United States Environmental Protection Agency Risk Reduction Engineering Laboratory Cincinnati, OH 45268 Research and Development EPA/600/S2-89/051 Jan. 1990 Project Summary Aerobic and Anaerobic Treatment of C. I. Disperse Blue 79 David A. Gardner, Thomas J. Holds worth, Glenn M. Shaul , Kenneth A. Dostal, and L. Don Betowski This study summarized here was conducted to determine the fate of C.I. Disperse Blue 79, one of the largest production-volume dyes, and select blodegradatlon products in a conventionally operated activated sludge process and in an anaerobic sludge digestion system. To achieve this objective, a pilot study was conducted from "November 1987 to February 1989. Two continuous-feed pilot-scale waaftewater treatment systems, one control and one experimental, were operated during the pilot study. Yfee experimental treatment system was fed screened, raw municipal wastewater dosed with a target concentration of 5 mg/L of active Ingredient in the commercial formulation of C. I. Disperse Blue 79. The control system was fed only the screened, raw municipal wastewater. After acclimation and after steady state conditions were reached, samples from each system were analyzed for the dye and related compounds. A bench-scale activated sludge system was also operated to assess the fate of dye degradation products from the anaerobic digester In an aerobic treatment system. This system was operated to simulate the recycle of digester supernatant to the head-end of a typical wastewater treatment system. The results of this extensive research project are presented In Volume I of the full report. Findings are presented regarding: (1) the development of an analytical procedure to determine C.I. Disperse Blue 79 in various sample matrices; (2) the effect of C.I. Disperse Blue 79 on the operation of an activated sludge system and an anaerobic digester; (3) the fate of the dye in the treatment systems; and (4) the detection of any degradation products in the systems. Laboratory and operating data collected during the study are presented In Volume II. This Project Summary was developed by EPA's Risk Reduction Engineering Laboratory, Cincinnati, OH, to announce key findings of the research project that Is fully documented In a separate report of the same title (see Project Report ordering Information at back). Introduction The fate of specific azo dyes in wastewater treatment facilities is unknown since few detailed studies have been reported in the literature. A better understanding of this large class of organic compounds is necessary to assess their effect on the environment and on human health. Several azo dyes and possible biodegradation products, such as aromatic amines, have been shown to be, or are suspected to be, carcinogenic. C.I. Disperse Blue 79, one of the largest production-volume dyes, is a water insoluble bromodinitroaniline- derived compound. The empirical formula for the bromomethoxy form of C.I. Disperse Blue 79 used in this study is C23H25BrN6010, the molecular weight is 625.4, and the structural formula is shown in Figure 1. The purpose of this study was to determine the fate of C.I. Disperse Blue ------- OCH3 NH-CO-CH3 Figure 1. C. I. Disperse Blue 79 79 and select biodegradation products in a conventionally operated activated sludge process and in an anaerobic sludge digestion system. Before testing, an analytical procedure for measuring C.I. Disperse Blue 79 concentrations was developed. Two continuous-feed, pilot- scale wastewater treatment systems (one control [Unit 1] and one experimental [Unit 2]) were operated at the Milwaukee Metropolitan Sewerage District (MMSD) South Shore Wastewater Treatment Plant. In addition to these pilot-scale systems, a bench-scale activated sludge system (Unit 3) was operated to assess the fate of dye degradation products from a digester in an aerobic treatment system. This system was operated to simulate the recycle of digester supernatant to the head-end of a typical wastewater treatment system. Experimental Procedures The two pilot-scale treatment systems were operated for the entire study, from November 1987 to February 1989. The bench-scale system was operated from November 1988 to February 1989. C.I. Disperse Blue 79 Extraction and Analysis Because a reliable method for dye analysis was needed to determine the fate of C.I. Disperse Blue 79 in the treatment system, an analytical procedure was developed. Various extraction methods and solvents were investigated to develop a suitable extraction procedure to prepare samples for high-performance liquid chromatog- raphy (HPLC) analysis. Pilot-Scale Treatment Systems Both pilot-scale activated sludge systems included a contact tank, a conical-shaped primary clarifier, an aeration basin, and a conical-shaped secondary clarifier. The contact tanks were installed to ensure the dye was mixed with the feed and to obtain a 30- min contact time between the raw wastewater and the dye. The primary and secondary clarifiers were approximately 49L, and the aeration tanks were approximately 185 L. The activated sludge basins were separated into three cells to operate as a plug-flow system. Peristaltic pumps supplied the screened, raw wastewater to the contact tanks. Gravity moved the wastewater from the contact tanks to the primary clarifiers, then to the aeration basins, and on to the secondary clarifiers. Activated sludge was wasted from the aeration basins via peristaltic pumps. Primary sludge was wasted manually once each day. The target hydraulic retention time (HRT) was 5.5 hr and the solids retention time (SRT), 7 days. The anaerobic digesters were cylindrical-shaped vessels constructed of clear PVC. Each digester had a total volume of 70 L with an operating volume of 39 L. The digesters were completely mixed and heated to maintain an operating temperature of 35°C. Gas production from the digesters was monitored with gas meters. Waste activated sludge and primary sludge from each activated sludge unit were mixed, thickened, and fed to the respective anaerobic digesters. The target SRT of the anaerobic digesters was 15 days and the target loading was 1.2 kg total volatile solids (TVS)/m3»day. The experimental treatment received screened, raw wastewater dosed with a target concentration of 5 mg/L of the active ingredient in C.I. Disperse Blue 79. The control system received only the screened, raw wastewater. After acclimation and steady state conditions were reached, the following samples from each system were analyzed for the dye and related compounds: influent, primary effluent, activated sludge effluent, primary sludge, waste activated sludge, digester feed, digester supernatant, and digester effluent. Bench-Scale Treatment System The bench-scale system was an activated sludge unit operated on a feed mixture prepared from the experimental system. The activated sludge unit consisted of a 6-L conical reactor, which served as the aeration basin; a 2-L inner cone for solids recycle; and a 125-ml clarifier tube for effluent clarification. Peristaltic pumps were used to deliver the feed and remove waste activated sludge from the unit. The feed mixture included primary effluent from the experimental system, supernatant from digester I preparation (primary and waste acti\ sludge thickening), and centrate centrifuging digested sludge from anaerobic digester. The mixture prepared to simulate the recycl digester supernatant and primary thickened waste activated sli supernatant to the head-end i treatment plant. Results and Discussion C.I. Disperse Blue 79 Extract/ and Analysis Initial experiments with two lie phase extractions did not yield acceptable procedure for C.I. Disp Blue 79; therefore, the approach changed to dispersing the aqu< sample in acetonitrile. A spectropt meter monitored the extrac procedure. C.I. Disperse Blue 79 cor trations in the extracts were anal' with the use of HPLC. Analytical spike and analyl duplicate analyses (part of the Qu Assurance/Quality Control (QA/ program) monitored the accuracy precision of the extraction procedure HPLC analyses. For the 57 analy duplicate analyses performed spectrophotometry, the average rel difference was 4.3% with a stan deviation of 6.7%. For the 29 anal} spike analyses performed spectrophotometry, the average s recovery was 105% with a stam deviation of 13.8%. For the se analytical duplicate measurem performed on the HPLC, the ave relative difference was 6.8% wil standard deviation of 9.1%. Tl replicate analyses performed on same sample had an average rel, difference of 4.8% and standard devi. of 3.1%. Field spike samples prepared 1 field duplicate samples ensured proper sample collection proced were used. The average rela difference for HPLC analyses was 1C with a standard deviation of 15.3%. Pilot-Scale Systems Operat/oi Operating data were collet throughout the entire period the syst were operated. Because of analy capacity limitations, however, the f scale activated sludge systems, anaerobic digesters, and the bench-s activated sludge system "Were' sampled over the entire period for analyses. Thus, all discussion: ------- }uipment performance include only data jring the time dye analyses were >nducted. Operating and analytical data for the lot-scale activated sludge units are mmarized in Table 1. The data for both lits 1 and 2 were similar. The average X)D value for Unit 2 was 73.5 mg/L and t for Unit 1 was 59.2 mg/L. Although 3 slightly higher effluent TCOD value • Unit 2 may have been caused by ding dye to the unit, the data indicate \ the overall performance of the perimental activated sludge system is not affected by this addition. tile 1. Summary of the Pilot-Scale Activated Sludge Systems' Operational and Analytical Data for the Dye Testing Period Parameter Unit 1 Unit 2 leddata rSS (mg/L TCOD (mg/L) TBOD (mg/L) NH3-N (mg/L) perational data HRT (hr) SRT (days) 238 364 182 22.5 5.28 5.94 211 375 177 20.7 5.28 5.87 ixed liquor data Temperature (°C) 20.0 20.0 pH (range) 6.8-8.0 6.8-7.6 DO, Cell 1 (mg/L) 2.4 2.8 DO, Cell 2 (mg/L) 3.6 3.5 DO, Cell 3 (mg/L) 3.6 3.9 TSS (mg/L) 3,030 3,060 02 Uptake rate 6.8-8.0 6.8-7.6 (mg/Lhr) 73.1 58.0 SSVI (ml/g) rimary effluent data rSS (mg/L) 134 139 NH3-N(mg/L) 22.7 21.7 nal effluent data TCOD (mg/L) TBOD (mg/L) TSS (mg/L) NH3-N (mg/L) 59.2 16 27 0.26 73.5 21 31 0.18 The anaerobic digester's operating id analytical data are summarized in ible 2. The feed, effluent, and )erational data indicate no significant ference between the two units. No Jverse affect was detected on the Deration of the experimental digester as result of adding dye. ye and Related Compounds esults The influent and effluent streams (feed ream, primary clarifier effluent, primary Table 2. Summary of the Anaerobic Digesters' Operational and Analytical Data for the Dye Testing Period Parameter Unitl Unit 2 Feed data 7SS (mg/L) TS (%) TVS (%) 21,300 2.42 1.81 21,400 2.42 1.82 Effluent data pH (range) 6.6-7.0 6.6-7.0 Temperature (°C) 35.0 35.0 TSS (mg/L) 12,700 12,200 TS (%) 1.46 1.48 TVS (%) 0.94 0.97 Operational data Alkalinity (mg/L) 2,930 2,820 Volatile acids < 51 < 50 (mg/L) Loading 1.21 1.22 (kg TVS/m3 day) TVS reduction (%) 47.8 46.4 Gas production 0.76 0.87 (m3/kg TVS destroyed) Percent CH4 in gas 58.9 57.9 sludge, waste activated sludge, and final clarifier effluent) from the experimental activated sludge systems were sampled and analyzed for C.I. Disperse Blue 79 and any related compounds to determine the fate of the dye in the treatment system. The average dye and TSS concentrations from the Unit 2 samples are summarized in Table 3. Influent and waste mixed liquor samples were analyzed from Unit 1. The dye was not detected in any of the control unit samples analyzed (i.e., no background concentration of dye was present in the raw municipal wastewater feed). The average dye concentration in the Unit 2 feed to the primary clarifier was 4.40 mg/L and the average final effluent concentration was < 0.93 mg/L, so that the average dye removal was greater than 79%. Although 5 of 19 analyzed effluent samples were below the 0.25 mg/L detection limit, the effluent dye concentration varied from < 0.25 mg/L to 3.70 mg/L. The variation in effluent dye concentration may have been caused by the variation in effluent TSS concen- tration. The correlation coefficient between TSS and dye concentrations in the Unit 2 effluent was determined to be 0.78. In addition, calculations performed on Table 3 data show that each gram of suspended solids in the waste activated sludge contained 30 mg of dye whereas each gram of suspended solids in the final effluent contained 33 mg of dye. These data indicate that the dye has a high affinity for the activated sludge solids. Approximately 21 % of the dye fed to the unit was in the final effluent; however, most of the dye was probably in the suspended solids in the effluent. The average final effluent TSS concentration was 28 mg/L. Lowering this TSS concen- tration by improving solids removal in the final clarifier could result in a lower dye concentration in the final effluent. Table 3. Average C. I. Disperse Blue 79 and TSS Concentration in the Unit 2 Experimental Activated Sludge Unit Samples C.I. Disperse Sample Location Feed Primary effluent Primary sludge Waste activated sludge Blue 79 (mg/L) 4.40 4.71 31.8 93.5 TSS (mg/L) 212 133 14,500 3,060 Final effluent < 0.93 28 Mass balance calculations were performed with the use of the measured dye concentrations and measured flowrates for each process stream. Mass balance calculations across the entire activated sludge system showed that an average of 86.5% of the dye contained in the feed stream was accounted for in the effluent streams. The primary sludge contained an average of 3.6% of the dye fed to the system; waste activated sludge, 62.3%; and final effluents, 20.4% (the percentages of the three streams do not equal 86.5% because of rounding off the individual values). Since most of the dye fed to the system was recovered and no other related compounds were detected, it can be concluded that no significant biodegradation of C.I. Disperse Blue 79 occurred in the activated sludge system. Feed sludge and effluent (digested sludge) samples from both the control and experimental digesters were analyzed for dye content. Detectable concentrations of dye were identified by HPLC-UV in 5 of 10 control-unit feed samples and in 4 of 10 effluent samples. The average concentrations were low, however, at < 1.45 mg/L for the feed samples and < 1.22 mg/L for the effluent ------- sample. The low level of dye in these control unit samples is negligible when compared with the much higher concentrations of dye in the experimental unit samples. The average experimental unit feed dye concentration was 443 mg/L, and the average effluent concentration was 7.86 mg/L. On the average, 98.2% of the dye contained in the feed sludge was degraded in the anaerobic digester. Thermospray ionization mass spectrometry was used to identify degradation products of C.I. Disperse Blue 79 in the anaerobic digester effluent. With this ionization technique, the parent dye was observable, but because of the electronegativity of many of the functional groups on the molecule (e.g., N02, Br), the sensitivity of the technique for this compound was poor. Four major degradation compounds were tentatively identified and found in significant amounts in the digester effluent. Their exact identity and amounts have not been verified because appropriate analytical standards were not available. The degradation products showed a better response than the parent dye by the ionization technique; the increased sensitivity probably resulted from either removal or reduction of the electro- negativity that was suppressing the sensitivity of C.I. Disperse Blue 79. The relative amounts of these compounds (A, B, C, and D having molecular weights of 283, 358, 400, and 478 daltons, respectively) measured in one set of samples are summarized in Table 4. The relative amounts of these compounds were greater in the digester effluent than in the digester feed. (Information about the structure of these compounds is discussed in detail in Volume I of the full report). Some of the potential degradation pathways of C.I. Disperse Blue 79 could liberate bromide from the dye compound. Anaerobic digester feeds and effluents were analyzed to determine if bromide was liberated in the digester. In one set of data from the control unit, sample analyses did not show any significant difference in bromide concentration between the feed and effluent (Table 5). In both sets of data from the experimental unit, however, the bromide concentration was much lower in the feed than in the effluent. The two feed samples averaged 2.50 mg/L, and the two effluents averaged 39.0 mg/L. For the one set of samples for which dye analyses are also available, the dye concentration was reduced from 250 mg/L to 9.46 mg/L. These data indicate that bromide was being liberated during dye degradation in the anaerobic digester. Bench-Scale Activated Sludge System During normal operation of a wastewater treatment system, the super- natant from sludge lagoons or other digester sludge thickening operations is returned to the head-end of the plant for treatment. The bench-scale activated sludge system (Unit 3) was operated to study the fate of dye degradation products from the anaerobic digester in an activated sludge system. The supernatant from the sludge thickening operation used to prepare the digester feed was mixed with centrate from centrifuging digester effluent and primary effluent to prepare feed for Unit 3. The supernatant was added to simulate the effluents produced from thickening waste activated sludge in a typical plant. The operating and analytical data from Unit 3 are presented in Table 6. The average HRT was 6.04 days, which was slightly higher than the Unit 2 value of 5.28 days; the average SRT for Unit 3 was 4.83 days, which was lower than the Unit 2 average of 5.87 days. The Unit 3 average SRT was lower than the target value of 7 days because of a relatively high average effluent TSS value of 40 mg/L. The bench-scale unit settling performance was more variable than that in the pilot units. The average effluent TCOD and TBOD values were also higher than the pilot unit values. The higher effluent values probably resulted from the higher TSS levels in the final effluent. The performance of Unit 3 with respect to TSS, TBOD, and TCOD removal was not as good as that of the pilot units but was typical of a bench-scale unit. Table 7 summarizes the dye data from the bench-scale unit feed, waste activated sludge, and final effluent sample analyses. The average feed dye concentration was 3.43 mg/L and the average effluent concentration was 1.32 mg/L, for a removal efficiency of 62%. The effluent concentration was probably high because of the relatively high TSS concentration in the final effluent. Mass balance calculations of the dye across Unit 3 showed that an average of 75.3% of the dye fed to the unit was accounted for in the effluents from the unit. The mass balance for Unit 2 showed 86.5% of the dye was recovered. Although the recovery from Unit 3 was slightly lower, it does not appear that significant degradation of the d occurred in the bench-scale activa sludge system. Degradation products of C.I. Dispe Blue 79 were also monitored in influent, effluent, and waste sludge fr Unit 3. Because no positive identifical was made of the by-products, quan cation was not possible. Some gen< observations can, however, be rm concerning the degradation produ based on relative amounts. The obser trend indicated that the concentration these compounds decreased across I 3. The final effluent samples alw; contained the lowest concentrations the degradation products, but because limited data, further conclusions can be drawn. Further evaluation of degradation products and their fate biological treatment systems may subject for further project work. Conclusions 1. The addition of C.I. Disperse Blue did not adversely affect the opera' of the pilot activated sludge unil that of the anaerobic digester. E the control and experimer activated sludge units produi effluents typical of munici wastewater treatment systems. anaerobic digesters achieved vol< solids reductions within the nor operating range for munici digesters. 2. No evidence of C.I. Disperse Blue degradation in the activated slu systems was found. Mass bala calculations showed that, on avers 86.5% of the dye contained in feed to the system was presen the effluent streams. 3. The majority of the C.I. Disp€ Blue 79 fed to the activated slu system was removed in the wi activated sludge. The average mass balance obtained around system was 86.5%; the dye partitioned in the effluent stream; follows: 3.6% in the primary slu< 62.3% in the waste activated slue and 20.4% in the final effluent. 4. The C.I. Disperse Blue 79 degraded in the anaerobic digei The dye concentration was redi from an average feed value of mg/L to an average effluent valu 15.0 mg/L, or a 97.4% reduction. 5. Possible degradation products ol dye were detected in the dige effluent. Although some prelimi measurements were made to ide the structure of these compounds ------- Table 4. Mass Spectrometric Analysis of Anaerobic Digester Samples Relative Sample Amounts (Arbitrary Units) Sample Location Control digester feed Experimental digester feed Experimental digester effluent A(MW 283) 26,000 24,000 568,000 B(MW 358) 7,300 6,400 264,000 C(MW 400) 7,400 6,300 365,000 D(MW 478) 1,700 43,000 225,000 Table 5. Bromide Analyses of Anaerobic Digester Samples Sample Location Experimental unit feed Experimental unit effluent Control unit feed Control unit effluent Experimental unit feed Experimental unit effluent Sample Date H5/89 1/5/89 217189 2/7/89 2/7/89 2/7/89 Bromide Concentration (mg/L) 1.75 40.8 0.85 0.66 3.24 37.1 C.I. Disperse Blue 79 Concentration (mglL) A * * *t 250 9.46 "Analysis not performed. Table 6. Summary of Activated Sludge Unit 3's Operational and Analytical Data Parameter Average Value Feed rss (mg/L) TCOD (mglL) TBOD (mg/L) Operation data HRT (hr) SRT (days) Mixed liquor data Temperature (°C) pH (range) DO (mg/L) TSS (mg/L) Final effluent data TSS (mg/L) TCOD (mg/L) TBOD (mg/L) 130 336 158 6.04 4.83 21.5 6.8-8.1 5.6 T,650 40 116 31 Table 7. Bench-Scale Activated Sludge System's C. I. Disperse Blue 79 Analytical Results C. I. Disperse Blue 79 TSS Sample Location (mg/L) (mglL) Feed Waste activated sludge Final effluent 3.43 37.6 1.32 14.5 2,150 53 ------- positive identification or quantification products from the anaerobic digester The full report was submitted i of the compounds was made. were destroyed when treated in a fulfillment of Contract No. 68-03-3371 I 6. Based on limited semi-quantitative bench-scale activated sludge Radian Corporation under th results, some of the dye degradation system. sponsorship of the U.S. Environment Protection Agency. ------- |