In contrast to sorghum, there is practically no public germplasm collection for Miscanthus, apart from a few collection activities by botanical gardens, research institutes and private companies. These collections comprise ornamental accessions and a small number of landraces [125].

In Europe, the private German nursery company TINPLANT has developed a large Mis­canthus collection of about 1050 accessions that includes several species: M. x giganteus, M. sinensis, M. sacchariflorus, M. tinctorius, M. condensatus, M. floridulus and some hybrids [126]. More recently, the French company Aelred has been developing a breeding program since 2010.

The Royal Botanic Gardens, Kew, in the United Kingdom has a collection of 125 plants of the genus Miscanthus and various associated species. Trinity College Dublin Botanic Gardens has a collection of Miscanthus accessions assembled from various national and international projects. Most of the material obtained by Trinity College Dublin was donated by various botanic gardens and research institutes throughout the world. The Agriculture Development and Advisory Service (ADAS) in the United Kingdom also holds a collection of Miscanthus.

Among research institutes, the Department of Agroecology and Environment at Den­mark’s Aarhus University holds a large collection of Miscanthus genotypes collected in part from Japan and established as part of the EU-projects EMI and BIOMIS. At the Institute of Biological, Environmental and Rural Sciences (IBERS) in Aberystwyth, UK, a Miscanthus breeding program that began in 2004, with germplasm assembled from European collec­tions based at academic institutions, horticultural companies and from Asia [127]. The collection includes accessions of M. sinensis, M. sacchariflorus and naturally occurring inter-specific hybrids and constitutes the largest Miscanthus collection outside Asia [128].

Miscanthus breeding and research activities at Wageningen UR Plant Breeding in The Netherlands and at INRA in Estrees-Mons, France, are focused on the development of new varieties suitable to bioenergy production and specific environments.

In Asia, the Institute of Botany from the Chinese Academy of Sciences in Beijing holds a large Miscanthus collection [9]. In 2010, the Institute developed a key for Miscanthus taxa, collecting 500 representative samples from each of the various species. The possibility of domesticating Miscanthus crops within China is also being investigated [129, 130]. In Japan, the Field Science Center for the Northern Biosphere has collected Miscanthus sinensis genetic resources from various parts of Japan (www. hokudai. ac. jp). In Korea 200 accessions of Miscanthus have so far been collected by the University of Seoul [17].

Finally, most of these germplasm collections are connected with Miscanthus breeding programs. Nevertheless, the development of miscanes, interspecific hybrids between species from the Saccharum and Miscanthus genus, is under investigation [13] and will also require the use of collections of Saccharum.

Biomass characteristics are important for both thermal and biochemical conversion and not all woods have the same properties. Therefore, ideal feedstock must be selected care­fully based on the properties for efficient and higher output. Moisture content, caloric value, proportions of fixed carbon and ash content are important for thermal processes (such as combustion, pyrolysis, gasification and torrefaction) while moisture content and cellulose/lignin ratio are important for biochemical processes (such as fermentation and anerobic digestion) [10].

The moisture content of Eucalyptus is higher than 50% (wet basis), which usually causes concerns among processors because it is more than in softwoods and other hardwoods, which are usually between 45 and 50% (Table 9.3). High heating value is comparable to other hardwoods but lower than pine, which is currently the primary species for energy

wood exported from the southeastern United States in the form of pellets. Bark is the major source of ash in woody biomass. High ash content along with the presence of metals such as silicon can cause fouling and slagging athigher temperatures [10,12]. This can significantly reduce the efficiency of power plants and increase operational costs. Ash content of woody biomass is less than 2%, whereas in grass species it can be as high as 5-7%. Ash content in Eucalyptus wood is approximately 1% (Table 9.3), but it can be further lowered by excluding bark.

Another chemical that is of major concern in woody biomass is chlorine. It is a corrosive element and in high concentrations in biomass can impact operations due to corrosion [12]. Corrosive action of chlorine can shorten the life of expensive equipment such as furnaces and boilers, requiring earlier replacement. Though short rotation woody crops are usually credited with higher chlorine content, study results show that chlorine content in Eucalyptus is usually less than 1% (Table 9.4).

Available data on chemical composition of Eucalyptus wood are mostly limited to pulping characteristics because of its wider use in the pulp and paper industry. With its potential to be used in the emerging bioenergy markets, other chemical properties are currently being studied (Table 9.5). Eucalyptus wood tends to have higher cellulose compared to other hardwoods. Compared to aspen, up to 9% more lignin has been recorded in Eucalyptus [14]. Access to sugar in lignocellulose biomass is still a challenge due to recalcitrance of cell wall [15] but, with appropriate pretreatment method under ideal conditions, Eucalyptus wood can be converted to biofuel [13,16,17]. Using biotechnology, lignin content can be manipulated in plants to increase sugar release. The U. S. Department of Energy’s National Renewable

Energy Laboratory study using ArborGen’s lignin-modified Eucalyptus (with only half the lignin content compared to unmodified plants) shows that low-lignin Eucalyptus can release up to 99% of sugar whereas conventional unmodified plants release only up to 40-50%.

Napier grass, [Pennisetumpurpureum (L.) Schum.], also known as elephant grass is native to equatorial Africa and is a major forage crop in the wet tropics of the world. It resembles sugar or energy cane in stature and in methods of propagation. It is considered a viable feedstock for bioenergy due to the perennial nature and yields similar to energy cane in Florida and Georgia [65].

5.4.2 Phylogeny, Growth, Yield and Chemical Composition

Napier grass is in the Panicoideae subfamily of the Paniceae tribe [66,67]. P purpureum is an allo-tetraploid with a base chromosome number 28 (2n = 4x = 28) with A and B genomes [68] and is phylogenetically similar to P. glaucum [69]. The A genome is homologous to pearl millet (P. glaucum L.) and is larger than the B genome, which controls the perennial growth habit of napier grass [70].

Napier grass is a tall bunch-type grass that produces a deep root system and rhizomes and can reach 6-7 m in height. The species is tropical in nature and performs well in climate zones 8 and 9 with temperatures of 30-35C [71]. Though frost will kill all aboveground plant material, well established plants will re-emerge in the spring if the soil does not freeze [70]. Napier grass responds well to irrigation and fertilizer but has drought tolerance due to a deep root system [72]. Napier grass has day length sensitivity and generally flowering will be initiated when day length is reduced to 11 hours or less. However, there appears to be an interaction on flowering initiation between day length and temperature [70].

Dry matter (DM) yield of napier grass varies due to cultivar and environment. The cultivar ‘Merkeron’ [73] was compared to the switchgrass cultivar ‘Alamo’ at three locations in

Georgia. Merkeron yielded 27 Mg ha-1 versus 15 Mg ha-1 for Alamo when averaged over six years [74]. Yields of between 30 and 60 Mg ha-1 yr-1 of DM have been reported for lines tested in southern and central Florida [75] and from 20-30 Mg ha-1 yr-1 in northern areas of the South [76]. Yields decreased significantly from the 31 Mg ha-1 yr-1 averaged for the first two years of a four-year study with no fertilizer inputs at Tifton, GA [65].

Napier grass leaves have high value as forage due to high crude protein and good digestibility [70]. The composition of napier grass leaves and stem vary widely. As a measure of forage quality, the percent in vitro dry matter digestibility (IVDMD) ranged from 35 to 60% for leaves and from 21 to 51% for stems among mature plants from a diverse germplasm nursery at Tifton, GA [77]. From the same nursery the percentage of neutral detergent fiber (NDF) ranged from 62 to 85%, and the percentage acid detergent fiber (ADF) from 34 to 60% for both leaves and stems. Lignin content is much greater in stems. The percentage acid detergent lignin (ADL) from leaves can range from 2 to 5% while stems have between 5 and 13% [77]. Hemicellulose tends to be higher in leaves than in stems [78]. Tall, high biomass genotypes generally average 20% leaf DM after a full season of growth. Leaf dry matter generally is negatively correlated with plant height; however, some tall accessions have relatively high leaf dry matter [79]. The protein content of napier grass leaves are the highest among perennial grasses [80], which may allow for production of plant protein-derived bioplastic polymers [81].

Compared with switchgrass, harvested napier grass after frost had a higher ash, nitrogen, potassium and phosphorus content in a four-year study at Tifton, GA [65]. It was found that with no addition of nutrients to plants, the uptake of potassium was dramatically reduced the second year while maintaining high yields, indicating that napier grass is a luxury consumer of potassium.

Biochemical conversion to ethanol has been shown to have potential for young napier grass plants. When four — to eight-week-old napier grass material is pretreated with esterase and subsequently cellulase, free sugar release is comparable to corn leaves [77] or bermuda grass [82]. Conversion of mature Merkeron by simultaneous saccharification and fermenta­tion (SSF) resulted in 107 mg g-1 of ethanol compared to 122 mg g-1 for Coastal bermuda grass [83].

The energy content of some pines can be increased by chemical treatment of stems two years before harvest [5]. The cost of the chemical (i. e., paraquat) is currently less than $0.004 per tree (does not include the cost of application). Injecting the chemical causes a wound response that increases the production of turpentine. The treatment also reduces the moisture content of the wood. As a result, the net energy yield of 20-year old Pinus elliottii logs may be increased by 13%. Organizations considering establishing bioenergy plantations with pines should consider this treatment because the reduction in moisture content should reduce the cost of transportation, since more logs can be loaded on a truck.

10.2 Harvesting

10.3.1 Harvest Age

The age when pines are harvested varies with the objective of the landowner. When the objective is based on economics (instead of biomass), rotations for pines in the southern United States are typically 20-35 years. Harvest rotations in Europe are typically longer and may range from 80-120 years [12].

When growing pines for biomass in research plots, the harvest age might be as short as 8-10 years [7, 13, 14]. However, the optimum economic rotation is determined by an economic analysis (Table 10.7). The economic rotation age for a biomass-only regime will

be similar to that used for a pulpwood-only regime. This is because the cost per delivered green Mg of pine is similar for both pulpwood and biomass.

The “biological rotation” for pine is the point where the periodic annual increment (PAI) crosses the mean annual increment (MAI). Typically, the optimal economic rotation is shorter than the “biological rotation.” The example given in Figure 10.2 illustrates why very short rotations of 5-10 years may not make sense in many cases. For example, when using a seven-year harvest rotation, two rotations of pine would capture less energy than one 15-year rotation. Because the costs for establishment and harvesting would be less, the final cost per MWh would be lower for the 15-year rotation. Some reports have incorrectly assumed the MAI for a 17-year rotation will be about the same as that obtained for an eight-year rotation. But according to the example in Figure 10.2, this flaw in logic would result in an 85% overestimation in yield.

Stand age (years)

(assumes 0% moisture content of wood)

(MAI at year 17 = 22 m3 /ha/yr)

Figure 10.2 With a zero percent discount rate and a relatively high growth rate of pine, the optimal (biomass — only) rotation will be the age where the current annual increment (thin line) and mean annual increment (thick line) cross. When the economic discount rate is positive, and there is only one price per Mg for the harvested wood, the optimum rotation age will typically be shorter.

10.3.2 Harvest Season

In temperate and tropical zones, pines may be harvested by hand year-round. In some cases, this is an advantage when biofuel plants are operated on a continuous basis. When using heavy machinery the soil sometimes becomes saturated, which reduces productivity and increases the risk of damage to the soil properties. To avoid temporary harvest disruptions caused by inclement weather, stockpiling wood can be done with little loss of energy. In some high latitude zones, the impact of logging machinery on soil is reduced during the winter when the soil is frozen.

The genetic variability present within sorghum is compiled in germplasm collections at the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) and the U. S. Sorghum Germplasm collection based in Griffin, GA, and Fort Collins, CO [82]. These repositories represent a valuable source of variation for future sorghum improvement. Most of this variation has limited direct use in temperate breeding programs due to photoperiod sensitivity. However, to make this diversity accessible, the sorghum conversion program was initiated in 1963 in the United States with a goal to introgress day-length insensitivity and dwarfing into exotic sorghum lines, making this germplasm accessible to all sorghum breeding programs [83]. During the active conversion program (approximately 40 years), over 700 photoperiod sensitive sorghum accessions were fully converted and released [84,85]. The lines have had a tremendous impact on sorghum improvements worldwide; it is difficult to find a sorghum grown anywhere (especially in temperate regions) that does not have sorghum conversion germplasm in its pedigree. In addition, conversion lines have been extremely important for genetic research and genomic analysis of sorghum for an array of traits ranging from adaptation, tolerance to biotic and abiotic stress, and improvement of quality and composition of grain, forage and eventually energy cultivars.

Shrub willows have several characteristics that make them an ideal perennial feedstock for the production of biofuels, bioproducts, and bioenergy: high yields that can be sustained in three to four-year rotations, ease of propagation from dormant hardwood cuttings, a broad underutilized genetic base, ease of breeding for several characteristics, ability to resprout after multiple harvests, and a chemical composition and energy content that is similar to other hardwoods. The similarity in the compositions of willow with other hardwoods allows these different sources of biomass to be blended together to provide a consistent year-round supply to end users.

A common misconception about willow biomass is that it makes a poor choice for the production of different forms of energy because its energy content is lower than other woody biomass and it has higher ash content. While the energy content of willow on a volume basis is lower than other hardwoods due to willow’s lower specific gravity, on a weight basis willow is similar to other hardwoods. The energy content of three-year-old willow stems averaged 19.4 MJ kg-1 [13] compared to 19.4-19.8 MJ kg-1 for northern hardwoods [14]. However, the specific gravity of shrub willow is lower than most other hardwoods. The mean specific gravity of three-year-old stems of different willow varieties in three trials ranged from 0.40 to 0.43 g cm-3 [15,16]. In both these studies there were significant differences in specific gravity among the willow varieties, with a range of 0.38-0.48 g cm-3. Similar aged hybrid poplar stems had a specific gravity of 0.35 ± 0.02 g cm-3 [15]. In contrast, the specific gravity of northern hardwoods is typically in the 0.5-0.6 g cm-3 range [17].

Willow biomass crops are grown using coppice management that utilizes willow’s natural ability to resprout. This results in biomass being produced and distributed across several stems. The number of stems produced and maintained in this production system varies among varieties and has ranged from 4.6 to 13.7 stems per stool after three years of regrowth following coppicing [18]. The production of multiple stems in this system results in the perception that the proportion of bark, and as a result the ash content, should be higher in willow biomass compared to other hardwoods. However, the bark on these smaller stems is thinner than on larger diameter stems, so the proportion of bark ranges from 4.2 to 6.6% on three-year-old stems [18]. Sampling procedures, age of plants, and stem diameter impact the proportion of bark such that other studies have reported that bark makes up 10-19% of the total biomass on two-year-old stems [19,20]. The ash content of willow biomass is also relatively low, ranging from 1.3 to 2.7% across multiple clones at one site [15, 18]. When ash content was examined in trials planted with the same suite of varieties at two different locations, there were differences in ash content, both between varieties and sites, but the mean ash content across the 18 varieties at each site was only 1.14 ± 0.09 and 1.93 ± 0.15% [16]. Clearly, soil and site characteristics have an effect on ash content, but sampling protocols and laboratory procedures may also have an impact. Reported values for ash content in willow are at or below levels reported for other sources of woody biomass, such as whole-tree chips or forest residuals [13].

Determining the composition of willow biomass is important to understand the potential for using willow biomass as a feedstock for liquid transportation fuels. Recent studies have shown differences among willow varieties and across sites. Cellulose content among 18 willow varieties at one site was 42.3 ± 0.3%, but was significantly lower (41.0 ± 0.2%) for the same suite of varieties grown at a second site. However, neither the lignin (22.1­22.3%) or hemicellulose (31.9-32.3%) content varied across the two sites. Other studies have found the ranges for cellulose (36-43%), hemicellulose (30-35%), and lignin (22­27%) to be slightly greater [16]. Further studies are underway to determine both genetic and site factors that may contribute to variations in biomass composition and its impact on energy production, particularly liquid transportation fuels.

Miscanthus genetic improvement programs are focused largely on biomass productivity and composition, along with traits that limit invasiveness, such as rhizome growth habit and reproductive sterility, and those enhancing tolerance to abiotic and biotic stress [3,24].

Plant height, stem diameter, lateness at panicle emergence, and growth rate are the main traits positively correlated to Miscanthus yield [26]. However, the heritability of these traits needs to be determined before their success in developing new varieties can be established.

For varieties propagated by rhizome or microplants, the clone will be the common varietal type. Seed propagated varieties could be developed in some species, but must be sterile to limit their invasiveness risk. Seed weight and seed pelleting methods will be important considerations of any seed-based varieties. The variety type is the clone when the propagation is vegetative. As for grasses, the synthetic variety type would be expected within seed-propagated Miscanthus species.

The low invasiveness potential of M. x giganteus and M. sinensis (Section 4.3.4, Invasiveness) and the high biomass yields of M. x giganteus make these species good candidates for Miscanthus breeding programs. Nevertheless, it is critical to ensure a low invasiveness by minimizing seed production and rhizome spread. The most effec­tive method would be to induce sterility (male and female) via triploidy (as in the current M. x giganteus). For seed-propagated varieties, sterility can also be obtained either by manipulating expression of plant hormones or cytotoxin genes in reproductive tissues. Breeding or selecting for late flowering and slow rhizome growth would also help to reduce invasiveness by minimizing seed production and rhizome spread. There is likely to be a large variation in flowering time within genotypes of M. sinensis due to its native distribution across wide range environments [110, 127].

Finally, developing triploid versions of M. x giganteus or M. sinensis will be important in reducing the invasiveness of new Miscanthus varieties [26]. The development of seed — based varieties is also under investigation by several institutes and companies. In addition, the breeding and productive potentials of M. floridulus and other Miscanthus species still need to be determined.

David B. South and Mathew Smidt

School of Forestry and Wildlife Sciences, Auburn University, U. S.A.

9.2 Introduction

Pine branches, pine logs, and residues from sawmills have been burned for energy for over a thousand years. In some places, demand for fuel and building materials was so great that expansive pine forests (Pinus spp.) were “utterly destroyed.” One example is in Scotland where 70% of the region was once covered by the “Great Wood of Caledon.” However, by 1500 the Pinus sylvestris forests were all but gone. As a result, in 1503 the Scottish Parliament passed an act to encourage tree planting. Today, pines are the most commonly planted tree genus in the world and, in some regions, pine acreage is increasing due to planting seedlings on former agricultural lands.

Pines make up about 28% of tree plantations for the ten countries with the largest areas of planted forests [1]. Pines are the most commonly planted genus in Asia, Australia, Europe and North America [2]. According to some, conifers outnumber non-conifers three to one in regards to fast-growing tree plantations [3]. Some reasons to explain the preference for pines are the ease of seed collection, the ability to store seed for long periods, the low cost of producing high quality planting stock, a consistently high level of plantation success, rapid early growth, and relatively high demand for pine sawlogs.

Although pine biomass is used to produce energy, pines are generally not planted for the sole purpose of bioenergy. There are four basic reasons for this: establishing a pine plantation requires an investment of money; several species of hardwoods from natural stands are a preferred source of firewood; returns are often higher when pine logs are merchandized and sold to pulp, sawmills, or pole mills; and coal and oil are more convenient sources of energy than wood. Historically, the economic returns for pure bioenergy plantations are lower than for plantations harvested and sold to pulp mills, sawmills or pole mills.

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

Diseases and insect pests have generally been minimal on napier grass grown in the United States. In Africa, a leafhopper (Recilia banda Kramer) has recently been identified as a vector to Napier stunt disease [93]. This disease is caused by a phytoplasma and has caused significant yield losses in East Africa [94,95]. Other diseases observed in Africa are sooty mold (Khuskia oryzae), generally associated with aphids and white mold (Beniowskia sphaeroidea) [87]. The most significant disease observed in the United States is eye spot (Helminthosporium ocellum Faris) [96], but resistance to this disease has been bred into the released cultivar Merkeron and resistance is found in many of the plant introductions. Insect damage is rare, partly due to thick pubescence on leaves and stems of most genotypes. However, two-lined spittlebug (Prosapia bicincta Say) has been shown to reduce stand loss of napier grass x pearl millet hybrids [97]. Weed control is essential at planting and spring re-emergence for good napier grass establishment and early growth. No herbicides are currently labeled for napier grass. Cutts et al. [98] studied 13 herbicides singly or in com­bination. Atrazine performed the best for most broad leaf weeds, pendimethalin for Texas millet (Urochloa texanum), and sulfentrazone for yellow nutsedge (Cyperus esculentus).

A “clearcut” harvest is used when all pines in the plantation are harvested. Bioenergy-only plantations will likely be harvested using a clearcut system. The harvesting costs (i. e., cost per MWh) are typically lower for a clearcut system than for a system that relies on one or two thinnings. This is because the time required to harvest one cubic meter of wood is typically higher when conducting a thinning, since the time required to move equipment to the location is doubled (one thinning) or tripled (two thinnings).

10.3.3 Thinning Harvest

In many countries, small and crooked pine trees are removed 5-10 years before the final harvest. The “thinned” trees are sold to pulp mills, oriented-strand board (OSB) mills or to energy plants. A “row thinning” occurs when every second, third, fourth (etc.) row is removed during the thinning. In cases where a hybrid stand has been established (where biomass rows are planted adjacent to sawtimber rows), each biomass row is removed at the first thinning.