- Email: [email protected]
Pilot-scale anaerobic co-digestion of municipal biomass waste: Focusing on biogas production and GHG reduction
Renewable Energy 44 (2012) 463e468
Contents lists available at SciVerse ScienceDirect
Renewable Energy journal homepage: www.elsevier.com/locate/renene
Technical note
Pilot-scale anaerobic co-digestion of municipal biomass waste: Focusing on biogas production and GHG reduction Xiao Liu a, *, Xingbao Gao a, b, Wei Wang a, Lei Zheng a, Yingjun Zhou c, Yifei Sun d a
School of Environment, Tsinghua University, Beijing 100084, China Chinese Research Academy of Environmental Sciences, Beijing 100012, China c Department of Urban and Environmental Engineering, Graduate School of Engineering, Kyoto University, Katsura, Nisikyo-ku, Kyoto 615-8540, Japan d School of Chemistry and Environment, Beihang University, Beijing 100191, China b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 April 2011 Accepted 24 January 2012 Available online 16 February 2012
A pilot-scale anaerobic co-digestion research study is presented to elucidate the feasibility of developing anaerobic digestion (AD) as an effective disposal method for municipal biomass waste (MBW) in China, focusing on biogas production and greenhouse gas (GHG) reduction. Food waste, fruitevegetable waste, and dewatered sewage sludge were co-digested in a continuous stirred-tank reactor for biogas production. Stable operation was achieved with a high biogas production rate of 4.25 m3 (m3 d)1 at organic loading rate of 6.0 kgVS (m3 d)1 and hydraulic retention time of 20 d. A total of 16.5% of lipids content was beneficial to the biogas production of the feedstock without inhibition to anaerobic digestion. Compared with the landfill baseline, GHG reduction is an important environmental benefit from MBW digestion. Therefore, anaerobic co-digestion is a promising alternative solution for MBW because it contributes significantly to the sound management of municipal solid waste in China. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Municipal biomass waste Anaerobic co-digestion Biogas production GHG reduction
1. Introduction With the rapid economic development and urbanization of China, almost all of its megacities are faced with the problem of municipal solid waste (MSW) disposal. In 2009, a total of 157.4 million tons of MSW was collected and transported nationwide. Of these, 89.0, 20.2, and 1.8 million tons were treated by landfill, incineration, and composting, respectively [1]. With increasing quantity, 50e60% of the MSW was biomass waste characterized by high water and biodegradable organic content. The municipal biomass waste (MBW) has led to serious adverse effects in traditional MSW treatment systems (i.e. landfill and incineration). The high water content may cause abundant production of leachate for landfill, and may cause unstable burning conditions and dioxin release from incineration. Also the high biodegradable organic content may cause the production and emission of greenhouse gas (GHG). From 2005 to 2010, the wastewater treatment capacity in China has increased from 60 million tons to 125 million tons which has resulted in the rapid increase of sewage sludge and the disposal of sewage sludge has also been a big problem in almost all the cities
* Corresponding author. Tel.: þ86 1062772814; fax: þ86 1062782910. E-mail addresses: [email protected], [email protected] (X. Liu). 0960-1481/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.renene.2012.01.092
in China. In some cities, the sewage sludge is simply deposited onto the wasteland without any treatment, which has caused serious pollution. Efficient MBW management technology is increasingly required due to environmental and economical concerns, such as climate change, eutrophication, and the diminishing resources of fossil energy and raw materials. Anaerobic digestion (AD) is considered as a sustainable option for the management of biomass wastes because the production of renewable energy and the recycling of nutrients [2]. Additionally, MBW separated from MSW and treated with AD can significantly reduce the load of traditional disposal facilities, and subsequently prolong their service life. It also decreases the secondary pollutants originated from the biodegradation of organic wastes during landfill, incineration and composting. AD has been employed in Western Europe since the 1980s, while up to 2010 one-hundred and ninety-five facilities have been constructed with a total annual capacity of 5.9 million tons [3]. De Baere concluded that AD facilities have captured the major market of waste treatment in the EU in the last decade, and will certainly continue developing in the future. AD is expected to become global utilised because of its environmental contribution, and energy benefits [4e6]. Codigestion of different types of MBW has been discussed in many references due to its potential of increasing biogas output and improving stability of anaerobic system [4,7e10]. Co-digestion also means more feedstock supply, which is especially needed by large-
464
X. Liu et al. / Renewable Energy 44 (2012) 463e468
scale plants where large quantities of MBW are required. MataAlvarez et al. summarized several cases of co-digestion of MBW, in both research and practical use [11]. Given that traditional technologies may cause severe environmental pollution for treating MBW, alternative environmentfriendly treatment technologies of MBW are much needed in China. Meanwhile, renewable energy recovered from the biomass waste has become a global concern, and bioenergy has been listed in the Chinese New Energy Promotion Plan [12]. The AD of MBW can reduce the GHG emission in two respects: reduction as compared with the baseline management, and reduction through providing alternative resources in terms of non-renewable fossil energy and materials. In 2009, 79% of treated MSW was contributed by the landfill, and landfill is considered as one of the main GHG emission sources in current China [1]. The development of AD can restructure the traditional MBW treatment and disposal system in China, and in turn achieve the reduction of GHG emissions. Along with other advantages mentioned earlier, AD has attracted much attention of both government and enterprises in recent years, and it can be expected that AD will be widely used in the near future in China. At present, the support policy from the central government has made every megacity receptive to building large-scale AD facilities in China. However, due to the differences of geographic locality and dietary habit, MBWs especially food waste in China has the typical characteristics of high water content (>85%) and high lipids content (>20% of dry basis) which is different from other countries, and few reports have discussed the AD of Chinese MBWs. Therefore, whether the Chinese MBWs fit for AD, and what is the suitable operation condition, these questions need to be answered before the construction and operation of the large-scale AD projects in China. Thus, a systematic research of AD of Chinese MBWs is urgently needed to provide fundamental technology parameters and promote the commercialization of AD technology. Finally, as the second biggest producer of carbon dioxide, China is facing great pressure in reducing GHG emissions, so the specific GHG reduction which can be achieved during AD of MBW is gaining interesting and needs to be quantified. In this study, a pilot-scale anaerobic codigestion research is presented to elucidate the feasibility of developing AD in China as an effective disposal method for MBW, focusing on the system performance and biogas production, and then GHG reduction of AD was analyzed compared with landfill. 2. Methods 2.1. Raw materials MBW used in this experiment comprised food waste (FW), fruitvegetable waste (FVW), and dewatered sewage sludge (DSS). FW was collected from a student canteen (capacity, over 1000 students for a seated dinner) at Tsinghua University, FVW from a wholesale market, and DSS from a municipal wastewater treatment plant (WWTP) in Beijing (Qinghe WWTP; Northern Beijing; capacity, 400,000 m3 d1). The inert materials in FW and FVW were manually separated (e.g., plastic, bone, wood, and others). FW and FVW were crushed to less than 3 mm size firstly by a food waste pulverizer, and then mixed with DSS at the ratio of 2:1:1. The mixed feedstock was kept at 4 C before use. During the experiment, a maximum of 80 kg d1 mixture was fed to the AD reactor. The characterization of raw materials is shown in Table 1 in terms of solid content, organic composition, and elemental composition. Volatile solid (VS), standing for organic content, accounts for 65e90% of the total solid (TS). VS can be divided into volatile dissolved solid (VDS) and volatile suspended solid (VSS). VSS/VS, representing the organic solid ratio in organic fraction,
Table 1 Characterization of raw materials.
Water content/% Total solid/g L1 Volatile solid/g L1 Suspended solid/g L1 Volatile suspended solid/g L1 VS/TS/% VSS/VS/% Lipids/%TS Protein/%TS Crude fiber/%TS C/% H/% N/% C/N
FW
FVW
DSS
Mixture (2:1:1)
83.4 166.3 26.7 149.0 24.3 72.8 14.3 68.8 12.0
93.8 62.2 16.0 50.8 11.2 35.7 14.2 29.6 11.2
84.5 154.9 18.1 101.9 10.8 151.7 21.4 98.5 12.8
88.7 142.1 9.3 117.3 7.8 91.7 15.0 74.0 12.2
89.6 46.2 21.8 16.8 5.6 48.2 7.3 2.8 17.4
81.6 58.3 2.9 13.2 15.3 42.0 6.1 2.4 17.4
65.8 96.6 10.3 34.3 7.1 37.2 5.5 5.9 6.3
82.5 63.1 16.5 20.8 6.4 45.4 6.6 3.5 12.9
accounts for 46.2%, 58.3% and 96.6% for FW, FVW, and DSS, respectively, indicating that organic solid plays an important role in the AD process, which has been proven to be the essential ratelimiting factor [13,14]. The organic compositions of the three raw materials differ significantly. For FW, the lipids content is the highest because of the Chinese traditional cooking style, whereas for FVW, the crude fiber is most abundant because of the cellulose content in fruits and vegetables. For DSS, the protein content is the highest compared with KW and VFR because of the high content of microorganisms. The C/N ratio of FW and FVW is suitable for AD; however, the C/N ratio of DSS is too low for AD. In the traditional AD of sewage sludge, slow degradation (>20 days), low organic loading rate, and the relatively low VS removal (30e40%) are often the disadvantages of the process because the digesters are operated with too low C/N ratios. Hence, in this study, DSS was co-digested with FW and FVW of higher C/N ratios to cover the adverse effects caused by its low C/N ratio. 2.2. Pilot-scale reactor start-up and operation Continuous stirred-tank reactor (CSTR) was used in the research. The reactor had a volume of 2 m3 (effective volume, 1.6 m3; height, 2.7 m; and diameter, 1.0 m). The feedstock of the reactor was a mixture of 50% of FW, 25% of FVW and 25% of DSS in the percentage of weight. Inoculums (digested sewage sludge) were collected from another municipal wastewater treatment plant (Xiaohongmen WWTP, located in southern Beijing), where the excess sewage sludge was treated by AD. A 1.6 m3 volume of digested sludge was collected and pumped into the reactor. The TS, SS, VS, and VSS contents of the inoculums were 27.0, 25.0, 10.4, and 9.7 g L1, respectively. The reactor was operated under mesophilic condition at 35 2 C by a water jacket. The reactors were constantly mixed using mechanical stirrers (100 rpm) with an agitation time of 15 min per two hours, and were fed once a day using a screw pump. 2.3. Analytical techniques The analysis of TS, VS, SS and VSS were based on the Standard Analytic Methods promulgated by the National Environmental Protection Agency of China (1989). The measurement of crude fiber and protein were according to ISO 6865:2000 and ISO1871:2009, respectively. Lipids content was determined using soxhlet extraction method according to ISO 6492:1999. The samples were filtered through 0.45 mm filters before the measurement of volatile fatty acids (VFAs) using gas chromatography (SHIMADZU GC-2010) with
X. Liu et al. / Renewable Energy 44 (2012) 463e468
a flame ionization detector and GDX-102 column (inlet, 200 C; oven, 170 C; and detector, 220 C). The various VFAs included acetic, propionic, iso-butyric, butyric, iso-valeric, and valeric acids. The biogas volume was measured with a wet-test gas flow meter, and the composition of the biogas was monitored by gas chromatography (SHIMADZU GC-2010) with a thermal conductivity detector and RT-Qplot column (column, 30 m 0.53 mm; flame ionization detector, 200 C; oven, 50 C; inlet, 200 C; carrier gas, hydrogen 10 ml min1; split ratio, 35:1, and injection volume, 100 ml). The C, H, and N were analyzed by an elemental analyzer (EAI CE-440). 2.4. Calculation of the GHG reduction The GHG emission of MBW treatment was evaluated using the analytical method of carbon footprint following the Guideline to PAS 2050 [15]. In this study, the carbon footprint of three scenarios was assessed which were baseline scenario (scenario 1), AD with power generation (scenario 2), and AD with bio natural gas (BNG) recovery (scenario 3). In China more than 70% of treated MSW was disposed by landfill, therefore, the baseline scenario was set as MBW landfill. The collection rate of landfill gas (LFG) is generally less than 60% in the developed countries, whereas to achieve a 20% gas recovery in China appears to be difficult [16]. In this study, an average LFG collection rate of 50% was set as the ideal condition resulting from the strict standards and laws concerning with the national energy saving and emission reduction strategy. Analyses of currently operating LFG-to-electricity projects in China show that power generated through LFG recovery and utilization is approximately 30 kWh per ton of waste. There is an evident difference compared with the 250e300 kWh per ton of waste via incineration so, the energy recovered from LFG is fairly limited. However, at present, power generation is a widely used method to obtain energy and reduce global warming potential of methane. Hence, LFG is captured to generate power in the baseline scenario. The methane potential for MBW in a landfill site was calculated following the 2006 IPCC Guidelines for National Greenhouse Gas Inventories [17], which suggested 0.5 as the fraction of the degradable organic carbon (DOCf) that can decompose under anaerobic condition of wet bulk waste. According to Olivier et al., an average DOCf about 40.9% (181 and 107 Nl kgVS1 for initial and aged waste, respectively) was observed in the BMP tests performed after 2-year anaerobic incubation of MSW under conditions of confinement and leachate percolation that replicate those found in real-scale bioreactor landfills [18]. As the total dry organic matter decreased simultaneously during the 2-year anaerobic incubation, the DOCf should be higher than 40.9% if the calculation was based on the same original organic matter. Therefore, 0.5 was used as DOCf to evaluate the methane potential from landfill sites in this study. Decomposable degradable organic carbon (DDOC) can be calculated from the results of anaerobic digestion as 0.0376 for the mixed feedstock. Hence, the average methane production in landfill for the raw materials was calculated as 19.3 m3 t1 according to following equation.
465
digestion, according to the mature technologies and product markets. From the experimental results, the methane production of the digested MBW was 46.6 m3 t1. For BNG recovery, a pressure swing adsorption system was considered. The power consumption for the purification of biogas is a significant GHG source. The capacity of an AD facility is set at 500 t d1 as a case study; thus, the methane production is 21,000 m3 d1 (875 m3 h1); this amount was used to evaluate the power consumption. The default GHG emission factors for power generation are based on a government report concerning low-carbon technologies with fossil fuel, which was released by the Department of Climate Change, National Reform and Development Commission (NDRC) of China in 2009 [19]. The GHG emission associated with power consumption for landfill and AD operations varies with the technological and operational conditions, and has a large uncertainty, so the ideal condition of no GHG emission of power consumption was sat in this study. Hence, the boundary of this evaluation included the direct emission to atmosphere and utilization process of biogas or LFG. 3. Results and discussion 3.1. Reactor performance The reactor was started at 15 Aug. 2009, and had been operated for more than 400 days with 4 different organic loading rates (OLR). Fig. 1 shows a time evolution of operational conditions of the CSTR in terms of daily biogas production, pH value, and OLRs. The reactor was started with OLR of 2.4 kgVS (m3 d)1 and hydraulic retention time (HRT) of 50 days (Phase I). Then, the OLR was increased gradually to 3.6 (Phase II), 4.8 (Phase III), and 6.0 kgVS (m3 d)1 (Phase IV) while HRT was decreased accordingly to 33, 25 and 20 days, respectively. Finally, the reactor was operated stably at OLR of 6.0 kgVS (m3 d)1 for more than 200 days. During the operation, the VS removal rate kept stable at around 65%. The pH value was 7.2e7.6, which is within the appropriate range for mesophilic digestion. Table 2 summarizes the operational conditions of the CSTR reactor. As can be seen that the pH value remains nearly constant, while the VFA concentrations increased slightly from 195 mg L1 to 500 mg L1with the increase of OLR from 2.4 to 6.0 kgVS (m3 d)1. Total alkalinities of the system were detected to be between 11.9 and 13.0 gCaCO3 L1, which has provided a high buffering capacity for the system. Analyses of pH values, VFA, biogas yield and methane yield showed that the OLR increasing procedure did not cause significant adverse effect on AD process. The VFA/alkalinity
22:4 12 0:0376 0:5 0:55 22:4 ¼ ¼ 19:3 m3 t1 12 103
DDOCm ¼ DDOC DOCf Cm
where DDOCm is the methane production of DDOC, m3 t1; Cm, methane concentration of biogas, using the default value of 55%. Power generation and BNG recovery are considered as available alternative technologies for the utilization of biogas from MBW
Fig. 1. Chronological plot of biogas production and pH values in CSTR system.
466
X. Liu et al. / Renewable Energy 44 (2012) 463e468
Table 2 Operational conditions of the CSTR anaerobic digestion system.
Table 4 Comparison of results with references.
Phase
_
II
III
IV
Amount of feedstock/kg d1 OLR/kgVS (m3 d)1 HRT/d Biogas production/m3 d1 VS reduction rate/% pH Total VFA/mg L1 Alkalinity/gCaCO3 L1
32 2.4 50 2.75 0.43 65.6 7.27 0.07 195 9 13.0 0.5
48 3.6 33 3.80 0.58 62.6 7.49 0.06 230 6 12.4 0.7
64 4.8 25 5.60 0.81 64.5 7.48 0.04 250 15 12.6 0.2
80 6.0 20 6.66 0.78 64.9 7.46 0.03 500 17 11.9 0.6
ratio can be used to evaluate the stability of the anaerobic system. When the ratio is less than 0.3e0.4 the process is considered stable and without the risk of acidification. Although VFA concentration increased slightly with the increased OLR, the VFA/alkalinity ratio was far lower than 0.3 with the highest ratio of 0.04 during the experimental period, which illustrated the high stability of the anaerobic system.
Feedstock
Conversion Reference OLR/kgVS HRT/d Biogas production/m3 rate/% (m3 d)1 (m3 d)1
FW:FVW:DSS ¼ 2:1:1 DSS and organic waste Slaughter waste, FVW and manure Fresh OFMSW Mechanically sorted OFMSW Source-sorted OFMSW
6.0 1.4 1.3
20 ee ee
4.25 0.7 0.3
66 60e70 50e67
This study [20] [21]
5.0 19.9
12 6
2.5 4.6
50 26
[22] [23]
4.2
14
2.7
67
[24]
OFMSW, organic fraction of municipal solid waste.
content of 16.5% lipids (dry basis), the biogas production of the feedstock have been increased compared with previously mentioned studies, without inhibition to the anaerobic digestion system.
3.2. Biogas production
3.3. GHG reduction evaluation
The biogas production rate, methane production rate, biogas yield of the feedstock and biogas yield per unit VS removed of four operational phases are listed in Table 3. The volumetric biogas production rate of the system was 4.25 m3 (m3 d)1 in phase IV while the volumetric methane production rate was 2.33 m3 (m3 d)1. Compared with several references listed in Table 4, this research achieved stable operation under relatively high OLR and short HRT, as well as high biogas production and VS removal rate. The biogas yield of the feedstock was 0.66e0.72 m3 kgVS1, corresponding to 45.4e51.4 m3 methane production per ton feedstock, which was 10e20% higher than the reported biogas yield for co-digestion of organic fraction of MSW (OFMSW) and DSS [25,26,10]. Gómez examined the co-digestion of primary sludge and fruitevegetable fractions of MSW with HRT of 37e47 days and OLR of 2.5e3.6 kgVS (m3 d)1, and the results showed that the anaerobic digester reached a biogas yield of 0.2e0.5 m3 kgVS1 [10]. In comparison to this process, the feedstock in this research has achieved a 10% higher biogas yield. The higher biogas production rate is suspected to be due to a better enrichment of the microorganisms in the CSTR system. Also a possible reason is that the feedstock has high lipids content of 16.5% and this is caused by the traditional cooking style of China. Researchers have shown that lipids have a higher methane production rate than protein and carbohydrates, and the lipids at proper content could significantly improve the anaerobic digestion of OFMSW and sludge by enhancing methane production. However higher lipids content may lead to system deterioration caused by the inhibitory effect of long chain fatty acids [27,28]. Basing on the stable operation of pilot-scale reactor for over 400 days, it can be proved that with the
A large amount of LFG, including methane and carbon dioxide, is generated in MSW landfill sites, contributing to global warming. USEPA estimated that the total anthropogenic emission of methane was 282.6 million tons in 2000 [29], of which 13% (36.7 million tons) was due to landfill emissions. For the raw materials used in this study, the average total methane emission was calculated at 19.3 m3 t1. The baseline technology was set as landfill with an average methane capture rate of 50%. The carbon footprint calculation for the landfill scenario is as follows: Landfill scenario (Scenario 1): Step 1: Direct emission of LFG to atmosphere
G11 ¼ Q1 =22:4 44 þ Q2 =22:4 16 F ¼ 19:3=0:55 0:45 0:5=22:4 44 þ 19:3 0:5=22:4 16 21 ¼ 160:3 kgCO2e t1 where Q1, Q2: the volume of methane and CO2; F, value of global warming potential of methane. Step 2: CO2 emission from power generation
G12 ¼ ðQ1 þ Q2 Þ=22:4 44 ¼ 19:3=0:55 0:5=22:4 44 ¼ 34:5 kgCO2e t1
Table 3 Biogas production of the CSTR anaerobic digestion system. Phase
_
II
III
IV
Biogas production rate/m3 (m3 d)1 Biogas yield/m3 kgVS1 CH4/% CH4 production rate/m3 (m3 d)1 CH4 yield/m3 kgVS1 CH4 yield of removed VS/m3 kgVS1 removed CH4 potential/m3 t1
1.72 0.36
2.23 0.46
3.50 0.90
4.25 0.38
0.72 0.15 56.6 1.5 0.97
0.66 0.13 57.4 0.8 1.36
0.73 0.19 58.7 2.3 2.05
0.72 0.06 55.9 2.0 2.33
0.41 0.67
0.38 0.61
0.43 0.67
0.39 0.62
48.7
45.4
51.4
46.8
Step 3: GHG offsetting by power generation
G13 ¼ Q1 C E=3600=1000 EF ¼ 19:3 0:5 35:9 0:3=3600=1000 0:8578 ¼ 24:8 kgCO2e t1
X. Liu et al. / Renewable Energy 44 (2012) 463e468
467
Table 5 Carbon footprint of baseline and AD process. Scenarios
GHG emissions calculation for each activity
Carbon footprint
Landfill (baseline): LFG capture ratio: 50% Power generation
LFG emission: 160.3 kgCO2e t1 LFG combustion: 34.5 kgCO2e t1 Power generation offsetting: 24.8 kgCO2e t1 Emission: 9.1 kgCO2e t1 Biogas combustion: 167.1 kgCO2e t1 Power generation offsetting: 120.1 kgCO2e t1 Emission: 17.5 kgCO2e t1 Power consumption: 10.3 kgCO2e t1 BNG recovery offsetting: 380.3 kgCO2e t1
170.0 kgCO2e t1
AD þ power generation
AD þ BNG recovery
where C, heat value of methane, 35.9 MJ m3; E, power generation efficiency, 0.3; EF, GHG emission factor of power generation with capacity more than 1000 MW, 0.8578 kgCO2e kWh1 was adopted. Therefore, the carbon footprint of Scenario 1 is: G1 ¼ G11 þ G12 G13 ¼ 170.0 kgCO2e t1. In the AD process, biogas is usually used in two ways. The most common way is power generation, and the other way is biogas purifying and separating into bionatural gas and use in terms of piped natural gas, compressed vehicle natural gas, raw chemical materials, etc. In this study, the methane production potential by AD was 46.8 m3 t1, whereas the biogas production potential of the digestate was 1.0 m3 t1 (treated as direct release into the atmosphere). The recovery rate of the purification and separation system was expected to be 95%, while gas impurities and system loss accounted for the missing 5%. The off gas was flared to avoid methane emission. Hence, the carbon footprint calculation of AD with power generation and BNG recovery can be conducted. AD with power generation (Scenario 2) Step 1: Direct emission of digestate biogas to atmosphere
352.5 kgCO2e t1
G31 ¼ 9:1 þ ðQ1 þ Q2 Þ 0:05=22:4 44 ¼ 9:1 þ 167:1 0:05 ¼ 17:5 kgCO2e t1 Step 2: Emission from power consumption The power consumption for the purification of biogas is an essential GHG source. The capacity of an AD facility is set at 500 t d1. Thus, the methane production is 21,000 m3 d1 (875 m3 h1). In the case of BNG recovery, a pressure swing adsorption system is adopted to purify the biogas. The rated power of the purification system is estimated around 250 kW according to some practical projects in China. Thus, the GHG emission from the power consumption can be calculated accordingly.
G32 ¼ 250 24 0:8578=500 ¼ 10:3 kgCO2e t1
Step 3: GHG offsetting by BNG recovery
G33 ¼ Q1 C=3600=1000 EF
G21 ¼ Q1 =22:4 44 þ Q2 =22:4 16 F
¼ 46:8 0:95 35:9=3600=1000 0:8578
¼ 1:0 0:45=22:4 44 þ 1:0 0:55=22:4 16 21
¼ 380:3 kgCO2e t1
¼ 9:1 kgCO2e t1
Therefore, the carbon footprint of Scenario 3 is: G3 ¼ G31 þ G32 G33 ¼ 352.5 kgCO2e t1. From the results summarized in Table 5, there is a GHG reduction up to 114 and 523 kgCO2e t1 for AD with power generation and BNG recovery compared to the baseline, respectively. For an AD facility with a capacity of 500 t d1, the GHG reduction will reach 20,800 and 95,400 tons CO2e annually for the two scenarios.
Step 2: CO2 emission from power generation
G22 ¼ ðQ1 þ Q2 Þ=22:4 44 ¼ 46:8=0:55=22:4 44 ¼ 167:1 kgCO2e t1
4. Conclusions
Step 3: GHG offsetting by power generation
G23 ¼ Q1 C E=3600=1000 EF ¼ 46:8 35:9 0:3=3600=1000 0:8578 ¼ 120:1 kgCO2e t1 Therefore, the carbon footprint G2 ¼ G21 þ G22 G23 ¼ 56.1 kgCO2e t1
56.1 kgCO2e t1
of
Scenario
2
is:
In a pilot-scale anaerobic co-digestion system of FW, FVW and DSS (capacity of 80 kg d1), stable operation was achieved with a conversion ratio of 64.5% at OLR of 6.0 kgVS (m3 d)1 and HRT of 20 d. The system had a biogas production rate of 4.25 m3 (m3 d)1 and a biogas yield of 0.72 m3 kgVS1 added, while the 16.5% lipids content was beneficial to the biogas production of the feedstock without inhibition to anaerobic digestion. The GHG reduction is considered an important environmental benefit of MBW digestion. Compared to the landfill baseline, GHG reduction reaches 114 and 523 kgCO2e t1 for AD with power generation and BNG recovery, respectively. Acknowledgements
AD with BNG recovery (Scenario 3): Step 1: Direct emission of the digestate biogas and off gas flaring into the atmosphere
This research was partially funded by the National High Technology Research and Development Program of China (No.
468
X. Liu et al. / Renewable Energy 44 (2012) 463e468
2008AA062401), Ministry of Science and Technology of China (2010DFA22770) and by China Postdoctoral Science Foundation funded project (201003105). References [1] National Bureau of Statistics of China. China statistical yearbook. Available from: http://www.stats.gov.cn/tjsj/ndsj/2010/indexch.htm; 2010. [2] Luste S, Luostarinen S. Anaerobic co-digestion of meat-processing by-products and sewage sludge e effect of hygienization and organic loading rate. Bioresour Technol 2010;101:2657e64. [3] Mattheeuws B. Anaerobic digestion state of the art 2010. NV: Organic Waste Systems. Available from: http://www.amb.cat/c/document_library/get_file?p_l_ id¼5694&folderId¼264924&name¼DLFE-7047.pdf; 2010. [4] Edelmann W, Schleiss K, Joss A. Ecological, energetic and economic comparison of anaerobic digestion with different competing technologies to treat biogenic wastes. Water Sci Technol 2000;41:263e73. [5] Di Stefano TD, Belenky LG. Life-cycle analysis of energy and greenhouse gas emissions from anaerobic biodegradation of municipal solid waste. J Environ Eng 2009;135:1097e105. [6] De Baere L. Will anaerobic digestion of solid waste survive in the future? Water Sci Technol 2006;53:187e94. [7] Hartmann H, Ahring BK. Anaerobic digestion of the organic fraction of municipal solid waste: influence of co-digestion with manure. Water Res 2005;39:1543e52. [8] Bouallagui H, Lahdheb H, Romdan EB, Rachdi B, Hamdi M. Improvement of fruit and vegetable waste anaerobic digestion performance and stability with co-substrates addition. J Environ Manage 2009;90:1844e9. [9] Holm-Nielsen JB, Seadi TA, Oleskowicz-Popiel P. The future of anaerobic digestion and biogas utilization. Bioresour Technol 2009;100:547e5484. [10] Gómez X, Cuetos MJ, Cara J, Morán A, García AI. Anaerobic co-digestion of primary sludge and the fruit and vegetable fraction of the municipal solid wastes: conditions for mixing and evaluation of the organic loading rate. Renew Energy 2006;31:2017e24. [11] Mata-Alvarez J, Mace S, Llabres P. Anaerobic digestion of organic solid wastes. An overview of research achievements and perspectives. Bioresour. Technol 2000;74:3e16. [12] Yao R, Li B, Steemers K. Energy policy and standard for built environment in China. Renew Energy 2005;30:1973e88. [13] Müller JA, Winter A, Strünkmann G. Investigation and assessment of sludge pre-treatment processes. Water Sci Technol 2004;49:97e104.
[14] Wang W, Hou H, Hu S, Gao X. Performance and stability improvement in anaerobic digestion of thermally hydrolyzed municipal biowaste by a biofilm system. Bioresour Technol 2010;101:1715e21. [15] BSI. Guide to PAS 2050: how to assess the carbon footprint of goods and services. London: British Standards; 2008. [16] Raninger B. Renewable energy for rural areas in China e GTZ training; August 2007. Shenyang, China. [17] IPCC. 2006 IPCC guidelines for national greenhouse gas inventories. Prepared by the national greenhouse gas inventories programme. Japan: IGES; 2006. [18] Olivier F, Gourc JP, Achour F, Morais J, Bayard R. Evolution of bio-physical and mechanical characteristics of MSW after 2 years incubation in a laboratoryscale bioreactor. Proc 10th international waste and landfill symposium, Sardinia, Cagliari, Italy; 2005. [19] Department of Climate Change. National Reform and Development Commission (NDRC) of China. Baseline emission factor of grid power generation projects of low-carbon technology and fossil fuels. Available from: http://cdm. ccchina.gov.cn/WebSite/CDM/UpFile/File2537.pdf; 2009. N, Ros M. Full-scale anaerobic co-digestion of [20] Zupan ci c GD, Uranjek-Zevart organic waste and municipal sludge. Biomass Bioenergy 2008;32:162e7. [21] Alvarez R, Lidén G. Semi-continuous co-digestion of solid slaughterhouse waste, manure, and fruit and vegetable waste. Renew Energy 2008;33: 726e34. [22] Krzystek L, Ledakowicz S, Kahle HJ, Kaczorek K. Degradation of household biowaste in reactors. J Biotechnol 2001;92:103e12. [23] Cecchi F, Mata-Alvarez J, Pavan P, Sans C, Merli C. Semidry anaerobic digestion of MSW e influence of process parameters on the substrate utilization model. Water Sci Technol 1992;25:83e92. [24] Mata-Alvarez J, Cecchi F, Pavan P, Llabres P. The performance of digesters treating the organic fraction of municipal solid wastes differently sorted. Biological Wastes 1990;33:181e99. [25] Sosnowski P, Wieczorek A, Ledakowicz S. Anaerobic co-digestion of sewage sludge and organic fraction of municipal solid wastes. Adv Environ Res 2003; 7:609e16. [26] Capela I, Rodrigues A, Silva F, Nadais H, Arroja L. Impact of industrial sludge and cattle manure on anaerobic digestion of the OFMSW under mesophilic conditions. Biomass Bioenergy 2008;32:245e51. [27] Zhu Z, Hsueh MK, He Q. Enhancing biomethanation of municipal waste sludge with grease trap waste as a co-substrate. Renew Energy 2011;32: 1802e7. [28] Davidsson Å, Lövstedt C, Jansen JC, Gruvberger C, Aspegren H. Co-digestion of grease trap sludge and sewage sludge. Waste Manage 2008;28:986e92. [29] USEPA. International analyses of methane emissions, www.epa.gov/methane/ intlanalyses.html; 2002.
Contents lists available at SciVerse ScienceDirect
Renewable Energy journal homepage: www.elsevier.com/locate/renene
Technical note
Pilot-scale anaerobic co-digestion of municipal biomass waste: Focusing on biogas production and GHG reduction Xiao Liu a, *, Xingbao Gao a, b, Wei Wang a, Lei Zheng a, Yingjun Zhou c, Yifei Sun d a
School of Environment, Tsinghua University, Beijing 100084, China Chinese Research Academy of Environmental Sciences, Beijing 100012, China c Department of Urban and Environmental Engineering, Graduate School of Engineering, Kyoto University, Katsura, Nisikyo-ku, Kyoto 615-8540, Japan d School of Chemistry and Environment, Beihang University, Beijing 100191, China b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 April 2011 Accepted 24 January 2012 Available online 16 February 2012
A pilot-scale anaerobic co-digestion research study is presented to elucidate the feasibility of developing anaerobic digestion (AD) as an effective disposal method for municipal biomass waste (MBW) in China, focusing on biogas production and greenhouse gas (GHG) reduction. Food waste, fruitevegetable waste, and dewatered sewage sludge were co-digested in a continuous stirred-tank reactor for biogas production. Stable operation was achieved with a high biogas production rate of 4.25 m3 (m3 d)1 at organic loading rate of 6.0 kgVS (m3 d)1 and hydraulic retention time of 20 d. A total of 16.5% of lipids content was beneficial to the biogas production of the feedstock without inhibition to anaerobic digestion. Compared with the landfill baseline, GHG reduction is an important environmental benefit from MBW digestion. Therefore, anaerobic co-digestion is a promising alternative solution for MBW because it contributes significantly to the sound management of municipal solid waste in China. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Municipal biomass waste Anaerobic co-digestion Biogas production GHG reduction
1. Introduction With the rapid economic development and urbanization of China, almost all of its megacities are faced with the problem of municipal solid waste (MSW) disposal. In 2009, a total of 157.4 million tons of MSW was collected and transported nationwide. Of these, 89.0, 20.2, and 1.8 million tons were treated by landfill, incineration, and composting, respectively [1]. With increasing quantity, 50e60% of the MSW was biomass waste characterized by high water and biodegradable organic content. The municipal biomass waste (MBW) has led to serious adverse effects in traditional MSW treatment systems (i.e. landfill and incineration). The high water content may cause abundant production of leachate for landfill, and may cause unstable burning conditions and dioxin release from incineration. Also the high biodegradable organic content may cause the production and emission of greenhouse gas (GHG). From 2005 to 2010, the wastewater treatment capacity in China has increased from 60 million tons to 125 million tons which has resulted in the rapid increase of sewage sludge and the disposal of sewage sludge has also been a big problem in almost all the cities
* Corresponding author. Tel.: þ86 1062772814; fax: þ86 1062782910. E-mail addresses: [email protected], [email protected] (X. Liu). 0960-1481/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.renene.2012.01.092
in China. In some cities, the sewage sludge is simply deposited onto the wasteland without any treatment, which has caused serious pollution. Efficient MBW management technology is increasingly required due to environmental and economical concerns, such as climate change, eutrophication, and the diminishing resources of fossil energy and raw materials. Anaerobic digestion (AD) is considered as a sustainable option for the management of biomass wastes because the production of renewable energy and the recycling of nutrients [2]. Additionally, MBW separated from MSW and treated with AD can significantly reduce the load of traditional disposal facilities, and subsequently prolong their service life. It also decreases the secondary pollutants originated from the biodegradation of organic wastes during landfill, incineration and composting. AD has been employed in Western Europe since the 1980s, while up to 2010 one-hundred and ninety-five facilities have been constructed with a total annual capacity of 5.9 million tons [3]. De Baere concluded that AD facilities have captured the major market of waste treatment in the EU in the last decade, and will certainly continue developing in the future. AD is expected to become global utilised because of its environmental contribution, and energy benefits [4e6]. Codigestion of different types of MBW has been discussed in many references due to its potential of increasing biogas output and improving stability of anaerobic system [4,7e10]. Co-digestion also means more feedstock supply, which is especially needed by large-
464
X. Liu et al. / Renewable Energy 44 (2012) 463e468
scale plants where large quantities of MBW are required. MataAlvarez et al. summarized several cases of co-digestion of MBW, in both research and practical use [11]. Given that traditional technologies may cause severe environmental pollution for treating MBW, alternative environmentfriendly treatment technologies of MBW are much needed in China. Meanwhile, renewable energy recovered from the biomass waste has become a global concern, and bioenergy has been listed in the Chinese New Energy Promotion Plan [12]. The AD of MBW can reduce the GHG emission in two respects: reduction as compared with the baseline management, and reduction through providing alternative resources in terms of non-renewable fossil energy and materials. In 2009, 79% of treated MSW was contributed by the landfill, and landfill is considered as one of the main GHG emission sources in current China [1]. The development of AD can restructure the traditional MBW treatment and disposal system in China, and in turn achieve the reduction of GHG emissions. Along with other advantages mentioned earlier, AD has attracted much attention of both government and enterprises in recent years, and it can be expected that AD will be widely used in the near future in China. At present, the support policy from the central government has made every megacity receptive to building large-scale AD facilities in China. However, due to the differences of geographic locality and dietary habit, MBWs especially food waste in China has the typical characteristics of high water content (>85%) and high lipids content (>20% of dry basis) which is different from other countries, and few reports have discussed the AD of Chinese MBWs. Therefore, whether the Chinese MBWs fit for AD, and what is the suitable operation condition, these questions need to be answered before the construction and operation of the large-scale AD projects in China. Thus, a systematic research of AD of Chinese MBWs is urgently needed to provide fundamental technology parameters and promote the commercialization of AD technology. Finally, as the second biggest producer of carbon dioxide, China is facing great pressure in reducing GHG emissions, so the specific GHG reduction which can be achieved during AD of MBW is gaining interesting and needs to be quantified. In this study, a pilot-scale anaerobic codigestion research is presented to elucidate the feasibility of developing AD in China as an effective disposal method for MBW, focusing on the system performance and biogas production, and then GHG reduction of AD was analyzed compared with landfill. 2. Methods 2.1. Raw materials MBW used in this experiment comprised food waste (FW), fruitvegetable waste (FVW), and dewatered sewage sludge (DSS). FW was collected from a student canteen (capacity, over 1000 students for a seated dinner) at Tsinghua University, FVW from a wholesale market, and DSS from a municipal wastewater treatment plant (WWTP) in Beijing (Qinghe WWTP; Northern Beijing; capacity, 400,000 m3 d1). The inert materials in FW and FVW were manually separated (e.g., plastic, bone, wood, and others). FW and FVW were crushed to less than 3 mm size firstly by a food waste pulverizer, and then mixed with DSS at the ratio of 2:1:1. The mixed feedstock was kept at 4 C before use. During the experiment, a maximum of 80 kg d1 mixture was fed to the AD reactor. The characterization of raw materials is shown in Table 1 in terms of solid content, organic composition, and elemental composition. Volatile solid (VS), standing for organic content, accounts for 65e90% of the total solid (TS). VS can be divided into volatile dissolved solid (VDS) and volatile suspended solid (VSS). VSS/VS, representing the organic solid ratio in organic fraction,
Table 1 Characterization of raw materials.
Water content/% Total solid/g L1 Volatile solid/g L1 Suspended solid/g L1 Volatile suspended solid/g L1 VS/TS/% VSS/VS/% Lipids/%TS Protein/%TS Crude fiber/%TS C/% H/% N/% C/N
FW
FVW
DSS
Mixture (2:1:1)
83.4 166.3 26.7 149.0 24.3 72.8 14.3 68.8 12.0
93.8 62.2 16.0 50.8 11.2 35.7 14.2 29.6 11.2
84.5 154.9 18.1 101.9 10.8 151.7 21.4 98.5 12.8
88.7 142.1 9.3 117.3 7.8 91.7 15.0 74.0 12.2
89.6 46.2 21.8 16.8 5.6 48.2 7.3 2.8 17.4
81.6 58.3 2.9 13.2 15.3 42.0 6.1 2.4 17.4
65.8 96.6 10.3 34.3 7.1 37.2 5.5 5.9 6.3
82.5 63.1 16.5 20.8 6.4 45.4 6.6 3.5 12.9
accounts for 46.2%, 58.3% and 96.6% for FW, FVW, and DSS, respectively, indicating that organic solid plays an important role in the AD process, which has been proven to be the essential ratelimiting factor [13,14]. The organic compositions of the three raw materials differ significantly. For FW, the lipids content is the highest because of the Chinese traditional cooking style, whereas for FVW, the crude fiber is most abundant because of the cellulose content in fruits and vegetables. For DSS, the protein content is the highest compared with KW and VFR because of the high content of microorganisms. The C/N ratio of FW and FVW is suitable for AD; however, the C/N ratio of DSS is too low for AD. In the traditional AD of sewage sludge, slow degradation (>20 days), low organic loading rate, and the relatively low VS removal (30e40%) are often the disadvantages of the process because the digesters are operated with too low C/N ratios. Hence, in this study, DSS was co-digested with FW and FVW of higher C/N ratios to cover the adverse effects caused by its low C/N ratio. 2.2. Pilot-scale reactor start-up and operation Continuous stirred-tank reactor (CSTR) was used in the research. The reactor had a volume of 2 m3 (effective volume, 1.6 m3; height, 2.7 m; and diameter, 1.0 m). The feedstock of the reactor was a mixture of 50% of FW, 25% of FVW and 25% of DSS in the percentage of weight. Inoculums (digested sewage sludge) were collected from another municipal wastewater treatment plant (Xiaohongmen WWTP, located in southern Beijing), where the excess sewage sludge was treated by AD. A 1.6 m3 volume of digested sludge was collected and pumped into the reactor. The TS, SS, VS, and VSS contents of the inoculums were 27.0, 25.0, 10.4, and 9.7 g L1, respectively. The reactor was operated under mesophilic condition at 35 2 C by a water jacket. The reactors were constantly mixed using mechanical stirrers (100 rpm) with an agitation time of 15 min per two hours, and were fed once a day using a screw pump. 2.3. Analytical techniques The analysis of TS, VS, SS and VSS were based on the Standard Analytic Methods promulgated by the National Environmental Protection Agency of China (1989). The measurement of crude fiber and protein were according to ISO 6865:2000 and ISO1871:2009, respectively. Lipids content was determined using soxhlet extraction method according to ISO 6492:1999. The samples were filtered through 0.45 mm filters before the measurement of volatile fatty acids (VFAs) using gas chromatography (SHIMADZU GC-2010) with
X. Liu et al. / Renewable Energy 44 (2012) 463e468
a flame ionization detector and GDX-102 column (inlet, 200 C; oven, 170 C; and detector, 220 C). The various VFAs included acetic, propionic, iso-butyric, butyric, iso-valeric, and valeric acids. The biogas volume was measured with a wet-test gas flow meter, and the composition of the biogas was monitored by gas chromatography (SHIMADZU GC-2010) with a thermal conductivity detector and RT-Qplot column (column, 30 m 0.53 mm; flame ionization detector, 200 C; oven, 50 C; inlet, 200 C; carrier gas, hydrogen 10 ml min1; split ratio, 35:1, and injection volume, 100 ml). The C, H, and N were analyzed by an elemental analyzer (EAI CE-440). 2.4. Calculation of the GHG reduction The GHG emission of MBW treatment was evaluated using the analytical method of carbon footprint following the Guideline to PAS 2050 [15]. In this study, the carbon footprint of three scenarios was assessed which were baseline scenario (scenario 1), AD with power generation (scenario 2), and AD with bio natural gas (BNG) recovery (scenario 3). In China more than 70% of treated MSW was disposed by landfill, therefore, the baseline scenario was set as MBW landfill. The collection rate of landfill gas (LFG) is generally less than 60% in the developed countries, whereas to achieve a 20% gas recovery in China appears to be difficult [16]. In this study, an average LFG collection rate of 50% was set as the ideal condition resulting from the strict standards and laws concerning with the national energy saving and emission reduction strategy. Analyses of currently operating LFG-to-electricity projects in China show that power generated through LFG recovery and utilization is approximately 30 kWh per ton of waste. There is an evident difference compared with the 250e300 kWh per ton of waste via incineration so, the energy recovered from LFG is fairly limited. However, at present, power generation is a widely used method to obtain energy and reduce global warming potential of methane. Hence, LFG is captured to generate power in the baseline scenario. The methane potential for MBW in a landfill site was calculated following the 2006 IPCC Guidelines for National Greenhouse Gas Inventories [17], which suggested 0.5 as the fraction of the degradable organic carbon (DOCf) that can decompose under anaerobic condition of wet bulk waste. According to Olivier et al., an average DOCf about 40.9% (181 and 107 Nl kgVS1 for initial and aged waste, respectively) was observed in the BMP tests performed after 2-year anaerobic incubation of MSW under conditions of confinement and leachate percolation that replicate those found in real-scale bioreactor landfills [18]. As the total dry organic matter decreased simultaneously during the 2-year anaerobic incubation, the DOCf should be higher than 40.9% if the calculation was based on the same original organic matter. Therefore, 0.5 was used as DOCf to evaluate the methane potential from landfill sites in this study. Decomposable degradable organic carbon (DDOC) can be calculated from the results of anaerobic digestion as 0.0376 for the mixed feedstock. Hence, the average methane production in landfill for the raw materials was calculated as 19.3 m3 t1 according to following equation.
465
digestion, according to the mature technologies and product markets. From the experimental results, the methane production of the digested MBW was 46.6 m3 t1. For BNG recovery, a pressure swing adsorption system was considered. The power consumption for the purification of biogas is a significant GHG source. The capacity of an AD facility is set at 500 t d1 as a case study; thus, the methane production is 21,000 m3 d1 (875 m3 h1); this amount was used to evaluate the power consumption. The default GHG emission factors for power generation are based on a government report concerning low-carbon technologies with fossil fuel, which was released by the Department of Climate Change, National Reform and Development Commission (NDRC) of China in 2009 [19]. The GHG emission associated with power consumption for landfill and AD operations varies with the technological and operational conditions, and has a large uncertainty, so the ideal condition of no GHG emission of power consumption was sat in this study. Hence, the boundary of this evaluation included the direct emission to atmosphere and utilization process of biogas or LFG. 3. Results and discussion 3.1. Reactor performance The reactor was started at 15 Aug. 2009, and had been operated for more than 400 days with 4 different organic loading rates (OLR). Fig. 1 shows a time evolution of operational conditions of the CSTR in terms of daily biogas production, pH value, and OLRs. The reactor was started with OLR of 2.4 kgVS (m3 d)1 and hydraulic retention time (HRT) of 50 days (Phase I). Then, the OLR was increased gradually to 3.6 (Phase II), 4.8 (Phase III), and 6.0 kgVS (m3 d)1 (Phase IV) while HRT was decreased accordingly to 33, 25 and 20 days, respectively. Finally, the reactor was operated stably at OLR of 6.0 kgVS (m3 d)1 for more than 200 days. During the operation, the VS removal rate kept stable at around 65%. The pH value was 7.2e7.6, which is within the appropriate range for mesophilic digestion. Table 2 summarizes the operational conditions of the CSTR reactor. As can be seen that the pH value remains nearly constant, while the VFA concentrations increased slightly from 195 mg L1 to 500 mg L1with the increase of OLR from 2.4 to 6.0 kgVS (m3 d)1. Total alkalinities of the system were detected to be between 11.9 and 13.0 gCaCO3 L1, which has provided a high buffering capacity for the system. Analyses of pH values, VFA, biogas yield and methane yield showed that the OLR increasing procedure did not cause significant adverse effect on AD process. The VFA/alkalinity
22:4 12 0:0376 0:5 0:55 22:4 ¼ ¼ 19:3 m3 t1 12 103
DDOCm ¼ DDOC DOCf Cm
where DDOCm is the methane production of DDOC, m3 t1; Cm, methane concentration of biogas, using the default value of 55%. Power generation and BNG recovery are considered as available alternative technologies for the utilization of biogas from MBW
Fig. 1. Chronological plot of biogas production and pH values in CSTR system.
466
X. Liu et al. / Renewable Energy 44 (2012) 463e468
Table 2 Operational conditions of the CSTR anaerobic digestion system.
Table 4 Comparison of results with references.
Phase
_
II
III
IV
Amount of feedstock/kg d1 OLR/kgVS (m3 d)1 HRT/d Biogas production/m3 d1 VS reduction rate/% pH Total VFA/mg L1 Alkalinity/gCaCO3 L1
32 2.4 50 2.75 0.43 65.6 7.27 0.07 195 9 13.0 0.5
48 3.6 33 3.80 0.58 62.6 7.49 0.06 230 6 12.4 0.7
64 4.8 25 5.60 0.81 64.5 7.48 0.04 250 15 12.6 0.2
80 6.0 20 6.66 0.78 64.9 7.46 0.03 500 17 11.9 0.6
ratio can be used to evaluate the stability of the anaerobic system. When the ratio is less than 0.3e0.4 the process is considered stable and without the risk of acidification. Although VFA concentration increased slightly with the increased OLR, the VFA/alkalinity ratio was far lower than 0.3 with the highest ratio of 0.04 during the experimental period, which illustrated the high stability of the anaerobic system.
Feedstock
Conversion Reference OLR/kgVS HRT/d Biogas production/m3 rate/% (m3 d)1 (m3 d)1
FW:FVW:DSS ¼ 2:1:1 DSS and organic waste Slaughter waste, FVW and manure Fresh OFMSW Mechanically sorted OFMSW Source-sorted OFMSW
6.0 1.4 1.3
20 ee ee
4.25 0.7 0.3
66 60e70 50e67
This study [20] [21]
5.0 19.9
12 6
2.5 4.6
50 26
[22] [23]
4.2
14
2.7
67
[24]
OFMSW, organic fraction of municipal solid waste.
content of 16.5% lipids (dry basis), the biogas production of the feedstock have been increased compared with previously mentioned studies, without inhibition to the anaerobic digestion system.
3.2. Biogas production
3.3. GHG reduction evaluation
The biogas production rate, methane production rate, biogas yield of the feedstock and biogas yield per unit VS removed of four operational phases are listed in Table 3. The volumetric biogas production rate of the system was 4.25 m3 (m3 d)1 in phase IV while the volumetric methane production rate was 2.33 m3 (m3 d)1. Compared with several references listed in Table 4, this research achieved stable operation under relatively high OLR and short HRT, as well as high biogas production and VS removal rate. The biogas yield of the feedstock was 0.66e0.72 m3 kgVS1, corresponding to 45.4e51.4 m3 methane production per ton feedstock, which was 10e20% higher than the reported biogas yield for co-digestion of organic fraction of MSW (OFMSW) and DSS [25,26,10]. Gómez examined the co-digestion of primary sludge and fruitevegetable fractions of MSW with HRT of 37e47 days and OLR of 2.5e3.6 kgVS (m3 d)1, and the results showed that the anaerobic digester reached a biogas yield of 0.2e0.5 m3 kgVS1 [10]. In comparison to this process, the feedstock in this research has achieved a 10% higher biogas yield. The higher biogas production rate is suspected to be due to a better enrichment of the microorganisms in the CSTR system. Also a possible reason is that the feedstock has high lipids content of 16.5% and this is caused by the traditional cooking style of China. Researchers have shown that lipids have a higher methane production rate than protein and carbohydrates, and the lipids at proper content could significantly improve the anaerobic digestion of OFMSW and sludge by enhancing methane production. However higher lipids content may lead to system deterioration caused by the inhibitory effect of long chain fatty acids [27,28]. Basing on the stable operation of pilot-scale reactor for over 400 days, it can be proved that with the
A large amount of LFG, including methane and carbon dioxide, is generated in MSW landfill sites, contributing to global warming. USEPA estimated that the total anthropogenic emission of methane was 282.6 million tons in 2000 [29], of which 13% (36.7 million tons) was due to landfill emissions. For the raw materials used in this study, the average total methane emission was calculated at 19.3 m3 t1. The baseline technology was set as landfill with an average methane capture rate of 50%. The carbon footprint calculation for the landfill scenario is as follows: Landfill scenario (Scenario 1): Step 1: Direct emission of LFG to atmosphere
G11 ¼ Q1 =22:4 44 þ Q2 =22:4 16 F ¼ 19:3=0:55 0:45 0:5=22:4 44 þ 19:3 0:5=22:4 16 21 ¼ 160:3 kgCO2e t1 where Q1, Q2: the volume of methane and CO2; F, value of global warming potential of methane. Step 2: CO2 emission from power generation
G12 ¼ ðQ1 þ Q2 Þ=22:4 44 ¼ 19:3=0:55 0:5=22:4 44 ¼ 34:5 kgCO2e t1
Table 3 Biogas production of the CSTR anaerobic digestion system. Phase
_
II
III
IV
Biogas production rate/m3 (m3 d)1 Biogas yield/m3 kgVS1 CH4/% CH4 production rate/m3 (m3 d)1 CH4 yield/m3 kgVS1 CH4 yield of removed VS/m3 kgVS1 removed CH4 potential/m3 t1
1.72 0.36
2.23 0.46
3.50 0.90
4.25 0.38
0.72 0.15 56.6 1.5 0.97
0.66 0.13 57.4 0.8 1.36
0.73 0.19 58.7 2.3 2.05
0.72 0.06 55.9 2.0 2.33
0.41 0.67
0.38 0.61
0.43 0.67
0.39 0.62
48.7
45.4
51.4
46.8
Step 3: GHG offsetting by power generation
G13 ¼ Q1 C E=3600=1000 EF ¼ 19:3 0:5 35:9 0:3=3600=1000 0:8578 ¼ 24:8 kgCO2e t1
X. Liu et al. / Renewable Energy 44 (2012) 463e468
467
Table 5 Carbon footprint of baseline and AD process. Scenarios
GHG emissions calculation for each activity
Carbon footprint
Landfill (baseline): LFG capture ratio: 50% Power generation
LFG emission: 160.3 kgCO2e t1 LFG combustion: 34.5 kgCO2e t1 Power generation offsetting: 24.8 kgCO2e t1 Emission: 9.1 kgCO2e t1 Biogas combustion: 167.1 kgCO2e t1 Power generation offsetting: 120.1 kgCO2e t1 Emission: 17.5 kgCO2e t1 Power consumption: 10.3 kgCO2e t1 BNG recovery offsetting: 380.3 kgCO2e t1
170.0 kgCO2e t1
AD þ power generation
AD þ BNG recovery
where C, heat value of methane, 35.9 MJ m3; E, power generation efficiency, 0.3; EF, GHG emission factor of power generation with capacity more than 1000 MW, 0.8578 kgCO2e kWh1 was adopted. Therefore, the carbon footprint of Scenario 1 is: G1 ¼ G11 þ G12 G13 ¼ 170.0 kgCO2e t1. In the AD process, biogas is usually used in two ways. The most common way is power generation, and the other way is biogas purifying and separating into bionatural gas and use in terms of piped natural gas, compressed vehicle natural gas, raw chemical materials, etc. In this study, the methane production potential by AD was 46.8 m3 t1, whereas the biogas production potential of the digestate was 1.0 m3 t1 (treated as direct release into the atmosphere). The recovery rate of the purification and separation system was expected to be 95%, while gas impurities and system loss accounted for the missing 5%. The off gas was flared to avoid methane emission. Hence, the carbon footprint calculation of AD with power generation and BNG recovery can be conducted. AD with power generation (Scenario 2) Step 1: Direct emission of digestate biogas to atmosphere
352.5 kgCO2e t1
G31 ¼ 9:1 þ ðQ1 þ Q2 Þ 0:05=22:4 44 ¼ 9:1 þ 167:1 0:05 ¼ 17:5 kgCO2e t1 Step 2: Emission from power consumption The power consumption for the purification of biogas is an essential GHG source. The capacity of an AD facility is set at 500 t d1. Thus, the methane production is 21,000 m3 d1 (875 m3 h1). In the case of BNG recovery, a pressure swing adsorption system is adopted to purify the biogas. The rated power of the purification system is estimated around 250 kW according to some practical projects in China. Thus, the GHG emission from the power consumption can be calculated accordingly.
G32 ¼ 250 24 0:8578=500 ¼ 10:3 kgCO2e t1
Step 3: GHG offsetting by BNG recovery
G33 ¼ Q1 C=3600=1000 EF
G21 ¼ Q1 =22:4 44 þ Q2 =22:4 16 F
¼ 46:8 0:95 35:9=3600=1000 0:8578
¼ 1:0 0:45=22:4 44 þ 1:0 0:55=22:4 16 21
¼ 380:3 kgCO2e t1
¼ 9:1 kgCO2e t1
Therefore, the carbon footprint of Scenario 3 is: G3 ¼ G31 þ G32 G33 ¼ 352.5 kgCO2e t1. From the results summarized in Table 5, there is a GHG reduction up to 114 and 523 kgCO2e t1 for AD with power generation and BNG recovery compared to the baseline, respectively. For an AD facility with a capacity of 500 t d1, the GHG reduction will reach 20,800 and 95,400 tons CO2e annually for the two scenarios.
Step 2: CO2 emission from power generation
G22 ¼ ðQ1 þ Q2 Þ=22:4 44 ¼ 46:8=0:55=22:4 44 ¼ 167:1 kgCO2e t1
4. Conclusions
Step 3: GHG offsetting by power generation
G23 ¼ Q1 C E=3600=1000 EF ¼ 46:8 35:9 0:3=3600=1000 0:8578 ¼ 120:1 kgCO2e t1 Therefore, the carbon footprint G2 ¼ G21 þ G22 G23 ¼ 56.1 kgCO2e t1
56.1 kgCO2e t1
of
Scenario
2
is:
In a pilot-scale anaerobic co-digestion system of FW, FVW and DSS (capacity of 80 kg d1), stable operation was achieved with a conversion ratio of 64.5% at OLR of 6.0 kgVS (m3 d)1 and HRT of 20 d. The system had a biogas production rate of 4.25 m3 (m3 d)1 and a biogas yield of 0.72 m3 kgVS1 added, while the 16.5% lipids content was beneficial to the biogas production of the feedstock without inhibition to anaerobic digestion. The GHG reduction is considered an important environmental benefit of MBW digestion. Compared to the landfill baseline, GHG reduction reaches 114 and 523 kgCO2e t1 for AD with power generation and BNG recovery, respectively. Acknowledgements
AD with BNG recovery (Scenario 3): Step 1: Direct emission of the digestate biogas and off gas flaring into the atmosphere
This research was partially funded by the National High Technology Research and Development Program of China (No.
468
X. Liu et al. / Renewable Energy 44 (2012) 463e468
2008AA062401), Ministry of Science and Technology of China (2010DFA22770) and by China Postdoctoral Science Foundation funded project (201003105). References [1] National Bureau of Statistics of China. China statistical yearbook. Available from: http://www.stats.gov.cn/tjsj/ndsj/2010/indexch.htm; 2010. [2] Luste S, Luostarinen S. Anaerobic co-digestion of meat-processing by-products and sewage sludge e effect of hygienization and organic loading rate. Bioresour Technol 2010;101:2657e64. [3] Mattheeuws B. Anaerobic digestion state of the art 2010. NV: Organic Waste Systems. Available from: http://www.amb.cat/c/document_library/get_file?p_l_ id¼5694&folderId¼264924&name¼DLFE-7047.pdf; 2010. [4] Edelmann W, Schleiss K, Joss A. Ecological, energetic and economic comparison of anaerobic digestion with different competing technologies to treat biogenic wastes. Water Sci Technol 2000;41:263e73. [5] Di Stefano TD, Belenky LG. Life-cycle analysis of energy and greenhouse gas emissions from anaerobic biodegradation of municipal solid waste. J Environ Eng 2009;135:1097e105. [6] De Baere L. Will anaerobic digestion of solid waste survive in the future? Water Sci Technol 2006;53:187e94. [7] Hartmann H, Ahring BK. Anaerobic digestion of the organic fraction of municipal solid waste: influence of co-digestion with manure. Water Res 2005;39:1543e52. [8] Bouallagui H, Lahdheb H, Romdan EB, Rachdi B, Hamdi M. Improvement of fruit and vegetable waste anaerobic digestion performance and stability with co-substrates addition. J Environ Manage 2009;90:1844e9. [9] Holm-Nielsen JB, Seadi TA, Oleskowicz-Popiel P. The future of anaerobic digestion and biogas utilization. Bioresour Technol 2009;100:547e5484. [10] Gómez X, Cuetos MJ, Cara J, Morán A, García AI. Anaerobic co-digestion of primary sludge and the fruit and vegetable fraction of the municipal solid wastes: conditions for mixing and evaluation of the organic loading rate. Renew Energy 2006;31:2017e24. [11] Mata-Alvarez J, Mace S, Llabres P. Anaerobic digestion of organic solid wastes. An overview of research achievements and perspectives. Bioresour. Technol 2000;74:3e16. [12] Yao R, Li B, Steemers K. Energy policy and standard for built environment in China. Renew Energy 2005;30:1973e88. [13] Müller JA, Winter A, Strünkmann G. Investigation and assessment of sludge pre-treatment processes. Water Sci Technol 2004;49:97e104.
[14] Wang W, Hou H, Hu S, Gao X. Performance and stability improvement in anaerobic digestion of thermally hydrolyzed municipal biowaste by a biofilm system. Bioresour Technol 2010;101:1715e21. [15] BSI. Guide to PAS 2050: how to assess the carbon footprint of goods and services. London: British Standards; 2008. [16] Raninger B. Renewable energy for rural areas in China e GTZ training; August 2007. Shenyang, China. [17] IPCC. 2006 IPCC guidelines for national greenhouse gas inventories. Prepared by the national greenhouse gas inventories programme. Japan: IGES; 2006. [18] Olivier F, Gourc JP, Achour F, Morais J, Bayard R. Evolution of bio-physical and mechanical characteristics of MSW after 2 years incubation in a laboratoryscale bioreactor. Proc 10th international waste and landfill symposium, Sardinia, Cagliari, Italy; 2005. [19] Department of Climate Change. National Reform and Development Commission (NDRC) of China. Baseline emission factor of grid power generation projects of low-carbon technology and fossil fuels. Available from: http://cdm. ccchina.gov.cn/WebSite/CDM/UpFile/File2537.pdf; 2009. N, Ros M. Full-scale anaerobic co-digestion of [20] Zupan ci c GD, Uranjek-Zevart organic waste and municipal sludge. Biomass Bioenergy 2008;32:162e7. [21] Alvarez R, Lidén G. Semi-continuous co-digestion of solid slaughterhouse waste, manure, and fruit and vegetable waste. Renew Energy 2008;33: 726e34. [22] Krzystek L, Ledakowicz S, Kahle HJ, Kaczorek K. Degradation of household biowaste in reactors. J Biotechnol 2001;92:103e12. [23] Cecchi F, Mata-Alvarez J, Pavan P, Sans C, Merli C. Semidry anaerobic digestion of MSW e influence of process parameters on the substrate utilization model. Water Sci Technol 1992;25:83e92. [24] Mata-Alvarez J, Cecchi F, Pavan P, Llabres P. The performance of digesters treating the organic fraction of municipal solid wastes differently sorted. Biological Wastes 1990;33:181e99. [25] Sosnowski P, Wieczorek A, Ledakowicz S. Anaerobic co-digestion of sewage sludge and organic fraction of municipal solid wastes. Adv Environ Res 2003; 7:609e16. [26] Capela I, Rodrigues A, Silva F, Nadais H, Arroja L. Impact of industrial sludge and cattle manure on anaerobic digestion of the OFMSW under mesophilic conditions. Biomass Bioenergy 2008;32:245e51. [27] Zhu Z, Hsueh MK, He Q. Enhancing biomethanation of municipal waste sludge with grease trap waste as a co-substrate. Renew Energy 2011;32: 1802e7. [28] Davidsson Å, Lövstedt C, Jansen JC, Gruvberger C, Aspegren H. Co-digestion of grease trap sludge and sewage sludge. Waste Manage 2008;28:986e92. [29] USEPA. International analyses of methane emissions, www.epa.gov/methane/ intlanalyses.html; 2002.