|
INTRODUCTION TO GRAPHITE
FURNACE ATOMIC ABSORPTION
CONSIDERATIONS IN ULTRA TRACE ANALYSIS
Performance Criteria
Differences in graphite furnace performance
characteristics require redefinition
of some basic AA performance criteria.Since the magnitude
of the graphite fur-nace
signal observed depends on analyte mass rather than
concentration,the term
‘‘characteristic mass’’is used as a measure of
the sensitivity of the furnace.Char-acteristic
mass is analogous to characteristic concentration for
flame AA except
that mass,rather than concentration,is related to
absorbance.The characteristic
mass of an analyte is defined as the mass of analyte in
picograms required to pro-duce
a peak height signal of 0.0044 absorbance or an
integrated peak area signal
of 0.0044 absorbance-seconds (A×s).
Similar to characteristic concentration for flame
AA,characteristic mass may be
used as an indicator of instrument optimization.Typical
characteristic mass values
for a properly adjusted instrument are usually given in
the instrument documen-tation.
Experimental values for characteristic mass (mo)can
be determined for
comparison by measuring the peak area absorbance of a
known mass of analyte
and calculating mo according
to the following equation.
mo (pg)=Sample
vol.(mL)x Analyte Conc.(mg/L)x 0.0044 AÃ s
Observed Peak Area (A×s)
Note that for this equation to be valid,the analyte mass
used to measure mo must
produce a signal in the linear range of the calibration
curve.By comparing the ex-perimental
value obtained from this equation to the reference value
given in the
instrument documentation,the state of instrument
optimization can be evaluated.
If the calculated mo is
significantly greater than the reference value,adjustments
should be made to improve the sensitivity of the
measurement.Calculated values.for mo which
are significantly less than the reference value,suggesting better than
specified sensitivity,may in fact be a warning sign of
analyte contamination in the
standard.
Another term characterizing graphite furnace instrument
performance is ‘‘detec-tion
limit’’.Similar to flame AA,the detection limit is an
indicator of the lower
limits of analyte detectability and is limited by the
instrument signal-to-noise ratio.
Graphite furnace detection limits are usually stated in
mass units (pg)rather than
concentration units,again reflecting the furnace
signal’s dependency on analyte
mass,rather than concentration.However,in real analytical
situations the detec-tion
limit will depend both on the system sensitivity
(characteristic mass)and the
maximum sample size which can be accommodated.
Graphite Furnace Applications
The sensitivity of graphite furnace atomic absorption
makes it the obvious choice
for trace metal analysis applications.Routine
determinations at the mg/L level for
most elements make it ideal for environmental
applications.Advances in instru-mentation
and techniques have made it possible to analyze very
complex sample
matrices,such as those frequently found in biological and
geological samples.
The microliter sample sizes used offer additional
benefits where the amount of
sample available for analysis is limited,as in many
clinical analyses.
COMPONENTS OF THE GRAPHITE FURNACE SYSTEM
The graphite furnace is made up of three major
components,the atomizer,the
power supply,and the programmer.The atomizer is located
in the sampling com-partment
of the atomic absorption spectrometer,where sample
atomization and
light absorption occur.The power supply controls power
and gas flows to the at-omizer
under the direction of the programmer,which is usually
built into the
power supply or spectrometer.A description of each of
these major components
follows.
The Graphite Furnace Atomizer
A basic graphite furnace atomizer is comprised of the
following components:
· graphite tube
· electrical
contacts
· enclosed
water cooled housing
· inert
purge gas controls
A graphite tube is normally the heating element of the
graphite furnace.The cy-lindrical
tube is aligned horizontally in the optical path of the
spectrometer and
serves as the spectrometer sampling cell.A few
microliters (usually 5-50)of sam-ple
are measured and dispensed through a hole in the center
of the tube wall onto
the inner tube wall or a graphite platform.The tube is
held in place between two
graphite contact cylinders,which provide electrical
connection.An electrical po-tential
applied to the contacts causes current to flow through
the tube,the effect
of which is heating of the tube and the sample.
The entire assembly is mounted within an
enclosed,water-cooled housing.Quartz
windows at each end of the housing allow light to pass
through the tube.The
heated graphite is protected from air oxidation by the
end windows and two
streams of argon.An external gas flow surrounds the
outside of the tube,and a
separately controllable internal gas flow purges the
inside of the tube.The system
should regulate the internal gas flow so that the
internal flow is reduced or,pref-erably,
completely interrupted during atomization.This helps to
maximize sample
residence time in the tube and increase the measurement
signal.Figure 5-1 illus-trates
one type of atomizer assembly,a longitudinally-heated
furnace.
The tube in Figure 5-1 is heated by passing electrical
current from the graphite
contacts at the ends of the tube through the length of
the tube.This type of furnace
is similar to the original design of Massmann,which is
the basis for most currently
available commercial graphite furnace systems.
The longitudinally-heated furnace has a major
liability.The electrical contacts at
each end of the tube must be cooled.As a result,there
must always be a tempera-ture
gradient along the length of the tube,the tube ends
adjacent to the electrical
contacts being cooler than the central portion.This
temperature gradient can cause
vaporized atoms and molecules to condense as they diffuse
to the cooler tube ends.
This may produce interferences,the most common type being
the incomplete re-moval
of analyte or matrix from the tube.Incomplete removal of
matrix during
pyrolysis can increase the magnitude of background
absorption during atomiza-tion.
Incomplete removal of analyte during atomization is more
serious.It creates
"carryover"or "memory",wherein a
portion of the analyte in the current sample
remains in the tube and contributes to the analytical
signal for the following sam-ple.
This produces erroneously high analytical results and
poor precision.
To minimize carryover,most longitudinally-heated furnace
heating programs use
one or more cleanout steps after the atomization step.A
cleanout step involves the
application for several seconds of full internal gas flow
and a temperature equal
to or greater than that used for atomization to remove
residual sample components.
While this technique works well for the more easily
atomized analytes,it is not
always successful with those analytes that require higher
atomization tempera-tures.
The use of a high temperature cleanout step may also
reduce tube lifetime.
The transversely-heated graphite furnace eliminates many
of the problems asso-ciated
with the longitudinally-heated furnace.The graphite tube
of the trans-versely-
heated furnace,shown in Figure 5-2,includes integral tabs
which protrude
from each side.These tabs are inserted into the
electrical contacts.When power
is applied,the tube is heated across its circumference
(transversely).By applying
power in this manner,the tube is heated evenly over its
entire length,eliminating
or significantly reducing the sample condensation
problems seen with longitudi-nally-
heated furnace systems.
An additional advantage of the transversely-heated
furnace is that it allows the use
of longitudinal Zeeman-effect background correction.As
described in Chapter 3,
longitudinal Zeeman offers all of the advantages of
transverse Zeeman correction
without the need to include a polarizer in the optical
system.This provides a sig-nificant
improvement in light throughput.
The Graphite Furnace Power Supply and Programmer
The power supply and programmer perform the following
functions:
· electrical power control
· temperature
program control
· gas
flow control
· spectrometer
function control
The power supply controls the electrical current supplied
to the graphite tube,
which causes heating.The temperature of the tube is
controlled by a user-specified
temperature program.Through the programmer the operator
will enter a sequence
of selected temperatures vs.time to carefully
dry,pyrolyze,and finally atomize
the sample.The program may also include settings for the
internal inert gas flow
rate and,in some cases,the selection of an alternate
gas.Certain spectrometer
functions,such as triggering of the spectrometer read
function,also may be pro-grammed
and synchronized with the atomization of the sample in
the furnace.
SUMMARY OF A GRAPHITE FURNACE ANALYSIS
A graphite furnace analysis consists of measuring and
dispensing a known volume
of sample into the furnace.The sample is then subjected
to a multi-step tempera-ture
program.When the temperature is increased to the point
where sample atomization occurs,the atomic absorption measurement is
made.Variables under op-erator
control include the volume of sample placed into the
furnace and heating
parameters for each step.These parameters include:
1)temperature final temperature during step
2)ramp time time for temperature increase
3)hold time time for maintaining final temperature
4)internal gas gas type and flow rate
In addition to the above,spectrometer
control functions can be programmed
to occur at specified times within the
graphite furnace program.While the
number of steps within each program
is variable,6 steps make up the typical
graphite furnace program.These
steps include:
1)Drying
2)Pyrolysis
3)Cool down (optional)
4)Atomization
5)Clean out
6)Cool down
Figure 5-3 illustrates a typical graph-ite
furnace program.The following
paragraphs will discuss each operator
controlled variable,and how they
may affect the analysis.
Sample Size
Since the graphite furnace signal depends on analyte
mass,the operator has an ef-fective
degree of control on measured absorbance by controlling
the sample vol-ume.
Larger volumes of sample solution contain more analyte
and result in greater
signals.The analytical range of furnace analysis can
therefore be controlled,to
some extent,by varying sample volume.For very low
concentrations,the maximum volume of analyte can be used,while for higher
concentrations,the sample
volume can be reduced.Smaller sample volumes can also be
used where sample
availability is limited or where background absorption is
excessively large.
The maximum volume of sample usable will depend on the
tube configuration.
Where the graphite platform is not used,sample volumes up
to 100 mL can be used,
depending on the type of tube and sample.With the
platform in place,a sample
volume of less than 50 mL is recommended.A convenient
sample volume for most
analyses is 20 mL.Where larger volumes are
required,i.e.,for improved detection
limits,multiple injections can be used with appropriate
drying and pyrolysis steps
between each injection to increase the effective sample
size.
The use of an autosampler is strongly recommended for
dispensing samples into
a graphite furnace.While skilled operators may obtain
reasonable reproducibility
by manual injection on a short term basis,autosamplers
have been proven to pro-vide
superior results.With many graphite furnace
systems,autosamplers can also
generate working standards from stock standard
solutions;add appropriate re-agents;
and provide method of additions analyses or recovery
measurements,all
automatically.
The Drying Step
After the sample is placed in the furnace,it must be
dried at a sufficiently low tem-perature
to avoid sample spattering,which would result in poor
analytical preci-sion.
Temperatures around 100-120 o C
are common for aqueous solutions.
Use of a temperature ‘‘ramp’’provides a variable
time over which the temperature
is increased.A longer ramp time provides a slower,more
‘‘gentle’’increase in
heating.When a platform is used,the temperature lag of
the platform versus the
tube walls provides a natural
‘‘ramping’’effect.Therefore shorter ramp times are
usually used with the platform.Longer ramp times are used
when the sample is
to be atomized from the tube wall.
After the temperature ramp,the furnace is held at the
selected drying temperature
until drying is complete.Since only a few microliters of
sample are used,the dry-ing
‘‘hold’’time is usually less than a minute.
During the drying process,the internal gas flow normally
is left at its default maxi-mum
value (250-300 mL per minute)to purge the vaporized
solvent from the tube.
The Pyrolysis Step
The purpose of the pyrolysis step (sometimes referred to
as the ashing,char or pre-treatment
step)is to volatilize inorganic and organic matrix
components selec-tively
from the sample,leaving the analyte element in a less
complex matrix for
analysis.During this step,the temperature is increased as
high as possible to vola-tilize
matrix components but below the temperature at which
analyte loss would
occur.
The temperature selected for the pyrolysis step will
depend on the analyte and the
matrix.Suggested temperatures normally are provided in
the documentation sup-plied
with the graphite furnace.The internal gas flow is again
left at 250-300 mL
per minute in the pyrolysis step,to drive off volatilized
matrix materials.For some
sample types,it may be advantageous to change the
internal gas,e.g.,to air or oxy-gen,
during the pyrolysis step to aid in the sample
decomposition.
The Pre-atomization Cool Down Step
With longitudinally-heated furnaces,it is frequently
advantageous to cool the fur-nace
prior to atomization since the heating rate is a function
of the temperature
range to be covered.As the temperature range is
increased,the rate of heating also
increases.The use of a cool down step prior to
atomization maximizes the heating
rate and extends the isothermal zone within the tube
immediately after heating.
The extended isothermal zone has been shown to improve
sensitivity and reduce
peak tailing for a number of elements,including those
which characteristically are
difficult to atomize in the graphite furnace.
A pre-atomization cool down step normally is not required
for transversely-heated
furnaces as the isothermal zone extends the length of the
tube with that type of
system.
The Atomization Step
The purpose of the atomization step is to produce an
atomic vapor of the analyte
elements,thereby allowing atomic absorption to be
measured.The temperature in
this step is increased to the point where dissociation of
volatilized molecular spe-cies
occurs.
The atomization temperature is a property of the analyte
element.By following
recommended procedures of analysis,it is usually possible
to use the temperatures
provided in the graphite furnace documentation without
further optimization.
Care should be taken to avoid the use of an excessively
high atomization tempera-
ture,as the analyte residence time in the tube will be
decreased and a loss of sen-sitivity
will occur.Also,the use of excessively high atomization
temperatures can
shorten the useful lifetime of the graphite tube.
For atomization,it is desirable to increase the
temperature as quickly as possible.
Therefore,ramp times normally will be set to minimum
values to provide the high-est
heating rate.It also is desirable to reduce
or,preferably,to totally interrupt the
internal gas flow during atomization.This increases the
residence time of the
atomic vapor in the furnace,maximizing sensitivity and
reducing some interfer-ence
effects.At the beginning of this step,the spectrometer
‘‘read’’function is trig-gered
to begin the measurement of light absorption.
The Clean Out and Cool Down Steps
After atomization,the graphite furnace may be heated to
still higher temperatures
to burn off any sample residue which may remain in the
furnace.An optional cool
down step then allows the furnace to return to near
ambient temperature prior to
the introduction of the next sample.With some systems,a
preset cool down step
is automatically included in each furnace cycle,and need
not be programmed sepa-rately.
FAST FURNACE ANALYSIS
The most time consuming portions of a graphite furnace AA
analysis are the dry-ing
and pyrolysis steps.Analysts have long sought some means
to reduce or elimi-nate
the time required for these pretreatment stages.For
years,an inability to
compensate for the high background signals generated with
incomplete pretreat-ment
and matrix-related interferences precluded reducing the
time required for
these steps.
The situation changed dramatically with the introduction
of two developments:
Zeeman effect background correction and Stabilized
Platform Graphite Furnace
(STPF)technology,which is described in detail in Chapter
6.Zeeman effect back-ground
correction is far superior to continuum source background
correction in
its ability to correct accurately for high levels of
background absorption.The use
of STPF technology provides almost constant
characteristic mass values inde-pendent
of sample matrix.The combination of these two techniques
was the key
to providing faster graphite furnace analyses.
In fast furnace analysis,the pyrolysis step and matrix
modification are usually
eliminated.The drying step is minimized by injection of
the sample onto a pre-heated
platform.Drying time is also minimized by using as small
a sample size
as possible consistent with the required analytical
precision and detection limits.
To further reduce the furnace program time,the cool down
step at the end of the
program is frequently reduced or eliminated.This is
possible since most furnace
systems typically require 20 seconds or more between the
end of a temperature
program and the point at which the next sample is ready
to be introduced into the
furnace.That 20 second period is usually sufficient to
allow the furnace to cool
reproducibly to the preheated drying temperature required
for the next sample.
Using these procedures,typical furnace program times
frequently can be reduced
from 2-4 minutes to 30 seconds or less per
determination,a considerable time sav-ings,
without sacrificing analytical precision or accuracy.
Fast furnace analysis techniques are not compatible with
all AA instrumentation.
To fully realize the benefits of fast furnace
analysis,the instrumentation used must
provide Zeeman background correction,be capable of
handling very high absor-bance
measurements and be compatible with the requirements of
STPF technol-ogy.
Also,the fast furnace technique may not be compatible
with all sample types.
Very complex matrices may still require at least a short
pyrolysis step and use of
a matrix modifier for optimum results.
MEASURING THE GRAPHITE FURNACE AA SIGNAL
Nature of the Graphite Furnace Signal
In flame atomic absorption,the absorption signal is
steady state.That is,as long
as solution is aspirated into the flame,a constant
absorbance is observed.For
graphite furnace analyses,however,the signal is
transient.As atomization begins,
analyte atoms are formed and the signal
increases,reflecting the increasing atom
population in the furnace.The signal will continue to
increase until the rate of atom
generation becomes less than the rate of atom diffusion
out of the furnace.At that
point,the falling atom population results in a signal
which decreases until all at-oms
are lost and the signal has fallen to zero.To determine
the analyte content of
the sample,the resulting peak-shaped signal must be
quantitated.
Peak Height Measurement
For many years,measuring the height of the transient
signal was the only practical
means for quantitating furnace results.The constantly
changing signal was moni-tored
on a strip chart recorder,and peak height was measured
manually in chart
divisions.Later instrumentation allowed peak height to be
measured directly by
electronic means.
While peak height does depend on the analyte
concentration in the sample,it is
also affected by other factors.Peak height is only a
measure of the maximum atom
population which occurred in the furnace during
atomization.If matrix compo-nents
in the sample affect the rate of atom formation,the
maximum atom popu-lation
and the peak height are also affected,as shown in Figure
5-4 for the
determination of lead in blood.While the two solutions
contain identical amounts
of lead (0.2 ng),the peak shapes and appearance times are
dramatically different.
This susceptibility to matrix effects makes graphite
furnace AA vulnerable to in-terferences
when peak height measurement is used for
quantitation.Therefore,
peak height measurements are seldom used with modern
graphite furnace AA sys-tems.
Peak Area Measurement
Modern instrumentation provides the capability to
integrate absorbance during the
entire atomization period,yielding a signal equal to the
integrated peak area,that
is,the area under the peak signal.If the temperature in
the furnace is constant dur-ing
the measurement process,the peak area will represent a
count of all atoms pre-sent
in the sample aliquot,regardless of whether the atoms
were generated early
or late in the atomization process.Integrated peak area
measurements (A .s)are
independent of the atomization rate,and are therefore
much less subject to matrix
effects as shown in Figure 5-4.As a result,peak area is
preferred for graphite fur-nace
analysis.
|
