15.2.1 Biogas

Anaerobic digestion is a complex process which involves several microorganism groups in a consortium which converts the organic matter of the biomass into biogas (a mixture of CO₂ and CH₄). Biogas contains traces of other gases such as hydrogen, hydrogen sulphide and carbon monoxide. The final mixture is treated and enriched in methane if it is to be used through the gas grid or as a transportation fuel.

There are four separate steps comprising the anaerobic digestion process: hydrolysis, acidogenesis (fermentation), acetogenesis and methanogenesis (Fig. 15.1). Hydrolysis is linked to acidogenesis as micro-organisms excrete extracellular enzymes which hydrolyse the complex organic matter into smaller compounds. These can be transferred through their cellular membranes and further degraded into a mixture of acids and alcohols. Hydrolytic enzymes include cellulase, cellobiase, xylanase and amylase for converting carbohydrates into sugars, protease for hydrolysing proteins into amino acids and lipase for degrading lipids into glycerol and long chain fatty acids (LCFA). In lignocellulosic materials, hydrolytic enzymes cannot access the polymers as these are amorphously bound with lignin. Special pretreatment methods are required for the initiation of hydrolysis (Section 15.3.4). Although hydrolysis is considered as a single step, it is actually a group of individual processes (enzyme production inside the microbial cell, diffusion through the membrane, adsorption on the polymer molecule, reaction, and enzyme deactivation). It is considered to be the slowest step of the whole process where the organic matter is in particulate form.

Acidogenesis involves multiple reactions for converting the products of hydrolysis into acids of low molecular weight and alcohol. The final composition of the acids and alcohol in this step depends on the quantities of sugars, amino acids and fatty acids produced from hydrolysis. In the case of sugar acidification, the microbes have the ability to shift their metabolism to more reduced organic metabolites, depending on the conditions including pH, hydrogen and partial pressure (Angelidaki et al., 2011). It is generally accepted that under conditions of low pH and high levels of hydrogen and formate, more reduced metabolites are produced, such as butyrate, lactate and ethanol.

The other two steps, acetogenesis and methanogenesis, are also linked as the hydrogen produced from acetogenesis must be scavenged by the methanogens if acetogenesis reactions are to proceed. Acetogenesis involves oxidation reactions of the acidogenesis products by micro-organisms which produce hydrogen to dispose of the electrons derived from oxidation. The production of hydrogen is thermodynamically feasible at low partial pressure (< 10^− 4 atm) (Harper and Pohland, 1987).

Formate is also used as an electron carrier. Hydrogen and formate are kept at low concentrations by micro-organisms such as methanogens, homoacetogens and sulphate reducers. Methane is the typical product of hydrogen and formate scavenging in the absence of sulphate. The syntrophic relationship between these micro-organisms is termed ‘interspecies hydrogen transfer’. Methanogenesis also involves the transformation of acetate into methane and this process accounts for 70% of the methane produced under stable conditions. Methanogens are generally sensitive to pH and ammonia, temperature changes and other factors. Methanogenesis is considered to be the step which limits the speed of the whole process (in the case of soluble organic matter), due to the sensitivity of slowly growing methanogens.

The main constituents of biomass mixture are methane and carbon dioxide. Its composition is affected by the biodegradability of the organic matter (feedstock) and the mean oxidative state of the carbon it contains. The lower the level of carbon in lipid and protein-rich feedstocks, the richer in methane the biogas will be. The methane yields from various types of feedstocks (energy crops, residues and wastewaters) are reported in Section 15.3.

15.2.2 Hydrogen

Hydrogen is a colourless and odourless gas and is the simplest element present in water, biomass and fossil fuels (gasoline and natural gas). It is considered to be an alternative, environmentally friendly fuel and is described as the ‘energy carrier’ of the future. It is also a very useful reagent for the production of a variety of chemicals. When used as a fuel, hydrogen produces water instead of greenhouse gases and has a high energy yield (122 kJ/g, which greatly exceeds that of hydrocarbon fuels). Hydrogen does not occur naturally as a gas but is always combined with other elements (carbon, oxygen, nitrogen) through chemical bonds, the breaking of which are energy intensive. It is mainly produced from hydrocarbons through steam reforming and from water by means of electrolysis. Biomass can also be utilised for hydrogen production. Recently, biological methods have been developed which are believed to be cost effective and to have a low environmental impact when compared to the thermochemical processing of biomass.

The biological methods for hydrogen production from biomass (biohydrogen production) include photo and dark fermentation in bio-reactors and bio-electrolysis in microbial electrolysis cells (MEC). The hydrogen yield from photo fermentation is low due to a limited range of substrates (acetate, etc.), the large bio-reactor surface which is needed for efficient light penetration, and the light saturation effect. In MEC, the biodegradable material is converted into hydrogen instead of electricity as in microbial fuel cells (Call and Logan, 2008). An MEC operates under anaerobic conditions (without oxygen in the cathode) and a small external voltage is applied to the cell, so that protons and electrons produced by the anodic reaction combine at the cathode to form hydrogen. MEC is a promising technology but it is still in its infancy and many microbiological, technological and economic challenges remain to be overcome.

Dark fermentation appears more promising in terms of yield and viability when compared to alternative bio-hydrogen processes. It is directly related to the acidogenic stage of the anaerobic digestion processes described in Section 15.2.1. The cost of dark fermentation hydrogen production is estimated to be 340 times lower than that of photosynthetic processes (Morimoto, 2002). Dark fermentation is not subject to the limitations of photo fermentation in terms of feedstock suitability, complex and expensive bio-reactor engineering and poor yield. However, higher yields and process efficiency are required if it is to become a sustainable process.

Hydrogen production generally disposes of excess electrons through the enzyme hydrogenase in certain strictly anaerobic micro-organisms (Clostridia, methylotrophs, rumen bacteria, methanogenic bacteria, archaea), facultative anaerobes (Escherichia coli, Enterobacter, Citrobacter), or even aerobes (Alcaligenes, Bacillus). Among the hydrogen-producing bacteria, Clostridium sp. and Enterobacter, are the most widely studied. Species of the genus Clostridium such as C. butyricum, C. acetobutyricum, C. beijerinckii, C. thermolacticum, C. tyrobutyricum, C. thermocellum and C. paraputrificum are examples of strict anaerobic and spore-forming micro-organisms which generate hydrogen gas during the exponential growth phase. In parallel, facultative anaerobes such as the species of genus E. coli and its modified strains and species of the genus Enterobacter, such as E. aerogenes and E. cloacae have also been used in hydrogen production. Extensive research has recently been carried out into hydrogen production at a high temperature, using thermophilic or hyperthermophilic bacteria.

The acidification of sugars under anaerobic conditions can yield high levels of hydrogen. The degradation of hexoses by mixed anaerobic microbial cultures has been extensively studied. It has been found that hydrogen and various metabolic products are produced, mainly volatile fatty acids (VFAs; acetic, propionic and butyric acid), lactic acid, and alcohols (butanol and ethanol), depending on the microbial species present and the prevailing conditions. The hydrogen yield may be correlated stoichiometrically with the final metabolic products by the principal reactions describing the individual processes of acidogenesis:

$\begin{array}{l}{\mathrm{C}}_6{\mathrm{H}}_{12}{\mathrm{O}}_6+2{\mathrm{H}}_2\mathrm{O}\to 2{\mathrm{CH}}_3\mathrm{COOH}+2{\mathrm{CO}}_2+4{\mathrm{H}}_2\hfill \\ {\mathrm{C}}_6{\mathrm{H}}_{12}{\mathrm{O}}_6\to {\mathrm{CH}}_3{\mathrm{CH}}_2\mathrm{COOH}+2{\mathrm{CO}}_2+2{\mathrm{H}}_2\hfill \\ {\mathrm{C}}_6{\mathrm{H}}_{12}{\mathrm{O}}_6+2{\mathrm{H}}_2\to 2{\mathrm{CH}}_3{\mathrm{CH}}_2\mathrm{COOH}+2{\mathrm{H}}_2\mathrm{O}\hfill \end{array}$

Hydrogen productivity (HP) is a parameter for assessing the yield of hydrogen production through dark fermentation. It is defined as the percentage of influent substrate electrons in the hydrogen gas produced in both gaseous and aqueous phases (Kraemer and Bagley, 2005). The above reactions show that the production of acetic and butyric acids leads to the simultaneous production of hydrogen, with the fermentation of hexose to acetic acid giving the highest theoretical yield of 4 mol of H₂/mol of hexose (HP = 33%). The conversion to butyric acid results in 2 mol of H₂/mol of hexose (HP = 17%), while during the production of propionic acid, hydrogen is consumed.

Conditions prevailing during dark fermentation should shift the metabolism of the microbial consortium towards acetate and/or butyrate production in order to achieve a high hydrogen yield. Clostridia sp. produces a mixture of acids, with butyrate exceeding acetate during the biological degradation of glucose (Mizuno et al., 2000). In practice, the production of more metabolic products (lactate or ethanol), accompanied by a negative or zero hydrogen yield, results in lower overall yields of hydrogen (HP: 10–20%). The shift of metabolism towards acetate may occur via different, non-hydrogen-yielding pathways. In mixed fermentation processes, the micro-organisms may select different pathways in converting sugars, as a response to changes in their environment (pH, sugar concentration, etc.). The absence or presence of hydrogen-consuming micro-organisms in the microbial consortium also affects the microbial metabolic balance and, consequently, the fermentation end products.

15.3.1 Anaerobic bio-reactors

The selection of reactor type is based mainly on the feedstock organic load and its solid matter content. Feedstocks with a high solid content are usually best converted to biogas in continuous stirred tank reactors (CSTRs) and plug flow reactors (PFR), while soluble feedstocks can be used in high-rate configurations. A brief discussion on the main bio-reactors which have been developed for biogas production follows.

There are several modifications of the above reactors, such as the UASB-filter reactor which combines UASB and anaerobic filter technology (Banu and Kaliappan, 2008) and the periodic anaerobic baffled reactor (PABR) which resembles a simple ABR which changes its flow pattern periodically due to the switching of influent and effluent points (Stamatelatou et al., 2009).

Apart from single-stage systems consisting of one of the main bio-reactor types described above, two or three stage systems are also widely used for the production of biogas from waste. When the waste contains a high portion of solids, the first stage serves as the hydrolytic-acidogenic step, while the second stage is the methanogenic step. The first stage requires a digester capable of handling high solid streams and operating at high retention times to enhance hydrolysis of the solids. The second stage can be a high-rate bio-reactor which further converts the first stage acidified effluent to biogas. Where the organic load of waste is mostly soluble and consists of carbohydrates, the first stage serves as the acidogenic step and takes place quickly, requiring a low hydraulic retention time. Both stages may be carried out in high-rate bio-reactors operating at different pH levels and different hydraulic retention times. The potential instability in single-stage systems, which is caused by the rapid acidification and volatile fatty acid accumulation, is therefore avoided. The different retention times required for the efficient operation of both stages determine the size of each bio-reactor. Depending on the requirements, the first bio-reactor may be larger or smaller than the second one.

Different temperature levels can be imposed on the two stages. Temperature phased systems are applied when a degree of hygienation of the effluent is required, especially in the anaerobic digestion of sewage sludge, manure, slaughterhouse wastewater and any other feedstock which may contain pathogens. The first stage is usually thermophilic, which inactivates the majority of pathogens (Song et al., 2004).

The two-stage configuration concept is also applied in the case of biohydrogen production. The acidogenesis step can be regulated to yield hydrogen along with volatile fatty acids and other metabolites. In this process, elimination of hydrogen scavenging micro-organisms, pH control and regulation of the substrate and product (hydrogen) concentration in the mixed liquor are the key operating factors directing the hydrogen efficiency of the first stage, as discussed in Section 15.3.2. The second stage is a typical methanogenic bio-reactor fed on the acidified effluent of the first stage.

15.3.2 Factors affecting biogas and biohydrogen technology

The composite nature of an anaerobic consortium, consisting of micro-organisms grown at different rates and optimal conditions, results in shifts in the balance of the population under the effect of environmental changes. It has been widely recognised that the most important factors affecting the anaerobic digestion process which may lead to failure, are the pH, the temperature, the presence of toxic or inhibitory substances and the hydraulic retention time (HRT).

The pH determines the concentration of weak acids and bases in undissociated form. These molecules can be transferred through the cellular membrane, thus changing the intracellular pH. The pH also influences the function of the extracellular enzymes which affect the hydrolysis rate. Biogas production takes place at neutral pH, although methane production is possible at lower or higher pH values. pH also influences the activities of hydrogen-producing micro-organisms, as it directly affects hydrogenase activity as well as the metabolic pathway followed. However, a wide range of pH values has been proposed as optimum for fermentative hydrogen production from different feedstocks. The pH range of 5–7.5 (Fang et al., 2002a; Calli et al., 2008) is usually reported as optimum, even though lower or higher pH values such as pH of 4.5 (Ren et al., 2007) and 9.0 (Lee et al., 2002) have also been proposed as optimal.

The temperature influences enzyme and consequently microbial activity. Generally, the higher the temperature, the higher are the microbial reproduction and substrate conversion rates. Because of the faster rates, the inhibition effects become more severe as the inhibitory factors evolve. The micro-organisms are categorised into three types according to the temperature range in which they grow: thermophilic (optimum above 50°C), mesophilic (optimum 30–40°C) and psychrophilic (optimum below 20°C). The appropriate adaptation of micro-organisms to temperature changes may allow them to function in more than one temperature range.

Elements or compounds found to be inhibitory and even toxic to anaerobic digestion are:

The hydraulic retention time (HRT), in combination with the concentration of feedstock in the influent of bio-reactors, affects the methane and hydrogen yield. In the case of biogas production, the selection of HRT and organic load are dependent upon the bio-reactor type used (Section 15.3.1). For hydrogen production from pure substrates such as glucose and sucrose, the most common values of HRT are in the range of 3–8 hours, with the lowest being 1 hour (Chang et al., 2002) and the highest 13.7 hours (Fang and Liu, 2004). For more complex substrates such as starch, an HRT of 15 or 17 hours is suggested as being necessary, due to the slow initial hydrolysis step (Hussy et al., 2003).

Hydrogen partial pressure is also important (as discussed in Section 15.2). Under stable biogas production conditions, it is sustained below 10⁻⁴ atm. In bio-reactors designed for biohydrogen production, the hydrogen levels are higher, but at levels higher than 60–100 Pa, hydrogen production is inhibited. Several methods of keeping the hydrogen partial pressure low have been studied. Mizuno et al. (2000) showed that sparging with nitrogen enhanced the hydrogen yield, while Voolapalli and Stuckey (1998) used a submerged silicone membrane dissolved gas extraction system, removing hydrogen and carbon dioxide from the reactor volume. Another potentially efficient method for removing hydrogen from the gas stream based on a heated palladium-silver membrane reactor has been proposed by Nielsen et al. (2001).

The inoculum added to the bio-reactor for biohydrogen production is of the greatest importance. When mixed microbial cultures are used for hydrogen production by dark fermentation, hydrogen-consuming bacterial activity should be suppressed, while the activity of the hydrogen-producing bacteria must be preserved. The most common practice is to pretreat the microbial biomass used to inoculate the bio-reactors. The pretreatment method relies on the spore-forming characteristics of hydrogen-producing Clostridium, which is ubiquitous in anaerobic sludge and sediment (Brock et al., 1994). When an anaerobic sludge is treated under harsh conditions, Clostridium is more likely to survive than the non-spore-forming bacteria, many of which are hydrogen consumers (Lay, 2001). Effective pretreatment processes include heating (100°C, 15 minutes), acidic or basic treatment (pH = 3, adjusted with ortho-phosphoric acid, 24 hours), aeration, chemical addition (chloroform, acetylene), and the application of an electric current (3–4.5 V). Where real biomass is used as a feedstock (wastewater, crops, etc.), it is advantageous to use the indigenous mixed microbial culture of the feedstock by applying operational conditions as proposed by Antonopoulou et al. (2008a, 2008b).

In general, the bio-reactors described in Section 15.3.1 can be utilised for fermentative hydrogen production in either batch or continuous mode. Batch mode operation is more suitable for research purposes, but at full scale, bio-reactors should operate on a continuous, or at least semi-continuous (fed or sequencing batch) basis. The continuous stirred tank reactor (CSTR) is the most common configuration, offering simple construction, ease of operation and effective homogeneous mixing as well as temperature and pH control. However, the solid retention time (SRT) is the same as the hydraulic retention time (HRT), resulting in a low concentration of microbial biomass and a low hydrogen production rate. The hydrogen-producing biomass in a CSTR could be self-granulated or flocculated under appropriate conditions (Fang et al., 2002b; Zhang et al., 2004). Another approach to increasing biomass concentration in a CSTR is to immobilise the biomass in bio-films or artificial granules made from a variety of support materials.

Another category of continuous flow reactors are those which permit the physical retention of the microbial biomass through flocculation, the formation of granules of self-immobilised microbes, microbial immobilisation on inert materials, microbial-based bio-films or retentive membranes (Hallenbeck and Ghosh, 2009). However, a potential problem posed by these types of reactors is the establishment of slow-growing methanogenic populations, due to extended retention of the biomass inside the reactor.

15.3.3 Pretreatment methods

Feedstocks with a high organic matter content that are difficult to degrade anaerobically generally contain lignocelluloses. Particular emphasis is therefore given to methods developed for enhancing biogas and hydrogen production from lignocellulosic feedstocks.

Lignocellulosic biomass contains complex carbohydrate polymers such as cellulose and hemicellulose, which are tightly bonded to lignin, the most recalcitrant component of plant cell walls. The higher the proportion of lignin, the higher the resistance to chemical and enzymatic/biological degradation. Softwoods usually contain more lignin than hardwoods and agricultural residues. However, regardless of the origin of the lignocellulosic biomass, some kind of pretreatment process is always necessary to break the lignin matrix and to de-polymerise the cellulose and hemicellulose. This facilitates the release of simple sugars (hexoses and pentoses) and results in a higher biofuel production yield. Pretreatment methods for lignocellulosic biomass can be divided into three main types: physical, chemical or physico-chemical and biological. It should be noted that most pretreatment methods have been limited to the experimental level as techno-economic evaluation is required to assess their cost in comparison to any resultant increase in energy gain.

Physical or mechanical pretreatment refers to milling which reduces the size of the particulate matter. The specific surface area of the solids and size of pores are therefore increased, the crystallinity and degree of polymerisation of the cellulose are decreased, and enzymes can more easily access the substrate to initiate hydrolysis. Mechanical pretreatment is always applied before any other kind of pretreatment. Delgenes et al. (2002) demonstrated that mechanical pretreatment by milling enhances methane production from 5 to 25%, while Menardo et al. (2012) showed that by reducing the size of wheat, barley, rice straw and maize stalks, the methane yields were increased by more than 80%.

During chemical or physico-chemical pretreatment, the lignocellulosic biomass is exposed to chemicals (acids, alkali or solvents) at ambient or higher temperature. The main effect is to alter the lignin structure and to dissolve the hemicelluloses. Several reviews have focused on this necessity and on the comparative study of these methods (Chandra et al., 2007; Hendriks and Zeeman, 2009).

Acid pretreatment may be performed with acids such as H₂SO₄, H₃PO₄, HNO₃ and HCl. During the pretreatment, the cellulose crystallinity is reduced, hemicelluloses are hydrolysed and furfural and hydoxymethyl furfural (HMF) are produced. Furfural and HMF are toxic to methanogens and hydrogen-producing bacteria (Gossett et al., 1982; Ramos, 2003; Vazquez et al., 2007). For this reason, the conditions for acid pretreatment are carefully studied. Dilute acids are preferred and used with short retention times (e.g., 5 min) at high temperature (e.g., 180°C) or relatively long retention times (e.g., 30–90 min) at lower temperatures (e.g., 120°C) (Table 15.1).

Alkali pretreatment involves the use of alkaline solutions such as NaOH, Ca(OH)₂ (lime) or ammonia to remove lignin and a part of the hemicelluloses by destroying the links of lignin and other polymers. Alkaline pretreatment can be generally classified into high and low concentration processes depending upon the concentration of the alkali (Mirahmadi et al., 2010). Table 15.2 lists some recent studies on alkaline pretreatment and its effect.

Thermal pretreatment takes place at high temperatures (from 150°C to 220°C). At temperatures above 160°C, the hemicelluloses are solubilised first, followed by lignin. Phenolics are released from the lignin solubilisation, some of which may be toxic to the micro-organisms (Gossett et al., 1982; Ramos, 2003). Compounds such as vanillin, furfural and HMF, which have a toxic effect on micro-organisms, may arise from the hydrolysis of hemicellulose. The degree of de-polymerisation and formation of inhibitory compounds depend significantly on conditions such as temperature. The thermal treatment technologies which have a positive effect on methane and/or hydrogen yield when applied to biomass are: steam explosion, liquid hot water and wet oxidation.

During steam explosion, lignocellulosic biomass is treated with high-pressure saturated steam for a short period of time, followed by rapid pressure relief. The temperature level of steam explosion is typically between 160 and 260°C and the duration of the process varies between a few seconds and several minutes (Sun and Cheng, 2002). The process has been thoroughly studied and applied at the laboratory and pilot scales. It is considered to be cost effective as a relatively low level of energy is required (Holtzapple et al., 1989).

Steam explosion has been studied in various biomass types. In plant biomass, increases in methane yield of 20% from wheat straw (Bauer et al., 2009) and up to 50% from salix (Estevez et al., 2012) have been recorded. The combination of steam explosion with chemical agents such as acids or bases has also been tested and shown positive results (Bruni et al., 2010; Teghammar et al., 2010). For example, the pretreatment of corn stover with steam explosion combined with acid resulted in high hydrogen yields of 3.0 mol mol hexose⁻¹, although steam explosion alone was also efficient (hydrogen yield of 2.84 mole mol hexose^− 1) (Datar et al., 2007).

Liquid hot water (LHW) or hydrothermolysis involves the contact of biomass with water under high pressure and temperature (200–230°C) for around 15 minutes. It has been applied to various biomass types such as corn stover (Zeng et al., 2007), food wastes and manure (Qiao et al., 2011).

In wet oxidation, the biomass is treated with water and air (or oxygen) at temperatures above 120°C (e.g., 148–200°C) for a period of around 30 min (Palonen et al., 2004; Schmidt and Thomsen, 1998). Lissens et al. (2004) used wet oxidation to improve anaerobic biodegradability and methane yields in several raw wastes such as food waste, yard waste and digested bio-waste treated in a full-scale biogas plant. The wet oxidation process increased methane yields by approximately 35–70% when compared to untreated raw lignocellulosic wastes.

Biological pretreatment refers to the use of whole micro-organisms or purified enzymes to disrupt the lignocellulosic matrix and enhance hydrolysis. Both fungi (brown, white and soft-rot fungi) and bacteria have so far been tested for the delignification of lignocellulosic biomass. Whiterot fungi such as Phanerochaete chrysosporium, Trametes versicolor, Ceriporiopsis subvermispora, and Pleurotus ostreatus are among the most effective micro-organisms for the biological pretreatment of lignocelluloses (Sun and Cheng, 2002).

Purified enzymes (ligninases such as laccase, lignin peroxidase and manganese peroxidase, cellulases or hemicellulases) have also been tested for lignin, cellulose or hemicellulose breakdown and solubilisation. Enzymatic pretreatment is usually combined with some physico-chemical method. For example, Talebnia et al. (2010) reported glucose yields of 98% after the enzymatic hydrolysis of wheat straw with thermal and chemical pretreatment. Other cases of biological pretreatment are reported in Table 15.3. The main advantages of biological pretreatment using whole micro-organisms are the low energy requirements, little (if any) chemical addition, and mild environmental conditions (low temperature and pressure). However, the main disadvantage is a slow conversion rate. Where purified enzymes are used in biological pretreatment, their high cost is the main disadvantage.

15.4.1 Energy crops

Energy crops consist of plants cultivated for biofuel production. Some of the crops used for biogas production, their harvesting, methane and electricity yields are reported in Table 15.4 as estimated by the FNR (2012). Energy crops may also be used as a co-substrate in co-digestion systems in combinations with feedstocks with a low energy content (e.g., manure).

Other biofuels, such as hydrogen, ethanol and biodiesel, can be produced from energy crops. Methane and hydrogen yields should be related to the lignin, cellulose and hemicelluloses content of the plants as lignin is considered to be resistant to degradation. Table 15.5 summarises the yields of methane and hydrogen from various energy crops and the yields of other biofuels are cited where available.

15.4.2 Crop residues

Crop residues include the biomass remaining after harvesting and the extraction of sugar, cereal, oil or any other material which cannot be further utilised in the food production chain. Examples of crop residues include sugar cane and sweet sorghum bagasse, corn leaves and stover, wheat straw, and forestry residues (hardwoods) such as wood trimmings or tree residues (Table 15.6). These are primarily of a lignocellulosic nature and are considered to be an abundant renewable source.

Recent data have shown the total global sugarcane production in 2007 to be 1.59 billion metric tonnes, with an average productivity of about 67 t/ha. The major sugarcane producing countries are India, Brazil, Philippines, China, USA, Mexico, Indonesia, Australia, Colombia, Brazil and India, which together produce almost 60% of the world’s sugarcane, with Brazil responsible for around 35% of total world production (McLaren, 2009).

In 2009–2010, Japan was the main contributor to total global rice production, amounting to 9.74 million tonnes (www.rice-trade.com, 2011). Biofuels such as hydrogen and methane generated from these feedstocks are characterised as ‘second generation’ as they do not compete with the production of food.

15.4.3 Manure

A mixture of several heterogeneous materials is implied by this designation. They include faeces, urine, hair or feathers, food, wastewater from livestock and poultry units and bedding materials (straw, sand, wood chips, etc.). This type of waste is largely produced in the initial steps of the food chain and involves meat and its products.

Manure is the waste most commonly treated by anaerobic digestion. This is because the anaerobic digestion of manure combines the energy exploitation of its organic content with its stabilisation, potential hygienation (if treated under thermophilic conditions) and odour control. Other benefits of the anaerobic digestion of manure include the reduction of greenhouse gases (which are emitted if manure is spread untreated), the conversion of organic nitrogen to ammonium nitrogen, which can be utilised in agriculture, and the production of a well-stabilised solid material possessing excellent fertilising properties. The biogas produced can be used to contribute to heating requirements in commercially available heating engines (boilers, heaters, etc.). Biogas from manure has also been used to generate electricity (Cantrell et al., 2008).

There is no single practice for the anaerobic digestion of manure. Digester loading, solid content and composition of the manure will be affected by the capacity of the unit and its feeding, by the collection method (scraping or flushing) and the bedding materials. The typical composition of manure from various sources is shown in Table 15.7 and includes the main characteristics of solid and nutrient content. The presence of sand is problematic as its tendency to settle on the bottom of the digester causes an accumulation of inert material which reduces the operating volume. Sand should therefore be removed prior to anaerobic digestion via sedimentation.

The anaerobic digestion of manure requires long hydraulic retention times for the hydrolysis of solids consisting mainly of lignocellulose. The hydraulic retention time is typically in the range 15–30 days under mesophilic conditions and 10–20 days under thermophilic conditions in a CSTR (Angelidaki et al., 2011).

The solid content of the influent manure also influences the process efficiency. The type of collection, scraping or flushing, affects the solid content of the final waste. In cattle livestock facilities, flushing uses 100–200 gallons (380–750 L) of water per cow per day (Burke, 2001). This causes considerable dilution of the total solids from 120 g L^− 1 to 8–15 g L^− 1 and bio-reactors such as the CSTR or plug flow may be used. The efficiency of these reactors varies between 40 and 45% reduction of the volatile solids at solid loading rates ranging between 3 and 6 kg VS m^− 3 d^− 1, despite the wide spread of the data (Fig. 15.3). The conversion efficiency for swine manure is slightly higher (Fig. 15.4). Poultry manure is high in solid concentration (Table 15.7). Although high solid anaerobic digestion requires small volume digesters and results in high specific methane production rates, this is not the case with poultry manure due to the high nitrogen level and inhibitory ammonia production. There are few pilot or full-scale studies on poultry manure. Solares et al. (2006) reported a 49% maximum volatile solid reduction in a 40 m³ thermophilic anaerobic digester operated at an HRT of 10 days with an influent solid concentration of 5.5% (the recommended influent solid concentration ranges between 5 and 6%; Singh et al., 2010).

The anaerobic digestion of manure is more stable under mesophilic than thermophilic conditions, although the conversion rate and pathogen reduction are better at high temperatures. Ammonia inhibition has been found to be more effective under thermophilic conditions (Angelidaki and Ahring, 1994) and this jeopardises the stability of the process. The degree of inhibition depends on the acclimatation of anaerobic micro-organisms and on the pH. Inhibition is caused in the range of 1.5–3 g N L^− 1 at pH above 7.4, while ammonia concentrations higher than 3 g N L^− 1 have been found to be toxic regardless of the pH level (Calli et al., 2005). The tolerance of methanogens to ammonia may vary as a result of micro-organism acclimatation, which is a time-consuming process. The ammonia may be reduced by diluting the manure with water or other waste (co-digestion); however, this requires high volume digesters and other wastes for co-digestion may not be available.

Physico-chemical methods such as chemical precipitation, adsorption on zeolite and clay and biological methods (nitrification-denitrification, Anammox) for ammonia removal have been studied and appear to be effective at solid concentrations of 0.5–3% (Chen et al., 2008). Ammonia may also be removed by stripping with an inert gas or biogas in two-stage or single-stage systems (Abouelenien et al., 2010). The viability of these approaches is unknown as no economic analysis has been performed.

The most common parameter for assessing the methane potential of a feedstock is its chemical oxygen demand (COD) concentration. The maximum methane generation is expected to be 0.35 m³ kg COD^− 1 converted at 0°C and 1 atm. However, the breakdown of the COD content of manure into its constituents reveals that the quantities of lignin (non-biodegradable) and the slowly degradable carbohydrates are 30% in cattle manure and 20% in pig manure (Møller et al., 2004). Based on the chemical composition analysis of carbohydrates, lipids and lignin, etc., Møller et al. (2004) assessed the theoretical methane yield as (L kg^− 1 VS converted) for cattle manure (468 ± 6 L kg ^− 1 VS), pig manure (516 ± 11 L kg^− 1 VS) and sow manure (530 ± 6 L kg^− 1 VS). Hill (1984) estimated a similar value for the theoretical methane yield (500 L kg^− 1 VS) of all types of manure. The actual methane yield (L produced per kg of VS added in a batch bio-reactor until no further methane is produced) is much lower due to minimally biodegradable lignin bounded formulations (e.g., lignocelluloses) and the inhibitors present, such as the ammonia produced during protein degradation. The values determined for various manure types were 148 ± 41 L kg ^− 1 VS (cattle), 356 ± 28 L kg^− 1 VS (pig) and 275 ± 36 L kg^− 1 VS (sow). The difference between the various manure types is obvious; pig and sow manure have a much higher methane potential than cattle manure as they contain more proteins and lipids (yielding biogas richer in methane) and less lignocellulosic material.

The discrepancy between the theoretical and actual methane yields also holds for the particulate fraction of manure subjected to separation through centrifugation, evaporation, flocculation and de-watering processes. This is not the case for the liquid fraction (Møller et al., 2004), where the theoretical yield is close to 506 ± 25 L kg^− 1 VS (real yield). Although the actual ultimate methane yield of the solid fractions is lower than that of the non-separated manure (a significant portion of the easily biodegradable organic matter is contained in the liquid fraction), the methane produced per volume of the particulate fractions is 2.6–3.6 higher than the pig non-separated manure and 5–6.6 higher than the sow non-separated manure. This indicates that treating the manure fractions separately will result in higher methane yields and specific production rates, as smaller digesters can be used for the particulate fraction and high-rate digesters for the liquid fraction.

A two-phase system for treating both fractions separately could therefore be beneficial. The concept of two-phase configurations could also be applied in the case of high solid manure with the aim of separating hydrolysis-acidogenesis from methanogenesis. In this case, a hydrolysis-acidogenesis bio-reactor should provide a high solid retention time for enhancing the hydrolysis, which is considered to be the rate-limiting step. Myint and Nirmalakhandan (2009) developed a leach-bed bio-reactor and increased the volatile fatty acid production while maintaining a pH between 4 and 5, which is considered to be the optimal condition for particulate hydrolysis. The enrichment of the liquid fraction in COD is achieved through recirculation of the leachate in the same leach-bed bio-reactor. The liquid fraction generated from the leach-bed can be fed to a high-rate methanogenic bioreactor.

A two-stage system was adopted by Kaparaju et al. (2009) based on CSTRs in series. Each CSTR maintains the whole anaerobic population (there being no separation between hydrolysis-acidogenesis and methano-genesis) but operates at different organic loading rates. The effluent of the first CSTR is the influent of the subsequent apparatus. A higher retention time is therefore achieved for the portion of solids that undergoes short-circuiting in the first reactor and leaves it earlier than the hydraulic retention time. It was found that the distribution of the volume of one CSTR to two CSTRs in series at ratios of 70:30 and 50:50 brought about an increase in biogas production of 16.4–17.8% when compared to the one step CSTR.

Although the anaerobic digestion of manure in developed countries is well established, biogas still remains an expensive source of renewable energy because of the high investment cost. This is particularly significant in developing countries. The payback period for a biogas plant of 500 kW of electrical energy is five years when cattle manure is used and 8.5 years when a mixture of cattle and chicken manure is used. However, small-scale biogas plants have a shorter payback period of 8–10 months. It seems that the main obstacle to widespread implementation of biogas technology can be overcome through efficient and simple digester construction and ease of operation and maintenance. State subsidies and tariffs for biogas are also significant incentives for investment (Avcioğlu and Türker, 2012).

The solid residue which remains after anaerobic digestion is another valuable product and may be used as fertiliser due to its high nitrogen and phosphorous nutrient content. However, the presence of pathogens could reduce its fertilising value. Pathogens are micro-organisms (bacteria, viruses, protozoa, etc.) which cause diseases in humans or animals and are abundant in manure. If manure is applied on land without prior treatment, it may contaminate surface water as some pathogens can survive for a long period of time. Pathogens may be adequately reduced as shown in the reports of Bendixen (1999) who studied the effect of duration and temperature on the anaerobic digestion or sanitation process in reducing the faecal streptococci which were selected as indicator pathogens. For example, mesophilic anaerobic digestion (37°C) achieves a pathogen reduction of less than 2 log₁₀ units (that is, if the pathogens present in the feedstock are 10⁵/g dry weight, they are reduced to 10³/g dry weight during the process). Thermophilic anaerobic digestion (55°C) achieves a pathogen reduction of 3 to 5 log₁₀ units. When a sanitation step (at 55°C) was included prior to mesophilic anaerobic digestion, the pathogen reduction was 3.6 log₁₀ units.

15.4.4 Food wastes

Wastes from the food processing industry are a significant source of energy due to their high organic compound content (carbohydrates, proteins, lipids). Much research has been conducted into optimising biogas and hydrogen production from this heterogeneous category of wastes.

The key point in biogas production from food wastes is to develop simple bio-reactor configurations and to design the process to overcome problems originating from specific characteristics of the waste. For example, wastes rich in proteins, such as fish processing residues, dairy wastewater and slaughterhouse wastewater produce ammonia which can be inhibitory at high levels when subjected to anaerobic digestion, as discussed in Section 15.3.2. The presence of cations (such as sodium from sodium chloride) may also be problematic. As salts are added during food processing, the wastes may be saline or hypersaline. A concentration of sodium exceeding 10 g/L may be strongly inhibitory to methanogens (Kugelman and McCarty, 1965; Rinzema et al., 1998). As in the case of ammonia inhibition, the acclimatation of anaerobic micro-organisms to saline conditions may increase their tolerance to sodium (Omil et al., 1995; Gebauer, 2004).

Lipids are another important constituent of food wastes and yield biogas with a high methane content when converted under anaerobic conditions. Food wastes rich in lipids come from the dairy, meat, fishing and edible oil industries. The production of long-chain fatty acids which are inhibitory to anaerobic micro-organisms and are slow to biodegrade, presents a problem in the degradation of lipids. This type of inhibition has been studied and found to be due to mass transfer limitation rather than to changes in the metabolism of the micro-organisms. The long-chain fatty acids tend to adsorb on surfaces and microbial cellular membranes, so preventing the transfer of substrates and nutrient into the cell. However, when no long-chain fatty acids are present, the activity of the micro-organisms is enhanced, indicating that the inhibition effect has not altered their metabolism (Pereira et al., 2003).

The degradation of long-chain fatty acids takes place through b-oxidation by syntrophic hydrogen-producing bacteria which require hydrogen-consuming methanogenic bacteria to maintain hydrogen at low levels, thus making the conversion thermodynamically feasible. Acetogenesis is slow due to the low energy release of corresponding metabolic reactions which cause a low growth rate in long-chain fatty acid degraders.

Various physico-chemical or biological pretreatment methods have been proposed for removing or converting inhibitory compounds. Another means of decreasing toxicant concentration is to dilute wastewaters with tap water, treated effluent or wastewater devoid of these contaminants. For example, phenolic compounds present in olive mill wastewater (OMWW) can be diluted by mixing with cheese whey or manure. This last option, known as co-digestion, is the most effective, as mixing different types of wastewater results in a better balance of nutrients and alkalinity. Co-digestion may also make anaerobic digestion possible throughout the year, given the seasonal nature of the various agro-industrial wastewaters (Carrieri et al., 1986, 1992; Angelidaki and Ahring, 1997; Lyberatos et al., 1997; Gavala et al., 1999; Angelidaki et al., 2002; Marques et al., 1998; Dareioti et al., 2009, 2010). The main characteristics and ‘niches’ in the production of biogas from various food waste types are discussed in Table 15.8, along with some typical methane yields.

Hydrogen production from food wastes has recently become more widespread, especially in the case of wastes rich in carbohydrates. Table 15.9 lists some case studies and records the carbohydrate content of each feedstock. Carbohydrates form a large portion of the organic matter (as expressed by the COD content of these wastes) as seen in Table 15.9, an indicator implying the suitability of these feedstocks for hydrogen and biogas production in a two-phase process.