JOURNAL OF COLLOID AND INTERFACE SCIENCE ARTICLE NO.
206, 94 –101 (1998)
Removal of Fluoride from Aqueous Solution by Using Alum Sludge M. G. Sujana, R. S. Thakur, and S. B. Rao1 Regional Research Laboratory (CSIR), Bhubaneswar-751 013, India Received December 15, 1997; accepted April 23, 1998
depending on its concentration. No other naturally occurring inorganic constituent in drinking water has been given so wide importance as compared to fluoride. In excess of 1.5 to 2.0 mg/L fluoride is known to cause permanent gray or black mottling of teeth enamel and the long-term intake of 3 to 10 mg/L may result in abnormal bone growth in both humans and animals (1). On the other hand, fluoride is recognized as an essential constituent in the human diet. Skeletal and dental problems can be prevented by maintaining fluoride concentration of about 1 ppm in the dietary intake. Several methods were tried for defluoridation of water, namely, adsorption, ion exchange, electrolysis, and precipitation. The different materials used for defluoridation include activated carbon, bone charcoal, tricalcium phosphate, synthetic ion exchangers, activated alumina, alum, and lime (2). In recent years, considerable attention has been devoted to the study of different types of low-cost materials such as tree bark, wood charcoal, saw dust, weeds, and other waste materials for adsorption of some toxic elements (3– 8). Alum sludge is a waste product generated during the manufacture of alum from bauxite by the sulphuric acid process. The sludge mainly consists of oxides of aluminium and titanium with small amounts of undecomposed silicates. Each of these oxides is known to possess adsorption and ion exchange properties (9). Raw alum sludge is highly acidic and is a pollutant. At present, economically viable methods for disposal or reuse are not known. In this context, we examined its use as an adsorbent for the removal of fluoride from polluted waters. The present studies are carried out with synthetic fluoride solutions with the objective of establishing process parameters. A few experiments were carried out with additional anions to simulate industrial waste waters.
The ability of treated alum sludge to remove fluoride from aqueous solution has been investigated. The studies were carried out as functions of contact time, concentration of adsorbent and adsorbate, temperature, pH, and effect of concentrations of other anions. The data indicate that treated alum sludge surface sites are heterogeneous in nature and that fits into a heterogeneous site binding model. The optimum pH for complete removal of fluoride from aqueous solution was found to be 6. The rate of adsorption was rapid during the initial 5 minutes, and equilibrium was attained within 240 minutes. The adsorption followed first-order rate kinetics. The present system followed the Langmuir adsorption isotherm model. The loading factor (i.e., the milligram of fluoride adsorbed per gram of alum sludge) increased with initial fluoride concentration, whereas a negative trend was observed with increasing temperature. The influence of addition of anions on fluoride removal depends on the relative affinity of the anions for the surface and the relative concentrations of the anions. © 1998 Academic Press Key Words: alum sludge; adsorption; chemical composition; fluoride; removal; anion competition.
Fluoride pollution in the environment occurs through two different channels: natural sources and anthropogenic sources. Fluoride is frequently encountered in minerals and in geochemical deposits. Because of the erosion and weathering of fluoride-bearing minerals it becomes a surface species. On the other hand, fluorine compounds are industrially important and are extensively used in semiconductors, fertilizers, aluminum industries, and nuclear applications. Toxic wastes containing fluorine/fluoride are generated in all industries using fluorine or its compounds as a raw material. Prominent among these is the aluminum smelter where fluorine gas is released into the atmosphere or treated as per the plant design specifications. The contamination resulting from spent pot liners is a major problem and mostly is not properly taken care of. Although small scale units treat the wastes by lime, large-scale industries must use elaborate treatment techniques before discharging the effluent. Fluoride in drinking water may be either beneficial or detrimental to health, particularly to infants and young children, 1
MATERIALS AND METHODS
Even though alum has widespread industrial and domestic applications, the waste produced from alum plants has caused serious concern to the society as well as environmentalists. The toxicity of alum sludge is caused by its high acidity. Alum sludge primarily contains Al2O3, SiO2, Fe2O3, TiO2, and SO2 4. In the absence of viable processes for utilization, these metal values are being discarded as waste. Use of industrial waste material as an adsorbent for the removal of toxic elements may not appear attractive, but alum sludge with its high metal
To whom correspondence should be addressed.
0021-9797/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
FLUORIDE REMOVAL FROM AQUEOUS SOLUTION
TABLE 1 Chemical Analysis of Dried Alum Sludge (Wt %) Fe2O3 Al2O3 TiO2 SiO2 SO22 4 LOI
7.18 47.2 20.65 1.6 3.2 19.0
composition could be a reasonable substitute for more conventional expensive adsorbents. The alum sludge sample used in this study was collected from the Visakhapatnam (India) Alum plant. The sludge collected was dried at 110°C for 24 h and crushed to yield a powder. This powder was washed repeatedly first with tap water and later with deionized water followed by drying in air. The sample thus washed was further calcined at 400°C for 3 h, sieved to ,100 mm particle size, designated as treated alum sludge (TAS), and stored in airtight containers until further use. The alum sludge sample was dissolved in conc. HCl (acid digestion) and filtered. The filtrate and residue were further analyzed by following the recommended analytical procedures (10). The chemical composition of the alum sludge given in Table 1 shows that the major components are alumina and titanium oxides. The mineral phase composition was determined by X ray diffraction method. The major mineral phases are boehmite, gibbsite, a-quartz, anatase, and hematite; this may vary from one sample to other depending on the type of bauxite used for the production of alum. The surface area of treated sample was determined by the Brunauer, Emmett, and Teller (BET) method at liquid nitrogen temperature using a Quantasorb (Quantachrome Corporation, USA) and found to be 119.4 m2/g.
taken and that remaining in the solution. All chemicals used are of analytical grade. All spectrophotometric measurements were made on Chemito 2500UV-Visible Spectrophotometer using 10-mm matched quartz cells. The pH of the solutions at the beginning and end of the experiments were measured, and the average values are reported. All pH measurements were made by an Elico Digital pH meter (model L1 120) using a combined glass electrode (model CL 51). The pH meter was calibrated with Orian Standard buffers before any measurement. The experimental parameters studied are adsorbent concentration (25– 800 mg/50 mL), contact time (2–240 minutes), initial fluoride concentration (5–35 mg/L), pH (3.5– 8.8), temperature (307–337 K at 10 intervals), and anion concentrations (synthetic solutions of nitrate, sulphate, phosphate and silicate). RESULTS AND DISCUSSION
Removal of fluoride as a function of adsorbent dosage (both treated and untreated) is shown in Fig. 1. It is evident that for the quantitative removal of 10 mg/L of fluoride, data clearly show that treated is more effective than untreated. This says that the treatment definitely improved the adsorption capacity with increased surface area and available binding sites. Effect of Adsorbent Concentration The influence of varying concentrations of sludge on the adsorption of fluoride at a particular pH is shown in Fig. 2a.
Adsorption Tests Adsorption tests were carried out with raw and treated alum sludge. Synthetic fluoride solutions were used in the adsorption experiments. Standard 0.01 M sodium fluoride was prepared in deionized water and diluted to exactly 100 mg/L. Experimental solutions were prepared by appropriate dilution. All adsorption tests were carried out at constant ionic strength of 0.1 M maintained with potassium chloride. A known amount of alum sludge and fluoride solution were taken in a 100-mL stoppered conical flask. Potassium chloride was added to maintain ionic strength, and pH was adjusted to the desired level with 0.1 M NaOH or 0.1 M HCl solutions. The final volume was made up to 50 mL with deionized water. The flask was agitated at constant speed in a thermostatic water bath at the designated temperature over a period of time and filtered on a Whatman No. 42 filter paper. The concentration of fluoride in the filtrate was determined by SPADNS (11) method and the percentage of fluoride adsorbed was calculated from the ratio of fluoride
FIG. 1. removal.
Effect of treated and untreated adsorbent concentration on fluoride
SUJANA, THAKUR, AND RAO
FIG. 2. (a) Percentage of fluoride adsorption and loading capacity for different alum sludge amounts. (b) The plot of log KD value as a function of alum sludge concentration.
The concentration of surface hydroxyl groups is related to sludge concentration through surface site density. Therefore percent of adsorption increased with sludge dose, whereas loading capacity decreased. A distribution coefficient KD reflects the binding ability of the surface for an element. The KD values of a system mainly depends on pH and type of surface. The distribution coefficient KD values for fluoride and alum sludge at pH 6 were calculated (12) with K D 5 C s/C w ~m3/kg!,
where Cs is the concentration of fluoride in the solid particles (mg/kg) and Cw is the concentration in water (mg/m3). It is seen that the distribution coefficient KD increases with an increase in alum sludge concentration, indicating the heterogeneous surface of the alum sludge. It is seen in Fig. 2b that KD increases with an increase in alum sludge concentration at constant pH. If the surface is homogeneous, the KD values at a given pH should not change with adsorbent concentration. Note that heterogeneous surface sites are reported on homogeneous solids also (13). Effect of Contact Time and Initial Fluoride Concentration The variation of fluoride adsorbed with time is shown in Fig. 3. It was observed that with a fixed amount of alum sludge, the amount of fluoride adsorbed increases with time as well as concentration. The loading capacity (i.e., the amount of fluo-
FIG. 3. Adsorption yield of fluoride as a function of time at different initial fluoride concentrations.
FLUORIDE REMOVAL FROM AQUEOUS SOLUTION
values were calculated from the slope of the plot and found to be 2.86 3 1022, 1.92 3 1022, 1.7 3 1022, and 1.5 3 1022 min21 for the initial fluoride concentration of 5, 10, 20, and 35 mg/L, respectively. Intraparticle diffusion. Besides the adsorption at the outer surface of the adsorbent, the adsorbate molecules may also diffuse into the interior of the porous adsorbent. This was studied by plotting the amount of fluoride adsorbed vs. the square root of time for different initial fluoride concentrations (Fig. 5). A linear relation was observed indicating the control of adsorption by intraparticle diffusion (16). The intraparticle diffusion rate constants were calculated from the slopes of the curves, and the values were found to be 2.4 3 1022, 4.5 3 1022, 6.2 3 1022, and 9.3 3 1022 mg/g/min0.5 for initial fluoride concentrations 5–35 mg/L, respectively. The adsorption isotherm. The adsorption isotherm for fluoride initial concentrations 5–35 mg/L at constant temperature (305 K) and pH 6 is shown in Fig. 6. The data correlated with linear form of Langmuir adsorption isotherm model C e/~ x/m!1/K 1K 2 1 C e/K 2,
FIG. 4. Lagergren plot for the removal of fluoride at different initial concentrations.
ride adsorbed per gram of sludge) increased with time and concentration and then attained a constant value after 2 h. The time to reach equilibrium conditions appears to be independent of initial fluoride concentrations. The adsorption of fluoride decreased from 100 to 60% by increasing fluoride concentration from 5 to 35 mg/L. Further, it was observed that the removal curves are smooth and continuous indicating the possibility of the formation of monolayer coverage of fluoride ion at the interface of alum sludge.
where Ce is the equilibrium concentration (mg/L), x/m is amount adsorbed at equilibrium (mg/g), and K1 and K2 are Langmuir constants related to equilibrium constant and adsorption capacity, respectively. The linear plot of Ce/(x/m) vs. Ce (Fig. 7) indicates the applicability of equation. The values of K1 and K2 were determined from the slope of the plot and
Adsorption Kinetics The kinetics of fluoride removal on alum sludge from aqueous solutions was studied for its application in the treatment of industrial effluents containing high fluoride levels. The rate constant of fluoride adsorption on alum sludge has been interpreted in terms of the adherence of fluoride on the active sites of the adsorbent as well as its intraparticle diffusion within the pores of the adsorbent (14). The specific rate constant Kr was determined using the Lagergren equation (15) log~q e 2 q! 5 log q e 2 K r/ 2.303Xt
where qe and q (mg/g) refer to the amount of fluoride adsorbed at equilibrium and at any time t. The plot of log(qe 2 q) vs. time presented in Fig. 4 shows the straight line curve indicating the applicability of the Lagergren equation and first-order kinetics. The adsorption rate constants Kr
Intraparticle diffusion at different initial fluoride concentrations.
SUJANA, THAKUR, AND RAO
FIG. 6. Adsorption isotherm for different initial fluoride concentrations on alum sludge.
found to be 1.098 L/mg and 5.394 mg/g, respectively. This clearly shows the formation of monolayer coverage of fluoride ion at the outer surface of the adsorbent.
FIG. 8. Adsorption of fluoride as a function of equilibrium time and temperature.
Effect of Temperature For an increase in temperature from 307 to 337 K, using 20 mg/L of initial fluoride concentration, an adverse effect was observed on the adsorption of fluoride. The percent of adsorption decreased from 85 to 72% at pH 6. Even though the fluoride adsorption at a given temperature increased with time, overall it decreased with an increase in temperature (Fig. 8). This may be happening because the rise in temperature increases the escaping tendency of the molecules from interface and thereby diminishes the extent of adsorption (i.e., decrease in adsorption capacity). This may also be the result of increased desorption by an increase in thermal energy of the adsorbate (17). The uniformity or nonuniformity of the surface sites of alum sludge is also studied by determining isosteric heats of adsorption as a function of adsorption density. By using Clausius–Clapeyron equation (18, 19), the isosteric heats of adsorption are calculated at two different temperatures: DH r 5 R [email protected]
~C 2/C 1!#/~1/T 2 2 1/T 1!
FIG. 7. Langmuir plot for the adsorption of fluoride on alum sludge at constant temperature (305 K) and pH (6).
where DHr is isosteric heat of adsorption (kJ/mole), R is the gas constant, C2 is the equilibrium concentration of the ion at temperature T2, and C1 is the equilibrium concentration of ion at temperature T1. In homogeneous surfaces, the isosteric heat of adsorption is independent of adsorption density. The decrease of isosteric heat of adsorption with increasing adsorption density (Figs. 9 and 10) shows that the surface of treated
FLUORIDE REMOVAL FROM AQUEOUS SOLUTION
FIG. 9. Adsorption density as a function of equilibrium concentration of fluoride at different temperature.
alum sludge is heterogeneous. These data are consistent with the distribution coefficient KD values (Fig. 2b) and suggest that the surface of treated alum sludge is heterogeneous. The decrease in the isosteric heat of adsorption with increasing adsorption density can be the result of different types of adsorption sites or the interaction of adsorbing ions (20).
FIG. 10. The isosteric heat of adsorption as a function of fluoride adsorption density.
the adsorption at the adsorbent and water interface. Hence, the adsorption of fluoride on alum sludge was studied at different pH values ranging from 3 to 9. The results are
Effect of pH The extent of adsorption of anions is strongly governed by the pH of the solution. Because anion adsorption is coupled with a release of OH2 ions, the adsorption is favored by low pH values (21). Alum sludge is a mixed adsorbent with different metal oxides. In humid environment hydroxylated surfaces of these oxides develop charge on the surface of aqueous solution. The specific adsorption of fluoride on metal oxides is modeled as a two-step ligand exchange reaction: 'SOH 1 H1 7 'SOH1 2 2 'SOH1 2 1 F 7 'SF 1 H2O,
which, combined, gives 'SOH 1 H1 1 F2 7 'SF 1 H2O,
where 'S represents a surface metal ion. The pH of the aqueous solutions is an important variable which controls
Effect of equilibrium pH on fluoride adsorption.
SUJANA, THAKUR, AND RAO
tions in the solution for the same doses of alum sludge. The affinity sequence for anion adsorption on treated alum sludge is in the following order phosphate $ silicate . sulphate . nitrate. Desorption Studies After the adsorption, the resulting fluoride-bearing sludge should be safe for disposal. The stability of this sludge from the point of view of fluoride resolubilization was studied, and the results are shown in Fig. 13. It is clear that partial desorption of fluoride may take place in strongly alkaline medium (pH above 8). Leaching tests were conducted at different time intervals while maintaining the pH constant (6.5); no release of fluoride was found up to 48 h. Comparative studies were also conducted under identical experimental conditions to examine the fluoride removal capability of alum sludge and other known adsorbents (Fig. 14). For the quantitative removal of 25 mg/L of fluoride in 50 mL solution, the adsorbent dosage of different materials is: alum sludge, 0.4 g; alumina, 0.5 g; and activated carbon, .1 g. CONCLUSIONS FIG. 12. sludge.
Effect of anion concentration on adsorption of fluoride by alum
presented in Fig. 11, which reveals that the adsorption of fluoride is maximum at pH 6. This is in agreement with fluoride removal studies on activated alumina by other workers (22). At a pH above 6, fluoride removal decreases sharply as a result of stronger competition from hydroxide ions on adsorbent surface. Adsorption is also found less in the acidic range; this may be as a result of the formation of weakly ionized hydrofluoric acid.
In the present work a simple, fast, and promising method for the treatment of fluoride from contaminated water is suggested. The alum sludge, a waste material from alum manufacture, containing different metal oxides with a het-
Effect of Competitive Ions The drinking or waste water contains many substances. If the relative binding ability of various anions on the sludge surface is known, their influence on the adsorption of fluoride can be estimated. Therefore, it was thought worthwhile to study the effect of competitive ions like sulphate, phosphate, nitrate, and silicate (as Si). Varying amounts of these solutions were prepared from their potassium salts and added to the tests with 20 mg/L fluoride. From Fig. 12, it is observed that the fluoride adsorption decreased from 85 to 62% in case of nitrate and 40% in sulphate. Increasing the dosages from 10 to 50 mg/L of these ions has not much effect on fluoride removal. Studies show that phosphates do not have adverse effect up to the 5-mg/L level (23). The present results show that the defluoridation with alum sludge in presence of phosphate and silicate at higher concentrations (10 –50 mg/L) has an adverse effect on fluoride removal. Both cases of equilibrium fluoride concentration increased with increase of phosphate and silicate concentra-
FIG. 13. Effect of pH on the desorption of fluoride from the surface of loaded alum sludge.
FLUORIDE REMOVAL FROM AQUEOUS SOLUTION
ACKNOWLEDGMENT The authors are grateful to Prof. H. S. Ray, Director, Regional Research Laboratory, Bhubaneswar, for giving permission to publish this paper.
FIG. 14. Effect of the concentration of different adsorbents on fluoride removal at constant time and temperature.
erogeneous surface, has shown a superior adsorption capability for fluoride ion. The adsorption seems to be a surface phenomena. Removal method is favored by the appropriate addition of sludge to fluoride-containing waste water at normal temperature (30°C). The optimum pH for fluoride removal was found to be 6. The adsorption followed firstorder rate kinetics, and data fit into linear form of Langmuir adsorption isotherm model. Studies on the influence of other anions showed that an increase of nitrate dosage from 10 to 50 mg/L has least effect on defluoridation as compared to sulphate. Silicate and phosphate, if present along with fluoride, inhibit fluoride removal by alum sludge. This may be a result of the higher selectivity of metal oxides present in the alum sludge toward silicate and phosphate ions. Disposal of the fluoride-bearing sludge may not pose environmental problems under normal weather conditions.
1. “Kirk Othmer Encyclopedia of Chemical Technology,” 3rd ed., vol. 10. Wiley-Interscience, New York, 1980. 2. Sujana, M. G., Thakur, R. S., Das, S. N., and Rao, S. B., Asian J. Chem. 9(4), 561 (1997). 3. Killedar, D. J., and Bhargawa, D. S., Indian J. Environ. Health 35(2), 81 (1993). 4. Szirmai Endre, Babusek Sandor et al. (Aluminiumpari Tervezo Vallalat (ALUTERV) Aquatech Kotnyezetve delmi es Muszaki Fejleszto Kisszovetkezel) Hung. Teljes Hu 47, 059 (cl.c01F7/74), 30 Jan (1989) 839, 27 Apr. 1987. 5. Arulanantham, A., Ramakrishna, T. V., and Balasubramanian, N., Indian J. Environ. Protection 12(7), 531 (1992). [English] 6. Shyma Nair, Jallam Geeta, and Pandey, G. S., Fluoride 23(1), 35 (1990). [English] 7. Bhargava, D. S., and Kelledar, D. J., Water Res. 26(6), 781 (1992). [English] 8. Wang Rongshu, Li Haiming, Na Ping, and Wang Ying, Water Qual. Res. J. Can. 30(1), 81 (1995). [English] 9. Amphlett, C. B., “Inorganic Ion Exchanges.” Elsevier, New York, 1964. 10. Vogel, A. I., “A Text Book of Quantitative Inorganic Analysis Including Elementary Instrumental Analysis,” Longmans, London, 1962. 11. Standard Methods for the Examination of Water and Waste Water, (Arnold E. Greenberg, Lenore S. Clesceri, Andrew D. Eaton, Eds.) p. 4. APHA, AWWA & WEF, Washington, 1992. 12. Sigg, L., in ‘‘Aquatic Surface Chemistry, Chemical Processes at the Particle Water-Interface’’ (Werner Stumm, Ed.) p. 325. John Wiley & Sons, New York, 1987. 13. Parida, K. M., Gorai, B., Das, N. N., and Rao, S. B., J. Colloid Interface Sci. 185, 335 (1997). 14. Orumwense, F. F. O., J. Chem. Tech. Biotechnol. 65, 363 (1996). 15. Namita Deo and Manzoor Ali. IJEP 12(1), 439. 16. Weber, W. J., and Morris, J. C., J. Soc. Eng. Am. Soc. Civil. Eng. 2, 31 (1963). 17. Panday, K. K., Gurprasad, and Singh, V. N., J. Chem. Tech. Biotechnol. 34A, 367 (1984). 18. Chiou, C. T., Shoup, T. D., and Porter, P. E., Org. Geochem. 8, 9 (1985). 19. Balistrieri, L. S., and Chao, T. T., Soil Sci. Soc. Am. J. 51(5), 677 (1987). 20. Moore, W. J., “Physical Chemistry.” Prentice-Hall, Englewood Cliffs, NJ, 1972. 21. Rude, P. D., and Aller, R. C., Am. J. Sci. 293, 1 (1993). 22. Schoeman, J. J., and MacLeod, H., Water SA 13(4), (1987). 23. Nawlakhe, W. G., and Bulusu, K. R., Indian J. Environ. Health 20(2), 156 (1978).