Potential biomass production depends on the accumulated amount of photosynthetically active radiation (SPAR) intercepted by the crop over the course of its growth and the efficiency with which the crop is able to convert this radiation into carbohydrates, which is known as the radiation use efficiency or RUE [31]. The contribution of each of these processes to biomass production is crop specific [32]. As a C4 plant, M. x giganteus has a naturally high RUE, which has been assessed at 4.09 [21], just short of its theoretical maximum of 4.6 gDM MJ-1 [33]. Water deficit has been shown to reduce the RUE of M. x giganteus by 30-80% [27] with M. x giganteus being more sensitive to drought than Micanthus sinensis [34].

Water use efficiency (WUE) of M. x giganteus has been shown to be higher in the United Kingdom and France than in the Mediterranean environment [27], with adult stands reaching between 9.1 and 9.5 gDM l-1 in the United Kingdom [35] and between 6 and 10 gDM l-1 in France [36].

M. x giganteus exhibits high nitrogen use efficiency (NUE) with 200 g g-1 determined in February for aboveground biomass and 180 g g-1 for the total crop including the annual increase in rhizome mass [37]. Minimum nitrogen content in the belowground biomass of a mature Miscanthus x giganteus crop ranged from 70 to 370 kgN ha-1 depending on harvest date and nitrogen treatments [28]. The transfer of nitrogen from the belowground biomass to the aboveground biomass at the beginning of growth can account for as much as 79% of the total nitrogen content of the belowground biomass [28] (Figure 4.3). Nutrient accumulation in the aboveground biomass peaks during late summer.

During crop senescence in autumn, nutrients are again remobilized but this time from the aboveground biomass to the belowground biomass (i. e. autumn remobilization). Strullu et al. [28] found that 42% of the maximum nitrogen content of the aerial organs of Mis- canthus x giganteus was relocated to the belowground biomass by an October harvest compared to 71% by a February harvest. It would appear, therefore, that only a small

proportion of plant nutrients are harvested during the winter harvest with the majority of nutrients being translocated to the rhizome and recycled to the soil.

The following step is the pretreatment of the fractionated material. The main goal of pretreatment is to overcome this lignocellulosic recalcitrance, to separate the cellulose from the matrix polymers, and to make it more accessible for enzymatic hydrolysis. Reports have shown that pretreatment can improve sugar yields to greater than 90% theoretical yield for biomass such as wood, grasses, and corn [8,9]. Pretreatment technologies for lignocellulosic biomass include thermal, (thermo)chemical, physical and biological methods or various combinations thereof [5, 9].

In general, pretreatment processes produce a solid pretreated biomass residue that is more amenable to enzymatic hydrolysis by cellulases and related enzymes than native biomass. Many pretreatment approaches, such as dilute acid and steam/pressurized hot water based methods, seek to achieve this by hydrolyzing a significant amount of the hemicellulose fraction of biomass and recovering the resulting soluble monomeric and/or oligomeric sugars. Other pretreatment processes, such as alkaline-based methods, are generally more effective at solubilizing a greater fraction of lignin while leaving behind much of the hemicellulose in an insoluble, polymeric form [10]. Most pretreatment approaches do not hydrolyze significant amounts of the cellulose fraction of biomass but enable more efficient enzymatic hydrolysis of the cellulose by removal of the surrounding hemicellulose and/or lignin along with modification of the cellulose microfibril structure [11]. Biological pretreatment uses microorganisms and their enzymes selectively for delignification of lignocellulosic residues and has the advantages of a low energy demand, minimal waste production and a lack of environmental effects [7, 12, 13]. It has been suggested that there will probably not be a general pretreatment procedure and that different raw materials will require different pretreatments [10]. Table 2.1 gives an overview of the different pretreatment technologies.

The choice of the optimum pretreatment process depends very much on the objective of the biomass pretreatment, its economic assessment and environmental impact. Tech­nological factors, such as energy balance, solvent recycling and corrosion, as well as environmental factors, such as wastewater treatment, should all be considered carefully when selecting a method [5]. Diverse advantages have been reported for most of the pre­treatment methods, which make them interesting for industrial applications. Only a small number of pretreatment methods has been reported as being potentially cost effective thus far. These include steam explosion, liquid hot water, dilute acid pretreatments, lime, and ammonia pretreatments [11, 16, 18,19]. The complete depolymerization of these renew­able feedstock in a cost-effective manner with minimal formation of degradation products represents a significant challenge for microbiologists and chemical engineers. Obstacles in the existing pretreatment processes include insufficient separation of cellulose and lignin (which reduces the effectiveness of subsequent enzymatic cellulose hydrolysis), formation of by-products that inhibit microbial growth and fermentation (e. g. acetic acid from hemi — cellulose, furans from sugars and phenolic compounds from the lignin fraction [20]), high use of chemicals and/or energy, and considerable waste production. Research is focused on converting biomass into its constituents in a market competitive and environmentally sustainable way [21].

Jacob J. Jacobson and Kara G. Cafferty

Idaho National Laboratory, U. S.A.

2.2 Introduction

In 1978, the United States enacted the Public Utilities Regulatory Policy Act (PURPA) giving small electricity producers (less than 80 MW) a natural monopoly by requiring electric utilities to purchase the small companies’ surplus electricity at a price equal to the cost the utility would have incurred by producing the electricity themselves. As a result, biopower experienced a threefold increase in grid-connected capacity, created 66 000 jobs, and had an industrial investment of $15 billion dollars during the next decade. Despite these historic advancements, biopower has not experienced further substantial growth. Currently, avoidance costs from electric utilities remain low due to the vast supply of natural gas and innovations in natural gas turbines. As a result, it is difficult for renewable fuels to compete and developments in renewable energy technology have slowed (Figure 3.1). However, interest in environmental sustainability has caused some state governments to implement renewable portfolio standards (RPSs) requiring that a minimum amount of renewable energy (wind, solar, biomass, or geothermal) be included in the electricity generation portfolio of each state. As of February 2012,30 states and the District of Columbia have enforceable RPS programs [2]. Despite these regulations, biomass makes up a small portion of the current power industry because of high biomass feedstock costs and low overall efficiencies.

Biopower produces less than 2% of the total electricity in the United States [1], while coal and natural gas supply over 65%; hydropower and nuclear make up the remainder. The primary reasons biopower contributes such a small percentage of the overall electricity production are the size and efficiency of its plants. The average size of biopower plants is

Cellulosic Energy Cropping Systems, First Edition. Edited by Douglas L. Karlen. © 2014 John Wiley & Sons, Ltd. Published 2014 by John Wiley & Sons, Ltd.

20 MW (maximum 75 MW); while a typical coal plant ranges from 100 to 250 MW. The small plant sizes (which lead to higher capital cost per kilowatt-hour of power produced) and low efficiencies (which increase sensitivity to fluctuation in feedstock price) have led to electricity costs of 8-12 c/kWh [3]. Therefore, for biopower to increase its contribution to the U. S. energy supply, plant size and efficiency must increase to be competitive with current fossil fuel technologies. Additionally, as biopower increases the demand for biomass supply will tend to increase the price of biomass. For the biopower industry to continue to expand the biomass must remain cost competitive.

Biomass is a desirable source of energy because it is renewable, sustainable, widely available throughout the world, and amenable to conversion. Biomass is composed of cel­lulose, hemicellulose, and lignin components. Cellulose is generally the dominant fraction, representing about 40-50% of the material by weight, with hemicellulose representing 20-50% of the material, and lignin making up the remaining portion [4-6]. Although the outward appearance of the various forms of cellulosic biomass, such as wood, grass, munic­ipal solid waste (MSW), or agricultural residues, is different, all of these materials have a similar cellulosic composition. Elementally, however, biomass varies considerably, thereby presenting technical challenges at virtually every phase of its conversion to useful energy forms and products.

Despite the variances among cellulosic sources, there are a variety of technologies for converting biomass into energy. These technologies are generally divided into two groups: biochemical (biological-based) and thermochemical (heat-based) conversion pro­cesses. Although there are specific technologies within each of these general categories, biochemical conversion technologies (i. e., enzymatic hydrolysis), generally operate on wet feedstocks with a high carbohydrate content at the time of conversion [7]. In contrast, ther­mochemical conversion processes (e. g., combustion, gasification, and pyrolysis), generally require a dry feedstock, low in ash content, and having a small, consistent particle size [8,9]. As a result of these generalizations, herbaceous feedstocks that are naturally higher in ash and carbohydrates are generally allotted to biochemical conversion, while woody

feedstocks with their lower ash content are directed to thermochemical conversion. Ther­mochemical conversion is more aptly used to create heat and electricity due to destruction of chemical bonds while biochemical processes are more suited to develop liquid fuels. With the exception of anaerobic digestion, biochemical conversion is not discussed in this chapter. The main thermochemical processes under which biomass can be converted into energy include:

• Combustion

• Gasification

• Pyrolysis

• Hydrothermal Liquefaction.

In general, the specific thermochemical process being used is determined by the operating air supply and temperature conditions. Combustion occurs in the presence of excess oxygen, gasification takes place when the quantity of oxygen is insufficient for stoichiometric requirements, and pyrolysis happens in the complete absence of air. As a result, gasification can actually be characterized as an intermediate between combustion and pyrolysis; an alternative between having an over-sufficient oxygen supply to biomass and its absolute absence from the process. The operating conditions required for the main thermochemical conversion processes are summarized in Table 3.1.

2.3 Combustion

Combustion or burning is the most common means of converting biomass to usable energy and heat. Historically, woody biomass, from timber harvesting, sawmills, and pulp and paper production, has been used to generate electricity and heat at co-located, direct-fired boilers. Agricultural residue, primarily from wheat and corn harvests, has also contributed to biopower production. These practices have grown the biopower industry into the third largest generator of renewable electricity in the nation, providing 12% of the United States’ renewable generation capacity in 2010 [11]. There are two mainstream methods of com­bustion, direct combustion and co-fired combustion.

The choice of biomass feedstock is a critical driver in determining key performance metrics of bioenergy — including economic viability, scale of production (both at individual facilities and in aggregate), and environmental impact. For commodities such as fuels or electricity, feedstock cost typically represents two-thirds of the product cost, or more [26]; therefore, selecting a cost-effective feedstock is essential. As is discussed in Part IV of this book, the logistics of growing, harvesting, storing, and transporting biomass — unique for a given feedstock type — affects the feasible size of the processing facility, which, in turn, impacts the overall sector scale. Each feedstock also has a particular set of environmental attributes — for example, water use, wildlife habitat, soil quality, and so on — that significantly affects the environmental performance of the bioenergy system.

In assessing the suitability of a biomass feedstock for a given conversion process, several material properties are important to consider, including: (1) moisture content; (2) energy density; (3) fixed carbon/volatile matter ratio; (4) ash content; (5) alkali metal content; and (6) carbohydrate/lignin ratio. The first five properties are especially important in thermochemical processing. For biological conversion, the first and last properties are of primary concern.

Only a few Miscanthus species — M. x giganteus, M. sinensis and M. sacchariflorus — have been investigated regarding biomass productivity and composition for breeding potential.

Miscanthus x giganteus, in particular, has demonstrated high productivity [39] in low input systems and a higher energy output:input ratio than maize [3]. However, Miscant­hus x giganteus has a narrow genetic diversity [16] and is not adapted to all climatic zones [40, 41]. It is crucial that the genetic diversity of the Miscanthus genus is investigated to determine if varieties suited to a broader range of environments can be developed.

Biomass yields increase each year in young Miscanthus plants, reaching a plateau after 2-5 years in M. x giganteus, depending on environmental conditions [3]. M. x giganteus and M. sacchariflorus took less time to reach a yield plateau than M. sinensis hybrids and M. sinensis genotypes [42]. During each growing season, yields peak during flowering [27] and then decline through the winter partly due to leaf loss [24] (Figure 4.4). In addition to time of harvest, biomass yield is influenced by environmental conditions [43] and genotype [24]. M. x giganteus and M. floridulus achieved higher aboveground biomass yields than M. sinensis and M. sacchariflorus in field trials in northern France [26]. In this study ploidy levels appeared to influence biomass production, with triploid and tetraploid forms of M. sinensis and M. sacchariflorus more productive than diploid forms. Zub et al. [26] also showed that plant height and shoot diameter were traits that contributed the most to biomass yield, regardless of harvest date or crop age.

Hydrolysis is the process by which water splits a larger molecule into two smaller molecules. In the case of the hydrolysis of polysaccharides to soluble sugars this is called “saccharifica­tion”. The goal of this process is the de-polymerization of cellulose and hemicelluloses into soluble monomer sugars (hexoses and pentoses). This can be accomplished by two different processes: (1) acid hydrolysis with a variety of low acid-high temperature or high acid-low temperature conditions being suitable to both the breakdown of the structure of the biomass and the release of free sugars, and (2) enzymatic hydrolysis after some sort of pretreat­ment which allows enzymatic attack of the polymers [22, 23]. The C6 dominated cellulose

can be enzymatically hydrolyzed by cellulases; for the C5 dominated hemicellulose the hemicellulases (such as xylanase) can be used.

Acid (sulfuric or hydrochloric) can serve both for disruption and hydrolysis of the cellulosic polymers and is currently seen as the most technologically mature method of sugar release from biomass. A major disadvantage of acid hydrolysis is the potential degradation of the released monosaccharides that leads to reduced sugar yields [13, 23]. Other drawbacks are the cost of acid, the requirement to neutralize the acid after treatment and the production of inhibitory by-products such as furfural and hydroxymetyl furfural [22, 24, 25].

Enzymatic degradation of lignocellulosic biomass on the other hand is very specific and side reactions, such as degeneration of sugars, do not occur. High yields are therefore possible. In addition, the mild conversion conditions lower the maintenance costs of the production plant [23]. High temperature and low pH tolerant enzymes are preferred for the hydrolysis due to the fact that most current pretreatment strategies rely on acid and heat [26]. In addition, thermostable enzymes have several advantages, including higher specific activity and higher stability, which improve the overall hydrolytic performance [27]. Ultimately, improvement in catalytic efficiencies of enzymes will reduce the cost of hydrolysis by enabling lower enzyme dosages [7].

Although acid hydrolysis methods have long industrial histories and are, therefore, more mature, enzymatic hydrolysis is seen as the most economically promising method for redu­cing costs while improving yields and a key to cost-effective production of monosaccha­rides [28]. Research is focusing on advanced screening processes of natural enzymes and developed man-made enzymes to increase the efficiency and improve enzymatic hydrolysis [29].

Direct-fired combustion is the most common technology currently used to generate elec­tricity. Direct combustion typically burns biomass in a boiler to produce steam that is

then used to power turbines to produce electricity (Figure 3.2). Direct-flred systems are typically smaller in size, 10-50 MW, due to the feedstock requirements and draw radius limitations [11]. In some cases, steam is released from the turbine at medium pressures and temperatures and is used for process heating and cooling to help improve the economics.

Biomass moisture content is defined as the amount of water in the biomass expressed as a percentage of the material’s weight; reporting on a wet basis is most common. Moisture content at harvest for woody feedstocks is usually 40-60% (wet basis); for herbaceous crops, it typically ranges from 10 to 70% (wet basis) depending upon the species, climate, geographic location, and stage of maturation. Biomass net energy density per unit mass decreases with increasing moisture content. Transport efficiency of biomass feedstock, therefore, decreases as moisture content increases. Storage of high-moisture biomass is also less efficient, both because of reduced energy density and increased probability of biological degradation, fire risk, and mold formation. Moisture content also affects down­stream processing, especially for thermochemical conversion. High-moisture feedstocks must be dried to levels of less than 50% for conventional combustion and less than 20% for gasification and pyrolysis. In biological processing for which some form of thermal pretreatment is used, moisture content can also significantly affect the energy efficiency of the process.

To date, Miscanthus biomass has been mainly used to generate renewable heat, electricity and combined heat and power. Its use in biofuel production is under investigation in several countries, as is its potential as a component in bio-based concrete materials and bio-based plastic composites (Section 4.4.1, Past and Current Projects). The composition of Miscanthus biomass must be optimized to suit to each end use (Table 4.1). However, the biomass composition of Miscanthus species varies widely and it is critical that new Miscanthus varieties are developed to provide consistent biomass compositions suited to specific industrial processes.

For biomass combustion, it is essential that the moisture, ash and mineral content of the biomass are minimized, as these reduce process efficiency [1, 43]. M. sinensis genotypes have a higher combustion quality because they contain lower contents of chlorine and potassium than M. x giganteus [55]. Delaying harvest from autumn until late winter can also improve the combustion quality of Miscanthus biomass because moisture, ash, potas­sium, chlorine and nitrogen contents are lowest at this time [43, 56]. Ranges of the main biomass components are illustrated for M. x giganteus (Table 4.1). The differing mineral and ash content of Miscanthus leaves and stems between clones or harvest dates provides an opportunity to manipulate biomass composition better suited to combustion [1, 57, 58].

Efficient biofuel production requires high levels of cellulose and hemicellulose and low lignin content. From biomass components, the cellulose shows the highest content at 41­52% of biomass dry matter (Figure 4.5), the hemicellulose displays 24-34% and the lignin varies from 8.8 to 12.6% [40, 47, 53, 56, 59,60]. Cellulose, hemicellulose and lignin

contents tend to increase between autumn and late winter in Miscanthus species. Between species, M. x giganteus and M. sacchariflorus species globally have higher cellulose and lignin contents and lower hemicellulose content than M. sinensis species (Figure 4.5). One exception to this is the work by Lygin et al. [61], who found that the M. sinensis “Grosse Fontaine” clone had higher cellulose content than M. x giganteus.

In summary, it would appear there is sufficient genetic variation in biomass productivity and biomass composition to breed Miscanthus varieties well suited to bioenergy use.

The released sugars can now be converted into a broad spectrum of biochemicals and biofu­els through fermentation. An enormous variety of microorganisms, such as yeasts, bacteria, or fungi, exist that can be added to the mixture of free sugars to be fermented into advanced biochemicals, including biofuels. Although organisms exist to break down virtually any organic material, six-carbon sugars, and especially glucose, are widely available in the plant and animal world. Hence, there is more experience fermenting six-carbon sugars (as present in cellulose) than the five-carbon sugars (as present in hemicellulose), but both are valuable fermentation feedstock, especially with recent advances in fermenting five-carbon sugars. Cost-effective processes will require the rapid, complete and simultaneous fermentation of all sugars. Therefore, new developments are focusing on optimizing the biochemical conversion pathway by integrating several processing steps. In the Simultaneous Saccha­rification and Fermentation process (SSF), cellulose hydrolysis and C6 fermentation are performed in one step. In the Simultaneous Saccharification and Cocurrent Fermentation process (SSCF), cellulose hydrolysis and the fermentation of both C5 and C6 sugars is performed. The ultimate step is the Consolidated BioProcessing (CBP), combining C5 and C6 hydrolysis and fermentation in one single process step (Figure 2.2).

Challenges faced are the inhibition of the yeast by the end-product, so lowering the yield, high distillation cost, formation of un-productive by-products such as acetates or furfural that cause inhibition of the fermentation process. In addition, hydrolysates of lignocellulose contain compounds that are inhibitory to most microorganisms. Tolerance to harsh environments, including elevated temperatures, high salt, and low pH, will be essential. Currently available strains are severely limited in pentose utilization and exhibit poor hydrolysate tolerance. New genetically modified microorganisms are being developed, designed in such a way that they are able to ferment different sugars, get round inhibition or tolerate harsh environments, as such leading to a higher overall yield [30].