TREATMENT OF DISTILLERY WASTEWATER

Abstract

The Sugar Cane molasses based alcohol distilleries using fermentation process generate about 10-15 dm3 of wastewater per dm3 of ethyl alcohol produced. The distillery wastewater (DWW) which is also called as distillery spent wash (DSW), slop, vinasse, stillage etc., is dark brown in colour and has very high chemical oxygen demand (COD) (60-200 kg/m3) and very high biochemical oxygen demand (BOD) (50- 75 kg/m3). High COD and BOD are due to the presence of reduced sugars, dissolved lignin, proteins, inorganics, etc. in high concentrations. The concentration-incineration and the anaerobic digestion are widely applied in distilleries to reduce the organic load of the effluent alongwith recovery of its energy content. Most of the distilleries in India use anaerobic treatment systems, usually upflow anaerobic sludge blanket (UASB) reactor and their different variants. Some distilleries also use biphasic (acidogenic followed by methanogenic) reactors. These systems are installed to recover maximum amount of energy in the form of methane rich-gas. These reactor systems operate at around 80 - 90% BOD removal efficiency and around 70% COD removal efficiency. Thus, the effluents of these treatment systems still contain very high COD (~ 30 -45 kg/m3) and very high BOD (~ 4.5 - 7 kg/m3). This anaerobically treated effluent is further treated aerobically, however, the effluent still does not meet the effluent discharge standards for sewers or surface water bodies. Therefore, efficient and cost effective process is needed to treat these effluents. The present study aims to explore suitable treatment methods for the removal of COD (and consequently also BOD) of the DWW and the biodigester effluent (BDE). The coagulation and flocculation at room temperature, and catalytic thermal treatment (or catalytic thermolysis) at moderate temperatures and autogenous pressures with subsequent catalytic/non catalytic wet oxidation have been undertaken for studies on BDE. For coagulation studies, various coagulants like AICI3, 11 FeCl3.6H20 and commercial poly aluminium chloride (PAC) have been used. For catalytic thermolysis and catalytic wet oxidation, different catalysts like CuS04 and metal oxides like CuO, MnO, ZnO and composite metal oxides have been used. Catalytic thermolysiswas also attemped for the DWW. The performance of the catalysts for the thermolysis and wet oxidation was first tested at atmospheric pressure in a 0.5 dm3 atmospheric pressure glass reactor (AGR). The catalysts whose performances were found to be good at atmospheric pressure study were selected for high pressure study in a 1dm3 stainless steel (SS-316) reactor (SSR). The AGR was equipped with a vertical condenser. The SSR was equipped with electrical heating, temperature indicator-cum-controller, liquid sampling port, pressure indicator, and a cooling coil. The reactor contents were agitated using a magnetic stirrer. After the start of an experimental run at a desired temperature, the effluent samples were withdrawn from the reactor at definite time intervals. The samples were filtered and the filtrate was analyzed for its COD and other parameters using standard methods. For coagulation and flocculation, 0.20 dm3 ofBDE was taken in a 0.50 dm3 glass beaker. The pH of the effluent was noted and the initial pH (pHo) was adjusted by adding 1Maqueous solution of NaOH / H2S04. A known amount of the coagulant was added to the effluent and flash-mixed for 5minutes, and, thereafter, gently mixed for 15 minutes. The effluent sample was then taken in a glass cylinder and kept quiescent for 8 h. The supernatant liquor was centrifuged and analysed for its COD value and other parameters. The catalysts were also characterized by XRD,AASand SEM. For screening the catalysts and the optimization of pHo for thermolysis, the experiments were conducted for 3 h treatment time (tR) at 100 °C and atmospheric pressure and at different initial pH (pHo) in the range of 1-10. The homogeneous copper sulphate and the heterogeneous metal oxide catalysts, such as, ZnO, CuO, Mn/Ce composite oxide and Cu/Mn composite oxides were used for the catalytic thermolysis. Since CuO was found to be the best among these catalysts, it was used in further m studies. The COD reduction due to thermolysis was studied for the distillery wastewater (DWW) and the biodigester effluent(BDE) at 80-100 °C and atmospheric pressure in the presence of various catalysts. The results obtained with CuO catalyst were: 47 % COD reduction and 68 % colour reduction for DWW and 61 % COD reduction and 78 %colour reduction for BDE at 100 °C, 4 kg/m3 catalyst loading in 12 h treatment period. The experimental data on the catalytic thermolysis exhibit two clearly distinct COD reduction phases: the first phase having fast rate followed by the second phase with a slower rate. During the first phase, it is expected that the large molecules in organic substrate break down to smaller molecules. The thermolysis of smaller molecules in the second phase appears to be relatively difficult due to recalcitrance of the residual dissolved organics and loss of catalysts. Catalytic thermolysis with 3 kg/m3 CuO at 100 °C and atmospheric pressure shows about 1.6 and 1.5 times higher COD reductions in comparison to that without using a catalyst for the DWW and BDE, respectively. Since CuO catalyst was found to be the best for thermolysis, further studies at moderate temperature and pressure were carried out using this catalyst. The thermolysis of BDE and DWW was performed in the presence of CuO catalyst in a 1dm3 SSR in a batch mode in the temperature range of100 - 140 °C and autogenous pressure (0.1 - 0.9 MPa) with CuO mass loading between (Cw) in the range of 2 - 5 kg/m3. The pHo is found to have profound impact on the efficiency of thermolysis on COD removal. pHo 2 and pHo 1 are found to be the optimum for thermolysis ofDWW and BDE, respectively. At 140 °C with 3kg/m3 Cw and pH0 2, a maximum of 60% COD ofDWWcould be reduced. At the same operating conditions and pHo 1, 70% COD reduction ofBDE is obtained. The removal efficiency for COD and colour is better than those reported byother investigators (Daga et al., 1986; Lele et al., 1989; Dhale and Mahajani, 2000). It is also seen that the COD reduction is quite fast during the transient preheating period (th) (i.e. the time taken for raising the temperature ofthe solution from ambient temperature to the treatment temperature) and iv the initial 2 h treatment time (tR) and, thereafter, the COD reduction proceeds very slowly. During preheating periods, thermal degradation/precipitation occurs. The catalytic thermolysis of both DWW and BDE follows a two step mechanism: the first step being much faster than the second step. The slurry obtained after the thermolysis has very good filtration characteristics. The removal of the molasses-derived colour and chemical oxygen demand from the biodigester effluent of a molasses-based alcohol distillery effluent treatment plant was also studied using inorganic coagulants, viz., FeCb, AICI3 and polyaluminium chloride (PAC). The coagulation/ flocculation yield about 55, 60 and 72 percent COD reductions and about 83, 86 and 92 percent colour reductions, with the use of 60 mM/1 AICI3, 60 mM/1 FeCb and 30 ml/1 of polyaluminium chloride, respectively, at their optimum initial pH. The pH of the effluent-coagulant mixture plays a very significant role in the coagulation/flocculation process, with pHo 5.5 being the optimum for PAC. Similar levels of COD and colour reductions were obtained by Migo et al. (1993), but the coagulant doses used were much higher. Superior performance of PAC may be due to its polymeric structure. The multivalent Al cations of the PAC coordinate with the anions present in BDE and result in complexation. The gel structure of the PAC also enmeshes the organics present in BDE. Thus, the complexation (and consequent precipitation) and the capture of the organics in the gel are responsible for the higher COD/colour reduction by PAC as compared to those obtained with FeCfj and AlC^.The solid residue, obtained by filtration and drying from the use ofPAC has specific energy of 13.4 MJ/kg and can be used as a medium energy fuel material. The filtration characteristics of the flocculated effluent are poor. High COD reduction of the wastewater by flocculation with PAC may be a good alternative and/or a supplementary method to the conventional aerobic treatment process of the biodigester effluent. The wet oxidation of the filtrate of the thermally treated BDE was also carried out. Various catalysts like CUSO4, CuO, Mn/Ce composite oxides and Cu/Mn composite oxides were tested for the COD reduction of the effluent at atmospheric pressure. The Cu/Mn(l/1) composite oxide was found to be the best for oxidation at atmospheric pressure, and therefore, it was selected for further studies at higher temperature and pressure. The effect of various parameters like pHo (1.5 - 10), temperature (120-180 °C), p0j (0.2-0.8 MPa) with catalyst mass loading (2 - 5kg/m3) on WO were also studied. The pHo was found to affect the wet oxidation considerably. Most of the catalysts have performed well at pH0 < 2 and pHo = 6-8. The thermally treated BDE still contains some proteins, lignin and carbohydrates. Several reactive groups including hydroxyl groups contained in lignin and carbohydrates, generally react at pH 0-2 and pH 6-8. For Cu/Mn (1/1) composite oxide catalyst the optimum COD and BOD reduction were 80 % and 87 %, respectively, and the colour reduction was 88 % at T = 140 °C, p0 = 0.4 MPa and Cw = 3 kg/m3. The increases in oxygen partial pressure upto 0.8 MPa, catalyst loading upto 5 kg/m3 and temperature upto 180 °C does not significantly increase the COD reduction. The wet oxidation (WO) of the filtrate with its residuel CuO was also found to considerably reduce the organic content of the wastewater with 69% COD reduction, 82 % BOD reduction and 83. % colour reduction at T= 140 °C, p0z =0.4 MPa. The two step processes: catalytic thermolysis at T = 140 °C, pHo =1, Cw(CuO) = 3kg/m3 followed by wet oxidation in the absence of any fresh catalyst (at T = 140°C, p0 = 0.4 MPa, pHo = 1.5)gave an overall 90.6% COD reduction, 96% BOD reduction and 95 % colour reduction which further increased to 94 % COD reduction, 97.8 % BOD reduction and 97 % colour reduction when wet oxidation was performed in the presence of Cu/Mn (1/1) composite oxide (3 kg/m3) catalyst. The wet oxidation was found to under go a two step process, first faster step followed by a slower second step. A first order reaction with respect to COD for wet oxidation with or without the addition of fresh catalysts in both the steps was observed. A first order two step COD reduction was also reported for wet oxidation ofDWW/BDE by VI previous investigators (Chowdhury and Ross, 1975; Daga et al., 1986; Lele et al., 1990; Dhale and Mahajani, 2000). In the present study, the reaction rate constants, are found to be higher than that obtained by these investigators. Mishra et al. (1995) have also reported first order two step kinetics for the oxidation of various organics. The order with respect to oxygen partial pressure p0 was evaluated as 0.734 for the first step reaction and 0.862 for the second step reaction for non-catalytic oxidation. For catalytic wet oxidation, it was found to be 0.292 for the first step and 0.826 for the second step of the reaction. On the basis of above results, it is concluded that the thermolysis and coagulation are the effective processes for the treatment of distillery effluents. The thermally treated BDE can be wet oxidized effectively.