UDK 544.723:666.322 Original scientific article/Izvirni znanstveni članek ISSN 1580-2949 MTAEC9, 47(4)481(2013) EVALUATION OF EQUILIBRIUM ISOTHERM MODELS FOR THE ADSORPTION OF Cu AND Ni FROM WASTEWATER ON BENTONITE CLAY OCENA MODELOV RAVNOTEŽNIH IZOTERM ZA ADSORPCIJO Cu IN Ni IZ ODPADNIH VOD NA BENTONITNO GLINO Saad A. Al-Jlil, Mohammad Sajid Latif National Center for Water Technology (NCWT), King Abdulaziz City for Science and Technology (KACST), P. O. Box 6086, 11442 Riyadh, Kingdom of Saudi Arabia saljlil@kacst.edu.sa Prejem rokopisa — received: 2012-12-03; sprejem za objavo - accepted for publication: 2012-12-14 A study was carried out to evaluate equilibrium isotherm models for the adsorption of Cu and Ni on bentonite clay from wastewater. The adsorption of Cu and Ni increased with an increase in the pH and was high at the pH values of around 7. The adsorption capacity of bentonite clay increased with an increase in the temperature. The maximum adsorption capacity of bentonite clay was 11.12 mg g-1 and 6.78 mg g-1 for Cu and Ni, respectively, at 20 °C. Among the three equilibrium isotherm models used, the Langmuir-Freundlich isotherm model described the experimental data well at different temperatures. Keywords: equilibrium isotherm models, adsorption, copper, nickel, bentonite clay, temperature, wastewaters Izvršena je bila študija ocene modelov ravnotežnih izoterm pri adsorpciji Cu in Ni iz odpadnih vod na bentonitno glino. Adsorpcija bakra in niklja narašča z naraščanjem pH in je velika pri pH vrednostih okrog 7. Adsorpcijska zmogljivost bentonitne gline se povečuje z naraščanjem temperature. Največja zmogljivost adsorpcije bentonitne gline je bila 11,12 mg g-1 in 6,78 mg g-1 za Cu in Ni pri 20 °C. Med tremi modeli ravnotežnih izoterm pa dobro opisuje eksperimentalne podatke pri različnih temperaturah izotermni model Langmuir-Freundlich. Ključne besede: modeli ravnotežnih izoterm, adsorpcija, baker, nikelj, bentonitna glina, temperatura, odpadne vode 1 INTRODUCTION Wastewater is an important resource of water for augmenting the existing inadequate fresh-water supplies for multiple uses other than drinking purposes. Among the different biological, organic and inorganic pollutants in wastewater, the presence of trace elements and heavy-metal ions above the permissible limits pose environmental hazards. According to FAO (1985), the maximum permissible limits of Cu and Ni in drinking water are 1 mg L-1 and 0.015 mg L-1, respectively.1 Previous studies showed that sewage water in Riyadh contains an appreciable amount of heavy metals.2 Frequently, industrial-waste effluents are discharged into the drainage channels.3 The sources of heavy metals in wastewaters, especially the industrial effluents, are metal plating and ceramics photography,4 household chemicals in sewage water,5 and corrosion of the pipes of drinking-water-supply networks as well as leaching of the chemicals from polyvinyl chloride (PVC) pipes.67 To minimize the environmental hazards, it is important to remove the heavy-metal ions from the wastewaters and maintain their concentration within the recommended limits before their land disposal as required by the National Regulatory Authority. Currently, among the various technologies for wastewater treatment, the use of bentonite clay is a very common and effective method for the removal of heavy-metal ions, especially Cu and Ni, from wastewaters. The main objective of this study was to test and evaluate the local, natural, Saudi bentonite clay for the adsorption of copper and nickel ions from wastewaters using three equilibrium isotherm models at different temperatures. 2 METHODOLOGY 2.1 Experimental part a) Material Adsorbent: The adsorbent used in this research was the local natural bentonite clay. Adsorbates: Copper-and-nickel-ion solution prepared from the copper sulfate and nickel nitrate purified and supplied by S. Define-Chem. limited (Laboratory Rasa-yan) was used in the experiment. b) Procedures Preparation of copper-and-nickel-ion samples Stock solutions of Cu and Ni metal ions with a concentration of 1000 mg L-1 were obtained. A mixture of Cu and Ni ions was prepared by mixing both metal ions having the same concentration and then diluted to obtain the concentrations (50-1000 x 10-6) required for the use in equilibrium experiments. After that, the initial concentrations of the metal-ion samples and the final concentration of the resultant solution (after the completion of the experiment) were diluted and analyzed with the atomic-absorption spectroscopy (model AAnalyst 700, PerKin Elmer, an atomic absorption spectrometer). c) Equilibrium experiments The equilibrium isotherm was determined by placing a constant mass of clay (1 g) with a solute solution 50 mL in glass bottles in a constantly agitating shaker. In each isotherm run, the solute-solution concentrations ranged from 50-1000 mg L-1 and the temperatures ranged from 0-20 °C during the study. The equilibrium experiments were run up to the state of equilibrium. After that the samples were filtered using filter papers, then diluted and the absorbance was measured using atomic absorption spectroscopy. The concentration of Cu and Ni ions was calculated from the absorbance using the calibration curve. The amount of metal ions adsorbed on the bentonite clay (adsorbent) was calculated with the mass-balance equation as follows: q e =- y (C 0 -C e ) M (1) where M is the adsorbent mass (g), V is the solution volume (L), qe is the adsorbed metal-ion concentration (mg/g), Co is the initial concentration of metal ions (mg L-1) and Ce is the metal-ion concentration in a bulk solution at equilibrium (mg L-1). 2.2 Multicomponent Equilibrium Isotherm Models Three equilibrium isotherm models were applied to interpret the experimental data for bentonite clay as an adsorbent at different temperatures to obtain the optimum values of the equilibrium parameters. These models are: a) Extended Langmuir isotherm model The extended Langmuir isotherm model was applied to determine the adsorption of copper and nickel ions on bentonite clay from wastewater. The extended Langmuir isotherm model can be written as follows:8 q el = q e2 = (2) (3) The extended Langmuir parameters Ki, bi, K2 and can be obtained by using the non-linear regression technique with equations 2 and 3. b) Langmuir-Freundlich isotherm model A combination of Langmuir and Freundlich isotherm models makes a new model called the Langmuir-Freundlich isotherm model.9,10 The Langmuir-Freundlich isotherm parameters Ki, bi, K2 and b2, ni and «2 can be obtained by using the non-linear regression technique with equations 4 and 5. This form can be written as follows: / , „»1 \ q el = q e2 = 1+ b1 c ;; + b2 C (4) (5) c) Multicomponent isotherm model The required five parameters of the multicomponent isotherm model are as follows:911 q el = q e2 = fkl C el ] \l + bl C nl + b2 C en2 J f k 2 C e 2_^ \l + bl C nl + b2 C en22 J (6) (7) The equilibrium constants Ki, bi, K2 and b2, «1 and «2 can be obtained with the non-linear regression technique with equations 6 and 7. 3 RESULTS AND DISCUSSION 3.1 Characterization of Bentonite Clay Saudi bentonite clay was analyzed with XRF (model JSX-3201, JEOL, an element analyzer). The chemical analysis is presented in Table 1. Physical parameters such as the BET surface area, the pore volume and the average pore width of bentonite clay were determined with a surface-area analyzer (model ASAP 2020, Micro-meritics). The particle density and porosity of solid materials were measured in the Micromeritics Material Analysis Laboratory (Norcross, Georgia, U.S.A.) using the gas pycnometer method (an Accupc i330 pycno-meter). The results are shown in Table 2. The XRD Table 1: Chemical analysis of the Saudi bentonite clay in mass fractions (w/%) Tabela 1: Kemijska analiza saudijske bentonitne gline v masnih deležih (w/%) Compound w/% (in clay) SiO2 55.0 ± 3.0 Al2O3 22.0 ± 2.0 TiO2 l.5 ± 0.25 Fe2O3 5.67 ± 0.5 MgO 2.30 ± 0.45 CaO <2.00 Na2O <2.00 K2O Ni ion. This can be concluded from the data presented in Table 3. Table 3: Copper and nickel properties9,10 Tabela 3: Lastnosti bakra in niklja9,10 Property Copper ion Nickel ion Ionic radius (nm) 0.072 0.069 Atomic mass 63.4 57.8 Coordination number 2, 4 4, 5 Electron configuration [Ar] 3d9 [Ar] 3d8 Electro negativity of the atom 1.90 1.91 This variability for more Cu ion adsorption than Ni on clay is clear from Table 3, which shows an unpaired electron for a Cu ion in addition to the Cu ion being paramagnetic.14 Consequently, the Cu ion can be attracted by a magnetic field resultant from the clay adsor-bent15 because the Ni ion is stable due to the absence of an unpaired electron. 3.3 Equilibrium Experimental Results a) Extended Langmuir isotherm model The extended Langmuir parameters Ki, bi, K2 and b2 were obtained with the non-linear regression technique from equations 2 and 3. The equilibrium parameters Ki, Table 4: Extended Langmuir constants for the Cu-ion adsorption on clay from a mixture of copper and nickel solution at different temperatures Tabela 4: Razširjene Langmuirove konstante za adsorpcijo Cu-ionov na glino iz mešanice raztopine bakra in niklja pri različnih temperaturah Temperature T/oC Ki/ (L/g) B1/ (L/mg) AARD/ % R2 X2 20 0.106 0.007 18.58 0.941 2.41 40 0.1306 0.0089 16.42 0.959 2.06 60 0.167 0.0114 16.88 0.94 2.346 80 0.176 0.012 15.91 0.95 3.41 Table 5: Extended Langmuir constants for the Ni-ion adsorption on clay from a mixture of copper and nickel solution at different temperatures Tabela 5: Razširjene Langmuirove konstante za adsorpcijo Ni-ionov na glino iz mešanice raztopine bakra in niklja pri različnih temperaturah bi, K2 and b2 were calculated with the non-linear regression technique and presented in Tables 4 and 5. b) Langmuir-Freundlich isotherm model The parameters of the Langmuir-Freundlich isotherm model such as Ki, bi, K2 and b2, ni and «2 were obtained applying equations 4 and 5. The equilibrium parameters are presented in Tables 6 and 7. Table 6: Langmuir-Freundlich constants for the Cu-ion adsorption on clay in a mixture of multicomponent from copper and nickel at different temperatures Tabela 6: Langmuir-Freundlich konstante za adsorpcijo Cu-ionov na glini iz multikomponentne mešanice za baker in nikelj pri različnih temperaturah Temperature T/oC Ki/ (L/g) bl/ (L/mg) n1 AARD/ % R2 X2 20 0.662 0.0028 0.482 13.51 0.959 0.526 40 0.9778 0.0093 0.4233 11.31 0.964 0.535 60 0.567 0.0046 0.5799 13.57 0.935 1.16 80 0.4623 0.0066 0.7434 13.002 0.969 0.932 Table 7: Langmuir-Freundlich constants for the Ni-ion adsorption on clay in a mixture of multicomponent from copper and nickel at different temperatures Tabela 7: Langmuir-Freundlich konstante za adsorpcijo Ni-ionov na glino v multikomponentni mešanici iz bakra in niklja pri različnih temperaturah Temperature T/oC K2 /(L/g) b2 (L/mg) n2 AARD/ % R2 X2 20 0.722 0.0135 0.375 12.533 0.892 1.5 40 0.879 0.0005 0.355 16.34 0.901 1.23 60 1.101 0.0300 0.3720 20.2412 0.885 1.386 80 0.1928 0.0108 0.788 17.087 0.874 16.08 Table 8: Multicomponent five-parameter isotherm constants for the Cu-ion adsorption on clay in a mixture of copper and nickel at different temperatures Tabela 8: Večkomponentna petparametrična izotermna konstanta za adsorpcijo Cu-ionov na glini iz mešanice bakra in niklja pri različnih temperaturah Temperature T/oC Ki/ (L/g) b1/ (L/mg) n1 AARD/ % R2 X2 20 1.714 18.098 0.0664 20.254 0.918 3.935 40 1.072 5.599 0.1479 16.71 0.947 2.142 60 1.343 4.761 0.0119 15.49 0.938 1.263 80 1.489 1.279 0.0088 9.057 0.958 0.6468 Table 9: Multicomponent five-parameter isotherm constants for the Ni-ion adsorption on clay in a mixture of copper and nickel at different temperatures Tabela 9: Večkomponentna petparametrična izotermna konstanta za adsorpcijo Ni-ionov na glini iz mešanice bakra in niklja pri različnih temperaturah Temperature T/oC K2/ (L/g) b2/ (L/mg) AARD/ % R2 X2 Temperature (oC) K2/ (L/g) b2 (L/mg) n2 AARD/ % R2 X2 20 0.557 0.0002 22.25 0.672 29.23 20 0.9066 0.0048 1.456 26.598 0.614 38.312 40 0.068 0.0002 23.07 0.755 25.92 40 0.5629 0.0042 1.397 21.79 0.769 24.854 60 0.0843 0.0001 21.76 0.808 35.15 60 0.6829 0.1277 0.909 20.309 0.8252 23.44 80 0.0865 0.0001 22.40 0.829 37.32 80 0.7455 0.4703 0.7206 20.232 0.873 9.521 c) Multicomponent five-parameter isotherm model The equilibrium constants for the model such as Ki, bi, K2 and b2, ni and «2 were obtained applying equations 6 and 7 as shown in Tables 8 and 9. 3.4 Estimation of the Best Fit The theoretical results and the experimental equilibrium data were compared using the average absolute relative deviation percent (AARD/%), the chi-square method (X2) and the adjusted coefficient of determination (R2). The AARD was calculated to determine the best fit between the theoretical and experimental results from the applied equilibrium isotherm models as summarized below. 100 N AARDI%= — V N t! q¡ -q¡ Q¡ (8) where N is the number of data points, q-Caic is the calculated amount of the metal-ion adsorption on clay and qiexp is the experimental amount of the metal-ion adsorption on clay for a given data point i. Also, the chi-square method (X2) was applied to evaluate the relationship between the theoretical and the experimental data from the equilibrium experiments. The chi-square (X2) formula is: N (4i - Qi,)2 (X) = ^-^ (9) where N is the number of data points, Qiexp is the experimental amount of the metal-ion adsorption on clay and qicalc is the calculated amount of the metal-ion adsorption on clay for a given data point i. In the case of the chi-square method (X2), the X2 has a small value, the result of the model is close to the result of the equilibrium experiment and vice versa. The adjusted coefficient of determination, R2, is normally used to evaluate the best fit. Significant differences were observed between the extended Langmuir isotherm model, the Langmuir-Freundlich isotherm model and the multicomponent five-parameter isotherm model. The Langmuir-Freund-lich isotherm model provided the best correlation with the equilibrium data. 3.5 Numerical Solution of the Non-Linear Isotherm Models A non-linear, least-square, data-fitting algorithm using the fminsearch function from MATLAB was used to find the optimum values of the isotherm models. Then the average absolute relative deviation percent (AARD/%), the adjusted coefficient of determination, R2, and the chi-square method (X2) were calculated for the experimental and theoretical values. The program was written using MATLAB and the flow chart is shown below: 4 CONCLUSIONS The study showed that pH is a significant factor in the adsorption processes as it causes electrostatic changes in the solutions. The optimum pH value for the Cu and Ni adsorption is around 7. The maximum adsorption capacity of clay increased with an increase in the temperature, which was 11.12 mg g-1 for copper and 6.78 mg g-1 for nickel at 20 °C in the experimental solutions. Among the various isotherm models, the Langmuir-Freundlich isotherm model described the experimental data very well. Acknowledgements The authors would like to thank King Abdulaziz City for Science and Technology (KACST) for the support and encouragement to carry out the study. 5 REFERENCES 1 M. J. Zamzow, J. E. Murphy, Removal of metal cations from water using zeolites, Separation Science and Technology, 27 (1992) 14, 1969-1984 2 A. S. Mashhady, Heavy metals extractable from a calcareous soil treated with sewage sludge, Environmental pollution (series B), 8 (1984), 51 3 A. M. Al-Rehaili, Municipal wastewater treatment and reuse in Saudi Arabia, The Arabian Journal for Science and Engineering, 22 (1997), 143 4 U. Forstner, G. T. Wittmann, Metal pollution in the aquatic environment, Springer verlag, New York 1979 5D. Jenkins, Effect of reformulation of household powder laundry detergents on their contribution to heavy metals levels in wastewater, Water Environment Research, 70 (1998) 5, 980 6 A. I. Alabdula'aly, Trace metals in Riyadh public water supplies, The Arabian Journal for Science and Engineering, 22 (1997), 165 7 M. Sadiq, G. Hussain, Drinking water quality in Saudi Arabia- an overview, The Arabian Journal for Science and Engineering, 22 (1997), 153 8 G. Mckay, Use of adsorbents for the removal of pollutants from wastewater, CRC, Press, Inc., 1996 9W. Fritz, E. U. Schuender, Simulation adsorption equilibria of organic solutes in dilute aqueous solution on activated carbon, Chemical Engineering Series, 29 (1974), 1279 10 A. I. Liapis, D. W. T. Rippin, A general model for the simulation of multi-component adsorption from a finite batch, Chemical Engineering science, 32 (1977), 619 11 C. M. Yon, P. H. Turnock, Multicomponent adsorption equilibria on molecular sieves, American Institute Chemical Engineers, Symposium, 67 (1971), 75 12 H. A. Elliott, C. P. Huang, Adsorption characteristics of some Cu(II) complexes on aluminosilicates, Water research, 15 (1981), 849-855 13 K. O. Adebowale, I. E. Unuabonah, B. I. Olu-Owolabi, The effect of some operating variables on the adsorption of lead and cadmium ions on kaolinite clay, Journal of Hazard Material, 134 (2006), 130 14 K. F. Purcell, J. C. Kotz, An Introduction to Inorganic Chemistry, Saunders College Publishing, Philadelphia 1980 15 K. H. Chong, B. Volesky, Metal biosorption equilibria in a ternary system, Biotechnology and Bioengineering, 49 (1996), 629