Ozone: Science and Engineering, 28: 3–8 Copyright # 2006 International Ozone Association ISSN: 0191-9512 print / 1547–6545 online DOI: 10.1080/01919510500479007 Ozonation of a Complex Industrial Effluent: Oxidation of Organic Pollutants and Removal of Toxicity A. A. Lima,1 A. F. Montalvao,2 M. Dezotti,1 and G. L. Sant’Anna Jr.1 1 2 Programa de Engenharia Quı́mica/COPPE/Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brazil Praxair White Martins, Gases Industrials, Rio de Janeiro, Brazil This work deals with the ozonation of a chemical industry wastewater, which contains many complex organic pollutants and presents high chloride content and toxicity. Batch experiments were carried out until ozone absorption reached 0.1 to 3 gO3/L. Ozonation promoted low to moderate levels of organic matter removal even when high ozone doses were applied. Organic matter removal tended to decrease when chloride content increased. Toxicity removal increased with the ozone dose up to 0.5 gO3/L. High or complete toxicity removal was attained after ozonation of the wastewater samples. Ozonation was also very effective in removing the polycyclic aromatic compounds found in the wastewater. Keywords Ozonation, Industrial Wastewater, Toxicity, Saline Effluents INTRODUCTION The chemical industries produce very complex effluents, which may contain a large and variable number of organic compounds, high salinity and sometimes toxicity. In such cases, the wastewater treatment process requires large equalization tanks and several treatment steps in order to produce a treated effluent with the required quality to be discharged in the natural receiving waters. Ozonation may be successfully used to remove nonbiodegradable or recalcitrant compounds that may impair toxicity to industrial effluents and has been applied to treat pulp and paper mill effluents (Zhou and Smith, 2000), landfill leachate (Baig et al., 1999), food industry wastewater (Beltran-Heredia et al., 2000), petrochemical industry effluents (Lin et al., 2001), wineReceived 05/27/2004; Accepted 11/15/2004 Address correspondence to G. L. Sant’Anna, Programa de Engenharia Quı́mica/COPPE/Universidade Federal do Rio de Janeiro, P.O. Box 68502, CEP 21945-970 Rio de Janeiro, RJ, Brazil. E-mail: lippel@peq.coppe.ufrj.br distillery wastewaters (Beltran et al., 2001) and textile industry effluents (Perkowski et al., 2000). Some industrial effluents present a high chloride content, which may impair the ozonation process. The effect of high chloride concentrations (above 10,000 mg/L), present in very complex chemical matrices, on the ozonation process, has not been deeply investigated. Chloride concentrations higher than 30,000 mg/L were shown to be deleterious to the ozonation of synthetic effluents containing phenol (Bessa, 2003). There is a general concern about the treatment of saline wastewaters since the production of such effluents will increase in the near future in view of the procedures adopted by the industries to reuse water and reduce water consumption. The aim of this work was to investigate the potential of the ozonation process to remove the acute toxicity of a complex industrial wastewater, presenting chloride concentration in the range of 2,000 to 27,000 mg/L. Removal of polycyclic aromatic hydrocarbons and improvement on the effluent biodegradability were also investigated. MATERIALS AND METHODS Wastewater Bayer industry supplied the industrial effluent used in this work. This chemical and pharmaceutical industry, located at Belford Roxo (RJ, Brazil), produces many chemicals, including polyurethane and agricultural and veterinary products. Grab samples of the industrial effluent were collected from the mixing vessel located downstream the equalization tanks, during an 8-month period. These samples were transferred to the laboratory and kept under refrigeration (4 °C) until use. Ozonation Tests were carried out in a pilot ozonation column (10 cm diameter; 100 cm height) made of Plexiglas. Gas was Ozonation of a Complex Industrial Effluent February 2006 3 injected in the bottom of the column through a stainless steel porous diffuser (10 mm pores). The ozonizer, manufactured by PCI (Charlotte, NC, USA), was fed with pure oxygen (99.9%) at a flow rate of 0.94 Nm3/h, generating a stream of 3% ozone. The maximum ozone production capacity was 40 g/h. The experiments were carried out in batch until ozone absorption reached 0.1, 0.2, 0.3, 0.5, 1, 2, and 3 g of O3 absorbed per liter of effluent. Ozone concentration was measured at the gas flow inlet and outlet. The amount of ozone absorbed was automatically calculated using the ozone analyzers (PCI) outputs and the software developed by White Martins Co. (Brazil). All the experiments were conducted at room temperature and at the original pH of the effluent. (HPLC) according to established methods (FEEMA, 1990; APHA et al., 1998). Acute toxicity was determined using the Microtox Analyzer, model M500 (Strategic Diagnostics, Newark, DE, USA), and the genetically modified luminescent bacteria K-70 PDB 101 (Bayer), using a period of exposition of 15 min. Toxicity results were expressed as EC50(%). RESULTS AND DISCUSSION Wastewater Characterization The characteristics of the raw wastewater samples, in terms of chloride concentration, organic matter content and toxicity, are shown in Table 1. There is a pronounced variation on wastewater characteristics from sample to sample. High levels of chloride (above 10,000 mg/L) were observed in most samples. The ratio BOD5/COD varied from 0.20 to 0.51 (average 0.34), indicating the presence of some recalcitrant organic matter in the effluent samples. The organic matter was almost completely dissolved in the wastewater since the suspend solids content was very low (< 20 mg/L). The contents of ammonium nitrogen and total phenols varied from 16 to 46 mg/L and 0.6 to 6.9 mg/ L, respectively. Most of the samples presented a pH of 12 and high levels of toxicity. Samples were considered severely toxic when EC values were smaller than 50%. Effluents with EC values below this limit are extremely harmful to the biological treatment performance, as established by the technical staff of Bayer Industry. All samples, except sample 4, presented EC values below that critical level, showing that toxicity is a common attribute of the industrial effluent under investigation. Biodegradation Tests To investigate the effect of ozonation on the biodegradability of the wastewater, batch assays were carried out as follows: 400 mL of raw effluent or ozonated effluent (ozone dose of 0.5 g/L) were placed into 1 L flasks, containing 0.1 L of adapted activated sludge. The sludge obtained from the industrial biological treatment unit was gradually acclimated to the ozonated effluents, filling the batch (fill and drain) reactor every day with increasing amounts of ozonated effluent in the feed mixture (ozonated plus raw effluent). Sludge adaptation was accomplished after 12 days of sequencing batch operation. Aeration and agitation was assured by the air bubbles generated in a porous diffuser placed in the bottom of the flask. The air flow rate was enough to keep the dissolved oxygen concentration in the liquid phase above 3 mg/L. At several time intervals samples were withdrawn from the flasks and filtered. The organic matter content of the filtered sample was determined as COD and DOC. Ozonation The effect of ozone dose on the removal of organic matter is illustrated in Figures 1a and 1b for two typical cases. The oxidation profiles were different for each sample but, in general, COD and DOC removal increased only slightly with an increasing amount of ozone absorbed. The ozone dose of 0.5 g/L (an adequate economical value) was selected to compare the results obtained for all samples. As shown in Table 2, removal Analytical Methods Chloride, chemical oxygen demand (COD), biochemical oxygen demand (BOD5), dissolved organic carbon (DOC), total phenols and ammonium nitrogen were determined according to the standard procedures (APHA et al., 1998). Polycyclic aromatic hydrocarbons (PAH) were determined by liquid chromatography TABLE 1. Sample 1 2 3 4 5 6 7 8 4 Characteristics of the Raw Industrial Wastewater Chloride (mg/L) COD (mg/L) 12,200 9,600 2,100 15,700 19,200 17,700 25,400 26,900 1,530 740 2,100 1,610 1,650 2,370 1,680 1,630 BOD5 (mg/L) A. A. Lima et al. 300 280 520 380 510 1,130 850 600 February 2006 DOC (mg/L) EC50 (%) pH 336 247 480 443 303 561 303 298 16.9 4.0 3.9 74.7 19.6 16.4 21.4 16.4 12.1 12.3 9.0 — 12.8 12.4 12.4 13.0 2500 600 500 2000 DOC (mg/L) COD (mg/L) 400 1500 1000 300 200 500 100 0 0 0 FIGURE 1. 0,5 1 1,5 0 0,5 1 Ozone Dose (g O3/L) Ozone Dose (g O3/L) (a) (b) 1,5 Typical oxidation results: a) COD variation with ozone dose b) DOC variation with ozone dose. Sample 3 ( ) and sample 6 (O).  45 40 COD removal (%) 35 30 25 20 15 10 5 0 0 5 10 15 20 Chloride concentration (g/L) 25 30 FIGURE 2. Effect of chloride concentration on COD removal – ozone dose: 0.5 g O3/L – each experimental point corresponds to a given sample. of DOC and COD ranged from 7 to 19% and 0 to 40%, respectively. As expected, COD removal decreased as chloride concentration increased, as shown in Figure 2. Even the ozonation of simple substances, like phenol, is negatively affected by chloride, as previously reported (Bessa, 2003). Chloride ions act as hydroxyl radical scavengers, which are predominant in high pH values (Lipczynska-Kochany et al., 1995). The pH did not vary significantly during the ozonation reaction. In the experiments employing higher ozone doses the maximum pH decrease was not higher than one unit. BOD5 removal was comprised between 0 and 56% (Table 2) and ozonation did not contribute to improve the biodegradability of the industrial wastewater. The ratio BOD5/COD of the ozonated samples was comprised between 0.18 and 0.43, with an average of 0.34, the same average value observed for the raw effluent samples. The results of the batch biodegradation tests, shown in Table 3, indicate that COD and DOC removals were similar for both effluents (raw and ozonated), confirming that ozonation, under the tested conditions, was not able to enhance the wastewater biodegradability. Total phenols removal was monitored in the experiments with samples 1 to 4. After ozonation (0.5 gO3/L) total phenols concentration was below 0.02 mg/L, the local discharge standard. Ammonium nitrogen was Ozonation of a Complex Industrial Effluent February 2006 5 TABLE 2. DOC, COD and BOD5 Removals Attained after Ozonation: 0.5 g O3 Absorbed Sample 1 2 3 4 5 6 7 8 DOC Removal (%) COD Removal (%) 13.1 13.0 17.7 15.3 19.1 8.2 14.2 6.7 9.2 25.0 40.6 22.8 29.3 8.9 13.7 n.d. BOD5 Removal (%) n.d.* 15.0 56.0 7.6 3.1 47.8 26.9 n.d.* * n.d. not detected. TABLE 3. COD and DOC Removals Attained in the Batch Biodegradation Tests after 24 and 48 Hours for the Raw Wastewater and the Ozonated Wastewater COD Removal (%) Sample 3 Raw Ozonated 4 Raw Ozonated 5 Raw Ozonated 6 Raw Ozonated 7 Raw Ozonated 8 Raw Ozonated DOC Removal (%) 24 h 48 h 24 h 48 h 64 50 65 60 72 50 73 63 40 70 70 70 42 59 68 72 50 38 54 44 70 62 81 83 50 55 60 64 32 39 64 68 50 62 — — 63 75 90 90 52 53 59 55 65 73 83 83 probably removed by gas stripping, which was favored by the high effluent pH. The removal of this pollutant varied from 9% to 53% (average of 27%). Toxicity was appreciably removed after ozonation. EC values of the ozonated samples are shown in Table 4. Complete removal of toxicity was observed in the ozonated effluent samples 4, 5 and 6. Toxicity was highly removed after ozonation of samples 1, 2 and 7. In general, toxicity removal increased with the ozone dose until 0.5 or 1.0 g O3/L, as illustrated in Figure 3 for sample 2. The removal of toxicity was probably the most important result of the present work. Considering the high level 6 A. A. Lima et al. TABLE 4. Sample Effluent Toxicity after Ozonation—Ozone Dose: 0.5 g O3/L 1 2 3 4 5 6 7 8 EC50 (%) 60.6 91.6 n.d.* 100 100 100 42.4 17.6 * n.d. not detected. of organic matter present in the industrial wastewater, moderate DOC removal, in the ozonation process, was expected. Furthermore, the organic matter would be removed at an appreciable extent in the biological treatment unit, since the ratio BOD5/COD was 0.34 on average. However, toxicity drastically affects the performance of the biological treatment process, impairing the quality of the biotreated effluent. Thus, it should be removed upstream of the biological treatment system. The PAH compounds found in some effluent samples were: chrysene, benzo(ghi)perylene and benz(a)anthracene. Ozonation promoted a very effective removal of polycyclic aromatic hydrocarbons (PAH) as shown in Table 5. In some samples, these compounds were removed to non-detectable levels. Stripping of PAH compounds could occur during ozonation. In the present work these compounds were not analyzed in the off-gas stream; neither was a blank test with an inert gas performed. It is probable that these compounds were not significantly stripped because they can interact with high molecular mass substances found in the wastewater (polymeric substances from the polyurethane production unit) and other organic macromolecules whose agglomeration is favored in saline environments. Evidences of oxidation of PAH compounds by ozone in aqueous solutions are found in the literature (Trapido et al., 1995; Yao et al., 1998). The stripping of naphthalene, a compound that is more volatile than the PAH substances used in our work, was considered to be insignificant in experiments of electrolytic aeration (Goel et al., 2003). These observations seem to indicate that, in the present work, oxidation was the preponderant mechanism for PAH removal. As expected, due to the complex effluent chemical composition, no relation between toxicity and PAH concentration was observed. These compounds were present in very low concentrations in the water matrix (effluent), which probably contains hundreds or thousands of other chemicals. Thus, for many industrial effluents, toxicity is a very complex response, that may be the result of countless interactions among many chemical species and compounds. CONCLUSIONS The characterization of the industrial wastewater revealed high levels of chlorides, presence of recalcitrant compounds (BDO5/COD ratio of 0.34) and high toxicity. February 2006 100 90 80 EC50 (%) 70 60 50 40 30 20 10 0 Raw 0.2 0.5 1.0 Ozone dose (g O3/L) 2.0 3.0 FIGURE 3. Variation of the effluent toxicity (EC50) with the amount of absorbed ozone (sample 2). TABLE 5. Removal of Some PAH Compounds by Ozonation— Ozone Dose: 0.5 g O3/L Sample 4 7 8 Compound Chrysene Benzo(ghi)perylene Benz(a)anthracene Chrysene Benz(a)anthracene Concentration (mg/L) Concentration Raw (mg/L) Effluent after Ozonation 0.134 0.225 0.031 0.077 0.044 n.d.* n.d. n.d. n.d. 0.012 *n.d. means non-detected – detection limits for chrysene, benzo(ghi)perylene and benz(a)anthracene are 0.025, 0.189 and 0.008 mg/L, respectively. be effective in removing the industrial wastewater toxicity, it is important to find candidate streams to be submitted to this oxidation process. Certainly, the flow rate to be processed will be significantly reduced, improving the viability of the ozonation technique for effluent toxicity removal. ACKNOWLEDGMENTS The authors express their gratitude to Bayer Industry, Belford Roxo, especially for the technical support given by Mrs. G. A. T. Fontoura and F. A. M Rodrigues and, to the financial support of the following Brazilian agencies: CNPq, CAPES, FUJB and Faperj. REFERENCES Ozonation promoted low to moderate levels of organic matter removal (COD, DOC and BOD5) even when high ozone doses were applied (1 to 3 g O3/L). For the ozone dose of 0.5 g O3/L the average removal efficiencies (8 samples) were: 13% (DOC), 19% (COD) and 20% (BOD5). Ozonation was not effective in improving effluent biodegradability, as confirmed in batch biodegradation tests. It was observed that the chloride content affected the ozonation performance. COD removal tended to decrease when chloride content increased. Effluent toxicity was effectively removed by ozonation. Complete toxicity removal was achieved for 3 effluent samples and high removals were attained for 3 other samples. Toxicity removal increased with ozone dose up to 0.5 g O3/L. Polycyclic aromatic compounds, although present in low concentrations in the raw wastewater samples, were effectively removed by ozonation, in most cases, to nondetectable levels. Identification of the specific industrial effluent streams that contribute to the overall wastewater toxicity is a necessary future investigation. Since ozonation showed to 1. APHA, AWWA and WEF , Standard Methods for the Examination of Water and Wastewater, 20th Ed., (New York, USA: 1998). 2. Baig, S., I. Coulomb, P. Courant, and P. Liechti, ‘‘Treatment of Landfill Leachates: Lapeyrouse and Satrod Case Studies’’, Ozone Sci. Engr., 21(1):1–22 (1999). 3. Beltran, J. F., J. F. Garcia-Araya, and P. M. Alvarez, ‘‘pH Sequential Ozonation of Domestic and Wine-Distillery Wastewaters’’, Water Res., 35(4):929–936 (2001). 4. Beltran-Heredia, J., J. Torregrosa, J. R. Dominguez, and J. Garcia, ‘‘Treatment of Black-Olive Wastewaters by Ozonation and Aerobic Biological Degradation’’, Water Res., 34(14):3515– 3522 (2000). 5. Bessa, E., Ozonation of Phenol in Saline Waters (in Portuguese), D.Sc. Thesis, COPPE, Universidade Federal do Rio de Janeiro (2003). 6. FEEMA (Rio de Janeiro State Environmental Agency), Me´todo de determinação de hidrocarbonetos aromáticos polinucleares (HAP) por cromatografia lı´quida de alta eficieˆncia, Normas Técnicas do Serviço de Análise Instrumental (1990). 7. Goel, R. K., J. R. V. Flora, and J. Ferry, ‘‘Mechanisms for Naphthalene Removal During Electrolytic Aeration’’, Water Res., 37(4):891–901 (2003). 8. Lin, C. K., T. Y. Tsai, J. C. Liu, and M.C. Chen, ‘‘Enhanced Biodegradation of Petrochemical Wastewater using Ozonation and BAC Advanced Treatment’’, Water Res., 35(3):699–704 (2001). Ozonation of a Complex Industrial Effluent February 2006 7 9. Lipczynska-Kochany, E., G. Sprah, and S. Harms, ‘‘Influence of Some Groundwater and Surface Constituents on the Degradation of 4-Chlorophenol by the Fenton Reaction’’, Chemosphere, 30:9–20 (1995). 10. Perkowski, J., L. Kos, and S. Ledakowicz, ‘‘Advanced Oxidation of Textile Wastewaters’’, Ozone Sci. Engr., 22(5):535–550 (2000). 11. Trapido, M., Y. Veressinina, and R. Munter, ‘‘Ozonation and Advanced Oxidation Processes of Polycyclic Aromatic-Hydro- 8 A. A. Lima et al. carbons in Aqueous Solutions — A Kinetic Study’’, Environ. Technol., 16(8):729–740 (1995). 12. Yao, J. J., Z. H. Huang, and S. J. Masten, ‘‘The Ozonation of Pyrene: Pathway and Product Identification’’, Water Res., 32(10):3001–3012 (1998). 13. Zhou, H. and D.W. Smith, ‘‘Ozonation Dynamics and its Implication for Off-Gas Ozone Control in Treating Pulp Mill Wastewaters’’, Ozone Sci. Engr., 22(1):31–51 (2000). February 2006