ANALYTICA CHIMICA ACTA

ELSEVIER
Analytica Chimica Acta 330 (1996) 25 l-257

Extraction as a method for preparation of vegetable samples for the determination of trace metals by atomic absorption spectrometry
Elzbieta Wieteska*, Anna Zi6ek, Anna Drzewihska
Military University of Technology, Institute of Chemistry, S. Kaliskiego 2, 01/489 Warsaw, Poland Received 24 October 1995; revised 9 April 1996; accepted 12 April 1996

An alternative, simple and rapid technique was developed for a quantitative isolation of the group of eight elements: Al, Ca, Cd, Cu, Fe, Mg, Pb and Zn from vegetable material samples in an open system with the use of acids: cont. HCl, dil. HN03 and hydrofluoric acids. Equivalence of both the extractants, HCl and HNOs, was found suitable for a quantitative isolation of Ca, Cu, Mg, Zn, Pb and Cd. For lead and cadmium, however, dil. HNOs proved to be more suitable because these elements are determined by the graphite furnace atomic absorption spectrometty (GF-AAS) technique. A quantitative dissolution of Al and Fe requires hydrofluoric acid as an additional extractant. The proposed method allows to obviate the organic matrix destruction stage, shorten the analyte dissolution time, reduce cost, and minimize hazards of loss and contamination. Validity and versatility of the method developed was verified by the analysis of standard reference materials.
Keywords: Atomic absorption spectrometry; Extraction; Plants; Trace elements

1. Introduction
Vegetable materials like pine needles, mosses, tree bark or leaves are often used as bioindicators for the determination of the degree of pollution of chemical elements in the environment. Analyses for the content of bioelements and toxic elements are also carried out to diagnose the reasons for abnormal plant development. In such studies conclusions are drawn on the basis of the results of a large number of samples. Hence, it is of significance that the analytical

procedures used should be rapid and simple, with no detriment to adequate accuracy and precision of determinations of the analytes. In the analytical procedures involving elimination of the matrix, the chemical pre-treatment of the sample is significant and frequently the most timeconsuming step. The issues related to the chemical pre-treatment of plant materials is the subject of this study. For the determination of metal elements in the materials mentioned the atomic absorption spectrometry (AAS) technique is widely used. The technique features a high accuracy and precision of the macro, micro and trace determinations of elements on the

* Corresponding

author. Fax: (48) 22 6669041.

0003-2670/96/$15.00 0 1996 Elsevier Science B.V. All rights reserved PII SOOO3-2670(96)00187-O

252

E. Wieteska et al./Analytica

Chimica Acta 330 (1996) 251-2.57

condition that the analyte sample is adequately prepared [l]. The AAS technique in its basic variants requires liquid samples. A major issue is the method selected for the preparation of the sample, by a process as simple as possible, which could secure the analyte to be quantitatively dissolved in the form suitable for a precise execution of instrumental analysis. For the preparation of samples of vegetal materials the dry ashing or wet digestion techniques are commonly used [2-51. The methods ensure the organic matrix to be eliminated and the analyte quantitatively dissolved, yet they are associated with some limitations and problems. The wet digestion procedure in an open system is laborious, expensive and associated with hazards of loss and contamination, as well as self-ignition of the sample. Another disadvantage of the procedure is the evolution of toxic gaseous products (nitrogen oxides) in considerable amounts. The dry ashing technique is discouraged in view of possible loss of volatile elements, the reactions of the analyte with the crucible material, difficulties involved in dissolving the combustion residue, and contamination hazard [6]. In view of the drawbacks involved in the classical techniques attempts are being made in finding some new procedures for the preparation of analytical samples. Recently the microwave technique is gaining popularity [7-lo]. Other proposals involve decomposition of the organic matrix in plants by laser, high pressure or ultrasound approaches [ll131. However, these techniques also have their weak points, the cost of the associated equipment being an added disadvantage. This paper reports the results of the attempts to develop a simple, rapid, inexpensive and possibly versatile technique of preparation of samples of vegetable material conveniently analysable by the AAS methods. The study was inspired by the reports on viability of a direct acid extraction of individual elements [ 11,121. The study was concerned with a quantitative, simultaneous isolation from vegetable material of the following group of the elements: Zn, Cu, Fe, Al, Ca, Mg, Pb, and Cd. The extraction in an open system was used with acids such as: cont. HCl and dil. HNOs. In the extracts obtained Pb and Cd were determined by the graphite furnace (GF)-AAS method; the remainder of the elements - by the flame AAS method.

The simplified method developed was compared for its efficiency with the classical pre-treatment procedures. The reliability of the procedure advanced was verified by the analyses of standard reference materials.

2. Experimental

2. I. Apparatus
All the atomic absorption measurements were carried out with a Per-kin-Elmer Model 2100 with a deuterium (D2) lamp for background correction. Cu, Fe, Zn were determined by the flame AAS in an acetylene-air flame; Al, Ca, Mg - in an acetylene nitrous oxide flame. The calcium absorbance was measured without background correction. The measurements were made under standard conditions for the most sensitive resonance lines of individual elements. For the determination of Pb and Cd a PerkinElmer Model HGA-700 atomization furnace was used with pyrolytic graphite tubes, Lvov platform and sampled with an AS-70 autosampler. The conditions for Pb and Cd determination are reported in Tables 1 and 2. 2.2. Reagents and reference materials Acids: HNOs (65%), HC104 (70%), HCl (36%), I-IF (40%) - Merck suprapur grade. Water: triply distilled in a quartz distiller, Standard solutions: Standard Zn, Cu, Fe, Al, Ca, Mg, Pb, Cd solutions of a 1 mg 1-l concentration were prepared from Riedel-de-Haen Fixanals. Working solutions were prepared in 1 M HNOs.

Table 1 Experimental

conditions

for cadmium

and lead determination Pb Cd 228.3 nm 0.7 nm 3mA 5 Pl 5 Pl

Wavelength Slit width Lamp current Sample aliquots Matrix modifier aliquots

283.3 nm 0.7 nm 7mA 5 Vl 5 Id

E. Wieteska et al./Analytica Chimica Acta 330 (1996) 251-257

253

(C)

1
2 3 4 5 6 7

90 110 400 850, 700 b 1800a, 1600b 2650a, 2500 b 20 programme programme

a Temperature b Temperature

Matrix modifier: NHaHzPOd, of a concentration of 4% by weight. Materials studied: pine needles collected from the woods in the Warsaw neighbourhood. Reference materials: certified NIST SRM 1575 pine needles and CTA-OTL-1 oriental tobacco leaves, and an alga II standard reference sample supplied by the Environmental Specimen Bank (ESB) in Jtilich, (D). 2.3. Sample preparation The pine needles, pulverized and homogenized, were air dried and then placed in a tight polyethylene vessel. Immediately before use the sample was dried to a constant weight at 85C. The reference materials were prepared in conformity with the instructions provided in the certificates. 2.4. Dissolution of analytes

2.4.2. Wet digestion Method 2: Weigh 0.5g sample into a quartz container, add 5cm3 of a HN03+HC104 mixture (4: l), heat gradually until the solution gets completely discoloured. Then evaporate to dryness. (A) Dissolve the residue in 1 M HNOs, filter into a 25 cm3 measuring flask, and make the volume up to the mark with 1 M HN03. (B) Treat the residue with 2 cm3 HF, heat until the acid is removed by evaporation. Dissolve the deposit in 1 M HNOs, transfer the solution into a 25 cm3 measuring flask, and make the volume up to the mark with 1 M HNO3. Polyethylene vessels were used throughout. 2.4.3. Dry ashing Method 3: Weigh 0.5 g sample into a quartz vessel, put it in a muffle furnace, and heat at 600C for 2h. (A) Dissolve the residue in 0.5 cm3 cont. HCl, add 1 M I-IN03, filter into a 25 cm3 measuring flask, and make the volume up to the mark with 1 M HN03. (B) Transfer the residue into a polyethylene evaporating dish, treat with 2cm3 HF and heat the mixture until complete evaporation of the acid. Dissolve the deposit in 1 M I-INOs, transfer the solution into a 25cm3 measuring flask, and make the volume up to the mark with 1 M HNOs. Blanks were run for all the methods used.

3. Results and discussion The extraction of Zn, Cu, Fe, Al, Ca, Mg, Pb and Cd from needle samples was performed in the open system with the use of various extractants, viz., cont. HCl and dil. HNOs. To evaluate the efficiency of the process, a comparison was made for the results of the analysis for the extracts and for the samples following a complete destruction of the organic matrix with the use of two mineralization techniques: the wet digestion and dry ashing procedures. The results are collected in Table 3. In view of a possible occurrence in pine needles of the compounds of the elements studied in combination with silicon, it was examined whether hydrofluoric acid was needed to be used so as to assure a complete dissolution of the analytes. The results of the respective solutions are reviewed in Table 4.

2.4. I. Acid extraction Method 1(A): Weigh 0.5 g sample into a quartz vessel and add 5 cm3 of cont. HCl or dil. (1 : 5) HN03, heat at 90C for 0.5 h. On cooling add 5 cm3 H20, filter into a 25 cm3 measuring flask, and make the volume up to the mark with water. Method l(B): Weigh 0.5 g sample into a PTFE vessel, add 2cm3 I-IF, heat at 90C until evaporated. Add 5 cm3 of cont. HCl or 5 cm3 (1: 5) dil. HNOs, and heat for 0.5 h. On cooling add 5 cm3 water, filter into a 25 cm3 polyethylene measuring flask and make the volume up to the mark with water.

254

E. Wieteska ef al./Analytica

Chimica Acta 330 (1996) 251-257

Table 3 Results of the analysis of needles in bearing various methods for digestion of analytes to solution without the use of HF (pg g- dry weight) Methods Elements Al Fe Mg Zn cu Pb Cd Caa Extraction HCI x&s (RSDB) 323k19 (5.9) 187f5 (2.9) 683f8 (1.2) 56f2 (3.6) 3.4zko.3 (8.8) 5.4hO.3 (5.6) 0.75f0.07 (9.4) 0.41f0.01 (2.4) HNOs xfs (RSD%) Wet asbing xfs (RSD%) Dry asbing xfs (RSD%)

270f20 (7.4) 78f3 (3.8) 688fll (1.5) 56f2 (3.6) 3.5f0.3 (8.5) 5.6f0.6 (10.7) 0.67f0.07 (10.4) 0.41f0.02 (2.4)

357*34 (9.5) 21115 (2.3) 690f7 (1.8) 56f2 (3.6) 3.6kO.2 (5.6) 5.5ztO.6 (10.9) 0.68f0.07 (10.3) 0.40f0.02 (5.0)

327f28 (8.6) 181+7 (3.6) 55f2 (3.6) 3.4f0.4 (11.7) 5.3f0.5 (9.4) 0.71f0.08 (11.2) -

Values in wt. 96. SAverage, n=lO; s=standard

deviation;

RSD=relative

standard deviation.

Table 4 Results of the analysis of needles in bearing various methods for digestion of analytes to solution with the use of HF (pgg- Methods Elements Al Fe Mg Zn cu Pb Cd Ca Extraction HCl xfs (RSD%) HNOs nfs (RSD%) Wet ashing xfs (RSDB)

dry weight)

Dry ashing xfs (RSDI)

514f23 (4.5) 208f7 (3.3) 692f12 (1.8) 56f2 (3.6) 3.4f0.3 (8.8) 5.8f0.6 (10.3) 0.76f0.08 (7.9) 0.40f0.02 (5.0)

509f18 (3.6) 205f6 (2.9) 678f15 (2.2) 57f3 (5.2) 3.4f0.3 (8.8) 5.7f0.5 (8.8) 0.68f0.07 (10.3) 0.40f0.02 (5.0)

515flO (1.9) 21 lf7 (3.4) 685flO (1.5) 56f2 (3.6) 3.6f0.2 (5.6) 5.8f0.7 (12.0) 0.72f0.09 (12.5) 0.41f0.02 (4.8)

508f13 (2.5) 215f4 (1.9) 55fl (2.2) 3.5f0.3 (8.5) 5.3f0.5 (9.4) 0.73f0.05 (6.8) -

values in wt. k. PAverage, n=lO; r=standard

deviation;

RSD=relative

standard deviation.

Consideration of the results shown in Table 3 shows that the extraction of analytes with the acids: cont. HCl and dil. HNOs (Method l(A)) ensures a quantitative dissolution process for six of the group of elements examined: Zn, Cu, Ca, Mg, Pb and Cd. For these elements our studies have demonstrated the equivalence of the extractants used. It should however, be emphasized that the Pb and Cd extraction with hydrochloric acid is not recommended because of interference of chlorides with the determination of these elements by the GF-AAS technique. Attempts to eliminate the effect of chlorides with the use of NI&HaPOd as the modifier brought about the desirable result in the Pb determination only. For Cd some positive results of determinations were obtained only upon a complete evaporation of the extract and dissolution of the residue in 1 M HNOs. For this reason dil. I-IN03 is more advisable

for the extraction of Pb and Cd from vegetable material samples. The use of HCl or dil. I-INOs (Method l(A)) is not sufficient for a complete isolation of Al and Fe from samples of the material under study. It should be realized that the use of nitric acid for the extractive isolation of iron is associated with the results that have a considerable error. Also, with the use of other techniques for sample preparation [ 14,151 too low results for the determinations of Fe were obtained, if the digestive agent was nitric acid alone. A quantitative dissolution of aluminium and iron by extraction with cont. HCl or dil. HNOs was achieved only upon sample pretreatment with HF (Method l(B)). In the case of aluminium the use of HF is critical for the determination of total Al both in the extracts and solutions following the dissolution process. Noteworthy is, however, that the efficiency of I-IF

E. Were&a et al./Adytica

Chimica Acta 330 (1996) 251-257

255

Table 5 Results of analysis of certified material NIST SRM 1575 pine needles after extraction Elements Certified value Found Method l(A) Fe Al cu Pb Cd Cad 2OOfl 545zt30 3.0f0.3 10.8f0.5 <0.5 c 0.41f0.02 Value determination 0.12*0.01 e 0.11*0.003f 72f13 66.2f3.8 f 178f4 84f6 b 298flO 3.1f0.2 10.7f0.5 0.34f0.05 0.4lfO.01 by 0.12f0.005 67f5 0.12f0.003 66f5

with acids and after wet ashing (ug g- dry weight)

Method l(B) 198f4 570flO 3.0f0.3 10.710.5 0.34f0.05 0.41f0.005

Method 2(A) 203f5 330f20 3.lf0.2 10.9f0.5 0.33f0.03 0.41f0.01

Method 2(B) 206f8 555f40 3.3f0.2 ll.lf0.5 0.31f0.04 0.41f0.01

Mg* Zn

0.12f0.002 68f4

0.12f0.002 7lf8

Results of the determinations b Results of the determinations agree Non-certified values. *Values in wt. %. White et al. [18]. Zunk [8].

in cont. HCl extracts. in dil. HNOs extractsFor

the remaining

elements the results of the determinations

in HCl and HNOs extracts

Table 6 Results of analysis of certified material Oriental Tobacco Leaves CTA-OTL-1 weight) Elements Certified value Found Method l(A) Zn cu Pb Cd Al Fe Ca Mg 31.7f1.2* 4470f210 in cont. HCl extracts. in dil. HNOs extractsFor 49.9f2.4 14.1f0.5 4.91f0.8 1.12f0.12 174Of290 49.2f0.7 14.4f0.3 4.53f0.5 1.20f0.10 375zt25 213f10b 79lf30 343f28 b 31.OkO.9 4524f180

after extraction

with acids and after wet ashing (ug g- dry

Method l(B) 48.9f0.9 13.9f0.5 4.86f0.4 1.22f0.10 1844fllO 1054fl20 31.5f0.7 4377f220

Method 2(A) 49.3fl.8 14.6f0.5 4.65f0.5 1.22fO.l 927f120 918f75 32.5f1.2 4539fl90

Method 2(B) 49.6f1.2 14.6f0.5 4.72f0.5 1.22zkO.l 1876flOO 1086fll8 32.1f1.3 4482fl50

Results of the determinations b Results of the determinations agree. Non-certified values. *Values in mg g-.

the remaining

elements the results of the determinations

in HCl and HNOs extracts

effect in the aluminium extractive isolation was achieved only if the sample was pre-heated with HF, followed by evaporation to dryness. Of significance here is the decomposition of aluminium-silicon compounds and the removal of silicon from the

solution under study. Analogous reported by other authors who used of sample preparation [16,17]. For iron the necessity of using treatment process is not so obvious

experience was various methods HF in the preand unequivocal

256

E. Wieteska et al./Analytica

Chimica Acta 330 (19%) 251-257 Table I Results of analysis of the control Julich (D) (ug g-i dry weight) Elements Reference

as for aluminium. The wet digestion without HF ensures a complete dissolution of iron. For both dry ashed and for extracted samples only the use of HF allowed to transfer iron quantitatively into solution. The foregoing facts suggest that the presence of HF facilitates dissolution of iron compounds that are hard to dissolve which need not necessarily be iron-silicon compounds. Following a preliminary treatment of the sample with I-IF, aluminium and iron may be quantitatively brought into solution both by the use of HCl and dil. I-INOs. For the remaining elements the use of I-IF in the acid extraction process is needless. The precision of the method advanced for the extractive isolation of the group of the elements studied, as estimated on the basis of the results of the analysis of pine needles, is satisfactory. The method advanced was tested on two standard reference materials, viz., NIST SRM 1575 pine needles and CTA-OTL-1 oriental tobacco leaves and a control sample of algae II supplied by the Environmental Specimen Bank in Itilich (D). A good agreement between the results of determinations of the elements with the values was obtained (cf. Tables 5-7) which has lent validity to the method.

sample Algae II from ESB in

value

Found Method l(A) 124f3 1.25f0.02 3.3f0.2 7.8-10.4

Zn Cd Pb cu

123f2 1.22f0.06 3.5f0.2 8.010.4

4. Conclusions In summary, the results of the study on vegetable material (pine needles) and on the standard reference materials, point out that the use of cont. HCl or dil. HNOs, with a previous treatment of sample with hydrofluoric acid, allows to obviate the operation of the destruction of the organic matrix and, at the same time, to dissolve quantitatively the following elements: Zn, Fe, Cu, Al, Ca, Mg, Pb and Cd. The proposed method can be used in several variants. The choice of suitable extractants depend on the nature of the elements to be determined, and on the instrumental analysis technique (flame or GF-AAS). For the extraction of Zn, Cu, Ca and Mg either cont. HCl or dil. HNOs acid may be used. For extractive separation of Pb and Cd it is advisable to use dil. HNOs as the elements are determined by the GFAAS technique. For a quantitative dissolution of aluminium and iron by acid extraction the use of I-IF is indispensable.

The method set forth for the extractive dissolution of Zn, Fe, Cu, Al, Ca, Mg, Pb and Cd from vegetable material may advantageously supersede the commonly used wet digestion and dry ashing techniques. The method features simplicity without any need of special equipment, short sample preparation time (ca. 45 mm), smaller amounts of reagents, thus a reduced contamination hazard, as well as lower costs of the analysis, accompanied by good precision and accuracy. The method may be of value in monitoring practice. Versatility of the method, apart from the analysis of several standard reference materials, were also confirmed by our studies on the extractive isolation of the analytes from tree bark and moss samples.

References

111 M. 121 G.
t31

141 PI 161 171 Bl [91 [lOI [111

Ihnat, J. Res. Natl. Bur. Stand., 93 (1988) 354. Knapp, B. Sadjadi and H. Spitzy, Fresenius Z. Anal Chem., 274 (1975) 275. I. Novozamsky, V.J. Houba, J. van der Lee, R. Van Eck and M.D. Mignorance, Commun. Soil Sci. Plant Anal., 24 (1993) 2595. G.N. Sokolova, Z.A. Shcherbakova and O.B. Shashova, Gig. Sanit., 5 (1989) 52. I.M. Skurikhin, J. Assoc. Off. Anal. Chem., 72 (1989) 286. M. Stoeppler (Ed.), Hazardous Metals in the Environment, Elsevier, Amsterdam, 1992. N.J. Miller-Ihli, J. Res. Natl. Bur. Stand., 93 (1988) 350. B. Zunk, Anal. Chim. Acta, 236 (1990) 337. S. Kokot, G. King, H.R. Keller and D.L. Massart, Anal Chim. Acta, 259 (1992) 267. M.A. Mateo and S. Sabate, Anal. Chim. Acta, 279 (1993) 273. M. Ahlgren, A. Kontkanen, K. Vattulainen and H. Vehvilainen, Analyst, 113 (1988) 285.

E. Wietesku et al./Analytica Chimica Acta 330 (1996) 251-257 [12] R.T. White, Jr., J. Assoc. Off. Anal. Chem., 72 (1989) 387.
[13] D.M. Kmuina, E.N. Savinova, T.V. Shumskaya, M.D. Alybaeva and A.V. Karyakin, Zh. Anal. Khim., 44 (1989) 567. W. Van Delft, H.J. Horstman, H. L-ers and G. Vos, Colloq. Atomspektrom. Spurenanal., 5 (1989) 603. [15] [16] [17] [18]

257

[14]

K. Topper and J. Kotuby-Almacher, Commun. Soil Sci. Plant Anal., 21 (1990) 1437. U. Bartels, CLB Chem. Labor. Biotech., 41 (1990) 640. C. Charlot, P. Ogereau and M. Texier, Analusis, 15 (1987) 386. R.T. White and G.E. Douthit, J. Assoc. Off. Anal. Chem., 68 (1985) 766.