Environ. Sci. Technol. 2003, 37, 1043-1047
Chromium(VI) Removal through Facilitated Transport Using CYANEX 923 as Carrier and Reducing Stripping with Hydrazine Sulfate F. J. ALGUACIL* AND M. ALONSO Centro Nacional de Investigaciones Metalu ´ rgicas (CSIC), Avda Gregorio del Amo 8, Ciudad Universitaria, 28040 Madrid, Spain
The transport of chromium(VI) through a flat-sheet supported liquid membrane (FSSLM) containing CYANEX 923 (mixture of phosphine oxides) as a carrier has been studied. The permeation of the metal is investigated as a function of various experimental variables: hydrodynamic conditions, concentration of chromium(VI) and HCl in the feed phase, CYANEX 923 concentration and diluent in the membrane, and strippant concentration in the receiving phase. By using hydrazine sulfate in the receiving phase, chromium(VI) is immediately reduced to the less toxic chromium(III). The aqueous mass transfer coefficient and the thickness of the aqueous boundary layer were calculated from the experimental results. The selectivity of CYANEX 923-based FSSLM toward different metal ions and the behavior of the system against other carriers is presented.
Introduction Nowadays membrane technologies (MTs) play an increasingly important role as unit operations for resource recovery, pollution prevention, energy production, and environmental monitoring and quality control with expected sales of membrane systems topping $1.5 billion by 2002 (1). From the membrane processes under consideration, liquid membranes (LMs) have been showing potential, and different configurations of LMs are being investigated for various applications. In the form usually designated as an immobilized liquid membrane (ILM) or supported liquid membrane (SLM), an organic solution (unmixable with water) containing an active complexing agent or carrier in a suitable diluent impregnates, by capillary action, the pores of a porous hydrophobic polymeric support. This membrane system is clamped between two cells or packed in fibers and modules, which separates the aqueous feed and the receiving phases. Various SLMs techniques in flat-sheet (FSSLM), hollow fiber (HFSLM), and spiral-wound configurations are under consideration. Whereas flat geometry is very useful for first laboratory studies or as an analytical tool, for industrial purposes, a flat geometry is less effective since the ratio of surface area to volume is too low. Hollow fiber and spiralwound modules are best suited to provide high surface areas to volume ratio. The removal of metal ions from dilute or concentrated solutions has received a great deal of attention for recovery of valuable metals or decontamination of effluents. Among the heavy metals, chromium, especially in its (VI) oxidation * Corresponding author e-mail:
[email protected]. 10.1021/es020585s CCC: $25.00 Published on Web 01/31/2003
2003 American Chemical Society
state, is considered one of the most toxic elements despite the broad use of this metal in the industry (2). Thus, its removal from the various and different chromium(VI)-bearing effluents is a primary target in industrialized countries. Several techniques (e.g., cementation, ion exchange, etc.) are being using in this role, and membrane technologies had also found here their particular application field (3). Thus, liquid membranes in various configurations had been used for the removal and recovery of chromium(VI) from aqueous phases (4-14), with quaternary ammonium salts as the most used carriers for chromium(VI) separation. Moreover, the recovery of chromium in the stripping or receiving phase as a less toxic form, such as chromium(III), is also of a practical interest, although no one of the above referenced membrane studies had investigated this approach. Since before scaling up the FSSLM, in either the form of hollow fiber or spiral-wound membrane systems, laboratory results are needed in order to design an efficient recovery process. On the other hand, CYANEX 923 has been suggested as a potential extractant for a number of metals including gold (15, 16), lanthanides and yttrium (17), zirconium and hafnium (18), iron(III) (19), and cadmium (20). Also the extraction of minerals acids by this extractant (21) is described; few data are available in the literature about the extraction of chromium(VI) by CYANEX 923, the extraction being quantitative in the 0.2-2 M HCl range, but no information about the extraction mechanism is given (19) although the extraction of chromium(VI) by tri-n-octylphosphine oxide (TOPO), which exhibits extraction properties similar to that of CYANEX 923, is reported (22). The predominate species extracted into the organic phase were H2CrO4‚(TOPO) and H2Cr2O7‚(TOPO)3. The present investigation deals with the removal of chromium(VI) from HCl medium in a FSSLM configuration using CYANEX 923 (industrial phosphine oxides mixture) as carrier and reducing stripping with hydrazine sulfate.
Experimental Section CYANEX 923 (CYTEC, Canada) was used as carrier for most experiments in the present investigation. CYANEX 923 is a mixture of various trialkylphosphine oxides. The composition of this reagent was determined and reported recently (23). Other carriers used were as follows: tributyl phosphate, abbreviated as TBP (Fluka); dibutylbutylphosphonate, abbreviated as DBBP (Albright and Wilson); and Aliquat 336 (Cognis, Germany). Aliquat 336 is a quaternary ammonium salt, usually shipped in the chloride form. Cumene (Fluka) was used as the main diluent, Solvesso 100 (aromatic diluent) was obtained from ExxonMobilChem, Iberia, Spain. Stock metal solutions were prepared by dissolving the required amount of K2Cr2O7 (Merck) in distilled water. All other chemicals used in the present study were of AR grade. The organic membrane phase was prepared by dissolving the appropriate volume of CYANEX 923 (carrier) in cumene (diluent) to obtain organic solutions of different concentrations. The polymeric support was impregnated with the carrier solutions by immersion for 24 h. The impregnated membrane was allowed to drip for 5 s before being placed in the FSSLM cell. Previous experiments had shown that prolonged immersion times (e.g., 36 and 48 h) of the membrane in the carrier solution had not influenced the permeation coefficient values obtained using 24 h. The flatsheet membrane used was Millipore Durapore GVHP4700 (USA) of 12.5 × 10-3 cm thick microporous polyvinylidene fluoride (PVDF) film with nominal porosity of 75% and effective pore size of 2.2 × 10-5 cm. VOL. 37, NO. 5, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1043
FIGURE 1. Schematic concentration profile for the supported liquid membrane. L, carrier. The batch transport experiments were carried out in a permeation cell consisting of two cubic compartments made of methacrylate and separated by the membrane. The exposed membrane area was 11.3 cm2, and the volume of the feed and receiving solutions was 200 mL (each). The experiments were performed at 20 °C at a mechanical stirring speed of 1600 rpm in the feed phase and 1000 rpm in the receiving phase, except in the experiments where the stirring speed was varied. Agitation was performed in both compartments by using cylindrical methacrylate impellers having a diameter of 2.4 cm. Membrane permeability was determined by monitoring chromium (or other metal) concentration by atomic absorption spectrophotometry (Perkin-Elmer 1100B, England) in the feed phase as a function of time. The permeation coefficient (P) was computed by the following equation (2426):
ln
[Cr]t [Cr]0
A ) - Pt V
(1)
where [Cr]t and [Cr]0 are the chromium concentrations in the feed phase at an elapsed time and time zero, respectively; A is the effective membrane area; V is the volume of the feed phase; and t is the elapsed time.
Results and Discussion In the transport of chromium(VI) through a SLM containing CYANEX 923 as a mobile carrier, the concentration profile across the permeation cell is schematically shown in Figure 1. Chromium(VI) in the feed solution diffuses toward the feed-membrane interface, where complex(es) formation between Cr(VI) and CYANEX 923 (Cr(VI)-L in the figure) occurs. The organic complex(es) then diffuse(s) through the membrane toward the membrane-stripping interface, where Cr(VI) is stripped and reduced to Cr(III) in the stripping solution. This reduction is quick, and it is characterized by the appearance of a greenish color in the aqueous solution. The stripping step regenerates the carrier and then diffuses back to the feed-membrane interface, after which the process is repeated. Accordingly with the corresponding chromium(VI) species distribution diagram (27), a tentatively mechanism that explained the chromium transport (feed phase) through the membrane can be represented by the next equation: H+ aq + HCrO4aq + qLm S H2CrO4‚qLm
(2)
where q is a stoichiometric coefficient (probably 1 (22)), and aq and m refer to the feed and membrane phases, respectively. In the stripping phase, the possible reaction occurs:
4H2CrO4‚qLm + 3N2H4‚H2SO4s f
24Cr3+ s + 3SO4s + 6OHs + 3N2 + 10H2O + 4qLm (3)
1044
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 5, 2003
FIGURE 2. Influence of the stirring speed in the feed phase on chromium permeability. Stirring speed in the receiving phase: 1000 rpm.
TABLE 1. Influence of Hydrazine Sulfate Concentration on Chromium Permeabilitya [hydrazine sulfate] (g/L)
P (cm/s)
1 2.5 5 7.5 10
8.3 × 10-3 9.5 × 10-3 1.6 × 10-2 1.6 × 10-2 1.6 × 10-2
a Aqueous feed: 0.01 g/L Cr(VI) in 0.5 M HCl. Organic phase: 0.50 M CYANEX 923 in cumene. Receiving phase: hydrazine sulfate solutions.
where s and m refer to the stripping and membrane phases, respectively. Such a process is called coupled cotransport, where the counterion concentration in the feed solution is used as the driving force for metal transport. Influence of the Stirring Speed in the Feed Phase. The influence of stirring speed was studied in order to optimize uniform mixing of both aqueous phases and to minimize thickness of aqueous boundary layer with feed and receiving conditions being maintained as 0.01 g/L Cr(VI) in 0.5 M HCl and 5 g/L hydrazine sulfate, respectively. The carrier concentration was 0.50 M in cumene immobilized on the Durapore microporous support. Results are shown in Figure 2. The permeability coefficient becomes independent of the stirring speed above 1200 rpm. Consequently, the thickness of the aqueous diffusion layer and the aqueous resistance to mass transfer were minimized, and the diffusion contribution of the aqueous species to the mass transfer process is assumed to be constant (26, 28, 29). Influence of the Stirring Speed and Stripping Agent Concentration of the Receiving Solution. The effect of varying the stirring speed (600-1600 rpm) in the receiving solution was also investigated using the same phases as above and maintaining a 1600 rpm constant speed in the feed solution. Results obtained show that maximum chromium permeability (P ) 1.6 × 10-2 cm/s) is obtained when a stirring speed of 1000 rpm in the receiving phase is used. Different concentrations of hydrazine sulfate were studied as stripping agents. The results are given in Table 1; it can be seen that, by using concentrations greater than 5 g/L, the chromium permeation was not improved, whereas this permeation decreases at lower concentrations. In the stripping process, hydrazine sulfate reduces Cr(VI) to Cr(III), which is not extracted by CYANEX 923. Thus, chromium is released to the receiving phase (see above). As a result of the previous experiments, 5 g/L hydrazine sulfate was selected as the stripping reagent concentration in the receiving phase.
FIGURE 3. Influence of the HCl concentration in the feed phase on chromium permeability. Aqueous feed: 0.01 g/L Cr(VI) and HCl. Organic phase: 0.50 M CYANEX 923 in cumene. Receiving phase: 5 g/L hydrazine sulfate solutions. Influence of the HCl Concentration in the Feed Phase. To study the influence of the hydrochloric acid concentration in the feed phase, experiments were performed at various HCl concentrations, keeping the carrier concentration in the membrane constant. The results are shown in Figure 3. As can be seen from the table, chromium(VI) permeability remains almost constant up to 1 M HCl and then decreases. This can be probably due to the existence of less extractable chromium(VI)-cloro-oxo complexes (i.e., CrO3Cl-) in the aqueous phase. Influence of the Carrier Concentration. The effect of carrier concentration on chromium permeation was studied. Experimental conditions were established as organic phases of Cyanex 923 in cumene, aqueous feed of 0.1 g/L Cr(VI) in 1 M HCl, and receiving phases of 5 g/L hydrazine sulfate. Results obtained show that upon the CYANEX 923 concentration range studied (0.125-0.75 M), no significant change in the permeation coefficient is revealed. This constant permeability value, Plim, known as limiting permeability, can be explained by assuming that diffusion in the organic membrane is negligible as compared with that for aqueous diffusion and that the permeation process is controlled by the diffusion in the stagnant film of the aqueous feed phase. Thus (25)
Daq Plim ) daq
(4)
and assuming a value of Daq) 10-5 cm2/s (30) and Plim ) 1.8 × 10-2 cm/s, the value of daq is 5.6 × 10-4 cm, this value being the minimum thickness of the stagnant aqueous diffusion layer within the present experimental conditions. The mass transfer coefficient in the feed phase is found to be 1.8 × 10-2 cm/s. Influence of the Diluent of the Membrane Phase. In many liquid-liquid extraction systems, the characteristics and type of the diluent of the organic phase may influence metal extraction; thus, it is also recognized that the membrane diluent chosen as a water-resistant barrier in any liquid membrane process exerts a major influence on membrane performance (31, 32) [e.g., enabling high membrane lifetimes to be obtained (33)]. The membrane diluent must fulfill the following requeriments: low mutual solubility between organic and aqueous phases, low volatility, surface tension lower than the critical surface tension, and high viscosity. It was concluded (34) that a diluent exhibiting a high tendency to solubilize water and a low organic/water interfacial area are the main cause of SLM instability. In the present work, the use of different diluents for the CYANEX 923/Durapore permeation system was investigated
for experiments carried out with feed phases of 0.02 g/L Cr(VI) in 1 M HCl, membrane phases of 0.50 M CYANEX 923 in each diluent and receiving phases of 5 g/L hydrazine sulfate. The results obtained show that for the present permeation system the change of the diluent influences chromium permeation by CYANEX 923, cumene (P ) 1.1 × 10-2 cm/s), and xylene (P ) 1.0 × 10-2 cm/s) being the diluents in which the highest permeation coefficients were obtained (toluene, P ) 7.5 × 10-3 cm/s and Solvesso 100 P ) 9.5 × 10-3 cm/s). Influence of Metal Concentration on Permeability of Chromium(VI). In Figure 1 it has been schematically shown how chromium(VI) diffuses through the membrane. The mass transfer of Cr(VI) crossing the membrane is described considering only diffusional parameters. The interfacial flux due to the chemical reaction has been neglected as the chemical reactions seems to take place at the aqueous feed solution-membrane and membrane-receiving phase interfaces, and previous studies suggest that chemical reactions can be considered to occur instantaneously relative to the diffusion process (35). Therefore, the Cr(VI) transport rate is determined by the rate of diffusion of chromium-containing species through the feed diffusion layer and the rate of chromium-CYANEX 923 (Cr(VI)-L in Figure 1) species through the membrane. Then, the flux of Cr(VI) crossing the membrane may be derived by applying Fick’s first diffusion law to the diffusion layer on the feed side and to the membrane (36). In the model, the diffusional fluxes at the feed aqueous boundary layer Jf and in the membrane phase Jm, represented by two equations, are functions of the respective diffusion coefficients of the species in each phase. Also, as the distribution coefficient of chromium between the membrane phase and the receiving phase is much lower than that between the feed phase and the membrane, the concentration of the metal-extracted complexes in the membrane phase on the receiving solution side may be considered negligible as compared with that on the feed solution side. If the chemical reaction expressed by the corresponding extraction reaction is assumed to be fast as compared with the diffusion rate, local equilibrium at the interface is reached. Thus, at steady state, Jf ) Jm ) J, and by combination of the various equations a final expression for the model is obtained (37, 38):
J ) P[Cr(VI)]0
(5)
where J is the initial flux, P is the permeability coefficient (calculated from eq 1), and [Cr(VI)]0 is the initial chromium concentration in the feed phase. Figure 4 shows values of the initial chromium flux J for experiments carried out with concentrations of chromium ranging from 0.01 to 0.08 g/L in the feed phase solution. It is shown that, under the experimental conditions, the initial flux is a strong function of the initial metal concentration in the feed phase, increasing its value when the metal concentration in the phase increases; it is also observed that experimental data fit reasonable well with the derived model for the present transport system. Behavior of the CYANEX 923 Carrier System As Compared to Other Potential Cr(VI) Carriers. The transport of chromium(VI) was also studied using other carriers to compare with results obtained using CYANEX 923. The carriers investigated were TBP (phosphoric ester), DBBP (phosphonic ester), and Aliquat 336 (quaternary ammonium salt). Results are summarized in Table 2, showing the best and comparable chromium permeabilities when CYANEX 923 and Aliquat 336 are used as carriers. In the case of the organophosphorus derivatives, the metal transport follows the apparent order: CYANEX 923 > DBBP > TBP, which can be interpreted in terms of the electron-donor properties of VOL. 37, NO. 5, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1045
FIGURE 4. Influence of initial metal concentration on chromium flux. Aqueous feed: Cr(VI) in 1 M HCl. Organic phase: 0.50 M CYANEX 923 in cumene. Receiving phase: 5 g/L hydrazine sulfate.
TABLE 2. Chromium Transport Using Various Carrier Systemsa carrier
P (cm/s)
CYANEX 923 ALIQUAT 336 TBP DBBP
1.8 × 10-2 1.7 × 10-2 2.1 × 10-3 2.5 × 10-3
a Aqueous feed: 0.01 g/L Cr(VI) in 1 M HCl. Organic phase: Carrier (0.50 M) in cumene. Receiving phase: 5 g/L hydrazine sulfate.
FIGURE 5. Chromium flux as a function of cycle number. Organic diluent: cumene. Receiving phase: 5 g/L hydrazine sulfate. Lifetime of the Supported Liquid Membrane. Figure 5 shows the results obtained in the experimental evaluation of the stability of the supported liquid membrane system used in this investigation. The results shown in the figure were obtained within five cycles run of Cr(VI) transport from feed solution containing 0.01 g/L Cr(VI) in 1 M HCl and through the membrane containing 0.50 M CYANEX 923 solution. As seen, the flux through the membrane under study remains practically constant within all transport cycles, which show the stability of the membrane system and indicate its possibilities in long-term removal experiments.
Acknowledgments TABLE 3. Chromium Transport in the Presence of Various Metalsa metal
Pmetal (cm/s)
SF
Cr(VI) Fe(III) Zn(II) Ni(II) Cu(II)
5.8 × 10-3 no transport 1.4 × 10-3 no transport no transport
quantitative 4.3 quantitative quantitative
We thank the CSIC for support. We also thank Mr. Bascones and Mr. Lo´pez for assistance in the technical and analytical work.
Nomenclature A
effective membrane area
[Cr]0
initial chromium concentration in the feed phase
[Cr]t
chromium concentration in the feed phase at an elapsed time
Daq
average aqueous diffusion coefficient of the chromium-containing species
daq
thickness of the aqueous feed boundary layer
J
chromium flux
Jf
diffusional flux at the aqueous feed boundary layer
Jm
diffusional flux in the membrane phase
a
Aqueous feed: 0.01 g/L of each metal in 0.5 M HCl. Organic phase: carrier (0.50 M) in cumene. Receiving phase: 5 g/L hydrazine sulfate.
the active substance of the carrier: higher in the case of the phosphine oxide and lower in the case of the phosphoric ester (TBP). Separation of Cr(VI) from Fe(III), Cu(II), Ni(II), and Zn(II). To examine the effect of several other metal ions generally accompanying Cr(VI), a systematic investigation was made of their influence on the overall permeation of Cr(VI). Thus, Fe(III), Cu(II), Ni(II), and Zn(II) were tested in the form of a mixture with chromium(VI). Table 3 shows the results obtained in these transport studies in which the separation factor of each base metal is defined as
SF )
PCr Pmetal
(6)
A selective transport of chromium(VI) is obtained within the experiments that allow recovery of pure Cr(III) from the receiving solution. This selectivity range can be correlate to the extraction properties of CYANEX 923 for each of the metal ions, with Cr(VI) extracted preferably to iron(III) (19). Zinc is also extracted at this hydrochloric acid concentrations range (39), but neither nickel(II) nor copper(II) are extracted by the reagent. The efficiency of chromium removal/ separation process can be easily improved using a more effective membrane unit (e.g., hollow fiber modules). 1046
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 5, 2003
P
permeability
SF
metal separation factor
V
volume of the feed phase
Literature Cited (1) Wiesner, M. R.; Chellam, S. Environ. Sci. Technol. 1999, 33, 360A. (2) Nriagu, J. O. In Chromium in the Natural and Human Environments; Nriagu, J. O., Nieboer, E., Eds.; John Wiley & Sons: New York, 1988; p 81. (3) Scott, K. Handbook of Industrial Membranes; Elsevier: Kidlington, 1997; p 691. (4) Alonso, A. I.; Urtiaga, A.; Irabien, A.; Ortı´z, I. Chem. Eng. Sci. 1994, 49, 901. (5) Guha, A. K.; Yun, C. H.; Basu, R.; Sirkar, K. AIChE J. 1994, 40, 1223.
(6) Zouhri, A.; Burgard, M.; Lakkis, D. Hydrometallurgy 1995, 38, 299. (7) Alonso, A. I.; Pentelides, C. C. J. Membr. Sci. 1996, 110, 151. (8) Ortı´z, I.; Gala´n, B.; Irabien, A. J. Membr. Sci. 1996, 118, 213. (9) Sastre, A. M.; Kumar, A.; Shukla, J. P.; Singh, R. K. Sep. Purif. Methods 1998, 27, 213. (10) Ortı´z, I.; Gala´n, B.; San Roma´n, F.; Urtiaga, A. M. In Proceedings of the Global Symposium on Recycling (Rewas’99); Gaballah, I., Hager, J., Solozabal, R., Eds.; TMS-Inasmet: Warrendale, PA, 1999; Vol. III, p 2173. (11) Alguacil, F. J.; Coedo, A. G.; Dorado, M. T. Hydrometallurgy 2000, 57, 51. (12) Youn, I. J.; Harrington, P. J.; Stevens, G. W. Solvent Extr. Ion Exch. 2000, 18, 933. (13) Vicent, T.; Guibal, E. Solvent Extr. Ion Exch. 2000, 18, 1241. (14) Alguacil, F. J.; Coedo, A. G.; Dorado, M. T.; Sastre, A. M. Hydrometallurgy 2001, 61, 13. (15) Alguacil, F. J.; Caravaca, C.; Martı´nez, S.; Cobo, A. Hydrometallurgy 1994, 36, 369. (16) Martı´nez, S.; Sastre, A.; Miralles, N.; Alguacil, F. J. Hydrometallurgy 1996, 40, 77. (17) Reddy, M. L. P.; Luxmi Varma, R.; Ramamohan, T. R.; SahuSushantha, K.; Chakravortty, V. Solvent Extr. Ion Exch. 1998, 16, 795. (18) El Amnouri, E.; Distin, P. A. Solvent Extr. Ion Exch. 1996, 14, 871. (19) Saji, J.; Prasada Rao, T.; Iyer, C. S. P.; Reddy, M. L. P. Hydrometallurgy 1998, 49, 289. (20) Rickelton, W. A. Solvent Extr. Ion Exch. 1999, 17, 1507. (21) Alguacil, F. J.; Lo´pez, F. A. Hydrometallurgy 1996, 42, 245. (22) Huang, T.-C.; Huang, C.-C.; Chen, D. H. Solvent Extr. Ion Exch. 1997, 15, 837. (23) Dziwinski, E.; Szymanowski, J. Solvent Extr. Ion Exch. 1998, 16, 1515.
(24) Danesi, P. R. J. Membr. Sci. 1984, 20, 231. (25) Rovira, M.; Sastre, A. M. J. Membr. Sci. 1998, 149, 241. (26) Kedari, C. C.; Pandit, S. S.; Misra, S. K.; Ramanujam, A. Hydrometallurgy 2000, 62, 47. (27) Puigdomenech, I. Medusa Program, Royal Institute of Technology, Stockholm, Sweden, 2002. (28) Sastre, A.; Madi, A.; Cortina, J. L.; Miralles, N. J. Membr. Sci. 1998, 139, 57. (29) Kedari, C. C.; Pandit, S. S.; Ramanujam, A. J. Membr. Sci. 1999, 156, 187. (30) Bermejo, J. C.; Alonso, M.; Sastre, A. M.; Alguacil, F. J. J. Chem. Res., Suppl. 2000, 9, 479. (31) Elhassadi, A. A.; Do, D. D. Sep. Sci. Technol. 1986, 21, 267. (32) Molinari, R.; Drioli, E.; Pantano, G. Sep. Sci. Technol. 1989, 24, 1015. (33) Dozol, J, F.; Casas, J.; Sastre, A. M. Sep. Sci. Technol. 1993, 28, 2007. (34) Danesi, P. R. Sep. Sci. Technol. 1984, 19, 857. (35) Zuo, G.; Orechho, S.; Muhammed, M. Sep. Sci. Technol. 1996, 31, 1597. (36) Danesi, P. R.; Horwitz, E. P.; Vandergrift, G. F.; Chiariza, R. Sep. Sci. Technol. 1981, 16, 201. (37) Rossel, A.; Palet, C.; Valiente, M. Anal. Chim. Acta 1997, 349, 171. (38) Shukla, J. P.; Sonawane, J. V.; Kumar, A.; Singh, R. K. Indian J. Chem. Technol. 1996, 3, 145. (39) Alguacil, F. J.; Martı´nez, S. J. Chem. Technol. Biotechnol. 2001, 76, 298.
Received for review February 11, 2002. Revised manuscript received December 2, 2002. Accepted December 6, 2002. ES020585S
VOL. 37, NO. 5, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1047
