2896
Ind. Eng. Chem. Res. 2000, 39, 2896-2901
Wet Air Oxidation of Desizing Wastewater from the Textile Industry Lecheng Lei, Xijun Hu,* Guohua Chen, John F. Porter, and Po Lock Yue Department of Chemical Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong
Wet air oxidation (WAO) was applied to treat the desizing wastewater from natural and manmade fiber processing in a 2-L autoclave reactor. The range of operating temperatures examined was between 150 and 290 °C. The partial pressure of oxygen ranged from 0.375 to 2.25 MPa at a reference temperature of 25 °C. Variations in chemical oxygen demand (COD) and total organic carbon were monitored during each experiment and used to evaluate the performance of the reaction process. Experimental results showed that WAO is an efficient method for the treatment of desizing wastewater. A higher COD removal was achieved under gentler reaction conditions with the aid of a catalyst. A two-parameter mathematical model involving a fast reaction and a slow reaction was used to describe the WAO reaction kinetics and to calculate the reaction activation energies. Introduction Wastewater discharged from the desizing processes of the textile industry is characterized by its very high chemical oxygen demand (COD) and is one of the main pollution sources in the textile industry. The desizing wastewater from natural fiber processing operations mainly contains starch, glucose, and yeast and has a COD value of 10000-20000 mg/L and biological oxygen demand (BOD) from 5000 to 10 000 mg/L. Man-made fiber desizing wastewaters typically have poly(vinyl alcohol) (PVA) as the major content, at COD levels of between 10 000 and 40 000 mg/L, as well as a relatively small amount of BOD (500-1000 mg/L). The low BOD/ COD ratio means that this kind of wastewater is very difficult to biodegrade. To comply with the current Hong Kong discharge regulations, factories have to reduce their COD and BOD values to 2000 and 800 mg/L, respectively, before discharging to sewers. In general, there are four methods available for wastewater treatment: chemical oxidation, biochemical oxidation, incineration, and wet air oxidation. Chemical oxidation is to oxidize the pollutants in the wastewater into carbon dioxide and water with strong oxidizers, such as ozone and hydrogen peroxide. The chemical oxidation process is relatively simple to operate but is only cost-effective for pollutants at low concentrations. It would become expensive for high COD wastewater such as that discharged from the desizing process. Biological oxidation is a commonly applied method in wastewater treatment. However, it is not suitable for wastewater with COD above 10 000 mg/L. The large quantity of sludge produced by this method also imposes further problems. Incineration may convert the pollutants effectively to innocuous end products at temperatures above 700 °C. It is not economically attractive unless the COD of the waste is above 100 000 mg/L because a large amount of energy will be wasted in boiling and heating the water (steam) if the organic concentration is not high enough. Wet air oxidation (WAO), which mineralizes the pollutants at elevated temperatures (200-350 °C) and pressures, has the * To whom all correspondence should be addressed. Email:
[email protected]. Fax: (852) 23580054. Tel: (852) 2358 7134.
ability to convert most organic compounds into carbon dioxide and water.1,2 Compared with traditional technologies, the wet air oxidation process has the advantage of high reaction rate, small quantity of secondary pollutants, and less space requirement. Thus the WAO process might be an economically feasible method for the treatment of desizing wastewater. Dating back to the early 20th century, wet air oxidation has been used for treating spent sulfite liquor from paper mills.3 And there are over 200 Zimpro WAO units in operation at more than 160 locations throughout the world.4 However, its application has been limited due to its severe operation conditions; i.e., it usually runs under high pressure and temperature. The purpose of this work is to investigate the WAO treatment of desizing wastewater, to lower the reaction temperature and pressure by adding a catalyst, and to study the reaction kinetics. Experimental Section A schematic diagram of the wet air oxidation (WAO) system, which has been used to treat dyeing and printing wastewater in our previous studies,5,6 is shown in Figure 1. The WAO system has a gas supply, a 2-L reactor, a wastewater charging system, and a liquid sampling point. The operation procedures of WAO process are described below. First the reactor was preheated to 60-70 °C. The wastewater (1.4 L) was then charged into the reactor using the water pump. The space occupied by gas inside the reactor was 600 mL. Pure nitrogen was used to purge the air inside the reactor at a total pressure of 1 MPa for 2 min after which the system was isolated. The reactor was heated to the desired reaction temperature (between 150 and 290 °C) which took between 1 and 2 h. Once the selected reaction temperature was reached, pure oxygen was introduced into the reactor and the reaction was assumed to start (time t ) 0). At designated time intervals thereafter, liquid samples were taken from the reactor and analyzed for COD and total organic carbon (TOC) contents. The system pressure is kept constant by supplying oxygen to balance the pressure drop caused by sampling. At the end of the reaction time, the heating jacket was turned off and the reactor was cooled for half
10.1021/ie990607s CCC: $19.00 © 2000 American Chemical Society Published on Web 06/20/2000
Ind. Eng. Chem. Res., Vol. 39, No. 8, 2000 2897
Figure 1. Schematic diagram of the wet air oxidation system.
Figure 3. Effect of oxygen on the WAO of cotton desizing wastewater at 290 °C. Figure 2. COD and TOC reductions during the reactor heating period for cotton desizing wastewater.
an hour. Tap water was then passed through the cooling coil to further cool the reactor to 60-70 °C. Pure nitrogen was passed through the reactor to discharge the treated wastewater. The reactor was filled with deionized water, and the stirrer was operated at 1000 rpm for 5 min. This final step was repeated three or four times until the discharged water was visually clean. The COD value is assayed with an HACH-DR2000 COD analyzer. The analytical agent, which consists of 0.05 g of HgSO4, 2.5 mL of 98% H2SO4, 0.5 mL of 1 mol/L K2Cr2O5, is mixed with 2 mL of wastewater sample and incubated at 150 °C for 2 h. COD is then assayed after cooling with the COD analyzer. All the chemicals used are in analytical purity grade. The desizing wastewaters investigated in this study were provided by two Hong Kong textile companies. The first was from a cotton desizing process and contained primarily starch and glucose. The COD, BOD, and TOC values for the cotton desizing wastewater were 10280, 3754, and 3558 mg/L, respectively. The second was from a man-made fiber desizing process, and poly(vinyl alcohol) (PVA) was its major component with COD, BOD, and TOC values of 12600, 620, and 3824 mg/L, respectively. Results and Discussion COD and TOC Removals Caused by Heating. Since heating the reactor to the desired reaction temperature took from 1 to 2 h, the COD and TOC variations during the heating period in the absence of oxygen were examined first. Figure 2 shows the COD and TOC reductions of the cotton desizing wastewater when it was heated from ambient temperature to 290 °C. The reduction or removal is defined as the difference between the initial COD or TOC value and its value at a given time divided by its initial amount. It was seen
that there was little change in COD or TOC when the temperature was below 240 °C. When the wastewater was further heated from 240 to 290 °C, there was a significant COD reduction. A COD reduction of 24% and a TOC removal of 8% could be achieved by simple thermal decomposition at 290 °C.7 This is the reason the COD or TOC reduction at zero reaction time has a finite number instead of zero in the following figures. The reduction of COD is more prominent than that of TOC during thermal decomposition. This is due to the pyrolysis producing hydrogen gas that might be lost from the solution at high temperature, which does not affect the TOC value. Effect of Oxygen on the WAO Process. However, by keeping the reactor at 290 °C for a longer timesup to 150 minslittle further reduction in COD or TOC was observed in the absence of oxygen, as shown in Figure 3. Figure 3 also shows that a significant increase in both COD and TOC removal (over 75%) was observed in the presence of oxygen at a partial pressure of 1.5 MPa (at a reference temperature of 25 °C)swhich is just the required stoichiometric amount to completely mineralize the organic carbon in the cotton desizing wastewater. Five different levels of initial oxygen partial pressure, 0.375, 0.75, 1.125, 1.5, and 2.25 MPa, all at a reference temperature of 25 °C, which correspond to 25%, 50%, 75%, 100%, and 150% of the theoretical requirement of oxygen, were tested to study the effect of oxygen partial pressure on the performance of the WAO process. The stoichiometric (theoretical) amount is defined as the amount of oxygen needed to chemically mineralize all the pollutants in the wastewater. The WAO treatment of natural fiber desizing wastewater under different initial oxygen partial pressures is shown in Figure 4. The COD and TOC removal rates were not surprisingly found to increase with oxygen partial pressure as the concentration of dissolved oxygen in the liquid phase is proportional to the corresponding oxygen partial pressure.
2898
Ind. Eng. Chem. Res., Vol. 39, No. 8, 2000
Figure 4. Effect of oxygen partial pressure on the WAO of cotton desizing wastewater at 240 °C.
In Figure 4 we noted that the COD removal exceeded the oxygen supplied at low oxygen partial pressure. For example, 48% of COD removal is achieved with 25% of oxygen supplied. The reason for this phenomenon is that the removal of COD is the overall result of oxidation reaction and thermolysis.7-9 Another reason is the regular additions of oxygen to maintain the reaction pressure after sampling so that in fact the COD removal is not in excess of the actual oxygen supply; i.e., the actual oxygen supply is more than 25%. This phenomenon begins to disappear when the oxygen supplied is more than 75% of the theoretical amount. Effect of Reaction Temperature on the WAO Process. The WAO treatment of natural fiber desizing wastewater was further examined at five temperatures, with the supply of oxygen fixed at 1.5 MPa, which is equivalent to the stoichiometric amount of 100% COD removal. The rates of WAO reaction and COD removal increase with the reaction temperature, as shown in Figure 5. There is no significant COD or TOC removal at a temperature of 150 °C. A COD removal of near 80% is obtained after 100 min of reaction at 290 °C. Catalytic WAO Process. Although the WAO reaction is very successful in the treatment of natural fiber desizing wastewater, it must be operated under relatively severe conditions in order to achieve a high COD removal rate. It is necessary to improve the process to reduce the reaction temperature and pressure. Adding some catalyst in the WAO process is one of the most popular options.5,10-12 In this study, transition metals in the form of sulfates and nitrates were used as the catalysts. The amount of catalyst added into the reaction system was 100 mg/L in terms of metal ion concentration. Again, the oxygen supply was fixed at the theoretical requirement, 1.5 MPa at a reference temperature of 25 °C. The reaction temperature was set at 240 °C. The catalytic wet air oxidation (CWAO) treatment of cotton desizing wastewater is shown in Figures 6 and 7. The addition of catalysts significantly enhances the
Figure 5. Effect of reaction temperature on the WAO of cotton desizing wastewater at 1.5 MPa partial oxygen pressure.
Figure 6. Effect of metal sulfates on the WAO of cotton desizing wastewater at 240 °C and 1.5 MPa partial oxygen pressure.
removal rates of COD and TOC. The catalytic activity of nitrates is slightly higher than that of sulfates, and Cu(NO3)2 is the best catalyst achieving a COD removal of 90%. WAO of Desizing Wastewater from a Man-Made Fiber Process. Having shown WAO to be a useful tool in the treatment of cotton desizing wastewater, its applicability to the treatment of desizing wastewater from a man-made fiber processing operation was examined. The initial COD, BOD, and TOC values of this
Ind. Eng. Chem. Res., Vol. 39, No. 8, 2000 2899
Figure 7. Effect of metal nitrates on the WAO of cotton desizing wastewater at 240 °C and 1.5 MPa partial oxygen pressure.
wastewater were 12 600, 620, and 3824 mg/L, respectively. The extremely low BOD/COD ratio (0.05) implies that this wastewater is very difficult to be treated by biological methods. Figure 8 shows the effect of temperature on the WAO treatment of this wastewater at an oxygen partial pressure of 2 MPa and four different reaction temperatures. A higher oxygen pressure (2 MPa) was used here because the chemical fiber desizing wastewater had a larger COD value so more oxygen was required to completely oxidize the waste content. The choice of 2 MPa was again based on providing the system with the required stoichiometric amount to mineralize the organic carbon in the wastewater. Similar to cotton desizing wastewater, the COD and TOC removals were improved by increasing reaction temperature. At 270 °C, 90% COD reduction and 80% TOC removal were achieved after 2 h. Figure 8 also shows the biodegradability change for the man-made fiber desizing wastewater. The biodegradability is defined as the BOD to COD ratio. After the WAO treatment, the biodegradability was significantly enhancedsmore than 80% was obtained at a reaction temperature of 270 °C for 2 h. An increase in the reaction temperature helps to enhance the biodegradability of wastewater. The increase in BOD to COD ratio is due to the decomposition of the large relatively biologically stable molecule (PVA) into smaller more biodegradable molecules. This suggests that WAO should be a suitable pretreatment step before biological degradation. As previously shown in Figure 2 for the natural fiber desizing wastewater, some degree of thermal decomposition was also observed for this system. Once again, different initial COD and TOC reductions were observed at different reaction temperatures. However, the biodegradability could not be improved by simply heating the wastewater. This is evident from Figure 8 where the BOD to COD ratios have the same initial value for different reaction temperatures.
Figure 8. Effect of reaction temperature on the WAO of manmade fiber desizing wastewater at 2 MPa partial oxygen pressure.
Reaction Kinetics. The reactions during WAO are very complex in terms of many intermediate and ultimate products.13 No attempt is made to provide a detailed reaction analysis on individual compounds. Instead a brief discussion on the kinetics study of COD reduction will be given. Since the reactor was well stirred and the oxygen supply was maintained at the stoichiometric requirement and more oxygen was given after each sampling, the mass transfer controlled regime is eliminated and the dissolved oxygen concentration is assumed constant for a given partial oxygen pressure. The rate data were modeled by first-order kinetics as in the following equation, with the “reactant” concentration given as COD
-
dCOD ) kCOD dt
(1)
where t is reaction time and k is the specific reaction rate constant which has the following temperature dependency
k ) k0 exp(-E/RT)
(2)
where k0 is the pre-exponential factor, E is the activation energy, R is the universal gas constant, and T is the temperature in kelvin. Integrating eq 1 gives
ln(COD0/COD) ) kt
(3)
where COD0 is the initial COD value. By plotting ln(COD0/COD) versus time, the slope is the specific
2900
Ind. Eng. Chem. Res., Vol. 39, No. 8, 2000
Figure 9. WAO of cotton desizing wastewater at 1.5 MPa partial oxygen pressure.
Figure 10. Effect of temperature on rate constants of cotton desizing wastewater at 1.5 MPa partial oxygen pressure.
reaction rate constant k. A typical plot of the WAO treatment of cotton desizing wastewater at a fixed partial oxygen pressure of 1.5 MPa and four different reaction temperatures (Figure 4) is shown in Figure 9. The data fit well into two straight lines for a given temperature, indicating that oxidation proceeds in two distinct steps: a fast initial reaction of large molecules decomposed into intermediate products, carbon dioxide, and water, followed by a slow reaction of further oxidizing the intermediate products into end products of carbon dioxide and water. To calculate the activated energies, eq 2 was transformed into a logarithmic form, which was plotted in Figure 10. As expected, the rate constant increases with increasing temperature, but this temperature effect is more significant for the fast reaction step. As the slow reaction, the reaction rate is not very sensitive to the change of temperature. This indicates that the fast reaction has a high activation energy, while the slow reaction has a small activated energy. The calculated activated energies are Efast ) 30 kJ/mol and Eslow ) 9.2 kJ/mol. It should be remembered that the rate constant in eq 1 is the product of the true specific rate constant and the dissolved oxygen concentration in water which is a function of the partial oxygen pressure, i.e. n
k ) k′[PO2]
(4)
The rate constant k can be reasonably assumed as
Figure 11. WAO of cotton desizing wastewater at 240 °C.
Figure 12. Effect of oxygen concentration on rate constants of cotton desizing wastewater at 240 °C.
constant for a given oxygen partial pressure only when oxygen is in excess. Otherwise as the partial oxygen pressure changes, the rate constant k may vary. This is shown in Figure 11 for the WAO of cotton desizing wastewater at 240 °C (Figure 4). The partial oxygen pressure varies from 0.375 to 2.25 MPa. The extracted rate constants versus partial oxygen pressure are shown in Figure 12. Since the stoichiometric oxygen requirement for complete oxidation of the cotton desizing wastewater is 1.5 MPa at 25 °C, it has a significant effect on the rate constant of the fast reaction if the oxygen partial pressure is less than this amount. However, when oxygen is in excess (higher than the stoichiometric oxygen demand), the rate constant becomes independent of the partial pressure of oxygen because the dissolved oxygen concentration can be considered as constant in this case. The slow reaction stage has a weaker dependency on the oxygen partial pressure for the whole pressure region considered indicating that the oxygen requirement of the slow reaction step is much less than that of the fast reaction. It also indicates that the intermediate products are difficult to be oxidized even at a large excess of oxygen. Conclusions Wet air oxidation is found to be an effective method for treating the desizing wastewater from both cotton and man-made fiber processing operations, one of the major COD contributors in the textile industry wastewater. More than 90% COD reduction and 80% TOC
Ind. Eng. Chem. Res., Vol. 39, No. 8, 2000 2901
removal can be obtained by properly choosing the reaction conditions. Increasing the reaction temperature and pressure helps to enhance the WAO efficiency. However, a certain portion of the organic pollutants could not be removed even at elevated temperature and prolonged reaction time. The performance of the WAO process can be improved by adding some suitable catalyst. WAO was also demonstrated to be a suitable pretreatment method before biodegradation because the BOD/COD ratio of desizing wastewater was significantly increased by WAO treatment. A preliminary investigation of reaction kinetics shows that wet air oxidation of desizing wastewater follows two steps, a fast reaction followed by a slow reaction stage. Acknowledgment The authors are grateful to the Industry and Technology Development Council of Hong Kong for their financial support. Nomenclature COD ) chemical oxygen demand, a parameter for wastewater characterization, mg/L COD0 ) initial COD value, mg/L E ) activation energy, kJ/mol k ) reaction constant k0 ) pre-exponential factor in the Arrhenius equation R ) gas constant, 8.314 J/(mol K) t ) reaction time, min T ) reaction temperature, °C or K TOC ) total organic carbon, a parameter for wastewater characterization, mg/L
Literature Cited (1) Dietrich, M. J.; Randall, T. L.; Canney, P. J. Wet Air Oxidation of Hazardous Organics in Wastewater. Environ. Prog. 1985, 4, 171-177. (2) Randall, T. L.; Knopp, P. V. Detoxification of Specific Organic Substances by Wet Oxidation. J. WPCF 1980, 52, 21172129. (3) Baillod, C. R.; Faith, B. M.; Masi, O. Fate of Specific Pollutants During Wet Oxidation and Ozonation. Environ. Prog. 1982, 1, 217-226. (4) Mishra, V. S.; Mahajani, V. V.; Joshi, J. B.; Wet Air Oxidation. Ind. Eng. Chem. Res. 1995, 34, 2-48. (5) Lei, L.; Hu, X.; Chu, H. P.; Chen, G.; Yue, P. L. Catalytic Wet Air Oxidation of Dyeing and Printing Wastewater. Water Sci. Technol. 1997, 35, 311-319. (6) Lei, L.; Hu, X.; Yue, P. L. Improved Wet Oxidation for the Treatment of Dyeing Wastewater Concentrate from Membrane Separation Process. Water Res. 1998, 32, 2753-2759. (7) Skaates, J. M.; Briggs, B. A.; Lamparter, R. A.; Baillod, C. R. Wet Oxidation of Glucose. Can. J. Chem. Eng. 1981, 59, 517521. (8) Pruden, B. B.; Le, H. Wet Air Oxidation of Soluble Components in Wastewater. Can. J. Chem. Eng. 1976, 54, 319-325. (9) Li, L.; Chen P.; Gloyna, E. F. Generalized Kinetics Model for Wet Oxidation of Organic Compounds. AIChE J. 1991, 37, 1687-1697. (10) Levec, J. Catalytic Oxidation of Toxic Organic in Aqueous Solution. Appl. Catal. 1990, 63, L1-L5. (11) Hu, X.; Lei, L.; Chu, H. P.; Yue, P. L. Copper/Activated Carbon as Catalyst for Organic Wastewater Treatment. Carbon 1999, 37, 631-637. (12) Chu, H. P.; Lei, L.; Hu, X.; Yue, P. L. Metallo-Organic Chemical Vapor Deposition (MOCVD) for the Development of Heterogeneous Catalyst. Energy Fuels 1998, 12, 1108-1113. (13) Hao, O. J.; Phull, K. K. Wet oxidation of Nitrotoluenesulfonic Acid: Some Intermediates, Reaction Pathways, and Byproducts Toxicity. Environ. Sci. Technol. 1993, 27, 1650-1658.
Received for review August 9, 1999 Revised manuscript received March 30, 2000 Accepted April 18, 2000 IE990607S