1

CHAPTER 1 BASIC PRINCIPLES

All atoms can absorb light at certain wavelengths corresponding to the energy
requirements of the particular atom. Sodium atoms, for example, absorb light very
strongly at 589.0 nm, because light at this wavelength has exactly the right energy
to transform the sodium atom to another electronic state. This electronic
transition is quite specific for sodium; atoms of any other element are different, so
their energy requirements are different and they cannot absorb light at this
wavelength. If the sodium atom is in the `ground state' we say that when a
sodium atom absorbs light it is transformed into an excited state; it is still a
sodium atom, but it contains more energy. There are several possible excited
states for the sodium atom and each has a particular energy. Usually these
energies are measured in relation to the ground state, and a particular excited
state for sodium may be 2.2 eV (electron volts) above the ground state. This
means that an atom in the excited state contains 2.2 eV more energy than a
ground state atom which, by convention, is ascribed an arbitrary energy of zero.
Some of the possible energy states for the sodium atom are shown in Figure 1.

Figure 1: Some possible energy states for the sodium atom.

Each transition between different electronic energy states is characterized by a
different energy, and hence by a different wavelength of light. These wavelengths
are sharply defined and when a range of wavelengths is surveyed, each
wavelength shows as a sharp energy maximum (a spectroscopic `line'). Atomic
spectra are distinguished by these characteristic lines, which can be contrasted
with the broader band spectra associated with molecules. Lines which originate
in the ground state atom are most often of interest in atomic absorption
spectroscopy; these are called resonance lines.

The atomic spectrum characteristic of each element, then, comprises a number of
discrete lines, some of which are resonance lines. Most of the other lines arise
from excited states, rather than from the ground state. Since these resonance
lines are much more sensitive and since most atoms in a practical atomizer are
found in the ground state, these excited state lines are generally not useful for
atomic absorption analysis.



2

The basic principles of atomic absorption spectroscopy can be expressed by
three simple statements:

1. All atoms can absorb light.

2. The wavelength at which light is absorbed is specific for a particular
chemical element. If a sample containing nickel, for example, together with
elements such as lead and copper is exposed to light at the characteristic
wavelength for nickel, then only the nickel atoms will absorb this light.

3. The amount of light absorbed is proportional to the concentration of
absorbing atoms.

An atomic absorption spectrophotometer (Figure 2) is simply an instrument in
which these basic principles are applied to practical quantitative analysis. The
instrument consists of:

Figure 2: Typical arrangement of atomic absorption spectrophotometer.

A light source to generate light at the wavelength which is characteristic of the
analyte element.
An atomizer to create a population of analyte atoms.
A monochromator to separate light at the characteristic wavelength from all
other light.
An optical system to direct light from the source through the atom population
and into the monochromator.
A light-sensitive detector.
Suitable electronic devices which measure the response of the detector and
translate this response into useful analytical measurements.



3

At its most basic level, the general analytical procedure is straightforward:

Convert the sample into solution.

Make up a solution which contains no analyte element (the analytical blank).

Make up a series of calibration solutions containing exactly known amounts of
analyte element (the standards).

Atomize the blank and standards in turn and measure the response for each
solution.

Plot a calibration graph showing the response obtained for each solution as
shown in Figure 3.

Figure 3. Calibration Graph.

Atomize the sample solution and measure the response.

Refer to the calibration graph, enter the graph at the point corresponding to
the response for the sample, and read off the concentration of the sample.

Fundamentally, then, quantitative analysis by atomic absorption spectroscopy is a
matter of converting samples and standards into solutions, comparing the



4
instrumental responses of standards and samples, and using these comparative
responses to establish accurate concentration values for the element of interest.
Basically this can be carried out using simple equipment and simple procedures.
Inevitably, however, there are aspects of the technique which are not quite as
simple and straightforward as this brief introduction suggests.


RELATIONSHIP BETWEEN LIGHT ABSORPTION AND ANALYTE
CONCENTRATION

The relation between light absorption and analyte concentration is defined in the
fundamental laws of light absorption:

1. Lambert's law

The portion of light absorbed by a transparent medium is independent of the
intensity of the incident light, and each successive unit thickness of the medium
absorbs an equal fraction of the light passing through it.

2. Beers law

Light absorption is proportional to the number of absorbing atoms in the sample.

The combined Beer-Lambert law may be expressed mathematically as:

I
t
= I
o
(10)
abc


Thus: log 10
I
I
abc
o
t
= ....................(1)

where: I
o
= incident radiation power

I
t
= transmitted radiation power

a = "absorption coefficient" (absorptivity)

b = length of absorption path

c = concentration of absorption atoms

that is, the absorbance is proportional to the concentration of the element for a
given absorption path length at any given wavelength.




5
Thus, if a sample of concentration c results in a certain measured absorbance,
another sample of concentration 2c will, according to the Beer-Lambert law, result
in twice this absorbance. The `absorbance' is a measure of the amount of light
absorbed by the atoms under given conditions, and this is the measurement of
interest; the Beer-Lambert law allows us to relate the measured absorbance to the
concentration of the analyte element in the sample.

In principle, it might be possible to calculate the concentration directly from
equation (1). In practice, however, the quantities a and b in equation (1) are
constants and are not normally determined. The simplest way to use atomic
absorption methods is to measure the absorbance for standard solutions where
the concentration is known, and then to compare these results with the
absorbance obtained for the unknown sample. Since the measured absorbance
depends directly on the concentration of the analyte element, this procedure gives
a simple and accurate method for determining the unknown concentration while
compensating for instrumental effects which make the theoretical calculation very
uncertain.

Conventionally, the calibration and comparison with standard may be done
graphically. A calibration curve is prepared for each situation, relating the element
concentration to the measured absorbance. Under the Beer Lambert law, this
graph will be linear and the concentration of the unknown samples may be
determined simply by interpolation on the graph (refer to Figure 3).

But, several factors relating to spectral effects and instrumental design can
combine to cause deviations from complete linearity, especially at higher
absorbances.



6

















Left Blank Intentionally



7
CHAPTER 2 ATOMIC ABSORPTION HARDWARE

1. LIGHT SOURCES

a) General
What are `continuum sources' and `line sources'? The most familiar
continuum source is the domestic incandescent bulb. This type of source
emits light over a wide wavelength range starting at about 300 nm and
extending far into the infra-red. Its spectral output characteristics are
shown in Figure 4.

Figure 4: Spectral characteristics of a continuum source.

In contrast, a line source emits light only at discrete wavelengths. Yellow
street lamps are familiar examples. The lamps contain sodium vapour, and
light is emitted at only two discrete wavelengths, 589.0 nm and 589.6 nm,
the well-known sodium `D' lines. The spectral output characteristics of a
sodium lamp are shown in Figure 5.

Figure 5: Spectral characteristics of a line source




8
For practical atomic absorption instruments, the light source is required to
emit light which is at precisely the same wavelength as that at which atomic
absorption occurs for the element of interest.

Conveniently, this can be done by using a light source which emits the
same atomic spectrum, since the atomic emission and absorption
wavelengths are identical. In most atomic absorption instruments, the light
source used is the hollow cathode lamp.

b) Hollow cathode lamps
A typical hollow cathode lamp (Figure 6) consists of a glass envelope
containing a cathode (a metal cup or cylinder containing the chemical
element to be excited) and a suitable anode. The sealed envelope
contains an inert gas, usually argon or neon, at a low pressure. When a
high voltage - up to 600 volts - is applied across the electrodes, positively
charged gas ions bombard the cathode and dislodge atoms of the element
used in the cathode. These atoms are subsequently excited by collisional
processes and the spectrum of that element can be produced. Some
cathodes can simply be machined from a bar of the pure element required -
nickel, iron, copper and so on. Some elements, however, are difficult to
machine, and cathodes for these elements are produced by powder
metallurgy. The cathode is usually surrounded by an insulating shield of
mica, ceramic material or glass. This ensures that the discharge is
confined to the interior of the cathode and results in a considerable
improvement in the intensity of the emitted spectral lines.

Figure 6: Hollow cathode lamp.

The anode can be an annular ring around the mouth of the cathode, or a
`flag' near the mouth of the cathode, or a wire or rod located in a
convenient position. In some designs, the anode also acts as a
mechanical support for the insulating shield.

The material used for the lamp window is important since it must transmit
the spectral lines of the element used for the cathode (Figure 7.)



9

Figure 7. Transmittance characteristics of quartz and Pyrex.

The fill gas is usually argon or neon at a pressure of 10 to 15 torr - about
1/50th of atmospheric pressure. Neon is preferred because it produces a
higher signal intensity than argon, but where a neon line occurs in close
proximity to the element's resonance line, argon is used instead.

Hollow cathode lamps are not, of course, the only light source capable of
producing the line spectra of chemical elements, but they are the most
universally accepted source for atomic absorption instruments.

OPERATING CONDITIONS

For practical operation, it is essential that the hollow cathode lamp is correctly
aligned in the optical path. This is, however, a simple matter of mechanically
aligning the lamp by means of one or two screws to obtain the maximum signal.
Once the lamp is correctly aligned, the operating current is then the only lamp
parameter requiring any attention. Manufacturers generally recommend a
suitable operating current. This is seldom highly critical, and small departures
from it will not affect analytical sensitivity to any great extent.

Larger departures from the recommended current can be more serious. If the
lamp current is too small, the analytical signal will require excessive amplification.
This will almost always produce undesirably high noise.

There are two adverse effects of operating a lamp at high current. The first of
these is that the resonance line becomes broadened and distorted. This



10
phenomenon is known as `self-absorption broadening' and is caused by atoms in
the discharge absorbing at the resonance wavelength emitted by similar atoms.
The result is that the absorption sensitivity is degraded and the calibration
curvature increases. Figure 8 illustrates the change in curvature for magnesium
when the lamp current is increased.

Figure 8: Effect of increased lamp current on calibration curvature

The second adverse effect is that lamp life is shortened as the operating current
is increased.

With these factors in mind, it is good operating practice to use the manufacturer's
recommendation as a starting point and then find by trial and error that current
which gives the optimum combination of signal-to-noise ratio and calibration
linearity. The major consideration is not to exceed the maximum rated current.

c) Deuterium Lamps

A continuum source does have an application in atomic absorption
spectroscopy - not to measure atomic absorption but to measure and
correct for the `background effects' caused by molecular species or by
scattering. This is called background correction (the causes of background
absorption are discussed later).

The source commonly used is a deuterium (or hydrogen) filled discharge
lamp which emits an intense continuum spectrum from below 190 nm to
about 425 nm (see Figure 9). This covers the region where most atomic
absorption lines occur and where the effects of background absorption are



11
most pronounced. A polyatomic gas (H
2
or D
2
) is used because a
continuum is produced rather than a line spectrum.

Figure 9: Spectral characteristics of deuterium lamp

The deuterium lamp is different to a hollow cathode lamp in construction and
operation. The lamp incorporates a heated, electron-emitting cathode, a metal
anode and a restrictive aperture between the two (see Figure 10). A discharge
current of several hundred milliamperes excites the deuterium gas. The
discharge is forced to pass through the small aperture, forming a defined area
of high excitation and hence high light emission. A quartz window transmits the
light to the spectrophotometer optical system.

The operating current for the deuterium lamp is usually automatically controlled
by the spectrophotometer - the operator simply switches the background
corrector system on and ensures that the lamp is correctly aligned.

Figure 10: Deuterium lamp.



12
2. MONOCHROMATORS

The function of the monochromator is to isolate a single atomic resonance line
from the spectrum of lines emitted by the hollow cathode lamp. In effect, it is an
adjustable filter which selects a specific, narrow region of the spectrum for
transmission to the detector and rejects all wavelengths outside this region.
Ideally, the monochromator should be capable of isolating the resonance line only
and excluding all other wavelengths. For some elements this is relatively easy,
for others it is more difficult.

Copper for example has a comparatively uncluttered spectrum with the nearest
line being 2.7 nm from the 324.7 nm resonance line (Figure 11).

Figure 11: Atomic Spectrum of copper near 324.7 nm

Nickel, on the other hand, has quite strong lines at 231.7 nm and 232.1 nm - one
on each side of the 232.0 nm resonance line (Figure 12).

Figure 12: Atomic spectrum of nickel near 232.0 nm




13
The ability to discriminate between different wavelengths (usually referred to as
resolution) is thus a very important characteristic of the monochromator.
Extremely sophisticated monochromators are available which can isolate
wavelength regions as small as 0.01 nm and less; however, for atomic absorption
spectrophotometers, a typical requirement is about 0.2 nm `bandpass'.
Most instruments use grating monochromators which operate in the range 190 -
800 nm. The slits of the monochromator determine the amount of the spectrum
that falls on the photomultiplier (band pass). It is generally advisable to keep the
slits as narrow as possible to lower the flame emission reaching the
photomultiplier and to reject unwanted lines from the hollow cathode lamp. The
minimum slit width is determined by the amount to which the resonance line signal
can be reduced before noise becomes too intolerable.

Figure 13: Optical geometry of monochromator (Czerny-Turner mounting)

If the slit is too wide, the light throughput will be high and the signal-to-noise ratio
may be excellent, but the resonance line may not be isolated from other lines and
the calibration may be badly curved. Conversely, if the slit is too narrow, the
resolution may be excellent but the signal-to-noise ratio may be unacceptably
poor because of the reduced light throughput.


Figure 14. Slit width, calibration linearity and signal-to-noise ratio



14

3. DETECTORS

The detector universally used in atomic absorption instruments is the
photomultiplier tube. No other device offers the same sensitivity over the
wavelength range required for atomic absorption analysis.

Figure 15. Electrode configuration of typical photomultiplier

The photomultiplier produces an electrical signal which is proportional to the
intensity of light at the wavelength which has been isolated by the
monochromator. This electrical signal is then amplified and used to provide a
quantitative measurement of absorption.

4. READOUT SYSTEMS

The signal from the detector is amplified if necessary and then used to drive some
device which gives a visible signal. Early instruments used a meter, calibrated in
absorbance, alternatively a chart recorder was used to obtain a visible record of
the detector output. Most modern instruments now include a computer. A useful
addition is the facility for averaging the signal over a period of time.

5. OPTICAL SYSTEMS

In atomic absorption spectrophotometers, the basic purpose of the optical system
is to gather light from the source, direct it through the analyte atom population,
and then direct it into the monochromator. Optical configurations may be either



15
single-beam or double-beam; mirrors or lenses can be used as the coupling
elements. The simple, single-beam mirror system illustrated in Figure 16 is used
in many instruments. The second mirror focuses the image of the lamp cathode at
the centre of the flame (or other atomization cell). The third mirror focuses this
image in turn on to a plane mirror where the beam is folded and passed to the
entrance slit of the monochromator. Lenses can be used to achieve the same
effect as shown in Figure 17.

Figure 16. Single-beam mirror system



Figure 17. Single-beam lens system


In the single-beam configuration shown, the light from the source traverses only
one path - through the flame. In this system it is necessary to measure the initial
intensity of the resonance line I
o
before inserting the sample and measuring the
transmitted intensity I
t
. The single-beam system therefore relies on the light
source remaining stable during the period of analysis, that is, intensity I
o
should
not drift or fluctuate while I
t
is being measured. With modern hollow cathode
lamps, I
o
will generally remain sufficiently stable after a warm-up period.
However, there are some analytical circumstances where any slight drift will be
inconvenient.




16

Figure 18. Double-beam system

A double-beam system such as the one illustrated in Figure 18 is designed to
correct for any drift in I
o
. Light from the source is directed at a beam splitter
consisting of a partially aluminized quartz plate. The sample beam is focused by
mirror M1, passes through the flame and is directed at the monochromator
entrance slit through mirror M2 and plane mirror M3. The reference beam
passing through the beam splitter is directed at the monochromator entrance slit
by a second array of mirrors such that this beam enters the monochromator
without passing through the flame.

A rotating reflecting chopper `C' is located near the monochromator entrance slit
where it alternately interrupts the sample beam to direct the reference beam into
the monochromator. In this way, intensities I
o
and I
t
are separately measured at
high frequency. For I
o
this high frequency of measurement provides
near-continuous monitoring so that corrections for variations in intensity can be
instantaneously applied. While the ratio I
o
:I
t
is being calculated electronically, the
calculation can be continuously corrected for variations in I
o
and the ultimate
analytical result will be unaffected by drift in source intensity.


6. BACKGROUND CORRECTION

Background interference occurs when radiation from the hollow cathode lamp is
attenuated by molecular species or solid particles in the observation zone (flame
or furnace).

Correction for this is relatively simple with the aid of a continuum source (usually a
deuterium lamp). Whereas the signal obtained when using a hollow cathode lamp is
the total absorbance (the sum of atomic absorption and background absorption), the



17
signal obtained from the continuum lamp is the background absorption only. In
the systems illustrated in Figures 19 and 20, the optical configuration is such that
radiation from both hollow cathode lamp and the continuum lamp coincide
precisely along the optical path through the observation zone.

Figure 19. Single-beam system with background corrector

The background signal is subsequently subtracted electronically from the total
absorbance signal and the analytical result is thus corrected for background
interference:

Hollow cathode lamp signal = atomic absorption + background
Deuterium lamp signal = background only
Electronically processed signal = atomic absorption only.

In the double-beam system illustrated by Figure 20, radiation from the continuum
source traverses the same sample and reference paths as radiation from the
hollow cathode lamp. The intensities of both sources can thus be concurrently
monitored. Any drift in the intensity of either source can be automatically
corrected for so as to maintain the accuracy of background correction.

Figure 20. Double-beam system with background corrector



18

























Left Blank Intentionally



19

CHAPTER 3 FLAME ATOMIZATION

GENERAL

It is worth stressing that successful atomic absorption analysis depends on
generating a supply of uncombined analyte atoms in the ground state and
exposing this atom population to light at the characteristic absorption wavelength.

This process consists of taking a solution of the analyte and heating it to a
temperature that is sufficient to dissociate the compound. Traditionally, the
thermal energy required has been supplied by a flame, although furnace
techniques are now widely used over a broad range of applications. But, the
flame is still the most widely used dissociation agency, and a thorough grasp of
the principles of flame atomization is fundamental to the understanding of atomic
absorption analysis.


1. THE BURNER-NEBULIZER SYSTEM

Burner-nebulizer systems used in atomic absorption instruments atomize the
analyte solution in successive stages (see Figure 21):

Figure 21. Flame atomization

The solution must first be converted into a spray of fine droplets by a pneumatic
nebulizer which typically has the shape and geometry illustrated in Figure 22.



20

Figure 22. Pneumatic nebulizer.

In operation, gas (usually the oxidant for the flame) flows through the nebulizer,
passes through the venturi section and draws solution into the venturi as a spray
of droplets. This high-speed spray strikes a carefully positioned obstacle (usually
a spherical glass bead) where the bigger droplets are shattered into smaller
droplets. Only about 10% of the solution is converted into sufficiently fine
droplets to be carried into the flame, the remainder is driven against the spray
chamber wall and drained off to waste. The spray and oxidant are mixed with the
incoming fuel and the complete mixture finally flows through the burner (Figure 23).

Figure 23. Cross-section of spray chamber/burner system

a) NEBULIZATION

The function of the nebulizer system is to convert the analytical solution
into an aerosol which is fully compatible with the requirements for effective
dissociation in the flame. Physically, these requirements appear to be met
by a population of droplets having median diameters between 5 and 7 m



21
and a maximum of 20 m. Obviously, small droplets are easier to dry and
vaporize than large droplets, and the emphasis is therefore on maximizing
the proportion of smaller droplets within the system. The position of the
glass bead relative to the nebulizer venturi is a critical factor in determining
nebulizer efficiency, droplet size, and analytical response (Figure 24). The
optimum position will depend on the physical characteristics of the solution
and it will be necessary to adjust the position of the bead accordingly.

Figure 24. Effect of changing position of glass bead

The pressure at which a nebulizer operates is an important determinant of
droplet size, overall nebulization efficiency and uptake rate. In practice,
the operating pressure is a fixed value established by the manufacturer as
the optimum for the particular nebulizer design. Note, however, that
short-term fluctuations in operating pressure should be avoided since this
can cause short-term variations in the atomization process and thus
degrade analytical precision.

The rate at which solution is aspirated into the flame will affect the
instrumental response as shown in Figure 25. It can be seen that the
absorbance/uptake rate curve tends to flatten at uptake rates higher than
about 6 mL per minute, and this is generally the practical maximum for
conventional fixed nebulizers. At higher uptake rates the system tends to
saturate, and the additional solution is not effectively atomized by the
flame. Furthermore, the cooling effect of the excess solution will worsen
any tendency for chemical interference to occur.



22

Figure 25. Effect of varying the uptake rate

Differences in uptake rates between samples and standards will clearly
affect analytical accuracy, and nebulization must be identical for all
samples and standards in a particular analysis. This can only be achieved
if the samples and standards have the same physical characteristics.
Nebulization efficiency will be influenced by the surface tension of the
solutions aspirated. Generally, as surface tension is lowered, the volume
of solution reaching the flame will increase and the analytical signal will
also increase. Note, however, that this prediction may not be fulfilled for all
solvent/solute systems in all nebulizers under all operating conditions.
The uptake rate, the size of the solution droplets, and their distribution are
governed by the prevailing aerosol dynamics, and this is subject to factors
such as the positioning of the glass bead (or baffle), the driving pressure of
the inflowing oxidant, and the geometry of the nebulizer. For practical
purposes, however, the analyst must recognize that differences in surface
tension between standards and the samples are undesirable, and surface
tension should be controlled:
a) With organic solvents or solvent mixtures - samples and standards
must be of exactly the same solvent composition.
b) With mixtures of aqueous and miscible organic solutions - the
presence of a small amount of a solvent such as ethyl alcohol will
modify the surface tension considerably, and the compositions must
therefore be carefully matched.



23

c) High salt concentrations will increase the surface tension of the
solution. The effect of nebulization is less severe than either (a) or
(b) but reasonable matching between samples and standards is
required.


2. FLAMES

Since the inception of atomic absorption flame spectroscopy, a variety of gas
mixtures have been used to produce the required flame. Many of the fuel/oxidant
combinations that were tried experimentally proved to be unsuitable in one way or
another for reasons of analytical usefulness, safety, cost or convenience. In
modern flame spectroscopy, air-acetylene and nitrous oxide-acetylene are almost
universally used for practical analyses. Entrained air flames employing hydrogen
were quite widely used, mostly in association with vapour generation techniques,
but modern vapour generation systems now rely on `absorption cells' heated
electrically or by an air/acetylene flame. Air-hydrogen and nitrous oxide-hydrogen
flames are still occasionally used for some applications, but modern practice
tends to favour their replacement with acetylene flames.

The air-acetylene flame is the most commonly used flame in practical atomic
absorption analysis. Operating temperature is about 2300

C.

The nitrous oxide-acetylene flame is considerably hotter and produces a
temperature of about 3000

C. It will atomize highly refractory compounds of

elements such as aluminium, silicon, vanadium and titanium as well as the rare
earth elements even though all these form highly refractory molecules in flames.

For each of these flames, the operating temperature and the intrinsic chemical
environment of the flame will depend on the fuel-oxidant ratio used. These
conditions are characterized as:

(i) Fuel-lean or oxidizing. This is the hottest flame.

(ii) Fuel-rich or reducing. This is the coolest.

(iii) The stoichiometric, or chemically balanced flame. Temperature is
about the middle of the range; the fuel-oxidant ratio is about
half-way between fuel-lean and fuel-rich.



24
3. FLAME REACTIONS

Although flame atomization systems have been in practical use for many years,
molecular dissociation in flames has only been studied to a limited extent, and the
exact mechanisms of decomposition within the flame are not yet fully understood.
In the simplest case, decomposition is most probably a direct progression from
analyte compound through the monoxide to atomic constituents although, of
course, some monoxides are more refractory than others. Where the chemistry
and thermodynamics are more complicated, decomposition is an indirect process,
and a variety of intermediate compounds may be formed in the flame before the
analyte is finally released in atomic form. Some of these intermediate compounds
are formed by reactions with species such as C, C
2
, CH, CN and NH which are
intrinsically present in chemical flames. Of these reactions, some promote
production of analyte atoms; other reactions result in the formation of refractory
compounds which inhibit the production of the analyte atoms. In some cases,
intermediate compounds may be formed by reactions between various species
present in the sample. Again, some of these reactions may be helpful in
promoting production of analyte atoms, others may inhibit the process.

From these general concepts it will be appreciated that it is difficult to predict
theoretically the decomposition chemistry for all elements under all analytical
circumstances. But, as a result of extensive practical experience, we can specify
which flame should be used for particular elements and indicate general reasons
for the use of different flames. For the purpose of this discussion it is convenient
to classify determinations into broad categories according to the relative difficulty
of decomposition and the general nature and extent of the intermediate reactions
which interfere with the production of analyte atoms.

(a) The air-acetylene flame is almost universally used for those elements
classified as easily atomized (copper, lead, potassium and sodium for
example).

In these simple decompositions, a high proportion of the available analyte
compound is readily converted to atoms in an air-acetylene flame (the
coolest flame in practical use). Interferences are negligible, and the
chemical environment within the flame (oxidizing, stoichiometric or
reducing) is not a critical factor.

(b) It will be noted that several elements can be determined in either the
air-acetylene or nitrous oxide-acetylene flame. The monoxides of these
elements are more difficult to decompose than those in category (a). An
air-acetylene flame is useful for these elements, but is not fully effective
across the analytical range.



25

If, for example, a solution of calcium nitrate is aspirated into an air--
acetylene flame, the nitrate will decompose to the monoxide, but only a
limited proportion or the monoxide will be finally converted to atoms and
the analytical signal will be correspondingly small. In the hotter nitrous
oxide-acetylene flame, however, a significantly higher proportion of
monoxide will be converted to atoms and a stronger analytical signal will be
obtained.

While both flames can be used for determinations in this category,
interferences in the air-acetylene flame can be severe, the chemical nature
of the flame is important, and it may be necessary to take the appropriate
counter measures described under category (e).

(c) Elements in this category invariably form compounds which are intrinsically
more refractory than those in category (b) and an air-acetylene flame will
not decompose them to any significant extent. All elements in this category
must be determined in the nitrous oxide-acetylene flame.

(d) Temperature is not always the main decomposition agency. Within each
flame type the chemical nature of the flame (oxidizing, stoichiometric or
reducing) will also have a profound effect on the decomposition behaviour
of many elements.

Consider the determination of molybdenum in an air-acetylene flame. If the
fuel-to-oxidant ratio is such that the flame is chemically balanced
(stoichiometric), decomposition will proceed as far as the monoxide but
very little of the monoxide will be converted to atoms. If the flame is now
made fuel-rich, it will be cooler than the stoichiometric flame, but it will be
strongly reducing and the monoxide will be reduced to molybdenum atoms.

Similarly, a stoichiometric nitrous oxide-acetylene flame produces a
relatively poor concentration of silicon atoms. On the other hand, a fuel-
rich nitrous oxide-acetylene flame is considerably more effective.

For determinations in this category, therefore, generation of atoms is not
merely a straightforward matter of temperature, but requires an appropriate
combination of temperature and chemical environment within the flame.

(e) Determinations in this category are characterized by interferences
produced by reaction of the analyte element with other species in the
sample.




26
From the example of calcium nitrate in the air-acetylene flame it will be
remembered that this flame can be used for practical analysis even though
a limited proportion of available monoxide is decomposed.

Suppose that the original calcium nitrate solution also contains a significant
amount of the silicate of another element. In this circumstance, the
reaction in the flame will result in the formation of a calcium silicate
complex. The concentration of free calcium atoms in the flame will now be
considerably lower because the silicate complex is much more refractory
than the monoxide and the analytical response will be poor. Interferences
of this and kind are extremely severe for magnesium, calcium, strontium,
and barium because in the presence of aluminium, silicon and
phosphorous, the formation of refractory aluminates, silicates and
phosphates will seriously hinder the production of analyte atoms.

Three counter-measures are available:

(i) A nitrous oxide-acetylene flame will minimize or remove such
interferences because the flame is hot enough to decompose the
compounds concerned.

(ii) A `buffer' element may be added to `compete' with the analyte
element for attachment to the interfering group, so that there will be
complete atomization even in lower temperature flames.

For instance, if a large excess of strontium is added to our original
calcium nitrate (+ silicate) solution, most of the silicate will attach to
the strontium (because of its greater concentration) so that the result
will be a much greater concentration of free calcium atoms in the
flame. Such additives are often called `releasing agents'.

(iii) The standard solutions can be accurately matched to the samples
with respect to the interfering element. This is not of course always
possible as it demands a knowledge of sample composition which
may not be available.

If such precautions are not taken, it is obvious that seriously
incorrect results may be obtained. If, for example, a series of
calcium standards were measured (with no silicate present) and a
calibration constructed, then the measurement of unknown calcium
samples containing a reasonable concentration of silicate would
yield very low results since the calcium atomic absorption reading in
the samples would be seriously depressed.



27
4. BURNER POSITION

From the discussion on decomposition, it is evident that the analyst can maximize
the population of analyte atoms in the flame given the correct flame type, suitable
flame stoichiometry, and appropriate solution chemistry. But, before effective
analytical measurements can be made, it is necessary to ensure that light at the
characteristic wavelength passes directly through the analyte atom population.

Free analyte atoms are not distributed evenly within the flame envelope and
under given flame conditions there will be a particular zone which is more densely
populated by analyte atoms than other parts of the flame. Obviously, the
maximum analytical signal will be obtained if light at the characteristic wavelength
passes directly through the maximum population zone. But the location of this
zone within the flame is not identical for all analyses, and there is no single fixed
position for the burner which would be automatically suitable for all
determinations. It is therefore, necessary to adjust the burner position for each
separate analysis so that the maximum population zone coincides with the optical
path (see Figures 26 and 27).

Figure 26. Burner adjustment (horizontal)


Figure 27. Burner adjustment (vertical).



28

All instruments incorporate simple controls which enable the analyst to move the
burner to the position at which maximum absorbance is obtained. It is also
important to remember that altering the fuel-to oxidant ratio, or changing the total
gas flow without altering the stoichiometry can raise or lower the location of the
maximum population zone within the flame.

Consequently, whenever flame conditions are changed, it is necessary to
re-adjust the height of the burner until maximum absorbance is again obtained.

The burner can also be rotated so that the optical path passes obliquely through
the atom population. The effect of this is to shift the working concentration range
upward.



29
CHAPTER 4 OPERATION AND OPTIMIZATION

A. SPECTRAA SOFTWARE/OPERATION

1. Introduction

This short guide to operating the Varian SpectrAA Series instruments is meant to
provide the novice user with a short summary of the most essential steps in
getting started, developing an analytical method and running the SpectrAA. The
very extensive HELP information in the operating software also provides useful
information to the user.
It is assumed that the SpectrAA has been installed and tested for correct
operation by a trained Varian Service Engineer, and that the user has been fully
made aware of all the safety aspects associated with the operation of the
SpectrAA and has been given preliminary instruction in the basic operation of
the instrument hardware and software.
It is also assumed that the user has some familiarity with Windows software and
the use of a mouse. Left mouse button click will be abbreviated to simple
click and, similarly, a right mouse button click to right click. A double mouse
button click will be indicated by double click. Double clicking a window in the
Instrument page will blow it up, and double clicking the display again will restore
it. Ctrl, and double click will show all 4 displays. To select a button, click on it
with the mouse, or press the Tab key to move the focus to the button and press
Enter.
The Help function is accessed from any window, page, or dialog box in the
SpectrAA software by pressing F1 or the Help pushbutton by clicking on the ?
and dragging this button to specific items in the software.



30

2. Getting Started

1. Switch on SpectrAA Instrument Power (SPS-5 Sample Preparation System,
SIPS, if necessary)
2. Switch on PC, Monitor, & Printer
3. Turn on acetylene, air, & if necessary nitrous oxide gas supply
4. Switch on Laboratory Exhaust system
5. After Windows software is loaded, click on SpectrAA Icon to load the AA
software
6. (or click on Start, Programs, SpectrAA)
7. Main Index Screen appears

The Main Index window contains four buttons: Worksheet, Reports,
Administration, and Exit.

The function of each button is as follows:

Worksheet .................. Opens the Worksheet Window
Reports....................... Opens the Reports window
Administration............. Opens the Administration window
Exit ............................. Shuts down the SpectrAA software



31

3. Developing a Method / Worksheet

The Worksheet window allows you to set up methods, sequences and labels, and
initiate sample analysis. The worksheet is the fundamental file type for SpectrAA,
containing one or more methods, sample labels, sequence information and
analytical results. You may load an existing worksheet or develop a new one.

The Worksheet window consists of four pages: Filing, Develop, Labels and
Instrument. To access a particular page, click on the appropriate page tab
appearing under the menu bar.

The Filing page
Use the Filing page to perform basic worksheet operations such as saving,
closing and renaming worksheets, creating a template from a worksheet and
loading a different worksheet. You can also view information about the worksheet
currently open.
The Develop page
Use the Develop page to add, delete, review and modify methods, modify
sequence parameters, change the order of methods and copy methods to the
Method Library.
The Labels page
Use the Labels page to set up solution labels, weights and volumes.
The Instrument page
Use the Instrument page to control the AA and initiate analysis. Data is presented
in the form of a spreadsheet, with one row per sample. The sample labels are
displayed in the left hand column with concentration results for each element



32
across the remaining columns.

a. Creating Worksheets

Click on Worksheet


Click on New Loads a new Worksheet Template

Enter File Name and click on OK

Click on New From - loads a new Worksheet using an existing worksheet as a
template.

Click on the Worksheet to be used as a template and click
OK

Enter new worksheet name, analyst, & comment and click
OK

All parameters are copied minus result data.

Click on Open Opens an existing worksheet to view results and edit data.





33
Element Selection

Click on Develop tab to select elements
Click on Add Methods to enter each element

Element selection

The default is to Load From Cookbook methods from Varians Analytical Methods
or select Method Library to select methods previously developed (see #4 below to
store methods to Method Library).

Choose Method Type needed check Flame, Furnace, Vapor, Zeeman and click
on elements to be analyzed in your method under Element Matrix and click on
OK. Hold down the ctrl key and click on the elements to enter more than 1
element at a time and click on OK. Also, the chemical symbol or 1
st
letter of an
element can be entered under Search, choose the element under Element
Matrix and click and OK.

1) Review click on Review to look at the method parameters
set up for each element

2) Up and Down click on Up and Down to toggle between elements

3) Delete click on Delete to delete elements

4) Copy to Library click on Copy to library to copy worksheet method
to library these methods are stored under Add
Methods and can be added to worksheet under Load
From and click on Method Library to choose stored
element methods.




34

4. Edit Methods

1. Click on Develop tab

2. Click on Edit Methods tab

The following Methods screen will appear:






35
a) Typical General Settings - common settings for elements in a method.

General Settings

Click on Develop, Edit Methods and then the following tabs:

Type/Mode- Under Sampling Mode select manual or autonormal
(for autosampler)

Under Instrument Mode select absorbance or
emission mode

Flame type & Gas flows will default to Varian Cookbook Methods

SIPS (when using SIPS)








36
b) Measurement Mode
Parameters selected are defaults- can change Measurement Mode and Calibration
Mode if necessary.
Select PROMT-(Precision Optimized Measurement Time)- integrates signal
until desired % RSD is achieved or until end of measurement time.
Integration Time- 10 seconds
Read delay-10 seconds
Calibration Concentration
Precision 1% RSD
Optical
UltraAA- if using UltrAA lamps
For all coded lamps- the default is Varians recommended current, wavelength
and slit- instrument will autoselect lamp position.
For uncoded lamps it is necessary to put in lamp position.
Select BG on for s < 250 nm (also, for high dissolved solids & low analyte)




37

c) SIPS Settings

MAKE SURE BOX IS CHECKED IN TYPE/MODE TAB TO UTILIZE SIPS

Nebulizer uptake rate -5mL/min (actual uptake rate). Prior to connecting
SIPS uptake rate should be measured at 5-6 mL/min.

Right pump- none
Or Bulk Standard (select for QC spikes)
Modifier-enter Pump speed (i.e. use 10,000 ppm KCl at 10%
pump speed to obtain 1000 ppm KCl in all solutions)

Manually Entered Standards to enter specific standard values.

Dilute sample by factor to have SIPS dilute samples. This is a predilution
factor before running samples. SIPS will always dilute overrange samples.


Conditioning Pump tubing- refer to Varians recommended SIPS Pump
Tubing Procedure for conditioning new pump tubing.

Click on Instrument tab. The Condition Pump Tubing drop down menu
can be accessed from the Instrument tab.



38

d) Entering Standards

Enter standard values into table (Enter Bulk std [ ] if using SIPS). Must
have d Manually Entered Standards in SIPS tab to enter own values.

Select Calibration Algorithm:

New Rational (least squares line of best fit abs/conc vs Abs
converted to Abs. Vs Conc) a/c = ma + na
2
+ oa
3


Rational cubic spline- fits from point to point

Linear c = m + na

Quadratic c = m + na + oa
2

Cubic c = m + na + na
2
+ oa
3







39
e) SPS-5 AutoSampler Setup

Rinse Rate- 1- will rinse after every sample

Rinse Time- 5 sec- rinse between samples

Enter SIPS Bulk Std Position (if using SIPS & SPS-5)

Probe Height- defines the downward limit of the autosampler probe, measured in
mm from the bottom of the tube. Set at 0 mm to ensure full immersion.




40
f) Alignment & Rinse of SPS-5 Probe

Flame Facilities can be accessed by clicking on Instrument tab and
then clicking on Instrument on the upper left page and clicking on
Flame Facilities to view the following screen:



Click on Align to perform the autosampler probe alignment.
Probe should be positioned over the alignment peg. If not, refer to the
alignment procedure in the SPS-5 Operation Manual.
Click on Rinse to return to the rinse solution.
If using SIPS, the pumps can be stopped by clicking on Stop Pumps.
Arm pressure is released by clicking on Unload.


g) Notes

Manually enter notes which can be printed out in report. Data will be
stored with your developed Method/Worksheet, i.e. details of sampling
procedures, sample preparation, etc.




41
h) Cookbook and QCP

Cookbook contains Varians Analytical Methods and can be displayed as
Notes or a Graph.
QCP
Click on Limits on left menu. on Enabled for each specific test to
be included in method.





42
QCP
Click on specific QCP tests on the left menu and Enabled and enter
Concentration Values, High Limits, Low Limits. Error Actions, and Rate where
applicable for each specific test.

User Defined Tests
You can add specific tests not included in the QCP protocol by clicking on User
Defined and entering parameters below:




43

QCP

Click on Test Info for each specific QCP test and observe the following screen:

The Test Name and Solution Name can be changed in this screen. Also, Click
on the black arrow on the far right of Report Value and Pass Test to change
parameters for reporting values and criteria for pass test. The Error Flag can
also be changed in this screen.

i) Setting up QC spikes

To set up an automatic addition of QC spikes run using SIPS-20 & SPS-5 enable
the Use SIPS checkbox on the Type/Mode page. Select in the SIPS screen
under Right Pump Bulk Standard. On the QC Tests Page enable the QC
spike test and enter the required spike concentration in the Conc. column, error
action, limits, and rate. Make sure the right pump tubing is in your bulk standard.
In the autonormal mode upon pressing Start, the autosampler probe will go into
the bulk standard position. The right pump turns on and then the left pump and
calibrates the two pumps against each other. The second pump will automatically
spike samples on-line and perform the selected error action, depending on the
rate specified in the method.



44
5. Sample Label Setup

Click on Labels tab to view the following screen:

Double click to enter sample labels, sample weights & volumes for corrections.

Choose import labels to copy sample labels, weight, & volume factors from
another worksheet.

Click on Total Rows to enter number of samples in Worksheet.

The user can also select
Nominal weight - set at 1 and Sample Weight
Nominal volume-set at 100 and Sample Volume
Select Auto Copy to automatically increment sample labels
Select Solution Type to enter QC functions by label (label driven QC
functions)

Weight/Volume Correction
Set Nominal Weights and Nominal Volumes to 1.
Example, set sample wt. = 1, sample volume = 10 & results will be
multiplied by 10.





45

6. Printing SPS-5 Loading Guide

Click on the Setup SPS Racks icon on the Labels tab. The following
screen appears:





This screen shows autosampler positions for samples and standards, rack types,
and number of racks for the samples setup in the method. Click on Loading
Guide to print out sample labels and positions in the autosampler racks.




46
7. FS Setup

Click on Fast Sequential icon in the Develop Tab

Fast Sequential Mode Box



Click on sort methods to have your methods (elements) sorted to run from
highest wavelength to lowest wavelength.







47
8. FS Wizard
Note the FS Parameters dropdown indicating these parameters to be the same
for all elements in a FS worksheet. In edit method these parameters can only be
set for the template element indicated in red- they will be applied to all
methods/elements in the worksheet. You can also change the template element
by clicking on the element you want to use as the template as indicated in red.

A Delay in minutes can be entered to allow time for warm-up of lamps and the
burner. The method can be changed to Start With Calibration, Reslope, Cal
Zero or Solution.
Click on Finish to indicate the completion of FS setup.





48

9. Editing Sequence Parameters

Click on Edit Sequence Parameters.
A delay in minutes can be entered before running worksheet to allow for lamp
warm-up and thermal equilibrium of the burner.

Click on Options
The QC Tests box is checked as the default. QC Tests box must be d before
running a worksheet to perform selected QCP protocol and the corresponding
error actions. In compliance with GLP requirements it will not allow the user to
change parameters after a worksheet is run. This box must be unchecked to
change parameters after a run (after data is collected the QC Tests box cannot




49
be unchecked).

Check boxes to leave Lamp On, Leave Flame On, and hear Continuous Alarm
after worksheet is complete. Weight/Volume Correction box to apply
factors.

10. Setup for Autoreports


The Reports tab allows the setup of an automatic report. Reports can be sent to a
LIMS system, and the option to generate the report during run, after run, or none
can be selected.

Solution data, report contents, signal graphics, and calibrations graphs can be
checked to customize report as per laboratory requirements.

Individual Method Notes specific for each element can be printed as well as
Sequence Notes included for entire Worksheet.




50
11. Instrument Optimization

Select Instrument tab to optimize instrument and run worksheet.

a) Lamp Optimization
Click on Optimize to optimize lamps & burner. Select method (element)
to optimize and click on OK. Click OK at analysis checklist prompt. The
following screen appears:
Click on Optimize Lamps and optimize lamps by adjusting the 2 screws on
the HCLs to obtain maximum signal (Rescale if necessary). It is
recommended practice to keep a log of the % gain for each new lamp.
Good lamps properly aligned should have a low value for % gain or EHT
and a fairly steady emission signal.




51
b) D2 Lamp Optimization

For D2 Background Correction

Click on Optimize Lamps to view the following screen:

If using background correction it will be necessary to optimize D
2
lamp only when
installed. Optimize lamps by adjusting the 2 silver screws to the left of the HCLs.
Optimize for maximum intensity for HCL and D2 lamps. Lamp signals should be
balanced.

c) Optimization of Instrument

Click on Optimize Signal to observe the following screen:





52
Click on Optimize Signal to optimize flame gases, burner head, and impact bead.
Flame gases can be adjusted by holding down and moving red arrows or
manually entering values.


Aspirate 5 ppm copper solution and adjust vertical, horizontal, and rotational
adjustment for burner head. Adjust impact bead to obtain desired sensitivity. A 5
ppm Cu standard should give an absorbance range of 0.5- 0.7 Abs. The
Sensitivity Check box will give you an indication of Abs values for the indicated
concentration for each element. Also optimize flame gases by moving red arrows
or entering the flow rates.

Click on OK when finished optimizing an element & click on Cancel when finished
optimizing method.






53
12. Running a Worksheet

Click on the Instrument tab to view the following screen:

Click on Select icon to highlight samples & elements to be analyzed. The default
is to run all samples and elements in the worksheet indicated by the brownish-red
highlighted color. It is necessary to deselect samples & elements if you do not
want to run them. Simply click on individual cells to deselect (they will be white)
or on top of the element columns to deselect entire elements.

Tagging samples

Samples can also be tagged indicated by diagonal slashed lines. In Reports,
click on Select tab and under solutions click on Tagged. Only tagged samples in
the method are reported.

Click on the START stoplight icon to run method. Click on Pause icon to pause
an autorun. Click on Continue to obtain dropdown list to continue with
Calibration, Reslope, Cal Zero, or Sample Solutions.

Click on Read to do individual manual reads. Manual reads are indicated in

Select
Samples
Run
Worksheet



54
report by an..m next to the result. Read allows you to randomly select and
measure any solution. Click on Stop to close this mode.

Click on Random icon to read a random sample or repeat a sample read during
an Autorun.

13. Editing Results & Viewing Calibration Results

Each standard result can be edited by masking 1 or more replicates or the mean.

Right click on the calibration standard to be edited on lower right screen of the
calibration curve.

Click on Edit Replicates. Click on either mean (to mask all replicates) or
choose single replicates to mask.

Click on Mask, Apply, and Close. Notice a ---e in the results indicating
results have been edited. By carrying out such editing you may save time in not
having to re-run that particular standard, or even having to prepare a new
standard and a new calibration.

Click on Overlay on the calibration screen to overlay calibration curves by clicking
on individual sample cells on the Worksheet screen (top left of Instrument
screen).



55

You can also edit the curve fit by right clicking on the calibration graph screen.
Click on Curve Fit and select desired calibration algorithm.

Right Click on Parameters to view calibration results and print out calibration
results. Correlation Coefficient and Characteristic Concentration are
displayed here.

Characteristic Concentration = Concentration * 0.0044/ Absorbance

Right Click on Datalog screen (bottom left of screen) to obtain printed report of
entire log or results only.


14. Printing Reports

The Reports window allows you to generate a report for the current worksheet or
any of the worksheets saved in the system.

The Reports window consists of four pages: Worksheet, Select, Settings, and
Report. To select a particular page, click on the appropriate page tab appearing
under the menu bar.

The Worksheet page
-Use the Worksheet page to select the worksheet results to include in the
report.





56
The Select page
-Use the select page to select the elements and solutions to include in the
report. Click on All, Tagged, of enter Label Range to indicate samples
printed out in report.

15. The Settings & Reports Pages

The Settings page
-Use the Settings page to specify the report style and content.

The Report page

Use the Report page to view and print the report, write to a Text File, or




57
export to a PRN file (Write to Text File and export as .csv file to open
worksheets in Excel).


16. Administration Settings

- The Administration window allows you to: transfer worksheets in and out of the
system (e.g. archiving old worksheets on floppy disk); delete worksheets and
library methods; and activate and modify password protection of the system.





58
Password Settings


*Varian cautions the use of Password Protection. Keep password in safe place- if
you forget the password the software must be reinstalled.


17. Customizing Autosampler Racks

Click on Customize Racks in the Administration icon to observe the following
screen:


Custom racks are set up from this page for the autosampler. Click on New to
setup new racks. The following screen appears:



59

Click on First Tube Position to setup probe at first tube and then Last Tube
Position to indicate position of the last tube on the rack. Click on Test icon to
test the setup by aligning probe and moving to autosampler tube positions.


18. On-line Help

The SpectrAA software contains extensive on-line Help, which serves as your
primary source of information on how to effectively use the software and the
instrument. The Help consists of contextual help and multimedia help.

The contextual help is accessed from any window, page, or dialog box in the
SpectrAA software by pressing F1 (the Help function key) or the Help
pushbutton (where available), and provides help specific to that screen.

The multimedia help (referred to as the Multimedia Practical Guide) contains
hardware-related information and other details to help you setup, operate, and
maintain your SpectrAA instrument.

To access the Help Video clips you must have the Varian Multimedia CD
inserted into your CD-ROM drive.



60

19. Instrument Shutdown

After finishing the analysis, rinse with 50 ml of DI water or dilute acid (0.1%-1%)
followed by DI water before exiting the system.

Click on Filing tab, Save, Close, and then Exit.

Or

From Instrument Window click on exit and Shutdown SpectrAA. The flame
will automatically shut down. If d in the sequence parameters, the flame and/or
lamps will automatically turn off at end of run.

All gas supplies should be turned off at the cylinders.


B. SPECTRAA OPTIMIZATION

Your AA cannot be expected to provide maximum analytical performance unless it
is first set up correctly. Although complete details for the set up and optimization
procedures are given in the help screens of the SpectrAA software the main
optimization parameters will be discussed in this chapter.

Before samples are analyzed, the hollow cathode lamp must be aligned and then
the flame signal is optimized to give maximum signal resolution.

A. Hollow Cathode Lamp alignment

A hollow cathode lamp should be aligned if:

You need the best possible signal to noise ratio

You are using a non-Varian lamp

You are using a position previously used for a non-Varian lamp

You have just installed the lamp.

1. Select the Instrument tab.

2. Press "Optimize" (the Optimize Method Select dialog box will appear).




61
3. Select the flame method you wish to optimize from the method list box and
press "Optimize". The Flame Optimization window will appear.

Note: The Analysis Checklist dialog may appear containing a list of hardware
items that you must set up and/or confirm. Press OK once these items are
confirmed to continue with the optimization.

4. Allow the lamp time to warm up (about 10 minutes). If the lamp is not
glowing review the method and make sure that the hollow cathode lamp is
fitted at the current active position.

5. Make sure nothing is in the optical path.

6. While watching the lamp signal bar on the display, slowly turn one of the
lamp adjustment knobs. If the signal decreases, turn the knob in the other
direction, until you find the maximum signal.

7. If the HC lamp signal is too small, first check that you have the correct lamp
for the current method, and that it is glowing. If so, press "Rescale". This
will bring the signal back into range for display. You should also press
"Rescale" if the signal becomes too large.

8. Repeat the previous step with the other adjustment knob.

B. Deuterium (D
2
) lamp alignment

1. Follow steps 1-4 in the above procedure choosing a method that uses
background correction. Allow the deuterium lamp to warm up for 30
minutes before use.

2. While watching the lamp signal bar on the display, slowly turn one of the
D2 lamp adjustment knobs (located on the front of the D2 lamp
compartment). If the signal decreases, turn the knob in the other direction,
until you find the maximum signal.

3. If the D2 lamp signal is too small press "Rescale". This will bring the signal
back into range for display. You should also "Rescale" if the signal
becomes too large.

4. Repeat with the second D2 lamp adjustment knob.




62
Note: The hollow cathode lamps and deuterium lamps should be aligned
separately such that each lamp bar is showing maximum energy. In many
cases this will mean that the length of the two bars will be different.

C. Flame signal optimization

1. Press InstZero to zero the instrument.

2. Use one of the Varian supplied burner cleaning and alignment strips (or a
business card) to locate the light path.

3. Rotate the burner, squeezing the prongs of the rotation handle, until the
slot is parallel to the light path.
4. Place the card halfway along the slot. Position the card with the vertical
line perpendicular to the slot, then adjust the burner height until the light
beam falls within the target area.

5. Light the flame. Wait until the flame is burning steadily before proceeding
to step 6.

6. Press the Optimize Signal button.

7. Aspirate a solution of analyte that will give an absorbance between 0.2 and
0.8.

8. Watch the signal bar and adjust the burner height using the outer knob on
the burner adjuster to obtain the maximum absorbance, but keep the
burner below the light path. The burner height controls the sensitivity and
will also influence atomization interferences.

9. Carefully move the burner horizontally by turning the inner adjustment knob
on the burner adjuster. In general, once optimized for maximum sensitivity,
this position can be used for all analyses.

10. Now rotate the burner using the tweezer control and adjust for maximum
sensitivity.

11. Adjust the impact bead position by gradually turning the impact bead
adjustment screw first clockwise, then counter-clockwise, to find the
maximum absorbance.

Note: If you are doing an Autorun you should optimize the burner position for one
of the methods prior to starting the run. For all elements except Cr and Si



63
the burner position will be suitable for each method. Cr and Si should be
analyzed separately as their determination is critically dependent upon the
burner position.

You can also optimize the gas flow rates if required.

12. Enter a value in the Gas Flow numeric entry fields or click and drag the
red marker to specify the gas flow rates to obtain the optimum absorbance
for the current method.

The black markers can also be moved with the mouse and their positions
saved when "OK" is pressed. Use them if you need to mark any gas flow
limits for the method.

13. When the signal and lamp values are satisfactory, press "OK". The system
is now optimized. You can now take readings, measure your calibration
standards, or perform a reslope according to your analytical requirements.

Tip: Enter the actual gain, absorbance and concentration you used in the
sensitivity check text field as a reminder of the required settings.

D. Autorun set up

An Autorun allows SpectrAA to measure a batch of samples sequentially. Only the
selected samples are measured.

1. Load an existing or new Worksheet.

2. Select the Labels page tab and enter your sample labels.

3. Enter the weight and volume of each sample (if required) and specify the
nominal weight and volume (if required).

4. Select the Develop page tab and add (and modify, if necessary) Worksheet
methods. Close the Methods dialog box when complete.

5. Press "Edit Sequence Parameters" (the Sequence Parameters window
will appear).

6. Select the Control page tab.

7. Select the Method type(s) you wish to display.




64
8. For each method displayed on the Control page, specify what you want to
happen at the start and end of each, i.e. do you want to delay the start of a
method? What actions do you want at the start and end of each method
and do you want to automatically clean the graphite tube at the start of a
furnace method?

9. Select the Options page tab.

10. Use Store signal graphics if you want to store the signal graphics for each
sample.

11. Define what is to happen at the end of the autorun in the Sequence
Completion group.

12. If the calibration for a method fails, the system needs to know what you
want to do - specify this in the On calibration error listbox.

13. If you are using a SIPS accessory, use On pump overrange to specify
what you want to happen when a SIPS pump purge fails after an overrange
sample. If you are not using SIPS, ignore this field.

14. If you want to use Quality Control Tests, select the 'QC Tests' checkbox.

Note: You should set up the QC test for each method in the Methods window.

15. If you entered sample weights and volumes in step 3, and wish to use them
to calculate the final sample concentration, select the 'Weight and volume
correction' checkbox.

16. Select the Reports page tab.

17. Use the parameters in Report contents to specify what your report should
contain.

18. Select Print report (in the Setup auto report output group) to print a report
to the printer. Choose to print the report out as each sample is measured
during the sequence or when the sequence is finished (Report group).

You can select to print the report directly to your PCs serial port (choose
Output results to PC port). Alternatively, you can export the report directly
to a PRN file (choose Export report as PRN file).

19 Select the Notes page tab.



65

20. If you want to include some additional text on your report, type it in on the
Notes page.

21. Close the Sequence Parameters window and return to the Labels page.

22. If you are using a SPS or PSD, use "Setup SPS Racks" or "Setup PSD
carousels".

23. Load your samples into the SPS or PSD (if you are using these
accessories).

24. On the Instrument page, select the samples you wish to analyze.

25. Press "Start" to start the autorun.



66

























Left Blank Intentionally



67
CHAPTER 5 FLAME INTERFERENCES

1. BACKGROUND ABSORPTION

One of the major virtues of atomic absorption spectroscopy lies in the fact that
absorption occurs at specific wavelengths which are characteristic of the chemical
element. Theoretically, any absorption which occurs can be directly attributed to
the presence of analyte atoms, and the degree to which absorption occurs is a
function of the density of the atom population. In practice, however, the analytical
signal can be attenuated by means other than atomic absorption. This will be
manifested as absorbance measurements which are higher than they should be
for a given atom population, and the analytical result will not be valid.

The major sources of background absorption are molecular absorption and light
scattering. Molecular absorption occurs when the flame is not hot enough to
decompose all compounds in the sample. The remaining molecules will then
absorb light, and if the absorption spectrum overlaps the analytical wavelength
some additional absorption will be measured at the same time as the analytically
valid absorption signal. The flame itself produces absorption, especially at low
wavelengths.

Figure 28 shows the wavelength range over which sodium chloride molecules will
absorb.

Figure 28. Broad band absorption (sodium chloride)




68
Figure 29 shows the characteristic sharp line for lead. If, for example, lead is
determined in the presence of large amounts of sodium chloride, the measured
absorbance will be the sum of the atomic absorption occurring at 217 nm and the
molecular absorption also occurring at 217 nm. To obtain analytically valid
results in this kind of situation, it is necessary to use a background corrector
system.

Figure 29. Sharp line absorption (lead)

Light scattering can occur when the sample has a high concentration of dissolved
solids. Here again, the flame may not be hot enough to completely decompose
the sample, and solid particles may remain in some areas of the flame. These
particles can reflect and scatter some of the incoming light, and this will be seen
as attenuation of the transmitted light although no absorption has taken place.

With flame atomization systems, both molecular absorption and light scattering
are insignificant unless measurements are made close to the limit of detection.
Nonetheless, there are some practical analytical situations at the lower
wavelengths when background absorption can be troublesome.


2. IONIZATION

Atomic absorption measurements depend upon the presence of free, neutral
atoms, having characteristic spectral properties. At elevated temperatures, atoms
can be ionized with a consequent change in their spectral response.

Ionization means the loss of one (or more) of the atom's outermost electrons.



69
Since this electron was involved in the energy level transitions that define the
resonance absorption line, no absorption can occur at this wavelength. The
degree of ionization is different for each element, depending on the energy
required to remove electrons. This energy can be supplied in various ways, but
for atomic absorption the major source is the heat of the flame. In the nitrous
oxide-acetylene flame particularly, many elements are at least partially ionized,
but the energy available is such that normally only one electron is removed and
the singly-charged ion is formed.

Note that the degree of ionization will increase as the concentration decreases to
that instead of the more normal calibration graph (nominally linear but in practice
curving toward the concentration axis at high absorbances) the shape of the
graph will be as illustrated by curve A in Figure 30.

Figure 30. Calibration curvature caused by ionization.

Note also that all of these ionization considerations apply only in pure solutions of
the particular element. The presence of any other element with an ionization
potential close to or lower than the analyte element will modify the extent of
ionization significantly. If a solution of sodium chloride (only) is sprayed into a
flame the equilibrium between neutral sodium atoms and sodium ions is
temperature dependent.
If an excess of potassium chloride is now introduced into the sodium chloride
solution the potassium will behave similarly.
Since both of these ionizations produce free electrons the law of mass action
causes the ionization equilibria to be displaced in a direction favouring neutral
atoms and consequently a greater absorbance will be obtained for a sodium
atomic absorption measurement.



70
The results of such behaviour are easy to see. If standards containing sodium
only are used, ionization will significantly reduce the atomic absorption readings.

If, then, a sample is measured which contains potassium as well, the degree of
ionization of the sodium atoms will be decreased due to the presence of the
potassium, and consequently a relatively higher atomic absorption reading will be
obtained.

In practice, the effective means of avoiding interference due to ionization is to
`buffer' the standards and samples with a high concentration of an easily ionized
element. Provided that the concentration of this element is much greater than the
analyte element, and its ionization potential is lower, essentially complete
suppression of ionization may be effected.

Due to their low ionization potential, sodium, potassium and caesium are the most
commonly used ionization buffers.



71
CHAPTER 6 ANALYSIS CONSIDERATIONS
The procedure used to measure the absorbance of a sample involves the
following steps once the analytical method has been developed:

(a) Preparation of sample solutions - the sample is dissolved in the correct
solvent system after any necessary decomposition procedures,
separations, additions and dilutions have been made. (Preparation of
standard solutions, usually by dilutions from stock solutions will also be
necessary for the measurement of concentration.)

(b) Adjustment of the instrument - the previously determined optimum
operating conditions must be achieved. This includes setting of
wavelength control, source current, slits, the detector and recording
system, the flow rates of gases if a flame is being used, and checking the
light path from source to monochromator entrance slit with respect to
burner height, etc.

(C) Zeroing the instrument on distilled water then determining the absorbance
of the blank.

(d) Aspirating the standard solutions and recording their absorbances.

(e) Nebulizing the sample solution and reading its concentration from the
calibration graph.

CALIBRATION PROCEDURES

There are two commonly employed approaches to calibration in atomic absorption
spectroscopy. These are the methods of direct comparison and standard
addition.

(a) DIRECT COMPARISON

Theoretical considerations lead to the prediction that the measured absorbance
will increase linearly with increasing analyte concentration in conformity with
Beer's law.

At low concentrations this prediction is generally fulfilled and absorbance does
increase almost linearly with the analyte concentration. As the concentration
increases, however, the correlation becomes increasingly curved. Traditionally,
the shape of this curve was deter- mined experimentally for each analysis.
Absorbance measurements were obtained for standards of known concentration,



72
and the analyte concentration in the sample was found by interpolation on the
curve. This graphical method is no longer required for routine analysis because
instruments can be calibrated to provide direct concentration measurements in
whatever units the operator chooses. Nonetheless, the graphical method is still a
valuable aid in establishing optimum working ranges and instrumental parameters
when developing or investigating an analytical method.

But, calibration linearity is not the only matter of analytical concern. The practical
analyst needs to know the capabilities of his spectrophotometer. For purposes of
comparison, three parameters are used to describe the analytical performance of
an atomic absorption spectrophotometer. All three are attempts to characterize
the useful range of measurement of the instrument for each element. The three
parameters are:

(a) Characteristic concentration

This is defined as the concentration, in solution, of the element to be determined
which will produce a change, compared to a blank solution, of 0.0044 absorbance
units (i.e. I per cent absorption) in the optical transmission of the atomic vapor at
the wavelength of radiation used.

Note that different instruments will produce different characteristic concentrations.
On any given instrument, however, there will be little variation from day-to-day
provided that the operating conditions are not changed and the instrumental
conditions are optimized The characteristic concentration is easily measured and
can there- fore be used as a convenient check that the instrument is performing
correctly. Also, it enables an analyst to calculate the absorbance which will be
produced by a particular concentration of a given element. In practice a
calibration graph is derived from measurements of a series of standard solutions.
Using the mathematically linear portion of the graph, read off the absorbance 'A'
given by concentration 'C'.

Then: characteristic concentration = C x 0.0044
A

(b) Detection limit

The detection limit is that concentration which can be detected with 95%
confidence. This is the concentration which gives an absorbance equal to twice
the standard deviation of a series of measurements at, or near, the blank level.
The practical virtue of the detection limit is that it indicates the concentration at
which useful measurements can be made. Since the detection limit is the
concentration giving a relative standard deviation of 50%, it is obviously not



73
possible to make precise measurements at this level. But, at ten times the
detection limit, the relative standard deviation should be about 5% which
establishes the lower limit of useful measurement.
(c) Working range
This is the range of concentration of analyte element that will produce suitable
absorbance values for practical analysis. It is impossible to be specific in defining
this range since analytical needs vary considerably. Generally, however, the
working range should lie between 0.I and 0.8 absorbance since this is generally
the region of optimum optical precision. But useful measurements can be made at
lower absorbances, and the overall working range can extend to a lower limit that
is about ten times the detection limit. For flame analyses, calculating the working
range is simple: Instrument manufacturers usually indicate the characteristic
concentration that can be expected for each element in aqueous solution. This
value can then be used to establish a nominal working range for the element of
interest. Since by definition the characteristic concentration yields an absorbance
of 0.0044, simple multiplication by 100 gives the concentration required to obtain
an absorbance of 0.44 (about the middle of the optimum range). Once this 'bench
mark' has been established, it is a simple matter to calculate the overall working
range together with any dilution or concentration factors which may be needed for
the specific analytical situation. In situations where the sample is complicated or
when little is known about the sample composition, it may be necessary to
establish the absorbance range by trial and error, and it is good practice to
produce an initial calibration graph before embarking on the final analytical
routine.

Adjusting the working range
Where the concentration range of interest is above the working range obtainable
at practical absorbances, the analyst can either dilute all solutions, or alter the
instrumental sensitivity.

In flame analyses, selecting alternative (less sensitive) wavelengths, rotating the
burner across the optical path, or doing both will allow determinations to be
carried out at significantly higher concentrations as shown in the following Table.

Burner
Position
Wavelength
nm
Relative
Sensitivity
Working Range
mg/L
240.7 1 0.06 15
Normal
304.4 16 1.0 240
240.7 20 1.2 300
Rotated
90
304.4 320 20 - 4800

TABLE: Working Concentrations (cobalt)




74
(b) THE METHOD OF STANDARD ADDITION

In this method, physical and chemical mismatch between samples and standards
is minimized because the standards are prepared from the actual samples.

The method assumes that the added analyte will be affected by interference in the
same way and to the same extent as the analyte in the sample. This is generally
valid for physical interferences. It is not always valid for chemical interferences
and in these situations the interference should be removed by appropriate means
(a hotter flame, or a releasing agent, for example).

The general procedure is to take several aliquots of sample, and then add
different quantities of the analyte element to each aliquot. After dilution to the final
volume, these solutions form a set of standards of different concentration. One
aliquot is diluted without any addition of analyte element. The quantities of analyte
added are based on the expected analyte concentration in the sample. In a typical
additions scheme the analyte additions are 50%, 100% and 150% of the expected
concentration.

The analytical result can be obtained by plotting absorbances against added
concentration and extrapolating the graph to zero as shown in Figure 31.
Alternatively, the instrument can be directly calibrated against the addition
standards. The built-in computer program will then calculate a least-squares
regression line through the calibration points and compute the sample
concentration directly by mathematical extrapolation.

Figure 31. Standard additions calibration

NOTE: The curve does not pass through the origin since the x-axis
represents the metal added and not the total metal concentration.



75

As a general rule, the addition standards must provide substantially linear
calibration since accurate regression cannot be obtained from non-linear
calibration points. It is essential to establish an accurate baseline from the
appropriate reagent blank.

In the most common applications of the standard additions method, each sample
must be analyzed individually against a set of standards which are specific for
that sample. There are some analytical situations in which a batch of samples can
be analyzed against the one set of addition standards, but this is only valid when
all samples in the batch are chemically and physically similar.

ACCURACY AND PRECISION

The accuracy of a determination is a measure of how close the value obtained is
to the "true" value. Unfortunately, the true value of a chemical constituent of a
substance cannot be known absolutely; hence the assessment of accuracy is, to a
large extent, subjective.

It is possible to develop standard reference materials that can be used to give an
approximation of the accuracy of a determination, i.e., a large number of samples
of the desired material are collected and a standardization programme involving
as many as possible different, but suitable, analytical techniques is followed. The
data is then processed statistically and a set of accepted values generated.

Precision of determinations is a measure of the degree of scatter obtained on
replication of the analysis of the sample. Frequently good precision is mistakenly
taken to indicate good accuracy. In many cases, sampling may account for the
greatest source of error. Sampling procedures seldom receive the same critical
study as have methods of analysis; however, a discussion of sampling is beyond
the scope of this course. Thus precision can vary considerably with
concentration, nature of the sample and element determined.

SENSITIVITY AND DETECTION LIMIT

The term sensitivity is used to indicate the slope of the calibration curve at the
stated concentration of the analyte. On the other hand, the detection limit is that
concentration of an element that gives a signal equal to some multiple, usually
twice, the standard deviation of the noise.

Sensitivity and detection limit for a given element depend on many factors. The
most important are the resonance wavelength chosen, the type of instrument, the
type of atomizer, the choice and adjustments of burner, flame and nebulizer, and



76
the type of solvent. In addition, the detection limit is dependent on the nature and
amount of dissolved matrix constituents in the solvent.

The concentration range over which the calibration curve is linear, or nearly so, is
not always very large. In general, atomic absorption studies are most used at
very low concentration Ievels to take advantage of the sensitivity of the method.
Measurements at higher concentration can only be made either after dilution of
the sample or after reduction of sensitivity of measurement.

SAMPLES AND STANDARDS

1. GLASSWARE AND OTHER CONTAINERS
All glass apparatus and containers used in analytical work must be carefully
selected to meet the requirements of its particular use. Although several types of
special-purpose glasses are available, borosilicate thermal-resistant types such
as Pyrex or Kimax, are generally satisfactory for all ordinary laboratory purposes
in water analysis. Borosilicate glass is especially suitable for storage of neutral or
acid solutions, for volumetric glassware, and for conducting reactions. Since
borosilicate glass is not entirely resistant to attack by strongly alkaline solutions,
bottles of polyethylene or Teflon must be used for storage of standard solutions of
silica, boron and the alkali metals.

All volumetric glassware, such as burettes, pipettes and volumetric flasks, must
be of borosilicate glass and must contain or deliver volumes within the tolerances
of the method. In addition, if such glassware is frequently used to measure
strongly alkaline solutions it must be recalibrated at regular intervals. Directions
for such calibration and testing of volumetric glassware are given by the National
Institute of Standards (1959) and in standard texts of quantitative analysis.

Evaporations may be carried out in glass, porcelain or platinum dishes. Platinum
is preferred if the mass of the residue must be determined accurately, because
the mass of platinum vessels is very constant.

Although platinum is one of the most resistant metals, it is not completely inert
and is subject to embrittlement. The following precautions are recommended:

Never put solutions containing tin, mercury or lead in a reducing environment
in platinum; if the free metal should be formed it will alloy with the platinum,
especially if heated.
Do not heat mixtures of hydrochloric acid with oxidizing substances, such as
nitrate or manganese dioxide; ferric chloride in hydrochloric acid attacks
platinum appreciably.



77

Place hot platinum vessels on a refractory material never on a cold metal
surface or on a dirty surface.
Use only clean platinum-tipped tongs to handle hot platinum vessels.

Coarse crystal growth and embrittlement may result from prolonged heating at
high temperatures, heating under reducing conditions, and heating phosphates
or sulphates in the presence of organic compounds. Embrittlement can be
counteracted by rubbing the platinum ware with moistened sea sand. Gentle
rubbing with sea sand cold works the metal and breaks down the coarse crystal
structure. Detailed instructions for the care and use of platinum are distributed
by manufacturers of these vessels and are described in textbooks of
quantitative analysis.

2. CHEMICALS AND SOLUTIONS

a) PURITY
Unless indicated to the contrary, all chemicals specified for use in the
analytical procedures shall conform to the specifications of the Committee
on Analytical reagents of the American Chemical Society. Chemicals used
for primary standards may be obtained from the National Institute of
Standards or from manufacturers marketing chemicals of comparable
purity.

Water used to dilute samples or to prepare chemical solutions shall first be
demineralized by passage through mixed cation-anion exchange resins or
by distillation. Its specific conductance at 25

C must not exceed 1.5

mho/cm, and it shall be stored in resistant glass or polyethylene bottles.

Carbon dioxide-free water may be prepared by boiling and cooling
demineralized water immediately before use. Its pH should be between 6.2
and 7.2.

Ammonia-free water may be prepared by passing distilled water through a
mixed-bed ion-exchange resin.

b) STANDARD SOLUTIONS
The concentrations of standard solutions are indicated as the mass of a
given element equivalent to, or contained in, 1 mL of solution. The
strengths of acids and bases are given in terms of molarities.

c) NONSTANDARD SOLUTIONS



78
The concentrations of nonstandard solutions are indicated in terms of the
mass of solute dissolved in a solvent and diluted to a given volume.
Unless specifically indicated otherwise, it is understood that the solvent is
demineralized water of required purity. Designation of concentration in
terms of percent is not used.

d) ACCURACY OF MEASUREMENT
Within the methods significant figures are utilized to define the accuracy of
weights and measures. Weighings will be accurate to the last figure
shown. For example, a mass designated as 4.532 g must be weighed
accurately to 0.0005 g, whereas a mass designated as 4.5 g must be
weighed accurately to only 0.05 g.

Required accuracy for measurement of volume in the analysis and
preparation of reagents is shown similarly. Standard solutions are always
prepared in and measured from volumetric glassware. The significant
figures given for such measurements are in practical agreement with the
tolerance limits for volumetric glassware used.

For example, "Add 2.0 mL of reagent" shows that a volumetric pipette must
be used for the addition, but add "2 mL" or "add 1.5 mL" shows that a
serological pipette may be used; "dilute to 1000 mL" shows that a
volumetric flask is essential, but "dilute to IL" permits the use of a
graduated cylinder.

Although pipettes are calibrated to deliver a specific volume at 20

C, the
error in measurement incurred by pipetting samples at room temperature is
insignificant for water analysis. One gram of pure water is contained in
1.002 mL at 20

C and in 1.007 mL at 38

C; the maximum error in volume

that will result from those temperature differences is only 0.5 percent.
Brine samples should be brought as near to 20

C as possible before
making dilutions for analysis because of density differences.


3. DILUTIONS OF WATER SAMPLES

The concentration of some inorganic constituents in a water sample very often will
exceed the working range as recommended in the application section of each
method. For example, sodium in brines will exceed the recommended range
several fold. Dilution is normally used to bring the concentration of any of these
constituents into the appropriate range. This procedure with proper technique is
satisfactory for dilutions of 1 to 1000. The following techniques must be observed:



79

No more than two sequential dilutions can be made on a sample; for example
(1 : 1000), 5.0 mL diluted to 500 mL and then 5.0 mL diluted to 50.0 mL.

The dilution procedure above is for single pass data only. If more than a 1 to
1000 dilution must be made, the sample must be analyzed in triplicate. This
will require making sequential dilutions of the sample in triplicate, and
reporting the average of the three determinations.



80
CHAPTER 7 AA MAINTENANCE AND SAFETY
For many analysts the routine measurement of metallic elements by AAS is often
hindered by the instability of their analytical results. This phenomenon usually
manifests itself in high relative standard deviations being obtained. Instrumental
instability was once strongly associated with poor electronic stability. The
technology today has so advanced with "state-of-the-art" electronics that most
causes of instability can be associated with either a lack of basic maintenance or
poor optimization. This section details some of the important points associated
with regular maintenance which can be easily carried out by the operator.

BURNER

soak in dilute soap solution, wash thoroughly with distilled water and dry
completely before use.
clean the Mark VI and VII burner slots with a cleaning card moistened with
"Brasso"

NEBULIZER

clean regularly. Aspirate the blank solvent (usually distilled water) at the end
of each day to thoroughly flush out the nebulizer.
use only the special wire provided for cleaning capillary (nebulizers are
precision manufactured components).
use a tantalum venturi for corrosive samples (all Varian nebulizers have Pt/Ir
capillaries).

GLASS BEAD
check surface regularly for cracks or abrasions.
check the in/out adjustment mechanism moves freely.

DRAIN
ensure clear plastic tubing is not too long.
ensure tubing travels downwards and not horizontally along bench.
flush regularly at the end of day.
ensure that liquid level in waste receptacle is lower than drain tube (this will
allow free drainage).
check liquid drains evenly down drain tube




81
SPRAY CHAMBER

check that the surface of the spray chamber is thoroughly wetted by the
solvent. The internal surface is `rough' to ensure this happens. Unstable
signals may otherwise result.
clean the inside of the chamber with detergent cleaner if the surface does not
wet completely.
check that the spray chamber drains correctly and that there is no obstacle or
solid matter in the drain tube. Clean thoroughly if necessary.
check 0-rings regularly
when changing from organic to aqueous and vice-versa, rinse the whole of the
interior of the spray chamber with acetone, then the respective solvent to be
used.

ENVIRONMENT
be sure no draughts are interfering with the flame - keep windows closed
always use the chimney provided with the instrument
ensure the venting system is positioned at the correct height above the bench
always use the front window guard - this will keep harmful glare away from
your eyes and help to keep the flame free from draughts

At the end of the day it is good practice to spray distilled water into the flame for
10-15 minutes. This will help to clean the entire burner/atomizer system.

Clean the instrument regularly, and also the surrounding bench - don't forget that
dust under your instrument will absorb corrosive vapours from the laboratory
atmosphere, and lead to corrosion.

Remember, your atomic absorption spectrophotometer is a sophisticated
instrument. Treat it as such.