Dec., 19581 PROCEEDINGS 655 Determination of Zinc and Other Elements in Plants by Atomic-absorption Spectroscopy BY D. J. DAVID (Division of Plant Industry, C.S.I.R.O., Canberra, A.C.T., Australia) Results of investigations into the application of atomic-absorption spectroscopy to the analysis of plant material for zinc, magnesium, copper and iron are given. For zinc and magnesium, the method is a t least as accurate and sensitive as other methods currently available, and is con- siderably better in both rapidity and freedom from interference by extraneous elements. For copper and iron, the method is insufficiently sensitive for general application in its present form. THE aim of the work described was to test the application of atomic-absorption spectroscopy, developed by Walshl and Russell, Shelton and Walsh,2 to the analysis of plant material for certain inorganic elements, particularly zinc.Most methods currently available for determining zinc in plant material involve pre- liminary chemical concentration with a solution of dithizone or another complexing reagent in chloroform or carbon tetrachloride. The dithizone extract can be subjected to arc-emission spectrographic analysis,s to further extraction to remove interfering elements and then photometric measurement of the zinc - dithizone complex4 or to evaporation, digestion and polarographic analy~is.~ Atomic-absorption spectroscopy has advantages over these methods in that the diluted plant digest is used directly and special precautions to avoid interference from other ions of plant origin are not necessary.Verdier, Steyn and Eve’s polarographic method6 and a method in which zinc is separated with Dowex 1 ion-exchange resin and then colorimetrically determined with Zincon7 both avoid the use of complexing reagents in organic solvents. When compared with the proposed method, however, they have the disadvantage that some chemical preparation of the solution is necessary before the zinc can be determined, and the polarographic method, at least, is inferior in sensitivity and accuracy. Calculations based on the data reported for the polarographic method6 suggest that the lower limit of determination of zinc in solution is about 1 p.p.m., but no estimate of accuracy is given at this level. At 31.5 p.p.m. of zinc in solution, a standard deviation of k1.26 p.p.m.is claimed. Six operations are necessary to prepare a solution for polarographic analysis by this method. DESCRIPTION OF APPARATUS A Lundegdrdh air - acetylene flame of the type described by Mitchel1,g into the base of which a fog of the sample solution is introduced, was placed on the optical axis between the slit of a flat-field Hilger medium-quartz spectrograph and a hollow-cathode discharge tube, which emitted intermittent light of the element to be determined at a frequency of 50 cycles per second. The plate holder of the spectrograph was replaced by a horizontal platform carrying a slit and photomultiplier assembly, which could be moved along guide grooves so that the slit remained in the focal plane of the spectrograph. During determina- tions, this exit slit was placed on the resonance line of the element to be determined and the transmitted light was picked up by an RCA 1P28 photomultiplier tube. The signal from the photomultiplier tube was fed to an a.c.amplifier tuned at 50 cycles per second, and the rectified output was measured with a millivoltmeter.656 DAVID : DETERMINATION OF ZIXC AND OTHER ELEMENTS [Vol. 83 Earlier apparatus described by Russell, Shelton and Walsh2 was similar in principle to that used in this investigation, but differed slightly in that a mechanical chopper was used to modulate the hollow-cathode beam and the signal from the ax. amplifier was rectified and fed into a pen recorder. The use of an intermittent hollow-cathode discharge and an a x . amplifier precludes all interference by flame-emitted light.As the apparatus described operates on the single-beam principle and the intensity of light emitted from the hollow-cathode discharge tubes is sensitive to slight fluctuations in mains voltage, an electronic a.c. voltage-stabiliser that delivered 240 & 1 volts was used to supply the hollow-cathode tubes. When the apparatus was used for analysis, the amplifier was adjusted to zero with the spectrograph-slit shutter closed and to full-scale deflection with the shutter open and a fog of pure water entering the base of the flame. The percentage reduction in reading when the water fog was replaced by a fog of sample solution was a measure of the absorption by the flame of the resonance line of the element to be determined.Adjustment to full-scale deflection was e-ffected by varying either the gain of the amplifier or the voltage applied to the photomultiplier stages. The air and acetylene pressures applied to the Lundegirdh flame assembly, which were kept constant by means of reducing valves during determinations, were 36 lb per sq. inch and 40 cm of water, respectively. Other equi-pment settings used are shown in Table I. TABLE: I INSTRUMENT SETTINGS FOR THE DETERMINATION OF ZINC, MAGNESIUM, COPPER AKD IRON Width of Width of Hollow-cathode Wavelength of Element entrance slit, exit slit, tube current, spectral line, mm mm mA A Zinc .. . . 0.10 0.2 10 2139 hfagnesium . . 0.15 0.!1 10 2852 Copper .. .. 0.20 0 4 20 3247 Iron .. .. 0.05 0.:2 50 3758 EXPERIMENTAL REPRODUCIBILITY OF THE METHOD- Atomic-absorption readings on thirty-nine portions of each of two zinc solutions in thirty-nine Lundegirdh spray bulbs gave results of 58.56 1.07 per cent.absorption at a zinc level of 10 p.p.m. and 7.26 +_ 0.94 per cent. absorption at a zinc level of 1 p.p.m. These variations, which are standard deviations of single determinations, may originate from variations in the dimensions of the spray Elulbs, from electrical variation, from variations in acetylene and air pressures or from inaccuracies when readings are made. A similar test for magnesium gave results of 10.37 & 0-33 per cent. absorption at a magnesium level of 0.5 p.p.m. and 43.15 i 1 4 1 per cent. absorption at a magnesium level of 5 p.p.m. The variation at the latter level includes a slight drift similar to that in the magnesium results shown in Table 11.TESTS FOR INTERFERENCES- A solution containing all the water-so1ubl.e major elements likely to be encountered in plant material was prepared by dissolving 10 g of potassium chloride, 2 g of sodium chloride, 4 g of calcium carbonate, 1 g of magnesium ox.ide, 3 g each of ammonium dihydrogen ortho- phosphate and ammonium sulphate and 0.4 g (of aluminium ammonium sulphate in sufficient hydrochloric acid to convert the calcium carbonate and magnesium oxide to chlorides. This solution was diluted to 1 litre with water, which gave a solution approximately equivalent to that obtained when 0.2 g of the ash from about 2 g of dry plant material (from an “average” pasture samples) is dissolved in a volume of 10 ml.Atomic-absorption measurements at 2139 A were made for zinc at six concentration levels in this solution and in water; the results were as follows- Amount of zinc present, p.p.m. . . . . . . 1 2 4 8 16 32 Absorption in water, % . . . . .. .. 10 18.5 34 69 86 100 Absorption in synthetic plant-ash solution, yo . . 11 21 35 60 88 100Dec., 1958; IN PLANTS BY ATOMIC-ABSORPTION SPECTROSCOPY 657 To study the interference of individual inorganic plant-elements on the absorptionat four levels of zinc, copper and iron, each of the ions Kf, Na+, Ca2+, hfg2+, AP+, SO,2- and PO:- was varied individually between zero and from two to ten times its concentration in the solution mentioned in the previous paragraph, the concentrations of the other ions remaining TABLE I1 EFFECT OF MAJOR ELEMENTS IN PLANT MATERIAL ON THE ATOMIC ABSORPTION All concentrations, except those for magnesium, must be divided by 100 for application to magnesium OF ZINC, IRON, COPPER AND MAGNESIUM Absorption of- Absorption of- Amount Possible of interfering element element present, % Sodium .. -!-:&8 10.78 r 0.0 Potassium . . ‘1 0.52 1.04 0.0 Calcium . . { 0.16 0.80 0.0 0.403 r 0.0 10,012 Amount Possible of interfering element element present, % 0.0 0.80 Calcium . . { 0.16 r 0.0 Magnesium { ::4: 0.403 0.0 0.115 Sulphur . . (,.023 0.012 0 p.p.m. of zinc, % 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4 p.p.m. ? f zinc, % 14.0 14.0 14.0 14.0 14.0 14.0 14.0 14.0 14.0 18.0 18.0 18.0 18.0 18.0 16.0 16.0 16.0 14.0 12.0 12.0 12.0 0 p.p.m.0 p.p.m. of of iron, copper, % % 0.2 0.4 0.0 0.0 0.0 0.4 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 Absorption of- A 80 p.p.m. 8 p.p.m. of of iron, copper, % % 4.0 5.8 4.0 6.0 4.0 5.4 4.2 5.8 4.0 3.8 4.0 5.8 3.6 5.6 4.0 5.8 4.2 5.6 2.0 6.0 1.4 6.2 1.6 6.0 1.8 6.0 1.8 6.0 1.8 6.2 1.8 6.0 1.4 5.8 1.4 4.6 3.6 4.6 2.8 4.4 3.6 4.2 6 p.p.m. of mag- nesium, 50.0 52.0 50.0 48.0 50.0 48.0 44.0 48.0 48.0 % - - - 46.0 44.0 44.0 42.0 44.0 42.0 44.0 42.0 44.0 1 p.p.m. of zinc, 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 6.0 6.0 6.0 6.0 4.0 4.0 4.0 4.0 4.0 4.0 % 20 p.p.m. 2 p.p.m. of of iron, copper, % % 0.8 1.6 0.8 1.6 0.6 1.8 0.8 1.8 1.0 1.8 1.0 1.8 1.4 1.8 1.2 1.8 1.2 1.8 0.6 1.8 0.4 2.0 0.4 1.8 0.6 1.6 0.6 1.6 0.4 1.6 0.4 2.0 0.4 2.0 0.6 1.4 1.2 1.2 1.2 1.2 0.8 1.2 Absorption of- 6 p.p.m. of mag- nesium, % 50.0 50.0 50.0 48.0 50.0 48.0 46.0 48.0 48.0 - - - 46.0 46.0 44.0 44.0 48.0 44.0 44.0 42.0 42.0 8p.p.m.160p.p.m. 16p.p.m. of of of zinc, iron, copper, % % Yo 30.0 7.0 10.6 30.0 7.0 10.6 30.0 7.0 10.6 30.0 7.0 10.4 30.0 7.0 10.4 30.0 7.0 10.6 30.0 7.4 10.6 30.0 6.4 10.4 30.0 6.6 10.0 40.0 3.2 12.0 36.0 3.4 12.0 40.0 3.2 11.6 44.0 3.8 12.2 46.0 4.0 13.7 48.0 3.6 14.0 44.0 4.0 12.0 43.0 4.0 12.0 45.0 3.6 11.7 24.0 6.0 8.2 24.0 6.0 8.2 24.0 6.6 8.6 -7 6 p.p.m. of mag- nesium, 60.0 50.0 50.0 50.0 50.0 50.0 46.0 46.0 46.0 % - - - 46.0 46.0 44.0 42.0 44.0 44.0 44.0 44.0 42.0 -7 6 p.p.m. of mag- nesium, 50.0 50.0 50.0 48.0 48.0 46.0 46.0 48.0 48.0 % - - - 46.0 44.0 44.0 44.0 44.0 42.0 44.0 42.0 42.0658 DAVID: DETERMIEATION OF ZINC AND OTHER ELEMENTS [Vol.83 unchanged. The details of zinc, copper and iron levels and the results of the experiment are shown in Table 11. A similar study of the interference of K+, Naf, Ca2+, A13+, SO,2- and PO,3- on magnesium absorption was made by measuring the magnesium absorption after hundredfold dilution of the solutions from the previous test. These results are also shown in Table 11, but it must be borne in mind that all solution concentrations, except those heading the magnesium columns, should be divided by 100 for application to the magnesium results. As a residual amount of sulphuric acid is present in a solution prepared by digesting plant material with a mixture of sulphuric, perchloric and nitric acids, the effect of high concentrations of sulphuric acid on the absorption of zinc, copper, iron and magnesium was investigated; the results are shown in Table 111.TABLE I11 EFFECT OF SULPHURIC ACID ON THE ATOMIC ABSORPTION OF COPPER AND MAGNESIUM Absorption Amount Absorption in presence of of element in absence of 2.5 per cent. v/v Element present, sulphuric acid, of sulphuric acid, p.p.m. % % r o 0.0 0.0 Zinc . . . . Iron . . . . Copper . . . . Magnesium . . 1 4 8 0 20 80 160 { ,i r o 4.0 9.0 18.0 0.0 0.6 3.0 6.2 0.0 1.0 4.0 8.2 0.0 2.0 9.0 16.0 0.0 0.6 2.6 5.6 0.0 0.8 4.0 7.4 0.0 72.0 96.0 98.0 ZINC, IRON, Absorption in presence of 10 per cent. v/v of sulphuric acid, % 0.0 2.0 8.0 14.0 0.0 0.4 2.0 4.2 0.0 0.8 3.0 5.6 0.0 60.0 94.0 97.6 ANALYSIS OF PLAST MATERIAL- In view of the fact that the only definite interference would be that of residual sulphuric acid from the digestion of plant material, the following procedure for preparation of the sample was investigated.Between 1 and 2 g of oven-dried plant material were digested in 8-inch x I-inch Quickfit test-tubes with 4 ml, accurately measured, of sulphuric acid - perchloric acid mixture (1 + 7 ) and about 15 ml of nitric acid. More nitric acid was added if the destruction of organic matter was incomplete after the mixture had been evaporated to a small volume. When the organic matter had been completely destroyed, the digest was heated strongly to the stage at which all the nitric and perchloric acids had been driven off and the 0.5 ml of sulphuric acid remained.The digest was cooled, and 9.7 ml of water were added to make the final volume to 10 ml. Ground-glass stoppers were placed in the test-tubes, which were then heated in a water bath for 30 minutes with intermittent shaking. After cooling and chilling to below room temperature with an ice -water mixture to prevent possible crystallisation after filtration, the solution was filtered through a Whatman No. 42 filter-paper, and 7-ml portions of the filtrate were transferred to Lundegkrdh spray bulbs for direct atomic-absorption analysis. Standards containing 0, 1, 2 , 4, 8, 16 and 32 p.p.m. of zinc in 5 per cent. v/v sulphuric acid were prepared, and 7 ml of each were placed in Lundegkrdh spray bulbs for absorption measurement and subsequent preparation of a calibration curve.The sample solutions were analysed first, and then the standards, after which the analysis of several of the earlier samples was repeated to ensure that no drift in sensitivity had occurred during the series of tests. A calibration curve, plotted from the measurements on the standards, was used to determine the concentrations of the sample solutions. A typical calibration curve for zinc is shown in Fig. 1.Dec., 19581 I N PLANTS BY ATOMIC-ABSORPTIOX SPECTROSCOPY 659 COMPARISOS WITH POLAROGRAPHIC AXALYSES- petiole samples from clover. met h0d.j Table IV shows polarographic and atomic-absorption results for zinc in stem, leaf and The polarographic analyses were carried out by Walkley's 4 1 A / L 2 E 15- L 0 " Absorption, % Fig.1. Atomic-absorption calibration curves for the analysis of plant material for zinc and magnesium. The lines used are Zn 2 1 3 9 ~ and Mg 2852.4: curve A, zinc; curve 13, magnesium TABLE IV Z I S C DETERMINATION BY ATONJC-ABSORPTION AND POLAROGRAPHIC METHODS Amount of zinc found in clover stem bv- atomic absorption, polarographp, p.p.m. p.p.m. 27.5 28.2 27.5 28.8 25.5 28.8 21.5 22.8 Amount of zinc found in clover leaf by- r - - u atomic absorption, polarography, p.p.m. p.p.m. 49.5 47.3 48.5 47.7 51.5 55.1 - - ;\mount of zinc found in clover petiole by- atomic absorption, polarography, p.p.m. p.p.m. 29.5 30.5 49.0 54.4 7 - - Mean of atomic-absorption results = 36.7 p.p,m, Alean of polarographic results = 38.2 p.p.m.TABLE V RECOVERY O F ZINC FROM MATERIAL OF PLANT ORIGIN BY ATOMIC ABSORPTION .Approximate weight of dry Sample sample, g Phalaris tops . . . . 0.8 Wheat heads . . . . 2.0 White clover leaf . . 0.8 White clover straw . . 1.8 Oat straw . . . . 1.3 Sheep faeces . . .. 0.9 Amount of zinc originally present, r g 127 44 82 90 30 123 Amount of zinc added, r g 118 118 118 118 118 118 Amount of zinc found, Recovery, Pg % 255 108 164 102 198 98 202 95 152 103 340 99660 DAVID: DETERMIXATION OF ZINC ASD OTHER ELEMENTS [Vol. 83 ~ C O V E R Y EXPERIMEKTS FOR ZINC- The results of a series of recovery tests for zinc on a variety of materials of plant origin are shown in Table V. These tests were carried out by coning and quartering samples of dry material, combining opposite quarters and digesting the two portions so obtained after an appropriate amount of zinc had been added to one of them.The digests were analysed by the proposed method. DISCUSSION OF IWSULTS Although the differences between zinc absorption in water and in synthetic plant-ash solutions are within those that could be expected from experimental error, it can be seen that the discrepancy is in the same direction at all zinc levels except the highest (see p. 656). This is probably due to slight contamination by zinc in the analytical-reagent grade salts used in preparation of the synthetic plant-ash solution. I t can be seen from Table I1 that the concentrations of iron and copper are approximately in the range that would be found in normal plant material that had been prepared for analysis by the procedure described.As the atomic-absorption measurements on these solutions were generally too low to be reliable, it is considered that both the proposed method and the apparatus, in their present forms, are unsatisfactory for determining copper and iron in plant material. The results shown in Table I1 for magnesium, although only applicable to one level of magnesium (6 p.p.m. in solution), suggest that no interferences of plant origin will occur during analysis. As plant digests prepared for zinc determination must be diluted one hundred times before magnesium is determined, 110 interference by residual sulphuric acid from the digestion reagents on magnesium absorption would be expected (see Table 111).Standards for plant analysis for magnesium can therefore be prepared by dissolving a mag- nesium salt in water only. A typical calibration curve for magnesium is shown in Fig. 1. Atomic-absorption readings for magnesium were found to be much more steady than those for zinc, owing, probably, to the effect of variation in supply voltage being less for magnesium emission than for zinc emission from the respective hollow-cathode tubes. Table I1 shows that a change in level of some of the absorption results occurs between the figures for calcium and magnesium interferences and also between those for sulphur and aluminium interferences. A shortage of Lundeggrdh spray bulbs made it necessary to analyse the solutions in batches of thirty-six, zinc, copper and iron being determined in one batch before analysis of the next batch.The change in level of results was caused by an alteration in the sensitivity of the apparatus when it was changed from analysis for one element to another and then re-set on the first element. The results for magnesium in Table I1 indicate that a discernible, but insignificant, drift occurred over the whole series of determinations. This is not surprising when it is considered that the series took more than 2 hours to complete. If the number of analyses in a batch of samples were kept to twenty or less, such a drift would not affect the results. As the results of polarographic and atomic-absorption analysis are in agreement, choice between the two methods must be based on rapidity and freedom from possible sources of contamination.Analysis by atomic-absorption spectroscopy is superior in both these respects. The recovery experiments indicate that the proposed procedure for digestion of plant material and subsequent dissolution of the zinc is satisfactory; they also give an additional check on accuracy. It can be seen from Fig. 1 that the curves for zinc and magnesium are approximately linear up to 18 and 8 p.p.m., respectively. Experience has so far shown that these upper limits of accurate analysis for zinc and magnesium are adequate for the analysis of plant material that has been prepared in the manner described. The lower limits of reasonably accurate analysis for zinc and magnesium in solution were found to be about 0.5 and 0.2 p.p.m., respectively . I thank Mr. C. H. Williams for the polarographic analyses quoted, Mr. A. Walsh for supplying the electronic equipment8 used and both for valuable discussion during the work.Dee., 19581 Ih’ PLANTS BY ATOMIC-ABSORPTION SPECTROSCOPY 661 1. 3. 4. 5 . 6. 7. 8. 9. > -. REFERENCES Walsh, .1., Spectrockim. Acta, 1966, 7, 108. Russell, B. J . , Shelton, J . P., and Walsh, A , , Ibid., 1987, 8, 317. Massey, H. F., Soil Sci., 1957, 83, 123. Cowling, H., and Miller, E. J . , I n d . EYg. Chem., Anal. Ed., 1941, 13, 148. \\‘alkley, A,, Australian J . E.zpt1. Biol. Med. Sci., 1942, 20, 139. Trerdier, E. T., Steyn, W. J . .L, and Eve, D. J . , J . Agric. Food Chem., 1957, 5, 354. Jackson, R. H., and Brown, J. G., Proc. Amer. SOC. Hort. Sci., 1986, 68, 1 . Mitchell, R. L., “The Spectrographic Analysis of Soils, Plants and Related Materials,” Tech. Comm. N o . 44, Commonwealth Bureau of Soil Science, 1948. Box, G. F. H., Russell, B. J., and TI7alsh, A , , to be published. Received June 5th, 1968