Removal of Cadmium and Zinc from Aqueous Solutions Using Red Mud

Red mud, an aluminum industry waste, has been converted into an inexpensive and efficient adsorbent. The product obtained has been characterized and ...
0 downloads 0 Views 186KB Size
Download PDF
Environ. Sci. Technol. 2002, 36, 3612-3617

Removal of Cadmium and Zinc from Aqueous Solutions Using Red Mud

of toxic metals from aqueous solutions. The results obtained for the removal of Cd2+ and Zn2+ are described in the current paper.

Materials and Methods VINOD K. GUPTA* AND SAURABH SHARMA Department of Chemistry, Indian Institute of Technology Roorkee, Roorkee-247 667 (Uttranchal), India

Red mud, an aluminum industry waste, has been converted into an inexpensive and efficient adsorbent. The product obtained has been characterized and utilized in batch and column operations for the removal of cadmium and zinc from aqueous solutions over a wide range of initial metal ion concentrations (1.78 × 10-5 to 1.78 × 10-3 M for Cd2+ and 3.06 × 10-5 to 3.06 × 10-3 M for Zn2+; contact time, 24 h) adsorbent dose (5-20 g/L), and pH (1.0-6.0). The removal of Cd2+ and Zn2+ was almost complete at low concentrations, while it was 60-65% at higher concentrations at optimum pH’s of 4.0 and 5.0, respectively, with 10 g/L of adsorbent in an 8-10 h equilibration time. The adsorption decreased with increase in temperature. Kinetic studies have been used to describe the mechanism of adsorption. Chemical regeneration of the columns has been achieved with 1% HNO3.

Introduction Environmental pollution by toxic metals is well-recognized and can be detrimental to living systems. Metals can be toxic pollutants that are nonbiodegradable, undergo transformations, and have great environmental, public health, and economic impacts (1, 2). Cadmium and zinc are toxic above permissible limits and can be introduced into water bodies from various sources (3, 4). The main techniques utilized to remove heavy metal ions from aqueous streams include ionexchange chromatography, reverse-osmosis, precipitation, and adsorption (5). However, in many situations, these processes do not work efficiently. For example, heavy metal precipitation produces sludges that must be treated and disposed of, normally at high cost (1). Similarly, the use of commercially available activated carbon is limited, especially in developing countries, because of its relatively high cost and the difficulties associated with its regeneration (6). As a result, recent research has focused on the development of low-cost carbon alternatives using various industrial wastes. Contributions in this area have been made by many researchers who have utilized a number of materials including, fly ash (7, 8), metal hydroxides (9), clay (10), soil (11), apatite (12, 13), bentonite (14, 15), lignin (16), charcoal (17), blast furnace slag (18, 19), chitosan (20), biomass (21, 22), peanut hull (23), bagasse pith (24), carbonaceous material (25), and bagasse fly ash (26). Aluminum industries generate red mud as a byproduct (27), which is currently being used in the manufacture of building materials and ceramics (28). We have utilized red mud, after converting it into an adsorbent, for the removal * Corresponding author phone: 0091-1332-85801 (work); fax: 00911332-73560; e-mail: [email protected], [email protected]. 3612

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 16, 2002

All of the reagents used were at least A.R. grade. Stock solutions were prepared by dissolving CdSO4 and ZnSO4 in double-distilled water. (a) Equipment. Atomic absorption spectra were recorded using an atomic absorption/emission spectrophotometer (PerkinElmer model 3100, Shelton, CT). X-ray measurements were performed with a Phillips X-ray diffractometer employing nickel-filtered Cu KR radiations. Surface areas were measured with a surface-area analyzer (model QS-7; Quantasorb surface area analyzer). pH measurements were made using a pH meter (model CT No. CL 46; Toshniwal, India). Scanning electron microscopy was performed using a Phillips SEM 501 electron microscope. The density of the adsorbent was determined using specific gravity bottles. (b) Material Development. Red mud obtained from Hindustan Aluminum Company (HINDALCO) (Renukoot, India) was in the form of clay-type waste residue composed of a fine fraction (mud) and a relatively coarse fraction (sand) with small granules. This material in the crude form showed poor adsorption properties. The material was, therefore, first treated with hydrogen peroxide at room temperature for 24 h to oxidize adhering organic matter and washed repeatedly with double-distilled water. The resulting material was dried at 100 °C, cooled, and again activated in air in a muffle furnace at 500 °C for 3 h. The final product exhibited the best adsorption capacity and optimum surface area. The product obtained at temperatures higher than 500 °C exhibited poor adsorption capacity, probably due to the collapse of surface functional groups on adsorbent. Conditions for activation were carefully controlled to obtain a product with batchto-batch reproducibility. The activated red mud was crushed into smaller particles and sieved to the desired particle size: 100-150, 150-200, and 200-250 B.S.S. mesh. The studies were carried out with the red mud of particle diameter 150200 B.S.S. mesh (0.089 mm). Finally, the product was stored in a vacuum desiccator until used. (c) Adsorption Studies. Experiments were conducted with metal ion solutions of varying concentration (1.78 × 10-5 to 1.78 × 10-3 M for Cd2+ and 3.06 × 10-5 to 3.06 × 10-3 M for Zn2+), maintained at the desired pH and temperature. Stoppered glass tubes containing metal ion solutions (10 mL) and known amounts of red mud (0.1 g, particle size 150-200 mesh) were stirred intermittently for a maximum period of 24 h. Triplicate runs were made for each sample to determine the reproducibility and relative deviation of the experiments. Preliminary studies showed that equilibrium was attained in 8-10 h; beyond this time, the adsorption of metal ions on the adsorbent material remained almost constant with the reproducibility and relative deviation of the order of (0.5% and (1.5%, respectively. It is pertinent to mention that the error bars for the figures were so small as to be smaller than the symbols used to plot the graphs. Adsorption studies were carried out at 30, 40, and 50 °C to determine the effect of temperature. The effect of pH was determined by studying the adsorption of metal ions at a fixed concentration over a pH range of 1.0-6.0. pH adjustment was made by dilute NaOH and dilute HCl. The pH was monitored at the beginning and at the end of the each experiment and no noticeable change in pH was observed. 10.1021/es020010v CCC: $22.00

 2002 American Chemical Society Published on Web 07/16/2002

(d) Kinetic Studies. For kinetic investigations, a batch technique was selected because of their simplicity. A number of stoppered glass tubes (50-mL capacity), containing 10 mL of solutions of metal ions of known concentration (8.89 × 10-4 M for Cd2+ and 1.84 × 10-3 M for Zn2+) were placed in an NSW-133 water-bath shaker. When the desired temperature was reached, 0.1 g of red mud was added into each tube, and the solutions were agitated by mechanical shaking. At certain times (30 min for the first two tubes and 60 min for all subsequent tubes) tubes were centrifuged to collect the adsorbent material. The supernatant was analyzed for metal content. It was determined that equilibrium was attained in 8-10 h. (e) Column Studies. A glass column (40 × 0.5 cm) was filled with red mud (0.5 g; B.S.S. mesh size 200-250) on a glass-wool support. The columns were loaded with solutions of Cd2+ (8.89 × 10-4 M) and Zn2+ (1.84 × 10-3 M) at pH 4.0 and 5.0, respectively. Effluent flow (0.5 mL/min) was adjusted with the help of a stopcock at the bottom of the column.

Results and Discussion Characterization of the Adsorbent Material. A 1.0-g sample of activated red mud was stirred with deionized water (100 mL, pH 6.8) for 2 h and left for 24 h in an airtight, stoppered, conical flask. An increase in pH to 7.5 was noticed. Activated red mud was found stable (did not dissolve, degrade, or change) in water, salt solutions, dilute acids, dilute bases, and organic solvents in the temperature range of 30-50 °C and pH range of 1.0-6.0. The composition of red mud as determined by chemical analysis (29) was Fe2O3, 38.80%; TiO2, 18.80%; SiO2, 9.64%; Al2O3, 17.28%; and Na2O, 6.86 wt %. The loss on ignition was found to be 7.34 wt %. The density and porosity were found to be 2.0 g/cm3 and 0.45% fraction, respectively. The d spacing values from X-ray diffraction suggested the presence of hematite, cancrinite, goethite, rutile, anataze, and quartz. The surface area of the adsorbent as calculated by the Brunauer-Emmett-Teller method was 108 m2/g. Scanning electron micrographs of activated red mud (Figure 1a,b) clearly reveal the surface texture and porosity of the material with a texture like aluminum silicates, distributed with heavy constituents such as iron. Adsorption Studies. The effect of pH on the removal of cadmium and zinc is depicted in Figure 2. The experiments were run in triplicate. The deviation in the results was of the order of (0.2% with a reproducibility of (0.5%. It can be seen from Figure 2 that the adsorption of cadmium and zinc increases with pH and reaches a maximum at pH 4.0 for Cd2+ and between pH 4.0-6.0 for Zn2+. The effect of pH can be explained by considering the surface charge on the adsorbent material. Among the various constituents of red mud, the ZPC (zero point charge) of silica is ∼2.3 and that of Fe2O3 is ∼8.6. The composite ZPC of the adsorbent was found to be 3.2. Thus, below a pH of 3.2 the surface has a high positive charge density and under these conditions the uptake of metal ions would be quite low due to electrostatic repulsion. With increasing pH (beyond ZPC), the negative charge density on the surface of adsorbent increases, thereby resulting in an enhancement of metal adsorption. The isotherm and other kinetic studies were carried out at the optimum pH of 4.0 for cadmium, and a fixed pH of 5.0 was chosen for zinc. The adsorption of cadmium and zinc decreases with a rise in temperature, indicating the process to be exothermic in nature. Experiments at lower concentrations (1.78 × 10-5 to 1.78 × 10-4 M for Cd2+ and 3.06 × 10-5 to 3.06 × 10-4 M for Zn2+) demonstrated the complete removal of both the metal ions (Figure 3a,b). The adsorption isotherms at higher concentrations are regular, positive, and concave to the concentration axis. The plot for Cd2+ adsorption (Figure 3a)

FIGURE 1. SEM micrographs of activated red mud at different magnifications, (a) 5000x (b) 80x.

FIGURE 2. Effect of pH on the adsorption of metal ions on activated red mud. shows that cadmium is nearly 100% adsorbed at lower concentrations (e3.56 × 10-4 M). Further, the adsorption of VOL. 36, NO. 16, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3613

TABLE 2. Thermodynamic Parameters for the Adsorption of Cd2+ and Zn2+ on Red Mud metal ion

-∆G° (kJ/mol) ((0.1) 30 °C 40 °C 50 °C

Cd2+ Zn2+

22.8 23.9

20.7 20.6

18.7 19.7

-∆H° (kJ/mol) ((0.1)

-∆S° (kJ/mol) ((0.02)

85.9 78.5

0.21 0.18

Chakravorti (30) expressed the feasibility of the process in terms of a dimensionless separation factor RL (eq 1)

RL )

FIGURE 3. Effect of temperature on the adsorption of (a) Cd2+ and (b) Zn2+ on activated red mud.

TABLE 1. Freundlich and Langmuir Constants of Cd2+ and Zn2+ on Red Mud Q0 × 104 b × 10-3 (mol/g) ((0.05) (L/mol) ((0.05) metal slope 1/n KF × 104 ion ((0.02) mol/g ((0.01) 30 °C 40 °C 50 °C 30 °C 40 °C 50 °C Cd2+ Zn2+

0.58 0.51

0.15 0.23

1.16 1.06 1.00 8.55 2.88 1.04 2.22 2.00 1.81 1.07 2.75 1.54

cadmium decreases with concentration (g8.89 × 10-4 M) and ultimately becomes constant. Similarly, the adsorption of zinc to red mud is nearly 100% up to concentrations e9.18 × 10-4 M (Figure 3b) and decreases with concentration up to g1.84 × 10-3 M. Almost complete removal at low concentrations of both metal ions and the steep increase in the early stages of the isotherms indicates a faster initial removal. Three parallel runs at lower concentrations gave standard deviations of (1.15% and (0.8% for complete removal of cadmium and zinc, respectively. These observations show the efficacy of red mud for the removal of cadmium and zinc from aqueous solution. Adsorption data for cadmium and zinc correlate well with the Freundlich (qe ) KFC1/n) and Langmuir (qe ) (Q°bC)/(1 + bC)) adsorption models (plots are not shown). Freundlich and Langmuir parameters obtained for cadmium and zinc are given in Table 1. The adsorption capacity KF is lower for the cadmium-red mud system than that for the zinc-red mud system. The values of slope (1/n), which reflect the intensity of adsorption, present the same trend. The value of Q° (maximum uptake) decreases with increases in temperature, thereby indicating the exothermic nature of the process. The values of Q° appear to be significantly higher for the zinc-red mud system in comparison to the cadmium-red mud system. Weber and 3614

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 16, 2002

1 1 + bC0

(1)

(b is the Langmuir constant and C0 is the initial concentration of the adsorbate) which was found to be 0.15 and 0.05, respectively, for cadmium and zinc. This indicates highly favorable adsorption (RL ,1) of cadmium and zinc on red mud. The thermodynamic parameters (18) for these systems are given in Table 2. The negative free energy values indicate the feasibility of the process and its spontaneous nature. The negative value of enthalpy change (∆H°) for the processes confirms their exothermic nature. Negative entropy of adsorption (∆S°) reflects the affinity of the material toward cadmium and zinc. Kinetic Studies. The experimental runs measuring adsorption with time revealed that the adsorption of Cd2+ and Zn2+ on red mud at pH 4.0 and 5.0, respectively, was quite fast. Typically, 35-45% of the adsorption capacity was realized within the first hour of contact. This initial rapid adsorption gives way to a very slow approach to equilibrium (reached in 8-10 h). It is pertinent to mention that the pH of the solution does not change during the adsorption process. The effect of particle size and the concentration of adsorbate and adsorbent were also studied, revealing that 10 g/L of the adsorbent (particle size 150-200 B.S.S. mesh) is sufficient for the removal of both Cd2+ and Zn2+ ions at 3.56 × 10-4 M and 9.18 × 10-4 M, respectively. Adsorption Dynamics. Rate Constant of Adsorption. Lagergren’s (31) rate equation (eq 2) was employed for studying the rate constant for the systems, where qe and q are amounts adsorbed at equilibrium and at time t, respectively, and Kad is the first-order rate constant

log(qe - q) ) log qe -

Kad t 2.303

(2)

The plots of log(qe - q) versus t were found to be linear for both cadmium and zinc, showing the first-order nature of the processes. The values of the rate constant, Kad, for each system as calculated from the respective plots were 3.47 × 10-3 and 2.14 × 10-3 min-1 for Cd2+ and Zn2+, respectively. Mass Transfer Study. The mass transfer model proposed by McKay et al. (32) have been employed for the determination of the surface mass-transfer coefficient βL for the adsorption of Cd2+ and Zn2+ on activated red mud. The straight line plots observed between ln[Ct/C0 - 1/(1 + mk)] and t show the validity of the model in the present study. The values of βL for different systems as determined from the plots were 4.52 × 10-4 and 4.20 × 10-4 cm/s for Cd2+ and Zn2+, respectively. These values indicate that the velocity of mass transfer of cadmium and zinc ions on red mud is quite high. Further, Bt values were obtained from Reichenberg’s table (33) and Bt versus t plots (Figure 4a,b) are used to distinguish between particle-diffusion and film-diffusion controlled

FIGURE 5. log Di versus 1/T plots of metal ions for activated red mud.

FIGURE 6. Breakthrough curves of metal ions for activated red mud.

FIGURE 4. Bt vs time plots of (a) Cd2+ and (b) Zn2+ for activated red mud. processes. The values of Di (effective diffusion coefficient) were calculated using the reported procedure (34). Adsorption kinetics of Cd2+ and Zn2+ on the red mud presents some interesting features. The Bt versus time plot (Figure 4a) for cadmium at e1.78 × 10-4 M does not pass through the origin, indicating that the rate-controlling process may be film diffusion. This is also supported by the fact that the diffusion coefficient is dependent on the concentration of the adsorbate. Similar plots at higher concentrations g4.44 × 10-4 M are curved. The same curves can, however, be resolved into two linear parts with different slopes, thereby, indicating a change in mechanism. The initial portion passes through the origin and has a smaller slope, thus reflecting that the effective diffusion coefficient (Di) is small (3.68 × 10-12 m2/s) and that in this region the particle-diffusion mechanism seems to be the rate-limiting step. In the later portion, the slope increases (the plot did not pass through origin), and consequently, the diffusion coefficient (Di) increases (8.65 × 10-12 m2/s). At this stage, the process must not be purely film-diffusion controlled and other factors such

as aggregation and electrokinetic interactions may also become effective. The Bt versus time plots (Figure 4b) for Zn2+ are linear at lower concentrations (e9.18 × 10-4 M) and do not pass through origin, thereby indicating the process to be film-diffusion controlled. However, at higher concentrations (g1.84 × 10-3 M), the process was initially filmdiffusion controlled (Di ) 1.57 × 10-12 m2/s), and during the later stage, the particle-diffusion mechanism also became operative, which was apparent from the deviation of the plots from linearity and with an increased effective diffusion coefficient (Di ) 5.30 × 10-12 m2/s). Figure 5 depicts the change in the rate of adsorption with temperature. Diffusion coefficient values (Di) for the cadmium-red mud and zinc-red mud systems at different temperatures are given in Table 3. It was found that the Divalues of both the systems decreased with an increase in temperature. Further, the log Di versus 1/T plot for both the metal ions are linear in nature. This permits the use of the Arrhenius-type equation (18) for the evaluation of preexponential factor D0 and energy of activation Ea from the intercepts and the slopes of the linear plots. D0 values are further used to calculate the entropy of the adsorption of cadmium and zinc on red mud. The values of these parameters are also listed in Table 3. VOL. 36, NO. 16, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3615

TABLE 3. Effective Diffusion Coefficients Di, D0, Ea, and ∆Sq Values for Cd2+ and Zn2+ at Different Temperatures Di × 1012 (m2/s) ((0.1)

metal D0 × 1013 Ea (kJ/mol) -∆S q ions 30 °C 40 °C 50 °C (m2/s) ((0.1) ((0.5) (J/(K‚mol))((0.5) Cd2+ Zn2+

1.3 1.6

1.0 1.4

0.9 1.1

1.1 2.0

13.8 16.2

146.1 140.7

After the metal ions (Cd2+ and Zn2+) were recovered, the spent column was washed with 50 mL of water in 10 mL increments, at a flow rate of 0.5 mL/min. It was again loaded with cadmium and zinc. The breakthrough capacities of cadmium-red mud system were found to be 37.9, 31.9, 30.2, 29.3, and 28.7 mg/g and for zinc-red mud system were 34.3, 27.0, 25.1, 24.1, and 23.8 mg/g in the subsequent regeneration cycles. The percent regeneration (RE%) as calculated using eq 4 was found to be 84.2%, 80.0%, 77.3%, and 75.8% for Cd2+ and 78.7%, 73.1%, 70.3%, and 69.3% for Zn2+ in subsequent cycles.

RE% )

Ar × 100 A0

(4)

(where Ar is the adsorptive capacity of the regenerated red mud and A0 is the original capacity of the virgin red mud). It was further observed that, after elution, if the column was again treated with 25-30 mL of 1% HNO3 solution (pH 5.5) at a flow rate of 0.5 mL/min and was washed with 100 mL of hot distilled water, it regained the sorption capacity of columns of virgin red mud.

Acknowledgments The authors are thankful to the Council of Scientific and Industrial Research (CSIR), New Delhi, India, for supporting the work.

FIGURE 7. Desorption curves of metal ions with 1% HNO3 for activated red mud. Column Studies. The approach of Weber (35) has been adopted for the design of fixed bed adsorbers for the removal of Cd2+ and Zn2+. Breakthrough curves (Figure 6) were used to calculate the column capacity. Breakthrough capacity Q0.5 (at 50% breakthrough) and at exhaustion point (Qe) expressed in mg of metal ion/g of red mud (24) was calculated using eq 3 [(Vmb is the volume at breakthrough (mL) and W is the weight of the adsorbent (g)].

CeVmb Q0.5 ) W

(3)

The column capacities at 50% breakthrough (Ce/Ci ) 0.5 where Ce and Ci are concentrations of the effluent and influent, respectively) were found to be 28.7 and 25.6 mg/g for Cd2+ and Zn2+, respectively. However, at the complete exhaustion point (Qe), the breakthrough capacities were 37.9 and 34.3 mg/g, respectively. Regeneration. The recovery of the adsorbed material, as well as regeneration of adsorbent, is quite important. Carbon columns are generally subjected to thermal regeneration where 5-10% of adsorbent is usually lost by attrition during each cycle and the recovery of adsorbate is also not possible. Elution of adsorbate with simultaneous chemical regeneration by a suitable solvent is a definite alternative to thermal regeneration and has been tried in the present studies by using mineral acids, salt solutions, and bases. The elution of cadmium and zinc was investigated under identical conditions of flow rate (0.5 mL/min), column length (40 cm), and column diameter (1 cm). The studies revealed that 1% HNO3 (60 mL) provided almost complete elution of cadmium (Figure 7). The first aliquot of 30 mL eluted 60.1% of the adsorbed cadmium, and the rest was desorbed in three increments (each of 10 mL). The recovery of the cadmium was 85%. In case of zinc-red mud system, 70 mL of 1% HNO3 was used (Figure 7). The first 20 mL elutes about 45% of the zinc metal ion and, with the other 50 mL in five increments, provided 96.3% desorption. 3616

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 16, 2002

Notation Ar

adsorption capacity of regenerated adsorbent (mg/g)

A0

original adsorption capacity of the virgin adsorbent (mg/g)

B.S.S.

British standard size

B

time constant

D0

pre-exponential factor (m2/s)

Di

effective diffusion coefficient (m2/s)

Ea

energy of activation (kJ/mol)

F

fractional attainment

∆G°

Gibbs free energy (kJ/mol)

∆H°

enthalpy change of the process (kJ/mol)

KF

Freundlich constant of solute (mol/g)

Kad

rate constant (L/min)

Q∞

amount of adsorbate adsorbed per gram of adsorbent (mol/g) at equilibrium

Qt

amount of adsorbate adsorbed per gram of adsorbent (mol/g) at time t

Q°

Langmuir monolayer capacity (mol/g)

Q0.5

column capacity at 50% breakthrough (C0/Ci ) 0.5) (mg/g)

Qe

total column capacity at exhaustion point (mg/ g)

RL

separation factor (dimensionless)

RE%

regeneration efficiency

∆S°

entropy of adsorption (kJ/(K‚mol))

∆Sq

entropy of activation (J/(K‚mol))

Vmb

volume of breakthrough (mL)

ZPC

zero point charge

n

Freundlich constant of solute

βL

mass transfer coefficient (cm/s)

Literature Cited (1) Brower, J. B.; Ryan, R. L.; Pazirandeh, M. Environ. Sci. Technol. 1997, 31, 2910-2914. (2) Cowan, C. E.; Zachara, J. M.; Resch, C. T. Environ. Sci. Technol. 1991, 25, 2213-2217. (3) Low, K. S.; Lee, C. K. Bioresour. Technol. 1991, 38, 1-6. (4) Gupta, V. K.; Rastogi, A.; Diwivedi, M. K.; Mohan, D. Sep. Sci. Technol. 1997, 32, 2883-2912. (5) Dean, J. G.; Bosqui, F. L.; Lanouette, K. H. Environ. Sci. Technol. 1972, 6, 518-524. (6) Lalvani, S. B.; Wiltowski, T.; Hubner, A.; Weston, A.; Mandich, N. Carbon 1998, 36, 1219-1226. (7) Diamadopoulos, E.; Loannidis, S.; Sakellaropoulos, G. P. Water Res. 1993, 17, 1773-1777. (8) Weng, C. H.; Huang, C. P. J. Environ. Eng. 1994, 120, 14701487. (9) Namasivayam, C.; Ranganathan, K. Water Res. 1998, 32, 969971. (10) Gagnon, C.; Arnac, M.; Brindle, J. R. Water Res. 1992, 26, 10671072. (11) Yin, Y.; Allen, H. E.; Huang, C. P.; Sparks, D. L.; Sanders, P. F. Environ. Sci. Technol. 1997, 31, 496-503. (12) Mandjiny, S.; Zouboulis, A. I.; Matis, K. A. Sep. Sci. Technol. 1995, 30, 2963-2978. (13) Chen, X.; Wright, J. V.; Conca, J. L.; Peurrung, L. M. Environ. Sci. Technol. 1997, 31, 624-631. (14) Pardes, E. G.; Sanchez, M. V.; Cruz, F. C.; Viciana, M. S.; Perez, M. F. J. Chem. Technol. Biotechnol. 1994, 59, 289-296. (15) Mellah, A.; Chegrouche, S. Water Res. 1997, 31, 621-629. (16) Srivastava, S. K.; Singh, A. K.; Sharma, A. Environ. Technol. Lett. 1994, 15, 353-361. (17) Mishra, S. P.; Chaudhary, G. R. J. Chem. Technol. Biotechnol. 1994, 59, 359-364.

(18) Gupta, V. K. Ind. Eng. Chem. Res. 1998, 37, 192-202. (19) Dimitrova, S. V.; Mehandgiev, D. R. Water Res. 1998, 32, 32893292. (20) Masri, M. S.; Reuter, F. W.; Friedman, M. J. Appl. Polym. Sci. 1974, 18, 675-681. (21) Chang, J.-S.; Law, R.; Chang, C.-C. Water Res. 1997, 31, 16511658. (22) Schneider, I. A. H.; Rubio, J. Environ. Sci. Technol. 1999, 33, 2213-2217. (23) Periasamy, K.; Namasivayam, C. Ind. Eng. Chem. Res. 1994, 33, 317-320. (24) Aly, H. M.; Daifullah, A. A. M. Adsorpt. Sci. Technol. 1998, 16, 33-38. (25) Srivastava, S. K.; Tyagi, R.; Pal, N.; Mohan, D. J. Environ. Eng. 1997, 123, 842-851. (26) Gupta, V. K.; Mohan, D.; Sharma, S. Sep. Sci. Technol. 1998, 33, 1331-1343. (27) Lopez, E.; Soto, B.; Asias, M.; Nunez, A.; Rubinos, D.; Barral, M. T. Water Res. 1998, 32, 1314-1322. (28) Li, L. Y. J. Environ. Eng. 1998, 124, 254-264. (29) Vogel, A. I. A text book of quantitative chemical analysis, 5th ed.; ELBS. Publication: London, U.K., 1972; p 417. (30) Weber, T. W.; Chakrovorti, R. K. J. Am. Inst. Chem. Eng. 1974, 20, 228-238. (31) Ho, Y. S.; McKay, G. Water Res. 1999, 33, 578-584. (32) McKay, G.; Allen, S. J.; McConvey, I. F.; Otterburn, M. S. J. Colloid Interface Sci. 1981, 80, 323-339. (33) Reichenberg, D. J. Am. Chem. Soc. 1953, 75, 589-597. (34) Boyd, G. E.; Adamson, A. W.; Meyers, L. S. J. Am. Chem. Soc. 1947, 69, 2836-2848. (35) Weber, W. J., Jr. Physicochemical processes for water quality control; Wiley-Interscience: New York, 1972.

Received for review January 22, 2002. Revised manuscript received May 28, 2002. Accepted May 30, 2002. ES020010V

VOL. 36, NO. 16, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3617