Public Disclosure Authorized ACS13221 v2 Public Disclosure Authorized ANNEXES WA S T E WAT E R Public Disclosure Authorized TO E N E R G Y A Technical Note for Utility Managers and Decision Makers on Urban Sanitation in East Asian Countries Public Disclosure Authorized January 2015 Contents TABLE OF CONTENTS ANNEX 1 ENERGY EFFICIENCY EFFORTS IN EUROPE AND USA 1 ANNEX 2 SLUDGE DIGESTERS & RENEWABLE ELECTRICITY POTENTIAL FROM BIOGAS IN USA AND EUROPE 9 ANNEX 3 CASE STUDIES 13 1. CASE STUDY 1: CAS + SLUDGE DIGESTION 14 1.1. BACKGROUND, PROCESS DESCRIPTION 15 1.1.1. DATA SOURCES 15 1.1.2. WASTEWATER MANAGEMENT 16 1.1.3. SLUDGE MANAGEMENT 16 1.1.4. ENERGY MANAGEMENT 17 1.2. ANALYSIS 17 1.2.1. WASTEWATER INFLUENT AND EFFLUENT 18 1.2.2. BIOGAS PRODUCTION & POTENTIAL FOR ENERGY GENERATION 20 1.2.3. OPERATION CAPACITY NEEDS, BIOGAS SAFETY 28 1.2.4. INSTITUTIONAL ASPECTS, ENERGY COSTS 27 1.2.5. GHG REDUCTION AND CDM CO-FINANCING 29 1.2.6. CAPEX STRUCTURE 32 1.2.7. OPEX STRUCTURE 35 1.2.8. VIABILITY OF INVESTMENT IN BIOGAS UTILIZATION 38 1.3. CONCLUSIONS FOR CAS + SLUDGE DIGESTION IN EAP COUNTRIES 39 2. CASE STUDY 2: TRICKLING FILTER + SLUDGE DIGESTION 56 2.1. BACKGROUND, PROCESS DESCRIPTION 57 2.1.1. DATA SOURCES 57 2.1.2. WASTEWATER MANAGEMENT 60 2.1.3. SLUDGE MANAGEMENT 60 2.1.4. ENERGY MANAGEMENT 60 2.2. ANALYSIS 60 2.2.1. WASTEWATER INFLUENT, EFFLUENT, AND OTHER PARAMETERS OF INTEREST 60 2.2.2. BIOGAS PRODUCTION & POTENTIAL FOR ENERGY GENERATION 63 2.2.3. OPERATION CAPACITY NEEDS, BIOGAS SAFETY 66 ANNEXES WASTEWATER TO ENERGY   i 2.2.4. INSTITUTIONAL ASPECTS, ENERGY COSTS 66 2.2.5. GHG REDUCTION AND CDM CO-FINANCING 66 2.2.6. CAPEX STRUCTURE 67 2.2.7. OPEX STRUCTURE 68 2.2.8. VIABILITY OF INVESTMENT IN BIOGAS UTILIZATION 68 2.3. CONCLUSIONS FOR TF + SLUDGE DIGESTION IN EAP COUNTRIES 69 3. CASE STUDY 3: UASB 72 3.1. BACKGROUND, PROCESS DESCRIPTION 73 3.1.1. GENERAL BACKGROUND, DATA SOURCES 73 3.1.2. WASTEWATER MANAGEMENT 73 3.1.3. SLUDGE MANAGEMENT 76 3.1.4. ENERGY MANAGEMENT 76 3.2. ANALYSIS 76 3.2.1. WASTEWATER INFLUENT, EFFLUENT, AND OTHER PARAMETERS OF INTEREST 76 3.2.2. BIOGAS PRODUCTION & POTENTIAL FOR ENERGY GENERATION 80 3.2.3. OPERATION CAPACITY NEEDS, BIOGAS SAFETY 84 3.2.4. INSTITUTIONAL ASPECTS, ENERGY COSTS 87 3.2.5. GHG REDUCTION AND CDM CO-FINANCING 88 3.2.6. CAPEX STRUCTURE 89 3.2.7. OPEX STRUCTURE 90 3.2.8. VIABILITY OF INVESTMENT IN BIOGAS UTILIZATION 92 3.3. CONCLUSIONS FOR UASB IN EAP COUNTRIES 93 4. CASE STUDY 4: COVERED ANAEROBIC PONDS 102 4.1. BACKGROUND, PROCESS DESCRIPTION 103 4.1.1. DATA SOURCES 103 4.1.2. WASTEWATER MANAGEMENT 103 4.1.3. SLUDGE MANAGEMENT 107 4.1.4. ENERGY MANAGEMENT 108 4.2. ANALYSIS 109 4.2.1. WASTEWATER INFLUENT, EFFLUENT, AND OTHER PARAMETERS OF INTEREST 109 4.2.2. BIOGAS PRODUCTION & POTENTIAL FOR ENERGY GENERATION 111 4.2.3. OPERATION CAPACITY NEEDS, BIOGAS SAFETY 114 4.2.4. INSTITUTIONAL ASPECTS, ENERGY COSTS 114 4.2.5. GHG REDUCTION AND CDM CO-FINANCING 115 4.2.6. CAPEX STRUCTURE 117 4.2.7. OPEX STRUCTURE 118 4.2.8. VIABILITY OF INVESTMENT IN BIOGAS UTILIZATION 119 4.3. CONCLUSIONS FOR COVERED PONDS IN EAP COUNTRIES 119 ANNEXES WASTEWATER TO ENERGY   ii 5. CASE STUDY 5: CO-DIGESTION OF ORGANIC WASTE 126 5.1. BACKGROUND, PROCESS DESCRIPTION 127 5.1.1. DATA SOURCES 127 5.1.2. WASTEWATER MANAGEMENT 127 5.1.3. SLUDGE MANAGEMENT 128 5.1.4. ENERGY MANAGEMENT 131 5.2. ANALYSIS 131 5.2.1. WASTEWATER INFLUENT, EFFLUENT, AND OTHER PARAMETERS OF INTEREST 131 5.2.2. BIOGAS PRODUCTION & POTENTIAL FOR ENERGY GENERATION 135 5.2.3. OPERATION CAPACITY NEEDS, BIOGAS SAFETY 138 5.2.4. INSTITUTIONAL ASPECTS, ENERGY COSTS 138 5.2.5. GHG REDUCTION AND CDM CO-FINANCING 138 5.2.6. CAPEX STRUCTURE 138 5.2.7. OPEX STRUCTURE 139 5.2.8. VIABILITY OF INVESTMENT IN BIOGAS UTILIZATION 141 5.3. CONCLUSIONS FOR CO-DIGESTION OF ORGANIC WASTE IN EAP COUNTRIES 141 6. CASE STUDY 6: ULTRASOUND SLUDGE DISINTEGRATION 148 6.1. BACKGROUND, PROCESS DESCRIPTION 149 6.1.1. DATA SOURCES 149 6.1.2. WASTEWATER MANAGEMENT 149 6.1.3. SLUDGE MANAGEMENT 149 6.1.4. ENERGY MANAGEMENT 153 6.2. ANALYSIS 153 6.2.1. WASTEWATER INFLUENT, EFFLUENT, AND OTHER PARAMETERS OF INTEREST 153 6.2.2. BIOGAS PRODUCTION & POTENTIAL FOR ENERGY GENERATION 154 6.2.3. OPERATION CAPACITY NEEDS, BIOGAS SAFETY 157 6.2.4. INSTITUTIONAL ASPECTS, ENERGY COSTS 157 6.2.5. GHG REDUCTION AND CDM CO-FINANCING 157 6.2.6. CAPEX STRUCTURE 157 6.2.7. OPEX STRUCTURE 158 6.2.8. VIABILITY OF INVESTMENT IN BIOGAS UTILIZATION 159 6.3. CONCLUSIONS FOR ULTRASOUND SLUDGE DISINTEGRATION IN EAP COUNTRIES 161 This document provides complementary information to the main report. In particular, annex 3 includes the detailed analysis and description of the case studies, which are summarized in section 2 of the main report. ANNEXES WASTEWATER TO ENERGY   iii ANNEXES WASTEWATER TO ENERGY   iv ANNEX 1: ENERGY EFFICIENCY EFFORTS IN EUROPE AND UNITED STATES ANNEXES WASTEWATER TO ENERGY   1 Efforts to Optimize Energy • Denmark 2000: DANVA Benchmarking Project Consumption in Europe (WERF 2010a) The activities related to energy optimization (frequently • Finland 2003: Finnish Benchmarking Project known as “energy efficiency”) at wastewater treatment (WERF 2010a) plants (WWTPs) in Europe started about fifteen years • Germany 2005: Rhineland-Palatinate ago and primarily focused on conventional activated sludge (CAS). Energy optimization entails two (MUFV-RP 2006) components: (a) minimization of energy consumption 2006: Bavaria (Graf 2008) and (b) maximization of biogas production, and they 2006: Baden-Württemberg include the following two methods: (DWA-BW 2010) • Benchmarking Most benchmarking projects focus on two aspects: • Energy manuals • Performance-related indicators, such as effluent The former provides up-to-date information on what quality, removal efficiencies, electricity consumption, is technically and financially feasible, and the latter and so on provides practical advice on how to implement energy- indicators, typically both pertaining to • Cost-related efficient designs and rehabilitate facilities to improve CAPEX and OPEX their energy balance. Generally, benchmarking results are difficult to transfer Benchmarking from one region to another. Local conditions, such as Benchmarking not only provides practical knowledge specific climate, specific discharge requirements, and about what is technically and financially feasible, specific unit costs often have quite different overall but it also encourages competition among and impacts in different regions and countries. This needs within WWTP utilities and leads to generally better to be considered when applying “lessons learned” performance and reduced cost for participants over from one region to another. time. The best known European benchmarking Most of the above benchmarking projects have been projects are the following (in chronological order, repeated ever since their start, typically annually or every indicating the year of project start): two years, so there is currently good and significant • Norway 1998: NORVAR (Norsk Vann BA) knowledge about the sanitation sector in many different (WERF 2010a) European regions, and it is interesting to see the development of project indicators over time. Gradual Dutch Benchmarking Project • Netherlands 1999: OPEX reductions, as depicted in figure A1-1, are to a (WERF 2010a) large extent connected to energy optimization efforts. Austria • 1999: Benchmarking in Sanitary Detailed benchmarking reports are not usually public Engineering (BLFUW 2001) documents. ANNEXES WASTEWATER TO ENERGY   2 Figure A1-1: Development of OPEX for large Austrian WWTPs participating in benchmarking Sources: Wett et al. 2007; Lindtner et al. 2012. Note: EUR/PE110/y x 16.67 = EUR/kg BOD5/y. Energy manuals Energy manuals usually include two main components: Energy manuals have an even longer history in Europe Technical, • energy-related guidebooks for than benchmarking projects. The first manuals were WWTPs, providing background information on developed in Switzerland in 1994, followed by various energy requirements at WWTPs and on energy manuals in Germany, trying to take specific conditions characteristics of all kinds of electro-mechanical and technological developments into account. The equipment 1994 Swiss Energy Manual for WWTPs was updated Practical • recommendations and strategies for most recently in 2010 and can now be considered as implementing energy optimization at WWTPs, representing the state of the art of knowledge on this covering minimization of energy consumption and subject. The European energy manuals that are worth maximization of biogas production and utilization mentioning are the following (in chronological order, indicating the year issued): The main drawbacks of these manuals are that they are (a) mostly written in German, which hampers • Switzerland 1994 (BUWAL 1994) international application, and (b) focused primarily • Germany 1999 North Rhine Westphalia on conventional activated sludge (CAS) systems under (MURL 1999) European climate conditions, and therefore not as 2008 German Association for Water, useful for other process technologies, such as upflow Wastewater, and Waste anaerobic sludge blanket (UASB), trickling filter, and (DWA-BW 2008) waste stabilization pond (WSP) technologies. This 2008 Umweltbundesamt (UBA 2008) does not mean some of the described subprocesses— for instance, sludge dewatering—could not also be • Switzerland 2010 (VSA 2010) applied together with technologies other than CAS. ANNEXES WASTEWATER TO ENERGY   3 Energy optimization • Energy cost at optimized WWTPs had been reduced by an average of 38 percent. A typical energy optimization process for WWTPs has two stages: • 33 percent of the cost reduction was due to improved efficiency; 67 percent was due to increased energy • Energy screening, in which an existing WWTP’s production from biogas. energy condition is checked by using seven indicators (see table A1-1). For each of these, a “target value” • Major efficiency increases were realized in the and an “ideal value” are defined. If any of the seven biological stage and with improved energy parameters fails to meet one or even several target management. values, a detailed analysis is recommended. • Detailed energy analysis, which is based on the Success stories following: Recent reports demonstrate WWTPs are capable of • Individual analysis of all treatment stages and all reaching a positive energy balance—that is, some electro-mechanical installations wastewater facilities do produce more energy than • Definition of (a) short-term measures and (b) they consume. The first example was reported by medium- and long-term measures Wett and others (2007), who described the Strass WWTP in Austria, a 167,000 PE60 CAS plant. An 8 • Evaluation of financial viability of recommended percent electricity surplus was achieved, even though measures the plant was practicing enhanced nutrient removal. Müller and others (2004) evaluated the energy These results had been reached by 2005 through a analysis of 344 WWTPs in North Rhine Westphalia, combination of several efficiency-focused measures, Germany, and produced several major findings: and they sparked widespread interest. This facility • Energy cost can be reduced by an average of 50 is used as a benchmark in Singapore, among other percent. countries (Cao 2011). In the United States it has been included as a case study in an analysis by the Water • Extrapolation of the findings in North Rhine Environment Research Foundation (WERF 2010a); Westphalia indicates a savings potential in Germany and Kang and others (2010) have developed a “four equal to US$4 billion–5 billion (EUR3 billion–4 steps to energy self-sufficiency” road map for U.S. billion) over a fifteen-year period. WWTPs. • Energy optimization is usually financially attractive Two more European facilities provide additional for WWTPs—that is, the NPV (net present value) examples of plants that have achieved positive energy of savings and investments is positive. balances without co-digestion of external waste: Keil For Switzerland, where more than two-thirds of all (2013) reports on the Bad Ischl WWTP in Austria, WWTPs have already undergone energy analysis, Müller a 100,000 PE60 plant that has produced an annual and others (2006) reported similar practical results: electricity surplus between 2 and 12 percent since 2009. Sandino and others (2013) report on Odense ANNEXES WASTEWATER TO ENERGY   4 WWTP, Denmark, a 385,000 PE60 plant that has had These results demonstrate that energy independence similar results. of CAS systems is a feasible target almost anywhere if co-digestion is applied. Without co-digestion, energy Recently, a strong trend has been observed toward independence of CAS systems is feasible, but it is much co-digestion of organic waste, which facilitates the more difficult to achieve. The successful examples achievement of energy independence. Case study 5 make clear that accomplishing it without co-digestion in this report describes a WWTP that achieved 100 requires know-how, a high degree of automation, percent electricity coverage through co-digestion. skilled operators, a clear target, and a commitment to Other successful examples are also cited. its implementation. ANNEXES WASTEWATER TO ENERGY   5 Table A1-1: State of the art indicators for energy screening of CAS, based on European experiences with WWTP benchmarking and energy optimization to date Source: VSA 2010. Note: kWh/PE/y x 16.67 = kWh/kg BOD5 /y. ANNEXES WASTEWATER TO ENERGY   6 Efforts to Optimize Energy Consumption in the (see, for example, WERF 2010a, 2010b, 2010c, and United States 2011a). WEF developed an energy road map (WEF Energy optimization at WWTPs is receiving 2012) (see figure A1-2) and subsequently specified increased attention in the United States, motivated this approach further (WEF 2013a). Generally, most by continuously increasing energy unit costs and large North American consulting companies have discussions about climate change and greenhouse gas been involved in this process in one way or another. (GHG) emissions. Many successful case studies make references to European or Asian facilities. Also, in recent years, an WERF and the Water Environment Federation (WEF) increased number of presentations at U.S. conferences have done considerable work in this area. WERF and workshops have dealt with energy efficiency at started off with a series of reports on best practices and WWTPs. case studies regarding energy efficiency at WWTPs Figure A1-2: Main topics of the U.S. energy road map and levels of progression Source: WEF 2012. ANNEXES WASTEWATER TO ENERGY   7 ANNEXES WASTEWATER TO ENERGY   8 ANNEX 2: SLUDGE DIGESTERS AND RENEWABLE ELECTRICITY POTENTIAL FROM BIOGAS IN THE UNITED STATES AND EUROPE ANNEXES WASTEWATER TO ENERGY   9 Sludge Digesters in the United cited by WERF (2012a) and Stone and Willis (2012), States and Europe stating that less than 20 percent of large WWTPs with According to Renewable Waste Intelligence (RWI anaerobic digestion employ combined heat and power 2013), 1,500 WWTPs (that is, less than 10 percent (CHP). The two studies present a survey on the use of the approximately 16,500 WWTPs in the United of anaerobic digesters and CHP that received over States) have anaerobic digesters, and only 250 of them 200 responses from thirty-six states, Australia, and use the biogas they produce (biogas from the other Canada. The respondents’ rates of anaerobic digestion 1,250 is flared off). This information matches a U.S. and CHP are presented in figure A2-1. Environmental Protection Agency (EPA) estimate Figure A2-1: Rated WWTP flow (MGD) versus biogas use Source: WERF (2012a) and Stone and Willis (2012) The figure shows that across the complete range of This analysis demonstrates that even small plants responding plant sizes from 1 million to more than 500 down to 1 MGD (10,000 PE60) can successfully argue million U.S. gallons per day (MGD; approximately a case for CHP in the United States. 10,000–7,000,000 PE60), anaerobic digestion (AD) is In the case of Europe, Meda and others (2006) being applied. Although larger plants are more likely present the results of a survey for Germany. This to be equipped with AD, many of the smallest have survey covered about two-thirds of total organic it as well. Most plants with anaerobic digestion also loading capacity installed in Germany and about have CHP. Only the intermediary group between 5 one-third of all WWTPs in absolute numbers. They and 35 MGD has a relatively high number of plants found that 76 percent of total PE60 capacity feature without CHP. anaerobic digestion, whereas 35 percent of WWTPs have it—that is, mainly medium and large WWTPs use anaerobic digestion (see figure A2-2). ANNEXES WASTEWATER TO ENERGY   10 Figure A2-2: Installed sludge stabilization technologies in Germany, based on number of WWTPs and on PE60 Source : Meda et al (2006) Renewable Electricity Potential from Biogas at this electricity production is substantial, it is only a WWTPs in the United States and Europe little over 10 percent of the actual consumption of all According to WEF (2011), more than 500 plants in U.S. WWTPs of about 40 million MWh/y (WERF the United States currently use anaerobic digestion 2010b). without CHP. If all of them were to install CHP In Germany, with 1 PE60 able to produce an average systems, they could generate approximately 340 of 15 kWh/PE60/y, the total electricity production megawatts (MW) of electricity. The total CHP potential from sludge equals about 1.8 million MWh/y potential from all WWTPs in the United States over at present. That is equivalent to more than 40 percent 1 MGD is quantified as 600 MW (Stone and Willis of the actual electricity consumption of all WWTPs of 2012). According to WERF (2012a), “WWTPs have 4.2 million MWh/y (DWA 2013b). As mentioned in the potential to generate an additional 200 to 400 the case of the United States, to further increase that MW of power from biogas.” percentage, utilities would have to combine the effects Assuming the 600 MW capacity is in operation all year of reduced electricity consumption at WWTPs with round, this would result in the production of about 5 measures to further increase biogas quantities. million megawatt hours per year (MWh/y). Although ANNEXES WASTEWATER TO ENERGY   11 ANNEXES WASTEWATER TO ENERGY   12 ANNEX 3: CASE STUDIES ANNEXES WASTEWATER TO ENERGY   13 CASE STUDY 1: CAS + SLUDGE DIGESTION ANNEXES WASTEWATER TO ENERGY   14 1.1. BACKGROUND, PROCESS DESCRIPTION 1.1.1. Data sources sludge (CAS) process. All CAS technology variations Case study 1 is based on publicly accessible data from have in common that they produce waste activated Europe, which are based on thousands of WWTPs, sludge. Extended aeration is excluded from in-depth so that sources such as benchmarking results for analysis in case study 1, since extended aeration does CAS and national/regional wastewater treatment sludge stabilization inside the aeration tanks and does statistics represent a solid data base. The core data not produce biogas for energy generation. for case study 1 were taken from Germany and The typical CAS technology used in this case study Austria, which have similar treatment requirements (see figure CS1-1) consists of preliminary treatment and technology standards. (screen, sand/fat removal), followed by primary 1.1.2. Wastewater management sedimentation tank (PST) and, subsequently, aeration The analysis made in this case study assumes the tank (AT) and secondary sedimentation tank (SST). reader is familiar with the conventional activated Figure CS1-1: Simplified flow scheme of CAS and mesophilic sludge digestion Thermal energy ANNEXES WASTEWATER TO ENERGY   15 1.1.3. Sludge management See table CS1-4 for general design parameters and key The assumption in the following analysis is that all characteristics of anaerobic digesters. sludge produced in wastewater treatment, both primary and secondary, will be thickened and subsequently 1.1.4. Energy management anaerobically digested. Thickening is important, since The biogas produced in the digesters can be utilized it reduces sludge quantities considerably and thus for the production of electric and/or thermal energy, allows for smaller digester designs, which produces in their combined form also called combined heat and CAPEX savings. Primary sludge thickens well and power (CHP) production. Prior to its utilization, the rapidly by gravity in the PST, up to some 3 percent gas is balanced in a gas holder and purified, as required. dry solids (DS). Therefore, separate thickeners for A flare is typically used to burn gas that cannot be primary sludge are frequently not necessary. Since utilized and for emergency situations. waste activated sludge rarely exceeds 1 percent DS, See table CS1-5 for general design parameters and key it requires separate thickening prior to digestion. To characteristics of biogas systems. maximize thickening efficiency and thus minimize digester cost, waste activated sludge is usually thickened mechanically to about 6 percent DS. Anaerobically digested sludge is generally dewatered prior to reuse or disposal. ANNEXES WASTEWATER TO ENERGY   16 1.2. ANALYSIS 1.2.1. Wastewater influent and effluent of Germany and Austria—DWA and OEWAV, Most benchmarking initiatives assume that respectively—compile these data in an annual WWTP participating WWTPs comply with legal effluent performance analysis. Table CS1-1 provides data for criteria and therefore do not present specific the year 2012 (DWA 2013a). information on influent and effluent wastewater characteristics. The national wastewater associations Table CS1-1: Average influent and effluent data of WWTPs in Germany and Austria in 2012       Germany Austria Number of WWTPs 5,917 906 WASTEWATER QUANTITY       Specific wastewater production m3/PE60/y 80 67     L/PE60 /d 219 184 WASTEWATER QUALITY       COD Influent mg/L 548 656   Effluent mg/L 27 44   Elimination % 95 93 BOD5* Influent mg/L 255 n.a.   Effluent mg/L 4.5 n.a.   Elimination % 98 n.a. Ntotal Influent mg/L 51 43   Effluent mg/L 9 9   Elimination % 82 79 NH4-N Effluent mg/L 1.2 1.2 NO3-N Effluent mg/L 6.0 5.7 Ptotal Influent mg/L 7.9 7.5   Effluent mg/L 0.7 0.7   Elimination % 91 91 Source: DWA 2013a. Note: 1 cap = 60 g BOD5/d in Germany and Austria; 1 PE60 = 1.0 cap in Germany and Austria; m3/PE60/y = m3/cap/y in Germany and Austria; L/PE60/d = L/cap/d in Germany and Austria. *Data from 2010. n.a. = not available. ANNEXES WASTEWATER TO ENERGY   17 With a chemical oxygen demand (COD) of about 1.2.2. Biogas production and potential for 600 mg/L, the raw wastewater is of average strength. energy generation This allows for efficient application of primary Project-specific biogas production rates can be sedimentation tanks (PSTs), which is advantageous calculated with the design values provided in table for high biogas production rates. On the other hand, CS1-5. However, this requires the prior definition of nitrogen reduction is high at these WWTPs; hence, sludge quantities and qualities. If these definitions are on top of high energy requirements for CAS, in not available, the summaries in figure CS1-2 of biogas general, there is additional energy consumption for the production for different conditions can be used for a oxidation of nitrogen. Under these nutrient reduction rough first estimate. The indicated ranges are a reliable conditions, WWTPs are usually not able to recover reflection of what is actually observed in practice, too. all 100 percent of their energy needs from the biogas. Figure CS1-2: Daily biogas production rate per PE60 at large WWTPs Source: VSA 2010. Note: L/PE60 /d x 16.67 = L/kg BOD5 /d. WWTPs with the best treatment performance Whereas WWTPs that only provide for carbon (C) (enhanced nitrogen (N) removal) show the lowest removal and include large PSTs can reach higher biogas production rates. This is due to short–retention biogas production rates of about 25–30 L/PE60/d, the time primary settling tanks (PSTs), usually required decisive question for EAP and other warm countries for efficient denitrification, and to high sludge age in will be in which loading range their CAS plants operate. the bioreactors, which leads to increased consumption Nitrification in warm climates is much faster than in of carbon, reduced quantities of fresh primary sludge, cold climates. Hence, even if a WWTP is designed and partially stabilized secondary sludge. A typical for carbon removal only, it will nitrify its wastewaters WWTP under these conditions might produce just unless due attention is paid to operating it in a loading about 20 L/PE60/d. range that does not yet lead to nitrification. ANNEXES WASTEWATER TO ENERGY   18 Note that the above figures hold true for WWTPs Methane’s calorific value of 10 kWh/m3 is important larger than 100,000 PE60 and are based on mesophilic to estimating the potential for energy generation digesters with a retention time of thirty days. For other from biogas production. Since biogas contains about conditions, consider the following: 60 to 70 percent methane, its calorific value usually ranges from 6.0 to 7.0 kWh/m3 after cleaning and • In cases of shorter retention times in the digester of moisture removal. The most common and economical just about fifteen to twenty days, the values of figure options for utilizing this resource are co-generation CS1-2 should be reduced by 10 percent. and microturbines. Consequently, the overwhelming • Likewise, small and medium WWTPs will produce majority of the WWTPs considered for case study 1 somewhat reduced biogas quantities, mainly due also use one or the other of these technologies. to reduced efficiencies caused by stronger impacts Based on the above described biogas production rates of peak loading periods. It is possible to assume 5 and typical electric efficiencies for CHP, as indicated percent lower values for a 50,000 PE60 plant and 10 in figure CS1-31, the potential for energy generation percent lower for a 20,000 PE60 plant. through CAS + mesophilic digestion is as shown in table CS1-2. Table CS1-2: Biogas and power generation potential of CAS + mesophilic digester at large WWTPs > 100,000 PE60   Retention time in PST (h)   0.5 1.0 1.5 2.0 Biogas production   - N elimination (L/PE60/d) 20 22 23 — - C elimination (L/PE60/d) 24 27 28 29 Electric efficiency CHP (%) 33 33 33 33 Thermal efficiency CHP (%) 50 50 50 50 Calorific value of biogas (kWh/m ) 3 6.3 6.3 6.3 6.3 Power generation   - N elimination (kWh/PE60/year) 15.2 16.7 17.5 — - C elimination (kWh/PE60/year) 18.2 20.5 21.2 22.0 Thermal energy generation   - N elimination (kWh/PE60/year) 23.0 25.3 26.4 — - C elimination (kWh/PE60/year) 27.6 31.0 32.2 33.3 Total energy generation   - N elimination (kWh/PE60/year) 38.2 42.0 43.9 — - C elimination (kWh/PE60/year) 45.8 51.5 53.4 55.3 Source: Authors’ calculation. Note: L/PE60/d x 16.67 = L/kg BOD5/d; kWh/PE60/y x 16.67 = kWh/kg BOD5/y. ANNEXES WASTEWATER TO ENERGY   19 Generally, the calculated values describe average of air from entering biogas systems. Micro-aeration conditions. For specific conditions, a certain amount of digesters, used for H2S removal, does not present of adjustment can be necessary—for instance, in the a safety problem. Confined areas can sometimes pose case of very large co-generation units with higher than risks for workers if methane and carbon monoxide assumed electric efficiency or of microturbines or of a accumulate. different calorific value of the biogas. Typically, those The Austrian wastewater association published a variations can affect the cited energy generation values compilation of all accidents related to biogas explosions by not more than about +/– 10 percent. in the previous twenty years (OEWAV 2006). The These values in table CS1-2 are confirmed by actual main recommendations that can be drawn from those benchmarking data from Austria (Lindtner 2011) (few) accidents are the following: and Germany (DWA-BW 2010; Graf 2010). These • When repairing a digester/gas holder/sludge pipe, practice data also show the electricity potential is not instruct workers very clearly about the necessary fully exploited at present. Of the biogas in the region, safety precautions. Particularly avoid any sources 20–30 percent still goes unused or is just utilized in of spark ignition, such as welding, metal work, burners for heat production. Notwithstanding, there is smoking, and so on when this ignition spark might a strong trend toward further increasing electric power get into contact with biogas. generation at WWTPs, particularly since energy unit cost is rising fast in Central Europe. • Always assign one person who is clearly in charge for supervision of work and safety precautions. 1.2.3. Operation capacity needs, biogas safety • At design, implementation, and operation stages, The key operational issues concerning anaerobic always adhere to standard safety precautions. digesters involve specific operational problems and Important precautions include those mentioned by safety concerns related to the management of explosive Germany’s wastewater association (DWA 1996): biogas. The most relevant aspects are those discussed below: • Maintain a slight overpressure in biogas systems to avoid entry of air. • Safety concerns • Design sludge feeding and abstraction based • Digester foaming on the sludge replacement principle: the sludge • Deposits in the digester entering the digester should push the same amount of sludge out of the digester. This will • Insufficient biogas treatment ensure the overpressure in the biogas systems can Safety concerns be maintained. Although accidents occur, they are very rare. Biogas • Take particular care when removing scum from operating conditions have a risk of explosion only the digester surface. This is an action during which when 5–15 percent biogas is mixed with 85–95 considerable amounts of biogas could be released percent air. It is important to prevent large quantities to the open air. ANNEXES WASTEWATER TO ENERGY   20 • Separate digester, gas holder, and biogas utilization Digester foaming through appropriate valves. Note that water-filled Digester foaming is occasionally observed in digesters foam traps are safer than mechanical valves, which at WWTPs. Once it appears, it can last from one tend not to be 100 percent foolproof. day to several weeks. Persistent foaming is rare. The • Construct gas pipes and gas installations from practical impact of foaming is that the foam tends to corrosion-resistant material only— for instance, clog the biogas system, and it can affect pumps and grade 1.4571 stainless steel or high-density mixing/heating devices. As a consequence, foaming polyethylene (HDPE). Take temperature effects results in extra manual repair and cleaning work. Also, on HDPE into due account. persistent foaming reduces the active volume in the reactor, thereby affecting the digester’s performance • Condensate removal must work automatically. through reduced VS (volatile solids) destruction, Manual condensate removal installations prove which in turn lowers biogas production (Kougias et unreliable in practice. al. 2013; Rodríguez-Roda et al. 2013; Moos 2012; • Favor pipe welding over pipe flanges. Also, Shimp et al. 2010). compensators should preferably be welded to the pipes. The following can cause foaming: • The ventilation of adjacent rooms and pipe • Filamentous microorganisms in the sludge that collectors should preferably be done through proliferate during wastewater treatment, most slight overpressure inside these rooms, rather than notably Microthrix parcivella, at low-loaded CAS through air suction out of them. plants. • Install and regularly maintain gas warning • Organic overloading of the digester, particularly in installations, particularly in those locations cases of high protein and/or ammonia concentrations where biogas accumulates. These installations in the digester feed. This rarely happens in are supposed to trigger an alarm before critical conventional digesters that exclusively digest sludge, concentrations are achieved. but it should be controlled at co-digestion facilities. • For any repair and maintenance work, use special • Small sludge surface area in the digester, leading to tools that cannot cause ignition sparks. intensified area-specific gas release. • Respect special operation and maintenance • The presence of surface active agents in the raw instructions from the providers of biogas and sludge, as well as surface active products of digestion, digester equipment. both of which tend to make the foam more stable. • Respect local legal requirements for biogas systems. ANNEXES WASTEWATER TO ENERGY   21 Operators should be aware that large biogas volumes, compared the phenomenon of digester foaming with as compared to reactor volume, are a major driving “what happens when a carbonated drink is poured force behind foaming. Shimp and others (2010) into a glass.” summarized the situation (figure CS1-3) and Figure CS1-3: Digester gas production rate relative to liquid volume Source: Shimp et al (2010) Typical countermeasures include the following: effective if filamentous microorganisms are causing the foaming. A typical dosage rate is about 1 g Al/ • Preventive, regular (microscopic) control of kgDS/d. microbiological characteristics of digester sludge • Waste activated sludge (WAS) chlorination • Temporary lowering of the liquid level in the digester • Dosage of chemical defoamers into the digester • Reduction of organic loading of the digester • Optimized mixing of the digester • More consistent raw sludge feeding to digesters instead of intermittent feeding in batches • Installation of foam level detection in the digester, possibly even combining it with automated response • Utilization of chemical precipitant containing (for instance, automatic lowering of the sludge level) aluminum, by dosing it into the wastewater train for P precipitation and/or by direct dosage into • Installation of special foam destruction installations, the digester. Poly-aluminum-chlorides (PAC) have such as water nozzles or mechanical devices at the become the product most commonly used for digester’s liquid surface this purpose. PAC dosage has proved particularly ANNEXES WASTEWATER TO ENERGY   22 • Installation of sludge disintegration prior to sludge destruction and biogas production are reduced. Since feeding. This proves particularly helpful where the the volume of deposits can be quite substantial—50 foaming is caused by filamentous bacteria. percent reductions are not unheard of—the negative effects on biogas production can also be substantial • Avoidance as much as possible of physical scum (see figure CS1-4). In addition, mixing systems can removal from digesters. After all, this measure carries be damaged, and pipes can be clogged. Moreover, the the risk of substantial quantities of biogas being deposits tend to solidify to the point where there is no released as well, which in turn increases the risk of easy way of removing them. Significant down times of explosion (see safety concerns, above). several weeks for digester cleaning and repair may be Deposits in the digester incurred as well. Deposits in a digester reduce its active reactor volume. Hence, sludge retention time shortens, and VS Figure CS1-4: Typical shape of deposits in a digester Source: Heumer 2012. ANNEXES WASTEWATER TO ENERGY   23 The following are causes of deposits in digesters: • Installation and proper maintenance of efficient stages for grit removal • Low efficiency of WWTP pretreatment screens. Even standard fine screens with 6 mm free open • Installation and proper maintenance of sludge spacing allow the passage of considerable quantities grinders or microstrainers, particularly for primary of solid materials. sludge, prior to sludge feeding into the digesters • Low efficiency of grit chambers. Any sand passing • An experts’ discussion is currently ongoing on the the grit chambers will inevitably end up in the overall advantages and disadvantages of giving up sludge, and a majority will deposit in the digesters. on enhanced biological P removal technologies from wastewater (bio-P) to minimize MAP • In the case of co-digestion, feeding of solid residues (struvite) formation. Dosing Fe into the digester into the digesters that were not efficiently removed for precipitation of orthophosphate has proved from organic waste successful in further reducing struvite formation. • MAP (magnesium ammonia phosphate, also called This also has the positive side effect of H2S removal struvite) formation in the digesters (see figure CS1-5) from biogas. • Some utilities have even installed struvite recovery • Inefficient mixing systems processes to turn the problem into a value-added • Lack of digester cleaning/maintenance resource. Figure CS1-5: Struvite deposits from a digester Once the problem of deposits manifests (and also in the event that an operator wants to assess a digester’s actual condition), there are several ways of approaching the issue: • Temperature effects. Digesters frequently have temperature sensors both near the bottom and in the upper half. Since deposits usually build up from the bottom, the temperature metered by the bottom sensor differs from that metered by the upper sensor. Ebner (2013) documented a case where the difference between the two sensors amounted to Source: Ebner 2013. more than 10°C. Such a temperature difference can be taken as an indication of deposits, but it does not allow any quantification of the problem. Typical countermeasures include these: • Tracer testing. This is the typical means of • Installation and proper maintenance of ever finer quantifying the active reactor volume. A tracer (for screens. An open spacing of 6 mm should be seen as example, lithium chloride) is fed to the digester, and the upper limit for any WWTP with digesters. Finer its response is analyzed. Two different approaches screens of 5 to 3 mm are preferable nowadays. are possible: ANNEXES WASTEWATER TO ENERGY   24 A. The tracer concentration is analyzed in the digester to derive both (a) the time required for complete digester effluent over time. From the characteristics and mixing and (b) the active reactor volume of the digester. duration of the tracer response in the effluent it is This type of tracer testing is generally completed within possible to derive both (a) short-circuiting currents less than twenty-four hours. It is marketed under the name and (b) the active reactor volume of the digester. This “VoluSense” (Ebner 2013). type of tracer testing usually takes days or weeks. A tracer study of an anaerobic digester with method B. An exactly defined quantity of tracer is added, the B costs on the order of US$5,000. Method A, since it digester is mixed without a feeding of fresh sludge, and takes more time, usually costs between US$5,000 and the tracer is analyzed in the external circulation pipe US$15,000, depending on the size of the digester and that is usually used for mixing fresh feeding sludge the extent of the analysis. with sludge from the digester. From the characteristics Typical tracer response curves of both methods are and duration of the tracer response it is possible presented in figure CS1-6. Figure CS1-6: Examples of tracer response methods A and B (Method A) (Method B) Sources: Nolasco et al. 2000; ARAconsult 2012. ANNEXES WASTEWATER TO ENERGY   25 • Divers. Divers can also be employed to assess the have the advantage not only of being able to provide size of deposits in digesters. Several companies have a reliable quantification of the required cleaning specialized in this field. When selecting a diving work and cost; they can also remove deposits, or at company, it is of particular importance to check their least take samples of them to assess their composition specific references. Diving in a digester’s environment and the difficulties of removal. The disadvantage of requires special training, experience, and equipment divers is that using them typically costs more than a (see figure CS1-7). Aside from safety considerations, simple reactor volume and mixing analysis. A diver there have been reports of inexperienced divers inspection usually costs around US$10,000. Divers greatly underestimating the deposits’ volume and specializing in this type of work may not be easily removal costs (KA-Betriebs-Info Editor 2011). found in EAP. Divers who are professionals in the field, however, Figure CS1-7: Divers at work in sludge digesters Source: www.umwelttauchservice.at. Once the degree of reduction of active reactor volume periods. It also allows faster removal of the deposits. and short circuiting are assessed, engineers must decide A typical diver emptying can take about two to three whether to remove deposits and improve mixing weeks, while complete emptying and restarting can conditions in the digester. The decision of cleaning take up to several months (Heumer 2012). However, the digester will depend on multiple factors, including the cost of diver cleaning is relatively high; it can degree of reduction of reactor volume, plant flexibility, quickly reach around US$100,000 per digester (Jilg and energy generation aspects, among several others. 2012; Heumer 2012). If cleaning is recommended, there are two options for • Complete emptying of the digester. As mentioned, removing deposits: removing deposits by emptying the digester usually • Divers. Divers can remove deposits without the takes more time (generally measured in weeks) than whole reactor being emptied. This is sometimes a using divers, both for the reactor cleaning itself and simple practical necessity, since emptying a digester for restarting the emptied digester and reaching means managing large sludge volumes in short time maximum biogas production again. Emptying also ANNEXES WASTEWATER TO ENERGY   26 requires a series of auxiliary installations, such as is sometimes required. When biogas treatment is special pumps, suction trucks, cranes, and mobile neglected, it may quickly lead to corrosion problems mechanical sludge dewatering to cope with the and to failure of the CHP (see figure CS1-8). Since sludge qualities and quantities in question. One the necessary repair costs could be high, all too often advantage of emptying is the opportunity for a visual biogas utilization is then stopped altogether. There inspection—by the operators themselves—of the are relatively cheap options for biogas treatment that pipes and installations located inside the digester, as would safely avoid these problems. well as the condition of the concrete walls. If the As described in Table CS1-5, two main parameters mixing system is damaged or needs maintenance, require special attention: H2S and siloxanes. The the digester can be out of service for a long time. table also describes typical threshold values and cost- Since emptying and cleaning a digester relies heavily effective treatment systems and their design. When on manpower, its cost will depend on labor costs these technologies are properly applied, the lifespan of and digester size. For a fair cost comparison with the biogas utilization can be expected to be long. use of divers, the cost of extra manpower provided by the operator and the loss of biogas and energy To some extent, slightly elevated concentrations of generation during this period must be included in certain substances can be acceptable without treatment. addition to contractor costs. In such cases, shorter maintenance intervals and more frequent lubrication oil changes usually compensate Insufficient biogas treatment for missing treatments. Such substitutions should For the utilization of biogas from digesters, it is always be made in accordance with the CHP suppliers’ necessary to treat the gas prior to its use (for details, see conditions; otherwise, the operator runs a high risk “biogas treatment” in table CS1-5). Foam trapping and of being left without a supplier guarantee if damage condensate removal are standard treatments that must should occur. be available at any plant. However, further treatment Figure CS1-8: Typical damage caused by insufficient H2S and siloxane removal: sulfate scaling in a heat exchanger and scaling of siloxanes Source: VSA 2012b. ANNEXES WASTEWATER TO ENERGY   27 1.2.4. Institutional aspects, energy costs Electricity supplies from WWTPs to the grid receive The utilization of biogas in the region of case study 1 EUR0.0589–0.0679/kWh (US$0.080–0.092/ (Austria and Germany) is most attractive if the biogas kWh). This is generally less than 50 percent of the is converted into electric power, since this is a valuable unit electricity cost the WWTPs have to pay when form of energy. It is therefore vital to understand purchasing from the public grid. Consequently, onsite the institutional and legal conditions under which generation and utilization of electricity from biogas at the current boom in biogas production and CHP is the WWTPs is more attractive than supplying it into taking place. Overwhelmingly, WWTPs in Europe the public grid. consume the generated electricity onsite. In those rare Austria also has a legally defined feed-in tariff for cases in which a surplus is generated, it is usually sold electricity from renewable sources. Details are defined in to the public grid at guaranteed feed-in tariffs that are Ökostrom-Einspeisetarifverordnung 2012 (ÖSET-VO paid per kWh supplied. A helpful survey of the actual 2012, By-Law on Renewable Electricity Feed-in Tariff). regulations on renewable energy generation in Europe In 2014, the feed-in tariff for electricity produced from can be found at http://www.res-legal.eu, which also WWTP biogas equaled EUR0.0594/kWh (US$0.080/ contains all relevant country-specific details. kWh), fixed for a period of thirteen years. In Germany, the Erneuerbare-Energien-Gesetz Other issues and conclusions related to German and (EEG), the Renewable Energy Sources Act, regulates Austrian regulation in 2014 include the following: the supply of electric power to the public grid from renewable sources. It was first introduced in 2000 and 1. The overwhelming majority of WWTP operators has been revised several times since, most recently in utilize power from biogas to cover their own 2012. The EEG defines minimum prices per kWh that electricity needs onsite. Supply to the public grid must be paid (for a period of usually twenty years) to is considered only in rare cases of surplus energy. the supplier of electric energy from renewable sources, Usually the latter is not financially attractive due such as wind-, solar-, hydro-, geothermal-, biomass-, to low feed-in tariffs. landfill-, and wastewater-based power generation. This 2. The feed-in tariffs for electricity from other law also covers biogas from WWTPs. Grid operators resources (for instance, solar and wind) are are obliged to give priority to renewable sources when considerably higher than those for biogas-based purchasing and transmitting electricity. Additionally, electricity and can go up to about EUR0.20/kWh the German Bank KfW is providing low-interest (US$0.27/kWh) in both countries. loans, usually with fixed interest rates of about 1 percent per year for ten-year loans, for investments 3. Domestic consumers currently pay an average into renewables. electricity tariff of about EUR0.20/kWh (US$0.27/kWh) in Austria and EUR0.26/kWh The initial minimum guaranteed feed-in tariff is (US$0.35/kWh) in Germany. gradually reduced over the years. The intention is 4. Industrial consumers, including WWTPs, pay to promote renewables strongly in the beginning, lower electricity tariffs than domestic consumers. but also to make them more efficient over time. WWTPs are typically charged about EUR0.10/ ANNEXES WASTEWATER TO ENERGY   28 kWh (US$0.135/kWh) in Austria and EUR0.15/ instance, as described for WWTPs in the section on kWh (US$0.20/kWh) in Germany. institutional aspects, above). 5. When the period of guaranteed feed-in tariffs • It uses a trading scheme for GHG emissions to ends, WWTPs usually have to cope with introduce a fair price for pollution. substantial reductions of their feed-in tariffs. Feed- The EU’s emissions trading scheme (ETS) operates in tariffs could be as low as EUR0.02–0.03/kWh under the Kyoto Protocol via the Clean Development (US$0.03–0.04/kWh). Only then do alternatives Mechanism (CDM), by which countries that have to grid supply for (rare) surplus energy, such ratified the Kyoto Protocol can invest in projects that as supply into natural gas pipelines, become reduce GHG emissions in developing countries. These attractive. investments can be traded to signatory countries, which can use these certified emissions reductions 1.2.5. GHG reduction and CDM co-financing (commonly known as carbon credits) to meet their General commitments under the Kyoto Protocol. European climate policy targets a drastic reduction of At present, 18 percent of projects registered as CDM greenhouse gas (GHG) emissions. To boost renewable worldwide are based on anaerobically digested biogas, energy and minimize energy inefficiencies, the with Brazil, Malaysia, Mexico, and the Philippines European Union (EU) uses three instruments: the countries with the most projects (Chamy 2013). • At the EU level, it defines emissions targets. Just The majority of projects originate from the industrial recently, on January 22, 2014, the new targets sector and not from digesters at municipal WWTPs. were announced by the EU Commission: (a) 27 Price of carbon credits percent of total energy is supposed to come from renewables by 2030 for the EU as a whole; (b) 40 The basic idea of the emissions trading scheme is percent reductions of GHG emissions until 2030, that GHGs are reduced where the cost to do so is as compared to emissions in 1990. lowest. However, the EU market was flooded with an excessive number of permits, so the price decreased, • At the national level, it provides financial incentives and in 2014 it fell to a low of about EUR5/ton CO2e and subsidies to achieve the emissions targets (for (US$6.8/ton CO2e); see figure CS1-9. Figure CS1-9: EU ETS carbon spot price NOTE: The EU ETS is just one specific emis- sions trading scheme prevailing in the region of case study 1. Besides ETS, there are others, and prices may vary from scheme to scheme. Not- withstanding, what all schemes have in common at present are (very) low prices per ton CO2e, some being even lower than those of ETS. Source: The Economist 2014. ANNEXES WASTEWATER TO ENERGY   29 As a consequence, the actual cost for cutting emissions A change would be possible if the real cost were, is considerably higher due to the combined effects of indeed, reflected by the cost of carbon credits. It subsidies. The Economist (2014) sets the cost at over would be an indication of what the real cost is when EUR150/ton CO2e (US$200/ton CO2e) under the assessing companies’ internal risk calculations. Many renewables program. Therefore, co-financing through companies use an “internal carbon price” per ton of ETS is not attractive for the time being. CO2e for planning purposes (see figure CS1-10). Figure CS1-10: Internal carbon price of selected companies Source: CDP = Carbon Disclosure Project (UK-based NGO), cited in The Economist 2013. The internal prices range from US$6/ton CO2e at Relevant aspects for CDM projects Microsoft to US$60/ton CO2e at Exxon Mobil. Under the principle of additionality, GHG emissions A particularly close look at those companies that from a CDM project must be reduced below those emit large amounts of GHG seems to indicate that levels that would have occurred in the absence of the a price of about US$40/ton CO2e reflects reality project. This poses uncertainties to any CDM project better than the results from ETS. If this price were to at WWTPs involving biogas utilization from new materialize, CDM co-financing could generally, and digesters. After all, without the digesters, and therefore also specifically for WWTPs, become a much more without the project, there would be no methane that appealing instrument in the future. could be reduced. Since methane is a stronger GHG ANNEXES WASTEWATER TO ENERGY   30 than CO2, already relatively low levels of methane alone. It also includes indirect impacts of the digester, utilization bring substantial carbon credits. such as reduced electricity consumption in (already existing) aeration tanks, where extended aeration is A second issue is the question of how to calculate the no longer required, since the sludge will be stabilized potential reduction in CO2e for a digestion project in the digesters after project implementation. The at a WWTP. Two main approaches can be applied total of electricity savings plus electricity generation individually or combined: reduction of electricity implies a reduced need for that specific amount of generation from fossil fuels and reduction of methane electricity generated from fossil fuels. The question emissions. is, then, how much of GHG emissions are caused by Reduction of electricity generation from fossil fuels electricity generation from fossil fuels? This depends is the classical approach when a WWTP is equipped on a country’s specific energy matrix. Different fossil with a new digester and a new CHP for the utilization fuels imply different GHG emissions in electricity of biogas. Since the sludge is considered a renewable, production (see figure CS1-11). Technology specifics any electricity generation from a sludge digester’s and country specifics influence the respective GHG biogas is from renewable sources. This consideration emissions for electricity production. is not limited to the electricity generation from biogas Figure CS1-11: Specific GHG emissions of electricity generation from fossil fuels in the world Source: World Nuclear Association 2011. Note: 1 ton CO2e/GWh = 1 g CO2e/kWh ANNEXES WASTEWATER TO ENERGY   31 Table CS1-3: GHG emissions per kWh for electricity generation in Germany and Austria, compared to European Union and the world Note: The table shows CO2 emissions from fossil fuels consumed for REGION 1990 2010 electricity generation, in both electricity-only and combined heat and gCO2/kWh gCO2/kWh power plants, divided by output of electricity generated from fossil Germany 607 461 fuels, nuclear, hydro (excluding pumped storage), geothermal, solar, wind, tide, wave, ocean, and biofuels. Both main activity producers and Austria 238 188 autoproducers have been included in the calculation. EU (27 member countries) 585 429 World 586 565 Source: IEA 2012. As it turns out (see table CS1-3), GHG emissions are methane generated in such ponds to the atmosphere. rather different, depending on the composition of a In Bolivia, covering the ponds to capture biogas would country’s or region’s energy production matrix. These be additional to business-as-usual (baseline situation). specifics always have to be considered in a CDM project. Thereby, mitigation of methane by collection and burning can be considered in the CO2e calculation, in Reduction of methane emissions can usually be addition to any other credits originating from electricity applied with existing WWTPs, where methane is generation with biogas, which displaces generation already being emitted into the open air and the CDM emissions elsewhere. project is introducing new installations that mitigate those emissions. Simple collection of the methane and 1.2.6. CAPEX structure subsequent flaring can render a project eligible for the Up-to-date cost curves have been developed for CAS CDM mechanism. However, the approach may also (Tectraa et al. 2010, 2011; Gretzschel et al. 2012) include a component of electricity generation from the reflecting the actual costs of all prevailing elements collected methane, which further increases its potential for sludge digestion in the area of case study 1 when for carbon credits. For instance, the covering of an open switching from extended aeration to mesophilic sludge sludge digester (but also of any other stages, such as digestion. The elements of the cost analysis include anaerobic ponds) and capture of the emitting methane the following: would fall into this category. • CAPEX of primary sedimentation tank (figure CS1- As pointed out before, due to the additionality criterion, 12) the new construction of a digester complete with CHP only takes into account the energy production • CAPEX of anaerobic digester (figure CS1-13) component under CDM, but not the methane • CAPEX of all other concerned items (intermediate elimination, since the methane is created and eliminated pumping station, primary sludge pumping station, by the project itself. An exception to this would apply primary sludge buffer tank, mechanical sludge in countries where business-as-usual conditions are to thickener, gas holder, gas flare, CHP, operation release biogas to the atmosphere without burning it. building, pipes, traffic areas, other costs; see figure For example, in Bolivia, most anaerobic ponds are not CS1-14) covered, hence the baseline in this country is to release ANNEXES WASTEWATER TO ENERGY   32 Figure CS1-15 presents the overall results from all three items together. Figure CS1-12: Specific CAPEX for primary sedimentation tanks Source: Tectraa et al. 2010, 2011; Gretzschel et al. 2012. Notes: CAPEX includes engineering and 19 percent value-added tax (VAT). EUR/PE60 x 16.67 = EUR/kg BOD5. Figure CS1-13: Specific CAPEX for Sludge Digesters Source: Tectraa et al. 2010, 2011; Gretzschel et al. 2012. Notes: CAPEX includes engineering and 19 percent value-added tax (VAT). EUR/PE60 x 16.67 = EUR/kg BOD5. ANNEXES WASTEWATER TO ENERGY   33 Figure CS1-14: Specific CAPEX for all other items concerned Source: Tectraa et al. 2010, 2011; Gretzschel et al. 2012. Notes: CAPEX includes engineering and 19 percent value-added tax (VAT). EUR/PE60 x 16.67 = EUR/kg BOD5. Figure CS1-15: Specific total CAPEX for technology change to mesophilic sludge digestion Source: Tectraa et al. 2010, 2011; Gretzschel et al. 2012. Notes: CAPEX includes engineering and 19 percent value-added tax (VAT). EUR/PE60 x 16.67 = EUR/kg BOD5. ANNEXES WASTEWATER TO ENERGY   34 Since the preceding figures do not explicitly present CAPEX for CHP, figure CS1-16 summarizes CHP cost, as indicated in different sources. Figure CS1-16: CAPEX for microturbines and co-generation Source: Wacker 2007; Geyer and Lengyel 2008; ASUE 2011. WERF (2010b) mentions an estimate for CAPEX of 1.2.7. OPEX structure US$4,124/kW (EUR3,050/kW) for microturbines. The same cost basis (Tectraa et al. 2010, 2011; No indication of the electric capacity behind this Gretzschel et al. 2012) that was presented in the value is provided. Nonetheless, the CAPEX level of previous section for CAPEX is used for the OPEX microturbines seems similar between the United assessment of case study 1. The following cost items States and Europe. are distinguished when introducing mesophilic sludge digestion. ANNEXES WASTEWATER TO ENERGY   35 OPEX increase: solids in digester as compared to extended aeration and (b) better dewatering properties of digested sludge • Extra OPEX of operation and maintenance (O&M) of additional installations • OPEX savings due to reduced electricity consumption • Extra OPEX of extra personnel needed to operate • OPEX savings due to use of electricity from biogas to additional installations reduce electric power purchases from the public grid OPEX decrease: Figure CS1-17 presents the quantified OPEX increases and decreases under the specific conditions • OPEX savings due to reduced sludge disposal cost, in Germany. Figure CS1-18 presents the overall brought about by (a) stronger degradation of volatile OPEX reduction. Figure CS1-17: OPEX increase and decrease caused by mesophilic digestion and biogas utilization Source: Tectraa et al. 2011. Notes: CAPEX includes engineering and 19 percent value-added tax (VAT). EUR/PE60 x 16.67 = EUR/kg BOD5. Figure CS1-18: Overall OPEX reduction caused by mesophilic digestion and biogas utilization Sources: Tectraa et al. 2011; Gretzschel et al. 2012. Notes: CAPEX includes engineering and 19 percent value-added tax (VAT). EUR/PE60 x 16.67 = EUR/kg BOD5. ANNEXES WASTEWATER TO ENERGY   36 An overall conclusion is that OPEX could be treatment (CHP) requirements, and treatment reduced by EUR5–6/PE60/y (in Germany or Austria) technology. A typical cost range is between EUR0.005 when introducing mesophilic digestion and biogas and EUR0.050/m3 biogas (EUR0.02–0.5/PE60/y, utilization, as compared to OPEX of extended based on biogas production rates from 10 to 30 L/ aeration. PE60/d, according to figure CS1-2). Micro-aeration or Fe dosage are on the lower end of that range, and The typical median total OPEX of CAS WWTPs in activated carbon adsorption is usually in the middle; the region varies between EUR11 and EUR21/PE60/y and the upper end of that cost range—sometimes even for large WWTPs of more than 100,000 PE60 and for beyond it—is required for cleaning and supplying plants of 10,000–50,000 PE60, respectively (OEWAV biogas into natural biogas supply pipelines. 2012). Hence, the introduction of sludge digestion and biogas utilization reduces total OPEX of the CHP plants concerned by about 15–50 percent. Many suppliers of CHP offer service contracts. Their As can be observed from figure CS1-17, O&M cost cost offers a perfect indicator of the actual OPEX of the new installations is not significant. Nonetheless, involved. ASUE (2011) analyzed sixty-one contracts the relative importance of this item could be different and derived cost curves for co-generation, while in other regions. Supplementary data are presented Geyer and Lengyel (2008) developed similar curves below concerning the respective O&M costs of biogas for co-generation and microturbines. VSA (2012b) treatment and of CHP. and Wacker (2007) cite such costs as well, confirming the more detailed cost information from Geyer and Biogas treatment Lengyel and ASUE. The results are summarized in OPEX of biogas treatment can vary within a wide figure CS1-19. range, depending on untreated biogas quality, Figure CS1-19: OPEX of CHP Source: Geyer and Lengyel 2008; ASUE 2011. ANNEXES WASTEWATER TO ENERGY   37 WERF (2010b) mentions OPEX of US$0.02/kWh The savings in operating costs can be estimated via an (EUR0.015/kWh) for microturbines. Although no average savings of some EUR5–6/PE60/y, as derived in indication of the electric capacity behind this value figure CS1-18. Within a period of fifteen to twenty is provided, the OPEX levels of microturbines in the years, this amounts to EUR75–120/PE60. This amount United States and Europe seem similar. is equivalent to CAPEX required for introducing digestion + biogas utilization for a WWTP size of 1.2.8. Viability of investment in 15,000–25,000 PE60 (see figure CS1-15). Generally, biogas utilization that means that in the region of case study 1, WWTPs Project financial viability is usually evaluated by with a capacity of more than 15,000–25,000 PE60 looking at life cycle cost. The lifespans of the various are financially more viable if they include digesters components differ, with civil works having a longer and biogas utilization. This is also what is currently lifespan (typically thirty years) than electromechanical concluded by various authors (Tectraa et al. 2010, equipment and installations (typically ten to fifteen 2011; Gretzschel et al. 2012; Dohmann and Schröder years). When weighing the relative extent of the various 2011). If the electricity unit cost increases further cost elements, an average lifespan of roughly fifteen to in the future, this financial threshold is expected to twenty years can be assumed. Viability thus means that decline to about 10,000 PE60 (see figure CS1-20). the financial return on the investment within those fifteen to twenty years is larger than the cost. Figure CS1-20: Future viability of mesophilic digestion versus extended aeration Mesophilic digestion Extended Aeration Source: Dohmann and Schröder 2011. ANNEXES WASTEWATER TO ENERGY   38 1.3. Conclusions for CAS + sludge digestion in unit electricity cost is lower, as well. Thus, the financial EAP countries value of the generated electricity is also lower. In EAP, the characteristics and quantities of sludge that The reduced CAPEX will consequently have to be can be expected from a CAS plant may be different balanced with reduced OPEX savings. Therefore, than in Europe. This change will become particularly the amortization periods of investment into sludge evident where wastewater dilution is high and where digestion and biogas projects in EAP are similar to or septic tanks continue to be used in large numbers. somewhat longer than in Europe. These effects imply a shift toward a higher percentage of WAS, compared to PS. This, combined with In terms of electricity coverage, a CAS plant + digester reduced total sludge quantities, will inevitably lead in EAP can usually be expected to be capable of to reduced biogas production, and thus to a reduced producing only part of the electricity required for the electricity generation potential. operation of that facility. Plants designed for carbon removal may achieve a higher coverage of up to 50– CAPEX of digesters may be lower in EAP than in 75 percent, whereas plants designed for nitrification/ Europe; civil works in particular may be cheaper. denitrification are expected to achieve a lower coverage Given that about 50 percent of total CAPEX is for of just 30–50 percent. electro-mechanical equipment, which is mostly imported and hence not much cheaper, the overall Success stories in EAP CAPEX reduction for a digester system in EAP as Successful examples of CAS + sludge digestion projects compared to Europe could be around 20 percent. are so far mainly limited to large cities and WWTPs. On the other hand, OPEX savings will be lower in For instance, Cao (2011) has reported on the Ulu EAP in absolute terms. Sludge disposal cost is usually Pandan Water Reclamation Plant in Singapore, which very low as compared to Europe, and in most cases the is based on CAS and achieving 63 percent N removal. ANNEXES WASTEWATER TO ENERGY   39 It is currently operating close to its design capacity of the digesters (Jiang et al. 2013). About 50 percent 361 MLD. The influent is normal strength wastewater of the actual biogas production of about 33,000 with TSS (total suspended solids) of about 300 mg/L m3/d is used for sludge drying and digester heating, and COD (chemical oxygen demand) of about 600 while the remainder is currently just flared (private mg/L. This plant generates 22,424 m3 biogas/d, communication 2014). equivalent to about 12.2 L biogas/PE60. This is on the An interesting case study was done for four WWTPs in lower end of the biogas expectation range indicated Chengdu, China, by Murray and others (2008). They in figure CS1-2, but it still shows the digestion analyzed a wide range of sludge treatment options, process works properly in EAP. Cao reported that including dewatering, lime addition, mesophilic through optimization measures, the plant is expected anaerobic digestion, heat drying, and incineration, to increase the present electricity coverage from 34 and various combinations thereof. Based on a life cycle percent to about 55 percent of its total needs. assessment, they concluded that “anaerobic digestion An interesting analysis of sludge management practices is generally the optimal treatment.” Even though the and options in China was prepared by ADB (2012). It authors based their analysis on the Chengdu case study, found that more than 80 percent of sludge is disposed they concluded the outcome should be representative of at landfills without prior stabilization. The official for many other WWTPs worldwide. policy tries to change this practice through improved In Vietnam, Yen So WWTP in Hanoi (200 MLD) volume reduction, stabilization, and safe disposal. Still, features sludge digestion, as well. It apparently worked for the time being, the number of anaerobic sludge well during commissioning and was producing biogas. digesters with biogas production/utilization is small. Digester operation was then stopped, however, due A few examples stand out, such as sludge digesters at to unresolved contractual issues between the private Sanjintan WWTP in Wuhan (Hubei Province), or operator and the project owner. Sibao WWTP in Huangzhou (Zheijang Province). Some of the few existing sludge digesters in East Asia Another example is the Bailonggang Municipal are suffering from problems associated with foaming WWTP in Shanghai, China, with a design capacity and deposits in the digester. It is assumed that this is of 2,000 MLD and 4.3 million PE, featuring eight typically the consequence of inadequate preliminary large, egg-shaped digesters with a total volume of treatment (screening, grit removal). To avoid such 99,200 m3; when it was successfully commissioned problems in the future, it may be worthwhile to in 2011, it was said to be the largest such sludge consider DBO (design-build-operate) contracts treatment facility in the world. This plant is receiving for WWTPs with sludge digesters. Under such a thin wastewater with just about 100 mg BOD5/L contract, experienced private companies in charge of and 100 mg TSS/L, the latter being overwhelmingly the complete WWTP design and operation would be inorganic (Enviro-Consult and Sogreah China 2007). expected to install proper components at all stages. As The biogas production is working well, with a yield case study 2, below, demonstrates, such an approach of 0.82 m3/kgVSSadded. Operational problems have can work well. been observed related to foaming and deposits inside ANNEXES WASTEWATER TO ENERGY   40 ANNEXES WASTEWATER TO ENERGY   41 ANAEROBIC SLUDGE DIGESTERS: TECHNICAL SUMMARY Table CS1-4: Anaerobic digesters: General design parameters and key characteristics DIGESTER Retention time: Fifteen to twenty-five days for large–small WWTPs at mesophilic DESIGN temperature 30–38°C (VSA 2012a; WERF 2010c; DWA 1996). Note that some guidebooks recommend shorter retention times down to about ten days (for instance, Metcalf and Eddy 2003), but this only serves to cover the most intensive period of biogas production; it is insufficient to stabilize the sludge properly under all operation conditions. With unheated digesters, the annual fluctuation of air temperature and its impact on digestion/biogas production should be analyzed carefully. In East Asia and other regions, biogas production can be successful even when temperature minimums go down briefly to about 25°C. Figure CS1-21 presents the digestion time requirements as a function of temperature. This allows for the necessary digestion time in case of non-mesophilic conditions. Figure CS1-21: Anaerobic digestion time required to achieve “stabilized sludge” in function of temperature Source: DWA 2003; Bauerfeld et al. 2009. Bauerfeld and others (2009) also highlight that unheated digesters are a research focus in Vietnam at Ho Chi Minh City’s universities (Nong Lam University; University of Technology), which are considering this as a meaningful technology, particularly for large WWTPs. ANNEXES WASTEWATER TO ENERGY   42 Given the long retention time of sludge in digesters, designers should avoid calculating sludge quantities for one-day peaks and apply those quantities to digester design. Rather, two-week or one-month peak values are adequate for digester design, since the typical twenty-day criterion already includes a safety margin of about five days against brief sludge peaks (for instance, caused by stormwater peaks, feeding only on five days per seven-day week, or unreliability of sludge production forecasts). Most of the biological degradation takes place within the first ten days (see figure CS1-25). Solids loading rate: (kgVS/m3/d) is sometimes recommended as a design parameter. However, it never prevails for digesters fed with primary sludge and WAS. Only in cases of co-digestion of additional organic waste does this parameter deserve more attention (see case study 5). Typical maximum permitted loadings are 1.5–4.5 kgVS/m3/d for small– large WWTPs at mesophilic temperature 30–38°C (VSA 2012a; Metcalf and Eddy 2003; Schmelz 2000; DWA 1996). DIGESTER Figure CS1-22: Typical shapes of anaerobic SHAPE AND CONSTRUCTION Shallow cylindrical Cylindrical Egg-shaped Source: Authors. Temperature control and adequate mixing are key to achieving adequate performance in anaerobic digesters. Usually, shallow digesters result in poor mixing, which in turn fosters grit settlement and accumulation of scum. On the other hand, the large surface area of cylindrical shapes provides for less intensive biogas emissions per unit area, which reduces the risk of scum formation. As inert material accumulates in the digester, the active volume is reduced, thereby reducing its performance. Reduced anaerobic digestion performance affects biogas production and subsequent energy generation. Egg-shaped digesters have the lowest outer surface to volume ratio and thus the minimum energy losses, which matters particularly in cold climate regions. They also feature good mixing properties, but they are more difficult to construct and usually require higher CAPEX than cylindrical shapes. ANNEXES WASTEWATER TO ENERGY   43 Construction is mostly done in reinforced concrete, but steel construction is sometimes used. The latter usually requires somewhat less CAPEX but is considered riskier in terms of corrosion and bursting. The state-of-the-art types of digester covers include floating, fixed, and membrane. The most common application is a fixed cover that provides free space between the digester roof and the liquid surface. The membrane cover is a relatively new development that combines its cover function with a gas holder on top of the digester. This may be attractive where land is in short supply and can offer cost advantages. Figure CS1-23 presents a combination of cylindrical and egg-shaped digester. Figure CS1-23: Cylindrical and egg-shaped digester Source: Fritzens WWTP, Austria. DIGESTER Mixing serves various purposes: (a) bringing substrate in contact with active biomass; MIXING (b) avoidance of sludge stratification and scum layer; (c) improved gas stripping; and (d) avoidance of dead volume. The following are the most widespread systems for digester mixing: • Mechanical low-speed mixers inside the digester • Biogas injection • Mechanical pumping with internal draft tube • Mechanical pumping with external pumps Different mixing systems have different efficiencies. No systematic comparison exists, however, and specific analysis is only sometimes done. For instance, Jenicek (2012) reports that replacing biogas injection with mechanical low-speed mixers inside Prague’s ANNEXES WASTEWATER TO ENERGY   44 digesters increased actual sludge retention time by 25 percent. Careful analysis is hence always recommended before deciding on a specific mixing system. DWA (1996) recommends that mixing systems have a daily mixing capacity of about 12–15 times the digester volume. Higher values are not deemed necessary as, for instance, recommended by Metcalf and Eddy (2003), who define a turnover time of tank contents of 20–30 minutes, which would imply a daily mixing capacity of 48–72 times the digester volume. In general, mixing is not required twenty-four hours a day, but it should always be done during feeding of raw sludge. DIGESTER To promote fast and intensive contact of fresh sludge with active biomass in the digester, FEEDING it is common practice to mix digester sludge with raw sludge and to feed the mixture into the digester. Some authors recommend mixing ratios of 2–4:1 between digester sludge and raw sludge, with a ratio of 1:1 considered the minimum requirement. Generally, to minimize sludge sedimentation in pipes it is recommendable to consider minimum values, for instance, DN ≥ 100 mm and v ≥ 0.8 m/s (DWA 1996). The sludge mixture usually passes through a heat exchanger before being injected into the digester. DIGESTER Digester heating serves two purposes: HEATING • Heating of (cold) raw sludge to mesophilic temperature • Compensation of heat losses through the walls, floor, and roof of the digester It is most common to employ external heat exchangers. In some cases internal heating devices inside the digester are also in use, but the maintenance of these units can prove problematic; therefore, internal heating is not a standard solution. The two most common types of external heat exchangers are heated tube-in-tube and spiral plate heat exchangers. In developed countries with cold winters, the thermal energy produced from CHP (about 50 percent of the biogas’s calorific value) is mostly sufficient to enable digester heating without additional external fuels. Only very cold periods may require additional heat sources. Consequently, in warm climates heating never becomes a real OPEX issue. It has even proved possible in warm climate countries to eliminate digester heating and digester insulation (see case study 2). In this case, it is important to ensure a minimum temperature in the digester, ideally >28–30°C. Below that value the biological activity reduces drastically (see figure CS1-24). Then, if lower temperatures persist for prolonged ANNEXES WASTEWATER TO ENERGY   45 periods, neither biogas production nor sludge stabilization will be satisfactory. The biological activity reduces at >40–45°C as well, but overheating is usually not an issue. Figure CS1-24: Relative digestion rate as a function of temperature Source: Haandel and Lettinga 1994. Table CS1-5: Biogas systems combined with anaerobic digesters: General design parameters and key characteristics BIOGAS DESIGN Volatile solids (VS) concentration of sludge: (a) primary sludge: average 75 percent (65–85 percent); (b) very high-loaded secondary sludge (sludge age ≈ 1 day): average 72 percent (65–80 percent); (c) secondary sludge of C elimination only (sludge age ≈ 5–10 days): average 70 percent (65–75 percent); (d) secondary sludge of N elimination (sludge age ≈ 10–15 days): average 68 percent (62–75 percent); (e) secondary sludge of extended aeration facility (sludge age > 20 days): average 65 percent (60–70 percent); (f) trickling filter sludge: average 70 percent (65–75 percent); (g) UASB sludge: average 55 percent (50–60 percent) (Bauerfeld et al. 2009; DWA 2003a; Buchauer 1996). Volatile solids destruction: (a) primary sludge: 55–60 percent; (b) secondary sludge: 30– 40 percent; (c) mixture of primary and secondary sludge: 40–50 percent (Kapp 1984; Roediger et al. 1990). ANNEXES WASTEWATER TO ENERGY   46 Figure CS1-25: Volatile solids (VS) destruction over time primary sludge prim. + sec. sludge secondary sludge Source: Kapp 1984. Note: Undigested primary or secondary sludge rarely has a VS content greater than 70 percent. The maximum VS destruction in a conventional mesophilic digester is about 50– 60 percent, as indicated in figure CS1-25. Hence, digested sludge under these conditions usually does not achieve a VS content below 50 percent. Typically, digested sludge has a VS content between 50 and 55 percent. A VS content above 60 percent indicates a malfunction whose causes should be investigated. Gas production: (a) primary sludge: 900–1000 L/kgVS destroyed; (b) secondary sludge: 700–800 L/kgVS destroyed; (c) mixture of primary and secondary sludge: 800–900 L/ kgVS destroyed (Kapp 1984; Roediger et al. 1990). Typical biogas characteristics: (a) methane CH4: 60–70 percent; (b) calorific value: 6.0–7.0 kWh/Nm3. BIOGAS Any biogas system should always contain the following: TREATMENT • A foam trap (for removal of foam and particles from gas) • Condensate removal (for removal of condensing gas humidity) If necessary, one or several of the following technologies might also be used for further biogas cleaning. The main substances that require improved treatment usually are H2S and siloxanes (Frey 2012; DWA 2011; WERF 2010b): ANNEXES WASTEWATER TO ENERGY   47 • Biological oxidation of H2S in digester (= micro-aeration of digester) • Precipitation of H2S by Fe dosage into digester • Adsorption reactors (Fe, activated carbon) • Wet scrubbers The first two options are the most common and economical solutions for H2S removal in medium- to large-sized plants, and sometimes the processes are combined. In smaller facilities, adsorption in Fe filters prove easier to operate and more economical. Activated carbon is nowadays the standard solution for siloxane removal. Wet scrubbers are only applied in special cases, since this is typically the most costly, albeit one of the most efficient, technologies. Table CS1-6 summarizes typical features of biogas treatment. Table CS1-6: Biogas treatment: typical parameters, requirements, implications Parameter Typical value Typical require- Implications in biogas* ment for CHP** Water n.a. <60–85 percent Corrosion humidity H2S (mg/m3) 434 <200–450 Corrosion, scaling, shorter intervals for percent lubrication oil and spark plug changes Total siloxanes 15 <2–6 Scaling and wear and tear in combus- tion chamber and catalyzer, shorter maintenance intervals Total chlorine 434 <100 Problems with corrosion and catalyzer and fluorine **According to VSA (2012b), DWA (2011), and information from various manufacturers. n.a. = not available. Micro-aeration: The air supplied into the digester should be introduced at several points just above the liquid level. This is where the desulphurizing microorganisms that utilize the oxygen have optimal conditions for their proliferation (Prochazka et al. 2013; Jäkel 2007). The required air quantity depends on the amount of H2S that should be oxidized. Figure CS1-26 presents the necessary dosages for aeration. Mercato-Romain and others (2013) also report on the low air supplies needed for efficient H2S removal. ANNEXES WASTEWATER TO ENERGY   48 Figure CS1-26: Minimum air dosage for H2S removal by micro-aeration in case of gas with 2000 ppm H2S Air dosage (liters / m3 gas) Biogas flow rate Source: Jäkel 2007. Precipitation of H2S by Fe dosage into digester: The required Fe dosage rate can be determined in two ways: by stoichiometric considerations or by introducing Fe so as to achieve a certain Fe/DS ratio in the digester. • Stoichiometrically needed are 1.75 g Fe/g S-eliminated. By using a typical overdosage of 100 percent (Procházka et al. 2013; DWA 1996), one can calculate the necessary dosage of Fe. • For safeguarding a typical H2S <200 ppm in the treated gas, a concentration of 30–40 g Fe/kg DS is usually necessary (Ries et al. 1992); see figure CS1-27. Figure CS1-27: H2S concentration in biogas as function of iron content in digester sludge, results from 21 WWTPs Source: Ries et al. 1992. ANNEXES WASTEWATER TO ENERGY   49 Adsorption to activated carbon (AC): Activated carbon cannot be designated for the removal of a specific substance alone. Rather, anything that can be adsorbed will be adsorbed to AC. Prior testing is hence indispensable if its exact cost needs to be calculated. Since H2S is frequently the dominant compound that is being adsorbed, as a rough first guide one can estimate the needed AC quantities according to a typical adsorption capacity of 0.2–0.5 kg S per kg AC. GAS HOLDER The gas holder is meant to balance biogas production with biogas utilization. The gas holder should not be confused with a “storage tank”; rather, it serves as a small “balancing tank.” The daily fluctuation of biogas production depends primarily on the digester’s sludge- feeding regime. If feeding is done only once a day, biogas production peaks after about two to three hours. For balancing this single peak, a holder volume of about 15–20 percent of daily production is usually sufficient (VSA 2010). Based on that background, gas holders are typically designed for a volume of about 10–30 percent of daily biogas production. The lower values prevail for large WWTPs; the higher values prevail for small WWTPs. But larger volumes of 50–80 percent are also sometimes recommended. However, whether the small amount of extra energy production facilitated by larger holder volume does, indeed, compensate for the substantial extra CAPEX is questionable. Technically, low-pressure gas holders are the typical standard today. Two main types dominate the market: • Double membrane biogas holder: This type of gas holder features two membranes: an outer membrane that is more robust and meant to protect against atmospheric conditions (sun, wind, rain, snow) and an inner membrane that contains the biogas. The space between these two membranes is connected to a blower that regulates the gas storage pressure. The higher the pressure, the better is the gas holder’s resistance against wind, but its energy consumption also increases. The typical compromise is a pressure range of some 20–80 mbar. The membrane material is usually a PVC-coated polyester fabric. Double membrane gas holders can be installed as standalones on the ground or on top of the digester. • Single membrane biogas bags: Gas bags are installed inside buildings or in specially constructed containers. Their shape can be easily adjusted to requirements. Usually they operate without pressure. ANNEXES WASTEWATER TO ENERGY   50 Figure CS1-28 shows two typical examples of biogas holders. Figure CS1-28: Double membrane gas holder and biogas bag Sources: Arrudas WWTP, Brazil (double membrane gas holder); Sattler, www.sattler-ag.com (biogas bag). FLARE A flare is an indispensable element of any WWTP with biogas production. It should be considered a safety installation, rather than a standard operating unit. The shorter its operation time, the higher the percentage of biogas utilization, and, thus, the greater the financial benefits. To fulfil its safety function, the flare must be able to cope with the maximum possible hourly biogas production rate. Underdesign is to be avoided. An example is depicted in figure CS1-29. Figure CS1-29: Gas flare Source: Onca WWTP, Brazil. ANNEXES WASTEWATER TO ENERGY   51 BIOGAS Options for biogas utilization include the following (WERF 2012a; Frey 2012; DWA 2011): UTILIZATION • Burner • Fuel cells • Stirling motor • Direct drive engines • Supply into natural biogas supply systems • Utilization as fuel for vehicles • Co-generation • Microturbines Burners only produce thermal energy. This form of energy is usually of substantially lower economic value than electric power, and it is rarely required in East Asia and other warm regions. Fuel cells have not yet reached a technical state of the art that would make their application easily possible anywhere. Even though they achieve relatively high electric efficiencies of >40 percent, they require extensive biogas treatment, they pose increased safety risks (H2), and their CAPEX is still too high for viable application. For instance, the U.S. EPA indicates that fuel cells will only be economically viable if prices decline (U.S. EPA 2006), and WERF (2010b) describes them as one of the most expensive CHP technologies. Microbial fuel cells, a novel development, require further research and have not yet reached the point where they could be recommended for wider application (Rulkens 2007; Kletke et al. 2010; WERF 2011b). The Stirling motor has very low electric efficiency of just slightly above 20 percent. It requires extensive maintenance of pistons and is only available in the market in very small units of about 10 kW. Direct drive engines may be considered for powering large consumers of electric energy, such as major pumps or aeration blowers. Even though it is sometimes stated that direct drive technology is cheaper and more efficient than co-generation or microturbines (Monteith et al. 2006), usually there are no substantial cost differences, and efficiencies are not, indeed, superior. Thus, even though they represent a robust technology, direct drive applications are not widely used. ANNEXES WASTEWATER TO ENERGY   52 Supply into natural biogas systems requires, as a matter of fact, a nearby biogas pipeline, which is frequently not available. Should this condition be met, it still faces a series of additional challenges: (a) technical: increase of CH4 content, very strict criteria for H2S and other quality parameters, pressure increase, need for constant supply; (b) legal; and (c) financial. Utilization as fuel for vehicles has been applied in various locations worldwide. However, when considering this option, several issues have to be resolved: compliance with normed fuel quality requirements, safety aspects, and, not least, the storage issue due to constant supply but only intermittent utilization. Hence, what is typically left as the most economical option for biogas reuse at WWTPs is combined production of electricity and heat (CHP) through co-generation or microturbines. Figure CS1-30 presents typical examples. Figure CS1-30: Co-generation and microturbine Sources: GE Jenbacher (co-generation); Capstone (microturbine). In co-generation, the biogas is burned in a motor, and a generator converts the mechanical energy into electric energy. In a microturbine, the biogas is also burned, but at substantially lower temperatures and at a surplus of oxygen. The burned gases move a turbine, from which a generator then produces the electric energy. The surplus of oxygen results in lower gas emissions (for example, less NOx, and CO), and the turbine shows less wear and tear than a motor. Co-generation is available on the market with electric unit capacities from some 30 kW up to several thousand kW. The maximum electric capacities of microturbines are lower, currently ranging from 30 kW to 600 kW per unit. ANNEXES WASTEWATER TO ENERGY   53 Figure CS1-31 presents typical electric efficiencies of these installations. They reflect average characteristics from products of numerous manufacturers available on the market. In CHP, about 50 percent of the biogas’s calorific value is converted into thermal energy, and about 10 percent is lost. No extra fuel is needed for the installations’ operation. Figure CS1-31: Average electric efficiencies of co-generation and microturbine at full load Sources: Values based on ASUE 2011 (for co-generation); VSA 2012b (for microturbines). Electric efficiencies at partial load, presented in figure CS1-32, only serve to demonstrate the different characteristics of co-generation and microturbines: while co-generation immediately loses efficiency under partial load, microturbines have a near-constant efficiency for awhile, and only then does the efficiency decline. Figure CS1-32: Exemplary electric efficiencies of co-generation and microturbine at partial load Source: VSA 2012b. ANNEXES WASTEWATER TO ENERGY   54 While co-generation is a well-known system, microturbines are not yet generally known, but they could offer substantial advantages in East Asian countries due to their simpler technology and lower maintenance needs. Their requirements for gas cleaning are less strict; after all, this system was originally developed for military purposes with a focus on robustness. Their main disadvantages are slightly higher CAPEX and lower electric efficiency of about 30 percent. The key differences between co-generation and microturbines are summarized in table CS1-7. Table CS1-7: Co-generation versus microturbines Criteria for comparison Co-generation Microturbine Electric efficiency ≈30–40 percent ≈30 percent Electric efficiency reduction at Less marked reduc- Marked reduction partial load tion Thermal efficiency ≈50 percent ≈50 percent Requirement toward gas quality High Medium–high Exhaust emissions to open air Higher emissions Lower emissions Noise emissions High Medium CAPEX Lower Higher OPEX Higher Lower Sources: VSA 2012b; Frey 2012; WERF 2010b; VSA 2010; Geyer and Lengyel 2008. ANNEXES WASTEWATER TO ENERGY   55 CASE STUDY 2: TRICKLING FILTER + SLUDGE DIGESTION ANNEXES WASTEWATER TO ENERGY   56 2.1. BACKGROUND, PROCESS DESCRIPTION 2.1.1. Data sources mainly domestic; hence, there is no interference from Trickling filters (TF) are an “old” technology that has industrial discharges that require special attention. been in use for many decades. In developed countries, The WWTP has been in operation for five years now, TFs used to be more widespread, but nowadays the and all conclusions about it can be based upon reliable total number of TFs has fallen below 10 percent of data. Biogas has so far been collected and metered but the total number of WWTPs (see, for example, figure is not yet utilized. Now that the biogas production I-8 in the main body of the report). The main reason yield is established, a biogas utilization project is in the TF technology was pushed back is its low flexibility pipeline. The plant is owned by ENACAL (Empresa for influencing treatment efficiencies. An operator can Nicaragüense de Acueductos y Alcantarillados), and influence the efficiency of TFs in some ways, but not the first five years of operation from which the data to the extent possible with CAS. have been drawn for this case study were contracted to Biwater International Ltd. Two key aspects stand out: the TF’s simplicity of operation and its low energy consumption. In a 2.1.2. Wastewater management CAS plant, approximately 60 percent of energy The TF technology applied in Managua is a classical consumption is required for aeration of bioreactors concept (see figures CS2-1 and CS2-2): preliminary (compare with case study 1); this energy item is treatment (screen, sand/fat removal) followed by eliminated completely in TFs that use natural air settling in the primary sedimentation tank (PST) draft instead. If this reduction in energy consumption and subsequent pumping onto the TFs. The treated is combined with mesophilic digestion and biogas wastewater and sludge flushed out from the TFs utilization, the goal of an energy-independent WWTP settle in the secondary sedimentation tanks (SSTs). can become realistic. This feature, combined with The effluent is subsequently discharged to the nearby simple operation (which in many environments is a Managua Lake. Part of the treated wastewater is big advantage in itself), is particularly attractive. recirculated to the TF pumping station to ensure a For case study 2, therefore, an example was selected of a certain minimum wetting rate of the filters (10–25 large modern TF plant in a country with a subtropical mm/pass) during periods when the influent flow rate climate: the main WWTP for Nicaragua’s capital, is insufficient for that purpose. The TFs themselves Managua, in Central America. This facility is designed have a filter depth of 5.4 meters and are filled with for about 1.1 million capita and was commissioned cross-flow plastic media with a specific surface area of in 2009. It has several interesting features, such as the 100 m2/m3. use of lamella settling tanks (both for PST and SST) Both PSTs and SSTs are equipped with lamella plates to reduce the plant’s footprint, anaerobic digesters to decrease their footprint. This was mainly done to operating at ambient temperature, and solar sludge minimize the size and cost of a flood protection dam drying. The municipal wastewater in Managua is ANNEXES WASTEWATER TO ENERGY   57 that surrounds the whole plant. The design of the For details of actual wastewater characteristics and lamella settlers is based on conventional criteria. other parameters, see tables CS2-1 and CS2-2. Figure CS2-1: Simplified flow scheme of TF system and sludge treatment at Managua WWTP Source: Authors. ANNEXES WASTEWATER TO ENERGY   58 Figure CS2-2: Aerial view of Managua WWTP Digesters Flare Dewatering Solar Drying Secondary Sedimentation Tanks Thickeners Trickling Filters Primary Sedimentation Tanks Screens Aerated grit and fat removal Source: ENACAL—Managua WWTP. ANNEXES WASTEWATER TO ENERGY   59 2.1.3. Sludge management surrounding the lack of heating. The biogas assessment The sludge from PSTs and SSTs is combined and has finally taken place for the years 2011–12 (FWT thickened by gravity to about 3 percent DS. For 2013b). Consequently, the biogas utilization project improved separation of solid and liquid phases at the is in the pipeline now, and its elements are already high temperatures in Managua, some polymers are defined as follows: added prior to thickening. The sludge is then digested • Foam trap in four closed digesters at ambient temperatures without any heating (the annual average air • Condensate removal temperature is ≈27.5°C, with monthly averages over • Optional: activated carbon filter for H2S and the year between 26° and 29°C). The digesters are siloxane removal the shallow cylindrical type (see table CS1-22), with • Gas holder (1000 m3) a ratio of diameter to sludge depth of about 2.2:1.0. They are constructed out of concrete without thermal • New flare insulation. Each digester is equipped with three • CHP: five microturbines with 200 kW each (total internal draft mixers. Digested sludge is dewatered in electric power = 1,000 kW) belt filter presses to about 30 percent DS and further dried in a solar drying facility to >90 percent DS. The • Pipes, valves, and so on, as required final product goes into agriculture/sanitary landfill. For details of actual biogas characteristics, see table Meyer-Scharenberg and Pöppke (2010) describe the CS2-2. For details on general design parameters and sludge treatment in more detail. key characteristics of biogas systems combined with A project is in the pipeline to install additional anaerobic digesters, see table CS1-5. mechanical thickeners to further increase the DS 2.2. Analysis content of the sludge prior to digestion, up to about 6 percent DS. This will halve the sludge volume to 2.2.1. Wastewater influent, effluent, and other digest and double the retention time in the digesters. parameters of interest The data presented in tables CS2-1 and CS2-2 are For details of actual sludge characteristics, see table taken from an assessment that took place for the two CS2-2. For details on general design parameters and years between January 2011 and November 2012 key characteristics of anaerobic digesters, see table (FWT 2013b). CS1-4. 2.1.4. Energy management The generated biogas is currently only flared. This was done in previous years to collect actual operation data regarding biogas generation in the unheated digesters because the forecast of biogas production was deemed unreliable in the design stage due to the uncertainties ANNEXES WASTEWATER TO ENERGY   60 Table CS2-1: Actual influent and effluent data of Managua WWTP, average data from 1/2011 to 11/2012 Managua       WWTP Number of WWTPs   1 Pop. equivalents avg. actual PE60 447,000   max. actual PE60 606,000 WASTEWATER QUANTITY     Specific wastewater production m /PE60/y 3 82     L/PE60/d 225 WASTEWATER QUALITY     COD Influent mg/L 505   Effluent mg/L 101   Elimination % 80 BOD5 Influent mg/L 248   Effluent mg/L 28   Elimination % 89 TSS Influent mg/L 259   Effluent mg/L 31   Elimination % 88 Ntotal Influent mg/L 27.6   Effluent mg/L 17.5   Elimination % 37 NH4–N Effluent mg/L n.a. NO3–N Effluent mg/L n.a. Ptotal Influent mg/L 3.7   Effluent mg/L 1.7   Elimination % 54 Source: FWT 2013b. Notes: 1 cap = 46.5 g BOD5/d in Nicaragua; 1 PE60 = 1.29 PE46.5 = 1.29 cap in Nicaragua; m3/PE60/y x 1.29 = m3/ cap/y in Nicaragua; L/PE60/d x 1.29 = L/cap/d in Nicaragua. n.a = not available. ANNEXES WASTEWATER TO ENERGY   61 Table CS2-2: Key characteristics of Managua WWTP       Actual Design GENERAL         Daily flow rate average m3/d 100,750 182,563   95 percentile 142,572 297,302 BOD5 load average kg/d 26,846 50,663 COD load average kg/d 55,469 101,326 TSS load average kg/d 28,178 30,398 BOD5 effluent average mg/L 28 —   90 percentile 42 90 COD effluent average mg/L 101 —   90 percentile 147 180 TSS effluent average mg/L 31 —   90 percentile 50 80 WASTEWATER TRAIN         PST retention time average h 1.37 —   5 percentile 0.79 0.50 TF volumetric load average gBOD5/m3/d 0.43 1.46 FST surface load average m/h 0.70 —   5 percentile 0.93 1.48 SLUDGE TRAIN         Thickened sludge average m3/d 800 795   kgDS/d 22,600 39,735 VS of raw sludge average % 67 75 Digester retention time average d 22.7 20.0   5 percentile 15.0 VS destruction in digester average % 49 50 VS of digested sludge average % 51 60 BIOGAS TRAIN         Biogas production average m3/d 7,159 13,038   95 percentile 9,466 Specific biogas product. average L / kgVSdestroyed 950 875 Calorific value biogas average kWh/m3 8.4 — H2S average ppm 130 —   max. ppm 210 — Air temperature average °C 29.5 — Note: Most of the above actual data refer to the period January 2011 to November 2012. The biogas data refer to January 2012 to October 2012. ANNEXES WASTEWATER TO ENERGY   62 2.2.2. Biogas production and potential for fluctuations need not be expected, and “normal,” energy generation undisturbed biogas production should be about 5–10 percent higher than the metered average during this Biogas production period. Hence, if there had been no problems with Figure CS2-3 shows biogas production in the digesters influent pumping, the documented average specific during the first ten months of 2012, which is mostly biogas production rate of 950 L/kgVSdestroyed over the constant. Just during two periods, in June and October whole period would have easily reached about 1000 respectively, is there a pronounced reduction in biogas L/kgVSdestroyed. This is on the upper end of the biogas production to almost zero. These reductions were production range indicated in table CS1-5. Thus, case caused by strongly reduced influent flow rates to the study 2 demonstrates that even unheated digesters can WWTP due to problems with the influent pumping be very efficient in warm climate environments. station. Under normal operating conditions, these Figure CS2-3: Daily biogas production at Managua WWTP Source: FWT 2013b. Potential for energy generation the proposed five microturbines with 200 kW each is assumed as 30 percent (see figure CS1-31). The calorific value of the biogas was determined by the Laboratorio de Geoquímica Geotérmica of the The potential for electricity generation is calculated and Ministerio de Energía y Minas, Managua, Nicaragua. It compared to the actual total electricity consumption of found an average value of 8.4 kWh/m3. Since this value Managua WWTP (figure CS2-4). The forecast average is surprisingly high, a more conventional value of 6.5 electricity production equals 426 MWh/month, which kWh/m was chosen for the analysis presented below 3 equals 5,110 MWh/year. Additionally, a potential for (see table CS1-5). Thus, it is possible that the actual thermal heat generation from CHP was calculated as potential of energy generation is as much as about 25 781 MWh/month, which equals 9,370 MWh/year percent higher than calculated. The electric efficiency of (FWT 2013b). ANNEXES WASTEWATER TO ENERGY   63 Figure CS2-4: Power generation potential of actual biogas production, compared to actual electricity consumption at Managua WWTP Source: FWT 2013b. Note: No electricity consumption data were available for the months of August and September 2012. From the results, the following conclusions can be • Only during two months (June and October) when derived: there were problems with the influent pumping stations was the consumption slightly higher than • Typically, the biogas’ electricity production potential the electricity potential. is 10–20 percent higher than the actual electricity consumption, even though in this case study less • The average electricity consumption equals 9–10 efficient microturbines are planned instead of more kWh/PE60/year at Managua WWTP. This is efficient co-generation. substantially lower than comparable values for CAS. On an annual basis, a WWTP based on • Microturbines are preferred over co-generation in TF technology and sludge digestion like that in this case for the following reasons: high efficiency at Managua can be operated with a positive electric partial load, somewhat lower requirements toward energy balance. The electricity supply from the gas quality, less emissions to air, less noise, longer public grid serves only as “safety net” for brief peak maintenance service intervals, and, most important, periods and during operational problems. lower OPEX (see also table CS1-7). ANNEXES WASTEWATER TO ENERGY   64 Table CS2-3: Biogas and power generation potential of TF + digester at Managua WWTP Retention time in PST   (h)   1.4 Biogas production   - N elimination (L/PE60/d) — - C elimination (L/PE60/d) 16.0 Electric efficiency CHP (%) 30 Thermal efficiency CHP (%) 55 Calorific value of biogas (kWh/m ) 3 6.5 Power generation   - N elimination (kWh/PE60/year) — - C elimination (kWh/PE60/year) 11.4 Thermal energy generation   - N elimination (kWh/PE60/year) — - C elimination (kWh/PE60/year) 20.9 Total energy generation   - N elimination (kWh/PE60/year) — - C elimination (kWh/PE60/year) 32.3 Source: Authors’ calculation. Note: L/PE60/d x 16.67 = L/kg BOD5/d; kWh/PE60/y x 16.67 = kWh/kg BOD5/y. When comparing the above results with international body of the report show European and U.S. energy experiences, the emerging picture might be confusing consumption results for TFs of 25 kWh/PE60/y and at first glance, and hence requires a closer look: 30–40 kWh/PE60/y on average, respectively. These average energy consumption values are still about • A basic “problem” is that most documented energy 25 percent lower than for CAS, but the differences data on TFs originate from facilities in developed from CAS are not very significant. countries, and very few are published about TFs in warm climate zones. The TFs in developed countries • Second, there are more efforts underway to optimize are adjusted to their local requirements, which usually CAS plants from the energy utilization point of call for C+N elimination in (at least seasonally) view, whereas TF plants are rarely subject to the cold climates. This usually implies additional same efforts. As TFs are classified as “low-energy electric power consumption for forced ventilation; technology,” operators usually think their energy- for additional denitrification stages; for higher saving potential is low. Only a few large-scale recirculation rates; or for separate installations for examples of energy optimization efforts at TFs are enhanced P removal. Figures I-6 and I-7 in the main reported in the literature. If optimization is carried ANNEXES WASTEWATER TO ENERGY   65 out, nonetheless, the gap between TF and CAS the public grid is not a dominant issue, but it is not widens. For instance, one recent study by Witzgall irrelevant, either. and others (2013) found that an optimized TF Nicaragua has no legal restrictions that would impede for C+N elimination (in the United States) only the supply of electric energy into the public grid. consumes half the electric power of a comparable Since this approach would be novel to the country, CAS system. however, no clear price has yet been set. This price • Third, case study 2 demonstrates that energy would depend on the result of negotiations between numbers from Europe and the United States should the owner of the WWTP (ENACAL) and the public not be uncritically applied to TFs in warm climates. supplier of electric energy (Gas Natural). Rather, when there is a requirement for C elimination ENACAL currently pays a low tariff of US$0.08/kWh only in warm climates, TFs can apparently operate for electricity purchased from the public grid. The with an electricity consumption of <10 kWh/PE60/y, conventional commercial tariff stands at US$0.32/ including sludge digestion and even sludge drying. kWh. This not only lowers the electricity bill in general, but it offers an even more attractive scenario: these FWT (2013b) therefore based all financial assessments facilities are indeed capable of fully covering their of investment in biogas utilization on three different own power needs through autogeneration of electric electricity cost scenarios: US$0.08, US$0.13, and energy from sludge at the WWTP. US$0.3/kWh, respectively. For results, see section • Fourth, it is important to understand that a 2.2.8. nitrification requirement does not fundamentally 2.2.5. GHG reduction and CDM co-financing change the electricity requirement of TFs. General Nitrification only implies a need for larger TFs, but the pumping head always remains the same. For general information on CDM co-financing and the price of carbon credits, see section 1.2.5. 2.2.3. Operation capacity needs, biogas safety For digester operation the same principles always Specifics of case study 2 apply, independent of digester location. All the As pointed out in section 1.2.5, due to the relevant operation and safety issues for digesters have “additionality” criterion, in the new construction of a already been discussed in much detail in case study 1 digester completed with CHP the energy production (see section 1.2.3). component under CDM should be considered, but not methane elimination, since methane is created 2.2.4. Institutional aspects, energy costs and eliminated by the project. The clear preference in Managua is for utilization of generated electricity onsite. But the numbers indicate Nicaragua’s specific energy mixture causes the GHG there could be small temporary power surpluses that emissions from electricity generation shown in table cannot be economically balanced in gas holders or CS2-4. batteries. Hence, the question of electricity supply to ANNEXES WASTEWATER TO ENERGY   66 With a forecast annual electricity production potential reduction is 2,350 tons CO2e/year for case study 2 at from biogas of 5,110 MWh/year, the potential GHG present, based on Nicaragua’s average energy mixture. Table CS2-4: GHG emissions per kWh for electricity generation in Nicaragua, as compared to the world REGION 1990 2010 gCO/kWh * gCO/kWh * Nicaragua 345 460 World 586 565 Source: IEA 2012. Note: The table shows CO2 emissions from fossil fuels consumed for electricity generation, in both electricity-only and combined heat and power plants, divided by output of electricity generated from fossil fuels, nuclear, hydro (excluding pumped storage), geothermal, solar, wind, tide, wave, ocean, and biofuels. Both main activity producers and autoproducers have been included in the calculation. 2.2.6. CAPEX structure CAPEX requirements for a biogas utilization project financial bids from potential suppliers. The outcome at Managua WWTP have been estimated by FWT (table CS2-5) can be considered quite realistic and (2013b). All components were based on actual reflects the local specifics prevailing in Nicaragua. Table CS2-5: CAPEX for biogas utilization project at Managua WWTP   CAPEX    US$ EUR Biogas pretreatment, gas holder, flare 877,500 650,000 CHP with microturbines 1,431,000 1.060,000 SUBTOTAL 2,308,500 1,710,000 Contingencies 10 percent 230,850 171,000 Consulting 10 percent 230,850 171,000 Price adjustment 5 percent 115,425 85,500 TOTAL 2,885,625 2,137,500 Source: FWT 2013b. The additional cost for an optional biogas treatment assumes an increase in influent load (and thus in biogas to remove H2S and siloxanes was estimated at US$1 production) of about 55 percent due to large increases million. For the time being, however, this was not in the sewer connection rate between 2013 and 2025. considered necessary, given the unproblematic results The influent pollution load to Managua WWTP will of the gas analysis available to date. thus increase to ≈700,000 PE60 on annual average and to ≈950,000 PE60 during peak periods. Note that the above CAPEX will be sufficient to cover the biogas project needs up to the year 2025. It ANNEXES WASTEWATER TO ENERGY   67 2.2.7. OPEX structure OPEX requirements for a biogas utilization project at see table CS2-6. All components were adjusted to the Managua WWTP were estimated by FWT (2013b); local specifics prevailing in Nicaragua. Table CS2-6: OPEX for biogas utilization project at Managua WWTP OPEX OPEX 2012 OPEX 2025   OPEX 2012   2025     US$/y EUR/y US$/y EUR/y Biogas pretreatment, gas holder, flare 37,800 28,000 37,800 28,000 CHP with microturbines 90,450 67,000 90,450 67,000 SUBTOTAL 128,250 95,000 128,250 95,000 Saving in case of electr. tariff = 1.86 C$/ kWh –278,100 –206,000 –483,300 –358,000 TOTAL in case of electr. tariff = 1.86 C$/ kWh –149,850 –111,000 –355,050 –263,000 Source: FWT 2013b. Consequently, if electricity unit cost remains US$150,000/y (EUR111,000/y) in 2012 and by unchanged at C$1.86/kWh, OPEX will reduce by US$355,000/y (EUR263,000/y) in 2025. 2.2.8. Viability of investment in biogas utilization Table CS2-7 summarizes several cost indicators from case study 2. Table CS2-7: Cost indicators for biogas utilization project at Managua WWTP     2012 2025 Average influent load PE60,avg 447,000 700,000 Peak influent load PE60,max 606,000 950,000 CAPEX US$ 2,885,625 — OPEX US$/y –149,850 –355,050 specific CAPEX US$/PE60,avg 6.46 4.12   US$/PE60,max 4.76 3.04 specific OPEX US$/PE60,avg –0.34 –0.51   US$/PE60,max –0.25 –0.37 Source: FWT 2013b and authors’ calculation. Note: 1 PE60 = 1.29 PE46.5 = 1.29 cap in Nicaragua; US$/PE60 x 16.67 = US$/kg BOD5. ANNEXES WASTEWATER TO ENERGY   68 The specific CAPEX related to the future peak these processes do not depend on location, but on influent pollution in 2025 is thus US$3.04/PE60,max the environmental conditions inside the digester. The (EUR2.25/PE60,max). The specific CAPEX related to reported biogas yield from the digesters in case study the future average influent pollution in 2025 is thus 2 confirms that unheated digesters in warm climate US$4.12/PE60,avg (EUR3.05/PE60,avg). countries can achieve comparable results to heated digesters in cold climates. Indirectly, this result also Relating the minimum OPEX savings to average demonstrates that the biogas potentials from CAS annual influent pollution leads to savings of US$0.34/ sludge and TF sludge are rather similar. PE60,avg (EUR0.25/PE60,avg) and US$0.51/PE60,avg (EUR0.38/PE60,avg) in 2012 and 2025, respectively. Case study 2 also confirms the feasibility of unheated It was concluded in FWT (2013b) that the project is sludge digesters in warm climates. financially viable. Total annual cost (CAPEX + OPEX) Compared to Nicaragua, the characteristics and of Managua WWTP is expected to decrease with the quantities of sludge that could be expected from a TF new biogas utilization project. plant in EAP are different. This change will become particularly evident where wastewater dilution is high 2.3. Conclusions for TF + sludge digestion in and where septic tanks continue to be used in large East Asian countries numbers. In EAP, reduced quantities of primary sludge The key characteristics of sludge digestion, VS from PSTs and reduced sludge quantities in general destruction, and biogas production, as already are expected. This, in turn, means a shift toward more described in case study 1, remain unchanged, as ANNEXES WASTEWATER TO ENERGY   69 TF sludge and less PS, resulting in reduced biogas Consequently, the amortization periods of investment potential. into sludge digestion and biogas projects in EAP need case-specific analyses. Another possible change relates to wastewater treatment requirements. The TF in Nicaragua is designed for Success stories in EAP carbon removal only. If in EAP additional nitrification No TF plants in EAP with publicly accessible operation is required, CAPEX, of course, will increase due to a data are known to the authors of this technical note. need for larger/more TF reactor volume, but OPEX will remain almost unchanged, since the same flow Since CAPEX levels in Nicaragua and EAP could be rate continues to be pumped to the same water head, of similar magnitude, similar financial conclusions and pumping is the dominant energy consumer. Only may be expected. The cost assessment, however, is also if denitrification is required, the energy requirements influenced by local wastewater specifics. Thus, in EAP, for additional mixing and recirculation will increase where the combined effects of wastewater dilution overall energy consumption. This might eventually and the use of large numbers of septic tanks prevail, lead to a situation where, with nitrification and these projects can be expected to be less financially denitrification, less than 100 percent of electricity appealing than they are in Nicaragua. Such tentative consumption can be covered from biogas. assessments can lead to wrong conclusions, as shown by the application example of the assessment tool that CAPEX could be similar in many cities in EAP and was explained in the main body of this report. A minor in Nicaragua. OPEX savings may be less in situations difference in influent characteristics can trigger a where wastewater dilution is high and sludge completely different outcome with respect to financial production is consequently lower, but they may also viability. Therefore, a sound assessment of individual be higher where no dilution prevails and power unit energy projects is highly recommended, particularly cost is higher than in Nicaragua (US$0.08/kWh). In where electricity unit cost is high, where extra organic several EAP countries, electricity unit cost is higher feedstock is available, and where safe sludge disposal than in Nicaragua. Thus, the financial value of the becomes a matter of concern. generated electricity is higher in these places as well. ANNEXES WASTEWATER TO ENERGY   70 ANNEXES WASTEWATER TO ENERGY   71 CASE STUDY 3: UASB ANNEXES WASTEWATER TO ENERGY   72 3.1. BACKGROUND, PROCESS DESCRIPTION 3.1.1. General background, data sources stage. This stage can use any aerobic technology, but Upflow anaerobic sludge blanket (UASB) technology the most common choice is ponds or trickling filter. was originally developed in the Netherlands. According CAS is only seldom used for that purpose. to Haandel and Lettinga (1994),“The steep increase in Case study 3 is based on a comprehensive analysis of energy prices in the 1970s reduced the attractiveness twenty-two UASB systems (FWT 2013c) in Minas of aerobic treatment systems and intensified research Gerais, Brazil, that was conducted in the course of efforts towards the development of systems with lower the project, “Despoluição da Bacia Hidrográfica energy consumption.” UASB reactors work well if a do Rio Paraopeba.” This project is co-financed by certain minimum wastewater temperature, ideally COPASA and KfW. All analyzed UASB systems are >20°C, is provided. Hence, the system is usually operated by COPASA (Companhia de Saneamento applied to industries with warm wastewater, and de Minas Gerais). UASB reactors for municipal wastewater are de facto used exclusively in countries with warm climates. The 3.1.2. Wastewater management first large-scale municipal UASBs were constructed The capacities of the investigated WWTPs cover in such places as Petregal (Brazil), Cali (Colombia), a wide range, from about 10,000 PE60 to almost 1 and Kanpur (India) in the 1980s. UASB is not yet million PE60. Most operate within their capacity, and widespread in the EAP countries focused on in this overloading is infrequent. report, even though the climate there would be well suited. Apart from several large-scale Indian Preliminary treatment at all the plants is by means applications, a UASB pilot in Singapore has clearly of screens and grit chambers. Specific fat removal is demonstrated that this technology would work well in never available. A pumping station is always included, EAP (Cao 2011). whether for influent pumping or pumping onto TFs. UASB technology digests both wastewater and sludge The twenty-two investigated WWTPs polish their in the same reactor without a supply of oxygen. Hence, anaerobic effluents from the UASB as follows: twelve no separate sludge digesters are required, as they were plants use trickling filters (TFs), four use ponds, one in case studies 1 and 2, and electricity consumption uses conventional activated sludge (CAS), and five is limited to wastewater pumping and supplementary have no polishing stage at all. Figure CS3-1 presents a installations, such as sludge dewatering. The anaerobic simplified flow scheme for the WWTPs of case study reactors are covered, and the biogas is collected. Since 3, and figure CS3-2 shows an aerial view of one of it is an anaerobic technology, which is usually not the investigated plants. Table CS3-1 provides key sufficient to meet required effluent standards, many characteristics of all the facilities. UASB systems feature a subsequent aerobic polishing ANNEXES WASTEWATER TO ENERGY   73 Figure CS3-1: Simplified typical flow scheme of UASB systems for case study 3 Source: Authors. Figure CS3-2: Aerial view on Onça WWTP, the largest investigated UASB plant (177 MLD) Screens, Grit chamber UASB Flare Sludge dewatering TF SST Source: COPASA—Onça WWTP. ANNEXES WASTEWATER TO ENERGY   74 For more details of actual wastewater characteristics and other parameters, see section 3.2.1. Table CS3-1: Key characteristics of UASB plants investigated for case study 3 Actual avg. an- WWTP Technology Capacity Capacity nual load     (m3/d) (PE60) (PE60) Curvelo UASB-TF 8,747 54,300 41,050 Janaúba UASB-ponds 3,854 20,157 25,280 Montes Claros_Vieiras UASB-TF 42,738 320,533 128,100 Tres Marias UASB-TF 3,065 20,673 16,183 Ipatinga_Rio Doce UASB 32,704 165,385 206,640 João Pinheiro UASB-TF 4,450 26,556 16,150 Lafaiete_Bananeiras UASB-TF 7,486 49,081 18,617 Alfenas UASB-TF 17,088 74,450 30,152 Caxambu UASB-ponds 6,963 27,450 6,103 Itajubá UASB 19,910 98,024 32,250 Lavras_Água Limpa UASB-ponds 6,786 20,976 14,967 Lavras_Ribeirão Ver- UASB-ponds 13,515 59,383 29,521 melho Pouso Alegre UASB 25,634 111,182 31,800 Varginha_São José UASB 10,289 39,287 22,883 Varginha_Santana UASB 18,985 79,990 29,367 Onça UASB-TF 176,861 930,833 577,917 Betim Central UASB-CAS 44,391 278,733 53,653 Nova Contagem UASB-TF 7,442 44,683 12,667 Vale do Sereno UASB-TF 2,354 10,210 9,263 Pará de Minas UASB-TF 12,994 74,000 52,317 São José da Lapa UASB-TF 3,228 20,600 7,450 Vespasiano UASB-TF 3,944 33,341 10,837 Average   21,519 116,356 62,417 Total   473,426 2,559,829 1,373,166 Source: FWT 2013c. Note: 1 cap = 54 g BOD5/d in Brazil; 1 PE60 = 1.11 PE54 = 1.11 cap in Brazil. ANNEXES WASTEWATER TO ENERGY   75 3.1.3. Sludge management table summarizes selected key parameters for design At most of the WWTPs analyzed in case study 3, and construction of UASB. sludge is produced in two stages: (a) in the UASB reactors and (b) in the polishing stage. Sludge from 3.1.4. Energy management the latter is generally conveyed back to the UASB In general, generated biogas is only flared at WWTPs. for stabilization. Hence, typical sludge withdrawal The biogas flow rate is only metered at a few of is only from the UASB reactor. This sludge has an the WWTPs of case study 3, and, unfortunately, average DS content of about 3.8 percent and can be these few flow meters deliver implausible results. considered stabilized (VS = 56 percent on average). General state-of-the-art design parameters and For that reason there is no particular need for further recommendations for UASB biogas systems are stabilization, and the sludge is just dewatered. Out compiled in table CS3-11. of the twenty-two plants, seven have centrifuges for sludge dewatering, while the other fifteen use sludge 3.2. Analysis drying beds. 3.2.1. Wastewater influent and effluent and For details of actual sludge characteristics and other other parameters of interest parameters, see section 3.2.1. Table CS3-2 and figures CS3-3, CS3-4, and CS3- 5 summarize the outcome of an assessment of the General design parameters for UASB reactors twenty-two UASB plants in case study 3 (FWT according to state-of-the-art recommendations are 2013c). The results cover the complete calendar year compiled in table CS3-10. It is assumed that the 2011, with data related to both wastewater and sludge interested reader is familiar with the background and characteristics presented. microbiological principles of anaerobic digestion. The ANNEXES WASTEWATER TO ENERGY   76 Table CS3-2: Actual influent and effluent data of UASBs of case study 3, average data from 2011       UASBs Number of WWTPs   22 Pop. equivalents avg. actual PE60 per UASB 62,417   max. actual PE60 per UASB 577,917 WASTEWATER QUANTITY     Specific wastewater production m /PE60/y 3 80     L/PE60/d 220 WASTEWATER QUALITY     COD Influent mg/L 697   Effluent mg/L 194   Elimination % 72 BOD5 Influent mg/L 297   Effluent mg/L 62   Elimination % 76 Ntotal Influent mg/L n.a.   Effluent mg/L 41   Elimination % n.a. NH4-N Effluent mg/L 38 Source: FWT 2013c. NO3-N Effluent mg/L 1 Notes: 1 cap = 54 g BOD5 /d in Brazil; 1 PE60 = 1.11 PE54 = 1.11 cap in Brazil; m3/PE60 /y x 1.11 = m3/cap/y in Brazil; L/PE60 /d x 1.11 Ptotal Influent mg/L 7.1 = L/cap/d in Brazil. “Effluent” and “elimination” related to total   Effluent mg/L 4.5 WWTP (not UASB only).   Elimination % 33 n.a. = not available. Figure CS3-3: BOD5 removal efficiencies at WWTPs of case study 3 Source: FWT 2013c. ANNEXES WASTEWATER TO ENERGY   77 Table CS3-3: Key average characteristics of UASBs of case study 3, actual versus design       Actual Design General         Daily flow rate average m3/d 12,093 21,519 BOD5 load average kg/d 3,745 6,981 COD load average kg/d 8,294 — TSS load average kg/d 3,493 — BOD5 effluent average mg/L 62 53 COD effluent average mg/L 194 — TSS effluent average mg/L 66 — WASTEWATER TRAIN         UASB retention time average h 15.8 7.8 UASB volumetric load average gBOD5/m3/d 580 1,080 UASB surface load average m/h 0.4 0.6 SLUDGE TRAIN         Sludge from UASB average m3/d n.a. n.a.   kgDS/d n.a. n.a. VS of raw sludge average % n.a. n.a. Digester retention time average d n.a. n.a. VS destruction in digester average % n.a. n.a. VS of UASB sludge average % 56 n.a. BIOGAS TRAIN         Biogas production average m3/d n.a. n.a. Specific biogas product. average L / kgVSdestroyed n.a. n.a. Calorific value biogas average kWh/m3 n.a. n.a. H2S average ppm n.a. n.a. Wastewater temperature average °C 23.6 n.a. Source: FWT 2013c. n.a. = not available. Sludge data quality at the investigated UASB systems Total sludge quantity equaled about 30 gDS/PE60/d. was generally not perfect, and not all values of interest These results, though, are just based upon a few values, were available at all facilities. Figures CS3-4 and CS3- since sludge quantities are poorly documented. 5 summarize important results as they were available. ANNEXES WASTEWATER TO ENERGY   78 Figure CS3-4: Dry solids content of UASB sludge at WWTPs of case study 3 Source: FWT 2013c. Note: Where no dry solids operation data are presented, none were available. Figure CS3-5: Volatile solids content of UASB sludge at WWTPs of case study 3 Source: FWT 2013c. ANNEXES WASTEWATER TO ENERGY   79 3.2.2. Biogas production and potential for consumers; hence, there was little incentive to invest in energy generation biogas utilization. The situation changed completely with a new federal law in 2012 allowing WWTPs to Biogas production consume at one site the same electricity they supply to Biogas flow metering was incomplete at the plants the grid at another (see section 3.2.4) at no cost. As a studied. The most reliable biogas quantification was consequence, COPASA has now begun the process of done in the same project area by the local university in assessing the biogas’s energy potential and the financial Minas Gerais (Universidade Federal de Minas Gerai implications of biogas utilization. It is to expect that [UFMG]) by Lobato and others (2011, 2012). The several energy utilization projects will be implemented results of that assessment have been summarized in in the years to come. table CS3-11. Its plausibility has also been confirmed Since no explicit, clearly defined biogas potential by results from other authors—for example, Noyola is available from the operators of the twenty-two and others (2006)—whose results are also provided UASBs, the subsequent analysis in case study 3 looks in that table. Here in this section, only the final at a potential biogas range between 7 and 17 L/ conclusion is presented: mean collectable biogas is 13 PE60/d, with 13 L/PE60/d considered the most likely L biogas/PE60/d (range 7–17 L biogas/PE60/d). This practical result. This range of typically collectable could be considered a good result, which is not much biogas resulted from the analysis of different influent below what can be collected in an anaerobic heated scenarios, as presented in table CS3-11. It can thus be state-of-the-art sludge digester. It is important to keep seen as a good reflection of conditions prevailing for in mind, however, that this result is only possible in a any of the twenty-two WWTPs of this case study. For properly designed and operated UASB. the calculation of the electricity generation potential, Potential for energy generation the following values were additionally used: calorific value of the biogas = 6.5 kWh/m3; electric efficiency To date, none of the analyzed WWTPs has been of CHP = 30 percent. Both these assumptions are producing electricity from biogas. Traditionally, the cautious, and an even higher electricity generation concept of energy generation from biogas had never potential might be quite feasible in cases of higher been applied to UASB in the region. Moreover, it used calorific value and/or higher CHP efficiency. The to be impossible to supply electricity from a WWTP results are summarized in table CS3-4. into the public grid. Most UASBs are not big energy Table CS3-4: Electricity production potential for a typical range of specific biogas production of 7–17 L/PE60/d at UASBs Electricity potential from biogas   (kWh/PE60/d)   7 L/PE60/d 10 L/PE60/d 13 L/PE60/d 17 L/PE60/d 5.0 7.1 9.3 12.1 Source: Authors’ calculation. Note: L/PE60/d x 16.67 = L/kg BOD5/d. ANNEXES WASTEWATER TO ENERGY   80 A comprehensive database was available on the electric and benchmark energy consumption values (FWT power requirements of all the analyzed UASB systems 2013c); see table CS3-5 and figure CS3-6. The classes in the year 2011. The in-depth assessment showed were defined as follows: a pronounced effect of economies of scale—that is, • Class 1: <10,000 PE60 larger WWTPs consumed less energy, and vice versa. For practical presentation of the results, the plants • Class 2: 10,000–50,000 PE60 were divided into four “classes” of different sizes, • Class 3: 50,000–100,000 PE60 and for each class it was possible to derive median • Class 4: >100,000 PE60 Table CS3-5: Electricity requirements of UASBs of case study 3     Electricity consumption   PE60 median benchmark     (kWh/PE60/y) (kWh/PE60/y) Class 1 <10,000 29 7.6 Class 2 10,000–50,000 13 2.5 Class 3 50,000–100,000 8 1.3 Class 4 >100,000 6 0.8 Source: FWT 2013c. Note: kWh/PE60/y x 16.67 = kWh/kg BOD5/y. The respective electricity requirements of the analyzed production potential from biogas in figure CS3-6. UASB plant classes are compared to the electricity Figure CS3-6: Power generation potential of realistic biogas production range, compared to actual electricity consumption of analyzed UASB plants Source: FWT 2013c. Note: kWh/PE60/y x 16.67 = kWh/kg BOD5/y. ANNEXES WASTEWATER TO ENERGY   81 From these results, the following conclusions can be requirements. Hence, the electricity consumption derived: at a specific plant appears to be dominated by the extent of wastewater pumping and by the electric • Actual electricity consumption at UASB plants efficiencies of the particular plant’s installations. shows a clear effect of economies of scale: larger There was only one notable exception: the UASB plants consume less energy, and vice versa. system that used a CAS system for polishing • Median electricity consumption of large plants featured elevated electricity consumption. However, equals 6 kWh/PE60/y; median plant sizes consume in that specific case, this was not so much dependent around 10 kWh/PE60/y; and small ones consume a on technology as it was a consequence of too-large median of almost 30 kWh/PE60/y. blowers running at much overcapacity, thereby • Benchmark values for the best performers have also needlessly consuming too much energy. been derived from the data. They are within 1–8 Figure CS3-7 clarifies further by looking at energy kWh/PE60/y for all classes. requirements for wastewater pumping in general. • No apparent correlation between specific electricity As it turns out, the energy requirement for pumping consumption and type of post-treatment could be wastewater up to five meters is typically <2 kWh/PE60/y, found, apart from the observation that the only CAS depending on the PE60 specific wastewater flow rate. polishing system had higher than average energy Figure CS3-7: Electricity requirement for wastewater pumping Source: Authors’ calculation. Note: kWh/PE60/y x 16.67 = kWh/kg BOD5/y. ANNEXES WASTEWATER TO ENERGY   82 The energy consumption values presented in figure Generally it appears that, for medium plant sizes CS3-6 are reasonable, and benchmark values of 1–8 of about 50,000 PE60 upwards, UASB plants could kWh/PE60/y are feasible. To date, no comprehensive operate with a positive energy balance on an annual energy data on UASB, comparable to the content basis if biogas production, collection, and utilization of figure CS3-6, have been published. These data worked normally—that is, these plants will be able are assumed to be the first benchmarking exercise to produce more energy than they consume. The for UASB. Electricity consumption data of UASB electricity supply from the public grid hence only plants are usually not well documented nor published serves as “safety net” for peak periods and during systematically. operational problems. Small UASB plants of less than 10,000 PE60, on the other hand, will only achieve a When comparing the above electricity consumption positive energy balance if they are optimized both in values for large plants to that of CAS (case study 1), terms of energy consumption and biogas production/ which equaled about 30–35 kWh/PE60/year, and that utilization. Yet, in reality, it is more likely that these of a large TF plant (case study 2), which equaled 9–10 plants will only recover about 30–50 percent of their kWh/PE60/year, UASB plants apparently show similar electricity requirements from biogas-generated sources. energy needs to TF plants, with a tendency to even The biogas and power generation potential of UASB is lower electric power consumption. summarized in table CS3-6. Note that in this case, the plants were all designed for carbon removal only. Table CS3-6: Biogas and power generation potential of UASB plants of case study 3   Key energy values of the investigated   UASBs Biogas production   - N elimination (L/PE60/d) — - C elimination (L/PE60/d) avg. 13 (7–17) Electric efficiency CHP (%) 30 Thermal efficiency CHP (%) 55 Calorific value of biogas (kWh/m3) 6.5 Power generation   - N elimination (kWh/PE60/year) — - C elimination (kWh/PE60/year) avg. 9 (5–12) Thermal energy generation   - N elimination (kWh/PE60/year) — - C elimination (kWh/PE60/year) avg. 17 (9–22) Total energy generation   Source: Authors’ calculation. - N elimination (kWh/PE60/year) — Note L/PE60/d x 16.67 = L/kg BOD5/d; kWh/PE60/y x - C elimination (kWh/PE60/year) avg. 26 (14–34) 16.67 = kWh/kg BOD5/y. ANNEXES WASTEWATER TO ENERGY   83 3.2.3. Operation capacity needs, biogas safety Scum formation The key operational issues concerning UASB reactors This is a rather specific phenomenon observed relate to specific operational problems and to safety in many UASB reactors worldwide. Only with a concerns due to the management of explosive biogas. renewed interest in optimized biogas collection and The following are the most relevant aspects: utilization, though, is it nowadays receiving due • Safety concerns attention. For many years it was just considered an operational hassle that was rarely investigated further. • Deposits in the digester However, now it has become standard knowledge • Insufficient biogas treatment that substantial scum formation under the three- • Scum formation phase separator hampers biogas collection and can eventually even bring it to a standstill. Figure CS3-8: Sand deposits inside UASB reactor Gas bubbles apparently have difficulty penetrating the rather compact scum layer. The gas accumulates underneath and then eventually manages to escape into the open air via the sedimentation compartment or through small cracks and/or openings in the three- phase separator. Consequently, it is not surprising that the flares at many UASBs went out of operation long ago after the initial operation period, since no gas arrived there anymore. The scum is a mixture of sludge, solids, and FOG Source: Morais et al. 2013. (fat, oil, and grease; Chernicharo et al. 2013). Its accumulation rate is relatively high. Morais and others The first three issues have already been discussed (2013) report a scum yield of 0.04 L/kgCODapplied. in case study 1. The arguments and information Chernicharo and others (2013) report scum yields of provided there also prevail for case study 3. For details, 0.20–0.24 L/m2/d, equivalent to 0.004 L/kgCODapplied. see section 1.2.3. This results in a scum buildup of about 1–10 mm per week under the three-phase separator. Hence, Deposits in the digester without scum removal, the scum layer will be several centimeters thick within a few weeks at the latest and As indicated, the same problems and counter-remedies will continue growing and compacting. Over time, prevail in principle for UASB as for mesophilic sludge manual removal becomes increasingly difficult, and digesters (described in section 1.2.3). For instance, removal by suction trucks proves time consuming and Morais and others (2013) describe sand accumulation costly. Two or three workers can be fully employed of 1.5 meters depth inside UASB reactors due to all year round at a medium or large UASB plant just malfunctioning of the grit chamber (see figure CS3-8). ANNEXES WASTEWATER TO ENERGY   84 removing the scum. Some impressions regarding scum If the scum is not removed from under the three-phase and its removal are presented in figure CS3-9. separator for prolonged periods, eventually it will also enter the sedimentation compartment and deteriorate even the effluent quality. Figure CS3-9: Scum formation in UASB reactors Scum removal at UASB Pará de Minas, Brazil Scum removal at UASB Onça, Brazil Scum in effluent compartment at Scum under three-phase separator Scum under three-phase separator UASB Nathay, Egypt at UASB Onça, Brazil at UASB Ipatinga, Brazil ANNEXES WASTEWATER TO ENERGY   85 Any counter-measure against scum can be based on through an aerated grit chamber. However, the one or both of the following elements: efficiency of these installations is considered unreliable under warm climate conditions. After all, temperature • Prevention/minimization of scum formation is an ideal “solvent” for FOG. The authors of the • Regular removal of scum present study believe a fine sieve, as described above, The prevention/minimization of scum must target its can have a similar effect: it is known that removing main constituents: (a) solids and (b) FOG. most of the solids also removes considerable quantities of the FOG attached to those solids. Thus, additional For improved solids removal, apparently even 6 mm fat removal stages are definitely recommendable but fine screens are insufficient. Many of the WWTPs might not be imperative if fine sieves are installed. investigated in case study 3 have screens, but they still suffer from scum. Therefore, it is recommended to use Scum removal must be automated if it is to work even finer screens. Ideally suited would be sieves with satisfactorily. Only if scum is removed regularly at openings of 1 mm or 2 mm. The 1.0 and 1.5 mm brief intervals, when it is still quite liquid, can the sieves, for instance, have proved very effective in the operator avoid its solidification and compaction. One elimination of almost all solids at MBR plants. system suggested by Chernicharo and others (2013) is depicted in figure CS3-10: a weir is installed in the FOG removal is more difficult but nonetheless upper part of the three-phase separator, and the water important. Parravicini (2012) describes the impact of level is kept 30 mm below the weir’s crest. Opening a FOG: these compounds attach to the sludge pellets valve connected to this zone causes the pressure under and thus reduce their specific weight. Consequently, the three-phase separator to drop and the water level the sludge particles float easily and are incorporated to increase. Scum flows over the weir and is removed. into the scum on the liquor surface. None of the plants The scum is sieved, with the liquor phase going to a in case study 3 features a specific fat removal stage drying bed and the solids going to a sanitary landfill. prior to the UASB stage. This also holds true for most The authors report that the minimum scum removal other municipal UASB plants worldwide. There are efficiency was 80 percent, while over 90 percent was conventional technologies to remove fat—for instance, most common. ANNEXES WASTEWATER TO ENERGY   86 Figure CS3-10: Scum removal technology for UASB Source: Suggested by Chernicharo et al. 2013. 3.2.4. Institutional aspects, energy costs autogenerated electricity. If one and the same operator The institutional background for biogas utilization is in charge of several plants, which is frequently the underwent a major change recently in Brazil. Before, case in Brazil, it can supply the excess power from it was legally impossible to supply electricity from a one WWTP into the public grid and withdraw it at WWTP into the public grid; consequently, interest another that does not dispose of electricity generation in biogas production and utilization was low, and from biogas. Hence, it has become relatively easy for a most biogas from UASBs was just flared. In 2012, the WWTP operator to utilize 100 percent of the electric situation changed completely: ANEEL, the national power generated from biogas at its own installations. energy agency in Brazil, introduced legislation that The financial OPEX gain of a WWTP operator allows the supply of electric power to the public producing its own electric power is thus equal to the grid from microgeneration sources (ANEEL 2012). cost of electricity purchased from the public grid. This Furthermore, a supplier to the public grid is now also unit cost per kWh is not exactly the same for any plant allowed to withdraw the same quantities of electric in Brazil, but the average of the twenty-two WWTPs power at another site, free of supply cost. investigated for case study 3 in 2011 was found to be This means an operator of a WWTP can now substitute US$0.25/kWh. the unit cost of its own electricity consumption through ANNEXES WASTEWATER TO ENERGY   87 3.2.5. GHG reduction and CDM co-financing turned out frequently to emit more GHG than aerobic For general information on CDM co-financing, price technologies, even though its energy consumption of carbon credits, and so on, see section 1.2.5. is low. The break-even point was calculated at an influent BODu of 300 mg/L for extended aeration An interesting question when summing up all and an influent BODu of 500–700 mg/L for aerobic GHG emissions of wastewater treatment is whether wastewater treatment with sludge age between five and anaerobic treatment emits more or less GHG than ten days. Only if influent BODu is higher than those aerobic treatment. Cakir and Stenstrom (2005) values does anaerobic treatment lead to lower GHG analyzed this issue and came up with an interesting emissions than aerobic treatment. Since municipal result. The thinner influent wastewater is, the more wastewaters frequently are not that concentrated, CH4 is lost in the treated effluent of anaerobic overall anaerobic systems emit more GHG than wastewater treatment. And since CH4 is a twenty-one aerobic technologies (see figure CS3-11). This picture times stronger GHG than CO2, this results in high would change, though, if the CH4 contained in the CO2e emissions from anaerobic wastewater systems. effluent of anaerobic systems were recovered. Consequently, anaerobic wastewater treatment has Figure CS3-11: Total CO2e emissions of wastewater treatment in function of influent BODu Source: Cakir and Stenstrom 2005. Note: Theta = sludge retention time; BODu = ultimate BOD. Specifics of case study 3 elimination, since the methane is created and eliminated by the project. As pointed out in section 1.2.5, due to the “additionality” criterion, the new construction of a UASB complete Brazil’s specific energy mixture causes the GHG emissions with CHP could only take into consideration the energy from electricity generation presented in table CS3-7. production component under CDM but not methane ANNEXES WASTEWATER TO ENERGY   88 Table CS3-7: GHG emissions per kWh for electricity generation in Brazil, as compared to the world Region 1990 2010 gCO2/kWh gCO2/kWh Brazil 55 87 World 586 565 Source: IEA 2012. Note: The table shows CO2 emissions from fossil fuels consumed for electricity generation, in both electricity-only and combined heat and power plants, divided by output of electricity generated from fossil fuels, nuclear, hydro (excluding pumped storage), geothermal, solar, wind, tide, wave, ocean, and biofuels. Both main activity producers and autoproducers have been included in the calculation. The annual electricity production potential from 3.2.6. CAPEX structure biogas of the twenty-two WWTPs of case study 3 Complete UASB plant equals 12.7 million kWh/year. This would allow a potential GHG reduction of 1,100 tons CO2e/year for CAPEX of seventeen of the twenty-two WWTPs of case study 3 in 2011, based on Brazil’s average energy case study 3 was available as total CAPEX for each and mixture. every plant. COPASA updated these CAPEX to 2012 cost levels, which were then related to the individual design capacity of each plant (FWT 2013a). This resulted in the cost curve presented in figure CS3-12. Figure CS3-12: CAPEX of UASB plants of case study 3 Source: FWT 2013a. Note: US$/PE60 x 16,67 = EUR/kg BOD5. ANNEXES WASTEWATER TO ENERGY   89 Comparing these values to international CAPEX • New preliminary treatment, complete with fine information leads to different results. For instance, sieves 1–2 mm and new aerated grit chambers for Libhaber and Orozco-Jaramillo (2012) mention sand and FOG removal: CAPEX = avg. US$7.5 CAPEX of US$20–40/capita for UASB reactors. (range US$4–11)/PE60 = avg. EUR5.5 (EUR3–9)/ Wagner (2010) reports about US$25–50/capita for PE60. UASB + ponds and specific UASB reactor cost of • Biogas system: collection, treatment, storage, and US$150–250/m3. WERF (2010b) cites CAPEX for utilization in CHP: CAPEX = avg. US$8.0 (US$5– UASB + post-treatment of US$35–60/PE. Sperling 15)/PE60 = avg. EUR 6.0 (EUR 4–11)/PE60. and Chernicharo (2005) indicate a range of about US$15–45/capita for UASB + post-treatment. • Generally, there was a clear effect of economies of scale: smaller plants were more expensive, per All in all, the literature points toward CAPEX of population treated, than larger ones. Additionally, 26°C) (Sperling and Chernicharo 2005) • Volumetric BOD load = <1,750–2,000 g BOD5/m3/d (Haandel and Lettinga 1994; Sperling and Chernicharo 2005); 1,650 at 20°C and 2,700 at 25°C (Libhaber and Orozco-Jaramillo 2012) • Upflow velocity: - <1.0 m/h (Haandel and Lettinga 1994; Libhaber and Orozco-Jaramillo 2012) - <0.6 m/h (WERF 2010b) - <0.5–0.7 m/h, max. 1.5–2.0 m/h (Sperling and Chernicharo 2005) • Sludge production: - UASB sludge: 0.2–0.4 gDS/gBOD5, influent (Haandel and Lettinga 1994; Sperling and Chernicharo 2005) - Co-digested secondary sludge from polishing stage; see information provided in tables CS1-4 and CS1-5. An average sludge retention time of ≈30 days can be assumed in UASBs. • Typical sludge DS = 3–5 percent (Sperling and Chernicharo 2005) • BOD removal efficiency: Is calculated by equations, such as the following: - Haandel and Lettinga 1994: EBOD = 100 x (1 – 0.68 x t–0.68) for T > 20°C - Sperling and Chernicharo 2005: EBOD = 100 x (1 – 0.70 x t–0.50) for T = 20–27°C - The BOD removal efficiency at T < 20°C is still subject to discussion. For instance, Singh and Viraraghavan (2003) recommend T = 15°C to reduce the efficiency calculated for 20°C by 10 percent. • SS removal efficiency: No reliable design tool exists, but an assumption of SS in the UASB effluent of 40–100 mg/L appears appropriate. ANNEXES WASTEWATER TO ENERGY   95 UASB Both rectangular and circular UASB reactors have been constructed. However, the SHAPE AND rectangular type is by far the most widespread shape nowadays. CONSTRUCTION Typical water depth in UASB is in the range of 3–6 m. The total reactor volume is usually split over several reactors. The individual reactor volume per unit recommended was <1,000 m3 (Haandel and Lettinga 1994) in the ’90s; nowadays it is 2,000–3,000 m3 per unit at large WWTPs. The environment in UASBs is highly corrosive. Consequently, most UASB reactors are constructed out of reinforced concrete, possibly with sufficient chemical resistance. It is imperative that all equipment that is not made from concrete, ranging from pipes, channels, distribution chambers, and manhole covers to the gas collection system, be made from noncorrosive materials such as PE, PVC, or GRP. The main construction differences of UASB found nowadays among large-scale plants relate to the influent distribution. INFLUENT It is of major relevance that the influent wastewater be evenly distributed at the bottom DISTRIBUTION of the UASB reactor. To that end, the total incoming flow is split into numerous subflows, each of which is then discharged at the UASB bottom through a specifically assigned pipe. The influence area of this supply pipe should be less than 5 m2, and ideally between 1.5 and 3.0 m2 (Sperling and Chernicharo 2005). With increasing flow rates, flow splitting is increasingly difficult to implement. Hence, influent distribution becomes a limiting factor in terms of the maximum treatment capacity UASB plants can achieve. For example, Onça WWTP (which is also part of this case study) is designed for a peak flow of 3.7 m3/s (avg. 2.1 m3/s), which is near the maximum flow range of what is practically recommendable. Wastewater splitting is usually done on the reactor surface, either by circular distribution devices or longitudinal (rectangular) devices (see figures CS3-14 and CS3-15). The former consist of circular distribution wells, and the latter use longitudinal free surface channels along which numerous distribution boxes are located. Each distribution box then receives its specific distribution pipe. To minimize clogging risks, the pipes should have a diameter of >75 mm. Since each pipe is hydraulically separated from the other, clogging can be easily spotted. In that case, pipe cleaning is done either with mechanical tools or high- pressure water. ANNEXES WASTEWATER TO ENERGY   96 Figure CS3-14: Typical distribution systems of influent to UASB Source: Haandel and Lettinga 1994. Figure CS3-15: Circular and rectangular inlet distribution at Betim Central WWTP (Brazil) and Cafeco WWTP (El Salvador) Source: COPASA—Betim Central WWTP, CAFECO WWTP. SLUDGE It is important for the operator to know the distribution of the sludge blanket inside the SAMPLING AND UASB and to optimize the discharge of sludge (as constant as with as high TS concentration DISCHARGE as possible). To that end, it is advisable to install sludge sampling and discharge pipes at different levels of the reactor. Sampling is usually done at several levels, 0.5–1.0 meters apart, while sludge discharge is often possible at just two levels: one near the reactor bottom and another 1.0–1.5 meters above that level (see figure CS3-16). ANNEXES WASTEWATER TO ENERGY   97 Figure CS3-16: Sludge sampling and discharge pipes at Montes Claros WWTP (Brazil) sludge sampling sludge discharge Source: COPASA—Montes Claros WWTP. Three-phase separator The three-phase separator is a key component of any UASB and is located in the upper zone of the reactor. Its purpose is to separate the three phases: biogas, liquor, and sludge. After all, all three phases move upward, since this is the wastewater flow direction, and gas bubbles rise as well. The three-phase separator (figure CS3-17) is constructed in such a way that the rising gas bubbles are collected and conveyed to the gas outlet pipe. This leads to a separation of gas and sludge particles, with the sludge particles consequently settling back into the sludge blanket. Still, some of the sludge particles may enter the settling compartment together with the treated effluent. Since the gas has already been effectively removed and cannot enter this compartment (as all vertical trajectories lead to the gas outlet), the sludge will be separated from the liquid phase here, and the particles will sink and glide back into the sludge blanket from this zone as well. Design and construction details (Sperling and Chernicharo 2005; Libhaber and Orozco- Jaramillo 2012) To facilitate the gliding back of solids from the sedimentation compartment to the digestion compartment, and for efficient gas collection underneath, the inclination of the three-phase separator should be at least 45°, and preferably greater than 50°. The sedimentation compartment itself should be 1.5–2.0 meters deep. Hydraulic retention time in the sedimentation compartment should be 1.5–2.0 hours on average (>1.0 [0.6] hours during daily [hourly] peak flows), and surface loading in the sedimentation compartment should be 0.6–0.8 m/h (<1.2 [1.6] m/h during daily [hourly] peak flows). The velocity in the apertures to the sedimentation compartment may not exceed 4.0 (5.5) m/h during daily (hourly) peak flows, and should be <2.0 m/h during average flow. ANNEXES WASTEWATER TO ENERGY   98 Figure CS3-17: Schematic section through an UASB reactor Source: Sperling and Chernicharo 2005. Three-phase separators can be constructed from quite different materials, ranging from coated metal sheets and plastic foils to concrete (see figure CS3-18). Figure CS3-18: Three-phase separators TREATED The treated effluent is collected on the reactor surface either through (a) launders with EFFLUENT overflow V-notch weirs (figure CS3-19) or (b) submerged perforated tubes. The former COLLECTION have the advantage of easier cleaning and control but are rather sensitive to exact leveling. The latter are more difficult to clean, if needed. Figure CS3-19: Treated effluent collection via launders with weirs at Pará de Minas WWTP (Brazil) Source: COPASA— Pará de Minas WWTP. ANNEXES WASTEWATER TO ENERGY   99 Table CS3-11: UASB biogas systems: General design parameters and key characteristics BIOGA S • Volatile solids concentration of sludge: UASB sludge: avg. 55 percent (50–60 DE S I GN percent) (Bauerfeld et al. 2009) • Volatile solids destruction of sludge from polishing stages in UASB: according to principles presented in table CS1-5 • Gas production: - Traditionally, the biogas yield has been estimated in UASB through a simple COD balance: CODbiogas = CODinfluent – CODeffluent – CODsludge, combined with the general gas equation and an assumed CH4 content in the biogas. This procedure is, for instance, described in Haandel and Lettinga (1994) and Sperling and Chernicharo (2005). However, in practice it turns out that this approach mostly leads to a more or less significant overestimation of the collectable biogas quantities. It is hence not recommendable. - Generally, it is of major importance to distinguish between biogas production and collectable biogas. There is always a considerable percentage of biogas that cannot be collected for various reasons. As has been found by Lobato and others (2011, 2012), the following requires careful analysis: COD converted into sludge; COD used for sulfate reduction; dissolved COD of CH4 lost with effluent; COD not converted into CH4 and lost with effluent; and other CH4 losses. Based on an elaborate study of several UASBs, and also taking typical parameter variations into account, the following conclusion can be derived: collectable biogas = 10–17 L/ cap/d (mean 14 L/cap/d). Behind these values are capita-specific pollution loads of an average 90–110 gCOD/cap/d. Hence, by including a small safety margin, it is possible to equate these capita-specific values with PE60-specific values. On that basis, the following collectable biogas yield can be assumed: 10–17 L biogas/ PE60/d (mean 14). - It is important to mention that the result of the above-described latest findings related to collectable biogas yields match earlier findings. Behind the above-cited PE-specific gas production numbers of Lobato and others (2011, 2012) lies a CH4 yield of 113–196 L CH4/kgCODremoved. This is rather similar to what had been found by Noyola and others (2006), who reported a range of 80–180 L CH4/kgCODremoved (≈7–16 L biogas/PE60/d). ANNEXES WASTEWATER TO ENERGY   100 - Conclusion: Based on the above described findings, the collectable biogas yield of municipal UASB can be expected to fall within a range of 7–17 L biogas/PE60/d, with average values of 13 L biogas/PE60/d (= 217 L biogas/kgBOD5/d, range = 115–280 L biogas/kgBOD5/d). • Typical biogas characteristics: (a) methane CH4: 60–80 percent; (b) calorific value: 6.0–8.0 kWh/Nm3 (Sperling and Chernicharo 2005) • COD mass balance: See typical example in figure CS3-20. Figure CS3-20: Typical COD mass balance for UASB Source: Lobato et al. 2011, 2012. Typically, less than 50 percent of influent COD is converted into CH4. The more than 50 percent of COD remaining is lost in the effluent, required for sulfate reduction, or converted into sludge. From the COD converted into CH4 only about two-thirds are actually collectable in the UASB, while one-third is lost. These losses can even go up to about 50 percent of the CH4 formed under unfavorable conditions. Under ideal conditions, they may be reduced to about 25 percent of the CH4 formed (Lobato et al. 2011, 2012). Similar results are presented by Chernicharo and others (2012). BIOGAS TREATMENT See table CS1-5. GAS HOLDER See table CS1-5. FLARE See table CS1-5. BIOGAS UTILIZATION See table CS1-5. ANNEXES WASTEWATER TO ENERGY   101 CASE STUDY 4: COVERED ANAEROBIC PONDS ANNEXES WASTEWATER TO ENERGY   102 4.1. BACKGROUND, PROCESS DESCRIPTION 4.1. Background, process description Case study 4 looks into the practical results from two cities in which covered anaerobic ponds have been 4.1.1. Data sources applied for many years: Santa Cruz in Bolivia and Waste stabilization ponds (WSPs) are a simple Melbourne in Australia. All operation data utilized in and widespread wastewater treatment technology. this case study were provided by the plant operators: Typically, they feature a series of ponds that cover SAGUAPAC (Cooperativa de Servicios Públicos Santa considerable areas of land. Apart from pumping the Cruz Ltda), www.saguapac.com.bo, and Melbourne wastewater into them, usually no electric energy is Water, www.melbournewater.com.au. required for running those systems. For the EAP countries on which this report focuses, WSPs are well Santa Cruz is a fast-growing, economically dynamic suited in terms of the warm climate. However, given city with a population of about 1.5 million, located that only municipal WWTPs are discussed here, the in the tropical lowlands of Bolivia at about 17° south land requirement is a serious constraint, and, in most latitude and about 400 meters altitude. Its annual cases, classical WSP systems may not be viable as average air temperature is around 23°C. municipal wastewater treatment systems. Melbourne is the capital of the state of Victoria. With The first, high-loaded stage of WSPs—anaerobic 4.3 million inhabitants, it is the second most populous ponds (APs)—could be a technology component city in Australia. Located at about 37° south latitude to consider in terms of energy generation potential. at sea level, it has an annual average air temperature of The subsequent phases of facultative ponds and around 20°C. maturation ponds are less relevant from that point of view. This has held particularly true from the moment 4.1.2. Wastewater management it was proved feasible to cover APs with plastic covers Santa Cruz, Bolivia and collect the biogas generated underneath. The AP thus turns into a combined wastewater treatment and SAGUAPAC operates four WWTPs, based on WSP digestion unit. Its land requirements are relatively technology with biogas collection in APs. Since two small as compared to complete WSP systems, and yet plants are located together at the same site, there are still more than 50 percent of organic load is removed. just two sites for biogas utilization. The key features Subsequent treatment stages will be needed to bring of those plants are summarized in table CS4-1, and effluent quality to required levels, but the effects of some pictures are provided in figures CS4-1, CS4-2, removing a considerable organic load free of energy and CS4-3. cost, thereby producing and collecting biogas that can be used for electricity generation, is an attractive combination. ANNEXES WASTEWATER TO ENERGY   103 Table CS4-1: Key characteristics of WWTPs in Santa Cruz, analyzed for case study 4 WWTP Technology Total ponds Anaerobic ponds water     number area number area depth     (nr.) (ha) (nr.) (m) (ha) System North             Norte 1 AP - FP - MP 5 20 2 3.5 3.0 Norte 2 AP - FP - MP 8 39 2 4.5 3.7 SUBTOTAL   13 59.0 4 — 6.7 System East             Este AP - FP - MP 12 50 3 4.5 3.8 Parque Indus- AP - FP - MP 6 12 3 3.5 1.8 trial SUBTOTAL   18 62.0 6.0 — 5.6 TOTAL             Average   8 30 3 4.0 3.1 Total   31 121 10 — 12.3 Source: SAGUAPAC 2014a. Note: AP = anaerobic pond; FP = facultative pond; MP = maturation pond. Both plants were equipped in 2009 with 2 mm rotary For details of actual wastewater characteristics and sieves for preliminary treatment. This was done to other parameters, see tables CS4-3 and CS4-4. For eliminate a maximum of coarse solid materials before details regarding general design recommendations for the wastewater reached the APs (SAGUAPAC 2014a). APs, see table CS4-12. Figure CS4-1: Covered APs with and without biogas at Santa Cruz WWTPs Sources: www.saguapac.com.bo; SAGUAPAC 2008. ANNEXES WASTEWATER TO ENERGY   104 Figure CS4-2: System North of Santa Cruz WWTPs Source: Google Earth. Figure CS4-3: System East of Santa Cruz WWTPs Source: Google Earth. Melbourne, Australia The history of Melbourne Water’s Western Treatment Plant goes back to 1897, when an infiltration facility Melbourne has two WWTPs—“Western Treatment was put into operation. Later, in 1936, the first Plant” and “Eastern Treatment Plant”—which serve treatment lagoon was added to act as a polishing 1.6 and 1.5 million people, respectively. While system for the infiltration effluents. In the following Eastern Treatment Plant, started up in 1975, is a years more lagoons were added and upgraded. In CAS system with a tertiary treatment stage, Western 1983, the network of lagoons and wetlands was Treatment Plant has a much longer history and a declared a Ramsar site, internationally renowned for unique treatment concept. ANNEXES WASTEWATER TO ENERGY   105 bird watching. The first modern lagoon was installed While 115E is a mere pond system, both 25W in 1986 (Hodgson and Paspaliaris 1996). Today all and 55E also include CAS systems for enhanced wastewater is treated through modern lagoons, and nutrient removal, commissioned in 2004 and 2001, the last filtration paddocks were closed in 2004. respectively. These CAS plants each receive effluents from pond number 4 and then discharge their effluents Nowadays, Western Treatment Plant actually into pond number 5 of their respective series of ponds comprises three WWTPs: the 25W, 55E, and 115E (Melbourne Water 2014). lagoon systems. Two of these plants—25W and 55E—receive raw wastewater, while 115E receives The key features of these plants are summarized in partially treated effluents from pond number 4 of table CS4-2, and some pictures are provided in figures 55E. Hence, only 25W and 55E currently operate a CS4-4 and CS4-5. covered anaerobic pond, or “anaerobic pot,” as it is referred to by the operator. Table CS4-2: Key characteristics of WWTPs in Melbourne, analyzed for case study 4 WWTP Technology Total ponds Anaerobic ponds water     number area number area depth     (no.) (ha) (no.) (m) (ha) Western Treatment Plant           25W AP - MAP - CAS - MP 11 264 1 2.9 8.0 55E AP - MAP - CAS - MP 11 206 1 2.1 9.3 115E MP 11 200 n.a. n.a. n.a. Average   11 223 1 2.5 8.7 Total   33 669 2 5.0 17.3 Source: Melbourne Water. Note: AP = anaerobic pond; CAS = conventional activated sludge; MAP = mechanically aerated pond; MP = maturation pond. n.a. = not available. Figure CS4-4: Western Treatment Plant, Melbourne Source: Google Earth. ANNEXES WASTEWATER TO ENERGY   106 Figure CS4-5: Covered APs at Western Treatment Plant, Melbourne Source: www.melbournewater.com.au. For details of actual wastewater characteristics and • Waste activated sludge (WAS) from the CAS systems other parameters, see tables CS4-3 and CS4-4. For Removed sludge is discharged into twenty-four sludge details regarding general design recommendations for drying pans (drying area = two hectares each), loaded APs, see table CS4-12. at 400 tons DS/ha. The consequent sludge drying capacity equals 19,200 tons DS/y. This is just about 4.1.3. Sludge management 50 percent of the sludge production. Therefore, a Santa Cruz, Bolivia significant proportion of sludge accumulates in the The sludge from anaerobic ponds has never been ponds. Work is currently underway to remove sludge removed to date (SAGUAPAC 2014b). This is an accumulated in lagoons and implement a new sludge astonishing feature, since the covers were already management strategy (refurbishment of old/disused installed in 2006–7, and a typical sludge removal drying pans, to enable loading of 750 tons DS/ha in interval for APs would be one to three years, at most. these drying pans) that will increase the sludge drying Apparently, the two-millimeter sieves installed prior capacity to 48,000 tons DS/y. During drying, not to the AP are having a positive impact on sludge only is water evaporation taking place, but about 20 removal intervals. percent of DS is also destroyed. Melbourne, Australia After drying, the sludge is stored in stockpiles. These are almost at capacity and will be reconfigured to contain Sludge originates from two sources (Melbourne an additional five years’ worth of solids production. Water 2014): Research into beneficial reuse of sludge is ongoing. • Anaerobically digested primary sludge from the 25W and 55E anaerobic ponds, including also For details of actual sludge characteristics, see table settled grit and screenings CS4-4. ANNEXES WASTEWATER TO ENERGY   107 4.1.4. Energy management • Flare Santa Cruz, Bolivia • Pipes HDPE, valves, and so on, as required While the covers were already installed in 2006–7, the • Control system flaring of biogas was installed later and only went into The biogas utilization project in the pipeline foresees operation in June 2009. For each of the two sites, there the following (Ghetti 2013): is a separate flaring station; these are called AT500 for System North and AT1000 for System East. • Co-generation: • AT500: two co-generators with 280 kWelectric each The only problems that have been observed are (a) (total = 560 kWelectric) inefficient condensate removal, which is particularly important due to the long biogas pipes to the flare, • AT1000: five co-generators with 280 kWelectric each and (b) air intrusion into the biogas system. Since gas (total = 1400 kWelectric) is collected by a suction system, the problem is not • TOTAL = 1,960 kWelectric so much the loss of biogas, but the reduced methane content in the gas. The problem was solved by • Electric installations; transformer for supply to thermo-welding of small fissures in the plastic cover, public grid implemented by a local company. • Gas scrubber At present, a project is in the pipeline to install a CHP For details of actual biogas characteristics, see table facility at each of the two sites to produce electric CS4-5. power. Melbourne, Australia Each of the systems actually available, both related Biogas from the 25W and 55E anaerobic ponds is sent to AT500 and AT1000, consists of the following to a power generation plant owned and operated by (Libhaber and Orozco-Jaramillo 2012; Libhaber the local energy company, AGL. 2010): About 99 percent of the electricity generated is used • Cover, made from HDPE, 1.5 mm, UV resistant, to operate the WWTP. The ratio between the plant’s 0.94 g/cm3 own production and purchased electric power is about • Floaters, counterweights, and inspection openings 3:1. The biogas thus supplies about 75 percent of the • Foam trap total WWTP’s electricity. Just a tiny 1 percent is also supplied to the public grid. • Condensate removal In fiscal year 2012–13, more than 10 percent of the • Suction compressors produced biogas was still flared. This is attributed to • Scrubbing tower (only for industrial WWTP, for a power station that is sized to maximize return on H2S removal) investment. Hence, particularly in summer when • Flow meter ANNEXES WASTEWATER TO ENERGY   108 biogas production is above average due to elevated WWTP. Table CS4-4 provides information on actual temperatures, some gas cannot be utilized and is flared. wastewater influent and effluent quality and sludge production. All numbers in both tables are based on Table CS4-13 summarizes general design private communications received from the WWTP recommendations for the biogas/energy component operators and relate to the year 2013. The presented of covered APs. quality data on Santa Cruz’s four facilities are flow 4.2. Analysis weighted, since those four plants treat different catchments with slightly different wastewater qualities. 4.2.1. Wastewater influent, effluent, and other parameters of interest Table CS4-3 provides key flow and load characteristics of the Santa Cruz WWTPs and Melbourne Western Table CS4-3: Key general characteristics of Santa Cruz WWTP and Melbourne Western WWTP, average data from the year 2013       Sta. Cruz Melbourne GENERAL         Daily flow rate average m /d3 118,000 482,000   max. month   127,000 560,000 BOD5 load average kgBOD5/d 48,000 274,000   max. month   61,000 300,000 Sources: SAGUAPAC and Melbourne Water—private communications 2014. ANNEXES WASTEWATER TO ENERGY   109 Table CS4-4: Actual influent and effluent data and sludge production of Santa Cruz WWTP and Melbourne Western WWTP, average data from the year 2013 Sta. Cruz Melbourne       WWTP WWTP Number of WWTPs   4 2 Pop. equivalents avg. actual PE60 802,000 4,569,000   max. month PE60 1,023,000 4,994,000 WASTEWATER QUANTITY       Specific wastewater production m3/PE60/y 54 39     L/PE60/d 147 105 WASTEWATER QUALITY       COD Influent mg/L 946 1,009   Effluent mg/L 197 32   Elimination % 79 97 BOD5 Influent mg/L 407 571   Effluent mg/L 60 4   Elimination % 85 99 Ntotal Influent mg/L 92 73   Effluent mg/L 66 21   Elimination % 28 72 NH4-N Effluent mg/L 2 5 NO3-N Effluent mg/L 0 15 Ptotal Influent mg/L 15.2 10.5   Effluent mg/L 4.4 9.0   Elimination % 71 14 SLUDGE PRODUCTION       Total annual DS production (tons DS/y) n.a. 37,000 DS content at removal from ponds (% DS) n.a. 6.55 Specific sludge volume produced (L/PE60/y) n.a. 124 Specific DS load produced (gDS/PE60/d) n.a. 22 Sources: SAGUAPAC and Melbourne Water, private communications 2014. Note: The Melbourne effluent values are from 25W and 55E. 1 cap = 40 g BOD5/d in Bolivia and 60 g BOD5/d in Australia; 1 PE60 = 1.50 PE40 = 1.50 cap in Bolivia and 1.0 cap in Australia; m3/PE60/y x 1.50 = m3/cap/y in Bolivia, and m3/PE60/y x 1.0 = m3/cap/y in Australia; L/PE60/d x 1.50 = L/cap/d in Bolivia, and L/PE60/d x 1.0 = L/cap/d in Australia. n.a. = not available. ANNEXES WASTEWATER TO ENERGY   110 4.2.2. Biogas production and potential for some general parameters that are of interest for the energy generation assessment of the collected biogas. Figure CS4- Biogas production 6 presents Santa Cruz’s annual biogas and CH4 fluctuation during the years 2009, 2010, and 2013. Table CS4-5 summarizes biogas data from both WWTPs of this case study. Furthermore, it contains Table CS4-5: Biogas data of Santa Cruz WWTP and Melbourne Western WWTP Sta. Cruz Melbourne 2010 2013 2012–13 Average wastewater temp. (°C) n.a. 27.0 20.8 Retention time in AP (d) 15 12 1 BOD5 in raw wastewater (mg/L) 430 407 571 COD in raw wastewater (mg/L) 972 946 1,009 COD/BOD5 (—) 2.3 2.3 1.8 COD removal efficiency (%) 63 n.a. 41 BIOGAS COLLECTION L/PE60/d 20.6 24.6 13.1 L/kgCODdestroyed 238 n.a. 297 CH4-content (%) 64 56 79 METHANE COLLECTION L/PE60/d 13.2 13.7 10.3 L/kgCODdestroyed 153 n.a. 235 Source: Authors’ calculation. n.a. = not available. Figure CS4-6: Daily biogas production and biogas CH4 content at Santa Cruz WWTP Source: SAGUAPAC. ANNEXES WASTEWATER TO ENERGY   111 Discussion: directions when comparing the two plants. While the former parameter presents Santa Cruz as the • When covering anaerobic ponds, it takes several plant with higher CH4 yield (13.7 versus 10.3 L/ months until biogas collection reaches steady state PE60/d), the latter suggests Melbourne has a higher conditions. In Santa Cruz it took half a year, from CH4 yield (153 versus 235 L/kgCODdestroyed). None March to August 2009, to reach that point. of the various parameters analyzed (wastewater • Biogas quantity always should be considered temperature, retention time in AP, BOD5 and together with its CH4 content. While this holds COD characteristics, or removal rates) provides an generally true for all biogas systems, it appears that explanation. covered ponds are especially susceptible to changes Conclusions: in CH4 content. The example from Santa Cruz demonstrates this quite clearly. It has been reported • For APs, it is recommended to prefer CH4 yields there that, due to small leaks, air is sucked into the over biogas yields, because intrusion of air can biogas. Even though these leaks have been repaired, significantly change the total biogas quantities. overall methane content to have fallen by about 20 • A suitable estimate can be obtained interpolating percent, from 64 percent CH4 in 2010 to 56 percent a CH4 yield of 10–14 L/PE60/d for average annual CH4 in 2013. wastewater temperatures of 21–27°C. • The increase of biogas quantities in Santa Cruz by 40 it is possible to use a CH4 yield of 350 • Alternatively, percent from 2010 to 2013 is thus explained as 20 L/kgCODdestroyed and assume losses of 40 percent for percent caused by air intrusion and the remaining concentrated wastewater with COD = 1000 mg/L. 20 percent attributed to an overall influent load For thinner wastewater, the loss assumption should increase of 20 percent in the same period. be increased according to dilution. • Furthermore, it is particularly important to distinguish between biogas production and biogas Potential for energy generation collection. As has been shown in case study 3 Based on the above-recommended CH4 yield of 10– (UASB), a considerable portion of the produced 14 L/PE60/d, a calorific value of CH4 of 10 kWh/m3, biogas is lost, either dissolved in the liquid effluent and an electric CHP efficiency of 30–35 percent, the or through imperfections in the collection system. expected electricity generation potential of covered AP Stoichiometrically, 1 kgCODdestroyed produces 350 L systems is about 11–18 kWh/PE60/d. CH4. The CH4 collected is 153 L CH4/kgCODdestroyed in Santa Cruz and 235 L CH4/kgCODdestroyed in At Santa Cruz WWTP, this electricity generation Melbourne. Hence, the losses observed at the potential is definitely higher than the energy WWTPs of case study 4 are 56 percent and 33 requirements for the existing pond system. No energy percent, respectively. consumption data were available for this case study, yet wastewater pumping is the main electric power • An interesting feature is that the CH4 yields for consumer. It is estimated that this requires about 0.5– L/PE60/d and L/kgCODdestroyed point in different 4 kWh/PE60/y, as demonstrated in figure CS3-7 for a ANNEXES WASTEWATER TO ENERGY   112 wide range of scenarios. Consequently, it is important to be operated in addition to the ponds. Figure CS4- to find an attractive use for this surplus. 7 presents the electricity balance for these conditions. (Note: The value for “kWh/y [potential]” includes the At Melbourne Western WWTP, the situation is flared gas.) different, since two energy-intensive CAS systems have Figure CS4-7: Power generation potential of actual biogas production, compared to actual electricity consumption at Melbourne Western WWTP in 2013 Source: Melbourne Water. Note: Data refer to 2013. Conclusions: only sufficient for partial electricity coverage, on the order of approximately 50–70 percent (for example, • The electricity generation potential from covered Melbourne Western WWTP). Note that the energy APs is about 11–18 kWh/PE60/d. consumption of this combination of covered AP + • This potential is sufficient to cover the needs of CAS requires less electric power than conventional wastewater systems with low energy consumption, CAS systems with PST (enhanced mechanical such as WSP (for example, Santa Cruz WWTP) or carbon removal, no separate sludge digester). TF (see case study 4). Table CS4-6 presents a summary of key biogas and • In cases of more energy-intensive treatment power generation parameters. technologies, such as subsequent CAS systems, the electricity generation potential from covered APs is ANNEXES WASTEWATER TO ENERGY   113 Table CS4-6: Biogas and power generation potential of covered AP at Santa Cruz WWTP and Melbourne Western WWTP   Sta. Cruz Melbourne Biogas production     - N elimination (L/PE60/d) — — - C elimination (L/PE60/d) 24.6 13.1 Electric efficiency CHP (%) 30 35 Thermal efficiency CHP (%) 50 50 Calorific value of biogas (kWh/m ) 3 5.6 7.5 Power generation     - N elimination (kWh/PE60/year) — — - C elimination (kWh/PE60/year) 15.1 13.2 Thermal energy generation     - N elimination (kWh/PE60/year) — — - C elimination (kWh/PE60/year) 25.1 18.8 Total energy generation     - N elimination (kWh/PE60/year) — — - C elimination (kWh/PE60/year) 40.1 32.0 Source: Authors’ calculation. Note: L/PE60/d x 16.67 = L/kg BOD5/d; kWh/PE60/y x 16.67 = kWh/kg BOD5/y. The biogas and electricity potential of Melbourne WWTP is assigned to “C elimination,” since it is not connected to the N removal of the downstream CAS system. 4.2.3. Operation capacity needs, biogas safety quantities of biogas involved at Santa Cruz WWTP, The key operational and safety issues are considered it was proposed by a study to install an electric power similar to those of UASB reactors: safety concerns, generation capacity of 1,960 kW (Ghetti 2013). deposits in the reactor, insufficient biogas treatment, The electricity potential of the collected biogas is larger and scum formation. Hence, for details see case study than the consumption at the WWTPs in question. 3, section 3.2.3. Therefore, SAGUAPAC intends to utilize the surplus electricity in water supply, which also falls under its 4.2.4. Institutional aspects, energy costs responsibilities. To supply the autogenerated electric Bolivia power from the two WWTPs to the water supply sites Any physical or legal entity has the right to produce requires using the public grid for transportation. Even and consume its own electric energy in Bolivia. As though it has never been done in Bolivia before, it is long as the installed electric power does not exceed legally possible to pay the electrical distribution utility 2000 kW, doing so requires only registration with the for the use of its network to transport SAGUAPAC`s “Autoridad de Fiscalización.” If the installed power electricity from its generation point (the anaerobic exceeds 2000 kW, the applicant requires a power lagoons) to other consumption points (water treatment generation license, which is more complicated and and pumping systems). In other words, SAGUAPAC costly to obtain. Given these legal boundaries and the would not sell the electricity it produces to the ANNEXES WASTEWATER TO ENERGY   114 electrical utility; it would pay for its transportation. 4.2.5. GHG reduction and CDM co-financing The electricity generation project currently in the pipeline thus provides for all necessary installations General for that purpose and includes an estimate for the cost For general information on CDM co-financing, the of utilizing the public grid in its financial assessments price of carbon credits, and related matters, see section (Ghetti 2013). For results, see section 4.2.8. 1.2.5. Also interesting is the question of how anaerobic The electricity tariff in Bolivia is different at different wastewater treatment compares to aerobic wastewater times of day and features various surcharges. Currently, treatment in terms of GHG emissions; for details, see the effective unit cost for electric power is US$0.065/ section 3.2.5. kWh (EUR0.048/kWh), as cited by SAGUAPAC Specifics of case study 4 (2014b). The assumed future electricity cost savings are based on a tariff of US$0.085/kWh (EUR0.063/ The “additionality” criterion (see section 1.2.5) means kWh), since a considerable portion of the electricity that new construction of a covered anaerobic pond production from biogas is assumed during periods plus biogas collection and utilization are not eligible when the unit cost is peaking (Ghetti 2013). under CDM for their methane mitigation effects. In cases like this one, in which the project both creates Australia and eliminates part of the methane, only the energy The priority in Australia is on utilization of the production component from a renewable source that generated electricity onsite for the operation of replaces fossil sources may be considered for CDM the WWTPs. Just a tiny 1 percent of the electricity on the plus side, while the additional CH4 emissions production is supplied into the public grid. This generated by the project, but lost dissolved in the pond is understood to be a consequence of issues with effluents, have to be considered on the minus side. balancing power production with supply. The CDM situation is different for Bolivia, where Generally, the typical average unit cost paid most existing ponds are not covered (baseline by Melbourne Water is about US$0.09/kWh scenario) and thereby produce methane emissions to (EUR0.067/kWh). the atmosphere. In the case of Bolivia, CDM carbon credits would apply both to the methane mitigation and the renewable energy generation. The country-specific energy mixtures shown in table CS4-7 derive from the GHG emissions for electricity generation in Australia and Bolivia. ANNEXES WASTEWATER TO ENERGY   115 Table CS4-7: GHG emissions per kWh for electricity generation in Bolivia and Australia, as compared to the world Region 1990 2010 gCO2/kWh gCO2/kWh Bolivia 307 423 Australia 817 841 World 586 565 Source: IEA 2012. Note: The table shows CO2 emissions from fossil fuels consumed for electricity generation, in both electricity-only and combined heat and power plants, divided by output of electricity generated from fossil fuels, nuclear, hydro (excluding pumped storage), geothermal, solar, wind, tide, wave, ocean, and biofuels. Both main activity producers and autoproducers have been included in the calculation. Table CS4-8 provides a quantitative assessment of this case study. The numbers thus highlight roughly the CO2e emission reduction, assuming both the which orders of magnitude to expect for GHG renewable electricity generation in CHP and the CH4 emission reductions from these installations. emission reduction would apply to either WWTP of Table CS4-8: GHG emission reduction due to electric power generation and CH4 mitigation at Santa Cruz WWTP and Melbourne Western WWTP   Santa Cruz Melbourne Electricity potential or production in CHP (kWh/y) 11,970,000 49,456,000 Substituted fossil CO2 emissions (gCO2/kWh) 423 841 GHG reduction by electric power generation (tons CO2e/y) 5,063 41,592 Methane emission reduction through flare and CHP (m methane/ 3 year) 3,990,000 17,224,000 Specific weight of methane (kg/m3 methane) 0.72 0.72 GHG factor of methane (—) 21 21 GHG reduction by CH4 emission reduction (tons CO2e/y) 60,329 260,427 TOTAL GHG reduction (tons CO2e/y) 65,392 302,019 Source: Authors’ calculation. ANNEXES WASTEWATER TO ENERGY   116 4.2.6. CAPEX structure This summary contains both all investments made Santa Cruz to date in the context of the biogas recovery project (SAGUAPAC 2014a) and the investments still The CAPEX estimate for the biogas utilization planned for CHP (Ghetti 2013). project in Santa Cruz is summarized in table CS4-9. Table CS4-9: CAPEX for biogas project at Santa Cruz WWTP   CAPEX   US$ IMPLEMENTED   Influent sieves 2 mm 822,336 Civil works for sieves 221,427 Covers on APs 1,733,275 Biogas collection system 312,558 Biogas flaring system 501,996 Civil works for gas flaring system 11,050 Electricity supply to flares 48,742 Low energy mixers/aerators downstream of APs 1,147,790 Other 12,000 SUBTOTAL 1 3,663,384 PLANNED   CHP: 7 x 280 kW electric 1,286,447 Electric installations and transformers 300,320 Civil works 60,000 Gas filters 37,577 Installation works 21,000 Contingencies 170,534 SUBTOTAL 2 1,875,878 TOTAL 5,539,262 Sources: SAGUAPAC 2014a; Ghetti 2013. Note: The cost for the covers of US$1.73 million, combined with the total covered pond area of 12.3 ha, leads to a unit cover cost of about US$14.1/m2. Melbourne No CAPEX information available. ANNEXES WASTEWATER TO ENERGY   117 4.2.7. OPEX structure of the biogas recovery project (SAGUAPAC 2014a) and the additional OPEX expected once the planned Santa Cruz CHP is installed, taking into account that the surplus The latest OPEX estimate for the biogas utilization power will be transferred through the public grid to project in Santa Cruz is summarized in table CS4- other SAGUAPAC facilities consuming electricity 10. This summary contains both the OPEX increase (Ghetti 2013). due to the investments made to date in the context Table CS4-10: OPEX for biogas project at Santa Cruz WWTP   OPEX   US$/y IMPLEMENTED   Electric energy 5,000 Maintenance of equipment 6,000 Maintenance of covers 49,000 SUBTOTAL 1 60,000 PLANNED   CHP: O&M 131,057 Additional manpower, insurance, administration 65,809 Fee for using the public grid 51,129 Electricity savings –897,930 SUBTOTAL 2 –649,935 TOTAL –589,935 Sources: SAGUAPAC 2014a; Ghetti 2013. Consequently, while to date OPEX has increased by US$100,000 per year, after the installation of CHP it will be reduced by about US$590,000 per year once it is working. Melbourne No OPEX information available. ANNEXES WASTEWATER TO ENERGY   118 4.2.8. Viability of investment in biogas utilization Santa Cruz Table CS4-11 summarizes several cost indicators from case study 4. Table CS4-11: Cost indicators for biogas utilization project at Santa Cruz WWTP       Average influent load PE60,avg 802,000 Peak influent load PE60,max 1,023,000 CAPEX US$ 5,539,262 OPEX US$/y –589,935 specific CAPEX US$/PE60,avg 6.91   US$/PE60,max 5.41 specific OPEX US$/PE60,avg –0.74   US$/PE60,max –0.58 Source: Authors’ calculation. Note: 1 PE60 = 1.50 PE40 = 1.50 cap in Bolivia; US$/PE60 x 16.67 = US$/kg BOD5. Overall, without interest, total CAPEX is covered 4.3. Conclusions for covered ponds in EAP through reduced OPEX (simple payback) in about ten countries years. The main reason for the long return period is The key characteristics of anaerobic ponds (APs) the low cost of electricity of US$0.085/KWh. described in case study 4, such as BOD5 elimination, VS destruction, sludge characteristics, and biogas However, one could also argue that the covers were production, can be transferred from Australia and necessary anyhow for bad odor elimination. This Bolivia to EAP countries without major changes. approach was taken in Ghetti (2013), which only Based on similar temperature conditions, similar looked at CAPEX and OPEX for CHP and rendered project-specific results can be attained. the project as financially very viable. Hence, under the given circumstances, there is a clear incentive for A possible change relates to wastewater treatment SAGUAPAC to proceed with the implementation of requirements. The AP systems in Santa Cruz, Bolivia, the CHP component. were designed for carbon removal only; the ones in Melbourne were later upgraded with nutrient removal Melbourne CAS stages downstream of AP. If, in EAP, additional No assessment of financial viability has been possible nitrification is required, more efficient polishing stages for Melbourne, since no CAPEX or OPEX information than just ponds will be needed. This will generally was disclosed by the operator. push up a WWTP’s CAPEX and increase its OPEX, ANNEXES WASTEWATER TO ENERGY   119 particularly where CAS is applied for nitrification. For Dilution of wastewater also has negative impacts polishing trickling filters (TFs), OPEX will stay the on the design of anaerobic ponds. One of the main same, if additional nitrification is required. After all, design parameters is retention time in those ponds. the pumping head does not change when more TFs Increasing flow rates thus lead to increasingly large need to be supplied to facilitate nitrification. (and more expensive) ponds. If denitrification is required as well, the additional The quantities of digested sludge will be lower in installations will, in any case, increase CAPEX, EAP than in Bolivia and Australia. This is particularly OPEX, and energy consumption further. This might expected where septic tanks continue to be used in eventually lead to a situation where, with nitrification large numbers. CAPEX is expected to be similar in + denitrification, less than 100 percent of electricity EAP to that in Bolivia, whereas it is expected to be consumption can be covered from biogas, even at higher in Australia. OPEX savings depend mainly on more efficient, large plants. The Melbourne case study electricity unit cost and may run anywhere from lower already shows that, if combined with CAS, about 50– to higher in EAP than in this case study. 70 percent of the overall energy consumption can be Success stories in EAP covered by the electricity from biogas. The Melbourne case also demonstrates a general problem: lack of In general, there are not many examples of covered carbon sources for denitrification after AP (how best APs worldwide. The two examples from case study 4, to solve this issue is still under consideration). This is in Melbourne (Australia) and Santa Cruz (Bolivia), are not to say that the AP and denitrification would not the best known plants of that type in Australia and work, but denitrification downstream of such efficient Latin America, respectively. BOD removal is generally challenging. It is questionable, though, if the same amounts of There have also been some reports on covered APs biogas can be collected when wastewater is diluted, as for piggery waste from New Zealand and on facilities is common in EAP. Due to the higher flow rate per in Vietnam: at two WWTPs in Da Nang (Phu Loc capita, a higher percentage of the generated methane WWTP and Ngu Hanh Son WWTP), existing can be lost dissolved via the liquid effluents, even if anaerobic ponds were covered as a result of public the same quantities of biogas are generated. This will concern over odor (World Bank 2013f). No biogas have a negative impact on the electricity generation is being collected, but interest exists to explore the potential, and it will increase the GHG emissions. It potential for wastewater-to-energy options. is also possible to assume a collectable biogas potential and, thus, electricity generation potential lower than the values indicated in case study 4, table CS4-6. ANNEXES WASTEWATER TO ENERGY   120 COVERED ANAEROBIC PONDS: TECHNICAL SUMMARY Table CS4-12 General design recommendations for covered anaerobic ponds AP DESIGN Retention time = >1 d, <5 d • Volumetric BOD load - <10°C: <100 gBOD5/m3/d - 10–20°C: <20 x T(°C) – 100 - 20–25°C: <10 x T(°C) + 100 • Sludge production: - 50–130 L/cap/y - 4–10 kgTS/cap/y • BOD5 removal efficiency: - <10°C: 40% - 10-25°C: 2 x T(°C) + 20 • COD effluent concentration: COD ≈ BOD5 x 2.5 • SS removal efficiency: 60–80 percent • N removal efficiency: 5 percent • P removal efficiency: 5 percent • FC effluent concentration: FCeff = FCinf / (1 + kT x t), with kT = 2.6 x 1.15(T-20), T: wastewater temperature (°C), t: hydraulic retention time (d) Sources: Based on FWT 2008; Sperling and Chernicharo 2005; Mara and Pearson 1998. ANNEXES WASTEWATER TO ENERGY   121 Figure CS4-8: Prevailing wastewater design temperature of ponds as function of air temperature Source: FWT 2008. The prevailing wastewater temperature can thus usually be assumed to be about 1–2°C higher than the coldest average monthly air temperature. AP SHAPE AND • Individual ponds should not be larger than 3 ha, to minimize wind impact. CONSTRUCTION • Length:width (at water surface) = 2:1, or larger • Water depth = 2.0–5.0 m • Internal embankment slopes with 1:2–1:3, external embankments with 1:1.5–1:2 • Minimum freeboard: 0.5 m (pond <1 ha), 0.5–1.0 m (pond with 1–3 ha) • Positioning: flow direction counter-current to dominant wind direction • If possible, install at least two parallel ponds for revision, and so on and bypass. • Inlet: always below water surface and above sludge level—for example, at mid-water depth • Outlet: scum wall 30 cm submerged into the water • Outlet flow velocity: <0.1 m/s at the most unfavorable point • Positioning: inlet and outlet in diagonally opposite corners of the pond • Bottom lining: clay, or plastic sheet (mostly HDPE, PVC) ANNEXES WASTEWATER TO ENERGY   122 SLUDGE • Sludge removal: mostly done with heavy machinery or with slurry pumps REMOVAL • Should be considered prior to construction PLASTIC COVER • Specialized companies for production and installation • Material: mostly HDPE, 1.5–2.5 mm, but also foam layer in the middle of two outer plastic layers (XR-5); polypropylene (PP) has been used. • UV light resistance • Anchored around the pond perimeter • Weighted pipes and so forth are used to hold the cover down. • Biogas migration below the cover must be possible. • Consider water level fluctuations in the pond. • Stormwater has to be drained safely. Some systems collect and remove the stormwater; others drain it into the pond water. • Typical lifespan: ≈ 20 years. Sources: Based on Libhaber and Orozco-Jaramillo 2012; Heubeck and Craggs 2009; NIWA 2008; DeGarie et al. 2000; Melbourne Water (www.melbournewater.com.au). ANNEXES WASTEWATER TO ENERGY   123 Table CS4-13: Covered anaerobic ponds: General design parameters and key characteristics of energy management BIOGAS • Biogas quantity DESIGN General: Few design references exist on biogas production in covered anaerobic ponds. The following approaches are mostly used for estimating the biogas potential in covered APs for municipal wastewater: (i) Biogas yield from influent BOD5 load: Daily biogas production (m3/d) = 0.3 x kgBOD5,influent/d (Libhaber and Orozco-Jaramillo 2012). The result from this calculation may be considered the collectable biogas quantity. Note: The above 300 L gas/kgBOD5 is equivalent to 150 L gas/kgCOD; assuming 60 percent COD destruction in the AP, this is equivalent to 250 L gas/kgCODdestroyed; based on 70 percent methane, this is 175 L CH4/kgCODdestroyed; compared to the theoretical maximum (see below) of 350 L CH4/kg CODdestroyed, this implies a biogas loss assumption of 50 percent. (ii) Biogas from COD load destroyed in AP: Daily biogas production (m3/d) = 0.350 x kgCODdestroyed/d. This is a stoichiometric relationship. It describes how much biogas is produced, but it does not say anything about how much of this biogas can indeed be collected. After all, much biogas is lost dissolved in the effluent. With a loss assumption of 50 percent, the collectable daily biogas quantity thus is 0.175 x kgCODdestroyed/d. (iii) Biogas from influent VS load: This approach is not recommended for municipal wastewater since it does not include the considerable dissolved influent organic load. It is more appropriate for industrial applications—for instance, in piggeries—where the bulk of the influent COD originates from VS load (Heubeck and Craggs 2009; NIWA 2008). Case study 4: The following has been found in this case study: (i) Biogas yield from influent BOD5 load: Collected methane = 9.8–13.7 L CH4/PE60/d. That is equivalent to 160–230 L CH4/kgBOD5. Based on an average CH4 content of 70 percent follows: Daily biogas production (m3/d) = (0.23 to 0.33) x kgBOD5,influent/d. (ii) Biogas from COD load destroyed in AP: Collected methane = 150–220 L/kgCODdestroyed. Compared to the above cited maximum of 350 L/ kgCODdestroyed, the losses thus amount to about 35–60 percent. ANNEXES WASTEWATER TO ENERGY   124 • Biogas quality A typical CH4 content from biogas of covered AP according to the literature is about 70 percent (Libhaber and Orozco-Jaramillo 2012). This case study found a range of 56–75 percent for CH4 content. BIOGAS TREATMENT See table CS1-5. GAS HOLDER (Not required; sufficient storage capacity below covers) FLARE See table CS1-5. BIOGAS UTILIZATION See table CS1-5. ANNEXES WASTEWATER TO ENERGY   125 CASE STUDY 5: CO-DIGESTION OF ORGANIC WASTE ANNEXES WASTEWATER TO ENERGY   126 5.1. BACKGROUND, PROCESS DESCRIPTION 5.1. Background, process description will present both initial results without co-digestion and subsequent results with co-digestion, making 5.1.1. Data sources even more apparent the drastic changes of energy Co-digestion of organic waste in sludge digesters has management due to co-digestion. been practiced in hundreds of WWTPs for many years. In Europe, this concept has been used for As long as mesophilic temperature conditions (30– more than three decades (WERF 2010b; Schmelz 38°C) are maintained in a digester and conventional 2000). With the recent trend toward maximizing design criteria are in place, the results from this case biogas yields and electricity production at WWTPs, study, and from the additional references cited in this technology has gained rapid and ever more this section, can be transferred to any other location widespread popularity in developed countries. worldwide. Indeed, it has evolved into the means most often used for substantially increasing the conventional 5.1.2. Wastewater management biogas yield of sludge digesters. For instance, in the Since 2005, wastewater treatment at Zirl WWTP United States, both WERF (2012a) and Willis and has been based on a classical CAS system, designed Stone (2012) recommend the “use [of ] alternative for enhanced P and N elimination, complete with feedstocks to increase biogas production” as a key preliminary treatment (screen, sand/fat removal), component in overcoming barriers to biogas use. followed by primary sedimentation tank (PST), bioreactor, and secondary sedimentation tank (SST). The traditional feedstocks used for co-digestion have been FOG (fats, oils, and grease) from fat traps from There was no need for a change of process technology restaurants and canteens and food waste from food in wastewater management due to the introduction service establishments. Nowadays, municipal organic of co-digestion. The only aspect to consider was the waste is also increasingly being used, as well as specific additional nitrogen (N) load in the filtrate from sludge industrial wastes, particularly from food industries. dewatering, caused by the degradation of the extra organic feedstocks. Yet, as it turned out, these additional Case study 5 presents the experiences of a medium- aeration requirements could be compensated for by sized municipal WWTP located in Zirl, Austria. It optimizing the automated control of aeration in the has a design capacity of 61,500 PE60 and 13,600 m3/d bioreactors. (www.avzirl.at). Its startup was in 1996 as an extended aeration (EA) facility. In 2005 it was equipped with Figure CS5-1 presents two pictures from the plant. mesophilic digesters, and wastewater treatment For details of actual wastewater characteristics and switched to CAS. Co-digestion started soon after, in other parameters, see tables CS5-1 and CS5-2. 2007, and has been practiced to date. The case study ANNEXES WASTEWATER TO ENERGY   127 Figure CS5-1: Two different views of Zirl WWTP Source: Abwasserverband (AV) Zirl et al. 5.1.3. Sludge management digestion is balanced in a gas holder, treated, and Sludge treatment at Zirl WWTP is based on finally utilized in CHP for the combined production a conventional concept, as presented in figure of electric and thermal energy. CS5-2: both primary and secondary sludge from Organic waste that is collected by a private company is wastewater treatment is thickened (gravimetrically co-digested in the WWTP’s single sludge digester. No and mechanically, respectively) and then digested in relevant changes were made to the sludge/biogas train a mesophilic digester at 35°C (net volume = 1,350 for that purpose. Only minor adjustments of existing m3). The digested sludge is dewatered in a screw press installations to serve optimal takeover and feeding of and then disposed of by an external company. Both organic waste were necessary. incineration and composting are currently applied to the digested and dewatered sludge. Biogas from ANNEXES WASTEWATER TO ENERGY   128 Figure CS5-2: Simplified flow scheme of co-digestion at Zirl WWTP Source: Authors. Several types of organic waste are being or have been co- All other organic feedstocks (from bakery, pizza digested at Zirl WWTP: producer, grocery stores’ waste, and municipal organic waste) undergo more intensive pretreatment. The • FOG (fats, oils, and grease) following installations are available for that purpose: • Leachate from composting (only from 2007 to tank: Trucks deliver the various types of • Reception 2010) organic waste and dump it into this tank. From • Biowaste from an industrial bakery here a spiral conveyor transports the waste to a • Biowaste from an industrial pizza producer hammer mill. • Packed grocery stores’ waste • Hammer mill: Manufactured by Wackerbauer, Germany, this mill is used for crushing coarse • Municipal organic waste organic material and reducing it in size and removing Management of FOG and of all other organic unwanted materials (in this case, plastics, wood, feedstocks is done separately. The former is buffered in stones, and metals, in particular). Liquors are added a heated holding tank to avoid solidification and then as necessary to obtain a suspension that can be easily constantly fed to the digester. The FOG installations pumped. Practice has shown that for the wastes in include mainly a FOG holidng tank of 10 m3, question, suspensions with a maximum DS of about completed with heating and grinding + feeding system 15–20 percent are still manageable. The liquors used to digester. for slurrying are typically of organic waste origin, as ANNEXES WASTEWATER TO ENERGY   129 well—for example, leachate from composting, and/ the WWTP, where it is unloaded into the biowaste or whey. The maximum grain size after the hammer holding tank. See figure CS5-3 for the hammer mill mill usually is 8–10 mm. and unloading of pretreated organic waste at Zirl WWTP. Figure CS5-4 shows typical characteristics • In this specific case, the reception tank and hammer of pretreated food waste. mill are located not at the WWTP site, but at the premises of a nearby private company that collects • Biowaste holding tank: Located at the WWTP, 110 the waste for co-digestion. The treated product is m3, complete with mixing and feeding system to then pumped to a suction truck and transported to the digester Figure CS5-3: Hammer mill and organic waste unloading at Zirl WWTP Spiral conveyor Container for unwanted materials Hammer mill Source: Abwasserverband (AV) Zirl et al. Figure CS5-4: Liquefied food waste prior to co-digestion at Zirl WWTP Source: Abwasserverband (AV) Zirl et al. ANNEXES WASTEWATER TO ENERGY   130 The combined thickened primary and secondary sludge • CHPs: (a) co-generation no. 1 with 105 kW electric from wastewater treatment and the pretreated organic power, 165 kW thermal power; (b) co-generation wastes are co-digested in a single closed digester (1,350 no. 2 with 75 kW electric power, 145 kW thermal m3) at mesophilic temperatures of about 35°C. The power; total electric power = 180 kW. Electric digester is of cylindrical type (see figure CS1-22), with efficiency of no. 1 is 34.7 percent and of the older diameter:sludge depth ≈ 13.0:12.6 m. It is constructed no. 2 just 25 percent. out of concrete, featuring thermal insulation and • Pipes, valves, and so on as required external heat exchangers for temperature control. The The electricity produced from biogas is primarily used digester is equipped with biogas injection for mixing. for the WWTP’s own energy requirements. Only Digested sludge is conditioned with polymers, a small percentage of the generated electric power dewatered in a screw press to about 25 percent DS, and is supplied into the public grid when production is then transported to composting or to an incineration above consumption levels. facility for thermal reuse. Co-digestion at Zirl WWTP has not changed biogas For details of actual sludge and feedstock characteristics, characteristics. Typically, the calorific value of the gas see tables CS5-2 and CS5-3. was and continues to be about 6.2 kWh/m3—that is, For details on general design parameters and key the biogas has 62 percent CH4 content. characteristics of anaerobic digesters, see table CS1-4. For details on general design parameters and key Specific impacts of co-digestion on sludge management characteristics of biogas systems combined with are summarized in table CS5-7. anaerobic digesters, see table CS1-5. For general 5.1.4. Energy management design parameters relevant to energy management At Zirl WWTP, the biogas produced in the digester is with co-digestion, see table CS5-9. balanced in a gas holder, treated, and finally utilized 5.2. Analysis in CHP for the combined production of electric and thermal energy. Almost no gas is flared. The main 5.2.1. Wastewater influent, effluent, and other elements of biogas management are the following: parameters of interest • Foam trap Wastewater • Condensate removal Table CS5-1 provides key flow and load characteristics • Two-stage activated carbon filter for H2S and of Zirl WWTP. Table CS5-2 provides actual siloxane removal information on wastewater influent and effluent quality. All numbers in both tables relate to the year • Gas holder (400 m3) 2012. • Flare ANNEXES WASTEWATER TO ENERGY   131 Table CS5-1: Key general characteristics of Zirl WWTP, average data from the year 2012       Actual Design GENERAL         Daily flow rate average m3/d 9,616 —   85 percentile   12,700 13,600 Pop. equivalents average PE60 42,950 —   85 percentile   54,200 61,500 Source: Abwasserverband (AV) Zirl et al. Table CS5-2: Actual influent and effluent data and sludge production of Zirl WWTP, average data from the year 2012       Zirl WWTP Number of WWTPs   1 Pop. equivalents avg. actual PE60 42,950   85 percentile PE60 54,200 WASTEWATER QUANTITY     Specific wastewater production m /PE60/y 3 82     L/PE60/d 224 WASTEWATER QUALITY     COD Influent mg/L 504   Effluent mg/L 29   Elimination % 94 BOD5 Influent mg/L 282   Effluent mg/L 5   Elimination % 98 Ntotal Influent mg/L 35.0   Effluent mg/L 10.7   Elimination % 69 NH4-N Effluent mg/L 2.8 NO3-N Effluent mg/L 6.5 Ptotal Influent mg/L 6.5   Effluent mg/L 0.6   Elimination % 91 SLUDGE PRODUCTION      Total annual sludge production (tons dewatered sludge/y) 2,322 DS content after dewatering (percent DS) 25 Specific DS load produced (gDS/PE60/d)   37 Source: Abwasserverband (AV) Zirl et al. Note: 1 cap = 60 g BOD5/d in Austria; 1 PE60 = 1.0 cap in Austria; m3/PE60/y = m3/cap/y in Austria; L/PE60/d = L/cap/d in Austria. ANNEXES WASTEWATER TO ENERGY   132 Return load from dewatering filtrate dewatering. Figure CS5-5 presents the total annual N return load registered at Zirl WWTP. An important question regards the additional nitrogen (N) return load via filtrate from sludge Figure CS5-5: N return load in filtrate from sludge dewatering at Zirl WWTP Source: Abwasserverband (AV) Zirl et al. Apparently, the N return load increased from about assessed on a case by case basis. twenty to thirty-three tons N/year between 2007 and At Zirl WWTP, there has been an increase in N return 2012. As a percentage of influent N load, the N return load, which was compensated for in practice by an load increased from about 17 percent to 30 percent. improved aeration control and automation system. Relatively speaking, N return load thus increased Notwithstanding, for fair comparison, the cost impact by 75 percent at Zirl WWTP. This is a rather large of this increased N return load must be considered increase, the size of which has not been found in all and is part of the OPEX assessment in section 5.2.7. the case studies. Svardal and Haider (2010) report on five WWTPs in the same region where co-digestion Sludge led to a wide variation of impacts on N return load, The main types of feedstock to the digester at Zirl ranging from no impact to a maximum increase of 14 WWTP are the following: percent, whereas Nowak and Ebner (2013) report on a case study where return N load increased by more • Sludge: Primary sludge (from gravity thickening after than 100 percent. In this last case, all raw co-digestates primary sedimentation) + WAS (from mechanical featured a high protein and nitrogen content. thickening after secondary wastewater treatment) In conclusion, generalizations are not possible. • Waste: FOG + organic waste mixture + leachate The additional N load always needs a case-specific from composting assessment. Thus, the impact on aeration requirements Table CS5-3 presents the typical characteristics of in secondary wastewater treatment also has to be these feedstocks. ANNEXES WASTEWATER TO ENERGY   133 Table CS5-3: Typical characteristics of feedstocks used at Zirl WWTP   SLUDGE WASTE Primary Organic waste Leachate from   WAS FOG sludge mixture composting DS (%) 5.4 6.1 6.0 18.3 5.0 VS (%) 4.4 4.4 5.7 15.5 2.5 COD (mg/L) 85,000 62,000 127,000 291,000 74,000 COD/VS (—) 1.9 1.4 2.3 1.9 2.9 Source: Abwasserverband (AV) Zirl et al. In terms of the distribution of VS and DS load added, the sludge represents 85 percent of loading. added to the digester, sludge (PS + WAS) makes up Figure CS5-6 shows the average VS, DS, and volume about two-thirds and organic waste is responsible for distribution for the year 2012. about one-third of total input. In terms of volume Figure CS5-6: Distribution of VS, DS, and volume input to co-digestion at Zirl WWTP in 2012 Source: Abwasserverband (AV) Zirl et al. ANNEXES WASTEWATER TO ENERGY   134 The additional waste feedstock produced an increase The two key loading parameters of the digester are the in dewatered sludge quantities by about 10 percent, typical retention time including waste, which was >20 which was attributed to the remains of waste feedstock days; and the organic loading equaled about 2.5 kg after digestion. This increase is less than what should VS/m3/d. Both those values are thus within the range be expected from calculations of sludge production, recommended in table CS1-4. according to table CS5-7. Apparently, the mutual digestion of sewage sludge and organic waste has 5.2.2. Biogas production and potential for enhanced the biological activities in the digester at energy generation Zirl WWTP, just as has been observed from other Biogas production case studies. Consequently, both sludge and biowaste are more efficiently degraded, and overall sludge Figure CS5-7 shows biogas production in the digester production is lower (and overall biogas yield higher) at Zirl WWTP. Startup of the digester was during than the sum of individual digestion of the same calendar year 2005. Hence, biogas production from substances. Regarding polymer consumption for sludge reached its full capacity without co-digestion sludge dewatering, the results are inconclusive: over only in 2006. In 2007 and 2008, only small quantities the years, specific polymer consumption has fluctuated of organic waste were co-digested. Consequently, within a range of 10–17 g polymer/kgDS. No clear biogas production increased, but not too much by attribution of higher or lower values to the impacts of then. Since 2009, large solids loads of organic waste co-digestion is possible. have been co-digested, and more biogas has been produced from this waste than from sludge ever since. Figure CS5-7: Daily biogas production at Zirl WWTP Source: Abwasserverband (AV) Zirl et al. ANNEXES WASTEWATER TO ENERGY   135 Potential for energy generation The increased biogas quantities changed the potential for electricity generation drastically. Figure CS5-8 demonstrates this clearly. Figure CS5-8: Power generation from biogas, compared to electricity consumption at Zirl WWTP Source: Abwasserverband (AV) Zirl et al. Initially, electricity generation from Zirl WWTP’s When analyzing the biogas yield as a function of input sludge sources covered only about 40 percent of VS, it turns out that biogas production is, in fact, higher its electricity consumption. Yet ever since 2009— than expected. Hence, a similar conclusion can be when co-digestion increased greatly—more electric drawn as that previously drawn for sludge production: power has been produced from biogas than the plant the mutual digestion of sewage sludge and organic consumes. It is worthwhile mentioning that the waste enhances the biological activities in the digester; operator is intentionally not increasing electricity consequently, both sludge and biowaste are more production further, since feed-in tariffs for electricity efficiently degraded, and overall sludge production is supplied to the public grid are low in Austria, and lower (and overall biogas yield higher) than the sum financially unattractive. Thus, the operator’s main of individual digestion of the same substances. A clear target is full electricity coverage of the WWTP’s needs. quantification of the effect for each individual input is not possible with the data available, but the overall trend is beyond doubt. ANNEXES WASTEWATER TO ENERGY   136 The specific electricity consumption at Zirl WWTP The specific biogas production prior to co-digestion in recent years has equaled about 25 kWh/PE60/y. was 17.5 L biogas/PE60/d. This value is also in line This is a good result as compared to average results with typical values presented in figure CS1-2 and in Europe, according to case study 1, but it is in line table CS1-2. With co-digestion, this value increased with actual recommendations for this plant size and N to about 45–50 L/PE60/d. elimination treatment target (compare annex 1, table Table CS5-4 presents a summary of key biogas and A-1: consumption target = 30 kWh/PE60/y; ideal = 23 power generation parameters. kWh/PE60/y). Table CS5-4: Biogas and power generation potential of co-digestion at Zirl WWTP   Biogas production and Power Retention time in PST (h)   Generation 0.5–1.0 Biogas production   - N elimination (L/PE60/d) 17.5 (sludge) + 30 (waste) - C elimination (L/PE60/d) — Electric efficiency CHP (%) 25 and 35 Thermal efficiency CHP (%) 50 and 55 Calorific value of biogas (kWh/m ) 3 6.2 Power generation   - N elimination (kWh/PE60/year) 11 (sludge) + 19 (waste) - C elimination (kWh/PE60/year) — Thermal energy generation   - N elimination (kWh/PE60/year) 18 (sludge) + 32 (waste) - C elimination (kWh/PE60/year) — Total energy generation   - N elimination (kWh/PE60/year) 29 (sludge) + 51 (waste) - C elimination (kWh/PE60/year) — Source: Authors’ calculation. Note: L/PE60/d x 16.67 = L/kg BOD5/d; kWh/PE60/y x 16.67 = kWh/kg BOD5/y. Zirl WWTP is not unique in its biogas production percent, on average. And many of the references from co-digestion; other WWTPs report similar cited, together with tables CS5-7, and CS5-9, present results. For instance, Schwarzenbeck and others further successful examples, albeit frequently with (2008) report on a municipal WWTP that increased smaller quantities of biowaste involved and thus less its electricity production with co-digestion to 113 pronounced increases of electricity coverage. ANNEXES WASTEWATER TO ENERGY   137 5.2.3. Operation capacity needs, biogas safety 5.2.6. CAPEX structure All issues generally associated with biogas safety and CAPEX requirements for the co-digestion project operation capacity needs have already been discussed at Zirl WWTP have been almost zero. Particularly in case study 1. For details, see section 1.2.3. since the pretreatment of the waste was to be carried No particular extra capacities or precautions are out by a private company, not much remained to required with co-digestion of organic waste. The only invest at the WWTP; the biowaste holding tank exception is that, in pretreatment, any substances and its periphery installations were already in place. that may settle in the digester need to be properly Therefore, only minor adjustments were made to the eliminated and/or reduced in size so they will leave existing infrastructure, implying a CAPEX of less than the digester together with the digested sludge. US$6,800. CAPEX requirements cannot be standardized for 5.2.4. Institutional aspects, energy costs other plants, either. Very much depends on the specific The institutional energy aspects relevant in Central organic waste, its specific pretreatment requirements, Europe have already been discussed in case study 1 for and the specific installations available that can be Austria, where Zirl WWTP is located. For details, see utilized. Still, the following thumb numbers can offer section 1.2.4. some orientation: Some other relevant institutional aspects are the • FOG installations: A typical concept for FOG co- following: digestion would include the installation of a FOG • Collection and pretreatment of the organic waste are holding tank that can be heated (in temperate carried out by a private company. climates only). This tank (for example, 30 m3), complete with connection for delivery tanks, • All investments required for waste collection and mixer in the tank, FOG grinding, feeding pump, pretreatment were made by the private company. and connecting pipe to digester, typically implies • Zirl WWTP receives the sludge free of charge. CAPEX on the order of US$68,000. The advantage of this cooperation for the private • Reception and holding chamber: Such a chamber is company is that it avoids the high cost of waste required when the organic waste is already pretreated disposal and has shorter transport distances. The and/or does not require pretreatment. In these cases, mixing of sludge and organic feedstock requires a larger reception installations, into which trucks can formal approval by the relevant authorities in Austria. dump their organic loads, may be required. These installations, complete with mixer, feeding, and so 5.2.5. GHG reduction and CDM co-financing on, cost around US$135,000. The same principles apply as already described in case • Mechanical pretreatment. Some types of organic study 1. For details, see section 1.2.5. waste, such as municipal organic waste, do require pretreatment. CAPEX of a hammer mill, as utilized for the biowaste that is co-digested at Zirl WWTP, ANNEXES WASTEWATER TO ENERGY   138 depends on its capacity. A large installation with a the benefit of a reduced electricity bill was balanced capacity of about fifteen tons/hour (approximately against the extra OPEX of co-digestion. Details are 25 m3/h), complete with all peripheral components, presented in table CS5-5. Furthermore, the structure could cost about US$675,000. of “additional cost” items, which increase OPEX, is presented in figure CS5-9. 5.2.7. OPEX structure The OPEX of the co-digestion in Zirl WWTP has been analyzed in detail for the year 2012. In this analysis, Table CS5-5: OPEX implications of co-digestion at Zirl WWTP in 2012 OPEX (US$) BENEFIT Reduced electricity bill –88,425 ADDITIONAL COST Additional sludge disposal cost 31,725 Additional polymer required for sludge dewa- tering 5,805 Additional aeration requirement due to addi- tional N return load 8,235 Additional maintenance (CHP, pumps, etc.) 14,985 Additional personnel cost 10,125 Subtotal 70,875 BALANCE TOTAL –17,550 Source: Abwasserverband (AV) Zirl et al. ANNEXES WASTEWATER TO ENERGY   139 Figure CS5-9: Structure of OPEX increasing cost items of co-digestion at Zirl WWTP in 2012 Source: Abwasserverband (AV) Zirl et al. While cost-saving effects result from a reduced percent of the electricity savings. This seems quite a electricity bill, the main cost-increasing effect is typical result for central European conditions. Svardal associated with the additional digested sludge (sludge and Haider (2010), after analyzing five WWTPs with disposal + polymers), which makes up about 55 co-digestion in the same region, conclude this ratio percent of all OPEX increases. may vary between 0 and 50 percent. Hence, specific local conditions have a significant financial impact, Related to the annual quantities of 3,300 m3 organic even in a narrowly defined region. waste, the total extra OPEX equals US$21.5/m3 for the organic waste added. The nearby Strass WWTP Tentatively, in regions where sludge disposal is not reported extra OPEX for co-digestion of US$30.4/ as expensive as at Zirl WWTP (unit disposal cost of m3 (Dengg 2003). Thus, as a general rough estimate, US$85/ton) and where personnel is cheaper, the OPEX additional overall OPEX of about US$27/m3 organic increases can be substantially less, and the overall cost waste added seems to reflect central European balance could be considerably more attractive. conditions well. It remains to be mentioned that in this calculation, the At Zirl WWTP, the benefits from electricity savings disposal cost of organic waste that would be incurred if outweigh the additional OPEX and lead to net savings it were not co-digested has not been factored into the of US$17,550. Related to the annual quantities of considerations, since Zirl WWTP receives the sludge 3,300 m3 organic waste, the net savings equal about free of charge from the private company. This could be US$5.4/m3 organic waste added. When comparing quite different at other locations. the overall financial benefits in electricity cost savings, Likewise, not included are the additional thermal at Zirl WWTP the actual net savings amount to 20 ANNEXES WASTEWATER TO ENERGY   140 energy gains from the increased biogas production, Moreover, it reduces the dependence of the WWTP since at Zirl WWTP it has no financial value. on outside power supply. Thus, even in the event of power blackouts or any other problems with the public 5.2.8. Viability of investment in biogas power supply, the WWTP can continue operating at utilization a normal level. In the case of Zirl WWTP, the application of co- Table CS5-6 summarizes several cost indicators from digestion makes economic sense; it achieves an annual case study 5. OPEX reduction of US$17,550 without additional investments. Table CS5-6: Cost indicators for co-digestion at Zirl WWTP     Case study 5 Average influent load PE60,avg 42,950 Average CAPEX EUR/PE60 0.1 Average OPEX reduction EUR/PE60/y –0.3 Average CAPEX US$/PE60 0.1 Average OPEX reduction US$/PE60/y –0.4 Source: Authors’ calculation. Note: 1 PE60 = 1.0 cap in Austria; US$/PE60 x 16.67 = US$/kg BOD5. In sum, co-digestion at Zirl WWTP is financially recycled, which has to be removed from the wastewater viable, and it increases operation safety. train where nitrogen standards exist. This implies a need for additional treatment capacity. A case-specific 5.3. Conclusions for co-digestion of organic assessment of its quantities and implications is always waste in EAP countries needed. But, as case study 5 has shown, nutrient What remains unchanged in EAP removal WWTPs might be able to cope with this (as compared to case study 5) requirement without any expansion works even when feeding large quantities of extra feedstock. Most likely, independent of location, after the introduction of co-digestion no changes will be Neither the widespread wastewater dilution in EAP required to the wastewater, the sludge, or the nor the continued use of septic tanks has an impact biogas train. on co-digestion. Wastewater treatment requirements matter only if Not much change is expected in the potential for enhanced nitrogen (N) removal is required. Via the increased biogas production when comparing European filtrate from sludge dewatering, an extra load of N is results with EAP results. After all, the digesters are ANNEXES WASTEWATER TO ENERGY   141 closed systems with a controlled environment that Success stories in EAP works comparably well at any location worldwide. The Co-digestion is not a traditional approach in Asia or additional waste will be partially destroyed, producing EAP countries. This is definitely connected to the additional biogas. The extent of that increase depends fact that anaerobic digestion in general is not yet on quantities and qualities of the extra feedstock fed widespread there. But the interest in co-digestion is into the digesters. Thus, the electricity generation growing. For instance, Chen and others (2013) report potential will also increase accordingly, to very much from Singapore on co-digestion experiments with food the same extent. waste. The biogas yield they found is on the upper CAPEX is usually required for pretreatment and end of the range indicated for this kind of feedstock in feeding of the extra waste. The necessary extent and table CS5-10. This clearly shows that the potential for cost are highly case sensitive and can range from very co-digestion exists in EAP. low amounts to substantial investments. In Vietnam particularly, co-digestion of fecal sludge OPEX savings are surprisingly low in case study 5. This with sewage sludge is considered of interest. The result may be drastically different in EAP, foremost Hanoi University of Civil Engineering (HUCE) is because the extra sludge disposal and manpower are conducting a research program on that topic at present much cheaper there. (Viet Anh et al. 2014); final results are expected next year. A preliminary result showed a wide variety of The regulatory framework in EAP is still in fecal sludge qualities, with some being quite suitable development. For instance, Vietnam only recently for co-digestion. The co-digestion of fecal and sewage introduced legislation defining feed-in tariffs for sludge also shows the same performance- enhancing electricity generated from biomass and from landfill effects highlighted in this technical note for other biogas; this legislation has also obliged the power co-digestion feedstocks in comparison to separate utility to buy such feed-ins. Another ongoing activity digestion. in Vietnam is the drafting of technical guidelines for energy recovery from sewage sludge. Such co-digestion of sewage sludge with fecal sludge could prove a promising combination for many locations in EAP, where large numbers of septic tanks will prevail for many years to come. Keeping in mind that fecal sludge quality does not only vary within the same city but also from region to region, more research is definitely needed. ANNEXES WASTEWATER TO ENERGY   142 CO-DIGESTION OF ORGANIC WASTE: TECHNICAL SUMMARY Table CS5-7: Co-digestion of organic waste in anaerobic digesters: General design parameters and key characteristics of sludge management DIGESTED The total digested sludge production from co-digestion can be calculated as the digested SLUDGE sum of its individual inputs. It is worth mentioning that this approach may lead to slight PRODUCTION overestimates of total digested sludge production. It has been confirmed by various studies and applications (Dengg 2013; Kusowski et al. 2013; Iacovidou et al. 2012; Johnson et al. 2011; Svardal and Haider 2010; Callegari 2010; Zupancic et al. 2008; Felde et al. 2005; Jansen et al. 2004), that the mutual digestion of sewage sludge and organic waste enhances the biological activities in the digester. Consequently, both sludge and biowaste are more efficiently degraded, and overall sludge production is lower (and overall biogas yield higher) than the sum of individual digestion of the same substances. Simple design tools for quantification of these effects without practical experiments are not available. Therefore, the common design approach is to add up the following components: • Digested sludge production from waste sludge • See table CS1-5. • Digested sludge production from organic waste The theoretical sludge production from organic waste digestion can be estimated by assuming a typical percentage for VS destruction (see table CS5-10). An estimate for the digested biowaste DS can then be derived from total biowaste DS input to the digester, minus VS destroyed. DIGESTED The practical implications of co-digestion for sludge dewaterability do not show a SLUDGE uniform trend. Both improved dewatering results and reduced polymer consumption for DEWATERING conditioning and reduced dewatered DS and increased polymer consumption are reported. For design considerations it is deemed acceptable to assume no relevant changes of sludge dewatering characteristics (that is, to assume the same DS and specific polymer consumption) after the introduction of co-digestion, as compared to prior conditions. FILTRATE The filtrate quality from sludge dewatering is affected by co-digestion. Whereas the QUALITY expected impacts on carbon parameters (BOD5, COD, DS) and phosphorus are mostly FROM SLUDGE considered negligible for municipal WWTPs, the additional nitrogen (N) that is released DEWATERING from the destroyed organic waste and returned to the wastewater train via the sludge dewatering filtrate may be worth considering. ANNEXES WASTEWATER TO ENERGY   143 If the WWTP in question is only designed for carbon removal, this additional N load is not crucial for compliance with the required effluent quality standards. However, due to the fast nitrification in warm climates, it will nonetheless increase energy consumption for aeration. If the WWTP is designed for enhanced N removal, the issue becomes even more important. In this case, the additional N load has to be duly considered in the design of the wastewater train. The additional N load can be estimated by assuming all N contained in the destroyed VS fraction of the organic waste ends up in the filtrate. (This is a safe assumption, since a minor percentage of this “destroyed” N will also be used for biomass synthesis.) For design purposes, it is hence important to know the N content of organic waste and the typical VS destruction in co-digestion. The former is presented in table CS5-8, and the latter can be taken from table CS5-10. If of interest, the additional P load that is returned to the wastewater train from organic waste co-digestion may be calculated the same way. Table CS5-8: N and P content of various types of organic waste Organic waste Ntotal Ptotal   (% of DS) (% of DS) FOG from grease traps 2 (1.5–3.7) 0.3 (0.1–0.7) Waste food from restaurants, canteens, 2 (0.6–5.0) 0.7 (0.1–1.5) etc. Bakery waste 2 0.7 Whey from dairy 1–2 — Agro-industrial waste (e.g., from sugar 1–13 0.5–2.6 mills, starch mills, breweries, etc.) Municipal organic waste 1.5 (0.5–2.7) 0.5 (0.3–0.8) Sources: Based on Traversi et al. 2013; ARAconsult 2009; Huber et al. 2007; Stabnikova and Wang 2006; Felde et al. 2005; DWA 2003b; Billmaier et al. 2001. ANNEXES WASTEWATER TO ENERGY   144 Table CS5-9: Co-digestion of organic waste in anaerobic digesters: General design parameters and key characteristics of energy management BIOGAS DESIGN Biogas production in co-digestion originates from different sources, which can be assessed separately and added together to come up with total biogas production: • Biogas from waste sludge See table CS1-5. • Biogas from organic waste The biogas production from organic waste can be estimated using the three different approaches described below. The method of choice usually depends on data availability. (i) Biogas yield from the digestion of kgVS Table CS5-10: Biogas yield from the digestion of kgVS Sources: FOG: Miot et al. 2013; Norgaard et al. 2013; Schafer et al. 2013; VSA 2012b; Johnson et al. 2011; Felde et al. 2005; DWA 2003b; Loll 2001; Billmaier et al. 2001; Schmelz 2000. Waste food: Chen et al. 2013; Schafer et al. 2013; VSA 2012b; Zupancic et al. 2008; Stabnikova and Wang 2006; Schmelz 2003; DWA 2003b; Loll 2001; Braun 2001. Bakery waste: ARAconsult 2009; Huber et al. 2007. Whey: Traversi et al. 2013; Baubüro and Syneco 2012. Agro-industrial waste: DWA 2003b. Municipal organic waste: VSA 2012b; Schmelz 2003; DWA 2003b; Loll 2001. (ii) Biogas yield from the digestion of kgCOD The methane yield can be calculated with a specific value of 350 L CH4/kg CODdestroyed (Nowak and Ebner 2013; Urban and Scheer 2011; Svardal and Haider 2010; DWA 2002). This methane yield may then be converted into biogas yield by taking the methane content of biogas (see table CS5-10) into account. For CODdestroyed as a percentage of CODtotal, similar percentages as indicated for VSdestroyed in table CS5-10 can be applied. ANNEXES WASTEWATER TO ENERGY   145 Of course, COD and VS of organic waste are interrelated. However, the ratio between these parameters is not a constant, since it depends on the material’s composition. Typical ratios are, for instance, COD/VS ≈ 2.0–2.5 for FOG, and COD/VS ≈ 1.4–1.6 for waste food (Urban and Scheer 2011; Svardal and Haider 2010). (iii) Biogas yield from the digestion of carbohydrate/proteins/lipids Table CS5-11: Biogas yield from the digestion of carbohydrates, proteins, and lipids   Biogas yield Methane content   (L/kg VS) (%) Carbohydrates 830 50 Proteins 720 71 Lipids 1,430 70 Source: DWA 2002. The biogas yields in table CS5-11 represent the theoretical maximum yield, if all VS were completely digested. The numbers presented are taken from DWA (2002), but quite similar theoretical values are cited by Miot and others (2013) and Urban and Scheer (2011). The practical biogas yields, though, are lower, since (a) not all VS is fully destroyed (compare table CS5-10); (b) the chemical composition of the substrates may vary within a certain range; and (c) a fraction of the substrate is always used for cell synthesis. Hence, when calculating the biogas yield for a specific organic waste, these additional aspects should be applied to reduce the biogas values derived from the data provided in table CS5-11. In design practice, usually only the very first of these impacts (a) is indeed phased into the calculations. The overall methane content from the digestion of a specific organic waste can be calculated as the weighted result according to the respective methane contents provided in table CS5-11. BIOGAS TREATMENT See table CS1-5. GAS HOLDER See table CS1-5. FLARE See table CS1-5. BIOGAS UTILIZATION See table CS1-5. ANNEXES WASTEWATER TO ENERGY   146 ANNEXES WASTEWATER TO ENERGY   147 CASE STUDY 6: ULTRASOUND SLUDGE DISINTEGRATION ANNEXES WASTEWATER TO ENERGY   148 6.1. BACKGROUND, PROCESS DESCRIPTION 6.1.1. Data sources ultrasound equipment are also available on the market. Sludge disintegration can be used for various purposes: Since ultrasound sludge disintegration is implemented reduction of foaming problems, reduction of sludge in closed reactors, treating conventional waste activated quantities, or increase of biogas production. In this sludge as can be found anywhere in the world, the case study, the focus is exclusively on the last. results from this case study can be transferred to any Several sludge disintegration technologies are available other location worldwide. on the market. The most common are based on mechanical, thermal, or chemical-thermal processes. 6.1.2. Wastewater management Generally, the thermal and chemical-thermal processes Wastewater treatment at all the WWTPs analyzed (for example, the Cambi process and Pondus process) in this case study is done by CAS. Since sludge are not deemed very appropriate for the EAP countries disintegration is applied in the sludge treatment train, on which this report focuses because they involve high- there are no changes to the wastewater train. Only one tech installations and require highly skilled operators. aspect may require attention: when using ultrasound The mechanical disintegration technologies include, sludge disintegration, the nitrogen (N) load in the among others, lysate centrifuges, agitator bead mills, filtrate from sludge dewatering increases. Those plants and ultrasound sludge treatment. Among these, the that require enhanced N removal will hence have to last is the most widely used. In Europe there are now deal with this additional N load in one way or another. about 100 WWTPs using this technology, and in Yet, since the additional loads are relatively small, the Asia it has gained ground, particularly in the south. impacts are small as well. Furthermore, it is a small installation that is very easy to operate. For these reasons, this system was selected 6.1.3. Sludge management for the present case study. All the WWTPs of this case study employ CAS technologies, combined with primary sedimentation. Case study 6 presents the experiences of eight Therefore, all these plants produce two types of sludge: municipal WWTPs with the application of ultrasound primary sludge and waste activated sludge (WAS). sludge disintegration to enhance biogas production in mesophilic sludge digesters. All the analyzed plants Ultrasound sludge disintegration (USD) for biogas are equipped with ultrasound installations from VTA enhancement is based on the principle that microbial Technologie GmbH (www.vta.cc), which is among cell walls are destroyed. The (easily digestible) cell the leading providers worldwide of these installations. liquors are released and contribute to additional Similar products from other manufacturers of biogas production. Hence, ultrasound disintegration 4 The disintegration (cell wall destruction) effect of ultrasound treatment is based on cavitation. Ultrasound oscillator elements are usually positioned inside a small disintegration reactor. The emitted ultrasonic waves cause a periodic compression and decompression in the sludge matrix. Locally this causes extreme conditions with high pressures and temperatures. Cavitation bubbles are formed and implode again. Depending on the energy applied, the extent of cell lysis of the sludge’s microorganisms can be controlled. ANNEXES WASTEWATER TO ENERGY   149 is exclusively applied to waste activated sludge, Typically, USD is located after thickening of WAS where microorganisms dominate the sludge mixture; and prior to digester feeding. The treated sludge treating primary sludge would not increase the biogas is subsequently fed into the digester. Figure CS6- yield, since the bulk of it is organic matter and not 1 depicts the typical positioning of USD in sludge microorganisms. treatment for biogas enhancement. Figure CS6-1: Simplified flow scheme of ultrasound sludge disintegration for biogas enhancement Source: Authors. Specific characteristics of the USD system analyzed in • DS content of 3–8 percent considered ideal for this case study are the following: optimum efficiency • Frequency used: 25 kHz Key characteristics of the WWTPs and their USD systems analyzed in this case study are summarized in • Typical retention time in the disintegration reactor table CS6-1. ranging from 60 to 180 minutes • Upstream flow of sludge inside the disintegration reactor ANNEXES WASTEWATER TO ENERGY   150 Table CS6-1: Key characteristics of WWTPs with ultrasound sludge disintegration investigated for case study 6 Sludge disinte-   WWTP Country Capacity Analysis grated       (PE60) (% of WAS) period 1 Villach WWTP Austria 200,000 30 1 year 2 Großostheim WWTP Germany 35,000 30–100 1 year 3 Halle Nord WWTP Germany 300,000 40–60 3 years 4 Miltenberg WWTP Germany 95,000 30 2 years 5 Roth WWTP Germany 65,000 80–100 0.5 years 6 Wasserfeld WWTP Italy 40,000 40–60 1 year Switzer- 7 Estavayer-le-Lac WWTP 110,000 70–100 0.2 years land Moossee-Urtenenbach Switzer- 8 40,000 50–100 1 year WWTP land ca. 60 percent of   AVERAGE   110,625 ca. 1.2 years WAS Source: VTA—private communication 2014. Note: 1 cap = 60 g BOD5/d in Europe. Figures CS6-2, CS6-3, and CS6-4 present, some photographs of implemented installations, and respectively, a schematic of the disintegration reactor, microscopic pictures of sludge before and after USD. Figure CS6-2: Schematic of an ultrasound sludge disintegration reactor Source: Eder 2007. ANNEXES WASTEWATER TO ENERGY   151 Figure CS6-3: Implemented ultrasound sludge disintegration installations ultrasonic oscillator ultrasonic oscillator new 20,000 operation hours Sources: VTA 2012; Eder 2007. Figure CS6-4: Microscopic analysis of ultrasound sludge disintegration Source: VTA 2012. The design of anaerobic sludge digesters remains Relevant aspects prevailing for USD installations are unchanged when introducing USD. Hence, for details summarized in table CS6-8. on general design parameters and key characteristics of anaerobic digesters, see table CS1-4. ANNEXES WASTEWATER TO ENERGY   152 6.1.4. Energy management 6.2. Analysis Two energy aspects have to be traded off when 6.2.1. Wastewater influent, effluent, and other analyzing USD: parameters of interest • Increased biogas, and thus increased energy Wastewater production from biogas No specific information regarding the wastewater • Additionally required energy input for the operation characteristics of the WWTPs investigated for this of USD case study is available. However, one can assume The crucial question is overall energy balance and rather similar values to those in table CS1-1, which the financial benefits compared to the CAPEX presented typical wastewater characteristics from the requirements, if it is positive; section 6.2 will analyze same region where the WWTPs of case study 6 are these issues. located. For details on general design parameters and key Sludge characteristics of biogas systems combined with USD focuses on improved sludge treatment. Table anaerobic sludge digesters, see table CS1-5. Typically, CS6-2 summarizes the observed changes in sludge no changes are required in the biogas train when USD characteristics at the WWTPs of this case study. These is installed. The increase in biogas is usually not so results relate to an average observation period of 1.2 large that expanded capacities are required. years, as summarized in table CS6-1. Specific, energy-related aspects of USD installations From the results of this case study, the following can be are summarized in table CS6-9. observed: Table CS6-2: Impact of ultrasound sludge disintegration on sludge characteristics Source: VTA—private communication 2014. ANNEXES WASTEWATER TO ENERGY   153 • Enhanced VS destruction during digestion: VS • Dewatered sludge production: There is a general trend destruction is clearly enhanced. VS destruction toward reduced sludge quantities of 12 percent on increases on average from 49 percent to 56 percent. average (range: 5–22 percent). This brings about a VS content in the digested sludge • Increased nitrogen (N) return load in filtrate that is about five (three to eight) absolute percentage from sludge dewatering: This aspect went mostly points lower than before USD. Comparable results unmentioned in the reports on the WWTPs of this are also cited in the literature (Schmelz and Müller case study. It is hypothesized that this is due to the 2004). combined effects of (a) little attention paid to this • Improved DS in dewatered sludge: Only three WWTPs return load and (b) no significant changes in return report improved dewatering characteristics: two cite load. Therefore, an increase of N return load by 5 an increase of absolute dewatered DS of 1 percent, percent, as found by Schmelz and Müller (2004), and one cites an increase of almost 3 percent. may be a realistic estimate. In all five other cases, dewatered DS is reported unchanged. Combining the observed reductions 6.2.2. Biogas production and potential for of dewatered sludge quantities (see below) with energy generation the observed VS destruction (see above) hints at an Biogas production improvement of dewatered DS by an absolute +1 Table CS6-3 and Figure CS6-5 show that USD percent DS. Hence, all in all, an assumption of 0–1 treatment increased the biogas production of the eight percent DS increase appears realistic. WWTPs of this case study by 24 percent on average • Reduced polymer conditioning in sludge dewatering: (range: 12–33 percent). This increase equaled about While some WWTPs report reduced polymer 100 (50–140) L/kgVSadded. consumption, others do not notice any changes. Some studies have even found increased polymer These results relate to an average observation period of consumption; for example, Schmelz and Müller 1.2 years, as summarized in table CS6-1. (2004) report an increase by 10 percent, so there is not a clear trend. For the time being, then, it is recommendable to assume no changes of polymer consumption after the introduction of USD. ANNEXES WASTEWATER TO ENERGY   154 Table CS6-3: Impact of ultrasound sludge disintegration on biogas production Biogas production Biogas production WWTP Biogas increase   without USD with USD     (L/kgVSadded) (L/kgVSadded) (L/kgVSadded) (%) 1 Villach WWTP 556 654 98 18 2 Großostheim WWTP 425 565 140 33 3 Halle Nord WWTP 423 475 52 12 4 Miltenberg WWTP 329 425 96 29 5 Roth WWTP 414 537 123 30 6 Wasserfeld WWTP 358 461 103 29 7 Estavayer-le-Lac WWTP 577 705 128 22 Moossee-Urtenenbach 326 400 74 8 WWTP 23   AVERAGE 426 528 102 24 Sources: Weimer et al. 2005; Ullrich and Eder 2006; Wernitznig and Eder 2006; Petril 2006; Rausch 2007; VTA 2010; Casazza 2010; Resch 2011; VTA—private communication 2014. Figure CS6-5: Impact of ultrasound sludge disintegration on biogas production Sources: Weimer et al. 2005; Ullrich and Eder 2006; Wernitznig and Eder 2006; Petril 2006; Rausch 2007; VTA 2010; Casazza 2010; Resch 2011; VTA—private communication 2014. ANNEXES WASTEWATER TO ENERGY   155 Potential for energy generation Energy consumption Since the biogas characteristics do not change with The additional power generation is partly consumed USD, the biogas increase by 24 percent (12–33 for the power requirements of USD installations, as percent) can be directly translated into additional depicted in figure CS6-6 and table CS6-4. energy generation of 24 percent (12–33 percent). Figure CS6-6: Energy consumption of ultrasound sludge disintegration Sources: Weimer et al. 2005; Ullrich and Eder 2006; Wernitznig and Eder 2006; Petril 2006; Rausch 2007; VTA 2010; Casazza 2010; Resch 2011; VTA—private communication 2014. Table CS6-4: Energy consumption of ultrasound sludge disintegration   Actual load Energy consumption for USD     (PE60) (kWh/d) (kWh/PE60/y)   1 Villach WWTP 120,000 245 0.75 2 Großostheim WWTP 35,000 90 0.94 3 Halle Nord WWTP 300,000 720 0.88 4 Miltenberg WWTP 60,000 163 0.99 5 Roth WWTP 50,000 147 1.07 6 Wasserfeld WWTP 28,000 40 0.52 7 Estavayer-le-Lac WWTP 90,000 600 2.43 Moossee-Urtenenbach 8 35,000 90 0.94 WWTP   AVERAGE 89,750 262 1.06 Sources: Weimer et al. 2005; Ullrich and Eder 2006; Wernitznig and Eder 2006; Petril 2006; Rausch 2007; VTA 2010; Casazza 2010; Resch 2011; VTA—private communication 2014. ANNEXES WASTEWATER TO ENERGY   156 On average, USD’s energy consumption amounts to It is worth mentioning that USD can help eliminate 1.06 kWh/PE60/year. digester foaming. Given that biogas production increased by about 24 6.2.4. Institutional aspects, energy costs percent (12–33 percent), this enables an additional The institutional aspects prevailing in Central Europe power production of 24 percent (12–33 percent). have already been discussed in case study 1 for The typical electric power generation of CAS plants Germany and Austria. For details, see section 1.2.4. with mesophilic digesters is about 15–18 kWh/PE60/y with N elimination (see table CS1-2). The increase of The situation in Italy and Switzerland, where some of electricity generation thus equals about 4 (2–6) kWh/ this case study’s WWTPs are located, is very similar PE60/y. It is possible to conclude that USD consumes to that in Germany and Austria. Country-specific roughly one-third of the electricity generated from the details of the actual regulations on renewable energy additional biogas. About two-thirds of the electricity generation are not presented here, but can be found increase remain as net benefit. online at http://www.res-legal.eu/. 6.2.3. Operation capacity needs, biogas safety 6.2.5. GHG reduction and CDM co-financing All issues generally associated with biogas safety and The same principles apply as already described in case operation capacity needs have already been discussed study 1. For details, see section 1.2.5. in case study 1. For details, see section 1.2.3. 6.2.6. CAPEX structure No particular extra capacities or precautions CAPEX required for USD depends on plant size and are required for the O&M of the ultrasound shows a clear effect of economies of scale. Figure CS6-7 installations. None of the WWTPs of this case provides the USD cost numbers for all WWTPs of this study reported any particular problems. The USD case study. CAPEX for all investments required for a automatically operates twenty-four hours a day, complete USD facility, including freight, installation, all year round. Operators usually just do routine startup, engineering, and taxes, is on the order of EUR inspections of pumping, mixing, and oscillator 3 (1–5)/PE60 (US$4 [1.4–6.8]/PE60). operation. Maintenance is limited to conventional pump and mixer maintenance and replacement of used oscillators after a couple of years. ANNEXES WASTEWATER TO ENERGY   157 Figure CS6-7: CAPEX of ultrasound sludge disintegration Sources: Weimer et al. 2005; Ullrich and Eder 2006; Wernitznig and Eder 2006; Petril 2006; Rausch 2007; VTA 2010; Casazza 2010; Resch 2011; VTA—private communication 2014. 6.2.7. OPEX structure OPEX of USD is characterized by two opposing trends: some implications increase OPEX, and some decrease it. Table CS6-5 summarizes both effects. Table CS6-5: Impact of ultrasound sludge disintegration on OPEX Item OPEX decrease OPEX increase Additional generation of about 24 per- Electric power required to operate USD: cent (12–33 percent) in electric power, Electric power typically about one-third of the addi- as compared to conventional CAS + tionally generated power digestion Maintenance cost for new USD instal- Maintenance — lations Manpower required for O&M of USD Manpower — installation: typically 50 hours per year (Wernitznig 2010) Reduction of sludge quantities by about 12 percent (5–22 percent) due to two Sludge reuse/dis- combined effects: (a) enhanced VS de- — posal cost struction during digestion; (b) improved DS in dewatered sludge Source: Authors’ summary. ANNEXES WASTEWATER TO ENERGY   158 While electricity cost, manpower cost, and sludge Under central European conditions, the dominant reuse/disposal cost can be easily assessed with local impact on OPEX comes from electricity. Taking all unit cost, maintenance can be quantified according to aspects into account, OPEX reduction found at the figure CS6-8. eight WWTPs of this case study turned out to be about EUR1.0 (0.3–1.7)/PE60/y (US$1.4 [0.4–2.3]/ PE60/y), depending on local conditions. Figure CS6-8: Maintenance cost of ultrasound sludge disintegration Source: VTA 2014. 6.2.8. Viability of investment in biogas utilization The amortization periods for investment into ultrasound sludge disintegration (USD) have been calculated by the various operators of the WWTPs in this case study. The outcomes are summarized in table CS6-6. All the results have been derived from NPV calculations, considering total life cycle cost. ANNEXES WASTEWATER TO ENERGY   159 Table CS6-6: Amortization periods reported for ultrasound sludge disintegration   WWTP Payback period     (years) 1 Villach WWTP ≈4 2 Großostheim WWTP ≈5 3 Halle Nord WWTP ≈5 4 Miltenberg WWTP ≈4 5 Roth WWTP ≈5 6 Wasserfeld WWTP ≈6 7 Estavayer-le-Lac WWTP ≈4 Moossee-Urtenenbach 8 ≈3 WWTP   AVERAGE 5 Sources: Weimer et al. 2005; Ullrich and Eder 2006; Wernitznig and Eder 2006; Petril 2006; Rausch 2007; VTA 2010; Casazza 2010; Resch 2011; VTA—private communication 2014. Table CS6-7 summarizes several cost indicators from case study 6. Table CS6-7: Cost indicators for ultrasound sludge disintegration at the WWTPs of case study 6     Case study 6 Average influent load PE60,avg 89,750 Average CAPEX EUR/PE60 3 (1–5) Average OPEX reduction EUR/PE60/y 1.0 (0.3–1.7 ) Average CAPEX US$/PE60 4 (1.4–6.8) Average OPEX reduction US$/PE60/y 1.4 (0.4–2.3) Source: Authors’ calculation. Note: 1 PE60 = 1.0 cap in Central Europe; US$/PE60 x 16.67 = US$/kg BOD5. At all eight WWTPs analyzed for this case study, it was concluded that the ultrasound sludge disintegration project is financially viable. Life cycle cost within the lifespan of new installations for USD (ten to fifteen years) was considered lower with the new installations than without. Average amortization is expected within about five years under European conditions. ANNEXES WASTEWATER TO ENERGY   160 6.3. Conclusions for ultrasound sludge In terms of financial viability, that means amortization disintegration in EAP countries periods may be somewhat longer in EAP than the Wastewater treatment requirements matter only if typical five years found in Europe. enhanced nitrogen (N) removal is required. Via the Success stories in EAP filtrate from sludge dewatering, an extra load of N is recycled. A case-specific assessment is always needed Applications of USD are mostly limited to Europe to of its quantities and implications. But, as case study 6 date. has shown, these implications are usually minor. In Shanghai, China, pilot tests have been done Neither the widespread wastewater dilution in EAP at Bailongang WWTP with ultrasound sludge nor the continued use of septic tanks has an impact on disintegration of sludge from chemically enhanced USD. After all, USD and digesters are closed systems primary treatment (CEPT). Not surprisingly, these with a controlled environment that works comparably results were not all that successful, since USD only well at any location worldwide. works well with biological waste activated sludge (WAS; VA Tech Wabag 2006). CAPEX is required for USD installations and should be similar in EAP and Europe. Recently, USD got a foothold in South Korea, where in 2011 a total of six large-scale USD facilities were OPEX savings are, in general, dominated by increased successfully commissioned. These plants are called electricity generation and reduced sludge quantities. Daegu Bukbu WWTP, Daegu Seobu WWTP, Daegu Since both these items feature lower unit cost in EAP Sincheon WWTP, Daegu Dalseo WWTP, Gangneung than in Europe, absolute OPEX reductions should be WWTP, and Gimhae WWTP. No operation data are lower in EAP. known to the authors of this technical note. ANNEXES WASTEWATER TO ENERGY   161 ULTRASOUND SLUDGE DISINTEGRATION: TECHNICAL SUMMARY Table CS6-8: Ultrasound sludge disintegration prior to anaerobic digesters for biogas enhancement: General design parameters and key characteristics of sludge management USD • Generally, ultrasound frequencies range from 20 kHz to 10 GHz. For sludge disintegration, CONSTRUCTION 20–40 kHz is commonly used. DETAILS • The typical lifespan of ultrasound oscillators = 45,000 operation hours. When running for twenty-four hours a day, that is equivalent to more than five years. The guaranteed lifespan usually is about 25,000 operation hours. • Modular systems, which can be easily extended • Low maintenance needs • Automated operation • Area requirement for disintegration reactor + pumps + control switchboard equal to about 5 m2/100,000 PE60 • Minimum required room height = 2.8 m • Noise emissions: max 65 dB DIGESTED General: The VS content in the digested sludge can be reduced by up to 25 percent (VTA SLUDGE 2012). Considering VS makes up about 50–60 percent of digested sludge, that means PRODUCTION overall sludge quantities may be reduced by up to 10–15 percent. Case study 6: About 15 percent (10–20 percent) more VS was destroyed in the digesters with USD than without USD. Therefore, in absolute terms, the percentage of VSdestroyed increased from 49 percent to 56 percent, on average. Furthermore, a reduction of dewatered sludge quantities of about 12 percent (5–22 percent) was observed. Since this reduction cannot be explained by the stronger VS destruction alone, it hints at an improvement of DS in dewatered sludge by an absolute +1 percent DS. DIGESTED General: A reduced VS content of the digested sludge should facilitate dewatering. Hence, SLUDGE the DS of dewatered sludge may increase, and dewatered sludge quantities (m3) to dispose DEWATERING of/reuse may be reduced. The combined effect of reduced VS content (see above) and of improved dewatered DS is said to enable an overall sludge reduction of up to 20 percent (VTA 2012). ANNEXES WASTEWATER TO ENERGY   162 Additionally, polymer conditioning for dewatering may be reduced by up to 20 percent (VTA 2012). Case study 6: The operators mostly did not report dewatered DS changes. Also, polymer consumption for sludge conditioning was usually not affected. Nonetheless, the reduction of dewatered sludge quantities hints at an increase of dewatered DS by an absolute +1 percent DS (see above item). This may not have been noticed by the operators, since it is a relatively minor impact that is not too apparent. FILTRATE General: The filtrate quality from sludge dewatering is affected by USD. Whereas the QUALITY expected impact on carbon parameters (BOD5, COD, DS) and phosphorus are mostly FROM SLUDGE considered negligible for municipal WWTPs, the additional nitrogen (N) that is released DEWATERING together with the cell liquors and is returned to the wastewater train via the sludge dewatering filtrate may be worth considering. If the WWTP in question is only designed for carbon removal, this additional N load is not crucial for compliance with the required effluent quality standards. Nonetheless, due to the fast nitrification in warm climates, it will increase energy consumption for aeration. If the WWTP is designed for enhanced N removal, the issue becomes even more important. In this case, the additional N load has to be duly taken into account in the design of the wastewater train. The additional N load caused by USD can be estimated by assuming 100 g of additional N per additional kgVSdestroyed (DWA 2009). Schmelz and Müller (2004) state that this leads to an increase of N return load by 5 percent. Case study 6: No quantifiable impacts on increased return loads were observed at the WWTPs of this case study. This could be seen as a confirmation of the cited minor N increase of only 5 percent. ANNEXES WASTEWATER TO ENERGY   163 Table CS6-9: Ultrasound sludge disintegration prior to anaerobic digesters for biogas enhancement: General design parameters and key characteristics of energy management ENERGY General: The energy benefits are directly proportional to the extra biogas generation. BENEFITS FROM USD Extra biogas: Typical increases in biogas quantities are up to 30 percent (VTA 2012). For the electricity generation from that extra biogas, apply the principles of table CS1-5. Case study 6: USD treatment increased biogas production of the eight WWTPs of this case study by 24 percent (range: 12–33 percent). This equalled an increase by about 100 (50–140) L/kgVSadded. Given the typical electric power generation of CAS plants with mesophilic digesters of about 15–18 kWh/PE60/y with N elimination (see table CS1-2), this equals an increase in electricity generation by 4 (2–6) kWh/PE60/y. General: The following electricity consumption is caused by USD: ENERGY REQUIREMENTS USD: Typical electricity requirements of ultrasound sludge disintegration vary between 0.5 FOR USD and 20 kWh/m3 sludge with 5 percent DS treated. This is equivalent to about 0.01–0.4 kWh/kgDS with an average DS = 5 percent. These values hold true if 100 percent of the sludge is, indeed, treated (Müller 2010; DWA 2009). • Peripheral installations (mixer, pumps): Typically, 0.9 kWh/m3 sludge is treated (DWA 2009). • Aeration of N return load: Additional aeration requirement due to N return load through filtrate = 1.95 kWh/kg Neliminated for fine bubble aeration (Müller 2010; DWA 2009). Case study 6: The overall energy consumption for all USD-related installations was found to be about 1.1 (0.5–2.4) kWh/actual PE60/y. This is equivalent to approximately 30 percent of the additional electricity generated from the increased biogas quantities (see above). BIOGAS TREATMENT See table CS1-5. GAS HOLDER See table CS1-5. FLARE See table CS1-5. BIOGAS UTILIZATION See table CS1-5. ANNEXES WASTEWATER TO ENERGY   164