《固体废物处理与资源化》课程教学资源（文献资料）The anaerobic digestion process of biogas production from food waste - Prospects and constraints

Bioresource Technology Reports 8(2019)100310 Contents lists available at ScienceDirect REPORT Bioresource Technology Reports ELSEVIER journal homepage:www.journals.elsevier.com/bioresource-technology-reports The anaerobic digestion process of biogas production from food waste: Prospects and constraints Sagor Kumar Pramanik",Fatihah Binti Suja,Shahrom Md Zain",Biplob Kumar Pramanik. Department of Civil and Structural Engineering Faculty of Engineering,Universiti Kebangsaan Malaysia,43600 UKM Bangi,Selangor,Malaysia School of Engineering,RMIT University,Melbourne,VIC 3000,Australia ARTICLE INFO ABSTRACT Keywords: The unrestrained release of huge quantities of food waste (FW)has become a significant concern because it Anaerobic digestion causes intensive environmental pollution.However,FW is a proper substrate that can be treated by anaerobic Biogas digestion (AD)due to its excellent biodegradability and high-water content.Studies have demonstrated that the Co-digestion AD process of tumning FW into biogas is an effective solution for FW treatment.This manuscript reviews the Food waste characteristics of FW,the biological process and biochemical reaction involved in the AD process,various op- Pre-treatment erational parameters and classification of the AD process,and the co-digestion and pre-treatment of the AD process for biogas production.Both co-digestion and pre-treatment processes could improve the FW hydrolysis rate and methane production.However,further improvement of this technology is required to assess its eco- nomic feasibility.Challenges and future perspectives of biogas production from FW are also discussed to improve the performance of AD technology. 1.Introduction 157,154,74.7,51,and 44 kg per person in Australia,America,Japan, Germany,United Kingdom,India,and China,respectively. The quantity of food losses and waste has grown enormously in the Since FW creates harmful impacts at the environmental level,ap- past few years because of the rapid growth of the world economy and propriate management and treatment of FW have become the major population.It is estimated that approximately 33.3%of food produced objective of numerous countries across the world.One of the main globally for human consumption is lost or wasted through the food environmental impacts of FW is associated with the embedded carbon supply chain(i.e.,1.6 gigatons of food per year)which has a production from the earlier life cycle phases of food before it became waste. value of $750 billion (Ma and Liu,2019;Slorach et al.,2019).The Moreover,activities related to food production such as agriculture waste of food is a non-productive use of scarce resources (land,water (including land-use change),processing,manufacturing,transportation, and fertiliser)and leads to environmental degradation (Gokarn and storage,refrigeration,distribution and retail have an embedded GHG Kuthambalayan,2017;Slorach et al.,2019).Baroutian et al.(2018) effect (Papargyropoulou et al.,2014).The same study reported that reported that the total amount of FW produced by a single person is agriculture is associated with approximately 22%of all GHG emissions 160-295 kg/year all over the world.Food is wasted during the food compared to livestock production (about 18%).The final disposal of FW supply chain.The chain includes agricultural production,processing, in landfills has also given rise to major environmental pollutions.Clercq distribution,consumption and post-harvest handling stages(Gustavsson et al.(2017)showed that considerable amounts of GHG including et al,2011;Papargyropoulou et al.,2014).It can be noted that de- methane (CH4)and carbon dioxide(CO2)are produced when FW is veloped countries tend to have major losses (70 to 80%)associated with disposed of in landfills.They showed that the emission of GHG into the the retail and consumer stages,whereas food wastage is higher at the atmosphere contributes to global warming;methane is a potent GHG immediate post-harvest stages in developing countries (Gokarn and with a greenhouse effect that is 25 times more powerful than CO2. Kuthambalayan,2017).Papargyropoulou et al.(2014)revealed that the Slorach et al.(2019)reported that global food loss and waste generate quantity of food losses and waste not only fluctuated between devel- annually 6.7%of total anthropogenic GHG emissions.Papargyropoulou oped and developing countries but also varied in low-income countries et al.(2014)demonstrated another environmental effect associated having poor producers and customers.A recent report developed by with FW,which is the disturbance of the biogenic phases of phosphorus Magnet (2018)noted that annual FW generation reached 361,278, and nitrogen,applied as fertilisers in agriculture. Corresponding author. E-mail address:biplob.pramanik@rmit.edu.au (B.K.Pramanik). https:/doi.org/10.1016/.biteb.2019.100310 Received 9 July 2019;Received in revised form 19 August 2019;Accepted 20 August 2019 Available online 21 August 2019 2589-014X/@2019 Elsevier Ltd.All rights reserved

Contents lists available at ScienceDirect Bioresource Technology Reports journal homepage: www.journals.elsevier.com/bioresource-technology-reports The anaerobic digestion process of biogas production from food waste: Prospects and constraints Sagor Kumar Pramanika , Fatihah Binti Sujaa , Shahrom Md Zaina , Biplob Kumar Pramanikb,⁎ a Department of Civil and Structural Engineering, Faculty of Engineering, Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysia b School of Engineering, RMIT University, Melbourne, VIC 3000, Australia ARTICLE INFO Keywords: Anaerobic digestion Biogas Co-digestion Food waste Pre-treatment ABSTRACT The unrestrained release of huge quantities of food waste (FW) has become a significant concern because it causes intensive environmental pollution. However, FW is a proper substrate that can be treated by anaerobic digestion (AD) due to its excellent biodegradability and high-water content. Studies have demonstrated that the AD process of turning FW into biogas is an effective solution for FW treatment. This manuscript reviews the characteristics of FW, the biological process and biochemical reaction involved in the AD process, various op￾erational parameters and classification of the AD process, and the co-digestion and pre-treatment of the AD process for biogas production. Both co-digestion and pre-treatment processes could improve the FW hydrolysis rate and methane production. However, further improvement of this technology is required to assess its eco￾nomic feasibility. Challenges and future perspectives of biogas production from FW are also discussed to improve the performance of AD technology. 1. Introduction The quantity of food losses and waste has grown enormously in the past few years because of the rapid growth of the world economy and population. It is estimated that approximately 33.3% of food produced globally for human consumption is lost or wasted through the food supply chain (i.e., 1.6 gigatons of food per year) which has a production value of $750 billion (Ma and Liu, 2019; Slorach et al., 2019). The waste of food is a non-productive use of scarce resources (land, water and fertiliser) and leads to environmental degradation (Gokarn and Kuthambalayan, 2017; Slorach et al., 2019). Baroutian et al. (2018) reported that the total amount of FW produced by a single person is 160–295 kg/year all over the world. Food is wasted during the food supply chain. The chain includes agricultural production, processing, distribution, consumption and post-harvest handling stages (Gustavsson et al., 2011; Papargyropoulou et al., 2014). It can be noted that de￾veloped countries tend to have major losses (70 to 80%) associated with the retail and consumer stages, whereas food wastage is higher at the immediate post-harvest stages in developing countries (Gokarn and Kuthambalayan, 2017). Papargyropoulou et al. (2014) revealed that the quantity of food losses and waste not only fluctuated between devel￾oped and developing countries but also varied in low-income countries having poor producers and customers. A recent report developed by Magnet (2018) noted that annual FW generation reached 361, 278, 157, 154, 74.7, 51, and 44 kg per person in Australia, America, Japan, Germany, United Kingdom, India, and China, respectively. Since FW creates harmful impacts at the environmental level, ap￾propriate management and treatment of FW have become the major objective of numerous countries across the world. One of the main environmental impacts of FW is associated with the embedded carbon from the earlier life cycle phases of food before it became waste. Moreover, activities related to food production such as agriculture (including land-use change), processing, manufacturing, transportation, storage, refrigeration, distribution and retail have an embedded GHG effect (Papargyropoulou et al., 2014). The same study reported that agriculture is associated with approximately 22% of all GHG emissions compared to livestock production (about 18%). The final disposal of FW in landfills has also given rise to major environmental pollutions. Clercq et al. (2017) showed that considerable amounts of GHG including methane (CH4) and carbon dioxide (CO2) are produced when FW is disposed of in landfills. They showed that the emission of GHG into the atmosphere contributes to global warming; methane is a potent GHG with a greenhouse effect that is 25 times more powerful than CO2. Slorach et al. (2019) reported that global food loss and waste generate annually 6.7% of total anthropogenic GHG emissions. Papargyropoulou et al. (2014) demonstrated another environmental effect associated with FW, which is the disturbance of the biogenic phases of phosphorus and nitrogen, applied as fertilisers in agriculture. https://doi.org/10.1016/j.biteb.2019.100310 Received 9 July 2019; Received in revised form 19 August 2019; Accepted 20 August 2019 ⁎ Corresponding author. E-mail address: biplob.pramanik@rmit.edu.au (B.K. Pramanik). Bioresource Technology Reports 8 (2019) 100310 Available online 21 August 2019 2589-014X/ © 2019 Elsevier Ltd. All rights reserved. T

S.K.Pramanik,et al. Bioresource Technology Reports 8 (2019)100310 There are several old technologies for treating different types of methane yields.Bong et al.(2018)reported that fruit and vegetable wastes (e.g.,animal manure,household food waste,agricultural crop waste has low lipid content but relatively high cellulosic content, residue,and organic industrial waste)around the world.Historical whereas FW and kitchen waste have high lipid content because of the evidence indicates that,the anaerobic digestion process is one of the presence of animal fat and oil.Studies have reported that fruit and oldest technologies.Granado et al.(2017)reported that the concept of vegetable waste had a lipid content of 11.8%,whereas FW and kitchen anaerobic digestion has been introduced around 1870 through the de- waste were reported to have 33.22%and 21.6%of lipid content,re- velopment of the septic tank system.In 1939,the first anaerobic di- spectively (Wang et al,2014;Yong et al.,2015).Y.Li et al.(2017a) gestion plant was constructed in the USA for treating organic fraction of reported that FW rich in lipids could produce a higher amount of me- municipal solid waste,whereas several anaerobic digestion plants were thane compared to carbohydrates and proteins.However,high lipid constructed over the last few decades in Europe (Karthikeyan et al., content can cause system failure due to the formation of long-chain 2018).The development of anaerobic digestion technology has im fatty acids.This occurs when the mass transformation of soluble or- proved very rapidly after the energy crisis in the 1970s (Deepanraj ganics into bacteria decreases due to the destruction of the cellular et al.,2014).Currently,anaerobic digestion technology is being used membrane (Leung and Wang,2016).Y.Li et al.(2017a)stated that FW not only for the treatment of organic wastes but also for the treatment rich in carbohydrate content will affect the carbon and nitrogen ratio of wastewater.For installing large-scale biogas plants,Germany and (C/N),and thus,nutrient restrictions and quick acidification could Switzerland are the pioneer countries in the global biogas industry occur due to increased organic matter (Karthikeyan et al.,2018).The United States has about 2100 current The total solids and volatile solids of each type of FW fall in the operational anaerobic digestion plants,whereas Asia has the largest ranges of 10.7%-41%and 10%-34.4%,respectively (Table 1),in- number of small-scale household anaerobic digesters that are used in dicating that water accounts for 60%-90%in fruit and vegetable waste, rural areas for lighting and cooking (Vasco-correa et al.,2018).They kitchen waste and FW.FW is considered to be a readily biodegradable also pointed out that socio-economic hurdle,existing infrastructure, organic substrate because of its large quantity of moisture content policymakers,technology availability and consistency are the main (Zhang et al.,2014).The characteristics of FW also define the relative differences in the development of anaerobic digestion plants around the quantities of organic carbon and nitrogen present in the FW.The C/N world. ratio for each type of FW was found to vary in the range of 12.7-28.84, AD is an excellent alternative for FW treatment compared to waste and displayed an acidic pH of 4.1-6.5 (Table 1).The methane pro- treatment,energy supply,and environmental protection (Leung and duction of every category of FW fall in the range of 346-551.4 mL/ Wang,2016).There are numerous benefits related to the AD process g VSadded (Table 1),which is higher compared to cow manure (233 mL/ such as decreased GHG emission,digestate for application in agronomy, g VSadded),grass silage (306 mL/g VSadded),and oat straw (203 mL/ small footprint production,and the generation of high-quality renew- g VSadded)(Huttunen and Rintala,2007). able fuel (Ariunbaatar,2014).However,the drawbacks of the AD process such as relatively high capital costs,long retention time,and 3.Biological process and biochemical reaction involved in the AD the required control of certain key parameters (e.g.,pH,temperature, process feed rate,alkalinity)prevents it from being widely implemented (Ariunbaatar,2014).Herein,the objective of this paper is to review the Anaerobic digestion is a biological process,which breaks down characteristics of FW,the biological process and biochemical reaction complex organic matter into simpler chemicals components in the ab- involved in the AD process,various operational parameters and clas. sence of oxygen.During this process,a gas that is mainly composed of sification of the AD process,and the co-digestion and pre-treatment of CH4 and CO2,also referred to as biogas,is produced as the end products the AD process for biogas production.This paper also discusses the under ideal conditions.A minor quantity of hydrogen sulphide (H2S). challenges and future perspectives of enhancing biogas production from ammonia (NH3),and other gases are also present when biogas is pro- FW and maintaining the AD process effectively. duced in the AD plant (Monnet,2003).The AD process can be divided into four stages i.e.hydrolysis,acidogenesis,acetogenesis,and metha- 2.Characteristics of FW nogenesis,as it is a multi-step biochemical process (Zhang et al.,2014). In the AD procedure,various kinds of bacteria degrade the organic FW is characterised by complex components and organic material. substance continuously in a multi-step method and via parallel reac- There are several types of FW such as fruit and vegetable waste, tions.Microorganisms play an essential role in the AD process,and the household and restaurant FW,brewery waste,and dairy waste (Xu bacterial groups are dissimilar among the phases of hydrolysis,acid. et aL,2018).Studies have found that the composition of FW varies ification,and methane production (P.Wang et al.,2018).Li et al. based on geographical changes,seasonal changes,cooking procedures, (2015)investigated the relationship between microbial community and consumption patterns (Meng et al.,2015;Xu et al.,2018).They structure and process stability and compared the microbial community reported that FW consists of various organic components such as pro- structure in both stable and deteriorative phase using 454-pyr- teins,carbohydrate polymers (starch,cellulose,hemicelluloses,and osequencing.They reported that bacteria are responsible for the de- lignin),lipids,and organic acids.Fisgativa et al.(2016)studied 102 gradation of FW to intermediate metabolites which can be later used by different FW samples and reported that the characteristic of FW dis- methanogens.They concluded that acid-producing bacteria such as played high coefficient of variance (CV).They indicated that the var- Acholeplasma and Actinomyces increased dramatically at deteriorative iations of 24%of the studied characteristics were described by the phase compared with stable stage,which may be the failure indicator in geographical change,seasonal change and the type of collection source. anaerobic digester treating FW.In hydrolysis,insoluble complex poly- They observed that FW has an average pH of 5.1(CV 13.9%),carbon mers comprising carbohydrates,proteins,lipids,and other organics are and nitrogen ratio of 18.5%(CV 31.8%),36%of carbohydrates (CV converted into smaller soluble molecules.It can be noted that hydro- 57.2%),26%of protein (CV 62.2%),15%of fats (CV 52.0%),and lysis is a comparatively slow stage and therefore can limit the rate of the biomethane potential of 460.0 NL CH4/kg VS(CV 19%) entire AD process,especially when FW is used as the feedstock(Kothari As mentioned in Table 1,the percentage range of degradable car- et al.,2014;Leung and Wang,2016;Ostrem,2004;A.Zhang et al. bohydrates,proteins,and lipids is (5.7%-53%),(2.3%-28.4%),and 2015;Zhang et al.,2014).In acidogenesis,monomers and dissolved (1.3%-30.3%),respectively.Meng et al.(2015)and Xu et al.(2018) compounds such as sugars,amino acids,and fatty acids resulting from indicated that carbohydrates and proteins have a higher hydrolysis rate hydrolysis are converted into simple molecules with a small molecular due to its rapid degradability compared to lipid.Thus,quickly de- weight such as volatile fatty acids(ie.,propionic,butyric,acetic acid), gradable carbohydrates and lipid-rich food wastes can produce high alcohols,and different kinds of gases (CO2,H2,and NH3)(A.Zhang

There are several old technologies for treating different types of wastes (e.g., animal manure, household food waste, agricultural crop residue, and organic industrial waste) around the world. Historical evidence indicates that, the anaerobic digestion process is one of the oldest technologies. Granado et al. (2017) reported that the concept of anaerobic digestion has been introduced around 1870 through the de￾velopment of the septic tank system. In 1939, the first anaerobic di￾gestion plant was constructed in the USA for treating organic fraction of municipal solid waste, whereas several anaerobic digestion plants were constructed over the last few decades in Europe (Karthikeyan et al., 2018). The development of anaerobic digestion technology has im￾proved very rapidly after the energy crisis in the 1970s (Deepanraj et al., 2014). Currently, anaerobic digestion technology is being used not only for the treatment of organic wastes but also for the treatment of wastewater. For installing large-scale biogas plants, Germany and Switzerland are the pioneer countries in the global biogas industry (Karthikeyan et al., 2018). The United States has about 2100 current operational anaerobic digestion plants, whereas Asia has the largest number of small-scale household anaerobic digesters that are used in rural areas for lighting and cooking (Vasco-correa et al., 2018). They also pointed out that socio-economic hurdle, existing infrastructure, policymakers, technology availability and consistency are the main differences in the development of anaerobic digestion plants around the world. AD is an excellent alternative for FW treatment compared to waste treatment, energy supply, and environmental protection (Leung and Wang, 2016). There are numerous benefits related to the AD process such as decreased GHG emission, digestate for application in agronomy, small footprint production, and the generation of high-quality renew￾able fuel (Ariunbaatar, 2014). However, the drawbacks of the AD process such as relatively high capital costs, long retention time, and the required control of certain key parameters (e.g., pH, temperature, feed rate, alkalinity) prevents it from being widely implemented (Ariunbaatar, 2014). Herein, the objective of this paper is to review the characteristics of FW, the biological process and biochemical reaction involved in the AD process, various operational parameters and clas￾sification of the AD process, and the co-digestion and pre-treatment of the AD process for biogas production. This paper also discusses the challenges and future perspectives of enhancing biogas production from FW and maintaining the AD process effectively. 2. Characteristics of FW FW is characterised by complex components and organic material. There are several types of FW such as fruit and vegetable waste, household and restaurant FW, brewery waste, and dairy waste (Xu et al., 2018). Studies have found that the composition of FW varies based on geographical changes, seasonal changes, cooking procedures, and consumption patterns (Meng et al., 2015; Xu et al., 2018). They reported that FW consists of various organic components such as pro￾teins, carbohydrate polymers (starch, cellulose, hemicelluloses, and lignin), lipids, and organic acids. Fisgativa et al. (2016) studied 102 different FW samples and reported that the characteristic of FW dis￾played high coefficient of variance (CV). They indicated that the var￾iations of 24% of the studied characteristics were described by the geographical change, seasonal change and the type of collection source. They observed that FW has an average pH of 5.1 (CV 13.9%), carbon and nitrogen ratio of 18.5% (CV 31.8%), 36% of carbohydrates (CV 57.2%), 26% of protein (CV 62.2%), 15% of fats (CV 52.0%), and biomethane potential of 460.0 NL CH4/kg VS (CV 19%). As mentioned in Table 1, the percentage range of degradable car￾bohydrates, proteins, and lipids is (5.7%–53%), (2.3%–28.4%), and (1.3%–30.3%), respectively. Meng et al. (2015) and Xu et al. (2018) indicated that carbohydrates and proteins have a higher hydrolysis rate due to its rapid degradability compared to lipid. Thus, quickly de￾gradable carbohydrates and lipid-rich food wastes can produce high methane yields. Bong et al. (2018) reported that fruit and vegetable waste has low lipid content but relatively high cellulosic content, whereas FW and kitchen waste have high lipid content because of the presence of animal fat and oil. Studies have reported that fruit and vegetable waste had a lipid content of 11.8%, whereas FW and kitchen waste were reported to have 33.22% and 21.6% of lipid content, re￾spectively (Wang et al., 2014; Yong et al., 2015). Y. Li et al. (2017a) reported that FW rich in lipids could produce a higher amount of me￾thane compared to carbohydrates and proteins. However, high lipid content can cause system failure due to the formation of long-chain fatty acids. This occurs when the mass transformation of soluble or￾ganics into bacteria decreases due to the destruction of the cellular membrane (Leung and Wang, 2016). Y. Li et al. (2017a) stated that FW rich in carbohydrate content will affect the carbon and nitrogen ratio (C/N), and thus, nutrient restrictions and quick acidification could occur due to increased organic matter. The total solids and volatile solids of each type of FW fall in the ranges of 10.7%–41% and 10%–34.4%, respectively (Table 1), in￾dicating that water accounts for 60%–90% in fruit and vegetable waste, kitchen waste and FW. FW is considered to be a readily biodegradable organic substrate because of its large quantity of moisture content (Zhang et al., 2014). The characteristics of FW also define the relative quantities of organic carbon and nitrogen present in the FW. The C/N ratio for each type of FW was found to vary in the range of 12.7–28.84, and displayed an acidic pH of 4.1–6.5 (Table 1). The methane pro￾duction of every category of FW fall in the range of 346–551.4 mL/ g VSadded (Table 1), which is higher compared to cow manure (233 mL/ g VSadded), grass silage (306 mL/g VSadded), and oat straw (203 mL/ g VSadded) (Huttunen and Rintala, 2007). 3. Biological process and biochemical reaction involved in the AD process Anaerobic digestion is a biological process, which breaks down complex organic matter into simpler chemicals components in the ab￾sence of oxygen. During this process, a gas that is mainly composed of CH4 and CO2, also referred to as biogas, is produced as the end products under ideal conditions. A minor quantity of hydrogen sulphide (H2S), ammonia (NH3), and other gases are also present when biogas is pro￾duced in the AD plant (Monnet, 2003). The AD process can be divided into four stages i.e. hydrolysis, acidogenesis, acetogenesis, and metha￾nogenesis, as it is a multi-step biochemical process (Zhang et al., 2014). In the AD procedure, various kinds of bacteria degrade the organic substance continuously in a multi-step method and via parallel reac￾tions. Microorganisms play an essential role in the AD process, and the bacterial groups are dissimilar among the phases of hydrolysis, acid￾ification, and methane production (P. Wang et al., 2018). Li et al. (2015) investigated the relationship between microbial community structure and process stability and compared the microbial community structure in both stable and deteriorative phase using 454-pyr￾osequencing. They reported that bacteria are responsible for the de￾gradation of FW to intermediate metabolites which can be later used by methanogens. They concluded that acid-producing bacteria such as Acholeplasma and Actinomyces increased dramatically at deteriorative phase compared with stable stage, which may be the failure indicator in anaerobic digester treating FW. In hydrolysis, insoluble complex poly￾mers comprising carbohydrates, proteins, lipids, and other organics are converted into smaller soluble molecules. It can be noted that hydro￾lysis is a comparatively slow stage and therefore can limit the rate of the entire AD process, especially when FW is used as the feedstock (Kothari et al., 2014; Leung and Wang, 2016; Ostrem, 2004; A. Zhang et al., 2015; Zhang et al., 2014). In acidogenesis, monomers and dissolved compounds such as sugars, amino acids, and fatty acids resulting from hydrolysis are converted into simple molecules with a small molecular weight such as volatile fatty acids (i.e., propionic, butyric, acetic acid), alcohols, and different kinds of gases (CO2, H2, and NH3) (A. Zhang S.K. Pramanik, et al. Bioresource Technology Reports 8 (2019) 100310 2

S.K.Pramanik,et al. Bioresource Technology Reports 8(2019)100310 Table 1 Characteristics and methane potential of some food waste reported in literatures. Source TS(%)VS(%)VS/TS (% C/N ratio pH Carbohydrates(%)Proteins (%Lipids(%) Methane yield (mL/g VS) Reference KW 24.9 23.1 92.8 18.24 49 17.3 23 501 J.Jiang et al.(2018) FVW 13.8 12.88 93.4 4.5 7.74 3.28 2.87 516 Edwiges et al (2018) FW 20 19.26 96.3 15.5 47.6 24.1 28.3 548.1 Li et al.(2018) 19.2 18.45 96.1 12.7 41.3 28.4 30.3 541.1 20.9 20.02 95.8 14 38.4 26.3 25.3 545.3 FW 10.86 10.22 94 15.18 4.16 5.71 2.29 1.31 460 Xiao et al (2019) 25.94 24.59 94.7 17.5 48 15.1 10.6 346.2 Shi et al.(2018) w 10.69 10.06 4.18 5.69 2.29 1.30 477-459 Xiao et al.(2018) 24.3 22.5 16 23.11 5.02 3.38 386.7-551.4 Liu et al.(2017) FW 19.1 18.53 97 17.7 10.8 4 37 536.19 Y.Li et al.(2017a) 17.2 16.72 97.2 13.4 1 2.9 441.23 20.5 19.78 96.5 17.8 3.6 5.3 531.30 19.7 19.05 96.7 16.8 10.1 53737 19.6 18.91 96.5 15.4 9.3 533.01 20.0 19.26 96.3 15.5 544.95 AFW 41 34.44 4 经 25 Naroznova et aL (2016) VFW 24 22.32 93 53 5 14 425 19.1 17.80 93.2 14.41 4.5 11.8 2.5 3.5 372.1 Li et al.(2016a) W 23.2 21.7 93.5 4.4 13.7 2.9 6.5 425.2 W.Zhang et al.(2015) 20.05 19.21 95.81 28.4 33.22 14.03 25.25 381 Yong et al.(2015) FW 29.4 28.01 95.3 14.2 4.1 18.1 19 529 Browne and Murphy (2013) FW 18.1 17.1 94 13.2 6.5 11.2 3.3 2.3 479.5 Zhang et al.(2011) Note:FVW:fruit and vegetable waste,KW:kitchen waste,FW:food waste,AFW:animal food waste,VFW:vegetable food waste. et al.,2015).In acetogenesis,acetogenic bacteria use volatile fatty acids utilisation rates,and bacterial development (Khalid et al.,2011).The for their growth and the growth of these bacteria is slow with a dou same study reported that cell energy fatigue and intracellular substance bling time of 1.5 to 4 days (Kothari et al.,2014).The concentration of leakage could occur due to this lower temperature.They also indicated products created in this phase differs according to the type of bacteria that AD operating under mesophilic conditions offers higher stability as well as culture circumstances such as temperature and pH(Ostrem, and requires lower energy cost compared to the thermophilic condition. 2004).Methanogenesis is the final step in the anaerobic digester,where AD operating under thermophilic condition provides several benefits the methanogens use acetic acid,hydrogen,and carbon dioxide to such as the higher growth of methanogenic bacteria at a higher tem- produce methane gas.Most methanogen bacteria require an optimum perature,reduced retention time,destruction of pathogens,enhanced pH range between 6.5 and 7.5 (Leung and Wang,2016).The four steps digestibility and better degradability of solid substrates,besides dif- of anaerobic biodegradation process such as hydrolysis,acidogenesis, ferentiating liquid and solid portions (Dobre et al,2014;A.Zhang acetogenesis,and methanogenesis,and the bacteria involved in each et al.,2015).However,the drawbacks of the thermophilic condition stage of the AD process,are schematically displayed in Fig.1. must be considered.There are several demerits of the thermophilic condition such as a greater amount of disproportion and higher energy 4.Parameters affecting the AD process of biogas production requirement because of the associated high temperature (A.Zhang etal.,2015) Different categories of bacteria are engaged in the AD process.The bacteria need an environment that has reached equilibrium to produce 4.2.pH and VFA biogas from FW.Leung and Wang(2016)reported that a stable process could be achieved if the bacteria stayed in an ideal condition.Fluc pH is the most significant parameter that affects the performance tuations in environmental conditions could interrupt the micro- and stability of an anaerobic digester.Microorganisms are sensitive to organism's equilibrium,which could perhaps impede or even close pH.This is because every group of bacteria needs a different pH range down the AD process(Ostrem,2004).Therefore,operational conditions for their growth (Appels et al.,2008).The ideal pH range for hydrolysis. (such as temperature,pH and VFA,the ratio of carbon and nitrogen, acetogenesis,and methanogenesis is almost 6.0,6.0-7.0,and 6.5-7.5 retention time,and organic loading rate)in the AD process need to be respectively (Leung and Wang,2016).Gerardi (2003)reported that the continuously observed and maintained within optimum ranges.The pH required for acid-forming bacteria and methane-forming bacteria effect and optimised range of these parameters on biogas production is>5.0 and 6.2,respectively,for acceptable enzymatic activity.Me- are explained in the following section. thanogenic bacteria display better performance in a pH range of 6.8-7.2 (Yadvika et al.,2004).CH4 production was found to be 75%more ef- 4.1.Temperature ficient with a pH of>5.0 (Yadvika et al.,2004).Krishna and Kalamdhad (2014)indicated that other main factors contribute to the Methane production is highly influenced by temperature,which fluctuation of pH such as alkalinity,volatile fatty acid (VFA),the affects bacterial performance within an anaerobic digester.Both me- quantity of CO2 production,and the concentration of bicarbonate thanogenic and volatile acid-forming microorganisms are affected by (HCO)during the AD process.They reported that the relationship temperature.Changes in temperature extensively influence the perfor between VFA and HCO3 concentrations should be controlled,as it helps mance of methanogenic microorganisms compared to the operating in adjusting the optimum pH during the AD process.For a stable and temperature (Gerardi,2003).The AD process can take place at various well-buffered digestion process,it is important to maintain at least temperatures,which are normally classified into three types,i.e.psy. 1.4:1 (molar ratio)of HCO3/VFA or a buffering capacity of chrophilic,mesophilic,and thermophilic temperatures.The mesophilic 70 meg CaCO3/L (Appels et al.,2008).Volatile fatty acids are short- and thermophilic temperature ranges are between 20 and 40C(usually chain fatty acids (acetic acid,propionic acid,butyric acid,and valeric 35'C)and 50 and 65'C(typically 45'C),respectively (Kothari et al., acid),which are the primary intermediate products produced from the 2014).A lower temperature reduces methane production,FW AD of FW (Zhang et al.,2014).Xu et al.(2014)and Shi et al.(2018)

et al., 2015). In acetogenesis, acetogenic bacteria use volatile fatty acids for their growth and the growth of these bacteria is slow with a dou￾bling time of 1.5 to 4 days (Kothari et al., 2014). The concentration of products created in this phase differs according to the type of bacteria as well as culture circumstances such as temperature and pH (Ostrem, 2004). Methanogenesis is the final step in the anaerobic digester, where the methanogens use acetic acid, hydrogen, and carbon dioxide to produce methane gas. Most methanogen bacteria require an optimum pH range between 6.5 and 7.5 (Leung and Wang, 2016). The four steps of anaerobic biodegradation process such as hydrolysis, acidogenesis, acetogenesis, and methanogenesis, and the bacteria involved in each stage of the AD process, are schematically displayed in Fig. 1. 4. Parameters affecting the AD process of biogas production Different categories of bacteria are engaged in the AD process. The bacteria need an environment that has reached equilibrium to produce biogas from FW. Leung and Wang (2016) reported that a stable process could be achieved if the bacteria stayed in an ideal condition. Fluc￾tuations in environmental conditions could interrupt the micro￾organism's equilibrium, which could perhaps impede or even close down the AD process (Ostrem, 2004). Therefore, operational conditions (such as temperature, pH and VFA, the ratio of carbon and nitrogen, retention time, and organic loading rate) in the AD process need to be continuously observed and maintained within optimum ranges. The effect and optimised range of these parameters on biogas production are explained in the following section. 4.1. Temperature Methane production is highly influenced by temperature, which affects bacterial performance within an anaerobic digester. Both me￾thanogenic and volatile acid-forming microorganisms are affected by temperature. Changes in temperature extensively influence the perfor￾mance of methanogenic microorganisms compared to the operating temperature (Gerardi, 2003). The AD process can take place at various temperatures, which are normally classified into three types, i.e. psy￾chrophilic, mesophilic, and thermophilic temperatures. The mesophilic and thermophilic temperature ranges are between 20 and 40 °C (usually 35 °C) and 50 and 65 °C (typically 45 °C), respectively (Kothari et al., 2014). A lower temperature reduces methane production, FW utilisation rates, and bacterial development (Khalid et al., 2011). The same study reported that cell energy fatigue and intracellular substance leakage could occur due to this lower temperature. They also indicated that AD operating under mesophilic conditions offers higher stability and requires lower energy cost compared to the thermophilic condition. AD operating under thermophilic condition provides several benefits such as the higher growth of methanogenic bacteria at a higher tem￾perature, reduced retention time, destruction of pathogens, enhanced digestibility and better degradability of solid substrates, besides dif￾ferentiating liquid and solid portions (Dobre et al., 2014; A. Zhang et al., 2015). However, the drawbacks of the thermophilic condition must be considered. There are several demerits of the thermophilic condition such as a greater amount of disproportion and higher energy requirement because of the associated high temperature (A. Zhang et al., 2015). 4.2. pH and VFA pH is the most significant parameter that affects the performance and stability of an anaerobic digester. Microorganisms are sensitive to pH. This is because every group of bacteria needs a different pH range for their growth (Appels et al., 2008). The ideal pH range for hydrolysis, acetogenesis, and methanogenesis is almost 6.0, 6.0–7.0, and 6.5–7.5, respectively (Leung and Wang, 2016). Gerardi (2003) reported that the pH required for acid-forming bacteria and methane-forming bacteria is > 5.0 and 6.2, respectively, for acceptable enzymatic activity. Me￾thanogenic bacteria display better performance in a pH range of 6.8–7.2 (Yadvika et al., 2004). CH4 production was found to be 75% more ef- ficient with a pH of > 5.0 (Yadvika et al., 2004). Krishna and Kalamdhad (2014) indicated that other main factors contribute to the fluctuation of pH such as alkalinity, volatile fatty acid (VFA), the quantity of CO2 production, and the concentration of bicarbonate (HCO3) during the AD process. They reported that the relationship between VFA and HCO3 concentrations should be controlled, as it helps in adjusting the optimum pH during the AD process. For a stable and well-buffered digestion process, it is important to maintain at least 1.4:1 (molar ratio) of HCO3/VFA or a buffering capacity of 70 meq CaCO3/L (Appels et al., 2008). Volatile fatty acids are short￾chain fatty acids (acetic acid, propionic acid, butyric acid, and valeric acid), which are the primary intermediate products produced from the AD of FW (Zhang et al., 2014). Xu et al. (2014) and Shi et al. (2018) Table 1 Characteristics and methane potential of some food waste reported in literatures. Source TS (%) VS (%) VS/TS (%) C/N ratio pH Carbohydrates (%) Proteins (%) Lipids (%) Methane yield (mL/g VS) Reference KW 24.9 23.1 92.8 18.24 – 49 17.3 23 501 J. Jiang et al. (2018) FVW 13.8 12.88 93.4 4.5 7.74 3.28 2.87 516 Edwiges et al. (2018) FW 20 19.26 96.3 15.5 – 47.6 24.1 28.3 548.1 Li et al. (2018) 19.2 18.45 96.1 12.7 – 41.3 28.4 30.3 541.1 20.9 20.02 95.8 14 – 38.4 26.3 25.3 545.3 FW 10.86 10.22 94 15.18 4.16 5.71 2.29 1.31 460 Xiao et al. (2019) FW 25.94 24.59 94.7 17.5 – 48 15.1 10.6 346.2 Shi et al. (2018) FW 10.69 10.06 94 – 4.18 5.69 2.29 1.30 477–459 Xiao et al. (2018) FW 24.3 22.5 92.6 23.11 5.02 – 3.38 – 386.7–551.4 Liu et al. (2017) FW 19.1 18.53 97 17.7 – 10.8 4 3.7 536.19 Y. Li et al. (2017a) 17.2 16.72 97.2 13.4 – 9.4 4.5 2.9 441.23 20.5 19.78 96.5 17.8 – 11 3.6 5.3 531.30 19.7 19.05 96.7 16.8 – 10.1 4.3 4.7 537.37 19.6 18.91 96.5 15.4 – 9.3 4.6 5 533.01 20.0 19.26 96.3 15.5 9.2 4.6 5.5 544.95 AFW 41 34.44 84 – – 52 12 25 – Naroznova et al. (2016) VFW 24 22.32 93 – – 53 5 14 425 KW 19.1 17.80 93.2 14.41 4.5 11.8 2.5 3.5 372.1 Li et al. (2016a) FW 23.2 21.7 93.5 4.4 13.7 2.9 6.5 425.2 W. Zhang et al. (2015) FW 20.05 19.21 95.81 28.4 – 33.22 14.03 25.25 381 Yong et al. (2015) FW 29.4 28.01 95.3 14.2 4.1 – 18.1 19 529 Browne and Murphy (2013) FW 18.1 17.1 94 13.2 6.5 11.2 3.3 2.3 479.5 Zhang et al. (2011) Note: FVW: fruit and vegetable waste, KW: kitchen waste, FW: food waste, AFW: animal food waste, VFW: vegetable food waste. S.K. Pramanik, et al. Bioresource Technology Reports 8 (2019) 100310 3

S.K.Pramanik,et al. Bioresource Technology Reports 8(2019)100310 Food Waste Complex polymers PH-S.5-60 Carbohydrates Proteins Lipids Clostridium,Acctivibrio, Clostridium,Proteus Clostridium. Fermentative Staphylococcus,Bacteroides Peptococcus,Vibrio. Micrococcus, Bacteroides,Bacillus Staphylococcus bacteria PH-6.0-7.0 Acidogenesis Monomers Sugars Amino acids LCFA Clostridium. Lactobacillus,Escherichia.Staphylococcus. Acidogenic Eubacterium limosum. Micrococcus,Bacillus,Sarcina,Veillonella, Streptococcus Pscudomonas.Desulfovibrio,Desulfuromonas. bacteria Desulfobacter,Selenomonas,Streptococcus VFAs and Clostridium,Syntrophobacter PH-6.0-7.0 Acetogenesis alcohols Syntrophomonas Acetogenic Clostridium,Syntrophobacter,Syntrophomonas bacteria Acetate oxidizing bacteria Acetate H2,C02 Homoacetogenic bacteria Methanobacterium,Methanocalculus Methanosarcina,Methanospirillum, Methanogenic PH-6.57.5 Methanogenesis Methanobrevibacter,Methanoplanus Methanosacta Methanoregula,Methanococcus, archaea Methanoculleus Acetotrophic CH4,CO2 Hydrogenotrophic methanogens methanogens Fig.1.Four biological steps and the respective bacterial groups engaged in every phase of the AD process.Information collected from Deepanraj et al.(2014), Gonzalez-Fernandez et al.(2015),Kothari et al.(2014),Leung and Wang (2016),and P.Wang et al.(2018). reported that methane production was inhibited completely when the for an effective AD process (Kothari et al.,2014).Zhang et al.(2014) concentrations of VFA fell in the range of 5800 to 6900 mg/L. reported that the C/N ratio greatly influences the stability of the AD process.This is because the optimal C/N ratio not only helps to main- 4.3.Carbon and nitrogen ratio tain a suitable environment,but it also helps to control proper nutrient balance through the development of microorganisms.The microbial The C/N ratio represents the relationship between the quantity of population could increase gradually if the quantity of nitrogen is low in carbon and nitrogen present in FW.An optimum C/N ratio is required the FW,and thus,more time will be required to decompose the existing

reported that methane production was inhibited completely when the concentrations of VFA fell in the range of 5800 to 6900 mg/L. 4.3. Carbon and nitrogen ratio The C/N ratio represents the relationship between the quantity of carbon and nitrogen present in FW. An optimum C/N ratio is required for an effective AD process (Kothari et al., 2014). Zhang et al. (2014) reported that the C/N ratio greatly influences the stability of the AD process. This is because the optimal C/N ratio not only helps to main￾tain a suitable environment, but it also helps to control proper nutrient balance through the development of microorganisms. The microbial population could increase gradually if the quantity of nitrogen is low in the FW, and thus, more time will be required to decompose the existing Food Waste VFAs and alcohols Carbohydrates Lipids Proteins Sugars Amino acids LCFA Acetate H2, CO2 CH4, CO2 Clostridium, Proteus, Peptococcus, Vibrio, Bacteroides, Bacillus Clostridium, Acetivibrio, Staphylococcus, Bacteroides Clostridium, Micrococcus, Staphylococcus Lactobacillus, Escherichia, Staphylococcus, Micrococcus, Bacillus, Sarcina, Veillonella, Pseudomonas, Desulfovibrio, Desulfuromonas, Desulfobacter, Selenomonas, Streptococcus Clostridium, Eubacterium limosum, Streptococcus Clostridium, Syntrophobacter, Syntrophomonas Clostridium, Syntrophobacter, Syntrophomonas Methanosarcina, Methanospirillum, Methanosaeta Methanobacterium, Methanocalculus, Methanobrevibacter, Methanoplanus, Methanoregula, Methanococcus, Methanoculleus Fermentative bacteria Acidogenic bacteria Acetogenic bacteria Methanogenic archaea pH= 5.5-6.0 Hydrolysis Acidogenesis pH= 6.0-7.0 Methanogenesis pH= 6.5-7.5 pH= 6.0-7.0 Acetogenesis Complex polymers Monomers Acetate oxidizing bacteria Homoacetogenic bacteria Acetotrophic methanogens Hydrogenotrophic methanogens Fig. 1. Four biological steps and the respective bacterial groups engaged in every phase of the AD process. Information collected from Deepanraj et al. (2014), Gonzalez-Fernandez et al. (2015), Kothari et al. (2014), Leung and Wang (2016), and P. Wang et al. (2018). S.K. Pramanik, et al. Bioresource Technology Reports 8 (2019) 100310 4

S.K.Pramanik,et al. Bioresource Technology Reports 8(2019)100310 18100021e1ac s1020ie100x (SIOZ)nM 【(0100ppn (ZIOZ)T 1 nrl 令 爱 总 202m 25 1215 胡 (SA 草 后 年房时 母 多 月 品 6活 開 1V1911 pnjs paas oydosaw a8pnIs paas a1qoaeuy a8pnjs a8emas oydosaW HISO snonuguoo-jwas TZ dosau :=8a.a 'snonupuoo-yas 10 pue I (snonunuoo 8615-0ML a8es-o3us -oM aseqd-ammeaduL omudosaw

Table 2 Comparison of different anaerobic digestion process for food waste. Process Reactor volume and type Inoculum Operating condition Methane yield (mL/ g VS) VS removal (%) Reference Single-stage 2 L semi-continuous CSTR, mesophilica Anaerobic sludge OLR = 5 g VS/L day, HRT = 20 days, pH = 7.7 494 74.7 Jo et al. (2018) Two-stage 0.5 L and 2 L semi-continuous CSTR, mesophilic OLR = 4 g VS/L day, HRT = 25 days, pH = 7.5 511 78.9 Single-stage 230 L CSTR, thermophilicb Anaerobic sludge OLR = 3.5 kg VS/m3 day, HRT = 20 days 450 93.6 Micolucci et al. (2018) Two-stage 200 and 760 L CSTR, thermophilic 550 96.2 Single-stage 4 L semi-continuous CSTR, thermophilic Mesophilic anaerobic sludge HRT = 30 days, pH = 8.31, TVFA = 0.87 g HAc/L 477 83.22 Xiao et al. (2018) Two-stage 2 L and 8 L semi-continuous CSTR, thermophilic HRT = 30 days, pH = 3.83 and 8.17, TVFA = 1.64 and 1.08 g HAc/L 459 82.02 Single-stage 1 and 20 L semi-continuous, mesophilic Anaerobic seed sludge OLR = 1.6–10 g VS/L, pH = 5.1–7.8 199 30.1 Zhang et al. (2017) Two-stage 249 44.2 Three-stage 307 83.5 Single-stage 5 m3 reactor, mesophilic Anaerobic sludge OLR = 3.79 kg VS/m3/day, pH = 7.32 380 96 Grimberg et al. (2015) Two-stage OLR = 0.78 kg VS/m3/day, pH = 5.2, 8.4 446 93 Single-stage 6 L CSTR, mesophilic Mesophilic seed sludge OLR = 2.4 g VS/L/day, pH = 7.77, HRT = 30 days 440 74.1 Wu et al. (2015) Temperature-phase two￾stage 1.5 L thermophilic CSTR, 6 L mesophilic CSTR OLR = 14.2 and 2.6 g VS/L/day, pH = 5.36 and 7.59, HRT = 3 and 12 days 440 80.1 Mesophilic 3 L continuous CSTR Anaerobic seed sludge OLR = 7.75 g VS/L/day, HRT = 10 days, pH = 7.64 350 – Q. Li et al. (2017) Thermophilic OLR = 5.19 g VS/L/day, HRT = 15 days, pH = 7.86. 407 Mesophilic 500 mL laboratory-scale bottle (semi￾continuous) Mesophilic sewage sludge OLR = 1.5 g VS/L/day, HRT = 20 days 371 94.7 Liu et al. (2017) Thermophilic Thermophilic sewage sludge OLR = 2.5 g VS/L/day, HRT = 20 days 541 93 Batch 75 L upflow anaerobic reactor, mesophilic Seed sludge OLR = 6.1 kg COD/m3/day, HRT = 30 days, pH = 7.5 266a 80.9b Park et al. (2018) Continuous OLR = 7.9 kg COD/m3/day, HRT = 13 days, pH = 7.5 326a 71.2b Batch 1 L glass digesters, mesophilic Anaerobic activated sludge OLR = 8 g VS/L, pH = 7.5 388 – Zhang et al. (2013) Continuous OLR = 10 g VS/L, pH = 7.1–7.7 317 Wet 6.0 L continuous reactor, mesophilic Mesophilic seed sludge OLR = 2.35 kg VS/m3/day, SRT = 20 days, pH = 7.39 370 80.1 Yi et al. (2014) Dry OLR = 9.41 kg VS/m3/day, SRT = 20 days, pH = 7.82 480 85.6 a L/kg COD. b COD removal. S.K. Pramanik, et al. Bioresource Technology Reports 8 (2019) 100310 5

S.K.Pramanik,et al. Bioresource Technology Reports 8 (2019)100310 FW,resulting in lower CH4 yield.In contrast,ammonia inhibition could On the other hand,higher OLR causes shorter HRT,which may result in occur,preventing microbial growth,especially if the concentration of microorganism washout,and this could lead to lower biogas production nitrogen is more than the microbial necessity (Kothari et al.,2014; (Leung and Wang,2016).Hu et al.(2018)reported that the thermo- Leung and Wang,2016).It is found that microorganisms use carbon philic reactor can be adapt to a wide range of OLR (3.0-14.4 kg-COD/ 25-30(Yadvika et al.,2004),25-35(Krishna and Kalamdhad,2014)or m/day)compared to the mesophilic reactor (OLR of 3.0-7.3 kg-COD/ 30-35(Leung and Wang,2016)times quicker than nitrogen during the m/day)in anaerobic FW treatment.Liu et al.(2017)pointed out that AD process.Therefore,the optimum ratio of C/N in the substrate should the optimal OLR on AD of FW under thermophilic and mesophilic be almost 20-30:1 (Yadvika et al.,2004)25-30:1(Krishna and condition was 2.5 and 1.5g-VS/L/day,respectively. Kalamdhad,2014)or 30-35:1 (Leung and Wang,2016).The co-diges- tion of organic substrates with animal manure,lignocellulosic biomass, 5.Classification of the AD process of FW and sewage sludge can be used to enhance the C/N ratios (Khalid et al., 2011). The classification of the anaerobic reactor when treating FW de- pends on the temperature,feeding type,and moisture content of the 4.4.Retention time FW.Single-stage and two/multi-stage processes have also influenced the performance of the AD process (Table 2)(Deepanraj et al.,2014). Retention time is one of the major parameters,which needs to be Mesophilic or thermophilic,batch or continuous process,or wet or dry repeatedly observed in the anaerobic digesters.Retention time is the digestion are the various categories of the AD process,which are further time needed for the complete degradation of organic matter or the described in the following sections. average time organic matter remains in a digester (Deepanraj et al., 2014;Mao et al.,2015).There are two important types of retention 5.1.Wet and dry digestion times involved in the AD system:solid retention time (SRT)and hy- draulic retention time (HRT).SRT is the average time that the bacteria The AD process can be categorised as wet or dry digestion based on spend in the digester,whereas HRT is the average time that the liquid the total solid concentration in FW.The anaerobic digestion of food sludge spends in the digester(Deepanrajet al,2014).According to Mao waste is termed as a dry process if the total solid concentration of the et al.(2015),bacterial development rate related to retention time de- food waste stays between 20 and 40%,whereas the anaerobic digestion pends on process temperature,substrate composition,and organic of food waste is considered as a wet process when the total solid con- loading rate;the shorter the HRT,the higher the value of organic centration of the food waste is 15%(Kothari et al.,2014;Deepanraj loading rate.A retention time of 10-40 days is necessary to treat or et al.,2014).Kothari et al.(2014)noted that most of the AD plants ganic waste in mesophilic temperature,while a lesser retention time constructed during the 1980s depended on the wet system,whereas could be used in thermophilic temperature (Kothari et al.,2014). new plants constructed in the last decade are mostly based on the dry Yadvika et al.(2004)described that high capital cost and large reactor process.They also noted that the HRT,OLR,and volatile solid removal volume are the main requirements for a longer HRT.However,shorter rate of the dry process were 14-60 days,12-15kg VS/m/day,and HRT offers insufficient time for the optimal degradation of the sub- 40%-70%,respectively,whereas the HRT,OLR,and volatile solid de- strate.Yadvika et al.(2004)noted that HRT varies with climate change struction rate of the wet process were 25-60 days,<5kg VS/m/day, For example,for tropical countries and in cold weather,HRT fluctuates and 40%-75%,respectively.It is important to note that a higher cost for from 30 to 50 days and 100 days,respectively.SRT maintains the digestate dewatering post-treatment and higher reactor volume are bacterial population in the reactor,which could result in waste stabi- required for the wet process compared to the dry process.Kothari et al. lisation (Dobre et al.,2014).A portion of the microbial community is (2014)noted a higher biogas production rate in the dry process com- removed every time sludge is withdrawn,resulting in a stable condition pared to the wet process.Yi et al.(2014)examined dry (total and the prevention of process failure (Appels et al.,2008).A long SRT solid =20%)and wet (total solid =5%)processes of AD during food helps permit biological acclimation to toxic compounds,and this can be waste treatment,and found that the CH4 production(0.48 L/g VS)and achieved by increasing both the volume of the digester and the con- volatile solid reduction (85.6%)were higher in the dry process com- centration of the bacteria (solids)(Gerardi,2003).Chen et al.(2018) pared to the wet process.They also found that the dry process allowed a observed that the CH4 yield reduced with higher SRT and the highest higher VFA concentration and OLR compared to the wet process,re- CH yield was achieved at an SRT of 6 days,compared to 7.5 days and sulting in a decreased possibility of inhibition of the AD technique.The 10 days.Fernandez-Rodriguez et al.(2014)reported that the maximum comparison of dry versus wet digestion process has shown that the dry CHa yield was achieved at an SRT of 5-8 days compared to SRTs of process had a more advantageous energy balance and economic per- 4 days and 3 days.They also indicated that SRTs lower than 4 days were formance compared to the wet process.However,complete mixing of inappropriate for a single-stage dry AD of organic fraction of municipal the waste is not possible in dry process,and,thus,the ideal contact of solid waste,which would render the process unbalanced. microorganisms and substrate cannot be guaranteed.Conversely,wet process offered several important benefits including greater flexibility 4.5.Organic loading rate over the type of feedstock accepted,dilution of inhibitory substances by process water and necessity of less sophisticated mechanical equip- Organic loading rate (OLR)is a significant operational parameter ment.Overall,dry digestion process was more cost-effective,and it had that affects the CH4 yield.OLR is defined as the amount of dry organic greater biogas productivity than wet process. solids,which can be fed into the digester per day per unit volume of digester capacity (Kothari et al.,2014).If the reactor is overfed beyond 5.2.Mesophilic or thermophilic digestion the suitable OLR,inhibitory substrates such as fatty acids could be ac- cumulated and CH4 production could be reduced.This is because micro The AD process can take place at various temperatures,usually bacteria cannot survive in an acidic condition in the AD system.System classified into mesophilic and thermophilic temperatures.As mentioned failure can also occur due to overfeeding.This is turn affects CH4 in Section 4.1,mesophilic and thermophilic digesters usually operate at production rate,which is highly dependent on OLR (Kothari et al., 20-40C (usually 35C)and 50-65'C (usually 45C),respectively. 2014).Hence,it is important to control the OLR of the digester.Leung Thermophilic microorganisms are more sensitive to temperature and Wang(2016)demonstrated that lower OLR and longer HRT might changes compared to mesophilic bacteria.Banks et al.(2008)compared prompt organic overload,and thus,CH4 production could be reduced the thermophilic condition with the condition for source-segregated This is likely due to the insufficient buffering capability in the digester. domestic FW.They found that the AD under thermophilic conditions

FW, resulting in lower CH4 yield. In contrast, ammonia inhibition could occur, preventing microbial growth, especially if the concentration of nitrogen is more than the microbial necessity (Kothari et al., 2014; Leung and Wang, 2016). It is found that microorganisms use carbon 25–30 (Yadvika et al., 2004), 25–35 (Krishna and Kalamdhad, 2014) or 30–35 (Leung and Wang, 2016) times quicker than nitrogen during the AD process. Therefore, the optimum ratio of C/N in the substrate should be almost 20–30:1 (Yadvika et al., 2004) 25–30:1(Krishna and Kalamdhad, 2014) or 30–35:1 (Leung and Wang, 2016). The co-diges￾tion of organic substrates with animal manure, lignocellulosic biomass, and sewage sludge can be used to enhance the C/N ratios (Khalid et al., 2011). 4.4. Retention time Retention time is one of the major parameters, which needs to be repeatedly observed in the anaerobic digesters. Retention time is the time needed for the complete degradation of organic matter or the average time organic matter remains in a digester (Deepanraj et al., 2014; Mao et al., 2015). There are two important types of retention times involved in the AD system: solid retention time (SRT) and hy￾draulic retention time (HRT). SRT is the average time that the bacteria spend in the digester, whereas HRT is the average time that the liquid sludge spends in the digester (Deepanraj et al., 2014). According to Mao et al. (2015), bacterial development rate related to retention time de￾pends on process temperature, substrate composition, and organic loading rate; the shorter the HRT, the higher the value of organic loading rate. A retention time of 10–40 days is necessary to treat or￾ganic waste in mesophilic temperature, while a lesser retention time could be used in thermophilic temperature (Kothari et al., 2014). Yadvika et al. (2004) described that high capital cost and large reactor volume are the main requirements for a longer HRT. However, shorter HRT offers insufficient time for the optimal degradation of the sub￾strate. Yadvika et al. (2004) noted that HRT varies with climate change. For example, for tropical countries and in cold weather, HRT fluctuates from 30 to 50 days and 100 days, respectively. SRT maintains the bacterial population in the reactor, which could result in waste stabi￾lisation (Dobre et al., 2014). A portion of the microbial community is removed every time sludge is withdrawn, resulting in a stable condition and the prevention of process failure (Appels et al., 2008). A long SRT helps permit biological acclimation to toxic compounds, and this can be achieved by increasing both the volume of the digester and the con￾centration of the bacteria (solids) (Gerardi, 2003). Chen et al. (2018) observed that the CH4 yield reduced with higher SRT and the highest CH4 yield was achieved at an SRT of 6 days, compared to 7.5 days and 10 days. Fernández-Rodríguez et al. (2014) reported that the maximum CH4 yield was achieved at an SRT of 5–8 days compared to SRTs of 4 days and 3 days. They also indicated that SRTs lower than 4 days were inappropriate for a single-stage dry AD of organic fraction of municipal solid waste, which would render the process unbalanced. 4.5. Organic loading rate Organic loading rate (OLR) is a significant operational parameter that affects the CH4 yield. OLR is defined as the amount of dry organic solids, which can be fed into the digester per day per unit volume of digester capacity (Kothari et al., 2014). If the reactor is overfed beyond the suitable OLR, inhibitory substrates such as fatty acids could be ac￾cumulated and CH4 production could be reduced. This is because micro bacteria cannot survive in an acidic condition in the AD system. System failure can also occur due to overfeeding. This is turn affects CH4 production rate, which is highly dependent on OLR (Kothari et al., 2014). Hence, it is important to control the OLR of the digester. Leung and Wang (2016) demonstrated that lower OLR and longer HRT might prompt organic overload, and thus, CH4 production could be reduced. This is likely due to the insufficient buffering capability in the digester. On the other hand, higher OLR causes shorter HRT, which may result in microorganism washout, and this could lead to lower biogas production (Leung and Wang, 2016). Hu et al. (2018) reported that the thermo￾philic reactor can be adapt to a wide range of OLR (3.0–14.4 kg-COD/ m3 /day) compared to the mesophilic reactor (OLR of 3.0–7.3 kg-COD/ m3 /day) in anaerobic FW treatment. Liu et al. (2017) pointed out that the optimal OLR on AD of FW under thermophilic and mesophilic condition was 2.5 and 1.5 g-VS/L/day, respectively. 5. Classification of the AD process of FW The classification of the anaerobic reactor when treating FW de￾pends on the temperature, feeding type, and moisture content of the FW. Single-stage and two/multi-stage processes have also influenced the performance of the AD process (Table 2) (Deepanraj et al., 2014). Mesophilic or thermophilic, batch or continuous process, or wet or dry digestion are the various categories of the AD process, which are further described in the following sections. 5.1. Wet and dry digestion The AD process can be categorised as wet or dry digestion based on the total solid concentration in FW. The anaerobic digestion of food waste is termed as a dry process if the total solid concentration of the food waste stays between 20 and 40%, whereas the anaerobic digestion of food waste is considered as a wet process when the total solid con￾centration of the food waste is < 15% (Kothari et al., 2014; Deepanraj et al., 2014). Kothari et al. (2014) noted that most of the AD plants constructed during the 1980s depended on the wet system, whereas new plants constructed in the last decade are mostly based on the dry process. They also noted that the HRT, OLR, and volatile solid removal rate of the dry process were 14–60 days, 12–15 kg VS/m3 /day, and 40%–70%, respectively, whereas the HRT, OLR, and volatile solid de￾struction rate of the wet process were 25–60 days, < 5 kg VS/m3 /day, and 40%–75%, respectively. It is important to note that a higher cost for digestate dewatering post-treatment and higher reactor volume are required for the wet process compared to the dry process. Kothari et al. (2014) noted a higher biogas production rate in the dry process com￾pared to the wet process. Yi et al. (2014) examined dry (total solid = 20%) and wet (total solid = 5%) processes of AD during food waste treatment, and found that the CH4 production (0.48 L/g VS) and volatile solid reduction (85.6%) were higher in the dry process com￾pared to the wet process. They also found that the dry process allowed a higher VFA concentration and OLR compared to the wet process, re￾sulting in a decreased possibility of inhibition of the AD technique. The comparison of dry versus wet digestion process has shown that the dry process had a more advantageous energy balance and economic per￾formance compared to the wet process. However, complete mixing of the waste is not possible in dry process, and, thus, the ideal contact of microorganisms and substrate cannot be guaranteed. Conversely, wet process offered several important benefits including greater flexibility over the type of feedstock accepted, dilution of inhibitory substances by process water and necessity of less sophisticated mechanical equip￾ment. Overall, dry digestion process was more cost-effective, and it had greater biogas productivity than wet process. 5.2. Mesophilic or thermophilic digestion The AD process can take place at various temperatures, usually classified into mesophilic and thermophilic temperatures. As mentioned in Section 4.1, mesophilic and thermophilic digesters usually operate at 20–40 °C (usually 35 °C) and 50–65 °C (usually 45 °C), respectively. Thermophilic microorganisms are more sensitive to temperature changes compared to mesophilic bacteria. Banks et al. (2008) compared the thermophilic condition with the condition for source-segregated domestic FW. They found that the AD under thermophilic conditions S.K. Pramanik, et al. Bioresource Technology Reports 8 (2019) 100310 6

S.K.Pramanik,et al. Bioresource Technology Reports 8 (2019)100310 displayed better volatile solid removal efficiency and biogas yield in operating conditions of low HRT (2-3 days)and acidic pH (5.5-6.5), compared to mesophilic conditions.They also found that the VFA in which the acidification stage must be maintained,whereas high HRT concentration of the thermophilic reactor (45,000 mg/L)was higher (20-30 days)and optimum pH(6-8)are required for the development than that of the mesophilic reactor (28,000 mg/L).Q.Li et al.(2017) of gradually growing methanogenic bacteria (Xu et al.,2018).Hagos also compared the thermophilic digestion and mesophilic digestion of et al.(2017)reported the several benefits of the two-stage process such FW and waste activated sludge mixture.They found that the thermo- as increased CH4 production,high OLR,improved process stability, philic system (407mL/g VSdded)yielded better CH production com- higher possibility of managing pathogens,and enhanced rate of volatile pared to the mesophilic system (350 mL/g VSadded).This was likely due solid removal efficiency.They also noted that complex maintenance, to the high transformation ratio from protein to ammonium in the high capital,and operational cost were the major drawbacks of the two- thermophilic reactor stage process.Xiao et al.(2018)compared the performance and energy recovery of single-stage and two-stage thermophilic anaerobic digestion 5.3.Batch and continuous system (TAD)of FW.They found that single-stage and two-stage TAD both exhibited better performance with an HRT of 30 days.However,single- A reactor that is fed at one time with fresh FW and the addition of stage TAD achieved slightly higher CH4 yield and volatile solid removal inoculum (such as anaerobic sludge)from other reactors followed by its rate compared to the two-stage TAD.They also noted that single-stage closing down for a certain period to degrade substrate anaerobically is TAD recovered more energy than the two-stage TAD.Another study by called a batch system.Batch digesters have several benefits such as Micolucci et al.(2018)found that the two-stage TAD of FW displayed being technically simple and having low investment cost and main- clearly enhanced biogas production and organic matter destruction tenance requirements and minimum parasitic energy loss (Kothari compared to the single-stage system.The volatile solid and augmen- et al.,2014).On the other hand,for the continuous process,fresh FW is tation removal rate of the two-stage process was increased by 17%and continuously loaded in the reactor and the same quantity of digested 16%,respectively.They also indicated that the digestate dewatering FW is removed.Studies have reported that biogas recirculation or post-treatment cost in the two-stage TAD was lesser than in the single- mechanical agitators could be used in the continuous process to agitate stage,whereas the two-stage TAD produced more net energy than the the FW and inoculum continuously (Kothari et al.,2014;Deepanraj one-stage TAD.Zhang et al.(2017)established a compact three-stage et al,2014).Biogas production can be kept almost constant and/or anaerobic digester for FW in which a three-stage anaerobic digestion continuous due to the constant input of FW.Park et al.(2018)studied process was used,consisting of three independent chambers including the effect of feeding mode for the AD of FW,and reported that the the hydrolysis stage,acidification stage,and wet methane-production continuous feeding of diluted FW yielded constant performance com- stage.They found that the three-stage AD obtained 24%-54%more CH pared to the batch feeding of undiluted FW.They also noted that the yield with an OLR of 10g VS/L compared to the one-stage and two- potential CH4 yield was 2.78 L CH4/L/day at an OLR of 8.6 kg COD/m/ stage AD.The three-stage AD also achieved a high volatile solid re- day during continuous feeding of diluted FW,whereas the potential moval efficiency of 83.5%. CH yield was 1.51 LCH4/L/day at OLR of 6.1 kg COD/m3/day during batch feeding of undiluted FW. 6.Factors that improve the anaerobic digestion process 5.4.Single-stage and two/multi-stage process 6.1.Co-digestion The single-stage process occurs when four metabolic phases-hy- drolysis,acidogenesis,acetogenesis,and methanogenesis-take place in Co-digestion is defined as two or more types of organic waste,which one reactor.Low OLR,long retention time,a pH range between 6 and 7, are mixed and treated simultaneously.This process is extensively low CH4 production,less investment,and less maintenance cost are the practiced nowadays to avoid the problems associated with the mono- major features of the single-stage process (Xu et al.,2018).The main digestion of FW(Shi et al.,2018).Anaerobic co-digestion has numerous limitation in the single-stage reactor is the presence of acidogenic mi. advantages which are displayed in Fig.2 (Mehariya et al.,2018;Tyagi croorganisms (i.e.,decreases pH)that result due to the quick acid- et al.,2018).However,there are a few disadvantages of co-digestion ification of FW during the acid formation phase,which disturbs the including transport costs (Mata-Alvarez et al.,2000).The effect of co- methanogenic bacterial groups.This is likely due to the lack of an op- digestion of FW with different substrates is summarised in Table 3.The timum environmental condition inside the single-stage reactor,which is mono-digestion of FW often results in low buffer capacity and a high C/ required for the growth of all participating bacteria (Deepanraj et al., N ratio.This is likely due to the high biodegradability of the substrate 2014;Xu et al.,2018).However,adjusting the feeding rate via mixing (Leung and Wang,2016).Therefore,animal manure can be considered and adding a buffer in the single-stage reactor could be a solution to as a suitable co-substrate due to its wide variety of nutrients and high manage the methanogenic microbial community in the reactor(Monson buffer capacity,which could enhance the maximum allowable OLR (up et al.,2007).The addition of buffer (NaHCO)reduced volatile solid to 10 kg VS/m /day)and offer a more stable environment for anaerobic and biodegradation time,which resulted in higher biogas productions bacteria (Xu et al.,2018).Sewage sludge is considered to be an ex- compared with the performances of the other non-chemically buffered cellent co-substrate because of its low organic load and high trace cultures studied (Abdulkarim and Evuti,2010).Chen et al.(2015) element content (Mehariya et al.,2018).Studies have reported that studied the effects of four alkalinity sources (i.e.lime mud from pa- large amounts of active bacteria present in sewage sludge are favorable permaking,waste eggshell,CaCO3 and NaHCO3)on the stability of for the development of various groups of microorganisms engaged in anaerobic digestion from food waste.They reported that adding a the AD process (Xu et al.,2018).The same study reported that CH4 buffer in the feeding stage promoted process stability and methane yield was enhanced by 1.4 times after the addition of sewage sludge production compared to the non-chemically buffered sample.It is worth into FW.The main substrate for co-digestion was primary sludge and noting that approximately 95%of Europe's full-scale plants usually waste activated sludge,where most of the plants operated at 35'C operate in a single-stage anaerobic process when treating organic waste temperature (except for the Camposampiero and Kurobe plants).The (Nagao et al.,2012). electricity generation of the East Bay MUD plant was 11 MW,which The two-process or multi-stage process occurs in separate anaerobic was higher than the Rovereto (0.4 MW),Camposampiero (0.4 MW), reactors.The first reactor is used for hydrolysis,acidogenesis,and Treviso (0.125 MW),Moosburg (0.38 MW),and Kurobe (0.095 MW) acetogenesis,whereas the second reactor is mainly employed for me- plants (Nghiem et aL.,2017). thanogenesis.Studies have shown that the acidification stage is evident

displayed better volatile solid removal efficiency and biogas yield compared to mesophilic conditions. They also found that the VFA concentration of the thermophilic reactor (45,000 mg/L) was higher than that of the mesophilic reactor (28,000 mg/L). Q. Li et al. (2017) also compared the thermophilic digestion and mesophilic digestion of FW and waste activated sludge mixture. They found that the thermo￾philic system (407 mL/g VSadded) yielded better CH4 production com￾pared to the mesophilic system (350 mL/g VSadded). This was likely due to the high transformation ratio from protein to ammonium in the thermophilic reactor. 5.3. Batch and continuous system A reactor that is fed at one time with fresh FW and the addition of inoculum (such as anaerobic sludge) from other reactors followed by its closing down for a certain period to degrade substrate anaerobically is called a batch system. Batch digesters have several benefits such as being technically simple and having low investment cost and main￾tenance requirements and minimum parasitic energy loss (Kothari et al., 2014). On the other hand, for the continuous process, fresh FW is continuously loaded in the reactor and the same quantity of digested FW is removed. Studies have reported that biogas recirculation or mechanical agitators could be used in the continuous process to agitate the FW and inoculum continuously (Kothari et al., 2014; Deepanraj et al., 2014). Biogas production can be kept almost constant and/or continuous due to the constant input of FW. Park et al. (2018) studied the effect of feeding mode for the AD of FW, and reported that the continuous feeding of diluted FW yielded constant performance com￾pared to the batch feeding of undiluted FW. They also noted that the potential CH4 yield was 2.78 L CH4/L/day at an OLR of 8.6 kg COD/m3 / day during continuous feeding of diluted FW, whereas the potential CH4 yield was 1.51 L CH4/L/day at OLR of 6.1 kg COD/m3 /day during batch feeding of undiluted FW. 5.4. Single-stage and two/multi-stage process The single-stage process occurs when four metabolic phases—hy￾drolysis, acidogenesis, acetogenesis, and methanogenesis—take place in one reactor. Low OLR, long retention time, a pH range between 6 and 7, low CH4 production, less investment, and less maintenance cost are the major features of the single-stage process (Xu et al., 2018). The main limitation in the single-stage reactor is the presence of acidogenic mi￾croorganisms (i.e., decreases pH) that result due to the quick acid￾ification of FW during the acid formation phase, which disturbs the methanogenic bacterial groups. This is likely due to the lack of an op￾timum environmental condition inside the single-stage reactor, which is required for the growth of all participating bacteria (Deepanraj et al., 2014; Xu et al., 2018). However, adjusting the feeding rate via mixing and adding a buffer in the single-stage reactor could be a solution to manage the methanogenic microbial community in the reactor (Monson et al., 2007). The addition of buffer (NaHCO3) reduced volatile solid and biodegradation time, which resulted in higher biogas productions compared with the performances of the other non-chemically buffered cultures studied (Abdulkarim and Evuti, 2010). Chen et al. (2015) studied the effects of four alkalinity sources (i.e. lime mud from pa￾permaking, waste eggshell, CaCO3 and NaHCO3) on the stability of anaerobic digestion from food waste. They reported that adding a buffer in the feeding stage promoted process stability and methane production compared to the non-chemically buffered sample. It is worth noting that approximately 95% of Europe's full-scale plants usually operate in a single-stage anaerobic process when treating organic waste (Nagao et al., 2012). The two-process or multi-stage process occurs in separate anaerobic reactors. The first reactor is used for hydrolysis, acidogenesis, and acetogenesis, whereas the second reactor is mainly employed for me￾thanogenesis. Studies have shown that the acidification stage is evident in operating conditions of low HRT (2–3 days) and acidic pH (5.5–6.5), in which the acidification stage must be maintained, whereas high HRT (20–30 days) and optimum pH (6–8) are required for the development of gradually growing methanogenic bacteria (Xu et al., 2018). Hagos et al. (2017) reported the several benefits of the two-stage process such as increased CH4 production, high OLR, improved process stability, higher possibility of managing pathogens, and enhanced rate of volatile solid removal efficiency. They also noted that complex maintenance, high capital, and operational cost were the major drawbacks of the two￾stage process. Xiao et al. (2018) compared the performance and energy recovery of single-stage and two-stage thermophilic anaerobic digestion (TAD) of FW. They found that single-stage and two-stage TAD both exhibited better performance with an HRT of 30 days. However, single￾stage TAD achieved slightly higher CH4 yield and volatile solid removal rate compared to the two-stage TAD. They also noted that single-stage TAD recovered more energy than the two-stage TAD. Another study by Micolucci et al. (2018) found that the two-stage TAD of FW displayed clearly enhanced biogas production and organic matter destruction compared to the single-stage system. The volatile solid and augmen￾tation removal rate of the two-stage process was increased by 17% and 16%, respectively. They also indicated that the digestate dewatering post-treatment cost in the two-stage TAD was lesser than in the single￾stage, whereas the two-stage TAD produced more net energy than the one-stage TAD. Zhang et al. (2017) established a compact three-stage anaerobic digester for FW in which a three-stage anaerobic digestion process was used, consisting of three independent chambers including the hydrolysis stage, acidification stage, and wet methane-production stage. They found that the three-stage AD obtained 24%–54% more CH4 yield with an OLR of 10 g VS/L compared to the one-stage and two￾stage AD. The three-stage AD also achieved a high volatile solid re￾moval efficiency of 83.5%. 6. Factors that improve the anaerobic digestion process 6.1. Co-digestion Co-digestion is defined as two or more types of organic waste, which are mixed and treated simultaneously. This process is extensively practiced nowadays to avoid the problems associated with the mono￾digestion of FW (Shi et al., 2018). Anaerobic co-digestion has numerous advantages which are displayed in Fig. 2 (Mehariya et al., 2018; Tyagi et al., 2018). However, there are a few disadvantages of co-digestion including transport costs (Mata-Alvarez et al., 2000). The effect of co￾digestion of FW with different substrates is summarised in Table 3. The mono-digestion of FW often results in low buffer capacity and a high C/ N ratio. This is likely due to the high biodegradability of the substrate (Leung and Wang, 2016). Therefore, animal manure can be considered as a suitable co-substrate due to its wide variety of nutrients and high buffer capacity, which could enhance the maximum allowable OLR (up to 10 kg VS/m3 /day) and offer a more stable environment for anaerobic bacteria (Xu et al., 2018). Sewage sludge is considered to be an ex￾cellent co-substrate because of its low organic load and high trace element content (Mehariya et al., 2018). Studies have reported that large amounts of active bacteria present in sewage sludge are favorable for the development of various groups of microorganisms engaged in the AD process (Xu et al., 2018). The same study reported that CH4 yield was enhanced by 1.4 times after the addition of sewage sludge into FW. The main substrate for co-digestion was primary sludge and waste activated sludge, where most of the plants operated at 35 °C temperature (except for the Camposampiero and Kurobe plants). The electricity generation of the East Bay MUD plant was 11 MW, which was higher than the Rovereto (0.4 MW), Camposampiero (0.4 MW), Treviso (0.125 MW), Moosburg (0.38 MW), and Kurobe (0.095 MW) plants (Nghiem et al., 2017). S.K. Pramanik, et al. Bioresource Technology Reports 8 (2019) 100310 7

S.K.Pramanik,et al. Bioresource Technology Reports 8(2019)100310 Biomass:Carbohydrates,Proteins,Lipids, Cellulose,Hemicellulose Low nitrogen Limitations of High Rich in nutrients content trace elements alkalinity content Low alkalinity High ammonia High carbon Co-Substrate content High nitrogen (animal manure content Deficiency ot Food Waste sewage nutrient contents Lignocellulosic Low carbon content wastes) High or anic Bulk density and Toxicity substrate dilution variability Decreased and lower ammonia feed inhibition volume Food waste+ Co- Optimum Substrate pH and Increased balance buffering C/N ratio Balance of capacity nutrients and moisture content Enhanced process stability and performance Increased biogas and energy production Nutrient rich residue can be used as fertilizer Better waste management process Fig.2.Primary benefits associated with co-digestion of food waste.Information collected from Mehariya et al.(2018)and Tyagi et al.(2018). 6.2.Pre-treatment (Liu et al.,2019).Karthikeyan et al.(2018)showed that the aim of using pre-treatment is to increase the biogas yield along with en- The performance of the AD process can be enhanced with the help hancement of the hydrolysis rate kinetics for proteins and lipids.They of a pre-treatment method.As mentioned in Section 3,among the also reported that the pre-treatment process could decrease harmful biological stages,hydrolysis is the main rate-limiting step in the AD of composites,which could disturb the AD process.In principle,pre- FW.This is because every stage directly influences both the food treatment can minimise the quantity of sludge and enhance the re- availability and mass transfer of the AD process.Zhang et al.(2014) duction of organic substances.Numerous pre-treatment methods have reported that the rate of hydrolysis is affected by particle size,com- been used for the AD of FW (Table 4).Among the pre-treatment position,and substrate component during the biodegradation of gran- methods include chemical,thermal,ultrasonic,mechanical,and biolo- ular substrates and high molecular compounds.According to gical methods,which are further discussed in the following paragraph. Ariunbaatar et al.(2014),FW needs to be pasteurised or sterilised be- Mechanical pre-treatment is aimed at enhancing the specific surface fore the AD process,and pre-treatment might be the best option for area and minimising the particle size of organic polymers (Ren et al. achieving more energy recovery along with eradicating the additional 2018).An increased surface area offers excellent communication be- cost for pasteurisation/sterilisation.Liu et al.(2019)reported that most tween the substrate and anaerobic microorganisms,which can in turn biogas plants apply thermal pasteurisation according to the appropriate increase the performance of the entire AD process (Ariunbaatar et al., regulations.Although the design and the operational approach of the 2014;Ren et al.,2018).The principle of ultrasonic pre-treatment de- thermal pasteurisation differ from one biogas plant to another,when it pends on the cavitation process,which is produced from massive hydro- comes to commercially viable technology,CambiTHPM and Biothelys" mechanical shear force through high-intensity ultrasonic waves and are the global leading provider of thermal hydrolysis,advanced anae- sludge disruption (Deepanraj et al.,2017).Increased ultrasonic pre robic digestion and biogas solutions for sewage sludge and organic treatment time has many physical and chemical impacts on the sub- waste management (Panigrahi and Dubey,2019).Regarding techno- strate such as free radical (e.g.,HO2.,OH.and H.)production,re- economic feasibility,6-25%of the total primary energy production can duction in partial pressure,micro-bubble production,disintegration of be achieved in European biogas plants through thermal pasteurisation the cell walls,moderate spreading,and solubilization of solid organic

6.2. Pre-treatment The performance of the AD process can be enhanced with the help of a pre-treatment method. As mentioned in Section 3, among the biological stages, hydrolysis is the main rate-limiting step in the AD of FW. This is because every stage directly influences both the food availability and mass transfer of the AD process. Zhang et al. (2014) reported that the rate of hydrolysis is affected by particle size, com￾position, and substrate component during the biodegradation of gran￾ular substrates and high molecular compounds. According to Ariunbaatar et al. (2014), FW needs to be pasteurised or sterilised be￾fore the AD process, and pre-treatment might be the best option for achieving more energy recovery along with eradicating the additional cost for pasteurisation/sterilisation. Liu et al. (2019) reported that most biogas plants apply thermal pasteurisation according to the appropriate regulations. Although the design and the operational approach of the thermal pasteurisation differ from one biogas plant to another, when it comes to commercially viable technology, CambiTHP™ and Biothelys® are the global leading provider of thermal hydrolysis, advanced anae￾robic digestion and biogas solutions for sewage sludge and organic waste management (Panigrahi and Dubey, 2019). Regarding techno￾economic feasibility, 6–25% of the total primary energy production can be achieved in European biogas plants through thermal pasteurisation (Liu et al., 2019). Karthikeyan et al. (2018) showed that the aim of using pre-treatment is to increase the biogas yield along with en￾hancement of the hydrolysis rate kinetics for proteins and lipids. They also reported that the pre-treatment process could decrease harmful composites, which could disturb the AD process. In principle, pre￾treatment can minimise the quantity of sludge and enhance the re￾duction of organic substances. Numerous pre-treatment methods have been used for the AD of FW (Table 4). Among the pre-treatment methods include chemical, thermal, ultrasonic, mechanical, and biolo￾gical methods, which are further discussed in the following paragraph. Mechanical pre-treatment is aimed at enhancing the specific surface area and minimising the particle size of organic polymers (Ren et al., 2018). An increased surface area offers excellent communication be￾tween the substrate and anaerobic microorganisms, which can in turn increase the performance of the entire AD process (Ariunbaatar et al., 2014; Ren et al., 2018). The principle of ultrasonic pre-treatment de￾pends on the cavitation process, which is produced from massive hydro￾mechanical shear force through high-intensity ultrasonic waves and sludge disruption (Deepanraj et al., 2017). Increased ultrasonic pre￾treatment time has many physical and chemical impacts on the sub￾strate such as free radical (e.g., HO2%, OH% and H%) production, re￾duction in partial pressure, micro-bubble production, disintegration of the cell walls, moderate spreading, and solubilization of solid organic Fig. 2. Primary benefits associated with co-digestion of food waste. Information collected from Mehariya et al. (2018) and Tyagi et al. (2018). S.K. Pramanik, et al. Bioresource Technology Reports 8 (2019) 100310 8

S.K.Pramanik,et al. Bioresource Technology Reports 8(2019)100310 5100 8100 8100 TE1 UM 8100 (SIoZ)eZ 910m 160U (810Z)'Tys ppay aumaw 53710 尉 品 127E 哥 品 厨 I'OZ N/5 608 N/ Hd 'I SE D.I Ss=amwadwal (Aep u)/SA-YIO 昏 apns apnjs aqorwuv qodow a8pnis (snonunuo-ruras) T6im11802.01 (10N》C:Fe (S 4277:1.77.1.59 ( (siseq SA)I:6 厨 aiensqns-00 +poispaod amuew 8d Md Aimns MO3 Arjep alseM pooy [edpoyun aIsEM a8eqqw aIsem o]mod apns paxtu sem uapIe8 Md 31SEM Ua +Md pueim +aseM pyos edpopunu jo uopwij u pnjs Md Mens +Md Mens 1aqM eM pood Mens 1EOUM SeM pood a8pnis paleanoe sem Md 9

Table 3 Comparison of operational parameters for anaerobic co-digestion of food waste with other substrates and their respective methane production. Feedstock + Co-substrate Mixing ratio Reactor type Inoculum Operating conditions Methane yield (mL/g VS) Reference Cucumber residues + pig manure + corn stover 3:5:2 (wet basis) 1 L glass reactor (batch) Anaerobic sludge Temperature = 35 °C, C/N = 14.5, VFA/alkalinity ratio = 0.76 g/L 305.4 T. Wang et al. (2018) FW + pig manure 50:50 (VS basis) 1 L glass digester (batch) Solid digestate Temperature = 37 °C, pH = 7.6–8.7 252 Y. Jiang et al. (2018) Municipal food waste + dairy cow slurry 67.8:32.2 (wet weight) 15 L CSTR (continuous) Anaerobic digestate OLR = 5.04 g VS/L, HRT = 17.5 days, temperature = 37 ± 2 °C, pH = 7.64 444.7 Morken et al. (2018) FW + sludge 1:3 (VS basis) Glass bottle (batch) N/A Temperature = 35 ± 2 °C, pH = 5.8–7.5, C/N = 6.29 435.5 Y. Wang et al. (2018) Potato waste + cabbage waste 1:1 (VS basis) 5 L CSTR (semi-continuous) Anaerobic granular sludge OLR = 3 kg VS/(m3 day), temperature = 37 ± 1 °C, C/N = 20.1 360 Mu et al. (2017) Kitchen waste + cow manure 1:1 (wet weight basis) 1 L lab-scale anaerobic digester (batch) Anaerobic sludge OLR = 8% TS, HRT = 45 days, temperature = 35 °C, pH = 7.5. 179.8 Zhai et al. (2015) FW + garden waste + mixed sludge 67.5:22.5:10 (VS basis) 7.5 L CSTR (semi￾continuous) Anaerobic sludge OLR = 2.55–7.79 g VS/L, HRT = 30–10 days, temperature = 55 °C. pH = 7.84–7.79 424–356 Fitamo et al. (2016) FW + straw 5:1 1 L bottle reactor (batch) Anaerobic granular sludge OLR = 5 g VS/L, HRT = 8 days, temperature = 35 °C, C/N = 30.9 392 Yong et al. (2015) FW + green waste 40:60 (VS basis) 500 mL glass bottle (batch) Anaerobic sludge OLR = 5 g VS feedstock, HRT = 24.5 days, temperature = 37 ± 1 °C, C/N = 15.8 272.1 Chen et al. (2014) Organic fraction of municipal solid waste + fruit and vegetable waste (FVW) 1:3 (VS basis) 2 L glass reactor (batch) Anaerobic sludge Temperature = 35 °C, pH = 7.4–8.2, C/N = 34.7, alkalinity = 860–986 mg/L HCO3− 396.6 Pavi et al. (2017) FW + cattle manure N/A 1 L glass digesters (batch) Anaerobic activated sludge OLR = 8 g VS/L, temperature = 35 ± 1 °C, C/N = 15.8, 388 Zhang et al. (2013) FVW + slaughterhouse waste + manure 11:8:7 (% wet weight) 2 L stainless steel digesters (semi-continuous) Anaerobic slurry OLR = 1.3 kg VS/(m3 day), HRT = 30 days, temperature = 35 ± 1 °C, pH = 7.4 320 Alvarez and Lidén (2008) Food waste + wheat straw 9:1 (VS basis) 6 L CSTR (continuous) Mesophilic anaerobic sludge OLR = 3 kg VS/(m3 day), temperature = 35 ± 1 °C, pH = 7.1–7.5 344 Shi et al. (2018) Food waste + wheat straw 5:5 (VS basis) 6 L CSTR (continuous) Thermophilic anaerobic sludge OLR = 3 kg VS/(m3 day), temperature = 55 ± 1 °C, pH = 7.1–7.5 370 Shi et al. (2018) FW + waste activated sludge 3:1 3 L CSTR (continuous) Anaerobic seed sludge OLR = 7.75 g VS/L/day, HRT = 10 days, temperature = 35 °C, pH = 7.64 350 Q. Li et al. (2017) FW + waste activated sludge 3:1 3 L CSTR (continuous) Anaerobic seed sludge OLR = 5.19 g VS/L/day, HRT = 15 days, temperature = 55 °C, pH = 7.86. 407 Q. Li et al. (2017) S.K. Pramanik, et al. Bioresource Technology Reports 8 (2019) 100310 9

S.K.Pramanik, et al. Bioresource Technology Reports 8(2019)100310 421022 'le 1a younf Suos pue uoow 100) (4102)m1 010 (q910Z)1 F1 (a8od ixau uo pammuo) anaq Jo]Md poieanaid jo Ausoiod pue ssauyano poAoudu pue %9-L q Iepuaod (9Od)asta pue 'Allenuauodxa poouequa uaoid apn Put '%8'ES-6'9 '%'ZSI-8'08 Aq pastaou!sem al uoponporduopnpoudpu siskroupfy Indoa feldsp pmo Il-asedrI pue 'MAd poieanun o1 pareduroo pastarou seM uoponpo o pareduo%o1 dn pastanu uoppnpord aueaw seM uoneziqn[os doD quauneanaid inom IM (HO)1/bau 86'99I paspurndo yly pue %56<SEM Suapyja [BAOWaI OOS uoponpord auenaw 'sisKopAy pue Aqerajemap plLA II asedrI+dV ioj Tu 681 (uw 8I=aun uop yuos) -63/3-120 PUE (wwO'10J)SA3/HD TUSIS 1131O837-T.7ge Iw/HD TU 891 T/H5 T 191 118P991:1820102-H4 8.A1PD9.·1155 'aInjeladtal oliydosau 212P2902H4 a8pnjs aeMas a8pns iemnuers aapns aEMas 41009 10-P.000m 151-1.0001 100-.0006 公00-000.09 公06-000.6 1100 D.091-02I SL THObOI I-d1 pue T-asedr1: /0770822-71:0000-04 uo(oapmndue pue)u PUE D.0S 'S'Hd 'I::I Jo e pasn aIam (q oJ wdI OSI sem uopesruos oJ pondde yds)punosen 1S2T.A71001 10

Table 4 Comparison of pretreatment methods to enhance AD of food waste. Pretreatment conditions Inoculum AD reactor type and condition Methane/biogas yield Effects of pretreatment References Acetic acid (0.2 M), 62.5 °C, and 30 min Anaerobic sludge 250 mL serum bottle (batch), pH = 7.0 ± 0.2, 37 °C, 86 days 53.58 mL CH4/g VSinitial Dilute acetic acid pretreatment improves the surface roughness and porosity of pretreated FW for better bacterial accessibility compared to untreated FW. Crystallinity index was increased by 56% and maximum recovery of sugar was 95%. Methane production enhanced by 10%. Saha et al. (2018) Thermal conditions: 70 min at 55–90 °C and 50 min at 120–160 °C Anaerobic seed sludge 250 mL glass bottles, 35 °C, 21 days The cumulative biogas production increased linearly. Thermal pretreatment displayed no major impact on the final content of protein, but it reduced the fat, oil, and grease (FOG) potential by 7–36% and improved the lag phase for degradation of protein (35–65%) and FOG (11–82%). The removal e fficiency of VS, FOG, and crude protein enhanced exponentially. Y. Li et al. (2017b) Ultrasound: specific energy applied for sonication was 1040 kJ/kg TS – 18.75 L induced bed reactor (continuous), 55 ± 1 °C, 40 days 520 L CH4/kg VS Ultrasound pretreatment improves the biodegradability and the feasibility of biogas plants. COD removal was > 90%. Methane production was enhanced by about 70%. They also concluded that lower specific energy was more e fficiencies that high specific energy. Ormaechea et al. (2017) Enzyme conditions: Lipase-I, Lipase-II, and Lipase-III were hydrolyzed in the conditions of 0–36 h, 50–2000 μL and 30–50 °C. Anaerobic seed sludge Scrum bottle (batch), 35 ± 1 °C, 60 days Maximum biomethane production was 189 mL for AF + Lipase II Enzyme pretreatment could improve hydrolysis rate and biomethane production rate. Methane production rate was increased by 80.8–157.7%, 26.9–53.8%, and 37.0–40.7% for AF, VO, and FG, respectively. Lipase-I and Lipase-II could display the optimal hydrolysis performance at 24 h, 1000–1500 μL and 40–50 °C. Meng et al. (2017) Ultrasonic (20 kHz and amplitude of 80 μm). Sonication times was 9, 18, 27 min and specific energy was 1175, 2380, 3560 kJ/kg TS Anaerobic digestate 1000 mL glass reactor (batch), 35 °C, 25 days Maximum methane yield = 237 mL/g VSin (sonication time = 18 min) Methane yield improved by > 80%. TS and VS reduction was increased compared to untreated FVW. Longer sonication time consumes more energy than shorter sonication time. Zeynali et al. (2017) Alkali concentration was 40–190 mEq/L, and time was 1–6 h Sewage sludge 2 L glass AD vessel (batch), mesophilic temperature, pH = 7.0, 35 days 864.2 mL CH4/g VSdestructed Methane production increased up to 20% compared to without pretreatment. COD solubilization was optimised at 166.98 mEq/L Ca(OH)2. Junoh et al. (2016) Commercial enzymes (carbohydrate, protein, and lipid were used at a rate of 1:2:1, pH 4.5, 50 °C and 150 rpm for 24 h) Granular sludge 2.7 L UASB reactor, 35 ± 2 °C, 75 days 0.35 L-CH4/g-SCOD Combination of enzymatic hydrolysis of FW and methane fermentation increases the hydrolysis rate by reducing HRT, resulting in high methane production. SCOD removal e fficiency was > 95% and high methane content (67–75%). Moon and Song (2011) Ultrasound (specific energy applied for sonication was 7500 kJ/kg TS) Mesophilic and thermophilic digestates 5 L CSTR, 36 °C and 55 °C Maximum methane production was 0.85 L CH4/L day at 36 °C and 0.82 CH4/ L day at 55 °C Ultrasound pretreatment not only improves energy production but also allows functioning at lower retention times. Methane production increase of up to 31% and 67% for mesophilic and thermophilic temperature compared to without pretreatment. Quiroga et al. (2014) Physical (2.5 mm–8.0 mm) Anaerobic sludge 2 L complete-mix anaerobic digesters (semi-continuous), 36 °C, 140 rpm 570 mL CH4/g VS (for 2.5 mm), 515 mL CH4/g VS (for 4.0 mm) and 465 mL CH4/g VS (for 8 mm) Lower food waste particle size improves digestate dewaterability and hydrolysis. Methane production enhanced by 10–29%. Agyeman and Tao (2014) Physical (Bead mill) 300 rpm Mesophilic anaerobic sewage sludge 2 L glass reactor (batch), 37 ± 1 °C, 80 rpm, 16 days 439 mLbiogas/g-TCOD Bead mill pretreatment successfully increased solubilization by almost 40% and methane yield by 28%. Excessive size reduction of the substrate caused VFAs accumulation, resulting in reduced methane yield. Izumi et al. (2010) 1000 rpm 503 mLbiogas/g-TCOD 1000 rpm 455 mLbiogas/g-TCOD 4000 rpm 470 mLbiogas/g-TCOD 20,000 rpm 455 mLbiogas/g-TCOD 40,000 rpm 404 mLbiogas/g-TCOD Thermal (120 °C) 10 min Anaerobic sludge 250 mL glass bottles (batch), 35 °C 112 mL CH4/mL Anaerobic biodegradability for liquid phase and methane production rate can be improved after thermal pretreatment. This pretreatment also helps to recycle the floating oil. Li et al. (2016b) 30 min 152 mL CH4/mL 40 min 168 mL CH4/mL 50 min 161 mL CH4/mL 60 min 129 L CH4/mL (continued on next page) S.K. Pramanik, et al. Bioresource Technology Reports 8 (2019) 100310 10