Atomic absorption spectrometry pregnant again after 45 years

1. Introduction

Atomic absorption spectrometry AAS is not a
technique that one gets excited about at first
glance. One has to come a little bit closer and
look twice in order to find out about its charm.
Maybe this is the reason why, after the first
complete description of the processes of absorption
and emission of radiation by atoms in flames
w x by Kirchhoff and Bunsen 1]3 in the 1860s, it
was optical emission that caught the interest of
spectroscopists, and not atomic absorption. And it
has taken almost a century until Alan Walsh
began to wonder why molecular spectra were
usually obtained in absorption and atomic spectra
w x in emission 4 . And he came to the conclusion
that there was no good reason for neglecting
atomic absorption spectra; on the contrary, they
appeared to offer many vital advantages over
atomic emission spectra, and so he started to
carry out his first experiments. Even though Alkew
x made and Milatz 5,6 published two papers about
AAS in the same year when Walsh’s first paper
w x 7 appeared, it was Alan Walsh who had the
vision of a bright future for this technique, and
who was supporting and ‘preaching’ AAS with the
dedication of a missionary, until it finally found
acceptance in the mid-1960s.
However, even with the support of a visionary
scientist such as Alan Walsh, AAS is not necessarily
exciting at first glance. When Alan, back in
1952, called his colleague John Willis, and showed
him his first experiments with the words ‘look,
that’s atomic absorption...’ the disappointing
w x response was only ‘so what?’ 4 . This lack of
interest continued when an atomic absorption
spectrometer was publicly demonstrated for the
first time in Melbourne in 1954. The only person
who got excited about AAS at first glance in these
early years was Boris L’vov, who decided to check
the validity of the author’s ideas immediately
after he had stumbled over Walsh’s first publicaw
x tion in 1956 8 , which qualifies him as another
visionary scientist who could recognize the importance
of a discovery that was disregarded by
almost everyone else. And L’vov should become
another dedicated missionary of AAS, after he
had succeeded to slowly escape the ‘splendid
isolation’ of the socialist system of his country.

But similar to the experience that Alan Walsh
had to make, the excitement of Boris L’vov was
not shared by his colleagues, and it is obvious that
Alan and Boris were the exception, whereas this
author was more the rule, if I may insert some of
my personal history. I remember very well my first
contact with AAS. It was when I applied for a
position as an application specialist for infrared
spectroscopy at Bodenseewerk Perkin Elmer in
1966, and they offered me a position as application
specialist for AAS instead. ‘Atomic what...?’
was my first question, and when I was digging into
Alan Walsh’s paper soon after, there was a complete
lack of excitement on my part. You aspirate
a solution into a flame and you get a result
immediately } something was missing; there is
no spectrum that needs interpretation as in
molecular spectroscopy. This technique was too
simple for me to be exciting, and I really don’t
know why I nevertheless accepted the position as
an application specialist for a technique that I did
not know anything about at that time. However,
the excitement came soon, I only had to meet
with the two missionaries of AAS, Alan Walsh
and Boris L’vov, the latter at the First Atomic
Absorption Spectrometry Symposium in Prague,
1967. It was at that conference when the spark of
excitement ignited a flame that would continue to
burn throughout my career, and that is reflected
in more than 250 publications that carry my name,
including a book on AAS that just appeared in its
w x third edition 9 .
It may well be the apparent simplicity of AAS
that made me wonder back in 1966 if there is
. anything of interest in this technique that caused
renowned scientists again and again to forecast
the end of AAS for the very near future. There
must have been a good reason for Alan Walsh to
write his article Atomic Absorption Spectroscopy
w x } Stagnant or Pregnant 10 back in 1974, i.e.
there must have been rumors in the air about a
forthcoming decay of AAS. But obviously, by that
time AAS had already given birth to a new kid,
 . the graphite furnace GF which, however, was
not yet well understood, and it required the input
w x of L’vov 11 , and the introduction of the Stabilized
 . Temperature Platform Furnace STPF concept
w x by Slavin et al. 12 before it could become the
driving force for AAS in the 1980s.

But at a time when AAS was in use in every
laboratory, and research groups all around the
world were unveiling atomization and interference
mechanisms of GF AAS, and the increasing
knowledge made this technique more and more
rugged, there was again a renowned scientist prew
x dicting the forthcoming end of AAS. Hieftje 13
applied a third order polynomial to the annual
number of publications on AAS, and predicted
from the extrapolation of that function that ‘with
the current rapid growth in ICP atomic emission
and ICP mass spectrometry and with new incursions
being made by methods based on glow-discharge
lamps, AAS is heading for difficult times.
If current trends continue, I would not be sur
prised to see the removal of commercial AAS
instruments from the marketplace by the year
2000.’ By that time, however, AAS was already
pregnant again, and the name of this new off-
 . spring was flow injection FI , a technique the
input of which can be considered revolutionary in
w x almost all aspects of AAS analysis 14 . It has
been shown repeatedly that FI is far more than
an elegant sample handling and sample introduction
technique, and the potential of FI has not at
all been exploited completely, as will be discussed
later in this paper.
Obviously only a few milestones in the development
have been discussed in this review of the
last 45 years of AAS } if 1954, when the first
atomic absorption spectrometer was exhibited in
w x Melbourne 10 , is taken as the starting point.
Zeeman-effect background correction, simultaneous
multi-element AAS, and the introduction of
solid-state detectors should at least be mentioned
as other important contributions. But now, at the
end of this millennium, isn’t it time to look for
new goals, new challenges, and send AAS to
retirement in the same way as this author was
sent to retirement, because AAS is an ‘established
technique’ that does not justify any more
research. Obviously this author has a different
opinion and is expecting a whole series of exciting
new developments, both in the field of application
and in instrumentation. AAS is clearly pregnant
again.

2. New fields of application


When new fields of application are discussed, it
is necessary also to consider alternate techniques
that might be capable of solving the same problem
equally well or even better. It must also be
kept in mind that there is no single analytical
technique that can solve all problems, and that
each technique performs best only in its optimum
working range, and that the results may deteriorate
quickly when it is used outside this range.
The practical analyst looks for a solution to his
analytical problems and, if he can choose, he
selects the analytical technique that offers the
best solution with respect to simplicity, time and
cost. Obviously, analytical aspects such as accuracy,
and frequently also detection power, play an
equally important role.
Inductively coupled plasma mass spectrometry
 . ICP-MS is undoubtedly the technique that is
most en ¨ogue today for trace element analysis,
and there is no doubt about its detection power,
 . its multi-element -mass capability, and its speed
of analysis. The possibility of doing isotope analysis,
in addition, offers a unique field of application
that is proprietary to MS. On the other hand,
ICP-MS is undoubtedly one of the relatively
expensive techniques, both in purchase price and
in running cost. ICP-MS also requires considerable
operator’s skill, and is certainly not free of
interferences, particularly in the presence of complex
and concentrated matrices. Because of its
popularity and competitiveness, ICP-MS will be
the prominent technique for comparison with the
new fields that AAS may enter soon.

2.1. FI on-line preconcentration and separation for
GF AAS
It has been mentioned earlier that the potential
of FI has not been fully exploited for its
application in AAS. One of these fields is
undoubtedly FI on-line preconcentration and separation
for GF AAS, using sorbent extraction in
packed microcolumns andror precipitation and
 . collection in knotted reactors KR . Although the
first papers using these techniques already apw
x peared in the early 1990s 15,16 , this idea was
never really supported by instrument manufacturers.
This is surprising because the combination of
FI and GF AAS offers ideal conditions for fully
automatic ultra-trace analysis in the low ngrl
range in complex matrices under normal routine
laboratory conditions. Detection limits of GF AAS
w x in samples such as seawater 17,18 or high-purity
w x reagents 19 are lowered by 2]3 orders of magnitude
with a relatively minor instrumental requirement,
only an insignificant reduction in sample
throughput, compared to direct GF AAS analysis,
and no need for clean-room facilities.
An example of the capability of this technique
Table 1
Determination of four elements in doubly deionized water
DDW and in NASS 2 open ocean sea water standard
reference material National Research Council Canada using
flow-injection on-line sorbent extraction separation and prea
concentration for GF AAS


is given in Table 1, which shows the determination
of four trace elements in a sea water reference
material by GF AAS after on-line sorbentextraction
preconcentration and separation. The
results demonstrate first of all that the technique
is capable of obtaining accurate results in the
ngrl range in a complex and concentrated matrix
such as sea water. Secondly it is shown that all
commonly occurring elements can be determined
 . in doubly deionized water DDW , making an
on-line purification of reagents a necessity.
Thirdly, by comparing the standard deviation obtained
in the sea water sample and in DDW it
becomes apparent that it is no longer the matrix
that determines the precision, but obviously the
 lamp flicker noise nickel is well known to be a
. noisy lamp . This means the matrix has been
separated completely and has no more influence
on the determination.
FI on-line preconcentration and separation
clearly brings GF AAS detection limits close to
those of ICP-MS, and it even surpasses the capabilities
of the latter technique when the matrices
are considered in which the detection limits are
obtained. Obviously, FI on-line preconcentration
and separation could also be coupled to ICP-MS,
but not with the same ease, considering the elu-
 . ents that are typically used organic solvents and
their volume which is usually -0.1 ml. Last, but
not least, the running cost for the GF AAS
approach is probably an order of magnitude lower
than that for the comparable ICP-MS system.

2.2. Solid sampling for GF AAS

Although Alan Walsh in his first approach proposed
‘that the sample is dissolved and then
w x vaporized in a Lundegardh flame’ 5 , which for
good reasons became the preferred technique in
the 1960s and 1970s, solid sampling is as old as
w x AAS 20 . Boris L’vov used a graphite furnace and
a few crystals of sodium chloride in his first
experiment back in 1957 to demonstrate the principle
of AAS, and several other groups developed
a variety of furnaces for solid sample analysis in
the following years. Fig. 1 shows an early application
of solid sample analysis that was carried out
in our laboratory using a prototype of what was
later called the HGA-70, and that was presented
at the International Atomic Absorption Spectroscopy
Conference in Sheffield, 1969. Although the
conditions were very primitive at that time, with
no reliable tools for solid sample introduction
into the furnace and no means for background
correction, the results were surprisingly good.
The direct analysis of solid samples, using all
kinds of furnaces and devices for sample introduction
was continued throughout the past 30
years, as reviewed in a recent book, edited by
w x Kurfurst 21 , but it was typically supported by ¨
isolated research groups only. This is surprising
because GF AAS is an ideal technique for direct
solid sample analysis, because it is very flexible
with respect to the sample size, which can range
from approximately 0.01 mg up to almost 100 mg
w x 22 , and also with respect to the form in which
the sample is presented. This is among other
things due to the way the sample is introduced,
e.g. on a platform, the thermal pretreatment in
the graphite furnace, and the long residence time
of the atoms in the atomizer. The introduction of
a commercial system for automatic slurry samw
x pling 23 , based on the extensive work of Millerw
x Ihli 24 , could have been the turning point, but
the acceptance was not as expected.
More recently, however, there appears to be an
increasing market requirement for solid sample
analysis, coming predominantly from the producers
of modern high-tech materials, such as hard
metals, superalloys and ceramic superconductors,
etc. All these materials have two things in common:
their quality depends extremely on their
purity with respect to a number of critical trace
elements, and they are very difficult to bring into
solution. It was particularly the group of Viliam
Krivan in Germany who did excellent pioneering
work in that field, demonstrating that detection
limits of solid sampling GF AAS are superior to
all other techniques available for this kind of
analysis, including ICP-MS, simply because the
risk of contamination associated with any acid
w x digestion technique, is avoided 22,25]27 . A typical
example is shown in Table 2 for the detection
limits achievable for a number of elements in the
analysis of powdered tungsten trioxide and tungsten
blue oxide, using a variety of analytical techw
x niques 27 .
Even more important for the practical analysis
is that the efforts of Krivan and his group have
resulted in the recent introduction of commercial
equipment that automates solid sampling GF AAS
and makes it accessible for routine application
 . Fig. 2 . It has been shown that modern furnace
technology using platform atomization in a transversely
heated graphite tube, and integrated
absorbance for signal evaluation, has increased
accuracy dramatically, making solid sample analysis
possible with calibration against simple aquew
x ous standards 22,25]27 .

2.3. Speciation analysis

In the 1960s, determination of summation
parameters such as total heavy metals in water by
sulfide precipitation and gravimetry, was replaced
by the determination of the individual elements,
as it was recognized that the various heavy metals
differ significantly in their toxicity. One of the
driving forces at that time was the introduction of
AAS which made these determinations easy. In
the meantime it is well established that the total
content of an element does not give sufficient
information, as several elements may be essential
for humans or animals, and may be toxic at the
same time, depending not only on their concentration,
but even more importantly on their oxidation
state or the chemical form in which they are
present. A typical example is chromium, which is
essential in its trivalent form, but carcinogenic in
its hexavalent compounds. Another example is
arsenic, the inorganic forms of which are well
known to be highly toxic, whereas organic com-
pounds, such as arsenobetain, which are found in
high concentration in some seafood, are non-toxic,
and may even be essential.
A limiting factor for the introduction of speciation
analysis into routine use is undoubtedly that
most of the research work in that field is currently
done using high-pressure liquid chromatography
 . HPLC as a separation technique, and ICP-MS
for detection. This combination, although very
powerful, is far too expensive for routine use,
considering that a fast multi-element system is
typically waiting 10]20 min for a few species of a
single element arriving at the detector. There is
no doubt that this kind of research has to be
carried out in order to gain all the knowledge
necessary for speciation analysis. However, for
routine purposes, a good percentage of speciation
analysis can be done without prior separation of
w x the individual compounds by chromatography 28 .
A simple procedure has, for example, been proposed
to distinguish between ‘toxic’, i.e. inorganic,
mono- and di-methylated arsenic, and non-toxic
arsenobetaine using hydride generation HG
w x AAS 29 . FI on-line column preconcentration
and separation is typically selective for one oxidation
state of an element only, and may hence be
used for the separation of redox species, as was
demonstrated for the differential determination
 . w x of Cr VI and total chromium in water 30 .
Another example of speciation analysis by on-line
separation and preconcentration is shown in Fig.
3 for an activated alumina microcolumn, which at
 . pH 2 selectively preconcentrates Cr VI , whereas
 . w x Cr III is retained selectively at pH 7 31 . There
is no question that this field of routine speciation
analysis is far from being completely explored.
In addition, AAS with a quartz tube atomizer is
an extremely attractive detector for gas chro-
 . matography gc , as already proposed back in
w x 1976 by Van Loon and Radziuk 32 . This door
was, however, closed by instrument manufacturers
with the introduction of digital electronics in
the 1980s that no longer permitted the recording
of a series of peaks over an extended period of
time, i.e. the duration of a chromatogram. It
should not be difficult to open this door again,
and AAS could then become a sensitive, highly
specific, and not too expensive detector for gc
w x 33 , and after post-column derivatisation, even
for HPLC 34]36 . This aspect of AAS as an
attractive detector for chromatography in speciation
analysis will be brought up one more time in
 connection with diode laser AAS see Section
. 3.2 .