369732

Talanta 52 (2000) 637643

Solid phase extraction and determination of lead in soil andwater samples using octadecyl silica membrane disks

modified bybis[1-hydroxy-9,10-anthraquinone-2-methyl]sulfide and flame

atomic absorption spectrometry

Mojtaba Shamsipur a,*, Farhad Raoufi b, Hashem Sharghi c

a Department of Chemistry, Razi Uni6ersity, Kermanshah, Iranb Department of Chemistry, Tehran Uni6ersity, Tehran, Iranc Department of Chemistry, Shiraz Uni6ersity, Shiraz, Iran

Received 6 November 1999; received in revised form 28 February 2000; accepted 20 March 2000

Abstract

A simple, reliable and relatively fast method has been developed to selectively separate and concentrate traceamounts of lead from aqueous samples for the measurement by flame atomic absorption spectrometry. By the passageof aqueous samples through an octadecyl-bonded silica membrane disk modified by a recently synthesized bis(an-thraquinone)sulfide, Pb2 ions adsorb quantitatively and almost all matrix elements will pass through the disk todrain. The retained lead ions are then stripped from the disk by minimal amount of acetic acid as eluent. Theproposed method permitted large enrichment factors of about 300 and higher. The limit of detection of the proposedmethod is 50 ng Pb2 per 1000 ml. The effects of various cationic interferences on the recovery of lead in binarymixtures were studied. The method was successfully applied to the determination of lead in soil and water samples. 2000 Elsevier Science B.V. All rights reserved.

Keywords: Lead(II); SPE; Octadecyl silica disks; Bis(anthraquinone)sulfide; Flame atomic absorption spectrometry

www.elsevier.com:locate:talanta

1. Introduction

The selective concentration, separation and de-termination of lead at trace levels is of specialimportance. This metal is among those heavy

elements that are of major interest in environmen-tal protection due to its cumulative toxicity andwell known health risk to animals and human[1,2]. Because of its widespread application as fueladditive, lead is emitted into the biosphere inconsiderable amounts [3]. The spectrophotometricmethods [4] and atomic absorption spectrometry[5] are among the most common methods em-ployed for the determination of lead in solutions.

* Corresponding author. Tel.: 98-831-723307; fax: 98-831-831618.

0039-9140:00:$ - see front matter 2000 Elsevier Science B.V. All rights reserved.

PII: S0 039 -9140 (00 )00390 -8

M. Shamsipur et al. : Talanta 52 (2000) 637643638

However, due to the presence of lead in environ-mental samples at low levels, its separation fromother elements present and also the use of apre-concentration step prior to the lead determi-nation is usually necessary.

Liquidliquid extraction and separation ofPb2 ion in the presence of classical [68] andmacrocyclic ligands [912] is frequently reportedin the literature. In order to reduce consumptionand exposure to hazardous organic solvents, dis-posal costs and extraction time, solid-phase ex-traction (SPE) can be used as an attractivealternative for the classical solventsolvent ex-traction methods [1315]. Recently, SPE diskswere successfully utilized for the extraction ofseveral organic [1619] and inorganic analytes[2023]. We recently modified octadecyl silicamembrane disks with some macrocyclic and clas-sical ligands for the extraction and determinationof ultra trace amounts of barium [21], mercury[22] and uranyl ion [23] by FAAS, CVAAS andspectrophotometry, respectively. We now reportan efficient and relatively rapid method for theselective extraction and concentration of Pb2

ions by octadecyl silica membrane disks modifiedwith a recently synthesized bis(anthraquinone)-sulfide derivative (bis[1-hydroxy-9,10-anthraquin-one-2-methyl]sulfide, BAS), as a highly selectiveand stable reagent, from soil and water samplesand its determination with FAAS. To the best ofour knowledge, modified octadecyl silica mem-brane disks have not been employed previouslyfor the separation and concentration of lead ionsfrom aqueous samples. It is noteworthy that, wehave recently reported a lead-selective membranepotentiometric sensor based on the anthraquinonederivative used in this study [24].

2. Experimental

2.1. Reagents

All acids used were of the highest purity avail-able from Merck chemical company and used asreceived. All organic solvents used were of HPLCgrade from Merck. Analytical grade nitrate saltsof lead, cadmium, mercury, cobalt, nickel, copper,zinc, magnesium, calcium, strontium, barium,lithium, sodium and potassium (all from Merck)were of the highest purity available and usedwithout any further purification except for vac-uum drying over P2O5. Bis[1-hydroxy-9,10-an-thraquinone-2-methyl]sulfide was synthesizedfrom commercially available initial reactants(Merck or Fluka) in \90% yield [25,26] and usedafter recrystallization from pure benzene and vac-uum drying. Doubly distilled deionized water wasused throughout.

2.2. Instruments

The lead determinations were carried out on aPerkinElmer 603 atomic absorption spectrome-ter with a hallow cathode lamp and a deuteriumbackground corrector, at a wavelength of 217 nm(resonance fine) using an airacetylene flame. TheAAS determinations of all other cations wereperformed under the recommended conditions foreach metal. The pH was determined with a Corn-ing ion analyzer 250 pH:mV meter with a com-bined glasscalomel electrode. Extractions wereperformed with 47 mm diameter0.5 mm thick-ness Empore membrane disks containing octade-cyl-bonded silica (8-mm particle, 60-A, pore size, 3M Co., St. Paul, MN) with a standard Millipore47-mm filtration apparatus.

2.3. Sample extraction

After placing the membrane disk in the filtra-tion apparatus, it was washed with 10 mlmethanol and then with 10 ml acetonitrile toremove all contaminants arising from the manu-facturing process and the environment. After dry-

M. Shamsipur et al. : Talanta 52 (2000) 637643 639

ing the disk by passing air through it for sev-eral min, a solution of 15 mg BAS dissolvedin 2 ml of chloroform was introduced to thereservoir of the apparatus and was drownslowly through the disk applying a slight vac-uum until the ligand penetrated the membranecompletely. The solvent was evaporated at50C.

After drying, the modified disk was washedwith 25 ml of water to pre-wet the surface ofthe disk prior to the extraction of Pb2 ionsfrom aqueous samples. Then 500 ml of the sam-ple containing 5 mg of Pb2 ions was passedthrough the membrane (flow rate20 mlmin1). After the extraction, the disk was driedcompletely by passing air through it for fewminutes. The extracted lead was then strippedfrom the modified membrane disk using 10 mlof 0.5 M solution of acetic acid (flow rate2ml min1). It is interesting to note that thereagent BAS is quite stable in acidic solutions,and it is sparingly washed out from the surfaceof the disks with the eluent used. Thus, eachmodified disk could sufficiently be used for atleast five times.

2.4. Lead ions in ri6er water and soil samples[27,28]

Accurately weighed 1 g of soil samples (lessthan 200 mesh, dried at 110C was poured intoa 250-ml beaker and 10 ml concentrated nitricacid was added to it. The mixture was gentlyheated under a hood until drying. After com-plete drying and cooling to room temperature, asecond 10-ml portion of concentrated nitric acidwas added and the procedure was repeated.Then 10 ml concentrated hydrochloric acid wasadded to the beaker and the mixture was gentlyheated until complete drying. After cooling, theresidue was dissolved in 10 ml of 1 M HCl andthe solution was then filtered into a 100-ml cali-brated flask, using a syringe filter (0.45 mm poresized). The sample was neutralized by properamounts of a 1 M NaOH solution and finallydiluted to the mark with water.

3. Results and discussion

It is well known that various hydroxyderivatives of 9,10-anthraquinone are able toform stable 2:1 (ligand-to-metal) complexes withalkaline earth, transition and heavy metal ions[2931], among which the resulting Pb2

complexes are of the most stable ones [29]. Due toits negligible solubility in water and the existenceof two 1-hydroxy-9,10-anthraquinone groups anda donating sulfur atom in its structure, we havepreviously used BAS as a very suitable ion carrierin construction of a Pb2 ion-selective electrode[24]. It should be noted that the log Kf value forthe formation of a 1:1 Pb2 BAS complex inacetonitrile solution, as estimated from theconductance measurements [32], was found to be\7. Thus, in this study we used the ligand as apotential modifier for the SPE and determinationof Pb2 ions by the membrane disks and FAAS.Some preliminary experiments were carried out inorder to investigate the quantitative retention ofPb2 ions by the octadecyl silica membrane disksin the absence and presence of BAS, after therecommended washing, wetting and conditioningprocedures were performed. It was found that,while the conditioned membrane disk itself doesnot show any tendency for the retention of leadions, the membrane disk modified by BAS iscapable to retain Pb2 ions in the sample solutionquantitatively (the test solution used contained 5mg lead in 25 ml water).

3.1. Choice of eluent

In order to choose the most effective eluent forthe quantitative stripping of the retained Pb2

ions by the modified disks, after the extraction of5 mg lead in 500 ml water, the lead ions werestripped with varying volumes of 1 M solutions ofdifferent acids and the results are summarized inTable 1. As seen, among four different acidsolutions used, 5 ml of 1 M acetic acid (or 10 mlof 0.5 M of the acid) can afford the quantitativeelution of lead from the disk. As is also obvious,the lower the acetic acid concentration used, thehigher the volume necessary for the quantitativeelution of the retained lead ions. It is noteworthy


that the acetate ion forms a highly stable complexwith Pb2 ion [33].

3.2. Effect of flow rate

The dependence of the uptake of Pb2 ions bythe modified disk on the flow rate was studied, theflow rate being varied from 1 to 30 ml min1. Itwas found that adsorption of the metal ion isquantitative and reproducible in this range. Simi-lar results for the extraction of organic [18,19] andinorganic [2123] species by octadecyl silica mem-brane disks have already been reported. The ratewas maintained at 20 ml min1 throughout theexperiments. On the other hand, quantitativestripping of Pb2 ions from the disk was achievedin a flow rate range of 0.54 ml min1, using 5ml of 1 M acetic acid. At higher flow rates (up to20 ml min1), quantitative stripping of leadneeded larger volumes of 1 M acetic acid. A flowrate of 2 ml min1 was chosen for further studies.

3.3. Effect of pH

Lead ion solutions were adjusted to the re-quired pH and passed through the modified disks.Then the metal ions were eluted from the disks by1 M acetic acid solution. Subsequent atomic ab-sorption analysis resulted in the percentage recov-eries of the eluted lead ions at various pH. ThepH range investigated was between 2 and 7. Theresults revealed that the percentage recovery oflead ions is nearly independent of pH. Higher pHvalues (\7) were not tested because of the possi-

bility of the hydrolysis of octadecyl silica in thedisks [22].

3.4. Effect amount of BAS

As the amount of loaded ligand on the mem-brane disk increases, the flow rate of the leadsolutions through the modified disk will decrease.Thus, the least amount of BAS necessary for thequantitative extraction of 5 mg Pb2 from a 500ml aqueous sample was studied by using varyingamounts of the reagent (i.e. from 5 to 25 mg). Theresults indicated that the extraction of lead isquantitative using above 15 mg of the ligand.Hence, subsequent extraction experiments werecarried out with 15 mg of BAS.

3.5. Analytical performance

The break-through volume of sample solutionwas tested by dissolving 5 mg of lead in 25, 50,100, 250, 500, 1000, 1500 ml water and the recom-mended procedure under optimal experimentalconditions was followed. In all cases, the extrac-tion by the modified membrane disk was found tobe quantitative. Thus, the break-through volumefor the method should be greater than 1500 ml.Consequently, by considering the final elution vol-ume of 5 ml and the break-through volume of1500 ml, an enrichment factor of 300 was easilyachievable.

The maximum capacity of the membrane diskmodified by 15 mg of BAS was determined bypassing 500 ml portions of an aqueous solution

Table 1Percent recovery of lead from the modified membrane disks using different stripping acid solutionsa

% RecoverybStripping acid solution

5 ml 15 ml10 ml 25 ml20 ml

8 12 15 16 17HNO3 (1 M)35 42HCl (1 M) 43 44 46

HBr (1 M) 60585448 52100 100CH3COOH (1 M) 100 100 100

68 100CH3COOH (0.5 M) 100 100 100

a Initial samples contained 5 mg Pb2 ion in 500 ml water.b RSDs obtained based on three replicate analyses were within 1.22.5.


Table 2Separation of lead from binary mixturesa

Amount takenDiverse ion % Found % Recovery ofPb2 ion(mg)

Li NAPDb2.1 98.9 (1.4)c

Na 2.3 NAPD 99.3 (1.7)NAPD2.0 99.1 (1.8)K

1.9Mg2 NAPO 96.4 (2)1.8Ca2 NAPD 101.2 (2.1)

3.7 (1.3)2.7 101.3 (1.6)Sr2

NAPDBa2 98.6 (1.4)3.0NAPD2.9 96.4 (1.3)Co2

2.6Ni2 NAPD 98.9 (1.6)NAPDCu2 99.8 (1.9)2.4NAPD2.2 97.3 (1.2)Zn2

1.6Cd2 NAPD 95.1 (1.1)3.0Hg2 NAPD 99.3 (1.2)

NAPD2.4 97.3 (1.4)Mn2

2.6Fe3 NAPD 98.4 (1.8)1.8Cr3 NAPD 98.3 (1.4)

NAPD3.0 99.3 (1.2)Al3

2.1VO3 99.2 (1.9)

a Initial samples contained 5 mg Pb2 and different amountsof diverse ions in 500 ml water.

b No adsorption, passes through disk.c Values in parentheses are RSDs based on three replicate

analyses.

[34,35], where Sb is the S.D. of the blank measure-ments (ten replicates) and m is the slope of thelinear calibration graph and K is a confidencefactor equal to 3. The LOD of the method wasfound to be 50 ng per 1000 ml. The reproducibil-ity of the procedure was found to be at the most2%.

In order to investigate the selective separationand determination of Pb2 ion from its binarymixtures with diverse metal ions, an aliquot ofaqueous solution (500 ml) containing 5 mg Pb2and milligram amounts of other cations was takenand the recommended procedures was followed.The results are summarized in Table 2. As seen,the lead(II) ions in the binary mixtures are re-tained almost completely by the modified mem-brane disk, even in the presence of up to about 3mg of diverse ions. It is interesting to note that,with the exception of Sr2 ions, the retention ofother cations by the modified disk is negligibleand they can be separated completely from thePb2 ion. However, in the case of Sr2 ion, itsretention by the disk is 3.7%.

In order to assess the applicability of the pro-posed method to real samples with different ma-trices containing varying amounts of a variety ofdiverse ions, it was applied to the determinationof lead in water and soil samples taken fromChaloos river (Chaloos, Iran) at different dis-tances from Alika lead mines (Chaloos, Iran). Theresults obtained for the water and soil samples bythe proposed method, after calibration by theexternal standardization method, and those ob-tained by inductively coupled plasma-atomicemission spectrometry (ICPAES) are summa-rized in Tables 3 and 4, respectively. As seen,there is a satisfactory agreement between the re-sults obtained by the proposed method and byICPAES. Moreover, as it is expected, the leadcontent of both water and soil samples decreasesby increasing distance from the lead mines.

4. Conclusions

BAS was used as a highly selective reagent tomodify the octadecyl silica membrane disks forthe successful SPE of Pb2 ions. In comparison

Table 3Determination of lead in water samples

Sample numbera Lead concentration (ppm)

Proposed method ICP

20.5 (1.8)1 21.0 (1.8)20.5 (1.4)20.1 (1.4)2

19.8 (1.8)3 20.0 (1.4)4 16.3 (1.2) 5 0.7 (1.4)

a Samples were collected from Chaloos river at distances of0.5, 0.7, 0.8, 2 and 15 km, respectively, from Alika lead mines.

containing 2000 mg lead through the disk, fol-lowed by determination of the retained metal ionsusing FAAS. The maximum capacity of the diskwas found to be 476 mg of Pb2 ions.

The limit of detection (LOD) of the proposedmethod for the determination of lead was studiedunder the optimal experimental conditions. TheLOD was obtained from CLODKbSbm1


Table 4Determination of lead in soil samples

Sample numbera Lead content of soil samples (mg g1)Lead concentration in final solution (mg ml1)

ICPProposed method

21.5 (1.4)1 214021.4 (1.2)11.3 (1.3)11.5 (1.4) 11502

3 4.0 (1.4) 4001.8 (1.2) 18045 1601.6 (1.1)1.3 (1.5) 1306

1.1 (1.2)7 1101.0 (1.5) 1008

0.9 (1.2)9 9010 0.2 (1.3) 20

a Samples were collected from Chaloos river at distances of 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 15 and 35 km, respectively, fromAlika lead mines.

with the previously reported organic reagents forPb2 ions [612], BAS possesses a high selectivityfor Pb2 ion over many other cations includingalkali, alkaline earth, transition and heavy metalions [24], it is quite stable at highly acidic mediaand can be easily synthesized from commerciallyavailable compounds at a high yield (i.e. \90%)[25,26]. The method is relatively rapid as com-pared with previously reported procedures for theseparation and determination of lead [612]. Thetime taken for the separation and analysis of leadion in 500 ml water samples is at the most 45 min.The method can be successfully applied to theseparation and determination of lead in realsamples.

References

[1] T.E. Berman, Toxic Metals and Their Analysis, Heydenand Son, London, 1980.

[2] C.D. Klassen, M.D. Amdur, J. Dull, Casarett and DullsToxicology, third ed., Macmillan, New York, 1986.

[3] J.O. Nriagu (Ed.), The Biochemistry of Lead in theEnvironment, Elsevier, Amsterdam, 1978.

[4] Z. Marczenko, Separation and Spectrophotometric Deter-mination of Elements, Ellis Horwood, London, 1986.

[5] B. Welz, Atomic Absorption Spectrometry, VCH, Am-sterdam, 1985.

[6] J. Mortatti, F.J. Krug, H. Bergamin, Energy Nucl. Agric.4 (1982) 82.

[7] Methods for Chemical Analysis of Water and Waste,EPA-600:4-79-020, Environmental Protection Agency, Re-search Triangle Park, NC, 1983.

[8] A. Arrebola Ramirez, D. Gazquez, I.M. de la Rosa, F.Moreno, Anal. Lett. 27 (1994) 1595.

[9] W. Szezepaniak, B. Jskowiak, Anal. Chim. Acta 140(1982) 261.

[10] Y. Sakai, N. Kawano, H. Nakamura, M. Takagi, Talanta33 (1986) 407.

[11] E.A. Novikov, L.K. Shpigun, Yu.A. Zolotov, Anal. Chim.Acta 230 (1990) 157.

[12] M. Kompani, M. Shamsipur, J. Sci. I. R. Iran 7 (1996) 13.[13] R.E. Majors, LC-GC 4 (1989) 972.[14] M. Moors, D.L. Massart, R.D. McDowall, Pure Appl.

Chem. 66 (1994) 277.[15] R.M. Izatt, J.S. Bradshaw, R.L. Bruening, Pure Appl.

Chem. 68 (1996) 1237.[16] E.R. Brawer, H. Lingeman, U.A.Th. Brinkman, Chro-

matographia 29 (1990) 415.[17] K.Z. Taylor, D.S. Waddell, E.J. Reiner, K.A. MacPher-

son, Anal. Chem. 67 (1995) 1186.[18] Y. Yamini, M. Ashraf-Khorasani, High Resolut. Chro-

matogr. 17 (1994) 634.[19] Y. Yamini, M. Shamsipur, Talanta 43 (1996) 2117.[20] L.B. Bjoklund, M. Morrison, Anal. Chim. Acta 343 (1997)

259.[21] Y. Yamini, N. Alizadeh, M. Shamsipur, Sep. Sci. Technol.

32 (1997) 2077.[22] Y. Yamini, N. Alizadeh, M. Shamsipur, Anal. Chim. Acta

355 (1997) 69.[23] M. Shamsipur, A.R. Ghiasvand, Y. Yamini, Anal. Chem.

71 (1999) 4892.


[24] H.R. Pouretedal, A. Forghaniha, H. Sharghi, M. Sham-sipur, Anal. Lett. 31 (1998) 2591.

[25] H. Sharghi, A. Forghaniha, Iran. J. Chem. Chem. Eng. 14(1995) 16.

[26] H. Sharghi, A. Forghaniha, J. Sci. I. R. Iran 7 (1996) 89.[27] R.M. Harrison, D.P. Laxen, Water Air Soil Pollut. 8 (1977)

387.[28] O. Hutzinger, The Handbook of Environmental Chem-

istry, vol. 3, 1980 Part A.[29] G.A. Qureshi, G. Svehla, M.A. Leonard, Analyst 104

(1979) 705.[30] P.L. Gutierrez, N. Nguyen, in: D. Dryhurst, K. Niki (Eds.),

Redox Chemistry and Interfacial Behavior of BiologicalMolecules, Plenum, New York, 1988.

[31] A.R. Fakhari, J. Ghasemi, M. Shamsipur, unpublishedresults.

[32] M.R. Ganjali, A. Rouhollahi, A.R. Mardan, M. Sham-sipur, J. Chem. Soc. Faraday Trans. 94 (1998) 1959.

[33] Ju. Lurie, Handbook of Analytical Chemistry, Mir,Moscow, 1975.

[34] ACS Committee on Environment Improvement, Anal.Chem. 52 (1980) 2242.

[35] J.D. Ingle, S.R. Crouch, Spectrochemical Analysis, Pren-tice Hall, Englewood Cliffs, NJ, 1988.

.

369732

Documents