Introduction Chromic acid is a highly toxic, carcinogenic oxidizing agent and is a major component of the effluents from leather tanning, chromium plating, galvanization, textile and dye industries [1]. In view of the environmental damage caused by it, chromium recovery and waste water recycling is one of the important problems of separation. Membrane separation processes have an edge over all the other separation methods (precipitation, exchange, evaporation, adsorption, and liquid-liquid extraction) as they do not require any addition of chemicals, require lesser energy, have a selective and nearly total recovery of chromates as well as produce purified water. Styrene can be copolymerized with many monomers like acrylonitrile and butadiene (ABS), acrylonitrile (SAN) and divinyl benzene (DVB) [2]. The SAN copolymers with 5-50% of acrylonitrile in styrene have greatly improved resistance to chemical attack by many reagents, better mechanical strength and excellent solvent resistance [2]. A number of membranes using SAN have been made for pervaporation [3, 4] due to their solvent stability but very little literature exists in the preparation of membranes from this polymer for nanofiltration separations [5] where these need to be used at high pressures for the separation of dyes. The membrane reported in this work is a ceramic supported copolymer membrane of styrene and acrylonitrile which is tough and scratchproof and the preparation procedure is given in the following discussion. Porous ceramic supports were prepared from a paste of a mixture of clays (kaolin, ball clay, pyrophallite, calcium carbonate, feldspar and quartz) in water and casted on a gypsum surface in the form of a circular disc followed by drying for 24h at room temperature, then 24h at 1000C and finally for 24h at 2500C to ensure the slow removal of water. The dried supports are then calcined at 9000C for 8h and then finally polished using a silicon carbide abrasive paper (No. C-220) to get smooth and flat ceramic discs of diameter 64mm and 2-3mm thickness. A mixture styrene (6.073g, 0.0607gmole), divinyl benzene (3%) and acrylonitrile (10%, 0.607g, 3.06710-3gmole) using dual initiators consisting of Benzoyl Peroxide (0.082g, 3.38810-4), azobisisobutyronitrile (0.043g, 2.6210-4gmole) with an accelerator Dimethyl aniline (0.025g, 2.0710-4gmole) in presence of N2 is copolymerized at 700C in a water bath for 1h to obtain a polymer syrup of about 8% conversion. A known amount (2g) of the polymer solution is spread over the wet support and is allowed to stand at controlled temperature and humidity (330C and R.H 55%) for at least one hour. After the unreacted monomer and the solvent are evaporated, the composite membrane is heated at 700C for 3 hours in an oven to further crosslink the membrane polymer. A mixture of NO and NO2 (called NOx ) is generated in a 1.5 liter two neck flask by reacting sodium nitrate (10g, 0.14 gmole) with sulphuric acid (36N, sp.gr.1.18, 25 ml) in the presence of ferrous sulphate (5g, 0.072 gmole) [6, 7]. The composite membrane is placed inside the bottle and the NOx (approximately 500ml) is introduced through the silicon septum of the reactor by a syringe. The reactor is kept into the oven at 800 C and after the completion of the nitration reaction, the reactor is cooled to the room temperature. The nitrated membrane is further refluxed with 50% hydrazine hydrate/water mixture for six hour in a water bath so as to reduce the NO2 groups to amine (NH2) groups as shown in the following reaction scheme and has also has been confirmed by FTIR. The aminated membrane is further modified by refluxing it with 2% v/v 1, 2 dichloroethane solution in ethanol for four hours. The modified membrane is quaternized by refluxing it with 5% solution of triethylamine (TEA) in ethanol for 4 h. This modified membrane is a strong anion exchange membrane having quaternary ammonium groups. The membranes have been fully characterized using Scanning Electron Microscopy (SEM), Fourier Transform Infrared Spectroscopy (FTIR), Atomic Force Microscopy (AFM), contact angle measurements, water content and molecular weight cut off experiments. The FTIR spectra confirmed the presence of presence of nitro and amine functional groups. The reduction in contact angle with each modification showed that the membrane became more hydrophilic. The pore diameter obtained using MWCO was found out to be 6.26, 8.32 and 11.2 nm for the unmodified, aminated and quaternized membranes, respectively. Figure 1 shows that polymer membrane is adhered to the support surface without penetrating the ceramic support. Pure water permeation data has been obtained for the modified, nitrated, aminated and quaternized membranes. Figure 2 shows the pure water flux varies linearly with pressure drop for the unmodified and the modified membranes and obeys the Darcy's law (Jv = LpP). From this figure, we determined the pure water permeability and found it to be 4.5365 X 10-8 (ms-1 KPa-1) for the unmodified membrane, 1.06121 X 10-7 (ms-1 KPa-1) for the nitrated membrane, 2.9978 X 10-7 (ms-1 kPa-1) for the aminated membrane and 8.64124 X 10-7 (ms-1 KPa-1) for the quaternized. We can infer from this figure that the water flux of nitrated and aminated membranes are much higher than the unmodified membrane due to the increased hydrophilicity and pore size of the modified membranes as has also been observed by contact angle measurements and morphological studies. Figure 3 shows that the chromium (VI) salt rejections as a function of pressure drop for the unmodified, nitrated, aminated and quaternized membranes. The observed rejection for the unmodified membrane increases with pressure drop but on modification the trend changes and the rejection at first increases and then falls with pressure. Therefore, observed rejection is no longer a proper parameter to express the membrane characteristics. The permeate flux was also found to be lower than the pure water flux which implies that the rejection results have to be corrected to include the effects of concentration polarization. Thus a procedure was developed as outlined in [8] to find the real rejection to characterize the membrane performance. It was found that the chromate concentration on the membrane surface was 2-3 times higher than the concentration in the bulk. The intrinsic rejection always increased with pressure even though the trends obtained for observed rejection were much different. It was also found that the intrinsic rejection of the aminated membrane was lower than the unmodified and nitrated membranes as opposed to the observed rejection. This could be attributed to the much higher flux obtained for these membranes. It is also evident that the quaternized membrane had the maximum intrinsic rejection and maximum flux also. This can be explained on the basis of interaction of fixed membrane charge sites with ionic solutes. It can be seen that the intrinsic rejection increases with the increase in the applied pressure, which is typically found in the separation of electrolytes through charged membranes. For nitrated membrane the rejection is found to be less (determined in repeated experiments) compared to the unmodified membrane and this is likely to be due to an increase in the pore size. Rejection data for different concentrations of chromic acid was obtained for the modified membranes. The experiments were conducted for the concentration of 1000ppm, 500 ppm and 100 ppm for the quaternized and the aminated membrane as shown in figure 4. It was found that the observed rejection increased with the decrease in concentration as has been reported in literature [9]. This is mainly because the effect of Donnnan exclusion reduces with the increase in concentration of the electrolyte. In the case higher electrolyte concentration, the membrane charge is shielded resulting in a lower effective charge of the membrane and hence lower retention.

References 1. J. L. Kroschwitz, Kirk-Othmer Encyclopedia of Chemical Technology, Wiley, New York, 1991 2. C. A. Brighton, G. Pritchard, G. A. Skinner, Styrene Polymers: Technology and Environmental Aspects, Applied Scince Publishers Ltd., London 3. S. Mandal, V. G. Pangarkar, Pervaporative dehydration of 1-methoxy propanol with acrylonitrile based co-polymer membranes prepared through emulsion polymerization: a solubility parameter approach and study of structural impact, Journal of Membrane Science, 209 (2002) 53-66. 4. S. K. Ray, S. B. Sawant, J. B. Joshi, V. G. Pangarkar, Development of new synthetic membranes for the separation of benzene-cyclohexane meixtures by pervaporation: A solubility parameter approach, Industrial and Engineering Chemistry Research, 36 (1997), 5265-5276. 5. M. Joshi, A.K.Mukherjee, B. D. Thakur, Development of a new styrene copolymer membrane for recycling of polyester fibre dyeing effluent, Journal of Membrane Science, 189 (2001) 23-40 6. S. Roy Chowdhury, P. Kumar, P.K. Bhattacharya, A. Kumar, Separation characteristics of modified polysulfone ultrafiltration membrane using NOx, Separation and Purification Technology, 24 (2001) 217-282 7. S. Sinha, A. Kumar, Preparation of high capacity chloromethylated strong base anion exchange resin using NOx, Separation science and Technology, 37(2002) 895-919. 8. S. Ghose, C. Bhattacharjee, S. Datta, Simulation of unstirred batch ultrafiltration process based on reversible pore-plugging model, J. Membrane Science. 169 (2000) 29-38. 9. B. Chaufer, M. R.-Baudry, L. Guihard, G. Daufin, Retention of ions in nanofiltration at various ionic strength, Desalination, 104 (1996) 37-46