394 Afialyst, June, 1968, Vol. 93, +@. 394-397 The Determination of Metals in Wool by Atomic-absorption Spectroscopy* BY F. R. HARTLEY AND A. S. INGLIS (Division of Protekra Chemistry, CSIRO, Wool Research Laboratories, Parkville (Melbourne), Victoria 3052, Australia) Atomic absorption has been shown to provide a simple and precise method for the determination of chromium, copper, mercury, tin, zinc, calcium, strontium and barium in wool. The method of standard additions is used for calcium and a calibration-graph method for the other metals. Both methods have an accuracy and reproducibility of about f 3 per cent. The wool sample is hydrolysed in constant-boiling hydrochloric acid, and the calibration graphs are prepared by using wool hydrolysates containing standard amounts of metal to compensate for physical and chemical inter- ferences caused by hydrochloric acid, amino-acids and ions present in the wool.THE wide use of metal salts in the treatment of wool has led to a need for rapid and precise methods for the determination of metals in wool. Recently, aluminium in wool has been determined by atomic-absorption spectroscopy.l Because that work showed that the method offered advantages in speed, convenience and specificity over existing methods,2 the present study was undertaken with a view to analysing wool for chromium, copper, mercury, tin, zinc, calcium , strontium and barium by atomic-absorption spectroscopy. In our previous work, the wool sample was hydrolysed by heating it overnight with constant-boiling hydrochloric acid in a sealed tube at 110" C, and the hydrolysate aspirated directly into the flame of the atomic-absorption spectrophotometer.It was possible to com- pensate simultaneously for both physical and chemical interferences by using solutions of wool hydrolysates to prepare calibration graphs. This technique necessarily required the wool to be free from metal ions, and for all of the metals, except calcium, this condition was obtained by repeated extraction with an aqueous 0-5 M EDTA solution. With calcium, although extraction with 0.5 M EDTA solution initially removed a considerable proportion of calcium ions, a small constant amount remained after repeated extraction. The method of standard additions3 was, therefore, used for this metal. APPARATUS- The apparatus described previously1 was used.Because of the relatively large size of the sample (about 0-3g), advantage was rarely taken of the very high sensitivity possible with atomic-absorption spectroscopy and, for all of the metals, except tin and barium, a reduction in sensitivity was necessary. This was achieved either by using a less sensitive line than that normally used or by rotating the b ~ r n e r . ~ This technique was preferred to dilution because it is quicker, although for zinc and strontium a dilution technique was necessary because rotating the burner did not reduce the sensitivity sufficiently, and other resonance lines of suitable sensitivity were not available. The operating conditions, shown in Table I, are not, therefore, necessarily the most sensitive, but those found to give the best results for the samples under test.REAGENTS- All of the reagents used were of analytical-reagent grade. Stock solutions (about 0.1 M) were prepared by standard methods.6 The wool used for the calibration graphs was prepared by shaking scoured wool (30 g) gently for 3 days with 800 ml of EDTA (disodium salt) solution (0.5 M). After rinsing it thoroughly in de-ionised water, the extraction with EDTA solution was repeated once more. The absence of the metal under test was confirmed by wet ashing,2 freeze-drying to concentrate the sample and subsequent analysis. * The work reported here is part of that presented to the 6th Australian Spectroscopy Conference, Brisbane, August, 1967. 0 SAC and the authors. EXPERIMENTALHARTLEY AND INGLIS TABLE I INSTRUMENTAL CONDITIONS FOR THE DETERMINATION OF METALS IN WOOL 395 Lamp current, Element Fuel - oxidant mA Chromium Acetylene - airt 10 Mercury Coal gas - air 4 Tin Hydrogen - air 5 zinc Coal gas - air 8 Copper Coal gas - air 6 Calcium Acetylene - nitrous oxide (10 : 1)s 10 Strontium Acetylene - nitrous oxide (4 : 1)s 10 Barium Acetylene - nitrous oxide (2 : 1) f 10 Slit width, P 100 50 20 50 300 25 100 50 Wave- 4254.3 4289.7 2942.1 2536.5 2354-8 2138.6 4226.7 4607.3 5535.6 Burner length, 5 5 5 10 10 10 10 5 5 5 cm 1 Burner :otation, 40" 10" 0" 0" 10" 0" 0" 0" 0" 0" Approxi- mate sensitivity* 7.5 10.9 5.7 2.9 4.2 $ 0.0 15 0.055 0.20 3-8 46 * Sensitivity is the number of milligrams of element per litre of water necessary to give 1 per f An oxidising flame was used to reduce the background noise observed with the more 5 Ratio of height of red "feather" to height of inner blue cone.cent. absorbance. sensitive reducing flame. Water contained hydrochloric acid (0.01 N) to prevent hydrolysis of tin(I1) chloride. ANALYSIS OF SAMPLES- Calibration-graph method-Prepare the samples and standards by hydrolysis in constant- boiling hydrochloric acid, as described previous1y.l Method of standard additiorcs-Hydrolyse the sample, as before,l and withdraw, by pipette, three 1-ml samples from the hydrolysate. Dilute one of the samples to the volume necessary to give a transmittance of 80 per cent. To the others add different, but accurately known, amounts of calcium ion (about 0.6 and 1.2 p.p.m.) before making up the volume with constant-boiling hydrochloric acid.Aspirate the three samples consecutively into the atomic-absorption spectrophotometer, measure the absorbance values and repeat the readings twice more to eliminate the effect of fluctuations in the flame or instrument conditions. Determine the magnitude of the background absorbance by measuring the absorbance of the nearby 4201 A line that is not absorbed by calcium. The concentration of calcium in the hydrolysate can be calculated6 from the three corrected absorbance readings in two independent ways. RESULTS AND DISCUSSION SAMPLE SIZE- It was found that if the wool samples were too small sampling errors were introduced, probably because the uptake of metal ions was not completely uniform among the fibres. TABLE I1 REPRODUCIBILITIES AND CONCENTRATION RANGES OF THE ANALYSES FOR METAL IONS IN WOOL Metal Chromium Copper .. Mercury . . Tin . . Zinc . . Calcium . . Strontium Barium . . Reproducibility, per cent. .. & 3.4 .. f 2.3 .. & 3-0 .. f 3.5 . . f 2.4 .. f 3.5 .. f 3-0 .. f 3.0 Typical concentration range, * mg of metal per g of dry wool 0.7, to 29 0.4 to 8 6 to 120 0.56 to 11 0.2 to 4$ 0-3, to 71 0.3 to 61 0.5 to 10 Minimum possible concentration, t 0.03 0.09 0-4 0.002 0.06 0-04 0.5 mg of metal per g of dry wool 0.56 * This concentration range was obtained by using the instrumental conditions given in Table I. A higher concentration of metal in wool than the maximum given here can be accom- modated, by rotating the burner more, by using a less sensitive resonance line or by dilution.The values given were obtained by diluting 1 part to 100 parts (for zinc), to 50 parts (for calcium) and to 10 parts (for strontium). Obtained by using conditions for maximum sensitivity. $ The range depended on the dilution used.396 HARTLEY AND INGLIS: DETERMINATION OF METALS [ArtdySt, VOl. 93 Thus samples of about 0.1 g gave reproducibilities of about &8 per cent., whereas samples of about 0.3 g gave reproducibilities of about +3 per cent. (see Table 11). Larger samples did not give better reproducibility. PRESENCE OF SOLID MATTER- The method used to prepare the samples gave dark brown solutions containing finely suspended matter that arose largely from decomposition of the amino-acid, tryptophan. However, neither the average absorbance reading nor the meter fluctuation at this reading was altered after filtration through a Millipore filter (0.20 p).ACCURACY AND REPRODUCIBILITY- Wool was analysed independently for calcium, chromium, copper and zinc. The results, shown in Table 111, indicated that the agreement between atomic-absorption spectroscopy and the independent method was good. The reproducibilities of the analyses, determined by analysing the same sample of wool six times, were generally about +3 per cent. (Table 11). The concentration ranges of metal in wool for which the method is applicable are shown in Table 11. TABLE I11 COMPARISON OF ATOMIC-ABSORPTION AND ALTERNATIVE METHODS OF ANALYSIS Metal content, mg per g of dry wool Independent Atomic Metal Independent method of analysis method absorption Calcium .. Nitric acid extraction; EDTA titration 0.068 0.069 Copper . . Hydrochloric acid extraction; EDTA titration 8.0 8-1 Chromium . . Wet ashing; iron(I1) titration of chromium(V1) 9.8 10.0 Zinc . . . . Wet ashing; EDTA titration 5.5 5.5 EFFECTS OF DIFFERENT MEDIA- As in the deterrnination of aluminium in wool,l it was found that the absorbance values in water, constant-boiling hydrochloric acid and wool hydrolysate solutions were in each instance different. This was shown to be largely because of the different surface tensions and viscosities of the solutions, which emphasised the need for samples and standards to be prepared in identical media. This was achieved automatically with the method of standard additions, and with the calibration-graph method by preparing the standard solutions in wool hydrolysates. VALIDITY OF THE METHOD OF STANDARD ADDITIONS- The method of standard additions, as normally used, requires a linear relationship between absorbance and concentration.Although this relationship was not linear for calcium in water when a nitrous oxide - acetylene flame was used, it was linear in wool hydrolysates in the concentration range 0 to 8 p.p.m. of calcium. Willis4 has pointed out that the method of standard additions is based on the assumption that the interfering substance alters the absorbance of the added metal to the same extent as it does that of the original sample. This may not always be so, particularly when only a small amount of interfering substance is present, but this condition was established in the present work by making two separate additions to each sample, thus providing a check on the linearity of the analysis in the presence of the interfering material.Although errors caused by instrumental drift are minimised, by measuring in quick succession solutions that are effectively sample and standard, the method of standard additions involves an extrapola- tion, which makes it inherently less accurate than the calibration-graph method, which involves an interpolation. It is, therefore, essential to correct for the background absorbance, which is caused either by absorbance by the flame or scattering of the light by solid particles in the flame, by measuring the absorbance at a nearby wavelength that is known not to be absorbed by the metal ion under examination.Modulated atomic-absorption apparatus should be used to avoid measuring flame emission.June, 19681 IN WOOL BY ATOMIC-ABSORPTION SPECTROSCOPY 397 SCOPE OF THE METHOD- The results obtained confirm the conclusions drawn from the earlier work1 that metals in wool can be readily determined by atomic-absorption spectroscopy. The method should be equally suitable for other insoluble protein materials, such as hair and hide. The advantage of being able to use a hydrolysis technique to decompose the protein material, rather than the more lengthy ashing or extraction techniques will still apply. This procedure has the additional advantage, as far as the biochemist is concerned, that the hydrolysate could also be used for other determinations, such as those involved in an amino-acid analysis. The authors thank Mr. B. J. Wallace for assistance with the experimental work. REFERENCES 1 . 2. 3. Hartley, F. R., and Inglis, A. S., Analyst, 1967, 92, 622. “Methods of Test for Textiles,” British Standards Handbook, No. 11, 1963. Elwell, W. T., and Gidley, J. A. F., “ Atomic-Absorption Spectrophotometry,” Second (Revised) Edition, Pergamon Press, Oxford, London, Edinburgh, New York, Toronto, Sydney, Paris and Braunschweig, 1966, p. 74. Willis, J. B., Meth. Biochem. Analysis, 1963, 11, 1. Vogel, A. I., “A Textbook of Quantitative Inorganic Analysis including Elementary Instrumental Analysis,” Third Edition, Longmans, Green and Co. Ltd., London, 1961. David, D. J., Analyst. 1962, 87, 576. Received January 29th, 1968 4. 5. 6.