Continuous Anaerobic Digestion of Food Waste and Paper Waste under Mesophilic-Dry Condition

ABSTRACT With increasing concerns regarding the limited capacity of landfill, conservation of resources, and reduction of CO2 emissions, dry anaerobic digestion of organic solid waste has recently been gaining considerable attention. However, there have been few reports on continuous operation and most have involved operation under thermophilic condition. In the present study, a continuous dry anaerobic digestion system treating a mixture of food waste and paper waste was operated under mesophilic condition. For easy injection of a solid type substrate, the feed was diluted six-fold with the sludge inside the reactor, and then, fed into the reactor. During the operation, hydraulic retention time (HRT) decreased as follows: 150, 100, 60, 40, and 30 d at a fixed substrate concentration of 30% total solids (TS), corresponding to a solid loading rate (SLR) of 2.0, 3.5, 5.0, 7.5, and 10.0 kg TS/m3/d, respectively. Up to 40 d of HRT, biogas production proportionally increased as SLR increased, but at 30 d of HRT, biogas production decreased. At further operation, instead of controlling HRT, substrate concentration was increased to 40% TS, which was found to be a better option for increasing the treatability. The system could achieve a stable CH4 production yield of 0.27 m3 CH4/kg TSadded and 0.25 m3 CH4/g CODadded, and over 75% of volatile solids (VS) reduction.

Journal of Water and Environment Technology, Vol. 8, No.3, 2010 Address correspondence to Sae-Eun Oh, Department of Environmental Engineering, Hanbat National University, Email: saeun@hanbat.ac.kr Received December 9, 2009, Accepted April 6, 2010. - 167 - Continuous Anaerobic Digestion of Food Waste and Paper Waste under Mesophilic-Dry Condition Dong-Hoon KIM*, Mo-Kwon LEE**, and Sae-Eun OH** * Bioenergy Research Center, Korea Institute of Energy Research, Daejeon 305-343, Republic of Korea ** Department of Environmental Engineering, Hanbat National University, San 16-1, Duckmyoung-dong, Yuseong-gu, Daejeon 305-719, Republic of Korea ABSTRACT With increasing concerns regarding the limited capacity of landfill, conservation of resources, and reduction of CO 2 emissions, dry anaerobic digestion of organic solid waste has recently been gaining considerable attention. However, there have been few reports on continuous operation and most have involved operation under thermophilic condition. In the present study, a continuous dry anaerobic digestion system treating a mixture of food waste and paper waste was operated under mesophilic condition. For easy injection of a solid type substrate, the feed was diluted six-fold with the sludge inside the reactor, and then, fed into the reactor. During the operation, hydraulic retention time (HRT) decreased as follows: 150, 100, 60, 40, and 30 d at a fixed substrate concentration of 30% total solids (TS), corresponding to a solid loading rate (SLR) of 2.0, 3.5, 5.0, 7.5, and 10.0 kg TS/m 3 /d, respectively. Up to 40 d of HRT, biogas production proportionally increased as SLR increased, but at 30 d of HRT, biogas production decreased. At further operation, instead of controlling HRT, substrate concentration was increased to 40% TS, which was found to be a better option for increasing the treatability. The system could achieve a stable CH 4 production yield of 0.27 m 3 CH 4 /kg TS added and 0.25 m 3 CH 4 /g COD added , and over 75% of volatile solids (VS) reduction. Keywords: dry anaerobic digestion, hydraulic retention time, mesophilic, methane INTRODUCTION In order to mitigate the effects of climate changes, the Kyoto Protocol was announced in 1997, dictating that industrialized countries should reduce their total greenhouse gas (GHG) emissions by 5.2% by the end of 2012 from the level of emissions in 1990. This target can only be met with a significant transition from fossil fuels to alternative energy sources that are cheap, renewable, and not causing pollution (Saxena et al., 2009). Tidal, geothermal, hydroelectric, and wind power could be the suitable candidates in some countries; however, they are not expected to become the dominant sources in the future (Zidansek et al., 2009). Meanwhile, biomass is widespread throughout the world, and thereby is not subject to world price fluctuations or supply uncertainties, in contrast with imported fuels. In addition, it is a carbon neutral resource in its life cycle (Fortman et al., 2008). The use of landfill has been the main final disposal method of organic solid wastes to date in most nations. However, as it creates a large amount of polluted leachate, emits GHG, and requires a long time (30-100 yrs) for degradation, the need to avoid landfill is now shared by all technical communities (Gioannis et al., 2009). Considering these aspects, the choice of a biological process seems obvious. In this regard, anaerobic digestion perhaps offers the best solution in terms of energy and mass balance (Pavan et al., 2000). Through anaerobic digestion, biomass including organic solid waste can be - 168 - stabilized in a closed reactor where biodegradation is highly accelerated relative to that in a landfill. Furthermore, clean biogas can be obtained, which can be utilized for heat or electricity generation. In addition, the effluent sludge may be composted or may be used for soil conditioning depending on its characteristics (Vallini et al., 1993). However, as the conventional anaerobic digestion method proceeds under a slurry state (<5% total solids – TS), a large amount of external water is required for diluting solids. This will not only increase the energy consumption for digester heating and feed slurry pumping, but also the volume of digester effluent that should be dewatered (Radwan et al., 1993). To overcome these drawbacks, dry digestion or “high-solid digestion” can be employed, in which a solid substrate having over 20% of TS concentration is directly fed to the reactor (Bolzonella et al., 2003). During the 1990s, dry digestion prevailed over wet digestion, and several commercialized dry digestion systems e.g. DRANCO (Six and De Baere, 1992), KOMPOGAS (Willinger et al., 1993), and VALORGA (Laclos et al., 1997) were developed. Various kinds of solid wastes such as agricultural residues, sludge cake, and organic fraction of municipal solid wastes have been treated. Recently, with increasing concern over the limited capacity of landfill, conservation of resources, and reduction of CO 2 emissions, dry digestion is gaining much attention. However, most researches have been limited to batch tests or investigation of the start-up period, and all the systems have been operated under thermophilic condition, based on thermophilic operation (50- 60°C) being favorable in terms of hydrolysis and microbial kinetics (Forster-Carneiro et al., 2007; Shuguang et al., 2007; Forster-Carneiro et al., 2008a; Forster-Carneiro et al., 2008b). It is important, however, to attain the maximum treatability of the system, and it is clear that the operation at mesophilic condition (30-40°C) would be less energy consumptive, thereby enhancing economic viability. In Korea, the amount of waste dumped to landfill has reached 10.5 million tons per year, accounting for 11.2% of total waste produced in 2007. Considering the increasing trend of waste production and the current remaining landfill capacity of 185 million tons, it is expected that the nation’s landfill will be filled within 10 years. Also, Korea is a highly energy-dependent country, fulfilling 97% of its energy consumption needs by import, and will be forced to reduce CO 2 emissions in the near future under the Kyoto Protocol. Therefore, implementation of anaerobic digestion for the treatment of organic solid waste is an urgent issue. In the present work, the performance of a continuous dry anaerobic digestion process under mesophilic condition was investigated. A mixture of food waste and paper waste, the main sources of municipal solid wastes, was used as a feedstock. During the operation, hydraulic retention time (HRT) and the solid concentration of substrate were controlled. - 169 - MATERIALS AND METHODS Feedstock and seeding source Food waste collected from a school cafeteria, and paper waste comprised of toilet paper, newspaper, and copy paper were shredded by a hammer crusher (TOP-03H) and cut crusher (TOP-03-CC), respectively, to a size less than 5 mm. Both crushers are manufactured by Korean Mechanics Engineering Corp. The mixing ratio of food waste and paper waste was 7:3 on a weight basis and a certain amount of water was added to adjust the TS concentration. The TS concentration of food waste and paper waste was 20% and 99%, respectively. Volatile solids content (VS/TS), total nitrogen (TN), and chemical oxygen demand (COD) concentration of the mixed feedstock were 94.5±0.9% (VS/TS), 0.014±0.004 g/g TS, and 1.09±0.10 g COD/g TS, respectively. As a seeding source, a mixture of dewatered sludge cake and anaerobic digester sludge taken from the same local wastewater treatment plant was used. The characteristics of these two different types of sludge are presented in Table 1. Dewatered sludge cake and anaerobic digester sludge were mixed at a 4:1 ratio by volume basis, resulting in initial TS and VS concentration of 17.3% and 7.6%, respectively. Table 1 - Characteristics of seeding inoculum used in this study Item Dewatered sludge cake Anaerobic digester sludge TS concentration (%) 20.2 5.8 VS content (VS/TS, %) 40.2 95.0 Alkalinity (g CaCO 3 /L) 16.3 2.8 TN (mg N/L) 5,200 1,900 Ammonia (mg NH 4 -N/L) 2,950 980 pH 8.5 7.6 Reactor operation As shown in Fig. 1, a horizontal-type cylindrical reactor was used for dry anaerobic digestion. The total volume of the reactor was 60 L with a diameter and length of 320 mm and 750 mm, respectively. The broth was agitated by four impellers at 25 rpm. Thirty liters of seeding source was added to the reactor, and purged with N 2 for 10 min in order to provide anaerobic condition. After three days of adaptation period (no feed injection), 0.27 L of substrate (30% TS), corresponding to 150 d HRT, was fed daily. There was no sludge injection until the inside sludge volume reached an effective volume of 40 L. At further operation, HRT was decreased to 100, 60, 40, and 30 d at a fixed substrate concentration of 30% TS. The substrate was added once daily until 60 d HRT, but the injection time was increased to twice daily at further HRT decrease. As a decline in system performance was observed at 30 d HRT, the HRT was again increased to 40 d for performance recovery. Substrate concentration was subsequently increased to 40% TS at 40 d HRT. In this study, for easy injection of a solid-type substrate, the feeding pump was turned on after the solid content of the substrate was reduced by dilution with some sludge inside the reactor. The same screw-type pumps were used for feeding the substrate and recycling the sludge inside the reactor. Further details are provided in Table 2. In order to optimize this process, 0.3 L (= 1Q) of substrate was mixed with 2-6 times recycled - 170 - sludge (2Q-6Q), and the pump was then turned on. This test was conducted during the 100 th -120 th day of continuous operation. All the systems were installed at a temperature-controlled (35±1°C) room. Fig. 1 - Schematic diagram of anaerobic dry digestion system Table 2 - Details of screw-type pump for feed injection and sludge recycling Model Output (W) Voltage (V) Frequency (Hz) Current (A) Starting torque (N·m) Rated torque (N·m) Max. speed (rpm) K9IP200FH 200 220 50 1.3 3.0 1.45 1,350 Analysis Measured biogas production was adjusted at standard temperature and pressure (STP), 0°C and 760 mmHg. The contents of CH 4 , N 2 , and CO 2 were determined by gas chromatography (GC; Gow Mac series 580) using a thermal conductivity detector and a 1.8 m  3.2 mm stainless-steel column packed with porapak Q (80/100 mesh) with helium as a carrier gas. The temperatures of injector, detector, and column were kept at 80, 90, and 50C, respectively. Volatile fatty acids (VFAs, C2-C6) and lactate were analyzed by a high-performance liquid chromatograph (HPLC; Finnigan Spectra SYSTEM LC, Thermo Electron Co.) with an ultraviolet (210 nm) detector (UV1000, Thermo Electron) and a 100 mm  7.8 mm fast acid analysis column (Bio-Rad Lab.) using 0.005 M H 2 SO 4 as a mobile phase. The liquid samples were pretreated with a 0.45 m membrane filter before injection to the HPLC. Alkalinity, pH and concentrations of TS, VS, COD, TN and ammonia were measured according to Standard Methods (APHA, at el. 1998). RESULTS AND DISCUSSION Continuous operation performance In order to treat solids with high concentration, highly concentrated biomass should also be prepared. Generally, however, wet digester sludge having less than 5% VS - 171 - concentration has often been used as a seeding source, requiring a long period to build up a highly concentrated microbial consortium (Forster-Carneiro, 2007; Fernandez et al., 2008; Montero et al., 2009). Instead, dewatered sludge cake having 8% VS concentration was used as the main seeding source in this study. It appears that this strategy was successful; CH 4 production was observed from the first day, and biogas production was stabilized within 30 d (Fig. 2). Time (day) 0 100 200 300 400 Biogas production (L/d) 0 50 100 150 200 250 Solid loading rate (kg TS/m 3 /d) 0 2 4 6 8 10 12 Biogas SLR HRT = 150d HRT = 100d HRT = 60d HRT = 40d HRT = 30d HRT = 40d HRT = 40d Substrate conc. = 30% TS Substrate conc. = 40% TS Fig. 2 - Daily biogas production of mesophilic-dry anaerobic digestion of organic solid wastes at various operating conditions However, in employing this strategy, special care should be taken with regard to ammonia inhibition, because dewatered sludge cake is rich in nitrogen, being degraded into ammonia during the digestion, and there does not exist any nitrogen removal mechanism in anaerobic digestion. In this study, fortunately, ammonia concentration during the start-up period did not exceed 3,500 mg NH 4 -N/L, and as the substrate was continuously supplied, its concentration gradually decreased to a range of 1,500-2,500 mg NH 4 -N/L. If the dewatered sludge cake contains a high concentration of ammonia, its removal by stripping or applying other tools prior to seeding is strongly recommended. Until the 336 th day of operation, HRT was decreased to 150, 100, 60, 40, and 30 d at a fixed solid concentration of 30% TS, corresponding to SLR of 2.0, 3.5, 5.0, 7.5, and 10.0 kg TS/m 3 /d, respectively. At the 171 st -175 th d period, immediately after the HRT transition from 60 d to 40 d, a sudden decrease of biogas production was observed. It - 172 - was suspected that this failure was related to the number of feeding time. Feeding of the substrate corresponding to 7.5 kg TS/m 3 /d at one time fell outside the treatable range. In order to recover the performance, the substrate injection time was increased to twice a day, which worked efficiently. Biogas production increased and stabilized with an average value of 143 L/d. When the HRT was decreased from 40 d to 30 d, there was an increase of biogas production, but it was not sustained. From the 250 th day, drastic biogas production drop was observed along with a decrease of CH 4 content in the produced biogas (Fig. 3). CH 4 content was in a range of 50-55% until 40 d HRT decrease, but it dropped below 50% at 30 d HRT. Also, the VS concentration in the reactor clearly showed an increasing trend, suggesting an overloading condition for solid hydrolysis. In addition, as shown in Fig. 4, both pH and alkalinity concentrations, which were maintained over 7.5 and 9,000 mg CaCO 3 /L, respectively, dropped significantly. Total organic acid concentration was lower than 200 mg COD/L until 40 d HRT, but it increased to 2,500 mg COD/L in which most of the acids consisted of propionate. This indicates that the balance was broken between the production of acids and their consumption by methanogenesis at 30 d HRT. Accumulation of acids decreased the pH, resulting in the failure of the whole system. The growth rate of the acidogenic bacteria is much higher than that of methanogenic archea, and the bacteria can be active at a weak acidic condition whereas the archea cannot. Therefore, the balance between the generation of acids and their conversion to CH 4 is important in a single-stage anaerobic digestion process (Mata- Alvarez et al., 2000; Ward et al., 2008). Time (day) 0 100 200 300 400 CH 4 content (%) 30 35 40 45 50 55 60 VS concentration (%) 6.0 6.5 7.0 7.5 8.0 8.5 9.0 CH 4 (%) VS (%) HRT = 150d HRT = 100d HRT = 60d HRT = 40d HRT = 30d HRT = 40d HRT = 40d Substrate conc. = 30% TS Substrate conc. = 40% TS Fig. 3 - Change of CH 4 content and VS concentration of the system - 173 - Time (day) 0 100 200 300 400 pH 6.5 7.0 7.5 8.0 8.5 Alkalinity (mg CaCO 3 /L) 4000 6000 8000 10000 12000 14000 pH Alkalinity HRT = 150d HRT = 100d HRT = 60d HRT = 40d HRT = 30d HRT = 40d HRT = 40d Substrate conc. = 30% TS Substrate conc. = 40% TS Fig. 4 - Change of pH and alkalinity of the system At further operation, HRT was increased again to 40 d in order to see the performance recovery. The productivity of biogas recovered, and all important indicators including CH 4 content, VS concentration, pH, and alkalinity showed a recovering trend. From the 337 th day, instead of controlling HRT, substrate concentration was increased to 40% TS in order to increase SLR. At this time, stable biogas production was observed for 100 days. The CH 4 content, pH, and alkalinity concentration did not show a decreasing trend. At first glance, VS concentration in the reactor seemed increasing, but this did not indicate a decrease of VS reduction efficiency as the substrate concentration increased. The VS reduction efficiency achieved in both feeding 30% TS and 40% TS of substrate was over 75%. The CH 4 production yield (MPY) is one of the important parameters determining the success of an anaerobic digestion system. The average MPY achieved at 10 kg TS/m 3 /d of SLR (40% TS fed at HRT 40 d) was 0.27 m 3 CH 4 /kg TS added . Based on the input COD, MPY was 0.25 m 3 CH 4 /kg COD added , indicating that 71% of the energy content in the organic solid waste was converted to clean bioenergy, CH 4 . The MPY value achieved in this study was comparable with the ones achieved in conventional wet digestion processes (Mata-Alvarez et al., 2000), and dry digestion process operating under thermophilic condition. Montero et al. (2008) continuously operated thermophilic dry digestion, and achieved 80% of VS removal efficiency and 0.24 m 3 CH 4 /kg COD added of MPY at 7.5 kg TS/m 3 /d of SLR. Solid-type substrate injection Recirculated sludge volume is an important parameter affecting the economic aspect of this system. When the recirculated sludge volume was 2Q, injection of feed was - 174 - impossible, as it clogged the pump line. When the recirculated sludge volume was increased from 3Q to 6Q, the required time to pump Q was reduced from 30.5 to 8.9 s. However, the total required time for feed injection was minimal when 5Q was recirculated, as shown in Fig. 5. In this study, the same pumps were used for recirculating and feeding, meaning that the consumption of electricity only depends on the total required time in running the pump. Therefore, 5Q recirculation was found to be the most economically feasible condition. The proper volume of recirculated sludge can vary according to the type of reactor, characteristics of the substrate and pumps used, and system performance, but it is obvious that this kind of test is very informative for scale-up, and to our knowledge, this is the first ever report of such an attempt. Recirculation Volume 3Q 4Q 5Q 6Q Time (sec) 0 20 40 60 80 100 120 140 160 Substrate injection Recirculation Total Fig. 5 - Required time for feed injection at various recirculation volumes CONCLUSIONS Continuous mesophilic-dry anaerobic digestion of food waste and paper waste was successfully conducted for 420 days. For easy injection of a solid type substrate, the feeding pump was turned on after the sludge inside the reactor was recirculated and mixed with the substrate. When the HRT was decreased to 30 d, corresponding to 10 kg TS/m 3 /d of SLR, system failure was observed. The change of CH 4 content, pH, VS and alkalinity concentration all reflected poor anaerobic digestion performance. The performance was recovered when HRT was increased to 40 d again. At further operation, instead of controlling HRT, substrate concentration was increased to 40% TS at a fixed HRT of 40 d. At this time, stable performance was achieved with a high MPY of 0.27 m 3 CH 4 /kg TS added and 0.25 m 3 CH 4 /g COD added , and over 75% of volatile solids (VS) reduction. The MPY of 0.25 m 3 CH 4 /g COD added indicates that 71% of the energy content in the organic solid waste was converted to clean bioenergy. In optimizing solid- - 175 - type substrate injection, the least time was required when the volume of recirculated sludge was five times that of the substrate. ACKNOWLEDGEMENT This study was supported by Korean Ministry of Environment as “The Eco-technopia 21 project”. REFERENCES APHA, AWWA and WEF. (1998). “Standard methods for the examination of water and wastewater”, 20th ed. Baltimore. American Public Health Association, 2, 57-59. Bolzonella D., Innocenti L., Pavan P., Traverso P. and Cecchi F. (2003). Semi-dry thermophilic anaerobic digestion of the organic fraction of municipal solid waste: focusing on the start-up phase, Bioresource Technol., 86, 123-129. Fernandez J., Perez M. and Romero L. I. (2008). Effect of substrate concentration on dry mesophilic anaerobic digestion of organic fraction of municipal solid waste (OFMSW), Bioresource Technol., 99, 6075-6080. Fortman J. L., Chhabra S., Mukhopadhyay A., Chou H., Lee T. S., Steen E. and Keasling J. D. (2008). Biofuel alternatives to ethanol: pumping the microbial wall, Trends in Biotechnol., 26(7), 375-381. Forster-Carneiro T., Perez M., Romero L. I. and Sales D. (2007). Dry-thermophilic anaerobic digestion of organic fraction of the municipal solid waste, Bioresource Technol., 98, 3195-3203. Forster-Carneiro T., Perez M. and Romero L. I. (2008a). Anaerobic digestion of municipal solid wastes: Dry thermophilic performance, Bioresource Technol., 99, 8180-8184. Forster-Carneiro T., Perez M. and Romero L. I. (2008b). Thermophilic anaerobic digestion of source-sorted organic fraction of municipal solid waste. Bioresource Technol., 99, 6763-6770. Gioannis G. D., Muntoni A., Cappai G. and Milia, S. (2009). Landfill gas generation after mechanical biological treatment of municipal solid wastes, Estimation of gas generation rate constants, Waste Manage., 29(3), 1026-1034. Laclos H. F., Desbois S. and Saint-Joly C. (1997). Anaerobic digestion of municipal solid organic waste: VALORGA Full-scale plant in Tilburg, The Netherlands, Wat. Sci. Tech., 36(6-7), 457-462. Mata-Alvarez J., Mace S. and Llabres P. (2000). Anaerobic digestion of organic solid wastes. An overview of research achievements and perspectives, Bioresource Technol., 74, 3-16. Montero B., Garcia-Morales J. L., Sales D. and Solera R. (2009). Analysis of methanogenic activity in a thermophilic dry anaerobic reactor: Use of fluorescent in situ hybridization, Waste Manage., 29, 1144-1151. Pavan P., Battistoni P., Mata-Alvarez J. and Cecchi F. (2000). Performance of thermophilic semi-dry anaerobic digestion process changing the feed biodegradability, Wat. Sci. Tech., 41(3), 75-81. Radwan A. M., Sebak H. A., Mitry N. R., El-Zanati E. A. and Hamad M. A. (1993). Dry anaerobic fermentation of agricultural residues, Biomass and Bioenergy, 5(6), 495- 499. - 176 - Saxena R. C., Adhikari D. K. and Goyal H. B. (2009). Biomass-based energy fuel through biochemical routes, Renew. Sust. Energ. Rev., 13(1), 167-178. Shuguang L., Tsuyoshi I., Masao U. and Masahiko S. (2007). Start-up performances of anaerobic mesophilic and thermophilic digestions of organic solid wastes, J. Environ. Sci., 19, 416-420. Six W. and De Baere L. (1992). Dry anaerobic conversion of municipal solid waste by means of the DRANCO process, Wat. Sci. Tech., 25. 295-300. Vallini G., Cecchi F., Pavan P., Pera A., Mata-Alvarez J. and Bassetti A. (1993). Recovery and disposal of theorganic fraction of MSW by means of combined anaerobic and aerobic biotreatments, Wat. Sci. Tech., 27(2), 121–132. Ward A. J., Hobbs P. J., Holliman P. J. and Jones D. L. (2008). Optimisation of the anaerobic digestion of agricultural resources, Bioresource Technol., 99(17), 7928- 7940. Willinger A., Wyder K. and Metzler A. C. (1993). KOMPOGAS-a new system for the anaerobic treatment of source separated waste, Wat. Sci. Tech 27(2). 153-158. Zidansek A., Blinc R., Jeglic A., Kabashi S., Bekteshi S. and Slaus, I. (2009). Climate changes, biofuels and the sustainable future, Int. J. Hydrogen Energy, 34, 6980- 6983. . 2010. - 167 - Continuous Anaerobic Digestion of Food Waste and Paper Waste under Mesophilic-Dry Condition Dong-Hoon KIM*, Mo-Kwon LEE**, and Sae-Eun OH**. anaerobic digestion process under mesophilic condition was investigated. A mixture of food waste and paper waste, the main sources of municipal solid wastes,