Sustainable strategies for anaerobic digestion of oil palm empty fruit bunches in Indonesia: a review

ABSTRACT

Indonesia has abundant oil palm empty fruit bunches (OPEFBs). However, high lignin content in OPEFBs may hinder their valorisation to bioenergy resources, as its degradation by microorganisms is rate-limiting. This paper aims to review OPEFB as feedstock in anaerobic digestion (AD) systems and sustainable strategies to enhance methane yield. This paper provides a comprehensive review of the availability of OPEFB, the prospective of AD technology, and compares various biological pre-treatment and co-digestion strategies for superior AD process. Improving the value-added benefits from OPEFBs and simultaneously making pre-treatment processes more sustainable is critical to wider application. Several scenarios were proposed by combining multi-stage pre-treatment of physical and mushroom cultivation pre-treatment prior to AD. This paper presents an original approach to establishing a commercially viable and sustainable AD of OPEFBs in Indonesia by integrating multi-product biorefinery and a circular economy. Further investigation is required to ensure cost-effective and eco-friendly configurations.

Abbreviations: ABR: Anaerobic Baffled Reactor; AD: Anaerobic Digestion; AF: Anaerobic Filter; AFEX: Ammonia Fibre Expansion; AI: Artificial Intelligence; AnMBR: Anaerobic Membrane Bioreactor; ANN: Artificial Neural Network; ASP: Activated Sludge Process; CMM: Cattle Manure and Maize Silage; COD: Chemical Oxygen Demand; CPO: Crude Palm Oil; CPW: Cocoa Pods Waste; CSTR: Continuous Stirred-Tank Reactor; DH: Dry Husk; DM: Dry Matter; DMF: 2,5-dimethylfuran; EM: Effective Microorganisms; EGSB: Expanded Granular Sludge Bed; FBR: Fluidised Bed Reactors; FFB: Fresh Fruit Bunch; FTIR: Fourier-Transform Infrared Spectroscopy; GHG: Greenhouse Gasses; HRT: Hydraulic Retention Time; IC: Internal Circulation Reactor; IoT: Internet of Things; Lac: Laccase Enzyme; L-AD: Liquid-AD; LiP: Lignin Peroxidase; LRAD: Low-Rate AD; MARS: Multivariate Adaptive Regression Splines; MCC: Microcrystalline Cellulose; MDF: Medium-Density Fibreboard; MEMR: Ministry of Energy and Mineral Resources, Republic of Indonesia; ML: Machine Learning; MnP: Manganese Peroxidase; MPA: Marine Predators Algorithm; MSW: Municipal Solid Waste; OFMSW: Organic Fraction of Municipal Solid Waste; OLR: Organic Loading Rates; OPEFBs: Oil Palm Empty Fruit Bunches; OPF: Oil Palm Fibre; OPKS: Oil Palm Kernel Shell; OPMF: Oil Palm Mesocarp Fibre; OPT: Oil Palm Trunk; P(3HB): Poly(3-Hydroxybutyrate); PHA: Polyhydroxyalkanoate; PLTBg: Biogas Power Plants; POM: Palm Oil Mill; POME: Palm Oil Mill Effluent; SCG: Spent Coffee Grounds; SRF: Solid Refused Fuel; SRT: Solids Retention Time; SS-AD: Solid-State AD; TS: Total Solid; TSI: Torrefaction Severity Index; UASB: Up-flow Anaerobic Sludge Blanket; VFA: Volatile Fatty Acids; VP: Versatile Peroxide; VS: Volatile Solid; W2E: Waste-to-Energy.

KEYWORDS:

• Anaerobic digestion
• bioenergy
• biological pre-treatment
• circular economy
• co-digestion
• multi-product biorefinery

1. Introduction

Globalisation of the oil palm industry for oil production has created significant forest losses and ecosystem degradation Thus, posing a new menace to biological diversity, particularly to the South East Asia countries such as Indonesia, Malaysia, and Papua New Guinea (Gatti et al. 2019). In Indonesia, every year, the oil palm industry produces a large quantity of residual biomass in the form of empty palm oil fruit bunches (OPEFBs). A large amount of OPEFBs has become a big challenge for oil palm industries in Indonesia (Nieves, Karimi, and Horváth 2011). Each OPEFB weighs around 3.5 kg with 130 mm thickness (Teh, Goh, and Kamarudin 2010); 17–-300 mm in length and 250–350 mm in width (Kerdsuwan and Laohalidanond 2011), thus making large dumping area a mandatory requirement within the industrial facilities. OPEFBs have now gained attention as a potential feedstock to produce bioenergy and bioproducts (Majesty and Herdiansyah 2019); as it is rich in organic materials with moisture content (60–70%) (Kim et al. 2013; Yacob et al. 2005). Different ranges of bioenergy (i.e. biogas, bioethanol, biohydrogen, biodiesel, solid refuse fuels, electricity, etc.) and high value-added products (i.e. xylitol, vanillin, activated carbon, etc.) can be produced from OPEFBs (Burhani and Triwahyuni 2019; Hambali and Rivai 2017; Harahap, Silveira, and Khatiwada 2019; Heryadi et al. 2019; Nuryadi, Pratomo, and Raksodewanto 2019; Parbowo, Ardy, and Susanto 2019; Rahmi et al. 2019; Yoo et al. 2019). In more detail, the potential valorisation of OPEFBs as bioenergy and high-value-added products, is shown in Figure 1.

Figure 1. Potential valorisation of OPEFBs as bioenergy and high value products (The figure is adapted from Kasivisvanathan et al. (2012); Vaskan, Pachón, and Gnansounou (2018); Ahmad et al. (2019); and Hafyan et al. (2020)). MCC: microcrystalline cellulose, MDF: medium–density fibreboard, PHA: Polyhydroxyalkanoate, P(3HB): Poly(3–Hydroxybutyrate), SRF: solid refused fuel.

Anaerobic digestion (AD) is a commercial conversion technology resulting in renewable energy production in methane (Appels et al. 2011). AD provides a pathway for a closed balanced carbon cycle concerning atmospheric carbon dioxide and utilises wastes and biomass as a significant renewable energy resource (Chynoweth, Owens, and Legrand 2001). Therefore, AD is considered one of the best ways to achieve sustainable energy development goals in many developing countries, including Indonesia. AD technology was a well-established technology applicable for first- to fourth-generation biomass, as shown in Table 1. With most of the country's consumption coming from fossil-based fuels and its inability to local oil production since 2004, there is an urgent need to adopt sustainable and renewable energy production, such as producing biogas from OPEFBs (Khalil et al. 2019). Previous studies suggest that implementing AD technology for biogas production provides better economic and environmental benefits (Suhartini et al. 2022a; 2022b). Various studies have also emphasised the opportunities for transforming OPEFBs into biogas under various operational conditions, operated in mono- and co-digestion systems (Chaikitkaew, Kongjan, and O-Thong 2015; Hidayat et al. 2020; Nieves, Karimi, and Horváth 2011; O-Thong, Boe, and Angelidaki 2012; Suhartini, Hidayat, and Nurika 2020; 2021; Suksong et al. 2017a; 2016; 2017b; 2020a; 2020b). However, several problems were found to influence the biogas production from OPEFBs by AD. Several studies have emphasised that operating AD of OPEFBs in mono-digestion systems without pre-treatment has low efficacy (Suhartini, Hidayat, and Nurika 2020; 2021; Suksong et al. 2017b). In addition, several studies have highlighted that operating AD of untreated OPEFBs under thermophilic temperature has superior performance than mesophilic temperature (Suksong et al. 2017a; Walter et al. 2015). In practice, this could be associated with higher energy and operational costs if it cannot maximise the net energy production (Suryawanshi, Chaudhari, and Kothari 2010). Furthermore, the low methane production was mostly found in anaerobic mono-digestion of untreated OPEFBs due to its complex lignocellulosic structure (Rohma, Suhartini, and Nurika 2021). Many studies have highlighted that OPEFB has low biodegradation in the AD process mainly due to its high lignin content (20.4%–28.8%) (de Gonzalo et al. 2016; Hidayat et al. 2020). Nieves, Karimi, and Horváth (2011) found that the higher the lignin content in biomass, the lower the biodegradability.

Table 1. Biomass resources and the conversion routes for biogas production.

Type of biomass	Raw materials	Conversation routes to biogas	Biogas yields	Advantages	Disadvantages
First generation	Plants as edible plants (i.e. rice, corn, wheat, barley, and sugar cane) and energy crops	Cultivation, Harvesting, Transportation, Storage, Physical pre-treatment (i.e. size reduction), AD	low	- Plant is rich in sugar and starch content - Has higher energy yield	- Conflict of food vs fuels - Required extensive agricultural land - Low environmental sustainability (not sustainable) - Increase greenhouse gas (GHG) and environmental issues
Second generation	Non-edible plants (i.e. grass and aquatic plants) and lignocellulosic biomass from food crop waste, forestry residues, agricultural waste, municipal and industrial waste	Collection, Transportation, Storage, Pre-treatments (i.e. physical combined with others); AD	medium to high	- Cheap, abundant, and renewable resources - No conflict over food vs fuels - Improve environment of waste valorisation - Lignocellulosic bioethanol has better GHG reduction potential - Reduce arable land use as plant can grow on marginal land or no land requirement for waste biomass resource	- Recalcitrant cell wall structure of lignocellulosic feedstock - Require pre-treatment of feedstock - Conversion technology hurdles - Cost of technology and pre-treatment
Third generation	Marine biomass, i.e. algae (macroalgae and microalgae)	Cultivation, Harvesting, Transportation, Storage, Pre-treatments (i.e. physical combined with others); AD	high	- High growth rate - No competition over food or feed plants - Has high carbohydrate composition, which higher than soybean (60-fold) or corn (10-fold) - Has high lipids content - No competition over use of agricultural/arable land or use of fresh water for cultivation - Better environment as CO2 sink	- Cost of technology - High operational cost (i.e. for cultivation and harvesting system) - Required innovation for efficient conversion - Technical limitation and process's difficulties
Fourth generation	Genetically modified plant (i.e. algae) and microorganisms (i.e. cyanobacteria, Bacillus anthracis, B. subtilis, Acinetobacter calcoacetius, Arthrobacter sp)	Algae: - Genetic modification for higher CO2 capture and lipid production, cultivation, harvesting, Pre-treatments (i.e. physical combined with others); AD - Microorganisms: - Genetic modification, CO2 fixation (photoautotrophic metabolism)	Much higher	- Fast growing with higher yields - Algae can be grown in wastewater streams thus reducing freshwater demand - No competition over food or feed plants - No competition over land use	- Cost of technology - Technical limitation and high investment - Difficulties in modification process - Complex process for searching of correct species

Sources: Abdullah et al. (2019); Aro (2016); Bateni et al. (2014); Dębowski et al. (2013); Dutta, Daverey, and Lin (2014); Ghosh et al. (2020); Grangeiro et al. (2019); Hofrichter (2002); Kumar et al. (2020); Lü, Sheahan, and Fu (2011); Martínez-Gutiérrez (2018); Moravvej, Makarem, and Rahimpour (2019).

Lignin functions as a protective component to maintaining a rigid structure in plants; thus, opening its compact structure increases the biodegradability (Nieves, Karimi, and Horváth 2011). Therefore, to increase the efficacy of AD of OPEFBs, the delignification is pivotal for releasing cellulose and hemicellulose. Furthermore, breaking the cellulose-hemicellulose-lignin bonds can also help to result in higher yields of reducing sugars (Prasad et al. 2019); and higher biogas or methane yields (Taherzadeh and Karimi 2008). Hence, pre-treatment of OPEFBs becomes essential to obtain maximum biogas/methane yield. Furthermore, the choice of pre-treatment is crucial for the overall success of the AD systems (Waluyo et al. 2018). Pre-treatment methods are generally divided into physical, physico-chemical, chemical, and biological pre-treatment, and their comparison is shown in Table 2.

Table 2. Summary and comparison of lignocellulosic biomass pre-treatment.

Method	Technique	Operational Definition	Purpose	Generated Results	Strengths	Weaknesses
Physical	Milling	Cutting lignocellulosic biomass into smaller pieces	Reducing the particle size and crystallinity to increase the surface area and decrease the polymerisation degree	Increasing the yield of fermentation products and producing no inhibitors	• Increasing the surface area to improve it can come into contact with the next enzyme / process • Decreasing the degree of polymerisation and crystallinity	• Requires high energy which makes the whole processes is expensive • Requires special equipment for extrusion, pyrolysis etc. • On a large-scale production, this technique may not be feasible
	Ultrasound/Ultrasonic	Radiating for 10 min with ultrasonic radiation at frequency in the range of 20–200 kHz	Disintegrating and breaking of the β-O-4 and α-O-4 bonds of lignin, which lead to the formation of cavitation bubbles by separating the polysaccharides and the lignin fraction of the biomass	Increasing pressure (100–5000 ATM) and temperature (500–15,000 K), thus changing the substrate's morphology. The cavitation causes disruption of the cell wall and dislocation of the middle layer, thus more cellulose can be accessed by enzymes
	Extrusion	Drenching biomass and pressing at a high temperature and certain frequency	Reducing the particle size, enhancing the particle unit, and increasing the physical properties of biomass	Changing the chemical structure and shortening the biomass fibre
	Pyrolysis	Heating biomass substrate at 200–600°C without or with oxygen	Triggering the chemical decomposition of organic matter, as well as the decomposition of biomass's chemical constituents.	Separating the lignocellulose composition. The resulted products are in the form of gas, oil and char, also called as syngas, bio-oil, and bio-char
Chemical	Acid	Dissolving biomass substrate in an acid solution with combination of low temperature and high acid concentration (concentrated) or high temperature and low acid concentration (diluted).	Dissolving hemicellulose to increase the porosity of biomass, thus cellulose is more accessible to enzymes	Hemicellulose is removed causing an increase in pore volume overall sugar yield. However, there is a risk of forming an inhibitor product. Using diluted acids is more promising than concentrated acids	• The use of alkaline effectively removes lignin • The use of acid effectively removes hemicellulose • High sugar yields • Lower cost than physical pre-treatment	• The formation of inhibitor compounds (i.e. 5-hydroxymethyl furfurals, aldehydes and phenolic acids) during acid pre-treatment • Require corrosion resistant media because of chemicals used • Alkaline pre-treatment requires a high downstream cost and not feasible for industrial scale
	Alkaline	Dissolving the substrate with alkaline solution (concentrated / liquid) with combination of at room temperature and a longer time or at high temperature and a shorter time	Breaking the bonds (i.e. esters, aryl-ether, -C-bonds) associated with lignin polymers and carbohydrates. Expanding the accessible surface area, removing acetyl groups, and reducing the crystallinity of cellulose.	Hemicellulose and some parts of lignin are dissolved, causing loss of some sugar available for fermentation and initiating inhibitory compounds
	Organosolv	Using organic solvents (i.e. methanol, ethanol, acetone, ethylene glycol, etc.), with or without addition of catalysts (i.e. organic, inorganic acid, alkaline).	Removing lignin and hemicellulose through solubilisation (cleavage of α-O-aryl bonds, β-O-aryl bonds, and 4-methylglucuronic acid ester bonds), thus increasing the surface area to reach cellulose	Lignin is able to be recovered as a value- added product and low energy use
	Ionic Solution	Using combination of solvents (i.e. cations and anions), which mixed with biomass, heated, and filtered with vacuum.	Interfering the non-covalent bonds of lignin, hemicellulose, and cellulose which seize for the hydrogen bonds between them.	Effective hydrolysis occurs and the ionic solution can be recycled. The process of recycling ionic solutions does not require a lot of energy
Physico-chemical	Ammonia fibre explosion	Mixing liquid ammonia at room temperature and high pressure, then held before release	Interfering lignin-carbohydrate bonds, enlarging biomass, decreasing cellulose crystallinity, and deacetylation of hemicellulose	The production of inhibitor compounds is low and the cellulose and hemicellulose content can be maintained	• The combination of physical and chemical can reduce damage due to pre-treatment, thus increase sugar yield • Feasible in large scale industry • The most likely method to be applied for various biomass • Less inhibitory and toxic formation	• High cost of technology as it requires specially designed equipment and a control environment • High capital investment • There is a possibility of partial degradation of hemicellulose and lignin due to a slight damage of substrate
	Steam explosion	The biomass particles are flowed with hot steam at 160–270°C for a few second to minutes, then released, immediately cooled down and the biomass became moist.	The sudden release aims to cause auto hydrolysis of the acetyl hemicellulose group, resulted in disruption and separation of the cell wall structure	Increasing the accessibility of cellulose, some hemicellulose and lignin are eliminated / dissolved. However, inhibitor compounds are formed, which may inhibit the conversion process and yield reduction
Biological	Fungi	Using brown-, white-, and soft-rot fungi to degrade lignin as the fungi can produce peroxidase and laccase enzymes	Remodel lignin into its constituent component using an oxidative ligninolytic enzyme system that attacks the phenyl ring in lignin and a hydrolytic enzyme system that degrades cellulose and hemicellulose to release fermentable sugars	Lignin is converted into aromatic compounds or bioethanol.	• Low energy and chemical requirement • The process can be carried out in mild environmental condition • Selective lignin and hemicellulose removal • No inhibitor compounds formation • Environmentally friendly process • The used enzyme cannot be reused	• A very slow rate of microbial growth makes the whole process of industrial scale is really expensive • High operational cost • Selection and a very efficient development of microbes for delignification are needed • Efficiency of degradation is still low • Microbes also utilise the biomass as the source of carbon metabolism; thus the expected sugar yield is low
	Enzymatic	Using pure and specific enzymes to degrade lignin	Catalyzing the oxidation of aromatic compounds in lignin, thus lignin can be degraded	Enzyme is able to remove biomass components as needed.
	Microbial	Using pure or consortia microorganism producing the respective enzyme to degrade lignin	In the hydrolysis step, the microorganisms, under aerobic condition, produce enzyme that effective to degrade lignin polymer into simple sugar (i.e. glucose)	Respective enzyme produced by microorganism is able to reduce lignin and increase methane yield
	Ensilaging	Using anaerobic/aerobic microbial to produce organic acids	The carbohydrates in the biomass samples are reduced to organic acids and alcohol	During the ensilaging process, organic acid is produced, pH is reduced, and WSC increases
	Mushroom	Using edible mushroom to convert lignocellulosic into simple sugar	Reducing the concentration of cellulose, hemicellulose and lignin into small molecular sugars, further used as carbon source for the growth of mushrooms	Cellulose, hemicellulose and lignin is converted into simple sugars, and thus increases readily available organic matter for AD process or increases mushroom yield

Sources: Baghbanzadeh et al. (2021); Hendriks and Zeeman (2009); Kumar et al. (2020); Pérez-Chávez, Mayer, and Albertó (2019); Prasad et al. (2019); Ravindran and Jaiswal (2016); Sieborg et al. (2020); Tanjore, Richard, and Marshall (2012); Zou et al. (2020).

Physical pre-treatment tends to increase the accessible area and size of pores and decrease the cellulose crystallinity and polymerisation degrees. However, it can often be slow, energy-intensive, and prohibitively expensive (Kristiani et al. 2015). Chemical pre-treatment utilises chemical compounds to alter the lignocellulosic components in the OPEFBs, thus reducing their recalcitrance for further utilisation processes. A typical chemical pre-treatment performed on OPEFBs is acid and alkaline pre-treatment. In alkaline pre-treatment, a large amount of alkaline (e.g. NaOH) is used to alter the intermolecular bonds in the lignocellulosic materials (i.e. xylan, hemicelluloses). This method removes the structural and compositional obstacles for processes (i.e. hydrolysis), improves the digestibility rate, and increases the yield of fermentable sugars available for further conversion (Sudiyani et al. 2013; Triwahyuni et al. 2015). Recent studies on chemical pre-treatment of OPEFBs have also been investigated concerning bioenergy production. These methods include the use of aqueous ammonia, peracetic acid/alkaline peroxide, phosphoric acid, bisulphite, acid hydrolysis, and organosolv pre-treatments (Pangsang et al. 2019). Most of these methods use a range of chemicals or organic solvents at high temperatures such as 100–200°C for a short period, either in the presence or absence of catalysts (Chin et al. 2019). Chemical pre-treatments obtain high sugar concentration for most of the methods employed; however, the process's washing and neutralisation steps make it non-feasible for the overall process economics. Furthermore, consideration of the inhibitory compound formation, cost of chemicals, and potential chemical wastewater treatment, limited its application. Also, many studies have emphasised the potential inhibitory effect of chemical pre-treatment residues on AD microbial consortia which leads to reduce biogas and methane yield (Liew et al. 2021; Rohma, Suhartini, and Nurika 2021; Stanley et al. 2022; Zhang et al. 2018; Zheng et al. 2014). For example, our previous study by Rohma, Suhartini, and Nurika (2021) found that the presence of excessive acid residues after acid pre-treatment may inhibit the AD process. Furthermore, Suhartini et al. (2022c) also reported that acid pre–treatment is not a feasible and sustainable option as it requires high operation costs for additional safety precautions, corrosion–resistant vessels, chemicals, safe disposal of waste chemical, and treatment of chemical wastewater.

Hence, the focus is more on chemical-free options such as hot-thermal water pre-treatment or biological pathways (Pangsang et al. 2019). Biological pre-treatment (alone or in combination with other pre-treatment) has been considered an economical pre-treatment method compared with other pre-treatment options (Saritpongteeraka et al. 2015). As shown in Table 2, despite some general weaknesses, biological pre-treatment may offer better economic and environmental benefits in terms of cost- and energy-saving, low use of chemicals, green operation, and no inhibitory compounds. Many studies have also highlighted that sustainable strategies for enhancing bioenergy production from lignocellulosic biomass should encompass appropriate green pre-treatment, intensification of conversion technology, and post-treatment operations (Ab Rasid et al. 2021; Bhutto et al. 2017; Kurian et al. 2013). The proper and cost-effective pre-treatment mays support the successful implementation of sustainable biorefinery of lignocellulosic biomass (Ab Rasid et al. 2021), including OPEFBs.

The utilisation of residues from the oil palm industry for biogas production using AD technology has been reported in Indonesia, Malaysia, and Thailand (as the top three oil palm producers). As already discussed, high concentrations of lignin and cellulose have negative synergy on biomass digestibility. These materials are inhibitory for biogas production as the biomass becomes less available for the microorganisms in the digester. Also, their further degradation leads to the accumulation of long-chain fatty acids, which are detrimental to AD (Temu et al. 2016). Therefore, green pre-treatments should be employed to achieve a cost-effective and sustainable biogas production via the AD process. Many studies have highlighted the various pre-treatments options for OPEFBs for bioenergy production. But, limited in-depth studies that focused on green pre-treatments (specifically biological methods) to support sustainable biorefinery of OPEFBs through biogas production in Indonesia. Therefore, this paper investigates the potential to implement biological methods for enhanced methane production from the AD of OPEFBs. This paper also investigates future perspectives on economically feasible and sustainable pre-treatment strategies for enhancing the efficacy of AD of OPEFBs. This paper presents a detailed analysis and comparison of key techniques in OPEFBs pre-treatments and AD, as well as the potential strategies proposed to support the established sustainable AD of OPEFBs in Indonesia.

2. OPEFB – availability and its potential for AD feedstock

2.1. Oil palm industry and the availability of OPEFBs

Oil palm grows mainly in Africa, Latin America, and Asia. Indonesia and Malaysia are the leading oil palm producers worldwide, followed by Thailand, Colombia, and Nigeria (Shahbandeh 2021; 2020). Indonesia has the largest certified palm planted area worldwide, accounting for 12.8 million hectares in 2018 (Hirschmann 2020). Oil palm plantation in Indonesia is divided into private plantations (49.81%), people's plantations (45.54%), and the country's estates plantations (4.65%). In 2018, the total production of oil palm in Indonesia was approximately 26.90 million tonnes, which was dominated by 5 (five) provinces, including Riau at 7.14 million tonnes (19.50%), Kalimantan Tengah at 5.76 million tonnes (15.74%), Sumatera Utara at 5.45 million tonnes (14.88%), Kalimantan Barat at 3.07 million tonnes (8.40%), and Sumatera Selatan at 3.04 million tonnes (8.31%), respectively (BPS-Statistics Indonesia 2018). These oil palms are mostly used for Crude Palm Oil (CPO) production, with a total value of 36.6 million tons in 2018 in Indonesia and predicted to increase annually (Hirschmann 2020). The global oil palm production was 56.38 million metric tons in 2012 and increased to 75.45 million metric tons in 2020 (Shahbandeh 2021; Statistica Research Department 2021). Similar trends were also observed in Indonesian oil palm production. The production increased from 26.02 million metric tons (in 2012) to 48.30 million metric tons (in 2020), giving an average increase of 53.08% annually. The global and Indonesian production of oil palm is shown in Figure 2. The figure indicates that Indonesia has abundant oil palm production, hence a high potential for residual waste, including OPEFBs.

Figure 2. Production of palm oil worldwide and in Indonesia from 2012 to 2019 (in million tons) (The figure is adapted from Sharbandeh (2021); (2020) and Statistica Research Department (2021)).

Furthermore, Suhartini et al. (2022c) reported more than 700 palm oil mills (POMs) in Indonesia, which is predicted to grow each year. Three big companies are producing CPO in Indonesia, including Golden Agri-Resources Ltd. (GAR) (produces 2.44 million tonnes CPO), Wilmar International (produces 1.97 million metric tonnes CPO), and IOI Group (produces 756.6 thousand metric tonnes) (Hirschmann 2020). Figure 3 illustrates the oil palm value chain and material flow (i.e. material balance) in a POM. It can be seen that various stakeholders are playing a pivotal role in the oil palm industry, from material acquisition to CPO manufacturing (Figure 3(a)). In Indonesia, tied/plasma, independent, and associated smallholders are critically important to ensure the oil palm's availability and sustainability supply to the industry, income generation for local farmers, and employment opportunities (Pacheco et al. 2017). Evidence also shows that the oil palm industry has contributed significantly to Indonesia's economy over the decades (Pramudya, Hospes, and Termeer 2017; Purnomo et al. 2020). Despite this, many have indicated a great environmental impact (Ayompe, Schaafsma, and Egoh 2021; Gatti et al. 2019; Meijaard et al. 2020; Teng, Khong, and Che Ha 2020). Considering the global commitment to bioeconomy and sustainability, the oil palm industries in Indonesia have made an effort to green and sustainable plantation and manufacturing processes (Ayompe, Schaafsma, and Egoh 2021; Boons and Mendoza 2010; Purnomo et al. 2020).

Figure 3. Palm oil value chain (a) (Adapted from Pacheco et al. (2017)), and the material flow in POM (b) (Adapted from Rizal et al. (2018)). CPO: crude palm oil, FFB: fresh fruit bunches, OPEFB: oil palm empty fruit bunch, OPF: oil palm fibre, OPKS: oil palm kernel shell, OPMF: oil palm mesocarp fibre, OPT: oil palm trunk, POM: palm oil mill, POME: palm oil mill effluent.

Indeed, an increase in CPO demand or production is followed by increasing the number of oil palm plantations and the waste resulting from POMs (Risdianto, Kardiansyah, and Sugiharto 2016). Figure 3(b) shows the material balance at a POM, indicating potential waste residual generation. A study reported that 1 ton of fresh fruit bunch (FFB) can produce approximately 0.2–0.3 tons of OPEFBs (Suksong et al. 2020a). Other studies reported that POMs produce OPEFBs with the value of 25% (w/w) of FFB (Santi, Kalbuadi, and Goenadi 2019); or about 22% (w/w) of FFB (Abdullah and Sulaiman 2013). While according to Rizal et al. (2018), as shown in Figure 3(b), about 23% of OPEFBs have resulted from each ton of FFB used in CPO production, with a 20% yield of FFB. Similarly, our previous study also suggested that, on average, a POM produces 21% CPO and 22.5% OPEFBs from processing 1 ton FFB (Suhartini et al. 2022c). Therefore, based on the CPO production of 48.30 million metric tons in 2020 in Indonesia (as shown in Figure 1), this equates to ∼241.5 million tons of FFB (input) and potential OPEFBs of 55.54 million tons. This data shows a huge availability of OPEFBs in Indonesia. This biomass can be considered a highly potential biomass resource to be further valorised as bioenergy or high value-added products.

Considering the current potential data of OPEFBs in Indonesia, further measures are essential to prevent any detrimental environmental impact from the direct disposal of OPEFBs. Furthermore, higher cellulose, hemicellulose, and lignin content in OPEFBs make them attractive to bioenergy production, via thermal or biological routes. For instance, to date, about 700 POMs (with an average production capacity of 30–45 FFB/h) across the country have utilised OPEFBs for local bioenergy generation. This is done via on-site biomass power plants (up to 3500 MW electricity) and biogas power plants (PLTBg) operated in co-digestion with palm oil mill effluent (POME) (∼700 MW electricity) (Suhartini et al. 2022c). Another potential valorisation of OPEFBs is bioethanol production due to their higher cellulose content in OPEFBs. With the combination of various pre-treatments (for delignification) and fermentation, OPEFBs were found to have higher bioethanol yields (Harahap, Mardawati, and Nurliasari 2020a; Mardawati et al. 2019; Suhartini et al. 2022c). Yet, limited studies have reported the commercial bioethanol plant using OPEFBs as feedstock in Indonesia. In 2019, about 0.40 million L/year of bioethanol is produced in the country, but mostly using molasses as feedstock (Ministry of Energy and Mineral Resources (MEMR), 2019). In addition, a high hemicellulose content in OPEFBs also makes it viable for the production of xylitol, an artificial sweeter used in food, pharmaceuticals, beverages, health and personal care, and animal feed industries (Harahap, Mardawati, and Nurliasari 2020a; Mardawati et al. 2022; 2018; Suhartini et al. 2022c). By combining pre-treatment and biological routes (i.e. enzymatic and yeast fermentation), hemicellulose is converted into xylose and later into xylitol (Mardawati et al. 2022; 2018; 2017; 2014; Suhartini et al. 2022c). Currently, the top three leading countries as xylitol producers are China, the USA, and India (Ahuja et al. 2020); still no commercial xylitol manufacturing is available in Indonesia (Suhartini et al. 2022c). High lignin content in OPEFBs also makes it feasible for direct valorisation as briquettes (Cabrales, Arzola, and Araque 2020; Nyakuma, Johari, and Ahmad 2012); solid fuel pellets (Munawar and Subiyanto 2014; Nyakuma, Wong, and Oladokun 2021); fibre- and particle-board or reinforced polymer composites (Ramlee et al. 2021a; 2021b; Zuhri, Sapuan, and Ismail 2009). These studies demonstrated that the chemical composition of OPEFBs may influence the best conversion routes for bioenergy or bioproducts.

2.2. Potential use of OPEFBs as AD feedstock

OPEFBs are non-wood lignocellulosic biomass, known as a high organic material resource (O-Thong, Boe, and Angelidaki 2012). OPEFBs have moisture content in the range of 66–69%, volatile matter of 86.5–87.7%, fixed carbon of 10.8–14.5%, ash of 3.7–5.3%, and a calorific value of 18.88 MJ/kg (Sukiran et al. 2017). Generally, OPEFBs contain carbon (48.72%), hydrogen (7.86%), nitrogen (0.25%), and oxygen (48.18%) (on a dry basis), respectively (Loh 2017). As lignocellulosic biomass, OPEFBs primarily consist of cellulose, hemicellulose, and lignin, which create a highly resistant and recalcitrant biomass structure (Rizal et al. 2018). Cellulose is a linear homopolysaccharide that contains crystalline and amorphous structures (Sukiran et al. 2017). Hemicellulose is a highly branched heteropolysaccharide, consisting of a wide variety of sugars (such as glucose, mannose, galactose, xylose, arabinose, 4-O-methyl glucuronic acid, and galacturonic acid residues) (Sawatdeenarunat et al. 2015). Anaerobic microorganisms can utilise cellulose and hemicellulose for biogas production (Suksong et al. 2016). Meanwhile, lignin, a complex polymer of phenylpropane units, is the most factor limiting the biodegradability of lignocellulosic biomass (Xu et al. 2019). Lignin encapsulates cellulose and hemicellulose by crosslinking through hydrogen and covalent bonds, forming a hydrophobic three-dimensional structure. This complex structure prevents plant cell walls from rapid degradation by microorganisms, thereby rendering biomass hard to be efficiently converted during AD (Ma et al. 2019). Li et al. (2019a) reported that the methane yield of cellulose was higher than hemicellulose, whereas lignin cannot be digested. However, the co-digestion of cellulose and hemicellulose increased methane yield and biodegradability. It suggested that biomass with high cellulose and low lignin content has higher biogas and methane production.

Thus, OPEFBs are potential renewable energy feedstock for biogas production via AD technology (Hidayat et al. 2020; O-Thong, Boe, and Angelidaki 2012; Suhartini, Hidayat, and Nurika 2020; 2021; Suksong et al. 2016; 2019b; 2020b). A study reported that AD of OPEFBs at a C/N ratio of 25–30 could have methane yields of 50–54 m³/ ton OPEFBs (Suksong et al. 2016). Another study suggests that methane yield from the AD of OPEFBs at operating conditions of TS (16% (w/w)), C/N ratio (30), and food to inoculum (F:I) ratio (2:1) was 0.358 m³ CH₄ /kg VS (Chaikitkaew, Kongjan, and O-Thong 2015). OPEFB is rich in carbon but low in nitrogen. In the AD system, the C/N ratio indicates the nutrient balance that anaerobic microorganisms require for growth (Korres et al. 2013). Furthermore, a study reported that without pre-treatment, the AD process shows a low conversion efficiency (<60%) (Suksong et al. 2017a). Our previous study, reported in Suhartini et al. (2021), that mono-digestion of untreated OPEFBs gave the lowest methane yield (0.185 m m³ CH₄ /kg VS) compared to other organic wastes (i.e. rice straw, maize straw, vegetable waste, jackfruit straw, banana peel, orange peel, apple peel, and pineapple peel). This was due to the high lignin content in the OPEFBs. Therefore, specifically for OPEFBs, pre-treatment to remove lignin or to extract cellulose and hemicelluloses is necessary to enhance biogas and methane yields.

3. Anaerobic digestion of OPEFBs

3.1. AD process and influencing parameters

Biomass conversion into energy sources can use biological processes by converting complex organic components into value-added metabolites. This transformation process is carried out in an anaerobic digester that uses microorganisms to degrade organic components into methane and other products (Ohimain and Izah 2017). This process is also defined as anaerobic digestion (AD), a biochemical process where the breakdown of complex feedstock releases energy into simpler units in the absence of oxygen (Zhang, Hu, and Lee 2016).

AD is a complex process consisting of four phases of degradation named hydrolysis, acidogenesis, acetogenesis, and methanogenesis (Deublein and Stenhauser 2011). The dominant biogas containing methane gas is produced in the methanogenesis process (Nieves, Karimi, and Horváth 2011; Suksong et al. 2020b). The operations inside an anaerobic digester are complex and very sensitive to small changes in conditions, where each intermediate step is governed by strict requirements (Horan, Yaser, and Wid 2018). Therefore, the condition of each process significantly affects biogas production. There is a reshuffling process of complex organic matter in hydrolysis, where carbohydrates become glucose, proteins become amino acids, and fats become fatty acids. In the acidogenesis phase, the monomer sugars are broken down into volatile fatty acids (VFA). In the acetogenesis phase, the VFA is used as a substrate for methanogenic bacteria in the methanogenesis process. The methanogenesis phase is the most crucial in AD because this process is susceptible and influenced by certain factors (such as raw material composition, feedstock rate, temperature, and pH) (Li et al. 2020; Van et al. 2020). Methanogenic bacteria are susceptible to oxygen, and low oxygen concentrations can kill all methanogenic bacteria (Neves, Oliveira, and Alves 2004). In addition to the methanogenesis phase, a dark fermentation process can also be carried out to produce H₂ gas. The stages in the AD process can be seen in Figure 4. Furthermore, methane gas produced from AD ranges from 50 to 70%, where the higher the methane produced, the greater the potential energy that can be used (Ohimain and Izah 2017). In the study of Zhang et al. (2007), biogas contains methane of 50–80% CH₄ and CO₂ of 20–50%, while in the study of Neves, Oliveira, and Alves (2004), the biogas contains 60–70% CH₄ and 30–40% CO₂.

Figure 4. Pathways of anaerobic digestion for biogas production (Adapted from Rasapoor et al. (2020); Li, Chen, and Wu (2019b); and Rabii et al. (2019)).

Biomass resources that contain lower (<40%) or higher water content are potential as feedstocks in the AD systems (Appels et al. 2011). Ideally, for higher efficacy performance of a dry (or high-solids) AD system, biomass should have a moisture content ranging from 20 to 40% (Rocamora et al. 2020); or any biomass with moisture content below 80% (or total solids/TS >20%) (Chiumenti, da Borso, and Limina 2018; Jansson et al. 2019). While biomass with moisture content above 85% (or TS < 15%) is better treated using a wet (or low-solids) AD system (Jansson et al. 2019). Compared with a wet-AD, operating a dry-AD system offers a lower water consumption, a lower water content in digestate (reducing digestate volume for storage and treatment), and a smaller reactor size (Chiumenti, da Borso, and Limina 2018; Jansson et al. 2019; Rocamora et al. 2020). However, a dry-AD needs a longer retention time, mixing for homogenisation, and often has a problem of the feedstock’ pumping than a wet-AD system (Jansson et al. 2019). Furthermore, Mata-Alvarez et al. (2014) added that AD could be operated either using single substrates (mono-digestion) or with the addition of co-substrates (anaerobic co-digestion). However, there were various disadvantages to anaerobic mono-digestion, such as the variability of substrate properties, seasonality, or the presence of inhibiting/toxic materials. Therefore, the study further added that co-digestion of two or more substrates is more economically feasible and can produce higher methane production than mono-digestion. Khalid et al. (2011) claimed that AD of organic solid waste might potentially be a better option to fulfil the future energy demand.

Various factors strongly influence biogas production, including nutrients, organic loading rates (OLR), hydraulic retention time (HRT), activity and population of microorganisms, presence of inhibitors/toxic substances, pressure, mixing conditions, substrates composition, alkalinity, pH, and temperature (Ohimain and Izah 2017). These operational parameters require constant monitoring and adjustment according to the biomass composition and the reactor conditions (Mao et al. 2015). Nutrients used in AD can be in the form of macro- and micro-molecules such as carbohydrates, proteins, fats, cellulose, nitrogen, sodium, magnesium, and calcium (Appels et al. 2011; Taricska et al. 2009). The growth and activity of microorganisms in AD reactors varies depending on each transformation process but the most crucial role in the formation of biogas/methane is methanogenic bacteria (Appels et al. 2011; Chen et al. 2014a; 2008). The development of these microorganisms must be avoided from any inhibitory substances such as light metal ions, heavy metals, and organic compounds (Chen et al. 2014a; 2008). The overall pressure inside the AD reactor is essential during methane production (Díaz et al. 2020; Lemmer et al. 2017). If the weight of the gas outside the reactor is greater than that inside the bioreactor, the pressure becomes negative, and the reactor explodes (Chen et al. 2014b; Suhartini, Heaven, and Banks 2014). The mixing and agitation processes may affect the even distribution of substrates and microorganisms, thus expanding the contact area of the substrate and microorganisms (Appels et al. 2011; Ohimain and Izah 2017). Chen et al. (2014a) reviewed various toxic materials (i.e. ammonia, chlorophenols, sulphide, halogenated aliphatics, long-chain fatty acids, heavy metals, and the emerging nanomaterials) that present in the feedstock can inhibit the AD process to various degrees. Those toxic materials have several effects on disrupting the activities of acetogens and methanogenic, delaying the formation of methane, reducing the methane concentration of biogas, and deteriorating the methanogenesis process. The most influential factors on biogas yield are pH and temperature because these factors significantly affect the growth and metabolism of certain microorganisms in the AD reactor (Appels et al. 2011; Taricska et al. 2009). For example, in acidogenesis and acetogenesis phases, microorganisms can grow optimally at pH < 6, while in the methanogenesis phase, the pH must be > 6. While the temperature may affect the microbial density, gas transfer rate, depositional properties, and solid residue from the substrate (Ohimain and Izah 2017). For instance, in the AD of OPEFBs, a study found that operating at thermophilic has superior performance than in mesophilic conditions. This was due to the improved efficacy of methanogens (i.e. Methanoculleus and Methanosarcina) (Walter et al. 2015).

Therefore, it can be seen from the above discussion that AD is a complex process and the performance of a feedstock is governed by several factors. Continuous monitoring is essential to facilitate effective biogas production.

3.2. Current practice of AD in Indonesia

Indonesia is one of the most populated countries, with increasing population and energy demand problems. With the current dependence on fossil fuel-based energy, Indonesia faces a huge challenge in innovating and developing new and renewable energy technology. Khalil et al. (2019) recommended transforming animal waste into biogas as a better option supported by an increase in the local livestock population, causing a large amount of animal organic waste available. Previously, Purwono et al. (2013) also stated that implementing a biogas digester is suitable for providing alternative energy in Indonesia's remote and marginal villages.

Several studies and practices of AD have been carried out in Indonesia, using various biomass feedstocks, including dairy manure (Usack, Wiratni, and Angenent 2014); animal waste (Khalil et al. 2019); sawdust (Ardhiansyah and Budiyono 2018); paper sludge (Priadi et al. 2014); landfill leachate (Kawai et al. 2012); POMEs (Halim et al. 2017; Harsono, Grundmann, and Soebronto 2014); fruit waste (Marendra et al. 2019; Sitorus and Panjaitan 2013); food waste (Suhartini, Lestari, and Nurika 2019; 2022b); microalgae (Salafudin et al. 2015); and macroalgae (Suhartini et al. 2022a). A previous study by Adinurani et al. (2013) found that Jatropha curcas Linn. capsule husk (JcL-CH) is a potential feedstock for producing biogas, either in a single- or two-stage AD system. Specifically, in the two-stage AD system, the methanogenesis reactor was equipped with a biofilm/bio-carrier made of palm fibre and glass wool. The study demonstrated that the two-stage AD system had superior performance to the single-stage AD, with 63.83% higher biogas production. It was also found that the methanogenesis reactor with glass wool biofilm was more effective than palm fibre. Their study also compared the quality and quantity of dry husk Jatropha curcas Linn (DH-JcL) in a single-stage AD system under batch and continuous conditions. The results showed that dry husk Jatropha curcas Linn in semi-continuous digester produced biogas with an average of 0.016 m³/kg DH-JcL and 0.01 m³/kg DH-JcL in the batch digester. The methane concentration in the semi-continuous digester was 83.15%, much higher than the batch digester of 68.61%. Another study by Priadi et al. (2014) reported that anaerobic co-digestion of paper sludge with cow manure has higher methane production (0.269 m³/kg VS) than mono-digestion of paper sludge (0.015 m³/kg VS).

Besides biomass, various studies have also reported the AD of industrial wastewater in Indonesia. For example, Kawai et al. (2012) demonstrated that AD of landfill leachate was an practical approach in transforming chemical oxygen demand (COD) into biogas, reducing COD concentration by up to 40%. Oktavitri et al. (2019) found that slaughterhouse wastewater can also be used as feedstock in the AD system with Chlorella vulgaris addition for biogas purification. By using this integrated AD system, COD concentration in the slaughterhouse wastewater was reduced by up to 63% and enhanced CO₂ removal in biogas by up to 7%. In addition, tofu-processing wastewater was also reported to have potential as an AD feedstock in Indonesia, either in continuous or batch AD systems (Darwin 2019). This study revealed that biogas produced from a continuous AD system was 0.123 m³/kg VS, which was much higher than the batch AD system (0.043 m³/kg VS). Such significant differences were possibly due to higher VS destruction found in continuous AD systems (76%) than in batch AD systems (57%). Additional benefits of implementing AD technology for treating tofu-processing wastewater were an increase in pH from 5 to neutral conditions (pH 6.6), following the industrial wastewater effluent discharge standard. A previous study by Syaichurrozi et al. (2016) reported that adding yeast (Saccharomyces cerevisiae) to the AD of tofu wastewater can increase biogas and methane production. Therefore, it can be concluded that currently, there is no established AD route for OPEFBs in Indonesia. However, there is a potential for it to be used in an industrial-scale process having similar biogas potential to the feedstock compared above.

3.3. Development in AD technology prospective for treating OPEFBs

Biogas and methane production has been developed and implemented globally via AD technology. AD reactors are commonly divided based on the production rates: low-rate and high-rate systems (De La Rubia, Villamil, and Mohedano 2019). A low-rate AD (LRAD) system is characterised by the presence of suspended biomass, requiring a long HRT and solids retention time (SRT). This system is recommended for media with high solids concentrations (i.e. mixed sewage sludge and organic fraction of municipal solid waste (OFMSW)). In such a system, the anaerobic conversion capacity is completely related to the growth rate of the microorganism. Because the growth rate is very low, a high reactor volume and continuous stirring are required to ensure homogeneity. The most suitable reactor in this type is a continuous stirred-tank reactor (CSTR) (De La Rubia, Villamil, and Mohedano 2019; Gerardi 2003). A high-rate AD (HRAD) system involves the retention of biomass, characterised by a relatively short HRT but a long SRT allowing a reduction in reactor volume. This reactor is usually used for municipal and industrial wastewater treatment. Biomass retention can be achieved by immobilising anaerobic sludge through granule reactors (i.e. up-flow anaerobic sludge blanket/UASB, expanded granular sludge bed/EGSB, and internal circulation reactor/IC), and biofilm (anaerobic filter/AF and fluidised bed reactors/FBR). This way, the biomass concentration increases, and the reactor can be operated at a high OLR without cleaning the biomass. Another way to increase the biomass concentration is to return/feedback concentrated sludge in a secondary purifier, similar to the activated sludge process (ASP), or by using physical or membrane filtration (i.e. anaerobic baffled reactor/ABR, anaerobic membrane bioreactor/AnMBR) (De La Rubia, Villamil, and Mohedano 2019; Fagbohungbe et al. 2015; Gerardi 2003; Van et al. 2020). Furthermore, according to Stamatelatou, Antonopoulou, and Michailides (2014), the selection of the AD reactor is mainly based on the total solid content (i.e. TS) and the organic content (i.e. volatile solid/VS) of the substrate feedstocks. A summary of the development in AD technology for biogas production is shown in Table 3. In the case of AD of OPEFBs, using a CSTR is recommended. In CSTR, agitation (or mixing) can homogenise nutrients, metabolites, and enzymes; thus, the contact surface area is faster and more expansive (Stamatelatou, Antonopoulou, and Michailides 2014). The CSTR also has a continuous and same rate of input feeding and effluent withdrawal to balance the microbial ability and population (Show, Yan, and Lee 2019). Other advantages of CSTR are uniformity in composition and temperature, the similar composition of inoculum and digestate (effluent), and stable operation (Foutch and Johannes 2003). The CSTR is also suitable for biomass with high solids concentration, such as OPEFBs. Therefore, this configuration is suitable for anaerobic mono- and co-digestion of OPEFBs with other organic waste (having TS > 20%). The system can also be integrated with pre-treatment facilities to enhance AD performance.

Table 3. Summary of prospective developed AD technology applied for biogas/methane production from OPEFBs.

System Rate	Parameters	Reactor type	Description	Advantages	Disadvantages
Low-rate	- TS is < 10% - Can be operated under single- and multi-stage AD system - Semi- and continuous feeding mode - High biogas production if high moisture feedstock - VS reduction is in the range of 50–70% - OLR is < 7 g VS m^3/day - High reactor volume requires higher heating - Not suitable for hydrophobic substrates - Digestate dewatering is highly required - Digestate has high nutrient content and less stable - Requires internal mixing equipment - Less pumping equipment is needed due to high moisture feedstock	Anaerobic lagoons (AL)	AL is an enclosed pool with a shallow depth and a large surface area. The temperature is not controlled and there is no agitation/mixing, thus the solids tend to accumulate on the bottom part of the pond.	- Low construction and operating costs - Simple operation and maintenance - Produce less digestate, hence lowering digestate handling and disposal cost - Saving energy - Effective for strong organic waste	- low efficiency and organic matter reduction - long retention time due to slow growth of anaerobic consortia - low biogas production - large area is needed - limitation on controlling the environmental condition
		Continuous stirred tank reactors (CSTR)	CSTR is a tubular reactor, in which the microorganisms are suspended in the digester through intermittent or continuous mixing.	- Has high efficiency due to continuous mixing to homogenises nutrient, metabolites and enzymes - Has good substrate–sludge contact, thus better mass transfer - Low potential for short-circuiting or clogging	- Requires high installation and operational high cost - High energy demand for mixing
		Plug-flow Digesters (PFD)	PFD is a tubular reactor which feeds raw materials from one end and moves in the opposite direction while being converted into biogas. The direction of flow can be vertical or horizontal. There is no agitation, thus the retention time should be set up long enough to get optimal results.	- Can be used to treat feedstock at high OLR and high TS concentration - Simple configuration and operation - Lower installation and operation cost, hence more economical than CSTR system - Lower energy demand - Low potential for short-circuiting or clogging - Has higher removal efficiency	- Potential of solid sedimentation at bottom of reactor - Difficult to move the feedstock from one side to other side of reactor - Lack in mixing, causing a low mass transfer - Has lower efficiency for treating low TS feedstock (< 10%) - Potential of scum-formation problems
High-rate	- TS is in the range of 10–40% - Can be operated under single- and multi-stage AD system - Batch, sequential batch, semi- and continuous feeding mode - low biogas production if low moisture feedstock - VS reduction is < 40% - Higher OLR, which is in the range of 7–15 g VS m^3/day - Small reactor volume requires lower heating - Suitable for hydrophobic substrates - Digestate dewatering is minimal - Digestate has less nutrient content and more stable - Mixing can be done via leachate and biogas recirculation, or via biogas and partial mixing - Advanced pumping equipment is needed due to low moisture feedstock	Anaerobic contact process (ACP)	ACP has similar shape to CSTR but there is an additional reactor (i.e. clier) for settling purposes. In the clier, there are two streams, include outflow (as effluent) and backflow (for biomass recycle to the main reactor). The separation in the clier is based on the gravity	- Can be used to treat concentrated wastewater with high concentration of suspended solids (1–2 kg BOD/m^3/day) - OLR is between 0.5–10 kg COD/m^3/day with HRT of 0.5–5 days - Has high efficiency in biodegradation of organic matter	- Requires post-treatment for meeting the standard of wastewater discharge - Long-term problems of inert materials accumulation causing reduction of active anaerobic microorganism - the high solids content may cause the separation process to run slowly
		Anaerobic filter (AF) (also known as fixed film digester or packed bed digester)	AF is an elongated containers filled with material that adheres to the container. The materials contain microorganisms that will provide high conversion rates at low HRT (<1 d). This reactor model can be modified in such a way both from the flow direction (i.e. up flow or downflow) and the flow rate (i.e. low or high).	- Can be used to treat any type of wastewater - Has high capability for biosolids retention - Adaptable for various type of filter media (synthetic or natural) - Ease of starting-up	- Potential of clogging if wastewater contain high solids - Problems of the stability of the filter material used - High installation cost - Longer time needed for microbial growth
		Anaerobic fluidised beds (AFB) or Anaerobic expanded bed system (AEBS)	AFB or AEBS are considered as the advanced AD technology, and it can be operated at OLR of > 40 kg COD/m^3/day. The digestion process occurred due to the presence of bacterial attached to mobile carrier particles (i.e. fine sand, plastic or pumice)	- The effluent can be circulated for influent dilution - Has good mass transfer - Less clogging and short circuiting due to large pore to bed expansion - Has high efficiency due to the high specific surface area of the carriers	- Difficult to achieve long-term stable operation - Difficult to control the formed biofilm, leading to operation problems
		Up-flow anaerobic sludge bed reactor (UASB)	UASB is an elongated container that contains microorganisms in the form of granules (placed as sludge blanket), thus easily settling. The bubbles created during biogas production may reduce the need of mixing. UASB is suitable for treating liquid substrates (or wastewater) with high amount of dissolved organic compounds.	- Use of granular or flocculent sludge blanket is reducing the needs of support medium - Can be feed using wide range of OLR - Has high COD removal efficiency - Can be used to treat high biomass content - Has short HRT and high SRT - Mixing is minimal - Saving energy	- Potential of granular sludge washout out/flotation/disintegration - Requires post-treatment for meeting the standard of wastewater discharge - Requires purification of biogas - Difficulties in start-up due to shock in temperature and OLR - Difficulties in controlling the bed expansion, hence limit the OLR applied
		Anaerobic baffled reactor (ABR)	ABR is a horizontal elongated rectangular container that is separated by several insulators (or compartment). Thus, the material can flow horizontally either on the surface or the bottom of the container. The goal is for the substrate to come into contact with the accumulated biomass at the bottom of the reactor.	- Reduce wash-out of bacteria - Has a simple design and, structure and operation - Has minimal sludge dewatering equipment - Can be used to treat low to high strength wastewater	- Bacteria has slow rate of movement

Sources: Bajpai (2017); Fagbohungbe et al. (2015); Gómez et al. (2019); Gould (2015); Stamatelatou, Antonopoulou, and Michailides (2014).

Furthermore, based on the configuration system, AD is divided into single-, two- and three-stage systems (Speece 2008; Taricska et al. 2009; Van et al. 2020). In a single-stage AD, all stages of the AD process occur in one reactor. In a two-stage, hydrolysis to acidogenesis stages is occurred in the first reactor, while acetogenesis and methanogenesis stages are in the second reactor. The advantages of a single-stage AD system include simple operation, easier maintenance, and low installation and maintenance costs. While the disadvantages include a low input rate, a longer retention time, a higher chance of explosion, difficulty in controlling pH, low stability, and a higher possibility of inhibitors accumulation (Aziz et al. 2020; Speece 2008; Taricska et al. 2009; Van et al. 2020). A two-stage AD system is suitable for feedstock material with high solids or lipid concentrations, has a more stable digestion process, a faster retention time, a high OLR, and a high efficacy for VS/COD reduction. The drawbacks of a two-stage AD system include the accumulation of hydrogen (may inhibit acidogenesis bacteria), separation of the reactor (may reduce the nutrients needed by methanogenic bacteria), high installation and construction cost for manufacturing of two interconnected reactors, and requirement of skilled personnel for its complex operation requires (Aslanzadeh, Rajendran, and Taherzadeh 2014; Aziz et al. 2020; Speece 2008; Taricska et al. 2009; Van et al. 2020). Three-stage AD system is aimed to separate acetogenesis and methanogenesis stages, thus increasing the hydrolysis rate of biomass feedstock and improving biogas/methane yield. However, this system has a more complex configuration and high cost; therefore, selecting reactor type, mode of feeding, and temperature for each stage is critically important (Van et al. 2020).

4. Sustainable strategies for improving the AD of OPEFBs

Integrating a green and sustainable approach in AD is essential for enhancing the biogas and methane production from OPEFBs. Due to the high lignin content in OPEFBs, strategies are required to increase biodegradability and speed up the degradation process. With the current movement toward biorefining and bioeconomy concepts, sustainable strategies for improving the efficacy of the AD process are essential. Hence, this paper highlighted various sustainable strategies potential to be applied prior to or within the AD of OPEFBs, including biological pre-treatment and co-digestion, as shown in Figure 5.

Figure 5. Sustainable strategies for improving the AD of OPEFBs, including the mechanism of each biological pre-treatment (Adapted from Abraham et al. (2020); Mishra et al. (2018); Mohd-Setapar, Abd-Talib, and Aziz (2012); Villa et al. (2020); Grimm and Wösten (2018); Wan Mahari et al. (2020); Pérez-Chávez, Mayer, and Albertó (2019); and Zou et al. (2020))

4.1. Biological pre-treatment of OPEFB for enhanced biogas/methane production

Various studies have emphasised the use of biological pre-treatment on OPEFBs as an alternative to chemical and thermal pre-treatment to increase further the efficacy of the AD process and the biogas and methane yields. These include fungi, microbes, enzymes, ensiling, and mushroom cultivation pre-treatment. Table 4 shows a comparison of these biological pre-treatments, while details on the examples of each pre-treatment are shown in Table 5.

Table 4. The comparison of each biological pre-treatment.

Biological pre-treatment	Operation cost	Process time	Performance	Toxicity
Fungal	Low cost and eco-friendly (Kainthola et al. 2021; Sulfahri et al., 2020); Vasco-Correa and Shah 2019; Wan and Li 2012)	• Up to 3-weeks (co-culture) (Asiegbu, Paterson, and Smith 1996) • 4–6-weeks (single-culture) (Asiegbu, Paterson, and Smith 1996; Ishola, Isroi, and Taherzadeh 2014; Isroi 2017)	• 8–16% lignin reduction (co-culture) and 4–14% lignin reduction (single-culture) (Asiegbu, Paterson, and Smith 1996); or up to 51% (Saha et al. 2016) • Hydrolysis efficiency: o Cellulose = 9.35–57.89% • Hemicellulose = 8.08–27.48% • Lignin = 10.6–12.5% (Ishola, Isroi, and Taherzadeh 2014; Montoya-Rosales et al. 2020; Suksong et al. 2020b) • 12–230% increase in biogas and methane production (Muthangya, Manoni Mshandete, and Kajumulo Kivaisi 2009; Tian, Fang, and Guo 2012) • Has low pre-treatment efficiency than enzymatic pre-treatment (Cai et al. 2021)	• No release of toxic compounds and less toxic effluent generation during the process (Rouches et al. 2016; van Beek et al., 2007) • Addition of inducer may inhibit of fungal growth and reduce biogas production (Kainthola et al. 2021)
Mushroom cultivation	Low-cost and eco-friendly (Aguilar-Rivera et al. 2017; Pérez-Chávez, Mayer, and Albertó 2019; Rosado, Kemmelmeier, and Da Costa 2002; Zadrazil 1997)	• 15–135 days depending on substrate and edible mushroom used (Pérez-Chávez, Mayer, and Albertó 2019) • 36 days for complete fruit body formation (Richard et al. 2020)	• 10–40.12% lignin reduction (Dimawarnita et al. 2022; Richard et al. 2020) • 25–35% cellulose and 8.5–28% hemicellulose reduction (Zou et al. 2020); or 49.56% hemicellulose reduction (Dimawarnita et al. 2022) • 17.7–180% biological efficiency (Aguilar-Rivera et al. 2017; Dimawarnita et al. 2022; Richard et al. 2020; Zou et al. 2020) • 4.3-fold increase in the release of reducing sugars (Mehta, Gupta, and Kaushal 1990) • 0.4- to 8-fold increase in biogas and methane production (Pérez-Chávez, Mayer, and Albertó 2019; Richard et al. 2020; Yunqin, Dehan, and Lishang 2010; Zou et al. 2020) • Production of multiple products: biogas, biofertiliser, and edible mushroom (Pérez-Chávez, Mayer, and Albertó 2019; Richard et al. 2020; Zou et al. 2020) • Economic profit of $0.592 from selling edible mushroom (Zou et al. 2020)	No release of toxic compounds and no generation of effluent (Fan et al. 2000; Sindhu, Binod, and Pandey 2016)
Microbial	Moderate-cost and eco-friendly (Cai et al. 2021)	12 h – 7 days incubation (Ali et al. 2020; Zhang et al. 2011)	• 2- to 3-fold increase in the release of reducing sugars (Wan and Li 2011) • More effective in degrading cellulose and hemicellulose compared to fungal pre-treatment (Banu et al. 2021) • Substrates degradation efficiency 56.35-69.2% (Ali et al. 2020) • 9.3–96.63% increase in methane yield (Ali et al. 2020; Poszytek et al. 2016; Yan et al. 2012; Zhang et al. 2011) • Has low pre-treatment efficiency than enzymatic pre-treatment (Cai et al. 2021)	• No release of toxic compounds (Cai et al. 2021; Mishra et al. 2018; Ravindran and Jaiswal 2016) • No release of pollutants (Cai et al. 2021)
Enzymatic	High cost and eco-friendly(Brémond et al. 2018; Carrere et al. 2016; Koupaie et al. 2019; Olatunji, Ahmed, and Ogunkunle 2021)	Fast process (i.e. within few hours) (Brémond et al. 2018; Schroyen et al. 2014; 2015)	• Up to 71.69% lignin reduction (Amin et al. 2010) • 24–280% increase in methane production (Brémond et al. 2018; Koupaie et al. 2019; Mahdy et al. 2015; Olatunji, Ahmed, and Ogunkunle 2021; Weide et al. 2020) • Release of phenolic and aromatic compounds (Schroyen et al. 2014; 2015) • Has higher pre-treatment efficiency than microbial and ensiling pre-treatment (Cai et al. 2021)	• No release of toxic compounds (Olatunji, Ahmed, and Ogunkunle 2021; Ravindran and Jaiswal 2016) • Potential release of effluent from the solvent used during the process (Mahdy et al. 2015; Schroyen et al. 2014; 2015) • No release of pollutants (Cai et al. 2021)
Ensilage	Low cost and eco-friendly (Cai et al. 2021; dos Santos et al. 2022)	10–90 days incubation (dos Santos et al. 2022; Janke et al. 2019; Wu et al. 2020)	• 8.70–18.16% lignin reduction, 4.58–23.73% hemicellulose reduction, and 2.50–24.96 cellulose improvement (Wu et al. 2020) • 9–119% increase in methane yield (dos Santos et al. 2022; Herrmann, Heiermann, and Idler 2011; Janke et al. 2019; Nolan et al. 2018; Vervaeren et al. 2010; Wu et al. 2020) • Has low pre-treatment efficiency than enzymatic pre-treatment (Cai et al. 2021)	• No release of toxic compounds (Cai et al. 2021; Mishra et al. 2018; Ravindran and Jaiswal 2016) • No release of pollutants (Cai et al. 2021)
Degree of Development (or Application)	TRL*	Advantages	Disadvantages	Refs
• Potential application at laboratory- and pilot-scale (Mustafa, Poulsen, and Sheng 2016; Rouches et al. 2019) • Full-scale application is difficult due to long incubation time and low lignin removal (Sayara and Sánchez 2019; Sindhu, Binod, and Pandey 2016)	3–6 (Dhouib et al. 2006; Mustafa, Poulsen, and Sheng 2016; Rouches et al. 2019; Tian, Fang, and Guo 2012)	- Different fungi species selectively degrading lignin and increase cellulose content for next step - Generates a highly delignified and cellulose-rich biomass - Other valuable compounds are released, such as aromatic compound from lignin degradation Low energy and chemicals requirement	- Need a longer incubation period - Considerable loss of holocellulose - Need sterile condition	(Agbor et al. 2011; Ishola, Isroi, and Taherzadeh 2014; Kainthola et al. 2021; Montoya-Rosales et al. 2020; Piñeros-Castro and Velásquez-Lozano 2014; Ravindran and Jaiswal 2016; Vasco-Correa and Shah 2019)
Industrial-scale (Pérez-Chávez, Mayer, and Albertó 2019; Zadrazil 1997)	7–9 (Pérez-Chávez, Mayer, and Albertó 2019; Zadrazil 1997)	- Degrading lignocellulosic material while producing high-value product (i.e. edible mushrooms) - Integrated process with mushroom cultivation and AD technology increased economic benefits of lignocellulosic material - Less energy consumption - Less chemicals used	- Requires additional nutrients such as S, N, O, and P. - Needs longer retention timeNeeds a controlled environmental condition to get the best results	(Agbor et al. 2011; Pérez-Chávez, Mayer, and Albertó 2019; Zou et al. 2020)
• Potential application at laboratory-scale (Cai et al. 2021; Wan and Li 2011) • Full-scale application is difficult due to longer incubation time, high cost of microbial culture, and slower process (Cai et al. 2021; Sayara and Sánchez 2019)	1–4 (Cai et al. 2021)	- Microorganisms degrade lignocellulosic material can be under aerobic or anaerobic conditions - Can be operated using pure or co-culture (consortia) microorganisms	- Using single species of microorganism has lower degradation performance - Need chemicals, additive, and nutrients - Need microorganism culture - Need sterile condition - Need storage	(Agbor et al. 2011; Baghbanzadeh et al. 2021; Bhutto et al. 2017; Ravindran and Jaiswal 2016)
Potential application at laboratory- and pilot-scale (Brémond et al. 2018) • Industrial application is hindered by expensive cost of enzyme and enzyme lifespan (Brémond et al. 2018; Carrere et al. 2016; Koupaie et al. 2019; Olatunji, Ahmed, and Ogunkunle 2021)	7–8 (Brémond et al. 2018)	- Results in high purity monomeric sugar yield (mainly glucose), - Has less impact on the environment - Needs mild reaction conditions	- High cost of enzyme - The technical and economic feasibility of the process has yet to be evaluated - High cost of additives - Higher energy demand - Needs advanced equipment for better resultsNeeds sterile condition	(Baghbanzadeh et al. 2021; Cai et al. 2021; Chiesa and Gnansounou 2014; Hamzah, Idris, and Shuan 2011; Ravindran and Jaiswal 2016; Sieborg et al. 2020)
Full-scale and industrial application (Borreani et al. 2018)	9 (Borreani et al. 2018)	- Much simple in operation - Lignocellulosic material can be degraded naturally or with the addition of microbial consortia - The technology has been widely used for energy crops, animal feed, or fertiliser - Can use anaerobic/aerobic microbial	- Requires additional nutrients such as S, N, O, and P. - Needs longer retention time - Needs a controlled environmental condition to get the best results - Need storage and additives	(Agbor et al. 2011; Cai et al. 2021; Milledge and Harvey 2016; Ravindran and Jaiswal 2016; Sieborg et al. 2020; Tanjore, Richard, and Marshall 2012)

Notes: *Technology Readiness Level (or TRL) classification are: 1 = basic, 2 = applied research, 3 = critical function or proof of concept established, 4 = laboratory testing or prototype validation, 5 = prototype system verified, 6 = pilot-scale demonstration, 7 = system incorporated in commercial design, 8 = system complete and qualified, 9 = system proven and full commercial deployment (Zabaniotou and Kamaterou 2019).

Table 5. Other studies on biological pre-treatment on OPEFB prior AD process.

Microorganism/Enzyme	Experimental Condition	Results	Refs.
Fungal pre-treatment
White-rot fungus Pleurotus floridanus	28 days incubation at 31°C	Digestibility of OPEFB's cellulose increased by 4-fold (from 4.66% up to 18.85%)	(Ishola et al. 2012)
White-rot fungus Pleurotus floridanus LIPIMC996	Addition of 200 µg Mn^2+/g OPEFB and 35 days incubation at 30°C	Lignin reduced up to 27.2%	(Harmini et al. 2013)
	Addition of 20 mM Nitrogen and 35 days incubation at 30°C	Lignin reduced up to 25%
White-rot fungi Phanerochaete chrysosporium	Four weeks incubation at 30°C	• P. chrysosporium degrades more polysaccharides than P. ostreatus • Lignin reduced up to 41%	(Piñeros-Castro and Velásquez-Lozano 2014)
White-rot fungi Pleurotus ostreatus	Four weeks incubation at 30°C	• Delignification without considerably affecting the cellulose fraction • Lignin reduced up to 50%
Aspernigillus niger USMAI1	Five days cultivation period	Fermentable sugar increased up to 140.6% (77 mg/g)	(Lim and Ibrahim 2013)
White-rot fungus Trametes versicolor TISTR 3224	Two weeks incubation at room temperature (28°C)	8.86% lignin and 5.68% cellulose degradation	(Kamcharoen et al. 2014)
White-rot fungus Trametes sp. BCC 8729		13.21% lignin and 20.85% cellulose degradation
White-rot fungus Phanerochaete chrysosporium ATCC 24725		18.79% lignin and 31.88% cellulose degradation
White-rot fungus Marasmius sp. BCC 9542		0.95% lignin and 3.14% cellulose degradation
White-rot fungus Xylaria sp. BCC 7749		1.15% lignin and 19.67% cellulose degradation
White-rot fungi of Pleurotus floridanus	28 days incubation	Biodegradation increased up to 95%	(Isroi 2017)
White rot fungi of Pycnoporus sanguineus	Four weeks incubation	Lignin reduced up to 90%	(Omar et al. 2017)
Oleaginous fungi Aspergillus tubingensis TSIP9	A. tubingensis TSIP9 was added after EFB, then soaked with 10% NaOH and boiled at 100°C for 15 min.	• Cellulose content increased from 40.13% up to 61.24%. • Lipid yield was 168.1 mg/g-EFB and cellulolytic enzyme was 16–20 U/g-EFB were generated as by-product.	(Intasit, Yeesang, and Cheirsilp 2019)
Schizophyllum commune ENN1	14 days incubation at 30°C	Sugar yield increased up to 230 mg/g and lignin reduced up to 54%	(Arbaain et al. 2019)
White-rot fungus Phanerochaete chrysosporium	14 days incubation	Increase in organic content. Further AD of treated OPEFBs resulted an increase in methane yield	(Hidayat et al. 2020)
Mushroom cultivation pre-treatment
Coprinus cinereus	The mushroom strain was cultivated on the mixed palm oil waste for 3 weeks. Afterwards, the spent mushroom substrate and the untreated palm oil waste were subjected to AD in AMPTS	• Lignin reduced by 23% • Methane yield was 0.287 m^3 CH4/ kg VS	(Temu et al. 2016)
Coprinus cinereus	Oil palm waste was cultivated for 21 and 17 days.	Methane yield increased from 0.33–0.34 to 0.43–0.49 m^3 CH4/kg VS.	(Mmanywa and Mshandete 2017)
n.a	Mushroom media medium from farmer was mixed by 20% cow dung for 65 days	• Reduction of lignin content by 2.67%, hemicellulose by 6.33% and cellulose by 6.55% • Methane yield was 40.69% in semi-wet fermentation and 38.71% in dry fermentation.	(Purnomo, Suprihatin, and Hasanudin 2018)
Straw mushroom (Volvariella volvacea)	The EFB was added inorganic and organic fertilisers after composted for 8 days. Afterwards, the EFB was subjected as mushroom medium for 4 days.	• Cellulose reduced from 24.12% to 23.51% • Hemicellulose reduced from 21.64% to 13.53% • Lignin reduced from 24.12% to 13.04%	(Triyono et al. 2019)
Oyster mushroom (Pleurotus spp.)	Oil palm mesocarp fibre (OPMF) was cultivated for 50 days	Lignin reduced to 10.63%	(Saidu et al. 2014)
Microbial pre-treatment
Effective Microorganism (EM)	In combination with chemical (phosphoric acid) and hydrothermal (160°C for 10 min).	Lignin and hemicellulose greatly reduced from ∼10% to 4.2% and ∼22% to ∼4.3%	(Shahriarinour et al. 2011)
Microorganism from genus of hydrogenotrophic (Methanoculleus and Methanobrevibacter) and genus of acetoclastic (Methanosarcina)	OPEFB was added to cattle slurry at 37°C and 55°C for 22 days to simulate SS-AD.	Methane yield increased from 89.9 to 94.7 mL CH4/g VS at 37°C and 72.1 to 211 mL CH4/g VS at 55°C	(Walter et al. 2015)
Bacillus sp.	n.a	Biogas production increased by 95%	(Salihu and Alam 2016)
Ganoderma boninense spores	OPEFB was incubated for 7 days under SS-AD.	Lignin reduced by ∼62%	(Jumali and Ismail 2017)
Clostridium acetobutylicum ATCC 824	50 g/ L OPEFB was added by 25 g/ L sugar and 15 FPU/g-substrate cellulose loading at 35°C.	OPEFB was slowly degraded but 94% fermentable sugar was completely metabolised	(Md Razali et al. 2018)
Bacillus subtillis	Co-digestion of POME and OPEFB augmented with 10% (v/v) B. subtillis loading under ambient temperature (27–30 °C) for 14 days.	Methane yield was 0.7 mL CH4/g VS	(Chin et al. 2019).
Enzymatic Pre-treatment
Lignin peroxides (LiP) enzyme	OPEFB to enzyme ratio was 1 (kg): 10,000 (μL), reaction time of 4 h and pH of 7.76	Lignin reduced by 71.69%	(Amin et al. 2010)
Manganese peroxidases (MnP) enzyme	OPEFB to enzyme ratio was 1 (kg): 10,000 (μL), reaction time of 5.76 h and pH of 6.	Lignin reduced by 67.94%
Laccase enzyme	5% (w/v) OPEFB was added 5–20 IU/g enzyme at pH 5 and 25 °C for 4 h with agitation 150 rpm.	Total sugar and weight loss increased from 2.4 to –3.5 mg/mL and 6.3 to 20.1%	(Ishmael et al. 2016)
Crude enzyme (CMCase and xylanase) from T. koningiopsis TM3	Addition of crude enzyme by 25 U/g substrate for 15 h.	Total sugar reduced by 29.2%	(Nutongkaew et al. 2019)
Enzyme mixtures (EMs) consists of cellulose, xylanase, and beta-glucanase	Five different dosage and formulation of EMs were added to agricultural wastes such as silage, straw and cattle manure for 60 days.	In 5 and 15 days, methane yield increased to 0.3% - 21.1% but gas yields were not detectable after 60 days in most cases	(Weide et al. 2020)

4.1.1 Fungal pre-treatment

In fungal pre-treatment, white-rot, brown-rot, and soft-fungi are mainly used to attack plant cell walls (i.e. lignin), composed of selective and non-selective lignin destruction, as explained by Kainthola et al. (2021). They added that lignin and hemicellulose are specifically degraded in the selective lignin destruction process, and only a small fraction of cellulose is degraded. In the non-selective lignin destruction process, all lignocellulosic materials are degraded simultaneously. For this purpose, the fungi produce ligninolytic enzymes such as lignin peroxidase (LiP), polyphenol oxidase, manganese peroxidase (MnP), and laccase (Lac) to degrade lignin. Brown-rot fungi mainly attack cellulose, while white and soft-fungi attack lignin and cellulose (Agbor et al. 2011). Thus, the fungal pre-treatment can degrade lignocellulosic material to fermentable sugars using their produced lignocellulolytic enzymes (Intasit, Yeesang, and Cheirsilp 2019; Zou et al. 2020). Factors influencing the fungal pre-treatment include substrate composition, fungal strain, incubation period, moisture content, temperature, pH, nutrient, oxygen concentration, and cultivation condition (Kainthola et al. 2021). Longer incubation time needed in fungal pre-treatment becomes the major constraint for full-scale application (Montoya-Rosales et al. 2020). Furthermore, fungal pre-treatment can be carried out as a single- or co-culture fungal strain, as shown in Figure 5.

Many studies have highlighted the successful application of single-culture fungal pre-treatment to alter the chemical composition and improve biodegradability. For example, Ishola, Isroi, and Taherzadeh (2014) reported that treating OPEFBs with white-rot fungus Pleurotus floridanus LIPIMC996 for four-week improved the biodegradation rate of OPEFBs by 4.5-fold. They further stated that after fungal pre-treatment, an increase of hemicellulose (27.48%) and a decrease in cellulose (34.34%) were observed. This indicated that Pleurotus floridanus LIPIMC996 could degrade the lignin content in OPEFBs. The results of Fourier-transform infrared spectroscopy (FTIR) analysis also confirmed their findings, indicating a reduction in the cellulose hydrogen bond of the fungal-treated OPEFBs. After fungal pre-treatment, the degradation of lignin in OPEFBs showed a positive impact on increasing bioenergy production (i.e. bioethanol). Kamcharoen et al. (2014) studied various white-rot fungi (i.e. Trametes versicolor TISTR 3224, Trametes sp. BCC 8729, Phanerochaete chrysosporium ATCC 24725, Marasmius sp. BCC 9542, and Xylaria sp. BCC 7749) for pre-treatment of OPEFBs. This study also indicated that fungal–treated OPEFBs have higher biodegradability than untreated OPEFBs, mainly due to lignin degradation. Suksong et al. (2020b) also found that pre-treatment of OPEFBs using single-culture Trichoderma reesei TISTR 3080 and Pleurotus ostreatus DSM 11191 increased methane production by 52% and 44%, respectively. This was parallel to higher hydrolysis efficiency in cellulose, hemicellulose, and lignin by 10.9%, 22.2%, and 10.6% (for T. reesei) or by 9.35%, 8.08%, and 12.5% (for P. ostreatus). The fungal metabolism, i.e. enzyme activity, was beneficial for breakdown and reducing the sugar content ranging from 33.8–43.5 mg/g. Their study also confirmed the potential benefit of fungal pre-treatment: the energy recovery of pre-treated OPEFBs accounted for 200% higher than untreated OPEFBs.

A case study in Indonesia by Isroi (2017) researched OPEFBs pre-treatment using white-rot fungi of Pleurotus floridanus and demonstrated that fungal pre-treatment altered the physical structure of OPEFBs to be more ‘breakable’. After pre-treatment, more lignin was degraded compared to cellulose or hemicellulose. This condition was mainly due to the ligninolytic enzyme production by Pleurotus floridanus, which contributes to the degradation of hemicellulose and lignin. The research also suggested that fungal pre-treatment of OPEFBs could enhance biodegradability, with the value of 95% over 28-day incubation, indicated by higher fermentable sugar production. Omar et al. (2017) also reported that using white-rot fungi (i.e. Pycnoporus sanguineus) to pre-treat OPEFBs resulted in greater damage to the lignin structure and a decrease in lignin content. The ratio of OPEFBs:fungi (60:40) generated the best degradation performance within a 4-week incubation period, indicated by 90% lignin removal. In another research by Intasit, Yeesang, and Cheirsilp (2019), oleaginous fungi Aspergillus tubingensis TSIP9 with cellulolytic enzymes were effective as a potent microorganism for solid-state fermentation with increased cellulose content after the treatment. Another recent study reported that pre-treatment of OPEFBs using an indigenous fungus (i.e. Schizophyllum commune ENN1) increased sugar yield (230 mg/g) and lignin removal (54%) (Arbaain et al. 2019). Our previous study also reported that fungal pre-treatment on OPEFBs using Phanerocheate chrysosporium improved organic matter and methane yield (Hidayat et al. 2020).

Besides single-culture fungi, co-culture fungi can also be used in biological pre-treatment. For instance, a previous study by Asiegbu, Paterson, and Smith (1996) reported that using co-culture fungal strains (i.e. Pleurotus sajor-caju, Phanerochaete chrysosporium, and Trametes versicolor) reduce higher lignin content (8–16%) within shorter time (3-weeks incubation) than single-culture (4–14% lignin removal for 6-weeks incubation). van Heerden et al. (2011) also found that P. sanguineus/A. flavipes co-culture could alter the chemical composition (i.e. cellulose, lignin, and hemicellulose) of woody biomass. Mohanan et al. (2014) compared the use of single- and co-culture fungal strains (from genus Trichoderma, Aspergillus, Pycnoporus, and an unidentified strain of F113) for pre-treatment of kitchen waste. This study found that the co-culture of Trichoderma sp. with other fungal strains has higher degradation of lignocellulosic material than the respective single-culture. Furthermore, Ming et al. (2019) studied the effect of co-culture fungal pre-treatment (using Aspergillus niger CBS 110.42 and CBS 554.65 and Phanerochaete chrysosporium CBS 246.84) on agricultural waste. The study reported that these co-culture fungal strains enhanced delignification by increasing the cellulose and xylan degrading activity (i.e. two-fold improvement). Saini and Sharma (2021) also agreed that using fungal co-culture significantly reduced the pre-treatment time. Therefore, it can be highlighted that co-culture fungal pre-treatment offers higher efficiency of delignification and reduces the incubation time required. Hence, co-culture fungal pre-treatment has strong potential to be further implemented to support a more effective and efficient bioenergy production. Sperandio and Ferreira Filho (2019) emphasised that implementing fungal co-culture may expand the biorefineries of lignocellulosic biomass for bioenergy and bio-based product generation.

Another potential route to enhance fungal pre-treatment is to combine it with other available pre-treatment. For example, Hidayatullah et al. (2021) studied a combination pre-treatment on OPEFBs using a single-culture fungal strain (Marasmius sp and Phanerochaeta chrysosporium) with hydrothermal. The study suggests that using Marasmius sp. (30-day incubation time) followed by hydrothermal pre-treatment (160°C, 15 min) has better delignification results. This combination produces higher lignin degradation (26.67%) with higher xylose (166 mg/g OPEFB) and glucose (283 g/g OPEFB) production than using Phanerochaeta chrysosporium. Kannaiyan et al. (2017) also found that a combination of hydrothermal and fungi pre-treatment increased deliginification rate, lignin degradation (i.e. by twofold), and reduced residence time. Castano, Crespo, and Torres (2019) stated that there is an interaction effect between incubation time and temperature on the delignification process of lignocellulosic biomass. Higher temperatures may promote higher degradation of complex sugars (or polysaccharide) (Lu et al. 2018). Thereby, it explains the higher performance of combining biological with thermal pre-treatment. Another example is combining fungal with dilute acid pre-treatment, which can enhance biogas production (Saini and Sharma 2021). A combination of two biological pre-treatments, such as anaerobic ruminal fungi with fermentative bacteria, increased methane yield from the AD of wheat straw and mushroom spent raw by 3- and 3.5-fold, respectively (Ferraro et al. 2018). In addition, using white-rot fungi combined with chemical or physical pre-treatment was effective for delignification of straw biomass (Ghaffar, Fan, and McVicar 2015). Combining physical pre-treatment (i.e particle size reduction with grinding) with fungal pre-treatment may provide better lignin bond disruption, thus producing a higher hydrolyzed product yield (Wagle et al. 2022).

Tian, Fang, and Guo (2012) stated that a combination of fungal pre-treatment with chemical pre-treatments (i.e. ionic liquids, supercritical CO₂, organic solvents, and diluted acids) could enhance the transformation of lignocellulosic biomass into bioenergy and bioproducts. With this combination, problems occurred during fungal pre-treatment can be resolved. Thus, these findings confirmed that combination pre-treatments might offer effective ways prior to the AD route. Such an approach could completely breakdown the lignocellulosic materials in biomass feedstock, hence higher biogas and methane production.

4.1.2. Mushroom cultivation pre-treatment

In mushroom cultivation pre-treatment, the edible mushroom is grown in lignocellulosic biomass to promote degradation by mycelium penetration (to increase the pore size) and by disrupting functional linkages between polysaccharides, cellulose, and lignin (Pérez-Chávez, Mayer, and Albertó 2019; Zou et al. 2020). This breakdown is occurred due to the presence of lignocellulolytic enzymes produced by the edible mushrooms (Zou et al. 2020). The edible mushrooms (i.e. Agaricus bisporus, Pleurotus ostreatus/oyster mushroom, Lentinula edodes) are known to be effective in transforming lignin and cellulose into simple sugars and small compounds, further used as food or carbon source for the mushrooms (Albertó 2017; Pérez-Chávez, Mayer, and Albertó 2019; Zou et al. 2020). In detail, mushroom cultivation pre-treatment is shown in Figure 5. Moreover, edible mushrooms can also be grown on compost or during the composting process, converting complex organic carbon into simpler forms and releasing many aliphatic compounds (O/N alkyl) (Triyono et al. 2019). One of the aliphatic compounds (i.e. humic acid) increases the ability to degrade lignin structure (Amir, Hophmayer-Tokich, and Kurnani 2016). Therefore, being fibrous in nature, a common practice found in the literature is using OPEFBs for mushroom cultivation. This practice is also seen as a different type of mushroom pre-treatment as it employs various mushroom species before utilising the biomass for further valorising routes, as shown in Table 5.

For instance, OPEFBs were potential as a substrate for oyster mushroom cultivation because they can stimulate the mushrooms’ growth (Sudirman et al. 2011). Mmanywa and Mshandete (2017) reported that treating oil palm waste with Coprinus cinereus resulted in higher methane production (0.43–0.49 m³ CH₄/kg VS) than untreated oil palm waste (0.33 −0.34 m³ CH₄/kg VS). These findings indicate that such unique approaches can provide food (i.e. edible and medicinal mushrooms) and fuel (i.e. biogas), thus consequently reducing the environmental problems created by oil palm waste. Purnomo, Suprihatin, and Hasanudin (2018) also adopted a similar approach to treating OPEFBs as mushroom cultivation media before feeding the biomass to the AD system. Their results also suggested that cultivating mushrooms can reduce lignin content by 2.67%, hemicellulose by 6.33%, and cellulose by 6.55%. Such an approach altered the lignocellulosic structure of OPEFBs, thus helping the microorganism consortia in the digester to easily degrade the organic matter into biogas or methane. After being treated as mushroom media, the study also reported a high biogas production from OPEFBs. This result agrees with a study by Triyono et al. (2019), who also stated that using OPEFBs as a growing media for mushroom cultivation can be used as an alternative approach to enhance lignin degradation and add more economic benefits through mushroom production. A study by Saidu et al. (2014) also demonstrated that oil palm mesocarp fibre could be used for oyster mushroom cultivation. Thereby, enhancing the degradation of complex material (i.e. lignin) and exposing more easily degradable organic material to microorganism consortia. Further digestion of treated oil palm mesocarp fibre produced higher biogas production than untreated biomass samples. Such finding also indicates that biological pre-treatment using mushrooms may increase the AD system's efficacy should the optimum conditions are achieved. Mamimin et al. (2021) studied straw mushroom (Volvariella volvacea) cultivation as a strategy to enhance biodegradability and methane production from OPEFBs. Similar to other studies, their results showed that mushroom cultivation pre-treatment improved biodegradability by twofold, with an increase in methane yield by 4.6-fold compared to untreated OPEFBs. The visual morphology of pre-treated OPEFBs also demonstrated that the biomass surface was fractured by the mycelia, allowing accessibility for microorganisms or enzyme to break down cellulose and hemicellulose of OPEFBs.

Several studies have also reported the combination of mushroom cultivation pre-treatments for better delignification results. Chen et al. (2022a) reported that Shiitake cultivation with an additional low concentration of nitrogen, following physical pre-treatment (i.e. particle size reduction), increased the delignification of wood biomass by 41.5%. Temu et al. (2016) studied the combined pre-treatment of mixed oil palm waste prior AD process. After physical pre-treatment (i.e. particle size reduction), biological pre-treatment using Coprinus cinereus (also known as edible and medicinal mushrooms) was employed. The results showed a decrease lignin content of the waste by 23%. FTIR analysis results further indicated this by the loss of CH₂ vibration, along with the removal of cellulose and hemicellulose. Their findings also showed that the methane yield increased to 0.287 m³ CH₄/ kg VS. There has been very limited information available on this strategy; hence further in-depth studies are essential.

4.1.3 Microbial pre-treatment

The microbial pre-treatment can be carried out using a single- and co-culture (i.e. microbial consortia) to degrade lignocellulosic material. However, a study by Baghbanzadeh et al. (2021) reported high efficacy of microbial pre-treatment using mixed culture than single species, as it can reduce metabolite repression and feedback regulation. Thus, various microbial species can simultaneously degrade lignin, cellulose, and hemicellulose under anaerobic or aerobic condition (Ravindran and Jaiswal 2016). The microbal pre-treatment is illustrated in Figure 5. Furthermore, in the existence of oxygen, facultative anaerobic and aerobic microorganisms can degrade lignin polymers. Microbial pre-treatment is more economical than enzyme pre-treatment but slower (Baghbanzadeh et al. 2021); and needs a large sterile area (Abraham et al. 2020). Abraham et al. (2020) stated the advantages of microbial pre-treatment, including low-energy consumption, no formation of inhibitory compounds, lignin, and hemicellulose hydrolysis, and cellulose structure alteration. Various studies also reported different microbes for pre-treating the OPEFB biomass, as shown in Table 5.

Several studies reported using a single-culture of microorganism as pre-treatment option to enhance the biodegradation of OPEFBs. For instance, Jumali and Ismail (2017) studied the microbial pre-treatment of OPEFBs biomass using Ganoderma boninense spores under a solid-state fermentation system. The study revealed that OPEFBs have higher lignin degradation after being treated with Ganoderma boninense spores for 7-day incubation, with ∼62% lignin removal. While only ∼5.1% lignin degradation was observed in untreated OPEFBs. This finding confirmed that increasing the number of microbes used and the incubation period can increase lignin removal. Thus, the microbial pre-treatment using Ganoderma boninense spores can be used as an alternative to enhance the bioconversion of OPEFBs biomass into fermentable sugars, resulting in more routes for valorisation. While Salihu and Alam (2016) reported that pre-treatment of OPEFBs using Bacillus sp. enhanced biogas production by 95%. A study by Nurika et al. (2022) also found that using a single culture of lignin-degrading bacteria (i.e. Comamonas testosteroni, Agrobacterium sp., Lysinibacillus sphaericus, and Paenibacillus sp) increase delignification in the range of 12.24% - 25.84%. This is also followed by improved methane production from 0.005 m³/kgVS (untreated OPEFBs) to 0.042 m³/kgVS (C. test), 0.016 m³/kgVS (Agrobacterium sp.), 0.021 m³/kgVS (L. sphaericus), and 0.029 m³/kgVS (Paenibacillus sp), respectively.

The biological pre-treatment was also researched towards adding co-culture microorganisms to improve biodegradation and biogas or methane production. Shahriarinour et al. (2011) stated that biological pre-treatment of OPEFBs with effective microorganisms (EM) significantly enhanced biodegradation due to increase in cellulose production by 13%. Another study found that pre-treatment using a bio-augmentation of Bacillus subtilis and mixed methanogens in varying ratios was highly effective and increased methane yield from the anaerobic mono-digestion of OPEFBs (Chin et al. 2020). They reported that Bacillus subtilis, a major industrial microorganism, can effectively degrade cellulose and adapt to aerobic and anaerobic conditions with a low nutrient requirement, fast-growth rate and industrially safe. Walter et al. (2015) found that pre-treatment of OPEFB using a combination of microorganisms from a genus of hydrogenotrophic (i.e. Methanoculleus and Methanobrevibacter) and genus of acetoclastic (i.e. Methanosarcina) can produce much higher biomethane. Shah et al. (2019) showed that using mixed cultures of Bacillus sp. to pre-treat rice straw increased the delignification rate (indicated by a reduced lignin content), further increasing methane production by 76%. Yan et al. (2012) reported that using mixed cultures of lignocellulolytic microorganisms as pre-treatment prior to the AD process improved methane production by 20%. This was possibly due to the higher efficacy of delignification, resulting in a higher breakdown of lignin, hemicellulose, and cellulose content. Ali, Abomohra, and Sun (2017) also found that co-culture microorganism pre-treatment improved the delignification of sawdust, resulting in more production of biogas (24.5%) and methane (72.6%), respectively. These findings highlighted the application of co-culture microorganisms as a pre-treatment option that offer a more efficient and greener route for better valorisation of lignocellulosic biomass. Another study of co-culture microbial pretreatment was also done by Suksong et al. (2019a), which resulted in the biodegradation efficiency of 62% (i.e. Clostridiaceae rich consortium) and 82% (i.e. Lachnospiraceae rich consortium), and produced maximum methane yields of 0.252 and 0.349 m³ CH₄/kg VS, respectively.

A combination of microbial pre-treatment with other green pre-treatment options has also been investigated. Md Razali et al. (2018) studied pre-treatment of OPEFBs using Clostridium acetobutylicum ATCC 824 with the addition of cellulase enzyme. The study revealed that OPEFBs were slowly degraded by Clostridium acetobutylicum ATCC 824. The addition of cellulase enzyme enhanced the delignification, indicated by increased production of fermentable sugars. Shahriarinour et al. (2011) stated that pre-treatment of OPEFBs with effective microorganisms (EM), in combination with chemical and hydrothermal pre-treatment, significantly reduced lignin content from ∼10% to 4.2%, thus improving the efficacy of biodegradation. Zhong et al. (2016) also found that combining co-culture microbial with fungi pre-treatment trigger a greater synergistic effect, resulting in higher biodegradation of lignocellulosic materials in wheat straw. This is parallel to an increase in biogas and methane production by 39.24% and 80.34%, respectively. They suggest that microbial-fungi pre-treatment offered a more complete degradation of lignocellulosic materials, resulting in a greater accumulation of easily degraded organic compounds. They claimed that combining biological pre-treatment could shorten the initial digestion period in the AD system and fasten the biogas and methane production. These findings suggest that a combination of microbial pre-treatment with other alternative pre-treatment may offer better delignification of OPEFBs, leading to higher efficacy in the AD process.

4.1.4. Enzymatic pre-treatment

Enzymatic pre-treatment has been highlighted as an effective pre-treatment to enhance the biodegradability of lignocellulosic biomass (Agabo-García et al. 2019); which employ a pure (or single) enzyme (Ravindran and Jaiswal 2016); and enzyme cocktail (Lopes, Ferreira Filho, and Moreira 2018). In detail, the enzymatic pre-treatment is illustrated in Figure 5. Various enzymes (i.e. lipases, glucanases, and proteases) are widely used in the pre-treatment of lignocellulosic biomass. However, other enzymes (i.e. Lac, MnP, and versatile peroxide/VP) oxidise specific components of the lignin structure in lignocellulosic biomass (Ravindran and Jaiswal 2016). For instance, MnP oxidises the phenolic structure of lignin, LiP targets non-phenolic components, and VP can oxidise phenolic and non-phenolic structures (Falade et al. 2017). MnP requires the presence of a chelating agent to stabilise Mn (III) and oxidise more reactive phenolic structures (Hofrichter 2002). Furthermore, removing phenol could enhance microbial growth, improving the fermentation process's efficacy and reducing time in the lag phase. The factor influencing enzymatic pre-treatment include biomass feedstock's source and characteristics (Agabo-García et al. 2019). Enzymatic pre-treatment also depends on the enzyme ratio, reaction time, temperature, and pH of the solution. The lignin degradation depends on these parameters, which must be optimised for desirable products (Amin et al. 2010). The main advantages of the enzymatic pre-treatment include high pure monomer sugar yields (mainly glucose), lower environmental impact, and mild reaction conditions compared to the chemical acid hydrolysis routes (Hamzah, Idris, and Shuan 2011); low-energy demand fast process, cellulose structure alteration, and partial hydrolysis of hemicellulose (Abraham et al. 2020). Furthermore, enzyme is selective to delignify the lignin content much faster than microbial pre-treatment (Rizal et al. 2018). However, the techno-economic feasibility of the process still needs to be assessed and verified if intended for the scale-up (Chiesa and Gnansounou 2014); due to the high cost of enzymes (Abraham et al. 2020).

Many studies have reported the positive impact of enzymatic pre-treatment (using a single enzyme) to enhance the biodegradability of OPEFBs. The reported enzymes on OPEFBs include a combination of Accellerase, Multifect Xylanase, Multifect Pectinase, cellulase, β 1–4 glucosidase, etc. (Ariffin et al. 2008; Hamzah, Idris, and Shuan 2011; Lau et al. 2010). A previous study by Amin et al. (2010) demonstrated that pre-treatment of OPEFBs using LiP and MnP enzymes reduced lignin by 71.69% and 67.94%, respectively. Another study comparing enzymatic hydrolysis and ammonia fibre expansion methods on OPEFBs found that, at the optimum conditions, the enzymatic pre-treatment could achieve 90% of the total cellulosic ethanol yield within the 72-hours of the pre-treatment (Lau et al. 2010). Ishmael et al. (2016) studied enzymatic pre-treatment on OPEFBs using the laccase enzyme. Their study reported that increasing laccase enzyme concentration from 5 to 20 IU/g has increased the delignification, indicated by an increase in total sugar from 2.4 to 3.5 mg/mL and in weight loss from 6.3% to 20.1%, respectively. This study also demonstrated that various factors might influence the efficacy of enzymatic pre-treatment of OPEFBs, including substrate's concentration, substrate's particle size, pH, incubation time, and temperature. In this study, for example, at a pH of 5, the laccase enzyme achieved the optimum lignin removal. Therefore, it can be said that using the laccase enzyme can be a safe option for pre-treatment of OPEFBs prior to AD or other potential bioenergy utilisation. Another study by Deb et al. (2019) reported that using cellulase enzyme in treating oil palm mixed waste and POME increased the hydrolysis rate and damaged the POME structure. Thus, it is easily accessible for degradation, leading to an increase in the overall production of sugars. A similar study by Weide et al. (2020) has also reported that adding cellulase enzyme to the AD process improved methane yields of agricultural wastes (i.e. silage, straw, and cattle manure), ranging from 0.3% to 21.1%.

The positive impact of cocktail enzymatic pre-treatment on improving biogas and methane yield from lignocellulosic biomass has also been highlighted. For instance, Weide et al. (2020) studied enzymatic pre-treatment on a mixture of cattle manure and maize silage (CMM) using two cocktail enzymes, including EM4 (cellulase + xylanase + beta-glucanase) and EM5 (xylanase + glucosidase + endo-pectinase). Their study found that using these cocktails enzyme improve methane yield from 0.559 m³/kgTS to 0.582 m³/kgTS (CMM + EM4) and 0.599 m³/kgTS (CMM + EM5), or by 11.0% and 6.8%, respectively. Another study reported that applying cocktail enzyme (i.e. endoglucanase + xylanase + pectinase) pre-treatment on sugar beet pulp silage produced 27.9% higher biogas production (Ziemiński and Kowalska-Wentel 2015). Treating spent hops with cocktail enzyme (i.e. endoglucanase + xylanase + pectinase) pre-treatment increased biogas production by 13% (Ziemiński, Romanowska, and Kowalska 2012). A previous study also showed that enzymatic pre-treatment on sorghum forage using a cocktail enzyme (i.e. endoglucanase + exoglucanase + xylanase) resulted in 15% higher methane yield (Rollini et al. 2014). Using cocktail enzyme (i.e. MnP + versatile peroxidase) could improve methane production from corn stover by 17% (Schroyen et al. 2014); while using a cocktail enzyme (i.e. cellulase + cellobiase) increase methane yield from switch grass by 38.9% (El-Mashad 2015). Improvement in biogas and methane production was because cocktail enzyme completely degraded the lignocellulosic materials, as their component requires different enzymes to degrade (Lopes, Ferreira Filho, and Moreira 2018). For example, cellulose compound in the cell wall of lignocellulosic materials requires cellulase enzyme to degrade and often need endoglucanases to hydrolyze into cellobiose, which further hydrolyzes with β-glucosidase to release glucose. Lignin needs enzymes (such as laccase) to be oxidised into phenolic and aromatic compounds. Furthermore, to improve the performance of enzymatic pre-treatment, applying a cocktail enzyme could be proposed to enhance the AD of OPEFBs. Yet, limited information is available, further studies need to be investigated for pre-treating OPEFBs since.

Improving the performance of the AD system using a combination of enzymatic with other pre-treatments has been highlighted in various studies. Ahmad Rizal et al. (2018), for example, studied the combination of physical (i.e. particle size reduction), thermal (i.e. superheated steam), and enzymatic (i.e. addition of laccase enzyme) pre-treatment significantly reduced lignin content of OPEFBs by 38.7%. This combination also resulted in a higher glucose yields (increase by 4.6-fold), which may enhance biogas and methane yield if used as feedstock in AD. Lau et al. (2010) studied ammonia fibre expansion (AFEX) and physical pre-treatment (i.e. particle size reduction) combined with enzymatic pre-treatment (i.e. cocktail enzyme of Accellerase, Multifect xylanase, and Multifect pectinase). Their study showed that, the combination pre-treatment, the hydrolysis yield increased by 90% over 7 h and the recalcitrance compounds could be reduced without damaging the fermentable materials. Rollini et al. (2014) also found that combining alkaline and cocktail enzymatic pre-treatment on sorghum forage improved methane yield by 37%, much higher than using cocktail enzymatic pre-treatment alone (15%). Combination enzymatic with other green pre-treatments needs to be explored further for OPEFBs.

4.1.5. Ensilaging pre-treatment

The concept of ensiling is not new but has been practiced by farmers and experimented with by the academic community on crop digestibility since the late 1970s (Coblentz and Akins 2018). Ensilaging technology is a conventional technology widely used to store animal feed (Ravindran and Jaiswal 2016; Tanjore, Richard, and Marshall 2012). The ensiling process is essentially simple yet can become complex due to numerous interactions among the biomass and the microbial species involved. The primary desired product from ensiling is the conversion of the biomass sugars into lactic acid in an anaerobic fermentation setup. Nevertheless, complications arise due to the aerobic conditions at the start and end of the process. Furthermore, simple sugars are not the only substrates available in the biomass that either get converted or become inhibitory to the ensiling bacteria (Mohd-Setapar, Abd-Talib, and Aziz 2012; Villa et al. 2020). They stated that the different phases of ensiling are classified mainly into six: phase 1 – initial aerobic phase, phase 2 – anaerobic fermentation phase, phase 3 – bring phase 2 to an end, phase 4 – lactic acid formation, phase 5 – material storage, and phase 6 – aerobic decomposition. In details, the ensiling process is illustrated in Figure 5. Typically used as a method for forage conservation, the investigation is proceeding on ensiling by researchers/academics or industry on diversified crops around the world. Ensiling offers the best means for preserving the nutritive value of given biomass from the point of harvest to storage until further valorisation (Muck 2012). Ravindran and Jaiswal (2016) reported that naturally occurring bacteria generate enzymes that degrade the lignocellulosic material into fermentable sugar suitable for animal feedstock during ensiling. Furthermore, they found that Lactobacillaceae is a dominant species in the microbial consortium during ensiling. Factors influencing ensiling pre-treatment include substrate composition, harvest time, additives, solid content, ensiling duration, moisture content, pH, and loss of dry matter (DM) (Sieborg et al. 2020). The study reported that the amount of butyric acid present during the incubation period required much attention, as indicated by the metabolic activity and significant losses to DM.

As discussed earlier, OPEFBs have high water retaining capacity, therefore, they can be degraded naturally using ensiling. Carrere et al. (2016) claimed that ensiling had been widely used to store energy crops; however, as AD pre-treatment, the ensiling advantages include low energy demand and scalability. Several studies, as shown in Table 5, have also reported the positive effects of ensiling on improving the digestability and enhancing the biogas or methane potential from various biomass feedstocks such as sorghum (Sambusiti et al. 2013a; 2013b); wheat straw (Rincón, Banks, and Heaven 2010; 2012); corn straw (Zhong et al. 2011); corn stalk (Menardo et al. 2015; Zhang et al. 2018; Zhao et al. 2016); grass (Abu-Dahrieh et al. 2011; Quiñones et al. 2011); maize (Evranos and Demirel 2015; Oliva-Merencio et al. 2015); tall fescue (Feng et al. 2018); and sugar beet (Ahmed and Kazda 2017). For instance, Mayulu (2014) studied that ensiling the oil palm industry residues increased their digestibility due to a reduction in crude fibre, as well as an increase in crude protein and nutrient content.

Several studies have emphasised the minimal effect of ensiling pre-treatment alone in improving AD performance. Therefore, combining different options is advisable for practical and commercial applications. A study by Vervaeren et al. (2010), for instance, highlighted that ensiling of maize substrate with the addition of biological additives (i.e. homo- and hetero-fermentative strains, yeasts, and enzymes) significantly improved the ensilage quality and slightly enhanced the biogas and methane production in the range of 10.1% to 14.7%. They further added that adding biological additives during ensiling may also improve the hydrolysis stage in the AD process (i.e. degradation of complex organic materials). Additional research has been carried out to ensure feasible application of ensiling pre-treatment. Yunqin, Dehan, and Lishang (2010) investigated that pulp and paper sludge pre-treated with combining mushroom cultivated combined enzymatic pre-treatment (e.g. laccase enzyme) increased delignification by 34.2%. While (Li et al. 2019a) reported that combining ensiling with enzymatic pre-treatment (i.e. using Acremonium cellulase) or with microbial pre-treatment (i.e. using L. plantarum A1) alone improved the delignification of corn stalk. However, a combination of ensiling with Acremonium cellulase and L. plantarum A1 performed has superior performance in the degradation of lignocellulose. Hence, this combination can be proposed as a suitable pre-treatment option for bioenergy production. Yet, there are limited studies and data available for OPEFBs. Thus, further investigation in this area is essential.

4.2. Co-digestion of OPEFBs

Co-digestion combines two or more substrates during the AD process as an alternative to increasing biogas production when mono-digestion is challenging to achieve maximum methane yield (Aziz et al. 2020). Various studies have emphasised that co-digestion is one of the sustainable approaches to enhancing biogas/methane yield and tackling waste problems. The addition of co-digestion substrates in the AD system could enhance the microbial’ presence and activity, add nutrient supply (that the main biomass is lacking), improve nutrient balance, improve potential detoxification from toxic substances, and provide additional carbon sources (i.e. readily biodegradable substances) (Chen et al. 2016; Jaman et al. 2022; Jin et al. 2018; Panichnumsin et al. 2010; Pastor et al. 2013; Suhartini et al. 2022a). Studies reported that co-digestion promotes co-metabolism between co-substrates with anaerobic consortia through the supply of carbon/energy sources or electron donors, thus enhancing the ability of the microorganism to degrade the organic matter present in the feedstock mixture (Gu 2021; Li et al. 2021a; 2021b; Suhartini et al. 2022a). Furthermore, co-digestion of OPEFBs could efficiently stabilise the carbon and nitrogen content in the feedstock to balance the C/N ratio to start up the digestion (Nurliyana et al. 2015).

Various studies have also emphasised co-digestion of OPEFBs with other biomass substrates as green strategies for improving biogas and methane yields, as shown in Table 6. For instance, in a study conducted by Suksong et al. (2019b), co-digestion of OPEFBs with sludge from liquid-AD (L-AD) was carried out using solid-state AD (SS-AD) under mesophilic and thermophilic conditions. Their study demonstrated that co-digestion operated at thermophilic temperature has superior performance (i.e. process stability and biogas/methane production) than at mesophilic temperature, giving the average methane production of 120 m³ CH₄/ton of OPEFBs. Liew et al. (2021) added that anaerobic co-digestion of OPEFBs with POME at a ratio of 0.6:1 under mesophilic and thermophilic conditions resulted in the highest methane production (i.e. more than 2-fold higher) compared to anaerobic mono-digestion of OPEFBs or POME. Their study suggests that POME (as substrate rich in nitrogen) provide additional nutrient to the microorganism consortia and the synergistic effect between the two substrates, thus enhancing the AD process stability. However, they reported that co-digestion of alkaline pre-treated OPEFBs (at 8% w/v of NaOH solution and a ratio of 1:19 (OPEFBs:NaOH)) with POME at a ratio of 0.31:1 and 0.45:1 did not significantly improved the methane yield. Their finding suggested that the mild concentration of NaOH during alkaline pre-treatment has no effective impact on enhancing delignification and biodegradation.

Table 6. Performance on anaerobic co-digestion of OPEFBs with biomass substrates from several studies.

Co-substrates type	Experimental Condition	Results	Refs.
POME	- Untreated and treated OPEFB - Pre-treatment with hydrothermal (230°C; 15 min) and NaOH (1% w/w)	- Increase methane yield by 25–32% or 0.276–0.340 m^3 CH4/kg VSadded (untreated OPEFBs) - Increase methane yield to 0.392 m^3 CH4/kg VSadded (treated OPEFB)	(O-Thong, Boe, and Angelidaki 2012)
POME	Mesophilic temperature (27–30°C), no pH control	Increase methane yield from 0.170 m^3 CH4/kg VSadded (mono-digestion) to 0.203 m^3 CH4/kg VSadded (co-digestion)	(Nurliyana et al. 2015)
Sludge from liquid-anaerobic digestion (L-AD)	Mesophilic and thermophilic temperature	Thermophilic has better methane yield (120 m^3 CH4/ton of feedstock mixture)	(Suksong et al. 2019b)
POME	- Mesophilic temperature - OPEFBs:POME ratio in the range of 0.25–0.3:1	Increase methane yield to 0.400 L/kg of substrate used	(Kim et al. 2013)
Chlorella sp	Anaerobic and aerobic condition	Biogas production rate increased in the range of 0.128–0.129 m^3/kg COD/day	(Ahmad et al. 2014)
POME	- OLR was in the range of 2–10 gVS - Size of OPEFB was in the range of 0.5–6 cm	- OPEFBs at size of 0.2 mm produced the highest methane yield of 0.320 m^3 CH4/kg VS (63–70% improvement) - POME:EFB (4.5–7.5 ratio and size OPEFBs of 3.3–6 cm) produce methane yield of 0.282 m^3 CH4/kg VS	(Saelor, Kongjan, and O-Thong 2017)
palm oil decanter cake (DC)	- Solid state anaerobic digestion (SS-AD) with RI/S of 1:3, and TS of 15%, - Mesophilic and thermophilic temperature, - Mixing ratio of 1:1 - 19:1 (on a VS basis)	Co-digestion of OPEFBs with DC increased methane yield I from 0.300 m^3 CH4/kg VS to the range of 0.322–0.414 m^3 CH4/kg VS	(Tepsour et al. 2019)
POME	- Alkaline pre-treatment was applied to OPEFBs - OPEFBs:POME ratio in the range of 0.15:1–0.75:1 - Mesophilic and thermophilic temperature	- At ratio of 0.6:1 (treated OPEFBs:POME) gave the highest methane yield of ∼60 mL CH4/ gVS (mesophilic) and ∼80 mL CH4/ gVS (thermophilic) - Co-digestion produced higher methane, with 2.36 times better performance than mono-digestion	(Liew et al. 2021)
Sewage chemical sludge (SCS) and Sewage biological sludge (SBS)	- SS-AD (50–60% TS) - Mesophilic temperature (35 ^oC) - Mixing OPEFBs:SCS or OPEFBs:SBS ratio in the range of 99:1, 95:5, 91:9, 87:13 and 83:17 (%w/w)	The highest methane yield at OPEFBs:SCS or OPEFBs:SBS ratio of 95:5, with value of 18.1 and 18.2 mL CH4/ gVS	(Suksong et al. 2017b)

Pre-treatment prior to anaerobic co-digestion is often required for higher biogas and methane yield. Mamimin et al. (2021) reported that anaerobic co-digestion of mushroom pre-treated OPEFBs with POME significantly enhanced biodegradability by 90.8% and increased methane yield from 0.281 m³CH₄/kgVS (mushroom pre-treated OPEFBs alone) to 0.405 m³CH4/kgVS. In addition, a previous study by O-Thong, Boe, and Angelidaki (2012) compared the methane production of mono- and co-digestion of OPEFBs with POME, with and without hydrothermal pre-treatment. OPEFBs were treated with hydrothermal (at 230°C for 15 min.) and NaOH (1% w/w), following mixing with POME at various ratios with an initial loading of 46 g VS/L. Their study reported that co-digestion of untreated OPEFBs with POME under thermophilic conditions enhanced biodegradability and increased methane production in the range of 25–32% (or 0.276–0.340 m³ CH₄/kg VS_added) at mixing ratios of 0.4:1–2.3:1. While co-digestion of treated OPEFBs with POME at a ratio of 6.8:1, the highest methane production was achieved at a value of 0.392 m³ CH₄/kg VS_added (or 82.7 m³ CH₄/ton) with an electrical potential of 330 kWh. A study reported by Kim et al. (2013) also shows a better co-digestion performance of treated OPEFBs (i.e. drying and grinding) with POME over mono-digestion, resulting in a methane yield of 0.4 L/kg of the substrate used. Moreover, Nurliyana et al. (2015) also highlighted that co-digestion of treated OPEFBs (i.e. drying and particle size reduction) with POME at a ratio of 24%:76% produced the highest methane (0.203 m³ CH₄/kg VS_added) with an electrical potential of 196 kWh.

The above findings confirmed that OPEFBs are highly potential as feedstock for AD. The biological pre-treatments, using co-culture microorganisms and cocktails enzyme, combined with other pre-treatment resulted in a higher reduction of lignin content and increased biogas and methane yields. These pre-treatments are needed to completely break down the complex structure of lignin bonds, allowing easier and faster access for microorganisms or enzymes to degrade cellulose and hemicellulose (Yang and Wyman 2008). Also, co-digestion of OPEFBs with other potential substrates may offer higher efficacy of the AD process due to additional nutrient and co-metabolism effect in the AD reactor. Thus, it can improve the microorganism's ability to effectively degrade and consume the organic materials in the mixed substrates, as highlighted in several studies (Gu 2016; Suhartini et al. 2022a; Zhang et al. 2021). Therefore, biological pre-treatments and co-digestion can be highlighted as alternative strategies to increase the biogas or methane yields of OPEFBs. Yet, these options still need in-depth investigation to find cost-effective, energy-saving, and eco-friendly commercial operations.

5. Potential future opportunities

The current status of OPEFBs for commercial application in the bioenergy sector is still ongoing research (Maryana et al. 2021; Purwandari et al. 2013; Risdianto, Kardiansyah, and Sugiharto 2016; Sudiyani and Hermiati 2010; Sudiyani et al. 2013). The oil palm oil mostly treated OPEFBs for private use, i.e. fertilizer via natural decomposition, steam for boilers, fuels for combustion machines, or burned (Nurdiawati et al. 2015; Srasri et al. 2022). But, OPEFBs have the potential to be used for bioenergy production, biofertilizer, and other bio-products. The findings for exploring the potential of OPEFBs are urgent due to the potential adverse impacts of underutilised or directly disposed of OPEFBs (Harahap et al. 2020b; Khangkhachit et al. 2021). One of the main sectors which are very close to human activity is energy and power utilisation; thus, using OPEFBs as feedstock on bioenergy production and collaborating with the well build technology such as the AD is advised (Liew et al. 2021; Mamimin et al. 2021; Tepsour et al. 2019). The use of biological pre-treatment is proposed, as the enzyme activity contribution and solid content decrement played key roles in degrading the lignocellulose layer. Hence, the feasible and sustainable pre-treatment selected should be based on the following consideration (Cai et al. 2021; Ravindran and Jaiswal 2016; Taherzadeh and Karimi 2008): (1) aid a high process efficiency; (2) avoid hemicellulose or cellulose destruction; (3) less or no formation of inhibitory compounds; (4) energy-saving process; (5) little or no chemical/additives addition; (6) shorter operation time; (7) no release of pollutants and toxic by-products; (8) cost-effective process (i.e. less cost for particle size reduction, for pre-treatment reactor construction or relevant equipment, for microbial culture or enzyme, for storage, etc.); (9) lower operational and capital cost; and (10) eco-friendly process.

In Indonesia, therefore, based on the aforementioned findings, it is suggested that AD of OPEFBs can potentially be applied using four scenarios. Prior to AD in all scenarios, pre-treatment of OPEFBs were essential to enhance delignification, which enhanced biogas and methane production. Considering various technical benefits and challenges stated in Table 4, thus mushroom cultivation pre-treatment is advised in all proposed scenarios. Compared with other biological pre-treatment, cultivation mushroom pre-treatment has recently been highlighted to have low capital cost, no reliance on chemicals or additives, less energy consumption, no inhibitory or toxic compound formation, mild operating condition, and scalability. Before mushroom cultivation pre-treatment, OPEFBs are subjected to physical treatment of particle size reduction. Previous studies have reported that particle size reduction of OPEFBs prior to any biological pre-treatment could advocate better delignification and biodegradation (Ahmad Rizal et al. 2018; Chen et al. 2022a; Temu et al. 2016).

Therefore, in all scenarios, the AD system was combined with physical pre-treatment (i.e. particle size reduction) and mushroom cultivated pre-treatment (i.e. edible mushroom). Then, it is followed by the biogas utilisation unit, digestate storage, digestate treatment, and digestate utilisation. Scenario 1 is anaerobic mono-digestion of OPEFBs with a single-stage configuration using a CSTR. Scenario 2 is anaerobic co-digestion of OPEFBs with other biomass substrates (with TS > 20%) using a single-stage CSTR configuration. Scenario 3 is anaerobic co-digestion of OPEFBs with biomass substrate (with TS < 20%) using a single-stage CSTR system. This scenario aimed to facilitate anaerobic co-digestion of OPEFBs with wastewater containing high organic pollutants (such as POME). For better homogenisation, mixing is essential before the feedstock entering the AD reactor after the combined pre-treatment. Furthermore, the moisture content in the feedstock mixture should also be controlled for better biogas and methane yields. Scenarios 2 and 3 could be applied if other local biomass substrates or wastewater can provide better synergistic effects, thus improving biogas and methane production. Scenario 4 is anaerobic co-digestion of OPEFBs using two-stage system of CSTR and UASB. In the first stage, using a CSTR bioreactor, with similar aims as stated in Scenarios 1 and 2. In the second stage, the UASB bioreactor is used because it is suitable for liquid substrates with high amounts of dissolved organic compounds and lipids. Therefore, the suitable feedstock used for this configuration is co-digestion of OPEFBs with POME. Using a two-stage configuration can produce biogas with the highest CH₄ yield. Furthermore, the stability of the AD process is easier to be achieved and efficiently reduces COD concentration in POME. Thus, with this scenario, it can tackle two waste problems, and at the same time, aiming to produce higher biogas or methane yields. The scheme of the proposed the sustainable AD of OPEFBs system is shown in Figure 6.

Figure 6. Proposed scenario for the sustainable AD of OPEFBs with integrated multi-product biorefinery concept in Indonesia.

Despite multi-stage pre-treatment, the proposed scenarios may offer feasible and sustainable of using OPEFBs for multi-product production. This is also consistent with the integrated multi-product biorefinery and circular economy concept. OPEFBs are utilised as feedstock for edible mushroom cultivation. Then, after mushroom harvesting, the spent OPEFBs substrates are used for feeding the AD digester to produce biogas. The resulting biogas can be converted into electricity and heat (using a combined and heat power unit) or into biomethane (using a biogas upgrading system). The residual digestate can be valorised as biofertiliser in oil palm plantations or nearby agricultural farm area. Thus, a closed-cycle of OPEFBs valorisation can be achieved with minimal or no waste generation. Previous studies have also emphasised that combining mushroom cultivation pre-treatment with AD technology may support sustainable integrated biorefinery and a circular economy of agro-industrial waste and agricultural residues through the production of multi-products (i.e. edible mushroom, biogas, biofertiliser, and vegetable production) (Kumar et al. 2022; Pérez-Chávez, Mayer, and Albertó 2019; Richard et al. 2020). Wagle et al. (2022) stated that multi-product biorefinery provide optimal valorisation of lignocellulosic biomass with less waste generation and carbon emission. They suggested that multi-product biorefinery could promote the sustainable biofuels production, trigger a more sustainable agricultural practice, strengthen energy and food security, and provide additional income to the local residents. However, emerging research on improving the process's efficacy, developing the plant's design, investigating the techno-economy and sustainability analysis, and examining the commercialisation prospects is critical. Hence, the sustainable and multi-product biorefinery of these proposed scenarios could maximise the value generation and the economic prospect of OPEFBs.

Furthermore, waste biorefineries offer sustainable solutions to limit the final disposal of waste by promoting their valorisation through bio-based production chains (i.e. bioenergy and value-added products) to foster long-term economic, social, and environmental benefits (Atabani et al. 2022a; 2022b; Hoang et al. 2022a; 2022b). Similarly, Hoang et al. (2022c) stated that waste-to-energy (W2E) facilitates a promising solution to global problems in energy supply and environmental pollution. They stated that integrated energy systems, efficient design, lean manufacturing system, sale of by-products, and the high-capacity plant should be incorporated with W2E technology to obtain maximum economic and environmental benefits. Moreover, various studies have also emphasised different solutions for the sustainable development of bioenergy and circular economy from waste and biomass. For instance, Yetri et al. (2020) found that cocoa pod waste (CPW) is highly abundant in Indonesia. Their study revealed that activated carbon monoliths from CPW have superior performance as a supercapacitor electrode. They claimed that this strategy could offer environmental benefits (i.e. reducing environmental pollution) and economic benefits (i.e. reducing imported carbon electrodes). Lee et al. (2022) reported that combining spent coffee grounds (SCG) with water chestnut shell biochar could offer energy-saving drying, thus optimising the valorisation of SCG as a coal co-firing fuel. Atabani et al. (2022a) evaluated the biorefining of SCG in combination with other biomass (i.e. coffee silverskin, pine sawdust, crude glycerol, coal fine, and saw) for fuel pellet production offers a higher combustion efficiency and a lower emission of CO₂ and NO_x. They suggested that this strategy may reduce the volume of organic wastes via the production of bioenergy and high-value-added products, thus enhancing the circular economy in the global world through waste biorefinery. Hoang et al. (2022b) also reported that carbohydrates extracted from various biomass/waste feedstock (i.e. agricultural residues, food crops, forest residues, municipal solid waste/MSW, and algae) are potential for the production of furan-based biofuels (i.e. 2,5-dimethylfuran (DMF)). This approach offers a green and prospective solution to produce bioenergy, reduce fossil fuel dependency, and preserve sustainable development globally.

Some management solutions for energy are considered essential to support sustainable bioenergy production. Taking a case study in Vietnam, sustainable bioenergy production via biomass gasification could be achieved if there are supportive Government policies in technology, infrastructure, investment, education, and finances (Hoang et al. 2022a). Furthermore, various approaches to improve biomass-based bioenergy production and management could be applied by incorporating digitalisation, system information, and intelligent infrastructure (Nguyen et al. 2021a; 2021b). For instance, the Internet of Things (IoT) application may offer a solution to improve energy efficiency, thus reducing the operational costs and the environmental impact (Nguyen et al. 2021a). Furthermore, Chen et al. (2022b) studied a type of artificial intelligence (AI), specifically machine learning (ML) approaches using multivariate adaptive regression splines (MARS) and artificial neural network (ANN) to predict the torrefaction severity index (TSI) of various torrefied biomass feedstock. Their study revealed that this method could potentially be applied to estimate the scaling-up of bioenergy production via torrefaction pathways. Another study by Bakır et al. (2022) also reported that the marine predators algorithm (MPA) could be implemented in bioenergy and fossil fuel-based energy generation. This approach offers a solution to forecast the greenhouse gasses (GHG) emissions, hence may assist the decision- and policy-maker in establishing an energy-efficient clean power plant in India. These findings suggest that using the framework of IoT or AI has emerged as an attractive concept to save energy while improving energy production at the same time.

However, some limitations exist to establishing sustainable bioenergy and circular economy via the AD of waste or biomass. These include uncertainties and competition of biomass feedstock, cost of technology, low level of social awareness, high capital and investment cost, and lack of supporting policies for bioenergy production (Hoang et al. 2022a). Other studies also pointed out some technological challenges over the implementation of AD technology, such as scalability, high energy and water requirement, inadequate control and monitoring of the AD process, performance instability, and contamination (Lin et al. 2018; Norrahim et al. 2022; Suhartini et al. 2022b). Other challenges in implementing of AD are post-treatment of digestate, relatively slow digestion, large investment, and lack of legislation for digestate utilisation (Lin et al. 2018). Moreover, other issues that may arise from the AD of palm oil biomass include potential odour, water or soil pollution, and GHG emission problems from digestate if not properly handled (Norrahim et al. 2022). Specific to OPEFBs, our previous study also found that the AD of treated OPEFBs (i.e. fungally, microbially, or chemically pre-treated) (Hidayat et al. 2020; Nurika et al. 2022; Rohma, Suhartini, and Nurika 2021) has higher biogas and methane potential compared with AD of untreated OPEFBs (Suhartini, Hidayat, and Nurika 2020; 2021). The findings suggested that pre-treatment is essential to enhance bioenergy production but may increase the operational cost, technology infrastructure, and energy requirement (Norrahim et al. 2022; Rizal et al. 2018). A further in-depth evaluation is increasingly critical to improve the conversion efficacy and commercialisation of OPEFBs via AD technology.

The future direction of valorising OPEFBs to biogas should focus on the integrated biorefinery and bioeconomy concept to achieve maximum valorisation with minimal or zero waste production. Fermoso et al. (2018) suggested that AD-based biorefinery, whether partially or integrated, offers great potential for effectively converting various low-value biomass or waste substrates into valuable products. Several strategies are proposed in various studies, for instance, the implementation of advanced control and monitoring system, the recovery of other valuable products (i.e. digestate pellet or biochars, organic acids) from AD process (Anukam et al. 2019); and the adoption of low-cost anaerobic digester technology (Mendieta et al. 2021). Furthermore, it is recommended to seek a more cost-effective AD process and technology through technology modification, operating parameters optimisation, and effective microorganism selection for pre-treatment and AD process (Anukam et al. 2019; Shrestha et al. 2017). Shrestha et al. (2017) added that, with respect to biological pre-treatment prior AD process, providing a similar natural ecosystem of the microorganism used during the pre-treatment is suggested for better lignocellulose degradation and higher biogas or methane yield. Future research should also focus on implementing multi-product biorefinery via AD as a central process. This approach offers many benefits, including the generation of eco-friendly energy, achievement of energy efficiency, and comprehensive waste management via production of bioenergy and bio-based products (Rekleitis et al. 2020); creation of the green downstream industry, improvement of local economies, creation of employment opportunities, and creation of new lucrative revenue (Abdullah and Hussein 2021; Ali et al. 2015). Hence, integrated biorefinery with zero-waste and circular economy consideration create a win-win solution to meet the three pillars of sustainable development goals (i.e. economic/profit, environment/planet, and social/people). Moreover, a recent study by Goswami et al. (2022) reported that applying big data analytics (BDA) with combination of IoT and ML promotes the creation of smart and intelligent bio-refineries. Such approaches offer solution to remotely supervise the operation, improve the operation's efficiency, and enhance savings in energy or operational cost. Yet, further research on comprehensive analysis, evaluation, and potential commercialisation of adopting smart and intelligent bio-refineries is critically required.

6. Conclusion

This review paper shows that OPEFBs can be a potential feedstock for the AD system in Indonesia, considering additional pre-treatment for improving delignification and enhancing biodegradation efficiency. In contrast to other pre-treatment, biological pre-treatment offers a green option for handling OPEFBs because it is low cost, less energy consumption, no release of toxic compounds, less pollutants generation, and environmentally safe. Co-digestion strategies may improve biomass digestibility, hence biogas/methane yield Among the discussed biological pre-treatment, the mushroom cultivation pre-treatment provides additional economic benefits from producing and selling of edible mushrooms. Based on the reviewed pre-treatment options, combining biological pre-treatment with other influential, low cost and eco-friendly pre-treatments may result in superior performance on delignification and promote higher biogas or methane production. Combining AD technology, either in mono- or co-digestion system, with multi-stage pre-treatment (i.e. physical followed by mushroom cultivation pre-treatment) could maximise the value of OPEFBs with multi-product generation (i.e. biogas, edible mushroom, and biofertiliser). Therefore, in the Indonesia context, this scenario is proposed for integrating multi-product biorefinery and circular economy with a zero-waste concept. Integrating multi-product biorefinery through AD of OPEFBs may offer feasible and sustainable valorisation routes with lower detrimental environmental impact. Further studies on optimising pre-treatment and AD process efficacy, as well as on the potential of full-scale to commercialisation are highly advised. Further work is also recommended to address key issues around technical, environmental, economic, legislation, and sustainable supply chains.

Acknowledgements

Authors would like to thanks the Ministry of Research, Technology, and Higher Education for the research funding provided through Penelitian Dasar Multi Tahun (Multi-Year Basic Research) Scheme (Grant Number: 108/E4.1/AK.04.PT/2021 and Contract Number: 1271.34/UN10.C10/TU/2021). The authors would also like to thank the Faculty of Agricultural Technology, Universitas Brawijaya, for supporting the manuscript's writing and publication.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Kementerian Riset dan Teknologi/Badan Riset dan Inovasi Nasional [grant number 108/E4.1/AK.04.PT/2021; contract number 127.34/UN10.C10/TU/2021].