Printer Friendly

Multi-element atomic absorption with on-line cation exchange.

In the 1980s, the areas of innovation in atomic absorption spectroscopy (AAS) were centered around enhanced performance and ease of use. Enormous strides were made in taking graphite furnace AAS (GFAAS) from a research tool to a routine laboratory analyzer. In addition, computer and software advances have made AAS a very versatile and easy-to-use technique. In the 1990s, indications are that AAS developments will be in increased productivity. Its advantages have made this technique more popular despite newer techniques, such as inductively coupled plasma optical emission spectrometry (ICP-OES) and ICP-mass spectrometry (ICP-MS) whose capabilities overlap with those of AAS. Some of these advantages are the simplicity of the instrumentation, the specificity of analysis and relative freedom from interferences, low cost of instrumentation, and the ruggedness of the technique. These are offset by the single element nature of the technique and its relatively poor linearity. Recent developments have already begun to eliminate these last two limitations of AAS.

Multi-element AA for Flame and Furnace

Much of the pioneering work in multi-element AAS has come from the laboratories of Harnley and O'Haver.(1-3) Their continuum source wavelength modulation AA has been shown to successfully measure many elements at the same time, using both flame and furnace atomization techniques. They have shown that few compromises need to be made in selecting conditions for multi-element analysis. The major limitation of this approach is that it is light limited below roughly 220nm where several very good analytical wavelengths exist. It is probably because of this limitation that no commercial version of this instrument has ever been introduced.

At the 1990 Pittsburgh Conference on Analytical Chemistry and Applied Spectroseopy in New York City, Thermo Jarrell Ash introduced the Smith-Hieftje 8000 automated AAS, an instrument capable of measuring up to four elements at a time by either flame or furnace AAS. This innovative system employs an eight hollow cathode lamp elevator mechanism which is comprised of two four-lamp arrays, one above the other. The elevator is computer controlled to position one array in the analytical position at a given time. The beam from each lamp in an array is directed through the atomizer in rapid sequence by a galvanometer-driven mirror (see Figure 1). A second galvanometer drive controls the grating inside the monochromator and is synchronized with the galvanometer mirror. Galvanometer drive mechanisms have the advantages of being very fast (the entire wavelength range from 190 to 900nm can be traversed in less than 20 milliseconds), reproducible, and wear-free having no belts or gears as other drive mechanisms do.

Little compromise needs be made in developing a multi element flame method. The major determinant is whether the element requires an air - acetylene or nitrous oxide - acetylene flame; the elements are categorized in Table 1. Ionization suppressants, when required for elements such as sodium and potassium, typically do not interfere with other elements.

Multi-element methods development for furnace AAS is a bit more complicated. Here the determining factor is the volatility of the element, and these are categorized in Table 2. The degree of compromise, which must be made, has been minimized by the recent introduction of somewhat universal matrix modifiers such as palladium nitrate and mixed palladium/magnesium nitrates.(4,5) Figure 2 shows a video display screen from the external computer of the SH-8000 system. Shown is a multi-element absorption profile for As, Se, Pb, and Tl atomized from a platform using a mixed palladiuni/magnesium modifier. For this method, the only program compromise was that the maximum pyrolysis temperature had to be lowered from 1150 to 800'C.

With the SH-8000, up to eight elements can be determined in unattended operation, with only two samplings of each unknown. This provides a dramatic improvement in productivity over the typical one-at-a-time AA. Appropriate applications for this technology include several laboratory types: the environmental laboratory whose low level furnace analyses have become the bottle neck of the lab; plating bath and geochemical laboratories which need to measure four to eight elements at flame AA levels.

Additional Capabilities of Simultaneous AAS In addition to the obvious productivity advantages of measuring four elements at a time, simultaneous AAS provides some other rather interesting possibilities. Extended linear range analysis can easily be carried out by measuring the absorption of a less sensitive wavelength at the same time as the primary one. In this way, the linear rage of AAS can be increased from three orders of magnitude to six.

Simultaneous internal standard analysis can be employed to compensate for physical variations in samples as has been done for many years with ICP emission instruments. In fact, the SH-8000 can be used in the atomic emission mode to measure up to twelve elements in an unattended run each with an off-line background correction point and each with an internal standard compensation. Interfacing a Multi-element AA with an Ion Chromatograph

Developments in sample pretreatment in ICP-OES(6,7) have recently been extended to AAS.(8) These approaches involve the interface of an ion chromatograph to the spectrometer to both concentrate the analytes and remove the interferences.

The principles of ion chromatography are well understood. A buffered sample is loaded onto a column packed with an appropriate cation exchange chelating resin. The anions pass through the column unretained while many of the cations are retained. In step two, the column is washed with buffer to remove those cations, only weakly held by the resin. In step 3, the column is stripped with acid to elute the cation analytes into a concentrated solution which is free of interferences. In this work, the column used was a commercial column from the Dionex Corp. and is known as the Met Pac CC-1 column. It is packed with macroporous iminodiacetate resin, has a high loading capacity (0.45 millequivalents), and can withstand acid or base concentrations up to 6 Molar. It is packed with macroporous iminodiacetate chelating resin, has a column. Table 2 categorizes the commonly found ions with respect to retention by the Met Pac column. It is interesting to note that arsenic and selenium exist as anions in solution and are therefore not retained. The interface is depicted in Figure 3. Its salient points are the gradient pumps which are used to convert the elution reagent between buffer and acid, the sample loop which provides a precise and reproducible measure of sample to be loaded onto the column, and of course, the Met Pac column itself. The output from V3 goes to this spedtrometer. The chelation process consists of five steps which are carried out automatically:

1) sample is buffered as it is drawn into the interface;

2) the buffered sample is loaded onto the column; the transition metals are retained; the alkali and alkaline earth metals are partially retained; and the anions are not retained at all;

3) the column is washed with buffer to remove the alkali and alkaline earth elements;

4) the transition elements are eluted to the detector via nitric acid; during this process, the resin is converted to the hydronium ion form;

5) ammonium acetate buffer then converts the resin back to the ammonium form and the system is now ready for the next sample.

The output of the IC was set to match the uptake rate of the flame atomizer nebulizer (4.0 ml/min). Figure 4 is a reproduction of the CRT of the external PC which drives the AAS system, and shows the absorbance versus time profiles for four elements measured simultaneously in a spiked seawater sample, using an air/acetylene flame as the atomizer. The absorbance measurement was timed to coincide with the elution of the analytes from the column. Unlike ion chromatography, which achieves specificity by separating the analytes in time prior to reaching the detector, in this approach, since the detector is analyte specific, all analytes can be eluted at the same time. Figure 5 demonstrates the reproducibility of the system using iron as the example. The peaks result from a spike of 0.25 ppm Fe on the clean NASS-2 seawater reference material. Figure 6 shows the same analysis performed direct without preconcentration. It is clear that the spike is no longer discernible above the seawater. The detection limits in seawater and the enhancement factors obtained using the IC/AA interface are given in Table 4 for the four elements. Fe, Pb, Cd, and Cu. On average, roughly a six-fold enhancement factor was found. Since the chelation/ extraction process adds several minutes to the analytical cycle, the fact that four elements can be measured simultaneously significantly enhances the productivity of this technique.

Because levels of trace elements in seawater tend to be well below the detection limits determined by flame IC/AA, the interface was configured with the automatic sampling device of the furnace atomizer accessory. The difficulties of measuring very low levels of metals in seawater by conventional GFAAS are well known. Figure 7 shows the results of such an attempt using the typical ammonium nitrate matrix modifier. The scan overlays show the straight seawater with overspikes of 1.0 and 10 ppb of Cd (although background absorbance levels were quite high, only the Smith-Hieftje corrected signals are shown). The furnace used was the CTF 188 (Thermo Jarrell Ash Corp.) along with the aerosol deposition module (ADM) for sample introduction. A schematic of the ADM is shown in Figure 8. The device utilizes a nebulizer/premix system to convert the sample into a fine mist similar to a flame AA. Sample is then transported to the graphite cuvette, following an appropriate rinse step to guarantee the absence of carryover. The deposit time determines the volume of sample introduced to the graphite cuvette.

Since levels of Cd as low as 10 to 20 parts per trillion are commonly found in seawater, some type of matrix separation appears to be needed. Because the ADM is a nebulizer based device, the interface to the IC was greatly simplified. The only modification required was to alter the angle of the ADM so that the nebulizer was pointing at a downward angle rather than upward. This was necessary since the eluent of the IC was continuous and would otherwise backup inside the nebulizer. For the furnace study only Pb and Cd were measured simultaneously. A time study was performed to determine the appropriate deposition time and the results are shown in Figure 9. After 30 seconds of deposit, no further increase was observed. A second study was performed to determine the optimum nebulizer uptake rate. Since sample volume was limited, the most efficient uptake rate was desired. As shown in Figure 10, the most efficient uptake rate was 1.6 ml/min. The flow rate of the IC was set to the same value. Figure 11 shows the scan overlays for Cd in seawater using the IC/GFAAS approach. Several interesting observations can be made from this figure: there is significant Cd contamination in the blank; the level of Cd in the seawater (0.029 ppb) and the 0.1 ppb spike are clearly measurable; and there is a change in peak shape due to the seawater, indicating the presence of some interferences. Figure 12 shows the reproducibility of three successive measurements of Cd in the NASS-2 seawater. Unfortunately, the contamination for Pb was even more severe preventing any accurate determination in the sub ppb range.

Methods of reducing contamination are being studied. One possible approach is to place secondary chelation columns in series in the reagent lines prior to the sample juncture. An alternate approach of placing a layer of chelation resin in the buffer reservoir is also being evaluated.


The commercial availability of an automated multi-element flame/furnace AA not only provides the analyst with a more powerful and efficient analytical option, but also alters the decision-making process involved in the appropriation of analytical instrumentation. Traditionally, as the number of routine analytes increases the advantages of ICP-OES for trace analysis and ICP-MS for ultra-trace analysis become more pronounced. As the popularity of multi element AAS continues to grow, the number of analytes necessary to justify the more expensive and complicated alternatives will increase.


1. J.M. Harnley, and J.S. Kane, Anal. Chem., 56, pp. 48-54 (1984).

2. J.M. Harnley, N.J. Miller-Ihli, and T.C. 0-Haver, Spectrochem. Acta, part B, 39B, 305 (1984).

3. S.A. Lewis, T.C. O'Haver, and J.M. Harnley, Anal. Chem., 56, 1066 (1984).

4. L.M. Voth-Beach, and D.E Schrader, J. Anal. At. Spectrose., 2, 45 (1987).

5. G. Schlemmer, and B. Welz, Spectrochim. Acta, Part B, (1986).

6. J. Riviello, and R. Manabe, Paper #353, 1990 Pittsburgh Conference in New York City.

7. R. Manabe, and J. Riviello, Paper #844, 1991 Pittsburgh Conference in Chicago, IL.

8. G.R. Dulude, and J.J. Sotera, Paper #766, 1991 Pittsburgh Conference in Chicago, IL.


I would like to acknowledge the helpful support of Dr. R. Manabe of Thermo Jarrell Ash Corp. and Dr. John Riviello of Dionex Corp.
COPYRIGHT 1991 Chemical Institute of Canada
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Dulude, Gerard R.
Publication:Canadian Chemical News
Date:Jun 1, 1991
Previous Article:75 years of chemistry at the NRC: one discipline marching to three drummers.
Next Article:E.W.R. Steacie and Science in Canada.

Related Articles
Atomic absorption spectroscopy: simple and effective.
Atomic Absorption's Surprising Market.
Surprisingly valuable: the hollow cathode lamp market.
Uncertainty in quantitative electron probe microanalysis.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters