrical graphite tube opened from
both ends and has a central hole for
sample introduction. It was found that
porous graphite results in poor
reproducibility since some of the
analyzed materials will diffuse through
porous graphite and will thus lead to a
history effect.
87
Therefore, the cylindrical graphite is made
from a special type of nonporous high
quality graphite called pyrolytic graphite. The
length of the cylinder is 2-5 cm and it has
less than 1 cm diameter. When the tube is
fixed in place electrical contacts are
achieved which are water cooled. Two inert
gas streams (argon) flow at the external
surface and through the internal space of the
tube to prevent oxidation and clean the tube
after each measurement. Usually, samples
are analyzed in triplicates where three
consecutive reproducible signals are
required for each sample..
88
89
Performance Characteristics of
Electrothermal Atomizers
Electrothermal atomization is the technique of choice
in case of small sample size. Also, higher
sensitivities than flames are ordinarily obtained.
Unfortunately, the analysis time is in the few minutes
range and the relative precision is in the range of 510% as compared to 1% in flame methods. In
addition, the linear dynamic range is usually small (~
two orders of magnitude) which requires extra
sample manipulation. It may be also mentioned that
better experienced personnel can achieve the merits
of the technique.
90
Atomic Absorption
Instrumentation
Atomic absorption instruments consist
of a source of radiation, a
monochromator, a flame or
electrothermal atomizer in which
sample is introduced, and a transducer.
91
Radiation Sources
Although radiation in the UV-Vis region is required, we
can not use broad band sources. This is because
even the best monochromators can not provide a
bandwidth that is narrower than the atomic
absorption line. If the bandwidth of the incident
radiation is wider than the line width, measurement
will fail as absorption will be only a tiny fraction of a
large signal which is difficult to measure and will
result in very low sensitivities (figure a). Therefore,
line sources with bandwidths narrower than that of
the absorption lines must be used
92
93
This can be achieved by using a lamp
producing the emission line of the
element of interest where analyte
atoms can absorb that line. Conditions
are established to get a narrower
emission line than the absorption line.
This can in fact be achieved by getting
an emission line of interest at the
following conditions:
94
1. Low temperatures: to decrease
Doppler broadening (which is easily
achievable since the temperature of the
source is always much less than the
temperature in flames).
2. Lower pressures: this will decrease
pressure broadening and will thus
produce a very narrow emission line.
95
This may suggest the need for a separate
lamp for each element which is
troublesome and inconvenient.
However, recent developments lead to
introduction of multielement lamps. In
this case, the lines from all elements
should not interfere and must be easily
resolved by the monochromator so
that, at a specific time, a single line of
one element is leaving the exit slit
96
Hollow Cathode Lamp (HCL)
This is the most common source in atomic absorption
spectroscopy. It is formed from a tungsten anode
and a cylindrical cathode the interior surface of
which is coated by the metal of interest. The two
electrodes are usually sealed in a glass tube with a
quartz window and filled with argon at low pressure
(1-5 torr). Ionization of the argon is forced by
application of about 300 V DC where positively
charged Ar+ heads rapidly towards the negatively
charged cathode causing sputtering. A portion of
sputtered atoms is excited and thus emit photons as
atoms relax to ground state. The cylindrical shape of
the cathode serves to concentrate the beam in a
limited region and enhances redeposition of
sputtered atoms at the hollow surface.
97
98
99
High potentials usually result in high
currents which, in turn, produce more
intense radiation. However, Doppler
broadening increases as a result. In
addition, the higher currents will
produce high proportion of unexcited
atoms that will absorb some of the
emission beam which is referred to as
self absorption (a lower intensity at the
center of the line is observed in this
case).
100
Electrodeless Discharge Lamps
(EDL)
An EDL is a sealed quartz tube containing a
few torr of an inert gas and a small quantity
of the metal of interest. Excitation of the
metal is achieved by a radiofrequency or a
microwave powered coil through ionization
of argon, due to high energetic
radiofrequency. Ionized argon will hit the
metal causing excitation of the atoms of the
metal of interest. The output power of the
EDL lamp is higher than the HCL lamp.
However, compared to HCL lamps, EDL
lamps are rarely used.
101
102
103
Emission in Flames
There can be significant amounts of emission
produced in flames due to presence of flame
constituents (molecular combustible
products) and sometimes impurities in the
burner head. This emitted radiation must be
removed for successful sensitive
determinations by AAS, otherwise a negative
error will always be observed. We can
visualize this effect by considering the
schematic below:
104
105
The detector will see the overall signal which is
the power of the transmitted beam (P) in
addition to the power of the emitted radiation
from flame (Pe). Therefore if we are
measuring absorbance, this will result in a
negative error as the detector will measure
what it appears as a high transmittance
signal (actually it is P + Pe). In case of
emission measurements, there will always be
a positive error since emission from flame is
an additive value to the actual sample
emission. It is therefore obvious that we
should get rid of this interference from
emission in flames.
106
Source Modulation
It turned out that excluding the emission
signal from flames can easily be done
by an addition of a chopper to the
instrumental design. The chopper is a
motor driven device that has open and
solid (mirrors in some cases)
alternating regions as in the schematic:
107
108
The function of the chopper is to chop the light
leaving the source so that when the incident
beam hits the chopper at the solid surface,
the beam will be blocked and detector will
only read the emitted signal from the flame.
As the chopper rotates and the beam
emerges to the detector, the detector signal
will be the sum of the transmitted signal plus
that emitted from the flame. The signal
processor will be able to subtract the first
signal from the second one, thus excluding
the signal from emission in flames.
109
This can be represented by the following
equations:
Signal 1 (Blocked Beam) = Pe
Signal 2 (Transmitted Beam) = P + Pe
Overall Difference Signal = (P + Pe) - Pe =
P (Corrected Signal)
This correction method for background
emission in flames is called source
modulation.
110
The schematic of the AAS instrument with source
modulation correction can be represented by the
following schematic:
111
It should be recognized that addition of extra
components to an instrument will decrease
the signal to noise ratio and addition of a
moving component is usually regarded as a
disadvantage due to higher need for
maintenance.
Another procedure which can overcome the
emission from flames is to use a modulated
power supply that will give fluctuating
intensities at some frequency (say for
example pulsed radiation at a specific
frequency).
112
The emission from flames is a continuous
signal but that from the source is modulated.
Now if we use a high pass RC filter, only the
fluctuating signal will be measured as signal
while the DC signal will be considered zero
as it can not pass through the electronic
filter. The high pass RC filter is a device
which uses a resistor and a capacitor the
impedance of which is inversely proportional
to the frequency of the modulated signal.
Therefore, only high frequencies will have
low impedance and can pass through the
capacitor while signals of low frequencies
will suffer very high resistance and will not
be able to go through the capacitor.
113
AAS Instruments
Instruments in AAS can be regarded as
single or double beam instruments.
Single Beam Atomic Absorption
Spectrophotometers
A single beam instruments is the same
as the one described above (source
modulation section) or generally:
114
The term
“spectrophotometer”
implies that the
instrument uses a
dispersive
monochromator
(containing a prism
or a grating). Also,
the detector is a
photomultiplier tube
in most cases.
115
Double Beam Atomic Absorption
Spectrophotometers
In this type of instruments, the incident beam
is split into two beams of equal intensity by a
chopper with the solid surface being a
mirror. One of the beams will traverse the
sample in the atomizer while the other is
considered as a reference. Detector signals
will be consecutive readings of both the
reference and sample beams. The ratio of the
reference to the sample beams is recorded to
give the final signal.
116
A schematic representation of a double
beam instrument is shown below:
117
It should be emphasized here that in the
absence of sample, Pr is not equal to P
since the reference beam traverses
through air while the other beam
traverses through the flame. In flames,
particulates and molecular species
scatter and absorb a portion of incident
radiation, which results in a lower
intensity of the beam. To act as a real
double beam, The AA
spectrophotometer reference beam
should pass through a reference flame.
118
But even if we do that, there are no
guarantees that both beams will be of
equal intensities because it is almost
impossible to obtain exactly equivalent
flames. It is therefore important to
understand that the excellent features
of a double beam configuration are not
achievable in AAS instrumentation.
119
Interferences in Atomic
Absorption Spectroscopy
There are two major classes of interferences
which can be identified in atomic absorption
spectroscopy. The first class is related to
spectral properties of components other than
atomized analyte and is referred to as
spectral interferences. The other class of
interferences is related to the chemical
processes occurring in flames and
electrothermal atomizers and their effects on
signal. These are referred to as chemical
interferences and are usually more important
than spectral interferences.
120
Spectral Interferences
1. Spectral line Interference
Usually, interferences due to overlapping lines
is rare since atomic lines are very narrow.
However, even in cases of line interference, it
can be simply overcome by choosing to
perform the analysis using another line that
has no interference with other lines.
Therefore, line interference is seldom a
problem in atomic spectroscopy.
121
2. Scattering
Particulates from combustion products
and sample materials scatter radiation
that will result in positive analytical
error. The error from scattering can be
corrected for by making a blank
measurement. Scattering phenomenon
is most important when concentrated
solutions containing elements that
form refractory oxides (like Ti, Zr, and
W) are present in sample matrix.
122
Metal oxide particles with diameters
larger than the incident wavelength will
make scattering a real problem. In
addition, samples containing organic
materials or organic solvents can form
carbonaceous (especially in cases of
incomplete combustion) particles that
scatter radiation.
123
3. Broad Band Absorption
In cases where molecular species from
combustion products or sample matrix are
formed in flames or electrothermal
atomizers, a broad band spectrum will result
which will limit the sensitivity of the
technique. It should be indicated here that
spectral interferences by matrix products
are not widely encountered in flame
methods. Even if matrix effects are present
in flames, they can be largely overcome by
adjusting various experimental conditions
like fuel/oxidant ratio or temperature.
124
Another method for overcoming matrix
interferences is to use a much higher
concentration of interferent than that
initially present in sample material, in
both sample and standards (this
material is called a radiation buffer).
The contribution from sample matrix
will thus be insignificant.
Spectral interferences due to matrix are
severe in electrothermal methods and
must thus be corrected for.
125
Background Correction Methods
a.
The Two Line Correction Method
In this method, a reference line from the source
(from an impurity in cathode or any emission
line) is selected where this line should have
the following properties:
1.
Very close to analyte line
2.
Not absorbed by analyte
If such a line exists, since the reference line is
not absorbed by the analyte, its intensity
should remain constant throughout analysis.
126
However, if its intensity decreases, this
will be an indication of absorbance or
scattering by matrix species. The
decrease in signal of the reference line
is used to correct for the analyte line
intensity (by subtraction of the
absorbance of the reference from that
of the analyte). This method is very
simple but unfortunately it is not
always possible to locate a suitable
reference line.
127
b.
The Continuum Source
Method
This background correction method is the
most common method although, for reasons
to be discussed shortly, it has major
drawbacks and fails a lot. In this technique,
radiation from a deuterium lamp and a HCL
lamp alternately pass through the graphite
tube analyzer. It is essential to keep the slit
width of the monochromator sufficiently wide
in order to pass a wide bandwidth of the
deuterium lamp radiation.
128
In this case, the absorbance by analyte
atoms is negligible and absorbance can
be attributed to molecular species in
matrix. The absorbance of the beam
from the deuterium lamp is then
subtracted from the analyte beam
(HCL) and thus a background
correction is obtained.
129
130
Problems Associated with Background
Correction Using D2 Lamp
1.
The very hot medium inside the graphite
tube is inhomogeneous and thus signal is
dependent on the exact path a beam would
follow inside the tube. Therefore, exact
alignment of the D2 and HCL lamps should be
made.
2.
The radiant power of the D2 lamp in the
visible is insignificant which precludes the
use of the technique for analysis of analytes
in the visible region.
3.
Addition of an extra lamp and chopper
will decrease the signal to noise ratio.
131
c.
Background Correction Based on
Zeeman Effect
Zeeman has observed that when gaseous atoms (but
not molecules) are placed in a strong magnetic field
(~ 1 tesla), splitting of electronic energy levels takes
place. The simplest splitting of one energy level
results in three energy levels, one at a higher energy,
another at a lower energy (two s satellite lines) and
the third remains at the same energy as the level in
absence of the magnetic field (central p line).
Furthermore, the p line has twice the absorbance of
a s line and absorbs polarized light parallel to
direction of the magnetic field while the two s lines
absorb light perpendicular to magnetic field.
132
Light from a HCL lamp will pass through
a rotating polarizer that passes
polarized light parallel to external
magnetic field at one cycle and passes
light perpendicular to field in the other
cycle. The idea of background
correction using this method is to allow
light to traverse the sample in the
graphite furnace atomizer and record
the signal for both polarizer cycles
using the wavelength at the p line.
133
First cycle: light parallel to field; the p
line of the analyte absorbs in addition to
absorbance by matrix (molecular matrix
absorb both polarized light parallel or
perpendicular to field)
Signal a = Ap + AMatrix
b.
Second cycle: light perpendicular to field;
the p line of analyte will not absorb light
perpendicular to field and s lines will also
not affect absorbance at the p line
wavelength. Only matrix will absorb.
Signal b = AMatrix
a.
134
The overall signal is the difference of the
two signals = Ap
Therefore, excellent background
correction is achieved using the
Zeeman effect. This background
correction method results in good
correction and is usually one of the
best methods available.
135
136
Chemical Interferences
These are interferences resulting from
chemical processes occurring in flames and
electrothermal atomizers and affect the
absorption signal. To quantitatively assess
the effects of the different chemical
processes occurring in flames, one should
regard the burnt gases as behaving like a
solvent. This is necessary since our
knowledge of gaseous state reaction
equilibria is rather limited. Chemical
interferences include three major processes:
137
Formation of Compounds of Low
Volatility
Anionic species forming compounds of low
volatility are the most important. The
formation of low volatility species will result
in a negative error or at least will decrease
the sensitivity. For example, the absorption
signal of calcium will be decreased as higher
concentrations of sulfate or phosphate are
introduced. Cations forming combined
products with the analyte will also decrease
the signal obtained for the analyte. For
example aluminum forms a heat stable
compound with magnesium.
138
Elimination of Low Volatility
Compounds
1. Addition of a releasing agent: cations that
can replace the analyte (preferentially react
with the anion) are called releasing agents.
In this case the analyte is released from the
compound of low volatility and replaced by
the releaseing agent. Lanthanum or
strontium are good releasing agents in the
determination of calcium in presence of
phosphate or sulfate. Also, lanthanum or
strontium are good releasing agents in the
determination of magnesium in presence of
aluminum since both can replace
magnesium.
139
2. Addition of a protective agent: organic
ligands that form stable volatile
species with analytes are called
protective agents. An example is
EDTA and 8-hydroxyquinoline which
will form complexes with calcium
even in presence of sulfate and
phosphate or aluminum.
3. Use of higher temperature is the
simplest procedure to try if it is
possible
140
Dissociation Equilibria
Dissociation reactions occur in flames
where the outcome of the process is
desired to produce the atoms of
analyte. For example, metal oxides
and hydroxides will dissociate in
flames to produce the atoms as in the
equations
MO = M + O
M(OH)2 = M + 2 OH
141
Remember that we are not working in solution
to dissociate the compounds into ionic
species. In fact, not much is known about
equilibrium reactions in flames. It should
also be remembered that alkaline earth
oxides and hydroxides are relatively stable
and will definitely show characteristic broad
band spectra (more intense than line
spectra), except at very high temperatures.
The opposite behavior is observed fro alkali
metals oxides and hydroxides which are
instable even at lower flame temperatures
and thus produce line spectra.
142
An equilibrium can be established for the
dissociation of compounds containing atoms
other than oxygen, like NaCl where:
NaCl = Na + Cl
Now, if the signal from a solution of NaCl was
studied in presence of variable amounts of Cl
(from HCl, as an example), the signal will be
observed to decrease as the concentration of
Cl is increased; a behavior predicted by the
Le Chatelier principle in solutions.
143
The same phenomenon is observed
when a metal oxide is analyzed using a
fuel rich flame or a lean flame. Signal
will be increased in fuel rich flames
since the dissociation of metal oxides
is easier due to less oxygen while the
opposite takes place in lean flames
(oxygen rich).
144
A good example on dissociation equilibria can
be presented for the analysis of vanadium in
presence of aluminum and titanium, fuel rich
flames result in higher absorbance signal for
vanadium since the little oxygen present in
flames will be mainly captured by Al and Ti,
thus more V atoms are available. However, in
lean flames, excess oxygen is present and
thus vanadium will form the oxide and
addition of extra Ti and Al will not affect the
signal.
145
Ionization Equilibria
Ionization in fuel/air flames is very limited due
to relatively low temperatures. However, in
fuel/nitrous oxide or fuel/oxygen mixtures,
ionization is significant. Therefore, at
higher temperatures an important portion
of atoms can be converted to ions:
M = M+ + e
K = [M+][e]/[M]
146
Ionization in flames may explain the decrease
in absorption signal for alkali metals at very
high temperatures where as the temperature
is increased signal will increase till an extent
at some temperature where it starts to
decrease as temperature is further
increased; a consequence of ionization.
Therefore, usually lower flame temperatures
are used for determination of alkali metals. A
material that is added to samples in order to
produce large number of electrons is
referred to as an ionization suppressor, the
addition of which results in higher
sensitivities.
147
Practical Details in AAS
1.
Sample Preparation
The most unfortunate requirement of AAS may
be the need for introduction of samples in
the solution form. This necessitates the
dissolution of the sample where in many
cases the procedure is lengthy and requires
very good experience. Care should be
particularly taken in order not to lose any
portion of the analyte and to make sure that
the reagents, acids, etc. used in the
dissolution and pretreatment of the sample
are free from analyte impurities.
148
I suggest that you follow exact
procedures for preparation of specific
samples for analysis by AAS. In some
cases where the sample can be
introduced directly to an electrothermal
atomizer without pretreatment (like
serum samples), definitely,
electrothermal atomizers will have an
obvious advantage over flame methods
which require nebulization.
149
Organic Solvents
1. Increased nebulization rate due to lower
surface tension of organic solvents which
produces smaller droplets as well as faster
evaporation of solvents in flames will result
in better sensitivities.
2. Immiscible organic solvents containing
organic ligands are used to extract metal
ions of interest and thus concentrate them
in a small volume (thus increasing
sensitivity) and excluding possible
interferences due to matrix components.
150
Calibration Curves
The absorbance of a solution is directly
proportional to its concentration but due to
the large number of variables in AAS, usually
this direct relationship may slightly deviate
from linearity. The standard procedure to do
is to construct a relation between the
absorbance and concentration for a series of
solutions of different concentrations. The
thus constructed graph is called a calibration
curve.
151
The unknown analyte absorbance is
found and the concentration is
calculated or located on the curve.
Neither interpolation nor extrapolation
is permitted to the calibration curve. A
sample can be diluted or the calibration
curve may be extended but always the
analyte absorbance should be within
the standard absorbance range
recorded. Usually, the concentration
axis has the ppm or ppb units.
152
Standard Addition method
Chemical and spectral interferences can be
partially or wholly overcome by the use of a
special technique of calibration called the
method of standard addition. In addition, the
use of this method provides better
correlations between standards and sample
results due to constant nebulization rates.
The method involves addition of the same
sample volume to a set of tubes or
containers.
153
Variable volumes of a standard are added
to the tube set followed by completion
to a specific volume. Now, all tubes
contain the same amount of sample but
different concentrations of analyte. A
plot is then made for the volume of
standard and absorbance. This plot will
have an intercept (b) with the y axis and
a slope equals m.
154
The concentration of the analyte can be
determined by the relation:
Cx = bCs/mVx
Where, Cx and Vx are concentration and
volume of analyte and Cs is the
concentration of standard.
One can only use two points to get the
analyte concentration using the
relation:
Cx = AxCsVs/(At –Ax)Vx
155
Detection Limits
Usually, atomic absorption based on
electrothermal atomization has better
sensitivities and detection limits than
methods based on flames. In general,
flame methods have detection limits in
the range from 1-20 ppm while
electrothermal methods have detection
limits in the range from 1-20 ppb.
156
This range can significantly change for
specific elements where not all
elements have the same detection
limits. For example, detection limits fro
mercury and magnesium using
electrothermal atomization are 100 and
0.02 ppb while the detection limits for
the same elements using flame
methods are 500 and 0.1 ppm,
respectively.
157
Accuracy
Flame methods are superior to electrothermal
methods in terms of accuracy. The relative
error in flame method can be less than 1%
while that for electrothermal method occurs
in the range from 5-10%. Also, electrothermal
methods are more susceptible to molecular
interferences from the matrix components.
Therefore, unless a good background
correction method is used, large errors can
be encountered in electrothermal methods
depending on the nature of sample analyzed.
158
Flame Photometry
The technique referred to as flame
photometry is a flame emission
technique. We introduce it here
because we will not be back to flame
methods in later chapters. The basics
of the technique are extremely simple
where a sample is nebulized into a
flame. Atomization occurs due to high
flame temperatures and also excitation
of easily excitable atoms can occur.
159
Emission of excited atoms is
proportional to concentration of
analyte. Flame emission is good for
such atoms that do not require high
temperatures for atomization and
excitation, like Na, K, Li, Ca, and Mg.
The instrument is very simple and
excludes the need for a source lamp.
The filter is exchangeable in order to
determine the analyte of interest and, in
most cases, a photomultiplier tube is
used as the detector.
160
161
162
b. Charge Transfer and Charge
Coupled Transducers
The photosensitive elements are, in contrary to
PDAs, arranged in two dimensions in both
charge injection devices (CID) and chargecoupled devices (CCD). Therefore, these are
very similar to photographic films. For example,
a commercially available transducer is formed
from 244 rows with each row containing 388
detector elements. This will add up to a twodimensional array holding 16672 detector
elements (pixels) on silicon chip that is 6.5 mm
by 8.7 mm.
163
164
165
The full description of the system and its
mechanism will not be covered here as
this is behind the scope of this course.
However, we should qualitatively know
that these important transducers function
by first collecting the photogenerated
charges in different pixels and then
measuring the quantity of the charge
accumulated in a brief period.
Measurement is accomplished by
transferring the charge from a collection
area to a detection area.
166
Optical Atomic Spectra
We have briefly described in an introductory
chapter that atomic spectra are usually
line, rather than band, spectra due to
absence of vibrational and rotational
levels. The existence of quantized
electronic energy levels explains the origin
of the observed line spectra and exact
locations of possible lines
167