Microsoft Word - contents_2.doc HUNGARIAN JOURNAL OF INDUSTRIAL CHEMISTRY VESZPRÉM Vol. 31. pp. 9-12 (2003) CHEMICAL UTILIZATION OF BIOMASS II: PREPARATION OF ADSORBENTS FROM WHEAT STRAW TO ELIMINATE CUPRIC AND ZINC IONS OF INDUSTRIAL WASTEWATERS J. DENCS, Z. MARTON, B. DENCS AND G. MARTON University of Veszprém, Cooperative Research Centre H-8201, Veszprém, POB 158. Hungary Agricultural residues represent a cheap and environmentally safe source of raw material for the preparation of ion exchangers and adsorbents that may be useful for removal of metals and/or colour from different wastewaters. Metal ions (Cd, Cr, Co, Cu, Fe, Pb, Mn, Ni and Zn) are known to contaminate the industrial effluents, water supplies, as well as mine waters. Fish and textile plants may apply this method to eliminate contaminants in case of low ion concentrations far away from their source. The simultaneous biosorption of Cu and Zn ions from aqueous solution by straw based oxidised cellulose was studied. The adsorption equilibrium was attained in 24 h at room temperature. The adsorption was found to fit the modified Langmuir type isotherms, and the parameters of the maximum adsorbent capacity and the adsorption intensity were calculated. Keywords: adsorbents, straw, bio-based products Introduction The sorption of heavy metal ions onto biosorbents from aqueous solution has been studied for single systems (Xie, Chang & Kilbane, 1966). The influence of a second metal has also been investigated (Trujillo, Jeffers, Ferguson & Stevenson, 1991). Chong and Volesky (1996) reported a description of a two metal biosorption equilibrium using Langmuir-type models. There have been several extended Langmuir models applied to multicomponent sorption systems. Jain and Snoeyink (1973) developed an extended Langmuir model by adding a term based on the hypothesis that in a multicomponent system adsorption occurs both with and without competition. A two-term expansion of the isotherm for competition equilibrium of the two components of a binary mixture has also been reported (Levan & Vermeulen, 1981). The present paper investigates the binary sorption of copper and zinc ions onto oxidised cellulose using three molar ratios of Cu:Zn. The model has been developed by incorporating an interaction factor correlation into the multicomponent Langmuir equation. This model was originally developed by Marton et al. (1981). Materials and Methods The batch sorption experiments for single- component systems were carried out using a constant temperature shaking water bath and a series of capped 125 ml conical flasks. In the sorption isotherm tests, 0.5 g of straw sample was thoroughly mixed with the metal ion solution (125ml). The concentrations of the two metal, copper and zinc ions system ranged from 0.2 to 3 mmol/dm3. A similar procedure was used for the bi-component studies. Three different systems were examined using copper and zinc in the molar ratios of 1:1, 1:2 and 2:1 (Cu:Zn) and in the concentration range of 0.5-4 mmol/dm3 for both Cu and Zn ions. 10 Materials Stock solutions of CuSO4 and ZnSO4 were prepared in distilled water. Wheat straw was obtained from a local agro-refinery with the fraction 60-80 mesh particle size. Pretreatment procedure consists of a cc. HNO3 and H3PO4 (1:1) treatment at 80 °C for 120 min and water washing. Preparation of oxidized cellulose (OC) Nitric acid (69.7%w/w) and phosphoric acid (85% w/w) were mixed in a 1:1 ratio. 5.0 g of straw sample was added to a 70 ml solution of the acid mixture. It was completely soaked and 1.0 g of sodium nitrate was added. An immediate formation of reddish brown fumes occurred. To prevent the release of these fumes into the air, the reaction container was covered with a Petri dish. The reaction mixture was allowed to react at room temperature, with occasional stirring using a glass rod, for 12, 24, 36 or 48 h. The reaction mixture, which appeared green, was diluted by adding an excess of water (about five times the volume of the reaction mixture). The diluted reaction mixture was filtered, and washed with water until the filtrate showed a pH of about 4. It was finally washed with acetone and then air-dried at room temperature. The dried OC was ball milled for 24 h and then sieved. The solid product of the reaction contained particles ranging between 74 and 105 µm was used for the adsorption investigation. Determination of carboxyl content This was performed according to the method described in the United States Pharmacopoeia (USP, 1995). Approximately 0.5 g of the sample was weighed and suspended in 50 ml of a 2% (w/w) calcium acetate solution for 30 min. The mixture was titrated with 0.1M NaOH using phenolphthalein as an indicator. The volume of NaOH solution consumed was corrected for the blank. The carboxyl content in the sample was calculated using the following relationship: 100 (mg)sampleofWeight VM49 w/w)(%groupsCarboxyl ⋅ ⋅⋅⋅ = where M is the molality of NaOH, and V is the volume of NaOH in ml consumed in titration, after correcting for the blank. The carboxyl content of our product was 18.2%. Isotherms Equilibrium sorption studies provide the capacity of the sorbent, which can be described by a sorption isotherm characterised by certain constants whose values express the surface properties and affinity of the sorbent. Sorption equilibrium is established when the concentration of sorbate in the bulk solution is in dynamic balance with that of the interface. Equilibrium relationships between sorbent and sorbate are described by sorption isotherms. The data were determined and analysed in accordance with some of the most frequently used isotherms. The Langmuir sorption isotherm (Langmuir, 1916) is the best known and the most often used isotherm for the sorption of a solute from liquid solution: Q C Qb 1 q C e 1e e += (1) where qe is the sorption capacity at the equilibrium solute concentration Ce (mmol/g); Ce is the concentration of sorbate in solution (mmol/dm3); Q is the maximum sorption capacity corresponding to complete monolayer coverage (mmol/g); bl is a Langmuir constant related to the energy of sorption (dm3/mmol). Butler and Ockrent (1930) developed a model for competitive sorption based on the Langmuir equation. This isotherm is applicable when each single-component obeys Langmuir behaviour in a single-component system. It is widely used to calculate the Langmuir constant Q, the amount of solute sorbed per unit weight of the sorbent, in the multicomponent systems. The common form for depicting the sorbate distribution is to correlate the amount of solute sorbed per unit weight of the sorbent with the residual solute concentration remaining in an equilibrium state. If there are two solutes present together in the sorption system the extended Langmuir isotherms are: 2211 111 1 CbCb1 CbQ q ++ = (2) 2211 222 2 CbCb1 CbQ q ++ = (3) where q1 and q2 are the amounts of solutes 1 and 2 sorbed per unit weight of the sorbent at equilibrium concentrations C1 and C2 respectively. 11 Q1 and Q2 are the maximum sorption capacities of 1 and 2, respectively. These are determined from single-component systems and, therefore, correspond to a monolayer coverage of the sorbent. b1 and b2 are the Langmuir constants and are a function of the energy of sorption of solutes 1 and 2, respectively. These are also determined from single-component systems. These Langmuir equations are simple extensions of the single- component isotherms to multicomponent sorption. They assume that each component adsorbs onto the surface according to the ideal solute behaviour under homogenous conditions with no interaction or competition taking place between the molecules. Extension of the Langmuir and BET theory for the description of multicomponent liquid-phase adsorption According to the original Langmuir hypothesis, the free area available for adsorption of component i is:         −στ= ∑ = n 1j iii0ii1 SQNzS (4) Let us assume that the ratio of free and occupied area of the adsorbent is: ( )∑∑ == δ−στ στ = n 1j ijjj0jj ii0ii n 1j j i QzN QzN S S (5) where δij is the surface-area-excess accounting for the difference in size and energy-state of the competing adsorbing components. By definition δij = 0 for each i = j, while for i ≠ j it may be a positive or negative number, or zero. By substituting Eq. 5 into Eq. 4 we obtain: ( )             δ−στ στ −στ= ∑ = = n 1i 1j ijjj0jj ii0ii i ii0iii QzNQzN S QNzS (6) Let us introduce: iii0ii CbNz =στ (7) Let us make use of the statement of BET theory, i.e. that at low concentration of the adsorbent component: i n 1i iii qCk1S         −= ∑ = (8) By substituting Eq. 7 and 8 into Eq. 6 we obtain: ( )         − ⋅ δ−+ = ∑∑ = = = n 1i ii n 1i 1j ijjjj i iii i Ck1 1 QCb Q 1 1 QCb q (9) Let us introduce: ( ) i ijjj ji Q Qb b δ− = (10) For two components Eq. 9 becomes: 22112211 111 1 CkCk1 1 CbCb1 CQb q −− ⋅ ++ = (11) 22111122 222 2 CkCk1 1 CbCb1 CQb q −− ⋅ ++ = (12) k1 and k2 in the equations 11 and 12 are so called interaction factors. Results and Discussion The data were examined by linear regression analysis using equations 11 and 12. Table 1 shows the modified Langmuir parameters for the adsorption of Cu(II) and Zn ions. Comparison of measured and calculated data (Fig 1) produced an excellent (>0.98) correlation coefficient for the bi- component system. Table 1: Modified Langmuir parameters for the adsorption of Cu(II) and Zn(II) for the eq. 11 and 12. Sample Q1 Q2 b1 b2 k1 k2 mmol/g dm3/mmol Untrd. straw 0.072 0.066 0.52 0.37 0.021 0.013 Oxidized straw 0.321 0.293 0.98 0.83 0.043 0.080 12 0 50 100 150 200 250 300 350 400 450 500 0 50 100 150 200 250 300 350 400 450 measured q1 (mmol/kg) pr ed ic te d q 1 (m m ol /k g) Cu:Zn = 1:2 - 2:1 Figure 1: The relationship between measured and predicted amount of adsorbed component of Cu ion in the presence of Zn ion. Conclusions A series of modified extended multicomponent Langmuir isotherms have been used to correlate binary copper-zinc equilibrium sorption study results using oxidised wheat straw. Because of complexity of the mechanism of biosorption a new interaction factor-combined equilibrium model has been developed. SYMBOLS A specific surface area of adsorbent, m3/kg b adsorption energy coefficient, m3/mol c1 concentration of i-th component, mol/m3 k1 BET constant, m2/mol N Avogadro’s number, 1/mol qi amount of adsorbed component, mol/kg Q ultimate uptake capacity, mol/kg R universal gas coefficient, m2 bar/K mol Si surface covered by i-th solute, m2 zi collison frequency factor of i-th solute σ0i actual area of one adsorbed molecule, m2 τi residence time on the surface of i-th solute, h SUBSCRIPTS i, j …, …, n component c calculated m measured REFERENCES 1. Xie J. Z. and Chang H. L.: (1996) Biores. Techn. 57, 127-136 2. Trujillo E. M; Jeffers T. H.; Ferguson and Stevenson, H. Q.: (1991) Environ. Sci. and Technol. 25, 1559-1565 3. Le Van M. D. and Vermeluen T.: (1981) Journ. of Phys. Chem. 85, 3247-3250 4. Marton G.; Dencs J.; Szokonya L. and Illés Z.: Hung. Journ. Ind. Chem. (1981), 9, 251-262