- Email: [email protected]
Fluoride removal using lanthanum incorporated chitosan beads
Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Fluoride removal using lanthanum incorporated chitosan beads Amit Bansiwal a,1 , Dilip Thakre a,1 , Nitin Labhshetwar a , Siddharth Meshram b , Sadhana Rayalu a,∗ a b
National Environmental Engineering Research Institute, India Department of chemistry, Laxminarayan Institute of Technology (LIT), RTM Nagpur University, Nagpur, India
a r t i c l e
i n f o
Article history: Received 13 April 2009 Received in revised form 13 July 2009 Accepted 15 July 2009 Available online 23 July 2009 Keywords: Fluoride removal Chitosan beads Thermodynamic parameter Mechanism Adsorption isotherm Kinetic
a b s t r a c t Highly selective material based on naturally occurring biomaterial namely chitosan has been designed for the defluoridation of water. Lanthanum incorporated chitosan beads (LCB) were prepared using precipitation method. The synthesis was optimized by varying different synthesis parameters namely lanthanum loading, complexation and precipitation time, strength of ammonia solution used for precipitation, drying time, etc. Lanthanum incorporated chitosan beads were characterized using SEM, FTIR, XRD and EDX. Surface area of LCB was observed to be 2.76 m2 g−1 . The equilibrium adsorption data fitted well to Langmuir adsorption isotherm and showing maximum fluoride adsorption capacity of 4.7 mg g−1 with negligible lanthanum release. Kinetic study reveals that adsorption of fluoride is fast and follows pseudo-first-order kinetics. The effect of pH was also studied and the best efficiency was observed at pH 5. Presence of sulphate, nitrate and chloride marginally affected the removal efficiency, however drastic reduction in fluoride uptake was observed in the presence of carbonate and bicarbonate. Negative value of change in free energy (G◦ ) and positive value of change in entropy (S◦ ) suggest the adsorption of fluoride by LCB is feasible and spontaneous process. Positive value of change in enthalpy (H◦ ) suggests the process of fluoride adsorption is endothermic in nature. Regeneration study reveals that 1 M ammonium chloride solution appears to be the promising regeneration media showing 81.22% regeneration. The adsorption capacity of LCB was similar in fluoride-contaminated ground water collected from Dhar district of Madhya Pradesh, India, as compared to simulated water. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Fluoride is one of the most abundant constituents occurring in groundwater worldwide and creates a major problem in safe drinking water supply. Widespread occurrence of fluoride above the prescribed limit in groundwater meant for human consumption has caused multidimensional health problems. The problem of excess fluoride in drinking waters is aggravating day by day, as more surveys to assess the ground water quality have been undertaken. Fluoride in water derives mainly from dissolution of natural minerals in the rocks and soils with which water interacts. The distribution and health effects of fluoride have recently been reviewed by Ayoob and Gupta [1]. According to authors, about 200 million peoples from 25 nations are facing the problem of excess fluoride in drinking water among which India and China are the worst affected. In India alone, endemic fluorosis is thought to affect around 1 million people [2] and is a major problem in 17 out of the country’s
∗ Corresponding author at: Environmental Materials Unit, National Environmental Engineering Research Institute, Nehru Marg, Nagpur 440020, India. Tel.: +91 7122247828; fax: +91 7122247828. E-mail address: s [email protected] (S. Rayalu). 1 Both authors have contributed equally. 0927-7765/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2009.07.021
22 states, especially Rajasthan, Madhya Pradesh, Andhra Pradesh, Tamil Nadu, Gujarat and Uttar Pradesh [3]. We have also conducted exhaustive water sampling in Dhar district of Madhya Pradesh, India to assess the distribution of fluoride in ground water. Fluoride concentrations in large number of samples fall above 1 mg l−1 (WHO prescribed limit for drinking water) and the maximum concentration found was 12.0 mg l−1 . The most important remedial action is the prevention of further exposure by providing safe drinking water. However in most of the areas source substitution may be impossible due to non-availability of alternate sources and therefore removal of excess fluoride is the only remedy. Several defluoridation techniques namely precipitation [4–6], membrane processes [7] electrolytic [8] ion-exchange [9] and adsorption onto various adsorbents have been evaluated both in field and lab. Owing to various drawbacks associated with conventional defluoridation techniques adsorption-based fluoride removal systems are emerging as the feasible option. Large numbers of adsorbents have been studied for the removal of fluoride from water based on alumina [10], other minerals and metal oxides [11–13], clays [13–15], carbons [16], zeolites [17] industrial/agricultural wastes [18], etc. Chitin is one of the most abundant biopolymer and it is estimated to be produced annually almost as much as cellulose. It has become of great interest not only as an under-utilized resource, but
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
217
2. Materials and methods Nomenclature qe C0 Ce V W Xm KL r Kf n k h qt kp G◦ H◦ S◦ Ko
equilibrium adsorbate capacity (mg g−1 ) initial concentrations of fluoride (mg l−1 ) equilibrium concentrations of fluoride (mg l−1 ) volume of the aqueous solution (l) mass of adsorbent (g) amount of adsorbate at complete monolayer coverage (mg g−1 ) Langmuir constant (affinity) 1 mg−1 dimensionless quantity Freundlich constants related to adsorption capacity (mg g−1 ) adsorption intensity (heterogeneity factor) second-order rate constant (min −1 ) initial sorption rate (mg g−1 min−1 ) amount of fluoride adsorbed per unit mass of adsorbent at time t (mg g−1 ) intraparticle diffusion rate constant (mg g−1 min−1/2 ) standard free energy change (kJ mol−1 ) standard enthalpy change (kJ mol−1 ) standard entropy change (kJ mol−1 ) sorption equilibrium constant
also as a new functional material of high potential in various fields. Chitosan is the N-deacetylated derivative of chitin, which results in higher nitrogen content and hence is a better chelating agent. Majeti and Kumar have reviewed the wide array of applications of chitin and chitosan in photography, cosmetics, environmental, biomedical, ophthalmology, food and nutrition, paper finishing, solid-state batteries, drug delivery systems, and also extensive biotechnological applications [19]. Extensive reports are also available on application of chitosan for water and wastewater treatment, particularly for the removal of heavy metals, since it has excellent chelating properties [20]. Extensive reports are available on defluoridation of water using chitin/chitosan. Ma et al. have reported magnetic chitosan particle for the removal of fluoride from water in batch mode [21]. Annouar et al. have also reported defluoridation of drinking water through adsorption of fluoride on chitosan followed by electrodialysis. However, there are only a few reports available on functionalization of chitosan using electropositive metals. In recent publication of our group we have reported defluoridation using chitin, chitosan and lanthanum-modified chitosan [22]. However, chitosan naturally occurs in the form of flakes or powder which are having limited utility particularly for column applications due to swelling, low mechanical strength, crumbling, etc. Attempts have also been made to overcome these drawbacks through formation of chitosan beads through various routes [23]. The present paper deals with the synthesis of lanthanum incorporated chitosan beads (LCB) and its application for the removal of fluoride from ground water. The synthesis was optimized by varying different synthesis parameters namely lanthanum loading, complexation and precipitation time, strength of ammonia solution used for precipitation, drying time, etc. The effect of various parameters on fluoride uptake capacity namely pH, adsorbent dose, initial fluoride concentration, presence of other anions and cations was also studied. Evaluation of LCB for the removal of fluoride from ground water collected from fluoride endemic regions of Dhar district of Madhya Pradesh, India was also carried out. Detailed physico-chemical analysis of the untreated and treated water was done to assess the potability of treated water.
2.1. Materials All the chemicals used throughout the study were of analytical grades and were prepared in double distilled water. 85% deacetylated chitosan was purchased from Chemchito Natural Products, Chennai, India. Lanthanum acetate was purchased from Himedia, Mumbai and sodium fluoride, sodium chloride, sodium bicarbonate, sodium nitrate, aqueous ammonia and acetic acid were procured from E. Merck India Ltd. Stock solution of 1000 mg l−1 fluoride was prepared from dried sodium fluoride and working fluoride solutions of required concentrations were prepared by diluting the stock solution.
2.2. Synthesis of LCB For the synthesis of LCB, 9 g of chitosan (85% deacetylated) was dissolved in 300 ml of acetic acid (CH3 COOH) solution (5%, v/v). In another beaker required quantities of lanthanum acetate was dissolved in 100 ml of distilled water. The lanthanum acetate solution was then added to the chitosan solution with stirring for 180 min. The lanthanum loading was varied from 1 to 20 wt%. The resulting La–chitosan solution was added drop wise into NH4 OH solution (50%, v/v) under vigorous stirring, using a syringe pump. The gel macro-spheres formed were allowed to stabilize in NH4 OH solution for 60 min. The beads were separated from the NH4 OH solution and washed with deionised water and dried at 50–80 ◦ C for 24 h in oven.
2.3. Batch adsorption studies Batch adsorption studies were carried out as follows: 50 ml of fluoride solution was mixed with different quantities of the adsorbent in PVC conical flasks. The flasks were kept on orbital shaker for 24 h at 30 ± 2 ◦ C to attain the equilibrium. The solution was then filtered and the fluoride concentrations were determined in filtrates and the amount of fluoride adsorbed was calculated from following equation: qe = (C0 − Ce ) ×
V W
(1)
where qe is the adsorption capacity (mg g−1 ) at equilibrium, C0 and Ce are the initial and equilibrium fluoride concentrations (mg l−1 ), respectively, V is the volume (ml) of solution and W is the mass (g) of adsorbent used. Besides determining the defluoridation capacities, the effects of other factors namely adsorbent dose, initial fluoride concentration, pH of the medium, presence of interfering ions normally present in drinking water were also studied. The effect of solution pH on fluoride uptake was studied by maintaining the pH by the addition of 0.1N HCl or 0.1N NaOH. The pH was varied from 3 to 9. The effects of various anions namely chloride, sulphate, nitrate, carbonate and bicarbonate were studied at optimal experimental conditions. Reusability studies of the LCB were also conducted to assess the cyclic use of the adsorbent. To study the reuse of adsorbent, required quantity of adsorbent was mixed with fluoride solution with initial concentration of 5 mg l−1 . After equilibrium, the adsorbent was separated by filtration and dried in air and was repeatedly used for fluoride adsorption to study the degree of reuse. Comparison of fluoride uptake in fluoride solution prepared in distilled water henceforth referred to as simulated water and fluoride-contaminated field water collected from Dhar district of Madhya Pradesh, India was also studied.
218
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
2.4. Methods of analysis Fluoride analysis was done by ion selective electrode method using ion meter (Thermoelectron, USA) coupled with fluoride ion selective electrode (Thermoelectron Model 9609 BNWP). TISAB III was used as buffer for maintaining the pH and background ion concentrations during measurement. pH was measured using the same ion meter coupled with pH electrode. The concentrations of various co-ions and other physico-chemical parameters of treated and untreated water were determined using standard methods [24]. Carbonate and bicarbonate alkalinity, total hardness and chlorides were determined titrimetrically, whereas the concentrations of nitrate and sulphate were determined spectrophotometrically using UV visible spectrophotometer (Chemito, Model UV-2100). The amount of lanthanum if any leached from adsorbent and other heavy metals was determined using ICP-OES (PerkinElmer, Model Optima 4100DV). Blank experiments were conducted through-
out the studies and majority of the experiments were repeated twice and it was observed that the experimental error was within ±2%. 2.5. Physical characterization The powder X-ray diffraction pattern of LCB before and after fluoride adsorption was recorded on Rigaku X-ray diffractometer in order study the structure of the adsorbent. The surface morphology of chitosan and LCB was observed using scanning electron microscopy (Jeol, JXA-840 A, Electron probe microanalyser, Japan). The elemental composition of LCB was determined by Energy Dispersive X-ray analysis. In order to determine the presence of functional groups in adsorbents the FTIR spectra of chitosan and LCB before and after fluoride adsorption in KBr pellets were recorded on Bruker, Model Vertex 70 spectrometer. The surface area of LCB was determined by BET surface area method.
Fig. 1. FTIR of: (a) chitosan, (b) LCB before adsorption, and (c) LCB after adsorption.
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
3. Results and discussion 3.1. Characterization of LCB 3.1.1. FTIR of LCB Fig. 1 shows the FTIR spectra of chitosan and lanthanummodified chitosan. From the FTIR spectra of chitosan it is observed that the band at 1571 cm−1 has larger intensity than the band at 1676 cm−1 suggesting effective deacetylation [25]. When chitin deacetylation occurs, the band at 1676 cm−1 decreases while the growth at 1571 cm−1 occurs, which indicates the presence of –NH2 group. The intensity of the band at 1571 cm−1 in chitosan was significantly reduced in LCB, which may be due to the coordination of lanthanum with –NH2 group. The coordination of chitosan with La+3 showed the involvement of amine and secondary hydroxyl group of chitosan in chelate formation. The band at 3695 cm−1 in chitosan is due to the stretching vibration of –NH group, which is shifted to 3610 cm−1 in LCB, in the 1571 cm−1 band to 1549 cm−1 and shift in 1059 cm−1 band which is due to secondary hydroxyl group to 1028 cm−1 supports the involvement of amine and secondary hydroxyl group in chelate formation [26]. The bands at 2924 cm−1 and 1376 cm−1 in chitosan and 2879 cm−1 and 1369 cm−1 in LCB were attributed to C–H stretching vibration in polymeric backbone and C–H bending, respectively. The band at 3294 cm−1 in chitosan is attributed to stretching vibration of hydroxyl group, which is shifted to 3056 cm−1 in LCB. The band at 1432 cm−1 in LCB can be assigned to the stretching vibration of C–N group of chitosan template. Also, the FTIR spectra of lanthanummodified chitosan show some new peaks at 943 cm−1 , 488 cm−1 ,
219
and 422 cm−1 . The band at 422 cm−1 can be assigned to vibration of La–O bond [27]. From the FTIR spectra of LCB after fluoride adsorption the loss of intensity was observed at 3610 cm−1 (–NH) in LCB before fluoride adsorption which may be due to the interaction of fluoride with the lanthanum coordinated to –NH group of chitosan. 3.1.2. Scanning electron microscopy of LCB The SEM micrographs of chitosan flakes and LCB are shown in Fig. 2. SEM micrograph of chitosan flakes Fig. 2a shows rough surface morphology. SEM micrograph of LCB (Fig. 2b) reveals dense and firm structure with minimum porosity and small oval shaped particles of lanthanum of diameter 2.5–3.0 m and length 4.0–5.0 m uniformly spread over the surface of chitosan. The surface area of chitosan was found to be 2.76 m2 g−1 . 3.1.3. EDX of LCB An EDX spectrum of lanthanum-modified chitosan and its percentage composition is given in Fig. 2c. The EDX spectra of LCB indicate the presence of lanthanum particles over the surface of chitosan. 3.1.4. XRD of LCB The X-ray diffraction pattern of chitosan and lanthanummodified chitosan are shown in Fig. 3. The X-ray pattern of chitosan shows the characteristic crystalline peak at 2 = 8◦ , 20◦ , 29◦ respectively, while these peaks (2 = 20.46◦ ) are less intensive in LCB which suggest that chitosan has a higher crystallinity than its complex. So the X-ray spectrum of LCB shows more amorphous nature, which allows a better accessibility to fluoride and thus a better
Fig. 2. SEM images of: (a) chitosan flakes, (b) LCB, and (c) EDX of LCB.
220
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
activity. The new peaks at 2 = 30◦ , 43.35◦ and 49.78◦ are observed which can be assigned to the lanthanum hydroxide formation. 3.2. Effect of lanthanum loading The optimization of synthesis of LCB has been done by varying the concentration of lanthanum from 1 to 22 (%) in order to study the effect of lanthanum loading on fluoride removal. Increase in lanthanum loading from 1 to 10 (%) resulted in increase percent removal of fluoride from 68 to 98 (%). A maximum of 98% of fluoride removal was achieved at the 10% lanthanum loading on Chitosan. Further increase in the lanthanum loading above 10% has no significant improvement in the fluoride removal capacity. So 10% lanthanum loading was optimal for the synthesis of LCB. 3.3. Preliminary adsorption experiment Preliminary adsorption experiment was carried out using bare chitosan and lanthanum-modified chitosan (LCB) at an initial concentration of 5.35 mg l−1 ; contact time 24 h to check the fluoride removal efficiency of both the adsorbent. The lanthanum-modified chitosan shows much higher fluoride removal efficiency as compared to bare chitosan. Therefore further adsorption studies were carried out using lanthanum-modified chitosan. 3.4. Effect of adsorbent dose The effect of adsorbent dose on fluoride removal at fixed initial concentration (5.34 mg l−1 ) is shown in Fig. 4. It was observed that the percent removal of fluoride increased with increase in the adsorbent dose while loading capacity (amount of fluoride adsorb per gram of the adsorbent) gradually decreases for the same. From the trend it appears that a maximum level (53–90%) of fluoride removal occurs with increase in dose from 0.2 to 0.4 g l−1 .
Fig. 4. Effect of adsorbent dose on fluoride removal by LCB-10 (Initial F concentration: 5.34 mg l−1 , and contact time: 24 h).
Thereafter, there was no significant improvement in the fluoride removal efficiency with increase of adsorbent quantity that may be due to low concentration of fluoride available at higher adsorbent dose. Also, negligible leaching of lanthanum ion 0.06–0.08 mg l−1 is observed with increase in adsorbent dose from 0.2 to 0.4 g l−1 . Therefore 0.4 g l−1 can be considered as optimum dose for further studies. 3.5. Effect of initial fluoride concentration The fluoride concentration in groundwater varies significantly due to seasonal, temporal and geological variations. The defluoridation technique should be able to remove fluoride from water at different initial concentrations. Keeping this in view the effect of initial fluoride concentration was studied. As expected the fluoride removal increases with increase in initial fluoride concentration which may be due to more availability of adsorbate and the active sites of the adsorbent have not reached saturation even at initial fluoride concentrations more than 20 mg l−1 . 3.6. Effect of pH Removal of fluoride through adsorption is highly dependent on solution pH as it can alter the surface charge of the adsorbent. Hence, in the present study the fluoride removal was studied at four different pH namely 3, 5, 7 and 9 and the results are shown in Fig. 5. It is apparent from the results that the pH has significant effect on sorption capacity and the maximum sorption capacity was observed at pH 5. Decrease in sorption capacities was observed below and above 5, which may be attributed to unfavourable surface charges and competition for adsorption sites due to excess anions at alkaline conditions. Above treated water samples at different pH were analysed on ICP in order to check the leaching of lanthanum ion from adsorbent. It revealed from the result that there is negligible leaching of lanthanum ions 0.084, 0.082 and 0.084 mg l−1 at pH 5, 7 and 9, respectively. Whereas significant lanthanum leaching from adsorbent is observed at pH 3 which may be due to the partial dissolution of biopolymer matrix at acidic pH. 3.7. Effect of co-anions
Fig. 3. XRD of: (a) chitosan flakes and (b) lanthanum-modified chitosan.
Beside fluoride the natural ground water always contains various other co-ions, which can compete for sorption sites and result in reduced efficiency of the adsorbent. To study the effect of coanions the fluoride uptake by LCB was studied in the presence of various anions namely like HCO3− , CO3 2− , Cl− , NO3− , and SO4 2−
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
221
or Ce 1 1 = + Ce qe Xm KL Xm
(3)
where qe is the amount of adsorbate adsorbed per unit weight of adsorbent at equilibrium (equilibrium uptake) (mg g−1 ) calculated using Eq. (1), Xm is the maximum adsorption capacity (mg g−1 ), KL is the Langmuir constant, and Ce is the equilibrium concentration of adsorbate in solution (mg l−1 ). The Freundlich adsorption equation is as follows: 1/n
qe = Kf Ce
log(qe ) = log k +
Fig. 5. Effect of pH on fluoride removal by LCB-10 (volume: 50 ml, temperature: 30 ± 1 ◦ C, adsorbent dose: 0.4 g l−1 , contact time: 24 h, and initial F concentration: 5.34 mg l−1 ).
and also cations like Na, Ca, Cu and Fe keeping initial fluoride concentration as 5 mg l−1 . As shown in Fig. 6 the removal efficiency reduces slightly in the presence of anions namely Cl− , NO3− , and SO4 2− whereas drastic reduction in the presence of HCO3− and CO3 2− was observed. The presence of HCO3− and CO3 2− increases the pH with consequent decrease in adsorption capacity. Also these anions appear to be the main competitors for active sites and have more affinity as compared to fluoride. As explained earlier, the removal efficiency reduces to a great extent in alkaline conditions, the observed trend of reduction in removal efficiency in the presence of HCO3− and CO3 2− is explainable since the presence of HCO3− and CO3 2− leads to alkaline conditions. 4. Equilibrium modelling The Langmuir and Freundlich adsorption models were used to obtain various adsorption parameters and to get an insight into the adsorption mechanism, The Langmuir adsorption equation is represented as follows: qe =
Xm KL Ce 1 + KL Ce
(4) 1 log(Ce ) n
(5)
where qe is the amount of adsorbate adsorbed per unit weight of adsorbent at equilibrium (equilibrium uptake) (mg g−1 ), Kf is the Freundlich constant, n = a Freundlich constant, which reflects adsorption intensity, Ce is the equilibrium concentration of adsorbate in solution (mg l−1 ). Results of experimental data were compared to the linear forms of the isotherms to determine which model most accurately described adsorption by the adsorbent. An estimation of the sorption capacity can be obtained from Langmuir constants (Xm and KL ) and the Freundlich constants (Kf and n). The fitness of Langmuir and Freundlich adsorption isotherms with the adsorption data obtained from experimental data is presented in Fig. 7 along with the values of respective correlation coefficients (R2 ). It is apparent from the results that better fit was observed for Langmuir adsorption isotherm, which is also evident from values of R2 , which are 0.96 and 0.93 for Langmuir, and Freundlich isotherms, respectively. Better fit of Langmuir isotherm model indicates the monolayer uniform adsorption. The values of Xm and KL obtained from Langmuir model are 4.7 mg g−1 and 0.023 l g−1 , respectively. Multilayer adsorption on non-uniform sites is the basis of the Freundlich model. The Freundlich model assumes an infinite supply of unreacted adsorbent sites and tends to represent heterogeneous materials better than other models. The higher Freundlich constant (Kf ) characterizes the adsorbent as more reactive, although the constant tends to be site and adsorbent specific. The constant n is an empirical constant and is dependent on the degree of heterogeneity in the adsorbing sites. The values of Kf and n obtained from Freundlich isotherms are 0.106 mg g−1 and 0.864, respectively. Since the value of n is close to unity it reflects favourable adsorption.
(2)
Fig. 6. Effect of different ions on fluoride removal by LCB-10 (volume: 50 ml, temperature: 30 ± 1 ◦ C, adsorbent dose: 0.4 g l−1 , contact time: 24 h, initial F concentration: 5.34 mg l−1 , and concentration of salts: 0.1 M).
Fig. 7. Adsorption isotherms for the removal of fluoride by LCB-10 (volume: 50 ml, temperature: 30 ± 1 ◦ C, initial F concentration: 5.34 mg l−1 , and contact time: 24 h).
222
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
5. Adsorption kinetics The effect of contact time on fluoride uptake at three initial concentrations namely 5.21, 9.75 and 14.8 mg l−1 using LCB was studied. It is apparent that fluoride uptake reached saturation after 70 min at all the three concentrations, indicating fast kinetics. The release of lanthanum ion with respect to time was also observed and it is revealed that there is negligible release of lanthanum ion 0.08–0.084 mg l−1 with the increase in contact time. Two main types of kinetic models frequently used to determine various kinetic parameters of the adsorption system are reaction-based models and diffusion-based models. To study the sorption kinetics pseudo-first-order and pseudo-secondorder kinetic models have been used. A simple pseudo-first-order kinetic model also known as Lagergren equation is represented as [28] log(qe − qt ) = log qe −
kad t 2.303
(6)
where qt is the amount of fluoride adsorbed at time t (mg g−1 ) and kad is the equilibrium rate constant of pseudo-first-order adsorption (min−1 ). The linearized plots of log(qe − qt ) vs t will give the rate constants. The pseudo-second-order model is also commonly used to predict the kinetic parameters linear for of which can be written as [28] t t 1 = + qt qe h
(7)
and h = kq2e
(8)
where qt is the amount of fluoride adsorbed at time t (mg g−1 ), qe the amount of fluoride adsorbed at equilibrium (mg g−1 ), and h
is is the initial sorption rate (mg g−1 min−1 ). The values of qe (1/slope), k (slope2 /intercept) and h (1/intercept) can be calculated from the plots of t/qt vs t. The linear plot of pseudo-first-order (Eq. (5)) is presented in Fig. 8a. The values of kad , k and h and correlation coefficients obtained from the linear plots are also presented in Table 1. It is apparent from the values of correlation coefficients that fitness of the pseudo-first-order model is better as compared to pseudo-second-order model indicating the applicability of Lagergren equation. Sorption of a liquid adsorbate on porous solid adsorbent can be modelled by pore diffusion models, which can be either particle diffusion or pore diffusion model. The particle diffusion model can be written as ln
C t
Ce
= −kp t
(9)
where kp is the particle diffusion coefficient (mg g−1 min). The value of kp can be obtained by slope of the plot between ln(Ct /Ce ) and t. The intraparticle pore diffusion model given by Weber and Morris [28] is also commonly used to characterize the sorption data. According to this model, if the rate limiting step is diffusion of adsorbate within the pores of adsorbent particle (intraparticle diffusion) a graph between amount of adsorbate adsorbed and square root of time should give a straight line passing through the origin. The equation can be written as qt = ki t 1/2
(10)
where ki is the intraparticle diffusion coefficient (mg g−1 min0.5 ), which can be obtained from the slope of plot of qt verses t1/2 . The plots of linear forms of particle diffusion and intraparticle pore diffusion models are given in Fig. 8b and c, respectively and the values of different parameters are given in Table 1. The
Fig. 8. (a) Lagergren plot, (b) particle diffusion plot, and (c) Weber–Morris plot for the fluoride removal by LCB at different initial concentrations (volume: 500 ml, temperature: 30 ± 1 ◦ C, and adsorbent dose: 0.4 g l−1 ).
values of R2 for particle diffusion model are closer to unity indicating that particle diffusion of adsorbate is contributing more towards rate determining step. However, in case of intraparticle diffusion model the lines are not passing through the origin, which reveals that the adsorption of fluoride on LCB-10 is a complex process involving surface adsorption, interparticle diffusion and intraparticle diffusion all contributing towards the rate of sorption.
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
223
Table 1 Lagergren, pseudo-second-order kinetic parameters, particle and intraparticle pore diffusion model parameter for different initial F concentrations. Co (mg l−1 )
Lagergren parameters −1
kad (min 5.27 (∼5) 9.75 (∼10) 14.8 (∼15)
0.085 0.081 0.079
)
2
R
0.998 0.995 0.988
Pseudo-second-order parameters k (g mg
−1
min)
−3
4.3 × 10 3.9 × 10−3 4.5 × 10−3
h (mg g
−1
min)
0.58 0.44 0.19
Particle diffusion model −1)
Intraparticle pore diffusion model
R
kp (min
R
kt (g mg−1 min0.5 )
R2
0.994 0.989 0.978
0.252 0.259 0.182
0.997 0.990 0.993
0.67 1.29 1.51
0.932 0.939 0.958
2
2
Volume: 500 ml, temperature: 30 ± 1 ◦ C, and adsorbent dose: 0.4 g l−1 .
6. Thermodynamic parameters
Chit-MOH + F− ⇔ Chit-MF + OH−
Thermodynamic parameters associated with the adsorption process, viz., standard free energy change (G◦ ), standard enthalpy change (H◦ ) and standard entropy change (S◦ ) were calculated using the following equations. The free energy of adsorption process is given by the equation:
where M represents La. At neutral pH reaction (16) is more prevalent in which hydroxyl group attached to lanthanum is then replaced with fluoride ions through exchange mechanism [32]. At alkaline pH, significant decline in the fluoride sorption capacity was observed, which may be due to the higher concentration of OH− ions resulting in the competition with fluoride for active sites on the adsorbent.
G◦ = −RT ln Ko
(11)
where G◦ is the free energy of sorption (kJ mol−1 ), T is the temperature in Kelvin, R is the universal gas constant (8.314 J mol−1 K−1 ) and Ko is the sorption equilibrium constant. The sorption equilibrium constant Ko for the sorption reaction was determined from the slope of the plot of ln(qe /Ce ) vs Ce at different temperatures and extrapolating to zero Ce according to method suggested by Khan and Singh [29,30]. The sorption equilibrium constant Ko may be expressed in terms of enthalpy change (H◦ ) and entropy change (S◦ ) as a function of temperature as shown below: ln Ko =
H ◦ S ◦ + RT RT
(12)
where H◦ is the heat of sorption (kJ mol−1 ) and S◦ is the standard entropy change (kJ mol−1 K). The values of H◦ and S◦ can be obtained from the slope and intercept of a plot of ln Ko vs 1/T and are found to be 10.48 kJ mol−1 and 0.11 kJ mol−1 K−1 , respectively. The negative values of G◦ −23.03, −24.15 and −25.24 (kJ mol−1 ) at 303 K, 313 K and 323 K, respectively and positive value S◦ suggest the spontaneous nature of fluoride adsorption. Also the positive value of H◦ suggests that the process of fluoride adsorption is endothermic in nature. 7. Mechanism of fluoride adsorption on LCB It has been reported in the literature that chitosan has the highest chelate forming ability as compared to other natural polymer. The nitrogen in the amino group of the chitosan acts as an electron donor and is presumably responsible for selective chelation with metal ion. The presence of the amino group in chitosan was confirmed by the FTIR analysis. [25] Theoretically upon hydroxylation, metal ions attached to the chitosan complete their coordination shell with OH group [31]. These OH group can bind or release proton depending on the initial solution pH, resulting in the development of surface charges. Under acidic condition, more positively charged sites develop resulting in the enhanced fluoride sorption capacity, which is evident from the results that maximum fluoride sorption capacity is observed at pH 5. At more acidic pH, significant decline in the fluoride sorption capacity was observed, which may be due to the formation of hydrofluoric acid. Following reaction mechanism is possible at acidic pH: Chit-MOH + H+ ⇔ Chit-MOH+ 2 +
Chit-MOH
2
−
+ F ⇔ Chit-MF + H2 O
(13) (14)
Total reaction can be written as Chit-MOH + H+ + F− ⇔ Chit-MF + H2 O
(15)
(16)
8. Reuse and regeneration Any adsorbent is economically viable if the adsorbent can be regenerated and reused for many cycles of operation. The studies on the effect of initial fluoride concentration indicate that the adsorbent has not reached saturation even when allowed to react with higher fluoride concentrations. To study the reusability of the LCB-10 adsorbent, the adsorption studies were conducted using the used adsorbent dried in oven at 70 ◦ C. The results show that fluoride removal decreases with successive reuse and reached to ∼4% after third reuse. This indicates exhaustion of fluoride removal capacity of material. After complete saturation, the fluoride loaded material was subjected for regeneration using different regeneration media like ammonium hydroxide, ammonium chloride and sodium chloride. It appears from the results that 1 M ammonium chloride shows a high regeneration capacity of 81.22% as compared to other regeneration media. Further, the regeneration studies in order to achieve 90–100% regeneration are in progress. 9. Field trials Considering the practical applicability of LCB-10 for defluoridation in actual field conditions, the LCB-10 was also tested with fluoride-contaminated ground water collected from fluoride affected regions of Dhar district of Madhya Pradesh, India. The untreated water had fluoride concentration of 10.2 mg l−1 . The detailed characteristics of untreated and treated ground water are presented in Table 2. It is evident from the results that LCB-10 effectively removes fluoride from field ground water and reduces the fluoride concentration to 0.24 mg l−1 with negligible release of lanthanum ion (0.067 mg l−1 ), which is well below the limits of Indian and WHO drinking water standards [34,35]. Other physico-chemical parameters of the treated water are also within permissible limits except which is marginally above the desirable limits of Indian drinking water standard but are within permissible limits in the absence of alternate sources [34]. Comparison of the fluoride removal by LCB-10 from field water and synthetic water shows that there is no significant difference between the fluoride uptake from field water and simulated water indicating LCB-10 can be exploited for defluoridation in actual field conditions, both at household and community level. The mechanism of fluoride adsorption through La incorporated chitosan appears to be through ligand exchange between fluoride ion and hydroxide ion coordinated with La (III) ion complexed with chitosan and has been discussed in detailed in our previous publication [33].
224
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
Table 2 Physico-chemical parameters of untreated and treated filed water. Parameter
Untreated sample
Treated with LCB-10
Turbidity (NTU) pH Total hardness as CaCO3 (mg l−1 ) Alkalinity as CaCO3 (mg l−1 ) Total dissolved solids (mg l−1 ) Chlorides (mg l−1 ) Fluorides (mg l−1 ) Nitrates (mg l−1 ) Sulphate (mg l−1 ) Iron (mg l−1 ) Cadmium (mg l−1 ) Chromium (mg l−1 ) Copper (mg l−1 ) Lead (mg l−1 ) Manganese (mg l−1 ) Zinc (mg l−1 ) Arsenic (mg l−1 ) Aluminium (mg l−1 ) Lanthanum (mg l−1 )
<1 7 60 37 477 292 10.2 <0.01 101 <0.01 <0.002 <0.002 <0.01 <0.002 0.01 0.25 0.008 0.023 0.001
8 5.99 80 261 723 262.9 0.24 0.7 25 <0.01 <0.002 <0.002 <0.01 <0.002 0.046 <0.005 0.007 <0.003 0.067
10. Conclusion The LCB-10 was synthesised for defluoridation by optimization of various synthesis parameters. These beads are showing excellent fluoride removal efficiency of 97% at pH 5 and overcome the drawbacks associated with the conventional adsorbents. The LCB-10 not only has much higher adsorption capacity 4.7 mg g−1 than simple lanthanum-modified chitosan reported in our previous publication [33] but also has numerous advantages namely high chemical and mechanical stability, high resistance to attrition, negligible lanthanum release, suitability for column applications, etc. Equilibrium data fitted well to Langmuir adsorption isotherm. Moreover the kinetics of fluoride adsorption on LCB-10 are very fast. Thermodynamic study reveals that fluoride adsorption using LCB-10 was spontaneous and endothermic in nature. It is possible to regenerate the adsorbent using 1 M ammonium chloride showing high regeneration efficiency of 81.22%. The LCB-10 has shown excellent removal efficiency both in simulated as well as field water collected from fluoride endemic regions, which proves its suitability for defluoridation of drinking water in actual field conditions both at household and community levels. The treated water quality also confirms to Indian and WHO drinking water standards. Acknowledgements Authors thankfully acknowledge CSIR, New Delhi for granting SRF to carry out the research work and constant support and guidance of Dr. T. Chakrabarty, Director NEERI. Financial assistance from UNIECF Bhopal, India for this work is gratefully acknowledged. References [1] S. Ayoob, A.K. Gupta, Fluoride in drinking water: a review on the status and stress effects, Crit. Rev. Environ. Sci. Technol. 36 (2006) 433–487. [2] S.P. Teotia, M. Teotia, R.K. Singh, Hydrogeochemical aspects of endemic skeletal fluorosis in India—an epidemiological study, Fluoride 14 (1981) 69–74. [3] A.K. Susheela, Fluorosis management programme in India, Curr. Sci. 77 (1999) 1250–1256.
[4] A.K. Yadav, C.P. Kaushik, A.K. Haritash, A. Kansal, R. Neetu, Defluoridation of groundwater using brick powder as an adsorbent, J. Hazard. Mater. 128 (2006) 289–293. [5] W.G. Nawlakhe, D.N. Kulkarni, B.N. Pathak, K.R. Bulusu, Defluoridation of water by Nalgonda technique, Ind. J. Environ. Health 17 (1975) 26–65. [6] H. Mjengera, G. Mkongo, Appropriate defluoridation technology for use in fluorotic areas in Tanzania, Phys. Chem. Earth 28 (2003) 1097–1104. [7] A. Lhassani, M. Rumeau, D. Benjelloun, M. Pontie, Selective demineralisation of water by nanofiltration application to the defluoridation of brackish water, Water Res. 35 (2001) 3260–3264. [8] X. Feng Shen, P.G. Chen, C. Guohua, Electrochemical removal of fluoride ions from industrial wastewater, Chem. Eng. Sci. 58 (2003) 987–993. [9] N.I. Chubar, V.F. Samanidou, V.S. Kouts, G.G. Gallios, V.A. Kanibolotsky, V.V. Strelko, I.Z. Zhuravlev, Adsorption of fluoride, chloride, bromide and bromate ions on a novel ion exchanger, J. Colloid Interface Sci. 291 (2005) 67–74. [10] S. Ghorai, K.K. Pant, Equilibrium, kinetics and breakthrough studies for adsorption of fluoride on activated alumina, Sep. Purif. Technol. 42 (2005) 265–271. [11] D. Mohapatra, D. Mishra, S.P. Mishra, G. Roy Chaudhury, R.P. Das, Use of oxide minerals to abate fluoride from water, J. Colloid Interface Sci. 275 (2004) 355–359. [12] M. Sarkar, A. Banerjee, P.P. Pramanick, A.R. Sarkar, Use of laterite for the removal of fluoride from contaminated drinking water, J. Colloid Interface Sci. 302 (2006) 432–441. [13] A. Tor, Removal of fluoride from an aqueous solution by using montmorillonite, Desalination 201 (2006) 267–276. [14] P.P. Coetzeel, L.L. Coetzee, R. Pukal, S. Mubenga, Characterisation of selected South African clays for defluoridation of natural waters, Water SA 29 (2003) 331–338. [15] X. Fan, D.J. Parker, M.D. Smith, Adsorption kinetics of fluoride on low cost materials, Water Res. 37 (2003) 4929–4937. [16] R.L. Ramos, J. Ovalle-Turrubiartes, M.A. Sanchez-Castillo, Adsorption of fluoride on aluminium impregnated carbon, Carbon 37 (1999) 609–617. [17] M.S. Onyango, Y. Kojima, O. Aoyi, E.C. Bernardo, H. Matsuda, Adsorption equilibrium modeling and solution chemistry dependence of fluoride removal from water by trivalent-cation-exchanged zeolite F-9, J. Colloid Interface Sci. 279 (2004) 341–350. [18] S. Chidambaram, A.L. Ramanathan, S. Vasudevan, Fluoride removal studies in water using natural materials, Water SA 29 (2003) 339–344. [19] N.V. Majeti, R. Kumar, A review of chitin and chitosan applications, React. Funct. Polym. 46 (2000) 1–27. [20] A.J. Varmaa, S.V. Deshpandea, J.F. Kennedy, Metal complexation by chitosan and its derivatives: a review, Carbohydr. Polym. 55 (2004) 77–93. [21] W. Ma, Y. Fie-Qun, M. Han, R. Wang, Characteristics of equilibrium, kinetics studies for adsorption of fluoride on magnetic-chitosan particle, J. Hazard. Mater. 143 (2007) 296–302. [22] S. Annouar, M. Mountada, A. Soufiane, A. Elmidaoui, M.A. Menkouchi Sahli, Defluoridation of underground water by adsorption on the chitosan and by electrodialysis, Desalination 165 (2004) 437–438. [23] Li. Nan, R. Bai, Copper adsorption on chitosan–cellulose hydrogel beads: behaviors and mechanisms, Sep. Purif. Technol. 42 (2005) 237–247. [24] Standard methods for the examination of water and wastewater, in: APHAAWWA-WPCF, 20th edition, 1998. [25] A.T. Pauline, J.I. Simionato, J.C. Garcia, J. Nozaki, Characterization of chitosan and chitin produced from silkworm chrysalides, Carbohydr. Polym. 64 (2006) 98–103. [26] A.J. Varma, S.V. Deshpande, J.F. Kennedy, Metal complexation by chitosan and its derivative: a review, Carbohydr. Polym. 55 (2004) 77–93. [27] D. Berger, N. van Landschoot, C. Ionica, F. Papa, V. Fruth, Synthesis of pure and doped lanthanum cobaltite by the combustion method, J. Optoelectron. Adv. Mater. 5 (2003) 719–724. [28] S. Meenakshi, N. Viswanathan, Identification of selective ion-exchange resin for fluoride sorption, J. Colloid Interface Sci. 308 (2007) 438–450. [29] S. Meenakshi, C. Sairam Sundaram, R. Sukumar, Enhanced fluoride sorption by mechanochemically activated kaolinites, J. Hazard. Mater. 153 (2008) 164–172. [30] A.A. Khan, R.P. Singh, Adsorption thermodynamics of carbofuran on Sn (IV) arsenosilicate in H+ , Na+ and Ca2+ Forms, Colloid Surf. 24 (1987) 33–42. [31] S. Ayoob, A.K. Gupta, P.B. Bhakat, V.T. Bhat, Investigation on the kinetics and mechanisms of sorptive removal of fluoride from water using alumina cement granule, Chem. Eng. J. 140 (2008) 6–14. [32] S. Jagtap, D. Thakre, S. Wanjari, S. Kamble, N. Labhsetwar, S. Rayalu, New modified chitosan based adsorbent for defluoridation of water, J. Colloid Interface Sci. 332 (2009) 280–290. [33] S. Kamble, S. Jagtap, N. Labhsetwar, D. Thakre, S. Godfrey, S. Devotta, S.S. Rayalu, Defluoridation of drinking water using chitin, chitosan and lanthanummodified chitosan, Chem. Eng. J. 129 (2007) 173–180. [34] Indian Standard Specifications for Drinking Water IS: 10500, 1991. [35] Guidelines for Drinking-water Quality, WHO, 1993.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Fluoride removal using lanthanum incorporated chitosan beads Amit Bansiwal a,1 , Dilip Thakre a,1 , Nitin Labhshetwar a , Siddharth Meshram b , Sadhana Rayalu a,∗ a b
National Environmental Engineering Research Institute, India Department of chemistry, Laxminarayan Institute of Technology (LIT), RTM Nagpur University, Nagpur, India
a r t i c l e
i n f o
Article history: Received 13 April 2009 Received in revised form 13 July 2009 Accepted 15 July 2009 Available online 23 July 2009 Keywords: Fluoride removal Chitosan beads Thermodynamic parameter Mechanism Adsorption isotherm Kinetic
a b s t r a c t Highly selective material based on naturally occurring biomaterial namely chitosan has been designed for the defluoridation of water. Lanthanum incorporated chitosan beads (LCB) were prepared using precipitation method. The synthesis was optimized by varying different synthesis parameters namely lanthanum loading, complexation and precipitation time, strength of ammonia solution used for precipitation, drying time, etc. Lanthanum incorporated chitosan beads were characterized using SEM, FTIR, XRD and EDX. Surface area of LCB was observed to be 2.76 m2 g−1 . The equilibrium adsorption data fitted well to Langmuir adsorption isotherm and showing maximum fluoride adsorption capacity of 4.7 mg g−1 with negligible lanthanum release. Kinetic study reveals that adsorption of fluoride is fast and follows pseudo-first-order kinetics. The effect of pH was also studied and the best efficiency was observed at pH 5. Presence of sulphate, nitrate and chloride marginally affected the removal efficiency, however drastic reduction in fluoride uptake was observed in the presence of carbonate and bicarbonate. Negative value of change in free energy (G◦ ) and positive value of change in entropy (S◦ ) suggest the adsorption of fluoride by LCB is feasible and spontaneous process. Positive value of change in enthalpy (H◦ ) suggests the process of fluoride adsorption is endothermic in nature. Regeneration study reveals that 1 M ammonium chloride solution appears to be the promising regeneration media showing 81.22% regeneration. The adsorption capacity of LCB was similar in fluoride-contaminated ground water collected from Dhar district of Madhya Pradesh, India, as compared to simulated water. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Fluoride is one of the most abundant constituents occurring in groundwater worldwide and creates a major problem in safe drinking water supply. Widespread occurrence of fluoride above the prescribed limit in groundwater meant for human consumption has caused multidimensional health problems. The problem of excess fluoride in drinking waters is aggravating day by day, as more surveys to assess the ground water quality have been undertaken. Fluoride in water derives mainly from dissolution of natural minerals in the rocks and soils with which water interacts. The distribution and health effects of fluoride have recently been reviewed by Ayoob and Gupta [1]. According to authors, about 200 million peoples from 25 nations are facing the problem of excess fluoride in drinking water among which India and China are the worst affected. In India alone, endemic fluorosis is thought to affect around 1 million people [2] and is a major problem in 17 out of the country’s
∗ Corresponding author at: Environmental Materials Unit, National Environmental Engineering Research Institute, Nehru Marg, Nagpur 440020, India. Tel.: +91 7122247828; fax: +91 7122247828. E-mail address: s [email protected] (S. Rayalu). 1 Both authors have contributed equally. 0927-7765/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2009.07.021
22 states, especially Rajasthan, Madhya Pradesh, Andhra Pradesh, Tamil Nadu, Gujarat and Uttar Pradesh [3]. We have also conducted exhaustive water sampling in Dhar district of Madhya Pradesh, India to assess the distribution of fluoride in ground water. Fluoride concentrations in large number of samples fall above 1 mg l−1 (WHO prescribed limit for drinking water) and the maximum concentration found was 12.0 mg l−1 . The most important remedial action is the prevention of further exposure by providing safe drinking water. However in most of the areas source substitution may be impossible due to non-availability of alternate sources and therefore removal of excess fluoride is the only remedy. Several defluoridation techniques namely precipitation [4–6], membrane processes [7] electrolytic [8] ion-exchange [9] and adsorption onto various adsorbents have been evaluated both in field and lab. Owing to various drawbacks associated with conventional defluoridation techniques adsorption-based fluoride removal systems are emerging as the feasible option. Large numbers of adsorbents have been studied for the removal of fluoride from water based on alumina [10], other minerals and metal oxides [11–13], clays [13–15], carbons [16], zeolites [17] industrial/agricultural wastes [18], etc. Chitin is one of the most abundant biopolymer and it is estimated to be produced annually almost as much as cellulose. It has become of great interest not only as an under-utilized resource, but
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
217
2. Materials and methods Nomenclature qe C0 Ce V W Xm KL r Kf n k h qt kp G◦ H◦ S◦ Ko
equilibrium adsorbate capacity (mg g−1 ) initial concentrations of fluoride (mg l−1 ) equilibrium concentrations of fluoride (mg l−1 ) volume of the aqueous solution (l) mass of adsorbent (g) amount of adsorbate at complete monolayer coverage (mg g−1 ) Langmuir constant (affinity) 1 mg−1 dimensionless quantity Freundlich constants related to adsorption capacity (mg g−1 ) adsorption intensity (heterogeneity factor) second-order rate constant (min −1 ) initial sorption rate (mg g−1 min−1 ) amount of fluoride adsorbed per unit mass of adsorbent at time t (mg g−1 ) intraparticle diffusion rate constant (mg g−1 min−1/2 ) standard free energy change (kJ mol−1 ) standard enthalpy change (kJ mol−1 ) standard entropy change (kJ mol−1 ) sorption equilibrium constant
also as a new functional material of high potential in various fields. Chitosan is the N-deacetylated derivative of chitin, which results in higher nitrogen content and hence is a better chelating agent. Majeti and Kumar have reviewed the wide array of applications of chitin and chitosan in photography, cosmetics, environmental, biomedical, ophthalmology, food and nutrition, paper finishing, solid-state batteries, drug delivery systems, and also extensive biotechnological applications [19]. Extensive reports are also available on application of chitosan for water and wastewater treatment, particularly for the removal of heavy metals, since it has excellent chelating properties [20]. Extensive reports are available on defluoridation of water using chitin/chitosan. Ma et al. have reported magnetic chitosan particle for the removal of fluoride from water in batch mode [21]. Annouar et al. have also reported defluoridation of drinking water through adsorption of fluoride on chitosan followed by electrodialysis. However, there are only a few reports available on functionalization of chitosan using electropositive metals. In recent publication of our group we have reported defluoridation using chitin, chitosan and lanthanum-modified chitosan [22]. However, chitosan naturally occurs in the form of flakes or powder which are having limited utility particularly for column applications due to swelling, low mechanical strength, crumbling, etc. Attempts have also been made to overcome these drawbacks through formation of chitosan beads through various routes [23]. The present paper deals with the synthesis of lanthanum incorporated chitosan beads (LCB) and its application for the removal of fluoride from ground water. The synthesis was optimized by varying different synthesis parameters namely lanthanum loading, complexation and precipitation time, strength of ammonia solution used for precipitation, drying time, etc. The effect of various parameters on fluoride uptake capacity namely pH, adsorbent dose, initial fluoride concentration, presence of other anions and cations was also studied. Evaluation of LCB for the removal of fluoride from ground water collected from fluoride endemic regions of Dhar district of Madhya Pradesh, India was also carried out. Detailed physico-chemical analysis of the untreated and treated water was done to assess the potability of treated water.
2.1. Materials All the chemicals used throughout the study were of analytical grades and were prepared in double distilled water. 85% deacetylated chitosan was purchased from Chemchito Natural Products, Chennai, India. Lanthanum acetate was purchased from Himedia, Mumbai and sodium fluoride, sodium chloride, sodium bicarbonate, sodium nitrate, aqueous ammonia and acetic acid were procured from E. Merck India Ltd. Stock solution of 1000 mg l−1 fluoride was prepared from dried sodium fluoride and working fluoride solutions of required concentrations were prepared by diluting the stock solution.
2.2. Synthesis of LCB For the synthesis of LCB, 9 g of chitosan (85% deacetylated) was dissolved in 300 ml of acetic acid (CH3 COOH) solution (5%, v/v). In another beaker required quantities of lanthanum acetate was dissolved in 100 ml of distilled water. The lanthanum acetate solution was then added to the chitosan solution with stirring for 180 min. The lanthanum loading was varied from 1 to 20 wt%. The resulting La–chitosan solution was added drop wise into NH4 OH solution (50%, v/v) under vigorous stirring, using a syringe pump. The gel macro-spheres formed were allowed to stabilize in NH4 OH solution for 60 min. The beads were separated from the NH4 OH solution and washed with deionised water and dried at 50–80 ◦ C for 24 h in oven.
2.3. Batch adsorption studies Batch adsorption studies were carried out as follows: 50 ml of fluoride solution was mixed with different quantities of the adsorbent in PVC conical flasks. The flasks were kept on orbital shaker for 24 h at 30 ± 2 ◦ C to attain the equilibrium. The solution was then filtered and the fluoride concentrations were determined in filtrates and the amount of fluoride adsorbed was calculated from following equation: qe = (C0 − Ce ) ×
V W
(1)
where qe is the adsorption capacity (mg g−1 ) at equilibrium, C0 and Ce are the initial and equilibrium fluoride concentrations (mg l−1 ), respectively, V is the volume (ml) of solution and W is the mass (g) of adsorbent used. Besides determining the defluoridation capacities, the effects of other factors namely adsorbent dose, initial fluoride concentration, pH of the medium, presence of interfering ions normally present in drinking water were also studied. The effect of solution pH on fluoride uptake was studied by maintaining the pH by the addition of 0.1N HCl or 0.1N NaOH. The pH was varied from 3 to 9. The effects of various anions namely chloride, sulphate, nitrate, carbonate and bicarbonate were studied at optimal experimental conditions. Reusability studies of the LCB were also conducted to assess the cyclic use of the adsorbent. To study the reuse of adsorbent, required quantity of adsorbent was mixed with fluoride solution with initial concentration of 5 mg l−1 . After equilibrium, the adsorbent was separated by filtration and dried in air and was repeatedly used for fluoride adsorption to study the degree of reuse. Comparison of fluoride uptake in fluoride solution prepared in distilled water henceforth referred to as simulated water and fluoride-contaminated field water collected from Dhar district of Madhya Pradesh, India was also studied.
218
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
2.4. Methods of analysis Fluoride analysis was done by ion selective electrode method using ion meter (Thermoelectron, USA) coupled with fluoride ion selective electrode (Thermoelectron Model 9609 BNWP). TISAB III was used as buffer for maintaining the pH and background ion concentrations during measurement. pH was measured using the same ion meter coupled with pH electrode. The concentrations of various co-ions and other physico-chemical parameters of treated and untreated water were determined using standard methods [24]. Carbonate and bicarbonate alkalinity, total hardness and chlorides were determined titrimetrically, whereas the concentrations of nitrate and sulphate were determined spectrophotometrically using UV visible spectrophotometer (Chemito, Model UV-2100). The amount of lanthanum if any leached from adsorbent and other heavy metals was determined using ICP-OES (PerkinElmer, Model Optima 4100DV). Blank experiments were conducted through-
out the studies and majority of the experiments were repeated twice and it was observed that the experimental error was within ±2%. 2.5. Physical characterization The powder X-ray diffraction pattern of LCB before and after fluoride adsorption was recorded on Rigaku X-ray diffractometer in order study the structure of the adsorbent. The surface morphology of chitosan and LCB was observed using scanning electron microscopy (Jeol, JXA-840 A, Electron probe microanalyser, Japan). The elemental composition of LCB was determined by Energy Dispersive X-ray analysis. In order to determine the presence of functional groups in adsorbents the FTIR spectra of chitosan and LCB before and after fluoride adsorption in KBr pellets were recorded on Bruker, Model Vertex 70 spectrometer. The surface area of LCB was determined by BET surface area method.
Fig. 1. FTIR of: (a) chitosan, (b) LCB before adsorption, and (c) LCB after adsorption.
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
3. Results and discussion 3.1. Characterization of LCB 3.1.1. FTIR of LCB Fig. 1 shows the FTIR spectra of chitosan and lanthanummodified chitosan. From the FTIR spectra of chitosan it is observed that the band at 1571 cm−1 has larger intensity than the band at 1676 cm−1 suggesting effective deacetylation [25]. When chitin deacetylation occurs, the band at 1676 cm−1 decreases while the growth at 1571 cm−1 occurs, which indicates the presence of –NH2 group. The intensity of the band at 1571 cm−1 in chitosan was significantly reduced in LCB, which may be due to the coordination of lanthanum with –NH2 group. The coordination of chitosan with La+3 showed the involvement of amine and secondary hydroxyl group of chitosan in chelate formation. The band at 3695 cm−1 in chitosan is due to the stretching vibration of –NH group, which is shifted to 3610 cm−1 in LCB, in the 1571 cm−1 band to 1549 cm−1 and shift in 1059 cm−1 band which is due to secondary hydroxyl group to 1028 cm−1 supports the involvement of amine and secondary hydroxyl group in chelate formation [26]. The bands at 2924 cm−1 and 1376 cm−1 in chitosan and 2879 cm−1 and 1369 cm−1 in LCB were attributed to C–H stretching vibration in polymeric backbone and C–H bending, respectively. The band at 3294 cm−1 in chitosan is attributed to stretching vibration of hydroxyl group, which is shifted to 3056 cm−1 in LCB. The band at 1432 cm−1 in LCB can be assigned to the stretching vibration of C–N group of chitosan template. Also, the FTIR spectra of lanthanummodified chitosan show some new peaks at 943 cm−1 , 488 cm−1 ,
219
and 422 cm−1 . The band at 422 cm−1 can be assigned to vibration of La–O bond [27]. From the FTIR spectra of LCB after fluoride adsorption the loss of intensity was observed at 3610 cm−1 (–NH) in LCB before fluoride adsorption which may be due to the interaction of fluoride with the lanthanum coordinated to –NH group of chitosan. 3.1.2. Scanning electron microscopy of LCB The SEM micrographs of chitosan flakes and LCB are shown in Fig. 2. SEM micrograph of chitosan flakes Fig. 2a shows rough surface morphology. SEM micrograph of LCB (Fig. 2b) reveals dense and firm structure with minimum porosity and small oval shaped particles of lanthanum of diameter 2.5–3.0 m and length 4.0–5.0 m uniformly spread over the surface of chitosan. The surface area of chitosan was found to be 2.76 m2 g−1 . 3.1.3. EDX of LCB An EDX spectrum of lanthanum-modified chitosan and its percentage composition is given in Fig. 2c. The EDX spectra of LCB indicate the presence of lanthanum particles over the surface of chitosan. 3.1.4. XRD of LCB The X-ray diffraction pattern of chitosan and lanthanummodified chitosan are shown in Fig. 3. The X-ray pattern of chitosan shows the characteristic crystalline peak at 2 = 8◦ , 20◦ , 29◦ respectively, while these peaks (2 = 20.46◦ ) are less intensive in LCB which suggest that chitosan has a higher crystallinity than its complex. So the X-ray spectrum of LCB shows more amorphous nature, which allows a better accessibility to fluoride and thus a better
Fig. 2. SEM images of: (a) chitosan flakes, (b) LCB, and (c) EDX of LCB.
220
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
activity. The new peaks at 2 = 30◦ , 43.35◦ and 49.78◦ are observed which can be assigned to the lanthanum hydroxide formation. 3.2. Effect of lanthanum loading The optimization of synthesis of LCB has been done by varying the concentration of lanthanum from 1 to 22 (%) in order to study the effect of lanthanum loading on fluoride removal. Increase in lanthanum loading from 1 to 10 (%) resulted in increase percent removal of fluoride from 68 to 98 (%). A maximum of 98% of fluoride removal was achieved at the 10% lanthanum loading on Chitosan. Further increase in the lanthanum loading above 10% has no significant improvement in the fluoride removal capacity. So 10% lanthanum loading was optimal for the synthesis of LCB. 3.3. Preliminary adsorption experiment Preliminary adsorption experiment was carried out using bare chitosan and lanthanum-modified chitosan (LCB) at an initial concentration of 5.35 mg l−1 ; contact time 24 h to check the fluoride removal efficiency of both the adsorbent. The lanthanum-modified chitosan shows much higher fluoride removal efficiency as compared to bare chitosan. Therefore further adsorption studies were carried out using lanthanum-modified chitosan. 3.4. Effect of adsorbent dose The effect of adsorbent dose on fluoride removal at fixed initial concentration (5.34 mg l−1 ) is shown in Fig. 4. It was observed that the percent removal of fluoride increased with increase in the adsorbent dose while loading capacity (amount of fluoride adsorb per gram of the adsorbent) gradually decreases for the same. From the trend it appears that a maximum level (53–90%) of fluoride removal occurs with increase in dose from 0.2 to 0.4 g l−1 .
Fig. 4. Effect of adsorbent dose on fluoride removal by LCB-10 (Initial F concentration: 5.34 mg l−1 , and contact time: 24 h).
Thereafter, there was no significant improvement in the fluoride removal efficiency with increase of adsorbent quantity that may be due to low concentration of fluoride available at higher adsorbent dose. Also, negligible leaching of lanthanum ion 0.06–0.08 mg l−1 is observed with increase in adsorbent dose from 0.2 to 0.4 g l−1 . Therefore 0.4 g l−1 can be considered as optimum dose for further studies. 3.5. Effect of initial fluoride concentration The fluoride concentration in groundwater varies significantly due to seasonal, temporal and geological variations. The defluoridation technique should be able to remove fluoride from water at different initial concentrations. Keeping this in view the effect of initial fluoride concentration was studied. As expected the fluoride removal increases with increase in initial fluoride concentration which may be due to more availability of adsorbate and the active sites of the adsorbent have not reached saturation even at initial fluoride concentrations more than 20 mg l−1 . 3.6. Effect of pH Removal of fluoride through adsorption is highly dependent on solution pH as it can alter the surface charge of the adsorbent. Hence, in the present study the fluoride removal was studied at four different pH namely 3, 5, 7 and 9 and the results are shown in Fig. 5. It is apparent from the results that the pH has significant effect on sorption capacity and the maximum sorption capacity was observed at pH 5. Decrease in sorption capacities was observed below and above 5, which may be attributed to unfavourable surface charges and competition for adsorption sites due to excess anions at alkaline conditions. Above treated water samples at different pH were analysed on ICP in order to check the leaching of lanthanum ion from adsorbent. It revealed from the result that there is negligible leaching of lanthanum ions 0.084, 0.082 and 0.084 mg l−1 at pH 5, 7 and 9, respectively. Whereas significant lanthanum leaching from adsorbent is observed at pH 3 which may be due to the partial dissolution of biopolymer matrix at acidic pH. 3.7. Effect of co-anions
Fig. 3. XRD of: (a) chitosan flakes and (b) lanthanum-modified chitosan.
Beside fluoride the natural ground water always contains various other co-ions, which can compete for sorption sites and result in reduced efficiency of the adsorbent. To study the effect of coanions the fluoride uptake by LCB was studied in the presence of various anions namely like HCO3− , CO3 2− , Cl− , NO3− , and SO4 2−
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
221
or Ce 1 1 = + Ce qe Xm KL Xm
(3)
where qe is the amount of adsorbate adsorbed per unit weight of adsorbent at equilibrium (equilibrium uptake) (mg g−1 ) calculated using Eq. (1), Xm is the maximum adsorption capacity (mg g−1 ), KL is the Langmuir constant, and Ce is the equilibrium concentration of adsorbate in solution (mg l−1 ). The Freundlich adsorption equation is as follows: 1/n
qe = Kf Ce
log(qe ) = log k +
Fig. 5. Effect of pH on fluoride removal by LCB-10 (volume: 50 ml, temperature: 30 ± 1 ◦ C, adsorbent dose: 0.4 g l−1 , contact time: 24 h, and initial F concentration: 5.34 mg l−1 ).
and also cations like Na, Ca, Cu and Fe keeping initial fluoride concentration as 5 mg l−1 . As shown in Fig. 6 the removal efficiency reduces slightly in the presence of anions namely Cl− , NO3− , and SO4 2− whereas drastic reduction in the presence of HCO3− and CO3 2− was observed. The presence of HCO3− and CO3 2− increases the pH with consequent decrease in adsorption capacity. Also these anions appear to be the main competitors for active sites and have more affinity as compared to fluoride. As explained earlier, the removal efficiency reduces to a great extent in alkaline conditions, the observed trend of reduction in removal efficiency in the presence of HCO3− and CO3 2− is explainable since the presence of HCO3− and CO3 2− leads to alkaline conditions. 4. Equilibrium modelling The Langmuir and Freundlich adsorption models were used to obtain various adsorption parameters and to get an insight into the adsorption mechanism, The Langmuir adsorption equation is represented as follows: qe =
Xm KL Ce 1 + KL Ce
(4) 1 log(Ce ) n
(5)
where qe is the amount of adsorbate adsorbed per unit weight of adsorbent at equilibrium (equilibrium uptake) (mg g−1 ), Kf is the Freundlich constant, n = a Freundlich constant, which reflects adsorption intensity, Ce is the equilibrium concentration of adsorbate in solution (mg l−1 ). Results of experimental data were compared to the linear forms of the isotherms to determine which model most accurately described adsorption by the adsorbent. An estimation of the sorption capacity can be obtained from Langmuir constants (Xm and KL ) and the Freundlich constants (Kf and n). The fitness of Langmuir and Freundlich adsorption isotherms with the adsorption data obtained from experimental data is presented in Fig. 7 along with the values of respective correlation coefficients (R2 ). It is apparent from the results that better fit was observed for Langmuir adsorption isotherm, which is also evident from values of R2 , which are 0.96 and 0.93 for Langmuir, and Freundlich isotherms, respectively. Better fit of Langmuir isotherm model indicates the monolayer uniform adsorption. The values of Xm and KL obtained from Langmuir model are 4.7 mg g−1 and 0.023 l g−1 , respectively. Multilayer adsorption on non-uniform sites is the basis of the Freundlich model. The Freundlich model assumes an infinite supply of unreacted adsorbent sites and tends to represent heterogeneous materials better than other models. The higher Freundlich constant (Kf ) characterizes the adsorbent as more reactive, although the constant tends to be site and adsorbent specific. The constant n is an empirical constant and is dependent on the degree of heterogeneity in the adsorbing sites. The values of Kf and n obtained from Freundlich isotherms are 0.106 mg g−1 and 0.864, respectively. Since the value of n is close to unity it reflects favourable adsorption.
(2)
Fig. 6. Effect of different ions on fluoride removal by LCB-10 (volume: 50 ml, temperature: 30 ± 1 ◦ C, adsorbent dose: 0.4 g l−1 , contact time: 24 h, initial F concentration: 5.34 mg l−1 , and concentration of salts: 0.1 M).
Fig. 7. Adsorption isotherms for the removal of fluoride by LCB-10 (volume: 50 ml, temperature: 30 ± 1 ◦ C, initial F concentration: 5.34 mg l−1 , and contact time: 24 h).
222
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
5. Adsorption kinetics The effect of contact time on fluoride uptake at three initial concentrations namely 5.21, 9.75 and 14.8 mg l−1 using LCB was studied. It is apparent that fluoride uptake reached saturation after 70 min at all the three concentrations, indicating fast kinetics. The release of lanthanum ion with respect to time was also observed and it is revealed that there is negligible release of lanthanum ion 0.08–0.084 mg l−1 with the increase in contact time. Two main types of kinetic models frequently used to determine various kinetic parameters of the adsorption system are reaction-based models and diffusion-based models. To study the sorption kinetics pseudo-first-order and pseudo-secondorder kinetic models have been used. A simple pseudo-first-order kinetic model also known as Lagergren equation is represented as [28] log(qe − qt ) = log qe −
kad t 2.303
(6)
where qt is the amount of fluoride adsorbed at time t (mg g−1 ) and kad is the equilibrium rate constant of pseudo-first-order adsorption (min−1 ). The linearized plots of log(qe − qt ) vs t will give the rate constants. The pseudo-second-order model is also commonly used to predict the kinetic parameters linear for of which can be written as [28] t t 1 = + qt qe h
(7)
and h = kq2e
(8)
where qt is the amount of fluoride adsorbed at time t (mg g−1 ), qe the amount of fluoride adsorbed at equilibrium (mg g−1 ), and h
is is the initial sorption rate (mg g−1 min−1 ). The values of qe (1/slope), k (slope2 /intercept) and h (1/intercept) can be calculated from the plots of t/qt vs t. The linear plot of pseudo-first-order (Eq. (5)) is presented in Fig. 8a. The values of kad , k and h and correlation coefficients obtained from the linear plots are also presented in Table 1. It is apparent from the values of correlation coefficients that fitness of the pseudo-first-order model is better as compared to pseudo-second-order model indicating the applicability of Lagergren equation. Sorption of a liquid adsorbate on porous solid adsorbent can be modelled by pore diffusion models, which can be either particle diffusion or pore diffusion model. The particle diffusion model can be written as ln
C t
Ce
= −kp t
(9)
where kp is the particle diffusion coefficient (mg g−1 min). The value of kp can be obtained by slope of the plot between ln(Ct /Ce ) and t. The intraparticle pore diffusion model given by Weber and Morris [28] is also commonly used to characterize the sorption data. According to this model, if the rate limiting step is diffusion of adsorbate within the pores of adsorbent particle (intraparticle diffusion) a graph between amount of adsorbate adsorbed and square root of time should give a straight line passing through the origin. The equation can be written as qt = ki t 1/2
(10)
where ki is the intraparticle diffusion coefficient (mg g−1 min0.5 ), which can be obtained from the slope of plot of qt verses t1/2 . The plots of linear forms of particle diffusion and intraparticle pore diffusion models are given in Fig. 8b and c, respectively and the values of different parameters are given in Table 1. The
Fig. 8. (a) Lagergren plot, (b) particle diffusion plot, and (c) Weber–Morris plot for the fluoride removal by LCB at different initial concentrations (volume: 500 ml, temperature: 30 ± 1 ◦ C, and adsorbent dose: 0.4 g l−1 ).
values of R2 for particle diffusion model are closer to unity indicating that particle diffusion of adsorbate is contributing more towards rate determining step. However, in case of intraparticle diffusion model the lines are not passing through the origin, which reveals that the adsorption of fluoride on LCB-10 is a complex process involving surface adsorption, interparticle diffusion and intraparticle diffusion all contributing towards the rate of sorption.
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
223
Table 1 Lagergren, pseudo-second-order kinetic parameters, particle and intraparticle pore diffusion model parameter for different initial F concentrations. Co (mg l−1 )
Lagergren parameters −1
kad (min 5.27 (∼5) 9.75 (∼10) 14.8 (∼15)
0.085 0.081 0.079
)
2
R
0.998 0.995 0.988
Pseudo-second-order parameters k (g mg
−1
min)
−3
4.3 × 10 3.9 × 10−3 4.5 × 10−3
h (mg g
−1
min)
0.58 0.44 0.19
Particle diffusion model −1)
Intraparticle pore diffusion model
R
kp (min
R
kt (g mg−1 min0.5 )
R2
0.994 0.989 0.978
0.252 0.259 0.182
0.997 0.990 0.993
0.67 1.29 1.51
0.932 0.939 0.958
2
2
Volume: 500 ml, temperature: 30 ± 1 ◦ C, and adsorbent dose: 0.4 g l−1 .
6. Thermodynamic parameters
Chit-MOH + F− ⇔ Chit-MF + OH−
Thermodynamic parameters associated with the adsorption process, viz., standard free energy change (G◦ ), standard enthalpy change (H◦ ) and standard entropy change (S◦ ) were calculated using the following equations. The free energy of adsorption process is given by the equation:
where M represents La. At neutral pH reaction (16) is more prevalent in which hydroxyl group attached to lanthanum is then replaced with fluoride ions through exchange mechanism [32]. At alkaline pH, significant decline in the fluoride sorption capacity was observed, which may be due to the higher concentration of OH− ions resulting in the competition with fluoride for active sites on the adsorbent.
G◦ = −RT ln Ko
(11)
where G◦ is the free energy of sorption (kJ mol−1 ), T is the temperature in Kelvin, R is the universal gas constant (8.314 J mol−1 K−1 ) and Ko is the sorption equilibrium constant. The sorption equilibrium constant Ko for the sorption reaction was determined from the slope of the plot of ln(qe /Ce ) vs Ce at different temperatures and extrapolating to zero Ce according to method suggested by Khan and Singh [29,30]. The sorption equilibrium constant Ko may be expressed in terms of enthalpy change (H◦ ) and entropy change (S◦ ) as a function of temperature as shown below: ln Ko =
H ◦ S ◦ + RT RT
(12)
where H◦ is the heat of sorption (kJ mol−1 ) and S◦ is the standard entropy change (kJ mol−1 K). The values of H◦ and S◦ can be obtained from the slope and intercept of a plot of ln Ko vs 1/T and are found to be 10.48 kJ mol−1 and 0.11 kJ mol−1 K−1 , respectively. The negative values of G◦ −23.03, −24.15 and −25.24 (kJ mol−1 ) at 303 K, 313 K and 323 K, respectively and positive value S◦ suggest the spontaneous nature of fluoride adsorption. Also the positive value of H◦ suggests that the process of fluoride adsorption is endothermic in nature. 7. Mechanism of fluoride adsorption on LCB It has been reported in the literature that chitosan has the highest chelate forming ability as compared to other natural polymer. The nitrogen in the amino group of the chitosan acts as an electron donor and is presumably responsible for selective chelation with metal ion. The presence of the amino group in chitosan was confirmed by the FTIR analysis. [25] Theoretically upon hydroxylation, metal ions attached to the chitosan complete their coordination shell with OH group [31]. These OH group can bind or release proton depending on the initial solution pH, resulting in the development of surface charges. Under acidic condition, more positively charged sites develop resulting in the enhanced fluoride sorption capacity, which is evident from the results that maximum fluoride sorption capacity is observed at pH 5. At more acidic pH, significant decline in the fluoride sorption capacity was observed, which may be due to the formation of hydrofluoric acid. Following reaction mechanism is possible at acidic pH: Chit-MOH + H+ ⇔ Chit-MOH+ 2 +
Chit-MOH
2
−
+ F ⇔ Chit-MF + H2 O
(13) (14)
Total reaction can be written as Chit-MOH + H+ + F− ⇔ Chit-MF + H2 O
(15)
(16)
8. Reuse and regeneration Any adsorbent is economically viable if the adsorbent can be regenerated and reused for many cycles of operation. The studies on the effect of initial fluoride concentration indicate that the adsorbent has not reached saturation even when allowed to react with higher fluoride concentrations. To study the reusability of the LCB-10 adsorbent, the adsorption studies were conducted using the used adsorbent dried in oven at 70 ◦ C. The results show that fluoride removal decreases with successive reuse and reached to ∼4% after third reuse. This indicates exhaustion of fluoride removal capacity of material. After complete saturation, the fluoride loaded material was subjected for regeneration using different regeneration media like ammonium hydroxide, ammonium chloride and sodium chloride. It appears from the results that 1 M ammonium chloride shows a high regeneration capacity of 81.22% as compared to other regeneration media. Further, the regeneration studies in order to achieve 90–100% regeneration are in progress. 9. Field trials Considering the practical applicability of LCB-10 for defluoridation in actual field conditions, the LCB-10 was also tested with fluoride-contaminated ground water collected from fluoride affected regions of Dhar district of Madhya Pradesh, India. The untreated water had fluoride concentration of 10.2 mg l−1 . The detailed characteristics of untreated and treated ground water are presented in Table 2. It is evident from the results that LCB-10 effectively removes fluoride from field ground water and reduces the fluoride concentration to 0.24 mg l−1 with negligible release of lanthanum ion (0.067 mg l−1 ), which is well below the limits of Indian and WHO drinking water standards [34,35]. Other physico-chemical parameters of the treated water are also within permissible limits except which is marginally above the desirable limits of Indian drinking water standard but are within permissible limits in the absence of alternate sources [34]. Comparison of the fluoride removal by LCB-10 from field water and synthetic water shows that there is no significant difference between the fluoride uptake from field water and simulated water indicating LCB-10 can be exploited for defluoridation in actual field conditions, both at household and community level. The mechanism of fluoride adsorption through La incorporated chitosan appears to be through ligand exchange between fluoride ion and hydroxide ion coordinated with La (III) ion complexed with chitosan and has been discussed in detailed in our previous publication [33].
224
A. Bansiwal et al. / Colloids and Surfaces B: Biointerfaces 74 (2009) 216–224
Table 2 Physico-chemical parameters of untreated and treated filed water. Parameter
Untreated sample
Treated with LCB-10
Turbidity (NTU) pH Total hardness as CaCO3 (mg l−1 ) Alkalinity as CaCO3 (mg l−1 ) Total dissolved solids (mg l−1 ) Chlorides (mg l−1 ) Fluorides (mg l−1 ) Nitrates (mg l−1 ) Sulphate (mg l−1 ) Iron (mg l−1 ) Cadmium (mg l−1 ) Chromium (mg l−1 ) Copper (mg l−1 ) Lead (mg l−1 ) Manganese (mg l−1 ) Zinc (mg l−1 ) Arsenic (mg l−1 ) Aluminium (mg l−1 ) Lanthanum (mg l−1 )
<1 7 60 37 477 292 10.2 <0.01 101 <0.01 <0.002 <0.002 <0.01 <0.002 0.01 0.25 0.008 0.023 0.001
8 5.99 80 261 723 262.9 0.24 0.7 25 <0.01 <0.002 <0.002 <0.01 <0.002 0.046 <0.005 0.007 <0.003 0.067
10. Conclusion The LCB-10 was synthesised for defluoridation by optimization of various synthesis parameters. These beads are showing excellent fluoride removal efficiency of 97% at pH 5 and overcome the drawbacks associated with the conventional adsorbents. The LCB-10 not only has much higher adsorption capacity 4.7 mg g−1 than simple lanthanum-modified chitosan reported in our previous publication [33] but also has numerous advantages namely high chemical and mechanical stability, high resistance to attrition, negligible lanthanum release, suitability for column applications, etc. Equilibrium data fitted well to Langmuir adsorption isotherm. Moreover the kinetics of fluoride adsorption on LCB-10 are very fast. Thermodynamic study reveals that fluoride adsorption using LCB-10 was spontaneous and endothermic in nature. It is possible to regenerate the adsorbent using 1 M ammonium chloride showing high regeneration efficiency of 81.22%. The LCB-10 has shown excellent removal efficiency both in simulated as well as field water collected from fluoride endemic regions, which proves its suitability for defluoridation of drinking water in actual field conditions both at household and community levels. The treated water quality also confirms to Indian and WHO drinking water standards. Acknowledgements Authors thankfully acknowledge CSIR, New Delhi for granting SRF to carry out the research work and constant support and guidance of Dr. T. Chakrabarty, Director NEERI. Financial assistance from UNIECF Bhopal, India for this work is gratefully acknowledged. References [1] S. Ayoob, A.K. Gupta, Fluoride in drinking water: a review on the status and stress effects, Crit. Rev. Environ. Sci. Technol. 36 (2006) 433–487. [2] S.P. Teotia, M. Teotia, R.K. Singh, Hydrogeochemical aspects of endemic skeletal fluorosis in India—an epidemiological study, Fluoride 14 (1981) 69–74. [3] A.K. Susheela, Fluorosis management programme in India, Curr. Sci. 77 (1999) 1250–1256.
[4] A.K. Yadav, C.P. Kaushik, A.K. Haritash, A. Kansal, R. Neetu, Defluoridation of groundwater using brick powder as an adsorbent, J. Hazard. Mater. 128 (2006) 289–293. [5] W.G. Nawlakhe, D.N. Kulkarni, B.N. Pathak, K.R. Bulusu, Defluoridation of water by Nalgonda technique, Ind. J. Environ. Health 17 (1975) 26–65. [6] H. Mjengera, G. Mkongo, Appropriate defluoridation technology for use in fluorotic areas in Tanzania, Phys. Chem. Earth 28 (2003) 1097–1104. [7] A. Lhassani, M. Rumeau, D. Benjelloun, M. Pontie, Selective demineralisation of water by nanofiltration application to the defluoridation of brackish water, Water Res. 35 (2001) 3260–3264. [8] X. Feng Shen, P.G. Chen, C. Guohua, Electrochemical removal of fluoride ions from industrial wastewater, Chem. Eng. Sci. 58 (2003) 987–993. [9] N.I. Chubar, V.F. Samanidou, V.S. Kouts, G.G. Gallios, V.A. Kanibolotsky, V.V. Strelko, I.Z. Zhuravlev, Adsorption of fluoride, chloride, bromide and bromate ions on a novel ion exchanger, J. Colloid Interface Sci. 291 (2005) 67–74. [10] S. Ghorai, K.K. Pant, Equilibrium, kinetics and breakthrough studies for adsorption of fluoride on activated alumina, Sep. Purif. Technol. 42 (2005) 265–271. [11] D. Mohapatra, D. Mishra, S.P. Mishra, G. Roy Chaudhury, R.P. Das, Use of oxide minerals to abate fluoride from water, J. Colloid Interface Sci. 275 (2004) 355–359. [12] M. Sarkar, A. Banerjee, P.P. Pramanick, A.R. Sarkar, Use of laterite for the removal of fluoride from contaminated drinking water, J. Colloid Interface Sci. 302 (2006) 432–441. [13] A. Tor, Removal of fluoride from an aqueous solution by using montmorillonite, Desalination 201 (2006) 267–276. [14] P.P. Coetzeel, L.L. Coetzee, R. Pukal, S. Mubenga, Characterisation of selected South African clays for defluoridation of natural waters, Water SA 29 (2003) 331–338. [15] X. Fan, D.J. Parker, M.D. Smith, Adsorption kinetics of fluoride on low cost materials, Water Res. 37 (2003) 4929–4937. [16] R.L. Ramos, J. Ovalle-Turrubiartes, M.A. Sanchez-Castillo, Adsorption of fluoride on aluminium impregnated carbon, Carbon 37 (1999) 609–617. [17] M.S. Onyango, Y. Kojima, O. Aoyi, E.C. Bernardo, H. Matsuda, Adsorption equilibrium modeling and solution chemistry dependence of fluoride removal from water by trivalent-cation-exchanged zeolite F-9, J. Colloid Interface Sci. 279 (2004) 341–350. [18] S. Chidambaram, A.L. Ramanathan, S. Vasudevan, Fluoride removal studies in water using natural materials, Water SA 29 (2003) 339–344. [19] N.V. Majeti, R. Kumar, A review of chitin and chitosan applications, React. Funct. Polym. 46 (2000) 1–27. [20] A.J. Varmaa, S.V. Deshpandea, J.F. Kennedy, Metal complexation by chitosan and its derivatives: a review, Carbohydr. Polym. 55 (2004) 77–93. [21] W. Ma, Y. Fie-Qun, M. Han, R. Wang, Characteristics of equilibrium, kinetics studies for adsorption of fluoride on magnetic-chitosan particle, J. Hazard. Mater. 143 (2007) 296–302. [22] S. Annouar, M. Mountada, A. Soufiane, A. Elmidaoui, M.A. Menkouchi Sahli, Defluoridation of underground water by adsorption on the chitosan and by electrodialysis, Desalination 165 (2004) 437–438. [23] Li. Nan, R. Bai, Copper adsorption on chitosan–cellulose hydrogel beads: behaviors and mechanisms, Sep. Purif. Technol. 42 (2005) 237–247. [24] Standard methods for the examination of water and wastewater, in: APHAAWWA-WPCF, 20th edition, 1998. [25] A.T. Pauline, J.I. Simionato, J.C. Garcia, J. Nozaki, Characterization of chitosan and chitin produced from silkworm chrysalides, Carbohydr. Polym. 64 (2006) 98–103. [26] A.J. Varma, S.V. Deshpande, J.F. Kennedy, Metal complexation by chitosan and its derivative: a review, Carbohydr. Polym. 55 (2004) 77–93. [27] D. Berger, N. van Landschoot, C. Ionica, F. Papa, V. Fruth, Synthesis of pure and doped lanthanum cobaltite by the combustion method, J. Optoelectron. Adv. Mater. 5 (2003) 719–724. [28] S. Meenakshi, N. Viswanathan, Identification of selective ion-exchange resin for fluoride sorption, J. Colloid Interface Sci. 308 (2007) 438–450. [29] S. Meenakshi, C. Sairam Sundaram, R. Sukumar, Enhanced fluoride sorption by mechanochemically activated kaolinites, J. Hazard. Mater. 153 (2008) 164–172. [30] A.A. Khan, R.P. Singh, Adsorption thermodynamics of carbofuran on Sn (IV) arsenosilicate in H+ , Na+ and Ca2+ Forms, Colloid Surf. 24 (1987) 33–42. [31] S. Ayoob, A.K. Gupta, P.B. Bhakat, V.T. Bhat, Investigation on the kinetics and mechanisms of sorptive removal of fluoride from water using alumina cement granule, Chem. Eng. J. 140 (2008) 6–14. [32] S. Jagtap, D. Thakre, S. Wanjari, S. Kamble, N. Labhsetwar, S. Rayalu, New modified chitosan based adsorbent for defluoridation of water, J. Colloid Interface Sci. 332 (2009) 280–290. [33] S. Kamble, S. Jagtap, N. Labhsetwar, D. Thakre, S. Godfrey, S. Devotta, S.S. Rayalu, Defluoridation of drinking water using chitin, chitosan and lanthanummodified chitosan, Chem. Eng. J. 129 (2007) 173–180. [34] Indian Standard Specifications for Drinking Water IS: 10500, 1991. [35] Guidelines for Drinking-water Quality, WHO, 1993.