重金属方面的英文文献

Analytica Chimica Acta 515(2004)

343–348

A continuous approach for the determination of Cr(VI)in sediment and soil based on the coupling of microwave-assisted water extraction, preconcentration, derivatization and photometric detection

S. Morales-Muñoz,J.L. Luque-Garc´ıa ∗, M.D. Luque de Castro

Department of Analytical Chemistry, University of Córdoba,Marie Curie Building, Campus of Rabanales, E-14071Córdoba,Spain

Received 19January 2004; received in revised form 19March 2004; accepted 19March 2004

Available online 25May 2004

Abstract

A dynamic system for the continuous leaching of Cr(VI)from sediment and soil based on both microwave assistance and iterative change

of the ﬂowdirection of the extractant through the sample cell has been developed. The microwave-assisted extractor has been coupled to a photometric detector through a ﬂowinjection interface in order to develop a fully automated method. The Cr(VI)extracted was monitored after derivatization with 1,5-diphenylcarbazide. Two approaches are proposed which differ in the inclusion of a preconcentration minicolumn packed with a strong anion exchange resin. A 0.04M ammonium buffer solution was used as extractant and 0.2g of sample—riversediment spiked with 50and 5␮g g −1for the method without preconcentration (methodA) and with preconcentration (methodB)—wassubjected to 8–14min of 300W microwave-assisted extraction. The within-laboratory reproducibility and repeatability were 2.6and 1.9for method A, and 4.0and 2.6for method B. The proposed methods have been compared with the reference EPA method 3060/7196.2004Elsevier B.V . All rights reserved.

Keywords:Microwave-assisted extraction; Hexavalent chromium; Sediment; Photometric detection

1. Introduction

The predominant use of chromium in industry (inpro-cesses such as plating, tanning, and paint and pigment production) unfortunately causes environmental concern. The toxicity of a metal depends on its oxidation state. In the case of chromium, Cr(III)is considered an essential microelement [1,2]while Cr(VI)is thought to be toxic [3]and carcinogenic. Thus, the identiﬁcationof Cr(VI)species in environmental samples is a problem of great concern be-cause of their toxicity to aquatic and terrestrial organisms, including humans. It is not sufﬁcientto give a total concen-tration of metal; instead, what is required to understand the potential toxicity of a sample is the concentration of Cr(VI)in the sample.

An ideal extraction method would extract the metal ef-ﬁcientlywithout converting metal ions from one oxidation state to another. Although much research has been focused

∗

Corresponding author:Tel.:+34-957-218615; fax:+34-957-218615. E-mail address:[email protected](J.L.Luque-Garc´ıa).

on the extraction and detection of chromium species in liq-uid samples such as natural and waste waters [4–6], the ex-traction of Cr(VI)from soil samples requires an additional

effort [7]. The difﬁcultyin determining Cr(VI)species in solid samples arises from the possible changes taking place in the chromium oxidation state.

The environmental protection agency (EPA)recog-nises four methods for sample preparation of hexavalent chromium:7195, coprecipitation; 7196, colorimetry with 1,5-diphenylcarbazide (DPC);7197, chelation/extraction;and 7198, differential pulse polarography. An ofﬁcialmethod, namely EPA method 3060, using alkaline diges-tion of Cr(VI)is also suggested for sample preparation of Cr(VI).A study was conducted by Gurknecht in 1983to evaluate the above four methods [8]. The study concluded that 7195and 7197methods were vulnerable to effects of matrix composition. The 7196colorimetric method based on the coloured complex formed between DPC and Cr(VI)[9]is one of the most sensitive and selective for Cr(VI)determination.

Several methods based on batch extraction have been proposed for the speciﬁcextraction of Cr(VI)from solid

344S. Morales-Muñozet al. /Analytica Chimica Acta 515(2004)343–348

samples [3,10–12]. There is only one case in the literature in which a dynamic extraction system has been used [13]. Almost all these methods are based on the use of ultrasound energy to accelerate the extraction step [3,10,11,13]. This energy has proved to be of great help in the extraction of Cr(VI)from solid samples without disturbing the species distribution.

Microwave-assisted extraction is also an expeditious, in-expensive and efﬁcientextraction technique. Over the past few years, this technique has been used in a discontinu-ous mode for accelerating the sample preparation step and avoiding some potential problems—namely,analyte losses and contamination [14,15]—relatedto conventional meth-ods. Dynamic microwave-assisted sample treatment has also been used [16–19], allowing the automation of the prelimi-nary step of the analytical process.

In the present research, a dynamic microwave-assisted ex-traction method has been developed and compared with the recommended EPA method 3060[21]. A commercial fo-cused microwave device (Soxwave-100)has been employed. The use of a dynamic extraction system [20]facilitates the automation of the whole analytical process. Thus, a fully automated approach in which a ﬂowinjection (FI)manifold is used as interface for the coupling of a microwave-assisted extractor with a photometric detector is proposed. The inclu-sion of a minicolumn packed with a strong anion-exchange (SAE)resin in the FI system allowed preconcentration of the Cr(VI)extracted before detection, thus allowing the analysis of samples with low levels of the target analyte.

2. Experimental

2.1. Instruments and apparatus

Microwave extraction was performed with a Soxwave-100focused microwave digestor (Prolabo,Fontenay-sous-Bois, France) with a maximum irradiation power of 300W. A TX 32device (Prolabo)was used for the control of the mi-crowave unit. A Gilson Minipuls-3low-pressure peristaltic pump (Gilson,Worthington, OH, USA) programmed for changing the rotation direction a preset intervals, three Rheo-dyne low-pressure selection valves (Rheodyne,Cotati, CA, USA), a laboratory-made chamber of Teﬂon(7cm ×7. 5mm i.d.) and Teﬂontubing of 0.8mm i.d. were used to build the leaching system.

Two Gilson Minipuls-3low-pressure peristaltic pumps, a Rheodyne Model 5041low-pressure injection valve, a laboratory-made minicolumn (3cm ×2mm i.d.) packed with SAE resin, a laboratory-made debbubler and Teﬂontubing of 0.8mm i.d. were used to construct the two dynamic man-ifolds. One of the channel of PP2was used as PP1.

A Waters model 490liquid chromatographic spectropho-tometer equipped with a low-volume ﬂow-celland Knauer x –t recorder was employed for the photometric monitoring of the Cr–DPCcoloured complex at 540nm.

2.2. Reagents and standards

Cr(VI)and Cr(III)working standard solutions were prepared from 1000␮g ml −1stock standard solutions prepared by K 2CrO 4and Cr 2O 3(Merck,Darmasdt, Germany). The Cr(VI)complexing reagent solution of 1,5-diphenylcarbazide (DPC)(Panreac,Barcelona, Spain) was prepared by dissolving 0.2g of DPC in 40ml of ethanol, diluting to 100ml with water and storing in a light-excluding bottle. Ammonia solution and ammonium sulfate (Panreac)were used for preparation of 0.04and 0.5M ammonium buffer (pH8) solution used as extractant and eluent, respec-tively. Ultrapure water from a Milli-Q system (Millipore,Canada, USA) was used throughout. All reagents were of analytical reagent grade. An SAE resin (Dowex1-X8-400; Sigma–Aldrich,Steinheim, Germany) was used for precon-centration of Cr(VI).The resin was supplied in the chloride form and was cleaned prior to use by slurring it with 3M HCl, allowing it to stand for 10min and then decanting off the acid. This procedure was repeated three times. After pouring off the last portion of the cleaning acid, the resin was slurried with 1M HCl and dried prior to use. 2.3. Samples

Two hundred and ﬁftygrams of river sediment as ma-trix, spiked with Cr(VI)to obtain a ﬁnalconcentration of 50␮g g −1, was used to carry out the optimisation study. A 250g portion of river sediment was spiked with 25␮g g −1of Cr(VI)and other 250g portion with 5␮g g −1of Cr(VI).These spiked levels were selected in order to obtain sedi-ments with environmentally representative concentrations. Six aliquots of 50g were spiked with Cr(VI)to obtain a ﬁnalconcentration in the sediment of 50␮g g −1in all them and with Cr(III)to a ﬁnalconcentration of:5, 10, 15, 20, 25and 30␮g g −1from the ﬁrstto the last aliquot. The sediment thus prepared was aged for three months in order to simulate the matrix–analyteinteraction in natural samples. Two types of natural contaminated soil (namely,clayey and slimy) were selected for validate the pro-posed method by comparison with the EPA method 3060[21].

2.4. Procedures

2.4.1. EPA method 3060/7196

2.4.1.1. Leaching step. A 50ml volume of extractant (asolution consisting of a mixture of 0.5M NaOH and 0.28M Na 2CO 3at pH 11.5) and 2.5g of sediment sample were poured into a beaker and heated to 90–95◦C on a hot-plate during 1h. The cooled extract (pH>12) was ﬁlteredthrough a 0.45␮m membrane ﬁlter,and then, the ﬁltratewas neutralised to pH 7.5with concentrated HNO 3and stored until the following step.

S. Morales-Muñozet al. /Analytica Chimica Acta 515(2004)343–348345

LEACHING STEP

DETERMINATION STEP

Fig. 1. Experimental set-up for development of the methods for Cr(VI)without (A)and with preconcentration step (B).LC, leacher-carrier; St, standard; PP, peristaltic pump; PPP, programmable peristaltic pump; SV , selection valve; W, waste; ME, microwave extractor; EC, extraction cell; R, refrigerant; MEC, microwave extraction controller; EX, extract; MC, mixing coil; RC, reaction coil; DB, debubbler; D, detector; E, eluent; IV , injection valve; SAEC, strong anion-exchange column.

2.4.1.2. Determination step. A proper amount of the neu-tralised ﬁltrateand 1ml of DPC solution (2mg ml −1) were poured into a 50ml calibrated ﬂaskand 1%HCl was used to adjust the solution to pH 2, necessary for the formation of the red–violetcomplex, which was monitored photomet-rically at 540nm.

2.4.2. Proposed procedure

Two different methods were developed using the mani-folds in Fig. 1, which differ in the inclusion of a preconcen-tration unit in Fig. 1B in order to achieve lower determina-tion levels.

2.4.2.1. Leaching step. An amount of 0.2g sediment was placed into the sample chamber. The closed system (witha total volume of 2ml) was ﬁlledwith the extractant (0.04M ammonium buffer solution) impelled by peristaltic pump PP1by maintaining selecting the leacher-carrier (LC)through valve SV1with the valve SV2in the closed-circuit position. The sample chamber was placed in the microwave vessel, which contained ∼=50ml of water, and irradiated at 300W for a preset time depending on the sam-ple matrix. During microwave irradiation, the direction of the extractant (at1.2ml min −1) was changed each 30s in an iterative manner, thus minimising both dilution of the extract and increased compactness of the sample in the ex-traction chamber, and avoiding overpressure in the system as a result. In addition, a closer sample–extractantcontact is achieved, thus accelerating the removal of the target an-alyte. After extraction, selection valve SV3was switched

and the extract was driven either to the preconcentration system (at0.2ml min −1) or directly to the detection system (at0.25ml min −1).

For introduction of the standards in the system, valve SV1selected the standard (St)channel with valve SV2in the closed-circuit position. In this way, the extraction sys-tem was ﬁlledwith a given standard. An amount of 0.2g of sediment without detectable level of the target analyte was placed into the extraction chamber in order to obtain a stan-dard volume equal to those of the extracts (whenextraction is performed). Once the system was ﬁlled,selection valves SV1and SV2were switched. In this way, the leacher-carrier drives to the waste the standard volume between both valves (SV1and SV2). Then, selection valves SV2and SV3were switched to the closed-and open-circuit positions, respec-tively, in order to drive the standard either to the preconcen-tration or detection system.

2.4.2.2. Preconcentration step. The extract from the closed system was driven (at0.2ml min −1) to a minicolumn packed with SAE resin where the analyte was retained. The minicolumn was located in the loop of an injection valve, thus allowing elution in the direction opposite to retention. Elution was carried out by passing through the minicolumn a 0.5M ammonium buffer stream at 0.25ml min −1. The elu-ate was driven to the spectrophotometer for determination after derivatization with DPC.

2.4.2.3. Determination step. The extract or the eluate from the preconcentration step was merged with an 1%HCl

346S. Morales-Muñozet al. /Analytica Chimica Acta 515(2004)343–348

stream and was driven to a mixing coil (1m ×0. 8mm i.d.). Then, the resulting stream was merged with 2mg ml −1DPC solution and driven to a reaction coil (1. 5m ×0. 8mm i.d.) in order to form the coloured complex. Both the HCl and DPC streams were pumped at 0.25ml min −1. Photometric detection was performed at 540nm. A debubbler was con-nected in-line to remove occluded air in the solid, which could give rise to parasitic signals at the detector.

3. Results and discussion

The order used for optimising the steps involved in the overall method was as follows:ﬁrst,the formation of the coloured complex was optimised for checking the other pre-vious steps; then, the variables affecting the leaching step and, ﬁnally,the preconcentration step, which is an optional step that provides a more sensitive alternative for the anal-ysis of soils with low levels of Cr(VI)(below1.2␮g g −1). 3.1. Optimisation of the coloured complex formation A multifactorial design methodology was used for the op-timisation of the main factors affecting the formation of the red–violetcomplex (namely,the ﬂowrates and the concen-tration of the HCl and DPC solutions and the ﬂowrate of the sample). The optimisation procedure and the results ob-tained were similar to those of Ref. [13]. The ranges assessed and optimum values found are shown in Table 1. 3.2. Optimisation of the continuous microwave-assisted extraction procedure

The variables optimised in the leaching step were the irradiation power, the irradiation time, the extractant ﬂowrate in the closed circuit during microwave irradiation and the time interval between successive changes of the ﬂowdirection of the extractant. A 0.04M ammonium buffer solution was selected as extractant as it had proved to be efﬁcient.The volume of extractant (2ml) corresponded

Table 1

Ranges and optimum values for the variables affecting the different steps Step Detection

Variable

Extract ﬂowrate (mlmin −1) DPC ﬂowrate (mlmin −1) Acid ﬂowrate (mlmin −1) DPC concentration (mgml −1) Acid concentration (%,v/v)Irradiation power (%)Irradiation time (min)

Extractant ﬂowrate (mlmin −1) Retention ﬂowrate (mlmin −1) Elution ﬂowrate (mlmin −1) Breakthrough volume (ml)

with the capacity of the closed circuit and was kept constant.

A univariate approach was used for the optimisation of the delay time –time interval between successive changes of the ﬂowdirection after the total volume of extractant had circulated once through the sample cell. The optimisation was developed under an irradiation power of 150W, a ﬂowrate of the extractant of 0.5ml min −1and 5min of irradi-ation time. After testing times of 10, 20, 30and 40s, the highest efﬁciencywas achieved using a delay time of 20s for changing the ﬂowdirection. Blockage of the sample cell was observed for longer times.

The irradiation power, the irradiation time and the ex-tractant ﬂowrate were optimised by the experimental de-sign methodology, as they were presumably interrelated (Table 1). A full two-level factorial design involving an overall of 23=8experiments plus three centred points was built for a screening study of the behaviour of the main factors affecting the extraction process [22]. The up-per and lower values given to each factor were selected from the available data and the experience gathered in the preliminary experiments.

The conclusions of the study were that the extractant ﬂowrate was not an inﬂuentialfactor in the range under study. However, the results showed better recoveries with the highest value tested. Thus, the optimum ﬂowrate was 1.2ml min −1. The irradiation time and the irradiation power were the key factors with a positive effect on the extraction efﬁciency.Higher values should be tested; however, con-cerning the irradiation power, the upper value of the design was the maximum power provided by the extractor used (300W) which yielded the best value. 3.3. Kinetics study

To determine the optimum extraction time for total re-moval of Cr(VI)as a function of the sample matrix, a study of the extraction kinetics was performed for spiked river sediments and natural contaminated soils. The other extrac-tion variables were ﬁxedat their optimum values. As can be

Tested range 0.25–1.750.25–1.750.25–1.752–81–2540–1002–160.2–1.20.1–1––

Optimum value

0.250.250.2521100See text

1.20.20.25>5

Leaching

Preconcentration

S. Morales-Muñozet al. /Analytica Chimica Acta 515(2004)343–348

347

C r V I e x t r a c t e d

)

10864200

Extraction time (min)

Fig. 2. Kinetics study of the extraction of Cr(VI)from different matrices.

seen in Fig. 2, total removal of Cr(VI)was obtained after 10min for the river sediment spiked at 50␮g g −1and 8min for the sediment samples spiked at 25and 5␮g g −1, while 12and 14min were necessary for slimy and clayey natural contaminated soils, respectively.

3.4. Optimisation of the preconcentration step

In order to obtain a more sensitive method, a preconcen-tration minicolumn was included in the FI manifold.

An SAE resin and 0.5M ammonium buffer were selected as sorbent material and eluent, respectively, based on the good results reported in the literature for the isolation of Cr(VI)from Cr(III)and other cations [3,12]. The elution ﬂowrate was not optimised as preconcentration was per-formed in the same manifold as detection and an extract ﬂowrate of 0.25ml min −1was previously optimised for a good development of the derivatization reaction. The reten-tion ﬂowrate was optimised in the range 0.1–1ml min −1and the results were that the recovery increased when the ﬂowrate decreased from 1to 0.2ml min −1and levelled off for lower values, so a ﬂowrate of 0.2ml min −1was selected for further experiments (Table 1). Samples volumes between 1and 5ml, which contained 5␮g g −1of Cr(VI),were passed through the minicolumn. The signal remained constant up to 5ml, so the breakthrough volume should be higher than 5ml.

3.5. Determination of Cr(VI)in the presence of Cr(III)In order to evaluate the feasibility of the proposed ap-proach for the extraction of Cr(VI)in the presence of Cr(III)without disturbing the species distribution, several sediment samples containing the same amount of Cr(VI)(50␮g g −1) and variable amounts of Cr(III)(from5to 30␮g g −1) were subjected to the extraction process. As can be seen in Table 2, the proposed approach does not disturb the species distribu-tion at Cr(VI)/Cr(III)ratios higher than 2:1.3.6. Features of the method

Calibration curves were obtained by using a linear plot of the peak area as a function of the standard con-centration of Cr(VI).The linear dynamic ranges for the two proposed methods were 1.5–7.5␮g ml −1—withcor-relation coefﬁcients(r 2) of 0.9894—formethod A; and 0.5–6.5␮g ml −1—withr 2=0. 9954—whenthe preconcen-tration step was included (methodB). The detection limits were 0.12␮g ml −1and 1.2␮g g −1, for the extract and soil, respectively, in the case of method A, and 0.012␮g ml −1and 0.12␮g g −1for method B.

The precision of the methods, expressed as within-laboratory reproducibility and repeatability, was studied in a single experimental set-up with duplicated for each method [23]. The experiments were carried out using 0.2g

Table 2

Recoveries obtained for Cr(VI)in the presence of Cr(III)Sample 1234567

Cr(VI)added (␮g g −1) [1**********]050

Cr(III)added (␮g g −1) [1**********]0

Cr(VI)recoveries (%)93.192.894.1101.8107.8110.4112.6

R.S.D. a (%)2.63.03.50.30.20.10.6

Relative standard deviation (n =3).

348S. Morales-Muñozet al. /Analytica Chimica Acta 515(2004)343–348

Table 3

Comparison of the proposed methods with the reference EPA method 3060Sample

Natural contaminated clayey soil Natural contaminated slimy soil

Method A a 7.32±0.166.91±0.21

Method B a 7.15±0.257.00±0.39

EPA method a 7.40±5.716.81±2.16

Amount of Cr(VI)extracted expressed as mean (␮g) ±relative standard deviation (n =3).

of spiked sediment containing 50␮g g −1for the method without preconcentration and 5␮g g −1for the method with preconcentration, under the optimum working conditions. The within-laboratory reproducibility and repeatability were 2.6and 1.9for method A, and 4.0and 2.6for method B. 3.7. Comparison of the proposed methods with the EPA method 3060

The optimised proposed methods were validated by com-parison with the EPA method 3060in terms of efﬁciencyand precision. The recoveries, expressed as an average of three extractions, were similar for the three methods (Table 3). However, the precision of the proposed methods was better than that obtained by the EPA method.

4. Conclusions

The automated approach based on the coupling of a microwave-assisted extractor with or without a precon-centration unit prior to a photometric detector allows the quantitative determination of Cr(VI)in spiked and natural sediment and soil samples. The dynamic microwave-assisted extraction based on the use of a closed extraction system with iterative change of the ﬂowdirection of the extractant during extraction overcomes the compactness of the sample that could cause overpressure of the system and favours a closer sample–extractantcontact, thus achieving an efﬁcientextrac-tion of Cr(VI)without disturbing the original species distri-bution. The proposed approach also allows obtaining results similar to those provided by the reference procedure, but in a shorter time (theextraction was performed in 10–14min versus 1h for the reference method) and with better precision.

Acknowledgements

Spain’sComisiónInterministerial de Ciencia y Tec-nolog´ıa (CICyT)is gratefully acknowledged for ﬁnancialsupport (projectBQU-2003-01333).

References

[1]A. Kortenkamp, M. Casadevall, S.P. Faux, A. Jenner, R.O.J. Shayer,

N. Woodbridge, P. O’Brien,Arch. Biochem. Biophys. 329(1996)199.

[2]S.L. Brauer, K.E. Wetterhahm, J. Am. Chem. Soc. 113(1991)3001. [3]J. Wang, K. Ashley, E.R. Kennedy, C. Neumeister, Analyst 122

(1997)1307.

[4]M.J. Marques, A. Salvador, A. Morales-Rubio, M. de la Guardia,

Fresenius J. Anal. Chem. 367(2000)601.

[5]N.K. Hamadi, X.D. Chen, M.M. Farid, M.G.Q. Lu, Chem. Eng. 84

(2001)95.

[6]Y .B. Martiúnez,N.A.F. Roig, F.J. López,F. Hernández,J. Chro-matogr. A 926(2001)265.

[7]M.J. Marques, A. Salvador, A. Morales-Rubio, M. de la Guardia,

Fresenius J. Anal. Chem. 362(1998)239.

[8]W. Gurknecht, R. Handy, L. Michael, E. Pellizari, L. Sheldon, M.

Umana, D. Whitaker, Evaluation of methods for determining hexava-lent chromium in extracts of industrial wastes, Contract No. 68-03-3099, US Environmental Monitoring and Support Laboratory, 1983. [9]US Environmental Protection Agency (EPA),Physical/Chemical

Methods, 3rd ed., Method 7196, Test Methods for Evaluating Solid Waste, US Government Printing Ofﬁce,Washington, DC, 1990. [10]J. Wang, K. Ashley, D. Marlow, E.C. England, G. Carlton, Anal.

Chem. 71(1999)1027.

[11]K. Ndung’u,N.K. Djane, F. Malcus, L. Mathiasson, Analyst 124

(1999)1367.

[12]G.P. Foy, G.E. Pacey, Talanta 51(2000)339. [13]J.L. Luque-Garc´ıa, M.D. Luque de Castro, Analyst 127(2002)1115. [14]H.M. kingston, L.B. Jassie, Introduction to Microwave Sample Prepa-ration, ACS Professional Reference Book, American Chemical So-ciety, Washington, DC, 1998.

[15]F.E. Smith, E.A. Arsnault, Talanta 43(1996)1207.

[16]E.S. Beary, P.J. Paulsen, L.B. Jassie, J.D. Fassett, Anal. Chem. 69

(1997)758.

[17]T. Duelcks, LaborPraxis 20(1996)68. [18]D. Lofty, Lab-Equip-Dig 30(1992)13.

[19]M.C. Herrera, M.D. Luque de Castro, J. Anal. Atom. Spectrom. 17

(2002)1530.

[20]A. Rios, M.D. Luque de Castro, M. Valcárcel,M. Grasserbauer, Int.

J. Environ. Anal. Chem. 66(1997)285.

[21]US Environmental Protection Agency (EPA),Test Methods for Eval-uating Solid Waste, Method 3060A, Alkaline Digestion for Hexava-lent Chromium, US Government Printing Ofﬁce,Washington, DC, 1996.

[22]Statgraphics Plus for Windows, Version 2.2, Rockville, MD, USA,

1992.

[23]D.L. Massart, B.G.M. Vanderginste, L.M.C. Buydens, S. De Jong, P.J.

Lewi, J. Smeyers-Verbeke, Handbook of Chemometrics and Quali-metrics, Part A, Elsevier, Amsterdam, 1997.