主题：Atomic absorption spectrometry pregnant again after 45 years

3. Instrumental developments


As in the previous chapter, this author does not
intend to consult a crystal ball to see the distant
future, but simply wants to analyze the ongoing
research for its potential to become attractive for
routine application within the next few years.
Obviously, as the instrumental developments, even
more than the applications, depend on the decision
of instrument manufacturers to invest in a
particular field, it is not easy to predict if those
developments that appear most promising to this
author, will really become commercial products.

3.1. FI gradient ratio calibration
Flow injection has been discussed several times
already in this paper as a revolutionary technique
for sample handling, sample introduction, and for
on-line preconcentration and separation for AAS.
However, there are many more application possibilities
for this technique, particularly because of
the very high repeatability of the transport
properties and the dispersion of the sample bolus
in the FI system. It has been demonstrated that
this high repeatability of all events over time can
be used successfully to correct for a number of
limitations of conventional AAS, including the
limited linear working range, and the influence of
w x concomitants on the analyte signal in F AAS 37 .
With this ‘gradient ratio calibration’ the entire
transient signal produced by the sample bolus is
stored in a computer and used for calibration.
The ratio of the absorbance of the calibration
solution and the test sample solution is formed at
the reading frequency of the spectrometer. If no
interference is present, and the measurement
solutions are all in the linear range, the ratio will
not change during the entire measurement. If the
maximum absorbance of a test sample solution is
outside the linear range, however, the absorbance
ratio changes as soon as the linear range is
exceeded. This is recognized and corrected by the
computer. The same situation occurs if an interference
is present which decreases with increas-
 . ing dilution dispersion , as shown in Fig. 4a for
the interference of phosphate on the determination
of calcium. If the computer program is used
to extrapolate the signal ratio over the entire
absorbance profile against absorbance As0, the
 interference can be eliminated by calculation Fig.
. 4b .
A major significance of this procedure is certainly
that not only can non-spectral interferences
in many cases largely be eliminated by calculation,
but more importantly that it can recognize
interferences. The possibility is thus given of issuing
a warning signal that permits the analyst to
thoroughly examine the trueness of the result
determined by the computer. This opens completely
new perspectives for quality assurance with
respect to trueness of analytical results. Techniques
like this one, which essentially only
require software development, and which belong
already in the area of expert systems, should
become more and more part of future instruments,
as computer capacity is no longer a limiting
factor.

3.2. Diode laser AAS

Tunable dye lasers have been proposed as radiation
sources for AAS already some 20 years ago
w x 38 . However, in spite of their extremely good
spectroscopic properties } they can be set to
virtually any atomic line between 213 nm and 900
nm with a bandwidth corresponding to the natural
width of an atomic line } they have not
found their way into common use because of a
number of practical and economic reasons: these
laser systems are expensive, frequently unreliable,
and difficult to operate. In contrast to dye lasers,
 . diode lasers DL appear to be more suitable to
 . one day replace hollow cathode lamps HCL and
 . electrodeless discharge lamps EDL . Semicon-
ductor laser diodes are nowadays mass produced
for compact disc players, laser printers, optical
data storage systems, and telecommunication
equipment, and hence they are cheap, reliable,
easy to operate and they have long lifetimes. A
number of these diodes have excellent spectroscopic
properties which make them attractive
w x sources for spectrochemical analysis 39 .
Firstly, the power of presently available
commercial DLs is between one and several orders
of magnitude higher than what is provided
by the best HCLs. In addition, DLs show exceptional
stability, both in terms of wavelength and
intensity. These two factors together make it possible
to measure extremely low absorbances if
optimal experimental conditions are realized and
the fundamental shot-noise limit is achieved. But
even in conventional ‘routine’ atomizers, such as
flames and furnaces, detection limits were
achieved that were 1]2 orders of magnitude lower
w x than those obtained with HCLs 39,40 .
Secondly, the typical linewidth of a commercial
DL is approximately two orders of magnitude less
than the width of absorption lines in flames and
furnaces. This makes possible the expansion of
the dynamic range of calibration to high analyte
concentrations by measuring the absorption on
the wings of the absorption line, where optically
w x thin conditions prevail 40 . In addition, a DL,
under normal operating conditions, emits one single
narrow line, which dramatically simplifies the
spectral isolation of the absorption signal. One
does not need the monochromator which is necessary
with HCLs for isolation of the analytical
line from the other spectral lines emitted by the
HCL, and the photomultiplier could be replaced
w x by a simple semiconductor photodiode 39,41 .
Thirdly, the wavelength of DLs can be easily
modulated at frequencies up to GHz by modulation
of the diode current. Wavelength modulation
of the DL with detection of the absorption at the
second harmonic of the modulation frequency, 2f,
 . greatly reduces low-frequency flicker noise in
the baseline, providing improved detection limits
w x 39,40 . In addition, wavelength modulation of the
radiation source provides an ideal correction of
non-specific absorption and significantly improves
the selectivity of the AAS technique.
The major limitation of DL AAS at this point
in time is that, although a commercial blue laser
diode was introduced earlier this year, the lower
wavelength limit for mass-produced DLs is approximately
630 nm, which means that even with
frequency doubling in non-linear crystals, the important
wavelength range for AAS of 190]315 nm
cannot be attained yet with this technique. However,
this need not necessarily be a major limitation
for the successful introduction of DL AAS
into routine application, as this technique, in the
opinion of this author, is ideally suited for dedicated
instruments for special purposes and as
low-cost detectors for gc or HPLC. One example
for such an application is the HPLC]DL AAS
 .  . system for speciation analysis of Cr III rCr VI
w x proposed by Groll et al. 42 ; another example is
the tungsten coil atomizer DL AAS system for
the determination of aluminum and chromium
w x described by Krivan et al. 41 , which is shown in
Fig. 5. Another very interesting aspect is that with
DLs as radiation sources, the determination of
non-metals such as halogens, sulfur, or even noble
gases comes within reach of AAS. All these elements
have long-lived excited states from which
strong absorption transitions can be induced by
w x the red and near-IR radiation of LDs 39 . Zybin
w x et al. 43 , for example reported about a determination
of chlorine by gc-microwave induced
plasma-DL AAS with a detection limit some two
orders of magnitude lower than the best values
obtained by optical emission spectrometry.
In the opinion of this author, DL AAS should
at this point in time not be considered a competitor
of the conventional AAS with HCLs and
EDLs, but as a very attractive expansion of the
capabilities of AAS that opens entirely new fields
of application at relatively low cost. The firstgeneration
instruments and modules for DL AAS
w x that became commercially available recently 44
offer an excellent opportunity to exploit the potential
of this technique not only in research but
already in routine application.

3.3. Continuum source AAS

In his early work of 1952, Walsh essentially
ruled out the use of a continuum source CS for
AAS because the required resolution of approximately
2 pm was far beyond the capabilities of
the best spectrometer available in his laboratory
at the time. He concluded that ‘one of the main
difficulties is due to the fact that the relationships
between absorption and concentration depend on
the resolution of the spectrograph...’ if a regular
w x monochromator is used 10 . Hence, line sources
 . LS , such as HCLs and EDLs were used almost
exclusively for routine application of AAS over
the past 45 years, as their stable, narrow-line
emission guarantees high analyte specificity, even
with the use of low-resolution monochromators,
as well as good detection limits, and a working
range of 2]3 orders of magnitude. The fact, however,
that the use of LSs in essence makes AAS a
 one-element-at-a-time technique although the simultaneous
determination of more than one ele-
. ment has in the meantime been realized has
stimulated researchers again and again to investiw
x gate the feasibility of CS AAS 45 .
Only recently, however, have these efforts
shown convincing results with the availability of
w x high-resolution echelle spectrometers 46 , solid- ´
w x state array detectors 47,48 , and continuum
sources that have a sufficiently high emission
intensity within the small spectral interval under
consideration over the entire wavelength range of
AAS. Fig. 6 depicts such a system with a Xenon
short-arc lamp, a double-echelle monochromator ´
 . DEMON , and a CCD detector, providing a
spectral resolution of approximately 2 pmrpixel
w x 49 . The resolving power of this setup is impressively
demonstrated in Fig. 7, which shows the
integrated absorbance spectrum around the
cadmium resonance line at 228.802 nm.
Obviously, instruments for CS AAS can be
designed in several different ways, and their fea
tures depend on their particular design. The setup
shown in Fig. 6 is still of the monochromator
type, i.e. designed for the determination of one
element at a time. However, it provides a variety
of advantages over conventional LS AAS. Firstly,
the atomic absorption can not only be measured
at the center of the absorption line with maxi-
.  mum sensitivity but also at its wings with
. reduced sensitivity , thus greatly increasing the
dynamic range to approximately 5]6 orders of
magnitude in concentration or mass. This,
together with detection limits that according to
w x recent reports 45 tend to be even better than
those obtained with LS AAS, and with the expectation
of further improvement, eliminates one of
the classical disadvantages of AAS } its limited
dynamic range.

Secondly, the setup shown in Fig. 6 with a CCD
detector provides a lot of information about the
spectral neighborhood around the analytical line,
which eventually results in a much more reliable
and accurate background correction, as essentially
any pixel or set of pixels may be used for
that purpose, depending on the nature of the
background. In addition, as this wavelengthresolved
absorbance is measured over time with
each pixel, a three-dimensional absorbance pattern
is obtained eventually, as shown in Fig. 8 for
the background signal of 5 ml of urine in the
vicinity of the palladium line at 247.642 nm. This
kind of structured background makes a correction
with a deuterium lamp impossible, and presents a
problem even for Zeeman-effect background corw
x rection 50 . In the case of the CS AAS, however,
it was possible to eliminate the background
almost completely by a least-squares background
correction with the normalized spectra of PO and
w x Na 50 .
To extend the capabilities of CS AAS to its full
capability, that is simultaneous multi-element
AAS, requires the replacement of the one-dimensional
array detector by a two-dimensional multiarray
detector. This would obviously have a significant
impact on the cost and the complexity of
the whole instrument, if all the above-described
features are to be retained. A sacrifice in information
obtained by significantly reducing the
number of pixels per array could be a compromise
which, however has yet to be investigated.
However, as also pointed out by Harnly
w x 45 , this author believes that the advantages of
CS AAS over LS AAS, even for the determination
of one element at a time, with respect to
detection limit, dynamic range, spectral information,
and background correction capabilities, are
so convincing that it is about time for CS AAS to
replace conventional LS AAS step by step. The
simultaneous multi-element version will then fol-
low with the increasing acceptance of CS AAS,
and as prices for two-dimensional array detectors
drop.

4. Conclusion


Atomic absorption spectrometry is without
doubt pregnant, and will give birth to new offspring
very soon, provided the instrument manufacturers
are prepared to give their assistance.
Some of the application-oriented innovations
could be incremented at a comparably small effort;
others, such as solid sampling for GF AAS
are already available, and only wait for their more
general acceptance. Diode laser AAS is still a
small niche market, but there is hope that the
most rewarding fields of application for this technique,
i.e. dedicated instruments, detectors for
chromatography, and determination of nonmetals,
will be exploited soon and introduced into
the market properly. Obviously, the manufacturers
of laser diodes have to recognize the importance
of the analytical and sensor market before
further progress can be expected. Continuum
source AAS, finally, has the greatest potential in
the opinion of this author, and it only requires
the decision of a courageous instrument manufacturer
to re-design the image of AAS, making it
young, bright, and prosperous again.