Comparison of pretreatment methods that enhance biomethane production from crop residues - a systematic review

A systematic literature review was conducted to compare the efficacy of biological, chemical, physical, and combined pretreatments in enhancing biomethane production from crop residues (CR). Three electronic databases viz., Science Direct, EBSCOhost, and PubMed were used to identify the studies in literature. The pretreatment methods were compared in terms of their advantages and disadvantages with reference to techno-economic aspects. The techno-economic aspects considered included rate of hydrolysis, energy use, effectiveness, cost, and formation of toxic compounds. A total of 3167 studies, covering the period 2014 - 2018, were screened for relevance to the study. Forty-four records (n=44) consisting of 36 research papers (n=36) and eight narrative reviews (n=8) met the inclusion criteria. The results show that physical and chemical methods are the most effective and fastest. These methods have limited utility due to high cost of resources, operation, and energy as well as formation of inhibitory by-products. Despite generation of toxic compounds, combined methods are regarded as fast and costeffective. Biological method is inexpensive, eco-friendly, and low energy-consuming. However, it is a nascent technology that is still developing. A combination of trends in research and development provide the best pretreatment alternative to improve the biomethane production from CR.


Introduction
Biogas is a renewable fuel with wide applications the world over (Weiland, 2010;Achinas et al., 2017;Alhassan et al., 2019). Over 60.8 billion m 3 of biogas are produced annually in the world (WBA, 2018). WBA data show that biogas production is increasing throughout the world. Global biogas production increased 3.7 times from 0.28 EJ to 1.31 EJ during 2000 to 2016 (WBA, 2018). Almost 54% of biogas is produced in the Europe. Africa accounts for only 0.03% of the annual global biogas production (WBA, 2018), yet it has vast resources for biogas production. Biogas is produced by anaerobic digestion (AD) of organic matter (OM). AD is a biochemical process whereby OM is degraded under anaerobic conditions by microbial consortia (Fitzgerald, 2013;Gould, 2015;Hagos et al., 2017;Mulat and Horn, 2018). It is an eco-friendly process and one of the most efficient methods for conversion of biomass to methane (CH4) (Horváth et al., 2016).
OMs vary in their potential to produce biogas by AD (Gould, 2015;Strong et al., 2016). Parameters used to estimate the potential of OM to produce biogas include anaerobic biogasification potential (ABP) and biochemical methane potential (BMP). These parameters allow direct evaluation of biogas yield, which can be achieved by the AD process (Jingura and Kamusoko, 2017). BMP is the maximum volume of CH4, which can be produced per gram of volatile solids (VS) in a substrate (Esposito et al., 2012). BMP provides an indication of the biodegradability of a substrate and its potential to produce CH4 via AD (Sell et al., 2010;Påledal et al., 2013). BMP is an important indicator of the quality of feedstock for biogas production (Triolo et al., 2013). Methods that can be used to determine BMP of feedstock were reviewed by Jingura and Kamusoko (2017). In their review, Jingura and Kamusoko (2017) indicated that the BMP test is a simple, repeatable, and inexpensive method.
The BMP of feedstock is affected by several biochemical characteristics (Gould, 2015;Strong et al., 2016;Jingura and Kamusoko, 2017). These include nutrient content, VS content, chemical oxygen demand (COD), biological oxygen demand (BOD), carbon to nitrogen ratio (C/N), and presence of inhibitory substances (Babaee and Shayegan, 2011; Kwietniewska and Tys, 2014). Amongst these characteristics, the C/N ratio plays a critical role in regulation of the microbial population of autotrophs and heterotrophs (Sepehri and Sarrafzadeh, 2018; Sepehri and Sarrafzadeh, 2019). Differences in biochemical characteristics make it possible to categorize feedstock on the basis of their BMP. This type of characterization places feedstock on different positions on a BMP spectrum. Feedstock at the lower end of the BMP spectrum are those, which present challenges in the AD process, whilst those at the upper end are highly biodegradable.
Different types of OM can be used as substrates for biogas production (Weiland, 2010;Påledal et al., 2013;Gould, 2015;Achinas et al., 2017). Feedstock for biogas production include animal manure and slurry, municipal solid waste, food waste, sewage sludge, and various types of crops and their residues (Demirbas and Balat, 2009;Achinas et al., 2017;Rabii et al., 2019). Over 200 billion tons of agricultural crop residues (CR) are produced annually in the world (Horváth et al., 2016;Patinvoh et al., 2017), presenting a vast resource for biogas production. CR are largely at the lower end of the BMP spectrum because of high lignocellulose content. Lignocellulose limits degradation by anaerobic bacteria (Wang, 2014;Achinas, 2017). CR are heterogeneous in nature limiting their use as feedstock for AD (Horváth et al., 2016). Sahito et al. (2013) reported a BMP range of 142 to 322 mL CH4 g VS -1 for various CR. By comparison, feedstock such as residual fats which are at the higher end of the BMP spectrum have BMP of up to 800 mL CH4 g VS -1 (Muzenda, 2014).
There are several pretreatment options that can optimize the biomethane production from CR (Ariunbaatar et al. . Most of the pretreatment methods have limited applications due to high energy demands, need for special equipment, and emission of several adverse by-products (Liu et al., 2019). The efficacy of pretreatment technologies varies from study to study (Amin et al., 2017;Karrupiah and Azariah, 2019). This heterogeneity limits the utility of information available in literature for planning and designing the AD of CR.
Few attempts have been made to compare the efficacy of pretreatment methods used on CR for the purpose of biogas production by AD. There are also limited aggregated results that provide a comparative analysis on the efficacy of the pretreatment methods with reference to CR. In view of these gaps, it is prudent to conduct a comparative analysis of the efficacy of pretreatment approaches by reviewing extant literature. This will provide empirical evidence on the comparability of existing pretreatment methods. Such information is useful for planning and enhancement of AD plants that utilize CR. To the best of our knowledge, this is the first systematic review study that has been undertaken to compare the efficacy of pretreatment methods in enhancing biomethane production from CR.

Procedure
The standard procedure on performing a systematic literature review (Kitchenham and Charters, 2007;Okoli and Schabram, 2010;Mittal et al., 2018) was used. The search period was January 2014 to November 2018. Most recent work on enhancing biogas production was published during this period (Prasad et al., 2017;Kougias and Angelidaki, 2018). Three databases were chosen on the basis of their availability in the university library and are among the top ten online research databases. These are Science Direct, EBSCOhost and PubMed. The search was delineated to online full-text journal articles. Gray literature covering government reports, conference proceedings, graduate dissertations and unpublished papers related to biogas was excluded. The challenges of searching gray literature were pointed out by Mahood et al. (2014) and Paez (2017).
A summary of the search protocol for identification and selection of articles for inclusion in this study is shown in Figure 1. The process for inclusion of a study was in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Moher et al., 2009;Forbes et al., 2018). The key search terms used for extracting the relevant articles are shown in Table 1. These were selected using the model of a concept map (Asiksoy, 2019). 'Biogas' and 'crop residues' were

Comparison criteria
The main variables of interest were advantages and disadvantages of each pretreatment method in terms of techno-economic aspects. The the major search terms. The major search terms were held as constant in order to eliminate articles not concerned with the use of CR for biogas production.
techno-economic aspects used were: rate of hydrolysis, energy use, effectiveness, cost input, and formation of toxic compounds. These were adapted from Wagner et al.

Identification and selection of articles for inclusion
A total of 3167 full text articles were screened for relevance to the study as shown in Figure 2. Disaggregation of the number of scooped articles by search terms is shown in Table 1. It is clearly shown that the majority of the publications (n = 1017) were extracted by using the search terms, "chemical pretreatment" and "crop residues" and "biogas".
Forty-four full text articles met the inclusion criteria. Thirty-six of the articles were research papers and eight were narrative reviews. The trend in the number of publications on pretreatment of CR for biogas production from 2014 to 2018 is shown in Figure 3. The number of publications regarding pretreatment of CR for biogas production rose between 2014 and 2017. This points to sustained research effort on the subject. This can be ascribed to the

Characteristics of selected articles
The number of selected articles and the authors are shown in Table 2. The articles are organized according to the four pretreatment methods. Various pretreatment techniques are used to depolymerize CR into simple components. These are shown in Figure 4. The key pretreatment methods are chemical, physical, biological, and a combination of these processes  "Pretreatment" and "crop residues" and "biogas" 384 214 6 604 "Chemical pretreatment" and "crop residues" and "biogas" 365 651 1 1 017 "Biological pretreatment" and "crop residues" and " biogas" 309 521 2 832 "Physical pretreatment" and "crop residues" and " biogas" 220 493 1 714 1082 Fig. 2. Summary of identification and selection of articles for the study inclusion process.   Table 3 provides information on the pretreatment methods with reference to rate of hydrolysis. According to information in Table 3, physical methods appear to be the fastest amongst the pretreatment methods. This is more so for microwave (MW) pretreatment (Wu et al., 2015;Kumar and Sharma, 2017). This is in agreement with Yuan et al. As shown in Table 3, studies considered in this review cited the major drawback of biological pretreatment as being a slow process (

Energy use
Seven studies indicated the main advantage of biological pretreatment as low energy consumption method ( Table 4)

Pretreatment method Observations Reference
Chemical -Chemical pretreatment has low energy needs Sahito

Effectiveness
Information in Table 5 suggests that chemical pretreatment is the most effective method. Song  As shown in Table 5, biological pretreatment is not as effective as the other methods. None of biological methods is efficient as standalone pretreatment method (Thomsen et al., 2016). For instance, no significant difference in methane yields between enzymatically-pretreated banana stems and non-treated stems was observed by Li et al. (2016). Paul et al. (2018) reported that fungal pretreatment of agricultural biomass did not improve methane production. However, pretreatment of rice straw with fungal strains such as Pleurotus ostreatus and Trichoderma reesei increased methane yield by 120% (Wagner et al., 2018). This variation in results is expected as biological pretreatment is still under development.
Physical pretreatment appears to be generally effective as shown in Table  5. For example, hot water pretreatment increased methane yield from rice straw by 222% (Ge et al., 2016) and MW method increased methane yield by 28% (Wu et al., 2015). Baeta et al. (2016) reported that autohydrolysis is a highly effective process. As reviewed by Den et al. (2018), MW pretreatment at 200 or 300°C cannot increase biogas production. Furthermore, an inverse relationship between temperature increase and biogas production was noted during MW pretreatment (Den et al., 2018). This is because high temperatures can lead to production of heat-induced inhibitors such as phenolics and furfural. As a result, MW pretreatment has been used in combination with chemical pretreatment at fairly low temperatures (Den et al., 2018). Table 5.
Comparison of pretreatment methods in terms of effectiveness.

Pretreatment method Observations Reference
Chemical Chemical pretreatment is an effective method  Table 6 compares the different pretreatments in terms of cost of operation. Despite reported to be less effective, biological pretreatment is considered to be variable in cost-effectiveness. Biological pretreatments were reported to be inexpensive (Mulakhudair et  is more economically and environmentally friendly than other pretreatment methods. During MP, microorganisms are partially exposed to O2 or air under moderate operating conditions such as temperature and pressure. MP has also minimal enzyme and energy requirements (Mustafa et al., 2018). The process is designed to promote the hydrolysis stage through stimulation of cell growth and activity (Wagner et al., 2018). MP is likely to be one of the most promising Table 6. Comparison of pretreatment methods in terms of cost. It is noted that, despite variations shown in Table 6, chemical and physical pretreatments appear to be predominantly expensive. Selected results show that chemical or physical pretreatments are expensive and not economically viable for biogas production from CR. Chemical pretreatments were reported to be expensive due to high costs of disposal of digestion residues (Wagner et  . Based on pretreatment of corn straw, Ca(OH)2 could be a better option than NaOH although their pretreatment costs were not highly variable (Song et al., 2014). Alkali pretreatment was favored over other pretreatments due to low operational costs (Ismail et al., 2017). The low cost of lime and ease of recovery from the waste make it a better pretreatment technology than other alkalis (Kumar et al., 2017). Comparison of pretreatment methods in terms of formation of toxic compounds.    Table 7 reported that chemical pretreatment causes formation of toxic compounds. As such, formation of inhibitory compounds is one of the disadvantages of chemical pretreatment. Production of inhibitors of methanogenic metabolism and growth can be ascribed to elevated levels of alkalinity and pH. The ultimate result is earlier process failure during biomethane production. The optimum pH range for methanogenic activity is 6.5 to 8.2. Alkaline addition provides buffering capacity and prevents inhibition during AD .

Formation of toxic compounds
As shown in Table 7, physical processes generally lead to formation of inhibitory compounds. For instance, thermal pretreatment at temperatures above 160°C may lead to partial degradation of polysaccharides and lignin to phenolic and heterocyclic compounds (Zieminski and Kowalska-Wentel, 2017). However, some of the studies (Baeta et al., 2016;Kumar and Sharma, 2017) show that autohydrolysis and mill pretreatment do not generate toxic compounds. Seven studies in Table 7 show that combined processes generate inhibitory compounds, especially SE pretreatment. Zheng et al. (2014) reported that the efficacy of SE and extrusion may be affected by production of fermentation inhibitory substances such as furfural and hydroxymethylfurfural (HMF) due to sugar and lignin degradation. This can explain the CH4 yield loss during high temperature SE pretreatment of late harvested hay observed by Bauer et al. (2014). Table 8 provides aggregated results of the five parameters for all the pretreatment methods. Putting together, biological methods have more techno-economic advantages across the five parameters compared to other methods. The advantages of biological pretreatments are tagged with low energy use, low cost, and ability to avoid formation of by-products that are toxic to methanogens. However, there is need to improve on the efficacy of biological pretreatment. Focus area should be enhancement of the rate of hydrolysis. Despite high effectiveness, the main limitations of chemical and physical methods are high energy use and cost, as shown in Table 8.

Conclusions
It is evident from this study that pretreatment methods used for CR are variable in their effects. As such, the multi-factor evaluation conducted in this study provides information that can assist selection of methods to use. Rate of hydrolysis, energy use, effectiveness, cost, and formation of toxic compounds are critical parameters that inform selection of a pretreatment method. Physical and chemical pretreatment methods have been utilized to some extent at industrial scale for delignification of CR to enhance biomethane production. However, these methods are energy intensive, expensive, not environmentally safe, and have the ability to generate toxic compounds including carboxylic acids, furans, and phenolic compounds which may be inhibit methanogenic activity. In comparison with other methods, biological pretreatment offers more techno-economic advantages. Biological pretreatment is regarded as inexpensive and low energy need process that can minimize formation of inhibitory compounds.  biological pretreatment is one of the most promising technologies for enhancing biomethane production of CR. Moreover, rigorous research is still needed for the development of novel microorganisms and more efficient pretreatment options for CR yielding potential results.