Accepted Manuscript Review Anaerobic digestion of lignocellulosic biomass: Challenges and opportunities Chayanon Sawatdeenarunat, K.C. Surendra, Devin Takara, Hans Oechsner, Samir Kumar Khanal PII: DOI: Reference:

S0960-8524(14)01362-5 http://dx.doi.org/10.1016/j.biortech.2014.09.103 BITE 13993

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

31 July 2014 19 September 2014 20 September 2014

Please cite this article as: Sawatdeenarunat, C., Surendra, K.C., Takara, D., Oechsner, H., Kumar Khanal, S., Anaerobic digestion of lignocellulosic biomass: Challenges and opportunities, Bioresource Technology (2014), doi: http://dx.doi.org/10.1016/j.biortech.2014.09.103

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Anaerobic Digestion of Lignocellulosic Biomass: Challenges and Opportunities Chayanon Sawatdeenarunat1, K.C. Surendra1, Devin Takara1, Hans Oechsner2 and Samir Kumar Khanal1*

1

Department of Molecular Biosciences and Bioengineering (MBBE), University of Hawai'i

at Mānoa, 1955 East-West Road, Agricultural Science Building 218, Honolulu, HI 96822, USA 2

University of Hohenheim, State Institute of Agricultural Engineering and Bioenergy,

Garbenstrasse 9, Stuttgart 70599, Germany * Author to whom all correspondence should be addressed to: E-mail: [email protected]; Tel: +1-808-956-3812; Fax: +1-808-956-3542

1

Highlights: •

Anaerobic digestion of lignocellulosic biomass can sustainably produce renewable energy.



Anaerobic co-digestion is a promising technology to improve digester performance.



Solid-state anaerobic digestion could efficiently digest high solid organic feedstocks.



Rumen microorganisms are an effective inoculum for the anaerobic digestion of lignocelluloses.



Anaerobic biorefinery could provide both bioenergy and valuable biochemicals.

2

Abstract Anaerobic digestion (AD) of lignocellulosic biomass provides an excellent opportunity to convert abundant bioresources into renewable energy. Rumen microorganisms, in contrast to conventional microorganisms, are an effective inoculum for digesting lignocellulosic biomass due to their intrinsic ability to degrade substrate rich in cellulosic fiber. However, there are still several challenges that must be overcome for the efficient digestion of lignocellulosic biomass. Anaerobic biorefinery is an emerging concept that not only generates bioenergy, but also high-value biochemical/products from the same feedstock. This review paper highlights the current status of lignocellulosic biomass digestion and discusses its challenges. The paper also discusses the future research needs of lignocellulosic biomass digestion.

Keywords: Anaerobic digestion; bioenergy; lignocellulosic biomass; solid-state digestion; rumen microorganism; anaerobic biorefinery

3

1

Introduction

In recent years, global energy demand has grown rapidly due to rising world populations and affluence (Surendra et al., 2014). Worldwide, energy consumption reached 524 QBtu in 2010, and is estimated to peak at 800 QBtu by 2040; corresponding to an average growth of 1.5% per year (EIA, 2013). Significantly, a large fraction of the world’s total energy demands (more than 84%) is supported by non-renewable fossil resources such as coal, oil, and natural gas. These resources are not only limited in supply but also have adverse effects on the environment due to the emission of greenhouse gases (GHGs) into the atmosphere (EIA, 2013). Bioenergy, especially biogas produced through the anaerobic digestion (AD) of renewable feedstocks, is considered to be one of the highly promising alternatives to fossil-derived energy due to several inherent and significant merits (Kaparaju et al., 2009; Cheng et al., 2011). Because of its advantages over conventional fossil-derived resources, AD has been adopted and integrated into society over the last century, with thousands of full-scale plants currently in operation worldwide. AD is suitable for converting non-sterile, diverse, complex feedstocks into energy-rich biogas. Many biodegradable feedstocks such as industrial wastewater, food wastes, animal manure, agri-wastes, sewage sludge, organic fraction of municipal solid waste, among others, have been employed as substrates for commercial biogas production. Such facilities illustrate the unique potential for bioremediation and waste stabilization with concurrent bioenergy production. More recently, lignocellulosic biomass, namely agri-residues and energy crops, have been gaining much attention as candidate feedstocks for producing bioenergy and biobased 4

products. Unlike conventional biorenewable feedstocks (i.e., sugar- and starch-based crops), lignocellulosic biomass do not directly compete with food or feed production. Moreover, high biomass yields even under low inputs of energy, water, fertilizers, and pesticides, make these crops ideal for biogas (and bioenergy) production (McKendry, 2002). The composition of lignocellulosic biomass, however, consists primarily of cellulose, hemicellulose, and lignin, and the interactions of these components create a highly resistant and recalcitrant biomass structure. Consequently, the hydrolysis of lignocellulose often becomes the rate-limiting step during traditional AD (Khanal, 2008). Several studies have focused on enhancing the digestibility of lignocellulosic biomass through physical, chemical, biological and hybrid pretreatments in the production of liquid fuels (primarily ethanol) via biochemical pathways (FitzPatrick et al., 2010; Takara and Khanal, 2011). Mechanical milling, steam explosion, hot water washing, acid and alkali pretreatments and ammonia fiber expansion, among others, have been employed as upstream unit operations to disrupt the complex structure of biomass, thereby increasing its porosity, removing lignin and/or hemicellulose, and reducing the overall crystallinity of the biomass structure to facilitate the biological conversion of biomass into bioenergy and biobased products (Monlau et al., 2013; Agbor et al., 2011). Many of these pretreatments, however, are economically and environmentally unfavorable due to the high cost of enzymes and the production of solid/liquid waste streams (Shrestha et al., 2008; Monlau et al., 2013; Alvira et al., 2010; Agbor et al., 2011; Kumar et al., 2009). AD is the naturally occurring, biological pretreatment of organic substrates carried out by robust, mixed culture microbial communities in the absence of oxygen (Khanal, 2008). The 5

consortium of microbes works synergistically to deconstruct recalcitrant biomass structures (like lignocellulose) into their respective fundamental components. In conventional bioprocessing strategies, the whole lignocellulosic feedstock is ground and fed into an anaerobic bioreactor to convert complex carbohydrates and organic matter into energy-rich biogas (Weiland, 2010). Though effective, this approach is time consuming and energy intensive, consequently limiting its application for large-scale bioenergy production from dedicated energy crops. An insightful study conducted by Yue et al. (2010) suggested that certain microorganisms present in the AD slurry may prefer specific biomass constituents over others. In particular, the authors found that the heterogeneous polysaccharide, hemicellulose, was broken down and metabolized before other structural components. By carefully adjusting the solids retention time (SRT), among several other operating conditions, the AD process may have the ability to promote methane (CH4) production from hemicellulose exclusively, while leaving behind cellulose and lignin in the fibrous solid residue. The removal of hemicellulose effectively destabilizes the recalcitrant biomass structure, thus allowing for the solubilization (i.e., saccharification) of cellulose by commercial enzymes in the downstream processes (Maclellan et al., 2013;Yue et al., 2011). Glucose, derived from the hydrolysis of cellulose, can serve as a substrate for producing drop-in biofuels via the carboxylate platform (Agler et al., 2011) or as a precursor for highvalue products such as bioplastics, succinic acid, fungal protein, etc. (FitzPatrick et al., 2010; Cherubini and Strømman, 2011). The organic acids produced through fermentative processes (where applicable) also have potential use in a number of chemical industries and products (e.g., resins, pesticides, fertilizers, etc.) (Cherubini and Strømman, 2011). Any 6

lignin remaining in the solid residue has little commercial value in current markets, but can be burned for in-house heat and electricity generation. Unique to an AD biorefinery approach, in contrast to conventional biofuel/bioenergy production, is the inherent generation of digestate (i.e., the nutrient-rich residue) resulting from the digested slurry. The digestate has important land-use applications and serves to improve nutrient retention in soil. The idealized AD biorefinery, as illustrated in Figure 1, is a rapidly emerging concept that can significantly improve the commercial viability and applicability of the AD process. The main purpose of this review is to highlight recent advances with respect to AD biorefinery. A particular emphasis has been placed on leading energy crops, the current status of energy crop digestion, and opportunities and challenges that are associated with energy crop digestion. Further research needs and recommendations are also discussed. 2

Lignocellulosic biomass: Composition

Lignocellulosic biomass is an abundantly available bioresource with an annual (global) yield of over 200 billion dry metric tons per year (Kumar et al., 2008). For example, U.S. alone produces about 1.37 billion dry tons of such biomass per year for biofuel production (Limayem and Ricke, 2012). Common examples of these renewable resources include agri- and forest residues, and dedicated energy crops (Cherubini, 2010). The basic structure of lignocellulose is comprised primarily of cellulose (35-50%), hemicellulose (2035%), and lignin (10-25%) (Liu et al., 2008), along with smaller quantities of other organic and non-organic compounds like proteins, lipids, and other extractives (Frigon and Guiot, 2010). Table 1 summarizes the typical composition of some commonly used lignocellulosic feedstocks. It is prudent to mention that the amounts of these constituents not only varies 7

between species, but can also vary due to growth conditions and maturation. Cellulose is the main constituent of virtually all plant cell walls, thus making this compound one of the most abundant (renewable) polymers on the planet. At the molecular level, cellulose (C6H10O5)n is a linear (unbranched) homopolysaccharide consisting of 10,000 to 15,000 Dglucose units linked by β(1→4) covalent bonds. The β configuration of the glucose residues creates a structure with physical properties that are very different from starch; another homopolysaccharide of glucose with α oriented bonds. The β(1→4) linkages of cellulose also make the polysaccharide nearly indigestible for most animals (except ruminants) since special enzymes, known as cellulases, are required to hydrolyze the covalent bonds. Hemicellulose, in contrast, is a highly branched heteropolysaccharide consisting of a wide variety of sugars (C-5 and C-6). The side groups extending off of the main hemicellulosic backbone preclude the polymer from forming crystalline structures reinforced by hydrogen bonding, unlike cellulose. The individual sugars of hemicellulose can differ considerably depending on the plant species, however, in general, the saccharification of hemicellulose typically produces a mixture of glucose, galactose, mannose, arabinose, xylose, and rhamnose. The last main constituent of lignocellulose, namely lignin, is a phenylpropane-based polymer with little value for bioenergy production, despite being the second most abundant polymer on the earth. Lignin is an essential part of the biomass structure as it provides mechanical support and water impermeability to the secondary cell walls of plants, but lignin also serves as both a physical and biochemical barrier that impedes most biomass-to-bioenergy conversion processes. 8

Table 1. The composition of selected lignocellulosic biomass and animal manure 3

Lignocellulosic biomass: Methane production potential

The AD of lignocellulosic biomass produces energy-rich CH4 gas. The yield of CH4 per unit area is often used to determine the energy productivity of a particular feedstock, and can vary significantly between species, as well as with maturity, geographical location, and inputs (water, fertilizer etc.) within the same species (Yang et al., 2013). The Biochemical Methane Potential (BMP) test is widely used to examine the anaerobic digestibility of organic substrates. The characteristics of selected energy crops with respect to BMP are summarized in Table 2. Table 2. The biomass yield and methane (CH4) production potential of selected lignocellulosic biomass The economic feasibility of AD is strongly contingent on the CH4 potential of the substrate. Higher CH4 production from a given feedstock directly corresponds to shorter payback periods (on investments) for commercial AD facilities. The feedstock composition is an important factor affecting both the CH4 yield as well as digester stability; which in turn is governed by the plant species, geographical location, and biomass maturity as discussed previously (Amon et al., 2007b). The authors correlated the effects of harvesting time with biogas production for whole maize (both stover and ear(s)), and found that the best harvesting age with respect to CH4 yield per hectare was at the end of wax ripeness (i.e., after 122 days). During this stage, the plant contained between 35-39% dry matter. At full ripeness (i.e., after 151 days), increases in CH4 production were minimal. This occurrence can likely be attributed to the carbon-to-nitrogen (C/N) ratio of the maize, reported as 42, 9

which was much higher than the recommended C/N ratio (i.e., 20-30) for AD (Chandra et al., 2012b). Additionally, the lignin content of the maize may have increased as the crop matured in the field. In general, CH4 production is known to be less from lignocellulosic crops which are high in lignin content (Agbor et al., 2011; Alvira et al., 2010; Astuti et al., 2009). Despite the lower conversion efficiency of the matured crop, however, the highest CH4 yield per unit area was observed for maize at full ripeness due to the significantly high volatile solids (VS) yield per cropping area. The increased VS content for older maize compensated for a lower CH4 yield per unit VS added (Schittenhelm, 2008). With respect to other candidate feedstocks, like cereal crops, for example, (e.g., wheat, triticale, and rye) harvesting should be conducted between the grain-in-the-milk stage and grain-in-the-dough stage to obtain the highest CH4 yield per unit area (Amon et al., 2007a). Similarly, for perennial grasses, the first cut should be conducted after the ear-emergence stage to optimize the CH4 yield (Amon et al., 2007a). As mentioned previously, the lignin content of the feedstock also affects the CH4 production potential. Triolo et al. (2012) reported that lignin contents greater than 100 g/kg VS was a critical point for AD, resulting in notably low CH4 potentials. A statistical model to predict CH4 yield based on the composition and structure of lignocellulosic biomass was developed by several researchers as shown in Table 3. These equations were primarily proposed to estimate the CH4 yield without conducting the time-consuming BMP test. Compared to other biomass characteristics, the lignin content was reported to be the most important factor affecting CH4 production (Gunaseelan, 2007); more than cellulose crystallinity (Monlau et al., 2012b). Triolo et al. (2011) concluded that the lignin fraction 10

alone could be used to predict BMP, however, the model produced slightly better predictions when the effects of cellulose were also accounted for. Importantly, there are still limitations in the modelling approach, particularly for inaccuracies observed with lipid-rich substrates. The model is also unable to incorporate different physical and chemical attributes of the feedstocks such as pH, particle size, moisture content and porosity among others (Monlau et al., 2012b). Table 3. Statistical models to predict Biochemical Methane Potential (BMP) of selected lignocellulosic biomass 4

Challenges in digesting lignocellulosic biomass

The low yield of quality biomass (i.e., biomass high in energy content and easy to convert into end product, CH4) is still one of the major challenges. Thus, one of the approaches of increasing the CH4 yield per unit cropping area is to improve the biomass yield. Selection of appropriate biomass species for a particular geographic location with improved photosynthetic efficiency, better efficiency for input utilization (e.g., fertilizer, and irrigation) and resistant to diseases and pests would improve the biomass yield (Sims et al., 2006). Further, conversion of lignocellulosic biomass into end product (i.e., CH4), is another major hurdle due to the complexity of lignocellulosic biomass structure. Thus, biomass should be harvested at the appropriate stage of maturity which would not only provide the good yield of biomass but also provide the biomass which could be converted into CH4 without intense biomass pretreatment. Additionally, lack of good digester for handling high solids feedstocks such as lignocellulosic biomass is another limitation in digesting lignocellulosic biomass. Significant part of knowledge in the current anaerobic 11

digester designs came from the wastewater treatment plant and was mainly developed for handling low solids feedstocks. Operating such a digester for digesting lignocellulosic biomass is energy intensive, especially in mixing biomass and ultimately results in lower net energy yield. Thus, an appropriate digester design capable of efficient handling of high solids feedstocks could enhance the energy balance of lignocellulosic biomass digestion to CH4. Finally, the comprehensive system for efficient utilization of both digested residue (i.e., solid residue after AD) and effluent is yet to be developed. The digested residue, usually rich in cellulose, can be broken down into simple sugar via enzymatic hydrolysis and utilized for producing high value products (as discussed in the later section). The effluent, on the other hand, is usually rich in nitrogen and other trace elements and needs to be treated before discharging into an environment. The combined AD and microalgae production has recently been developed to utilize the nutrients in the effluent by algae. Lipids from such produced algae could be utilized for biodiesel production and the residue after lipid harvest could be further fed into the digester for CH4 production. The effluent after the algae production can be recycled for operating anaerobic digester.

5

Current technology in lignocellulosic biomass digestion

5.1

Anaerobic co-digestion

As alluded to earlier, the C/N ratio of feedstocks is critically important to facilitate the conversion of lignocellulosic biomass to CH4 (Wu et al., 2010). Dedicated energy crops are often rich in carbohydrates, but are low in nitrogen (Giuliano et al., 2013; Ye et al., 2013). Thus, the mono-digestion of energy crops alone may result significantly low CH4 yield 12

(i.e., biogas high in carbon dioxide (CO2) but low in CH4 content) if an optimal C/N ratio of 20-30 is not properly maintained. Another significant demerit of mono-digestion of energy crops is the lack of essential trace elements such as iron, cobalt, nickel, molybdenum, selenium, and tungsten. These metals, though present in miniscule quantities, are considered vital for sustaining methanogens. Thus, the supplementation of nutrients and trace elements enhances CH4 yields in addition to improving digester stability. For example, CH4 yield from Napier grass reportedly increased by 40% when nickel, cobalt, molybdenum and selenium were added to the reactor (Demirel and Scherer, 2011). The mono-digestion of conventional AD substrates, such as animal waste, is also not recommended as it can result in digester instability caused by ammonia toxicity from the rapid degradation of organic nitrogen such as urea and protein (Abouelenien et al., 2014). Thus, the co-digestion of carbohydrate-rich lignocellulosic biomass with nitrogen-rich animal waste has significant implications in maintaining an optimal C/N ratio for commercial CH4 production with renewable feedstocks (Giuliano et al., 2013). Several studies to date have demonstrated the successful anaerobic co-digestion of livestock wastes and lignocelluloses as illustrated in Table 4. The establishment and maintenance of an appropriate C/N ratio was one of the key factors surrounding a successful co-digestion. Ye et al. (2013) reported the co-digestion of rice straw and swine manure in a series of batch experiments. The CH4 yield increased by an impressive 71% compared to the monodigestion of rice straw when the quantity of swine manure-to-rice straw was adjusted to 2:1 (on a VS basis). The C/N ratios of the co-substrates (i.e., mixture of swine manure and rice straw) and mono-substrate (i.e., rice straw) were 21.7 and 47, respectively. In an earlier 13

study, a C/N ratio of 20 resulted in the highest biogas yields during the co-digestion of swine manure and three lignocellulosic substrates; namely, wheat straw, corn stalk and oat straw (Wu et al., 2010). The volume of biogas produced reportedly increased by 11, 8, and 6-folds when compared with AD of swine manure as the control, respectively. Chicken manure is rich in organic nitrogen, compared to most other animal wastes. Consequently, during mono-digestion, AD systems are often prone to suffer from ammonia toxicity. Li et al. (2014) investigated the co-digestion of chicken manure and corn stover using batch and continuously-stirred tank reactors (CSTR), where C/N ratio was adjusted to 20. The authors found that the CH4 yield obtained from their batch experiment achieved 62% of the calculated theoretical yield. The CSTR reactor was operated at an organic loading rate of 4 kgVS/m3/day with stable performance without VFAs accumulation. This enhancement in biogas production strongly supports the importance of an optimal C/N ratio for better CH4 yields. Table 4. The anaerobic co-digestion of selected lignocellulosic biomass and animal manure 5.2

Solid-state anaerobic digestion (SS-AD)

In general, AD is classified into three important groups based on their operating total solids (TS) contents namely; liquid (L-AD), semi-solid (S-AD), and solid-state (SS-AD) with respective TS concentrations of less than 10%, 10-20% and more than 20% (Karthikeyan and Visvanathan, 2012; Cui et al., 2011). The criteria is rather loosely defined, however, as Brown et al. (2012) refers to AD with TS contents less than 15% and more than 15% as LAD and SS-AD, respectively. The L-AD is typically suitable for substrates which are high in moisture content, such as domestic and industrial wastewater (Xu and Li, 2012). For 14

dilute waste streams (e.g., wastewater with TS content

Anaerobic digestion of lignocellulosic biomass: challenges and opportunities.

Anaerobic digestion (AD) of lignocellulosic biomass provides an excellent opportunity to convert abundant bioresources into renewable energy. Rumen mi...
745KB Sizes 0 Downloads 9 Views