ABSTRACT Adsorptive removal of selenate from aqueous phase using ZrIV-loaded orange waste gel was investigated. Adsorption kinetics showed that equilibrium was attained within 2 and 4 h for 12 and 23 mg/L of SeVI, respectively. A pseudo-second-order kinetic model was found to characterize the adsorption kinetics for all the initial selenium concentrations tested. Adsorption isotherms at various pH values were examined and the data fitted well with Langmuir model. The maximum sorption capacity was found to be 25 mg/g at optimum pH. The adsorption system was found to be favorable with separation factors between 0 and 1. Competitive adsorption by coexisting anions showed that chloride barely affected SeVI adsorption while phosphate and sulfate impeded SeVI adsorption. A ligand exchange mechanism was inferred for SeVI adsorption.
Journal of Water and Environment Technology, Vol. 8, No.4, 2010 Address correspondence to Katsutoshi Inoue, Department of Applied Chemistry, Saga University, Japan, Email: inoue@elechem.chem.saga-u.ac.jp Received April 30, 2010, Accepted August 30, 2010. - 313 - Study of Zr IV -loaded Orange Waste Gel for Selenate Adsorption Biplob Kumar BISWAS * , Katsutoshi INOUE * , Hidetaka KAWAKITA * , Hiroyuki HARADA * , Keisuke OHTO * , Shafiq ALAM ** * Department of Applied Chemistry, Saga University, Honjo 1, Saga 840-8502, Japan ** Department of Engineering and Applied Science, Memorial University, Canada ABSTRACT Adsorptive removal of selenate from aqueous phase using Zr IV -loaded orange waste gel was investigated. Adsorption kinetics showed that equilibrium was attained within 2 and 4 h for 12 and 23 mg/L of Se VI , respectively. A pseudo-second-order kinetic model was found to characterize the adsorption kinetics for all the initial selenium concentrations tested. Adsorption isotherms at various pH values were examined and the data fitted well with Langmuir model. The maximum sorption capacity was found to be 25 mg/g at optimum pH. The adsorption system was found to be favorable with separation factors between 0 and 1. Competitive adsorption by coexisting anions showed that chloride barely affected Se VI adsorption while phosphate and sulfate impeded Se VI adsorption. A ligand exchange mechanism was inferred for Se VI adsorption. Keywords: adsorption, ligand exchange, orange waste, removal, selenate. INTRODUCTION Selenium (Se) is a naturally occurring element, which is one of the inorganic contaminants that have become environmental concerns lately. It is introduced in the environment from different sources, both natural and anthropogenic, such as from the activities related to thermal power stations, combustion of fossil fuels, roasting and refining of sulfide ores. Selenium has an ambivalent behavior ranging from being essential to highly toxic, depending on species, oxidation state and concentration. Selenium deficiency has been reported to be linked to certain endemic diseases in China (Tan and Huang, 1991), whereas selenium enrichment in soil and water has been implied as a major factor resulting in severe teratogenic symptoms in wildlife at Kesterson National Wildlife Refuge, California (Presser and Ohlendorf, 1987). Thus, an important feature of selenium is the very narrow margin between nutritionally optimal and potentially toxic dietary exposures for animals, which necessitates a clear knowledge of the processes affecting selenium distribution in the environment. Selenium can be present in the aqueous environment in various forms such as elemental selenium (Se 0 ), selenite (SeO 3 2- ), selenide (Se 2- ), selenate (SeO 4 2- ) and organic selenium. Hydrogen selenide and selenium oxides have been known to be public health hazards (Lo and Chen, 1997). Selenite is present in reducing environments while selenate in oxidizing environments. Selenite is difficult to be dissolved in water and easily adsorbed by soil colloids. As a result, selenate becomes the major species of selenium in water. Numerous treatment technologies such as ion exchange, reverse osmosis, nanofiltration, solar ponds, chemical reduction with iron and microalgal-bacterial treatment have been reported for selenium removal from contaminated waters. One of the important Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 314 - processes to regulate the concentration of selenium in aqueous system is the adsorption on solid surfaces. Although much research has been devoted to study the effect of reaction time, pH and competitive anions on selenium adsorption on layered double hydroxides (LDH), activated carbon, activated alumina and magnetite, only a few literatures are reported on adsorption using natural waste materials. Therefore, the present study aims to investigate the adsorption behavior of chemically modified orange waste, an available agro-industrial waste material, in aqueous system as a function of various adsorption parameters such as pH, time and initial selenate concentration. MATERIALS AND METHODS Materials Selenate and zirconium solutions were prepared by dissolving sodium selenate decahydrate (Na 2 SeO 4 ·10H 2 O, Aldrich, USA) in deionized water and zirconium oxychloride octahydrate (ZrOCl 2 ·8H 2 O, Wako, Japan) in 0.1 M hydrochloric acid, respectively. However, the preparation of the adsorption gel (Zr IV -loaded saponified orange waste (SOW)) has been described in detail elsewhere (Biswas et al., 2007). About 100 g of orange waste was mixed along with 8 g of Ca(OH) 2 and crushed to make small particles. The mixture was transferred into a beaker and a substantial amount of deionized water was added, which was then stirred for 24 h at about 200 rpm at room temperature in order to facilitate the saponification. The pH of this suspension was maintained constant at around 12.5 by adding sodium hydroxide solution. After stirring, the suspension was repeatedly washed with deionized water until neutral pH by means of decantation and finally filtered to obtain a wet gel, which was dried in a convection oven for about 48 hours at 70ºC to produce a dry gel. The specific surface area of this gel was measured as 7.25 m 2 /g by using Belsorp 18PLUS-SP (BEL. JAPAN. INC.) according to the BET method. The SOW gel was further modified by loading with Zr IV to facilitate ligand exchange adsorption of selenate. Approximately 3 g of SOW gel was equilibrated with 500 mL of a 0.1 M zirconium solution at pH 2.11 for 24 h. The gel was then filtered and washed with deionized water until neutral pH, followed by drying in vacuum and finally sieved to produce a particle size fraction between 75 and 150 μm for the adsorption tests. The presence of Zr IV ion in the gel was confirmed by taking energy dispersive X-ray spectroscopy (EDS) (Biswas et al., 2009) and the amount of zirconium loaded was calculated from the difference in the metal concentration in the solution before and after loading. However, due to such chemical modification, the gel became aqueous-insoluble, which is a distinct benefit to operate this gel for a prolonged time (Dhakal et al., 2005). Selenate sorption studies Batch adsorption tests were carried out to determine the adsorption behavior for selenate on Zr IV -loaded SOW gel. The kinetic studies of Se VI adsorption were carried out by first combining a 400 mg gel together with 240 mL of measured concentration of Se VI solutions into a 300 mL Erlenmeyer flask. The suspensions (pH 3) were stirred at 150 rpm and the temperature was kept constant at 30˚C. Reduction in selenium concentration was measured at varying time. Adsorption isotherms were measured at different pH. These tests were carried out by mixing 20 mg of gel together with 10 mL of selenate solution of different concentrations in 50 mL conical flasks. The flasks were then shaken at 150 rpm for 24 h in a thermostated shaker maintained at 30ºC after which Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 315 - 0 5 10 15 20 0 300 600 900 1200 1500 q (mg-Se VI /g) Time (min) 23 mg/l 12 mg/l Fig. 1 - Adsorption kinetics of different initial selenium concentrations at pH 3 the suspensions were filtered and analyzed. Earlier, using a similar procedure, the effect of pH on Se VI adsorption was examined in a series of experiments where the initial Se VI concentration was maintained constant at different pH. Analysis The pH and concentration of selenate in the test solutions were measured by using DKK-TOA model HM-25G pH meter, calibrated with buffers of pH 4.0 and 7.0 on a regular basis, and a Shimadzu ICPS-8100 ICP/AES spectrometer, respectively. The linearity of calibration of ICP/AES reaches from 2 μg/L up to 1000 μg/L with a detection limit of about 1 μg/L for measured species. Reproducibility of all adsorption experiments was confirmed by repeating the same adsorption test. RESULTS AND DISCUSSION Kinetic studies Sorption kinetics of Se VI is depicted in Fig. 1 where it is shown that the sorption capacity of Se VI has been increased from 7.3 to 13.1 mg/g with the increase in the initial Se VI concentration from 12 to 23 mg/L. It is also shown that the initial selenium adsorption is very fast and then the sorption capacity does not appreciably change with time, which finally reaches plateau at 2 and 4 h for 12 and 23 mg/L, respectively, under the operating conditions. The time-dependent adsorption results (up to 120 min) of Se VI were analyzed using pseudo-first-order and pseudo-second-order models (Ho and McKay, 1998). The simplified kinetic equations of pseudo-first-order and pseudo-second-order reactions are as follows: t k qqq ete 303.2 log)log( 1 11 −=− (1) 2 22 2 1 e et qk q t q t += (2) where k 1 and k 2 are the rate constants for pseudo-first-order (1/min) and pseudo-second-order adsorption (g/mg/min), respectively, while q t and q e are the amounts of selenate adsorbed (mg/g) at time t and at equilibrium, respectively. The mg/L mg/L Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 316 - y = 0.137x + 0.61 R² = 0.99 y = 0.076x + 0.56 R² = 0.99 0 5 10 15 20 0 30 60 90 120 150 t/q t Time (min) 12 mg/l 23 mg/l y = 1.27x + 2.4443 y = 0.57x + 2.6596 0 3 6 9 12 15 03691215 q t (mg/g) t 0.5 (min 0.5 ) 23 mg/l 12 mg/l Fig. 2 (a) - Second-order kinetic plot for the adsorption of selenate on the gel Fig. 2 (b) - Determination of intra-particle diffusion rate constant Table 1 - Pseudo-second-order kinetic parameters for the adsorption of Se VI C in (mg/L) q e2(cal) (mg/g) k 2 (g/mg/min) R 2 q e(exp) (mg/g) 12 7.3 3.07×10 -2 0.99 7.3 23 13.2 1.03×10 -2 0.99 13.1 kinetic parameters can be evaluated from the linear plots of )log( 1 te qq vs t and t/q t vs t. It was found that the data fitted well with the pseudo-second-order equation (Fig. 2 (a)) with respect to adsorption capacity and correlation coefficients (Table 1). At the same time, the rate was found to be a function of the initial selenium concentration. However, this analysis does not provide any information regarding the transport of solutes inside adsorbents. Figure 1 clearly shows that there are different mechanisms involved in the sorption process: film diffusion in the first step and intra-particle diffusion in the second step, which eventually leads to final equilibrium. In order to identify the diffusion mechanisms during the sorption process, the experimental kinetic data (up to 60 min) were tested against the intra-particle diffusion model (Weber and Morris, 1963) as shown in Fig. 2 (b). The initial rate of intra-particle diffusion can be defined as follows (Eq. 3): 5.0 tkq it = (3) where k i is the intra-particle diffusion rate constant (mg/g/min 0.5 ), which can be determined from the slope of the straight line portion of q t vs t 0.5 . However from Fig. 2(b), k i was found to increase with the increase in the initial Se VI concentration and was evaluated to be 0.57 and 1.27 mg/g/min 0.5 for 12 and 23 mg/L of selenate, respectively. The increased slope value (k i ) can be explained by the growing effect of concentration gradient as driving force. Moreover, the slopes did not pass through the origin, which indicated that the adsorption mechanism is complex and both the surface adsorption and intra-particle diffusion contribute to the rate-determining step. A similar observation was reported by Teng et al. (2009) for fluoride removal by hydrous manganese oxide-coated alumina. Effect of pH on selenate adsorption Since pH, in general, is considered as an important parameter that controls the mg/L mg/L mg/L mg/L - Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 317 - adsorption, the adsorption edges at two different concentrations of selenate were studied (Fig. 3) at different pH values ranging from 0.5 to 8. At low pH, the adsorption of selenate increases with the increase in pH and reaches the maximum (from pH 1 to 2), which finally decreases with further increase in pH. Selenate can exist in different ionic species depending on the solution pH. The dominant species in the above-mentioned pH range are HSeO 4 - and SeO 4 2- ions which can be adsorbed on the gel by substituting hydroxyl ions from the coordination sphere of the immobilized zirconium ions. Since zirconium tends to be extensively hydrolyzed, a number of hydroxyl ions are therefore available for ligand exchange with selenate anions (Cotton et al., 1999). Such adsorption can be explained by ligand exchange mechanism as shown in Fig. 4. This mechanism has been further confirmed by the increase in pH after adsorption. However, leaking of zirconium was found to be very insignificant at pH > 1. Hence, an optimum pH for the system operation was selected to be 1.5 as it yields the highest selenate removal. When the initial concentration of Se VI increased, the adsorption curve shifted to the left because the number of adsorption sites remained fixed. If the gel provides a sufficient number of adsorption sites, Se VI adsorption density will increase with higher surface loading, but this phenomenon will be constrained by the concentration of adsorption sites. Therefore, the adsorption density on the gel in relatively high concentration solution will not be greater than that in a more dilute solution. This is in good agreement with the phenomenon reported by Kuan et al. (1998) for selenium removal using aluminum-oxide-coated sand. However, in both cases the sorption of Se VI decreases with the increase in pH, which could be due to the increasing amount of OH - in the solution resulting to an intermolecular competition between OH - and selenate. 0.0 1.0 2.0 3.0 4.0 5.0 0 20 40 60 80 100 012345678 Leaking of zirconium (mM) Selenate removal (%) Equilibrium pH 0.11 mM (1) 5.5 mM (2) Zr(IV) leakage (1) Zr(IV) leakage (2) Fig. 3 - Effect of pH and initial concentration on the adsorption of Se VI O OH OH C O O O OZr + OH - H 2 O - A OH - H 2 O n A - OH - O OH OH C O O O OZr + OH - H 2 O - HO OH - H 2 O n Fig. 4 - Inferred mechanism for Se VI adsorption on the gel via ligand-substitution Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 318 - 0 5 10 15 20 25 30 0 200 400 600 800 q (mg/g) C e (mg/l) pH 1.5 pH 2.2 pH 3 (a) pH 3 y = 0.07x + 1.1 R² = 0.99 pH 2.2 y = 0.05x + 0.10 R² = 0.99 pH 1.5 y = 0.04x + 0.22 R² = 0.99 0 5 10 15 0 40 80 120 160 200 C e /q C e (mg/l) pH = 3 pH = 2.2 pH = 1.5 (b) Fig. 5 - (a) Adsorption isotherms at different pH values and (b) corresponding Langmuir plots Isotherm study The experimental data obtained for different initial selenate concentrations at constant temperature and pH were taken for isotherm study. The distribution between the adsorption gel and selenate at equilibrium was described by the Langmuir isotherm model as depicted in Figs. 5 (a) and (b). This model is widely used under the assumption that maximum adsorption occurs when the surface of the adsorbent is covered (monolayer) by adsorbate. The Langmuir equation applied for this study is given below (Eq. 4): m e me e q C bqq C += 1 (4) where C e is the equilibrium concentration (mg/L), q e is the amount adsorbed at equilibrium (mg/g), q m is the adsorption capacity for Langmuir isotherms and b is the Langmuir isotherm constant (l/mg). The adsorption capacity of the gel was evaluated to be 25, 20 and 14 mg-Se VI /g at pH 1.5, 2.2 and 3.0, respectively. The essential feature of the Langmuir isotherm can be expressed in terms of a dimensionless constant termed separation factor (R L ). In order to predict the adsorption efficiency, the separation factor was calculated by using the following equation (Viswanathan and Meenakshi, 2008): 0 1 1 bC R L + = (5) where b is the Langmuir isotherm constant (l/mg) and C 0 is the initial concentration (mg/L) of selenate. The significance of separation factor is that if R L > 1, the adsorption is unfavorable; if 0 < R L < 1, the adsorption is favorable; and if R L = 0, the adsorption is irreversible. The value of R L for an initial concentration of 10 mg/L was found to be 0.355, 0.474 and 0.611 at pH 1.5, 2.2 and 3.0, respectively, which signifies that the adsorption module is favorable at all pH values tested. Competitive adsorption of Se VI with other anions onto the Zr IV -loaded SOW gel To assess the competing effects of other anions on Se VI removal by the Zr IV -loaded (mg/L) (mg/L) Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 319 - 0 20 40 60 80 100 024681012 Removal (%) IMR Selenate and chloride Selenate and sulfate Selenate and phosphate Fig. 6 - Effect of Cl - , SO 4 2- and PO 4 3- on the adsorption of Se VI SOW gel, batch adsorption tests were carried out by adding Cl - , SO 4 2- and PO 4 3- at various initial molar ratios (IMR), where IMR is the ratio of molarity of other anions to that of selenium. The effect of the presence of Cl - , SO 4 2- and PO 4 3- on the adsorption of Se VI at pH 2 (Fig. 6) shows that Cl - does not interfere with selenium adsorption while SO 4 2- and PO 4 3- reduce the extent of selenium removal. The difference in the reduction of selenium removal in the presence of such anions can be related to their different complex formation characteristics with the gel. Since Se VI exists in solution as SeO 4 2- whose chemical properties resemble those of SO 4 2- , its adsorption onto the gel was affected by non-specific adsorption between the gel and sulfate (Lo and Chen, 1997). However, anion such as phosphate is bound relatively strongly with the adsorption sites via complex formation on inner-sphere, which is barely affected by ionic strength (Zhang et al., 2008). This is the reason why the removal efficiency of Se VI decreases significantly with the presence of sulfate and phosphate. CONCLUSIONS From the kinetic study, it was observed that adsorption kinetics is governed not only by surface adsorption but also by intraparticle diffusion. The adsorption isotherms fitted well with the linear form of Langmuir equation and a maximum sorption capacity of 24 mg-Se VI /g was observed. The trend of higher selenium sorption at acidic pH is due to the predominance of SeO 4 2- and HSeO 4 - in the aqueous phase. Although, the Zr IV -loaded SOW gel exhibited a good selenate removal at lower pH range, the reduction in sorption capacity at higher pH suggests a prospect for adsorbent regeneration. The competitive adsorption of SO 4 2- and PO 4 3- with respect to Se VI is more evident than that of Cl - . REFERENCES Biswas B. K., Inoue K., Ghimire K. N., Ohta S., Harada H., Ohto K. and Kawakita H. (2007). The adsorption of phosphate from an aquatic environment using metal-loaded orange waste, J. Colloid Interface Sci., 312, 214-223. Biswas B. K., Inoue J., Kawakita H., Ohto K. and Inoue K. (2009). Effective removal and recovery of antimony using metal-loaded saponified orange waste, J. Hazard. Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 320 - Mater., 172, 721-728. Cotton F. A., Wilkinson G., Murillo C. A. and Bochmann M. (1999). Advanced Inorganic Chemistry, sixth ed. John Wiley and Sons, Inc., Singapore. Dhakal R. P., Ghimire K. N. and Inoue K. (2005). Adsorptive separation of heavy metals from an aquatic environment using orange waste, Hydormetallurgy, 79, 182-190. Ho Y. S. and McKay G. (1998). A comparison of chemisorpion kinetic models applied to pollutant removal on various sorbents, Trans. IChemE., 76B, 332-340. Kuan W-H., Lo S-L., Wang M. K., and Lin C-F. (1998). Removal of Se(IV) and Se(VI) from water by aluminum-oxide-coated sand, Water Res., 32, 915-923. Lo S-L. and Chen T-Y. (1997). Adsorption of Se(IV) and Se(VI) on an iron-coated sand from water, Chemosphere, 35, 919-930. Presser T. S. and Ohlendorf H. M. (1987). Biogeochemical cycling of selenium in the San Joaquin Valley, California, USA, Environ. Manage., 11, 805-821. Tan J. and Huang Y. (1991). Selenium in geo-ecosystem and its relation to endemic diseases in China, Water, Air, Soil Pollut. 57, 59-68. Teng S-X., Wang S-G., Gong W-X., Liu X-W. and Gao B-Y. (2009). Removal of fluoride by hydrous manganese oxide-coated alumina: Performance and mechanism, J. Hazard. Mater., 168, 1004-1011. Viswanathan N. and Meenakshi S. (2008). Selective sorption of fluoride using Fe(III) loaded carboxylated chitosan beads, J. Fluorine Chem., 129, 503-509. Weber W. J. and Morris J. C. (1963). Kinetics of adsorption on carbon solution, J. Sanit. Eng. Div. Am. Soc. Civ. Eng., 89, 31-59. Zhang N., Lin L-S. and Gang D. (2008). Adsorptive selenite removal from water using iron-coated GAC adsorbents, Water Res., 42, 3809-3816. . Eng., 89 , 31-59. Zhang N., Lin L-S. and Gang D. (20 08) . Adsorptive selenite removal from water using iron-coated GAC adsorbents, Water Res., 42, 380 9- 381 6 (mg/L) Journal of Water and Environment Technology, Vol. 8, No.4, 2010 - 319 - 0 20 40 60 80 100 024 681 012 Removal (%) IMR Selenate and chloride Selenate and