You are on page 1of 6

Journal of Colloid and Interface Science 288 (2005) 3035 www.elsevier.


Adsorption of a model anionic dye, eosin Y, from aqueous solution by chitosan hydrobeads
Sudipta Chatterjee a , Sandipan Chatterjee a , Bishnu P. Chatterjee a , Akhil R. Das b , Arun K. Guha a,
a Department of Biological Chemistry, Indian Association for the Cultivation of Science, Jadavpur, Kolkata 700032, India b Polymer Science Unit, Indian Association for the Cultivation of Science, Jadavpur, Kolkata 700032, India

Received 13 December 2004; accepted 17 February 2005 Available online 8 April 2005

Abstract The process of sorption is being increasingly used for ecofriendly and economic remediation of textile dye efuents. The present model study deals with the adsorption of a model anionic dye, eosin Y, from wastewater using conditioned chitosan hydrobeads. Conditioning reduced the pH sensitivity and maintained the maximum sorption capacity of the beads near pH 8. To understand the chemicophysical characteristics of the adsorption process we studied, the kinetics and isotherm behavior of the system. It was observed that temperature played a signicant role in the process. The Langmuir model was found to be most appropriate for the description of the adsorption process. The kinetic results followed a second-order equation. It was observed that 1 g of chitosan adsorbed 76 mg of eosin Y. The dye was desorbed from the beads by changing the pH of the solution, and the conditioned chitosan beads were reused ve times without any loss of mechanical and chemical efcacy. 2005 Elsevier Inc. All rights reserved.
Keywords: Bioremediation; Anionic dye; Chitosan hydrobeads; Chemicophysical study

1. Introduction Worldwide annual production of dyes is around 7 105 tons, 510% of which is discharged into waste streams by the textile industries [1]. The majority of these dyes are of synthetic origin and toxic in nature with suspected carcinogenic and genotoxic effects [24]. The dye-bearing efuent, when discharged into water bodies, affects photosynthesis, aquatic life, and also humans. A wide range of conventional treatment techniques such as chemical coagulation, activated sludge, trickling lter, carbon adsorption, and photodegradation have been investigated extensively for removing dye from aquatic bodies. The adsorption process produces an efuent that is free
* Corresponding author. Fax: +91 33 2473 2805.

E-mail address: (A.K. Guha). 0021-9797/$ see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2005.02.055

of harmful substances like carcinogenic aromatic amines produced by microbial degradation [1,5] and free radicals by photodegradation using UV light. The activated carbon generally used for dye adsorption is very expensive. Biosorptionuse of waste biomaterials as sorbentsis a newly developed technique [69] for the removal of harmful substances from water bodies, but suffers serious limitations in the case of anionic dyes as most of the biomaterials contain negatively charged cellulosic moieties, which lower adsorption due to coulombic repulsion. Recently, chitosan, a biopolymer of glucosamine, showed a higher capacity for adsorption of colorant than activated carbon [10]. No and Meyers [11] demonstrated that swollen beads of chitosan exhibit superior sorption capacity compared with akes. Moreover, the beads can be easily separated from the treated bulk for repeated use. The ability of the anionic dye to adsorb onto chitosan beads can be attributed to the surface charge

S. Chatterjee et al. / Journal of Colloid and Interface Science 288 (2005) 3035


which depends on the pH of the operating system. Chitosan has a positively charged surface below pH 6.5 (point of zero potential), and reducing the pH increases the positivity of the surface [12,13], thus making the sorption process pH sensitive. Chitosan hydrobeads lose their integrity as a result of partial dissolution in acidic solvent, making them unsuitable for reuse. Crosslinking of chitosan with different bifunctional reagents [1416] has been tried to increase the integrity of the beads but has resulted in poor adsorption. Thus, it is of interest to increase the integrity of the chitosan hydrobeads, as well as sorption properties, at acidic pH. We studied the efcacy of adsorption of a water-soluble model anionic dye, eosin Y, by sulfate-conditioned chitosan hydrobeads and the chemicophysical nature of adsorption to gather more information on the interaction of conditioned chitosan and anionic dyes. Eosin Y (CI acid red 87), a coal tar xanthene dye, was chosen as the model anionic dye to avoid environmental hazards during investigation, as this dye is not specically listed as toxic by different health agencies.

which is far longer than required to reach equilibrium. The beads were separated from the solution by ltration through glass wool. The initial and equilibrium concentrations of the dye were estimated spectrophotometrically. The experiment was repeated at different pH values (4.012.0) of the dye solution. 2.5. Equilibrium sorption experiment An equilibrium sorption study was carried out with conditioned chitosan beads as described above. The concentration of the dye in the solution was allowed to vary from 1 to 500 mg L1 . The experiment was conducted at pH 6.0. The same experiment was also repeated at two different temperatures (viz. 40 and 50 C). 2.6. Kinetic experiment One and one-half grams of conditioned chitosan beads was added to different 250-ml Erlenmeyer asks, each containing 50 ml of aqueous solution of eosin Y. The pH of the solution was adjusted to 6.0, and initial dye concentrations were 20, 50, 100, and 200 mg L1 . The experiment was conducted at pH 6.0 and 30 C, and the asks were agitated (50 rpm) for 48 h. At regular intervals, the concentration of unadsorbed dye was determined. 2.7. Desorption experiment After conducting the equilibrium study with 200 mg L1 eosin Y, dye-loaded conditioned chitosan beads were collected by ltration through glass wool and used for desorption experiments. The beads were transferred to different 250-ml Erlenmeyer asks, each containing 50 ml water, and the pH was adjusted to 10.0, 11.0, and 12.0. The asks were agitated at 50 rpm up to 20 h at 30 C. The concentration of the eluted dye was determined at different time intervals.

2. Materials and methods 2.1. Chemicals Chitosan was prepared from shrimp shell by a modied Hackmann procedure as described earlier [17]. Eosin Y was purchased from BDH, India, and all other chemicals were procured from E. Merck, Germany. 2.2. Preparation of chitosan hydrobeads Chitosan beads were prepared by dropwise addition of degassed chitosan solution (2% w/v) in 7% v/v AcOH to an alkaline coagulating mixture [H2 O:MeOH:NaOH = 4:5:1 w/w] as described by Mitani and co-researchers [18]. The beads were collected by ltration and washed with water until neutral. The beads were conditioned using 0.1 M (NH4 )2 SO4 [19]. 2.3. Estimation of eosin Y The concentration of eosin Y in the solution was determined from the calibration curve drawn by measuring the absorbance at 517 nm in a Varian UV/vis spectrophotometer. 2.4. Effect of conditioning Fifty-milliliter aqueous solutions of eosin Y (500 mg L1 ) were taken in different 250-ml Erlenmeyer asks. The pH of the dye solutions was adjusted to 4.0. In one set of experiments, 1.5 g of swollen chitosan beads, and in another set of experiments the same amount of conditioned beads, was added. The asks were agitated for 48 h (50 rpm) at 30 C,

3. Results and discussion 3.1. Effect of pH The effect of the conditioning of chitosan on dye adsorption at different pH is illustrated in Fig. 1. It appears that conditioned and unconditioned chitosan beads adsorbed equal amounts of eosin Y (76 mg/g dry chitosan) at pH 4.0. In the case of unconditioned beads, dye adsorption decreased sharply with increasing pH. The adsorption of eosin Y on the conditioned beads, on the other hand, remained practically unchanged up to pH 8.0; beyond this range a sharp decrease was noted. Adsorption of the anionic dye eosin Y by chitosan can be attributed to the interaction between the protonated amine group of chitosan (RNH2 ) and the anionic group of the dye (DCO2 Na) molecule:


S. Chatterjee et al. / Journal of Colloid and Interface Science 288 (2005) 3035

Fig. 1. Adsorption of eosin Y by conditioned and nonconditioned chitosan beads at different pH. Results represent averages of ve replicated experiments.

Fig. 2. Langmuir plot of eosin Ychitosan bead equilibrium experimental data. R represents the correlation coefcient. P represents the probability (that R = 0), and the equation represents the regression equation of the data.

RNH2 + H3 O+ RNH+ 3 + H2 O,
+ DCO DCO2 Na 2 + Na , + RNH+ 3 + DCO2 RNH3 DCO2 . H2 O

surface with uniform energy. The linear form of the model is given by log qe = log KF + 1/n log Ce , (1)

Decreasing the pH of the solution makes more protons available to protonate the amine group of chitosan, with the formation of a larger number of cationic amines. This results in increasing dye adsorption by chitosan due to increased electrostatic interactions. Conditioning of chitosan with ammo nium sulfate would probably form RNH+ 3 SO4 NH4 , which undergoes interaction with dye anion following the mechanism proposed by Muzzarelli and Rocchetti [19] in the case of oxyanion adsorption by sulfate conditioned chitosan:
RNH2 RNH+ 3 NH4 SO4 . 0.1 M (NH4 )2 SO4

where qe and Ce are the sorbate concentrations at equilibrium in the solid (mg g1 ) and liquid (mg L1 ), respectively; 1/n is the heterogeneity factor; and KF is the Freundlich constant. The sorption process under investigation does not obey the Freundlich isotherm model. No linear relationship is found between the sorbet concentrations at equilibrium in the solid and liquid phases respectively (gure not shown). On the other hand, in the Langmuir model it is assumed that the thickness of the adsorbed layer is monomolecular in nature. The linear form of the isotherm, 1 Ce Ce = , + qe Q0 b Q0 (2)

Adsorption of the colored anion on the chitosan surface probably proceeds through the ion exchange reaction
RNH+ 3 NH4 SO4 + DCO2 + RNH3 DCO2 + NH4 SO 4.

3.2. Effect of temperature Conditioned chitosan beads adsorbed eosin Y most efciently at 30 C. With a rise in temperature, uptake falls signicantly, indicating the exothermic nature of the process. Below 30 C, the process is slow and requires a long time to achieve equilibrium. A comprehensive understanding of the nature of the interaction of adsorbate with adsorbents is essential for the most effective use of the adsorbent. To formulate a pragmatic program for design, operation, and optimization, a correlation of the equilibrium data with the different relevant models of isotherm is a prerequisite. In the present study the experimental data for the eosin chitosan equilibrium isotherm were compared with those for the Freundlich and Langmuir isotherm models. The Freundlich isotherm explains adsorption on a heterogeneous

has been used to analyze the experimental data, where Ce and qe are the equilibrium dye concentrations in the aqueous (mg L1 ) and solid (mg g1 ) phases, respectively; Q0 and b are Langmuir constants related to the capacity of adsorption and energy of adsorption, respectively. Plots of Ce /qe versus Ce for eosin Y at three different temperatures are presented in Fig. 2. The linear form of the isotherms over the whole concentration range is conspicuous and the corresponding correlation coefcients are satisfactory (R 1.0). These strongly suggest that the present sorption process occurs at a single surface and follows the Langmuir sorption model closely. Langmuir monolayer saturation capacity Q0 and energy of adsorption b are listed in Table 1. The general shape of the equilibrium curves along with the sharp curvature reects the characteristics of Langmuir equilibrium with a high degree of irreversibility. The essential features of the Langmuir isotherm can be expressed in terms of the dimensionless equilibrium parameter RL , which is dened as RL = 1/(1 + bC0 ), where b is the Langmuir constant as described above and C0 is the initial dye concentration. RL values within the range 0 < RL < 1 indicate favorable adsorption. RL values calculated using C0

S. Chatterjee et al. / Journal of Colloid and Interface Science 288 (2005) 3035


Table 1 Langmuir adsorption isotherm and second-order rate constants Constant in Langmuir model T ( C) Q0 (mg g1 ) b (L g1 ) RL (using C0 = 200 mg L1 ) 0.0360 0.0487 0.0495 Constant in second-order model Initial concentration C0 (mg L1 ) 20 50 100 200 h (mg g1 min1 ) 0.1603 0.1779 0.2030 0.5000 qe (calculated) 12.50 33.33 50.00 100.00 k2 10.3 104 1.6 104 0.8 104 0.5 104

30 40 50

80.84 75.82 72.83

0.1345 0.0977 0.0960

Table 2 Thermodynamic properties of the conditioned chitosaneosin Y adsorption process Concentration (mg L1 ) Initial, C0 (mg L1 ) 50 Final, Ce (mg L1 ) 5.120 8.010 8.220 16.678 18.880 22.460 44.720 49.820 53.620 88.240 92.330 96.430 30 40 50 30 40 50 30 40 50 30 40 50 8.770 5.242 5.083 4.996 4.297 3.450 2.240 2.010 1.795 1.267 1.166 1.070 T ( C) Kc Thermodynamic parameter G0 (kJ mol1 ) 5.467 4.311 4.366 4.052 3.672 3.119 2.303 1.758 1.473 0.596 0.387 0.171 H0 (kJ mol1 ) 17.10 S0 (J mol1 K1 ) 51.8










as high as 200 mg L1 are well within the dened range (Table 1) and indicate the acceptability of the process. Thermodynamic parameters such as change in free energy ( G0 ), enthalpy ( H 0 ), and entropy ( S 0 ) were determined using the following equations [20,21]: KC = CAe /Ce , G0 = RT ln KC , log KC = H0 S0 , 2.303R 2.303RT

where KC is the equilibrium constant, CAe is the amount of dye (mg) adsorbed by each gram of adsorbent from 1 L of the dye solution at equilibrium, Ce is the equilibrium concentration (mg L1 ) of the dye in the solution, T is the solution temperature (in K), and R is the universal gas constant. H 0 and S 0 were calculated from the slope and intercept of vant Hoff plots of log KC versus 1/T . The results are summarized in Table 2. The G0 values indicate that the adsorption process is spontaneous in nature and more favorable at lower concentrations of dye compared with higher concentrations. The appreciably low free energy values indicate saturation of the process. The enthalpy values suggest that the reaction is exothermic and thus less favorable at higher temperatures.

Fig. 3. Effect of contact time on eosin Ychitosan bead equilibrium at different initial dye concentrations: (a) 200 mg/L; (Q) 100 mg/L; (") 50 mg/L; (2) 20 mg/L. Results represent averages of ve replicated experiments.

3.3. Kinetic study The kinetic behavior of the adsorption process was studied at 30 C using different initial dye concentrations (Fig. 3). It appears from the gure that the adsorption capacity of the beads increases with increasing initial dye concentration, but the time required to reach equilibrium in


S. Chatterjee et al. / Journal of Colloid and Interface Science 288 (2005) 3035

the process is independent of initial dye concentration as reported by previous researchers [22,23]. The process is very fast initially and then slowly reache equilibrium within 20 h. To determine the rate-controlling and mass transfer mechanism, kinetic data were correlated to linear forms of the pseudo-rst-order rate model, log(qe qt ) = log qe k1 t, 2.303 (3)

and the second-order rate model, 1 t 1 = + t, qt h qe

2 h = k2 q e ,

(4) (mg g1 )
Fig. 4. Second-order plot of adsorption of eosin Y onto chitosan beads using different initial dye concentrations. Symbols are described in the legend to Fig 3. R represents the correlation coefcient, P represents the probability (that R = 0), and the equation represents the regression equation of the data.

where qe and qt are the amounts of dye adsorbed at equilibrium and at different intervals, respectively; k1 (min1 ) and k2 (g mg1 min1 ) are the pseudo-rst-order and second-order rate constants; and h represents the initial adsorption rate (mg g1 min1 ). By correlation of the kinetic data with the above two rate models, it was found that the plot of t/qt against time using different initial dye concentrations gives straight lines (Fig. 4) with high correlation coefcients (R 1.0). This indicates that the present sorption system follows predominantly the second-order rate model and the overall process appears to be controlled by chemisorption [24,25], whereas the high degree of nonlinearity (gures are not shown) and poor correlation coefcient suggest the inability of the pseudo-rst-order model to describe the kinetic prole of the adsorption process. The average values of the rate constants are determined from the intercepts of the curves and are given in Table 1, which lists k2 , h, and qe (calculated) as a function of initial dye concentration. The second-order rate constant decreases with an increase in initial dye concentration, while the initial adsorption rate increases with an increase in initial dye concentration. 3.4. Desorption study Fig. 5 illustrates the desorption of eosin Y from chitosan beads at different pH with time. Eosin has been found to be desorbed from the loaded beads just by increasing the pH of the eluant to the alkaline range. The reaction responsible for the desorption is
RNH+ 3 DCO2 + NaOH RNH2 + DCO2 Na + H2 O.

Fig. 5. Desorption of eosin Y from dye-loaded chitosan beads at different times: (Q) pH 12.0; (!) pH 11.0; (2) pH 10.0. Results represent averages of ve replicated experiments.

4. Conclusions 1. Conditioning of the chitosan beads with ammonium sulfate reduces the pH sensitivity of the process. 2. Adsorption of the eosin dye on chitosan beads follows the Langmuir model. 3. The process is exothermic in nature. The highest efciency of the adsorption process is observed at 30 C. 4. The sorption process is very fast initially, attains equilibrium within a few hours, and follows the second-order kinetic rate model. 5. The treated chitosan bead adsorbate is recyclable.

The desorption process is observed to be reasonably fast initially and then slowly to attain equilibrium after about 16 h. The rate of desorption is found to increase with an increase in pH of the eluant. About 65% of the dye is desorbed from the beads at pH 12.0 within 2 h, whereas 98% can be eluted at pH 11.0 or 12.0. The chitosan beads used in this study were reused for ve cycles without any loss of their sorption capacity (data not shown).

Acknowledgment Sandipan Chatterjee acknowledges the Council of Scientic and Industrial Research (CSIR), Government of India, for providing a Senior Research Fellowship.

S. Chatterjee et al. / Journal of Colloid and Interface Science 288 (2005) 3035


[1] Y. Wong, J. Yu, Water Res. 33 (1999) 3512. [2] P.C. Vandevivere, R. Bianchi, W. Verstraete, J. Chem. Technol. Biotechnol. 72 (1998) 289. [3] W.K. Walthall, J.D. Stark, Environ. Pollut. 104 (1999) 207. [4] K. Al-sabti, K. Ecotoxicol. Environ. Saf. 47 (2000) 149. [5] T. Panswad, W. Luangdilok, Water Res. 34 (2000) 4177. [6] K.R. Ramakrishna, T. Viraraghavan, Water Sci. Technol. 36 (1997) 189. [7] Y.S. Choi, J.H. Cho, Environ. Technol. 17 (1996) 1169. [8] A.K. Bhattacharya, C. Venkobachar, J. Environ. Eng. 110 (1984) 110. [9] S.E. Bailey, T.J. Olin, R.M. Bricka, D.A. Adrian, Water Res. 33 (1999) 2469. [10] H. Yoshida, A. Okamoto, T. Kataoka, Chem. Eng. Sci. 48 (1993) 2267. [11] H.K. No, S.P. Meyers, Rev. Environ. Cont. Toxicol. 163 (2000) 1. [12] L. Jin, R. Bai, Langmuir 18 (2002) 9765.

[13] X. Zhang, R. Bai, J. Colloid Interface Sci. 264 (2003) 30. [14] Z. Cao, H. Ge, S. Lai, Eur. Polym. J. 37 (2001) 2141. [15] D. De-Capitani, A.A. Angelis, A. Crescenzi, G. Masci, A.L. Segre, Carbohydrate Polym. 45 (2001) 245. [16] Y. Kawamura, H. Yoshida, S. Asai, H. Tanibe, Water Sci. Technol. 35 (7) (1997) 97. [17] S. Chatterjee, A.K. Guha, B.P. Chatterjee, Ind. J. Chem. Technol. 10 (2003) 350. [18] T. Mitani, N. Fukomuro, C. Yoshimoto, H. Ishii, Agr. Biol. Chem. 55 (9) (1991) 2419. [19] R.A.A. Muzzarelli, R. Rocchetti, Talanta 21 (1974) 1137. [20] C. Namasivayam, R.T. Yamuna, Environ. Pollut. 89 (1995) 1. [21] G.C. Catena, F.V. Bright, Anal. Chem. 61 (1989) 905. [22] M.S. Chiou, H.Y. Li, Chemosphere 50 (2003) 1095. [23] M.S. Chiou, P.-Y. Ho, H.Y. Li, Dyes Pigments 60 (2004) 69. [24] G. McKay, Y.S. Ho, Process Biochem. 34 (1999) 451. [25] G. McKay, Y.S. Ho, Water Res. 33 (1999) 578.