14.5.5.1 Sodium hydroxide pretreatment

Alkali pretreatment processes generally do not hydrolyze hemicellulose as extensively as acidic pretreatments, but can be effective at removing lignin, which can lead to an increase in the enzymatic digestibility of alkali pretreated solids. Several studies on alkali pretreatment using sodium hydroxide have been reported and reviewed (11, 12, 14). This pretreatment approach causes swelling of fibers, leading to an increase in internal surface area, reduction in the degree of polymerization, a decrease in crystallinity, separation of the structural linkages between lignin and carbohydrates, and disruption of lignin structure (11). The effectiveness of sodium hydroxide pretreatment has been correlated to feedstock lignin content, with high lignin feedstocks, especially softwoods, showing poor performance using this approach (14). Dilute sodium hydroxide pretreatment has been shown to be quite effective on low lignin (10-18% lignin content) straw feedstocks (58), but the cost effectiveness of this pretreatment approach has not been thoroughly evaluated.

14.5.5.2 Ammonia pretreatment

In addition to the rapid expansion AFEX pretreatment process, which utilizes ammonia to achieve both chemical and physical changes to biomass, there are a number of additional ammonia pretreatment processes. The simplest ammonia pretreatment process involves a relatively low-temperature soaking (ambient temperature up to 90°C) using aqueous ammonia (various strengths up to 29 wt% NH4OH) at solids loadings of 10-50% and res­idence times from a few hours to up to 1 day (14, 59, 60). In these processes, up to 80% delignification has been reported on feedstocks such as wheat straw and corn stover, with much lower extents of hemicellulose solubilization. However, good enzymatic digestibility of the remaining cellulose and some of the remaining hemicellulose can be achieved us­ing commercial cellulase preparations (60). There is little evidence that this pretreatment approach would be effective on woody biomass types, especially softwoods. As with other alkaline pretreatment approaches, the augmentation of cellulase enzymes with hemicellu — lase and other accessory enzyme activities could improve the enzymatic saccharification of ammonia-pretreated biomass.

A percolation-type ammonia pretreatment process, known as ammonia-recycled percola­tion (ARP), has also been investigated (10,14,19,61). The ARP process passes dilute aqueous ammonia (<15% NH3) through a packed bed of biomass at temperatures of 150-170°C. Because of the flow-through nature and ammonia-based chemistry, ARP pretreatment of corn stover can achieve very high lignin removal (above 80% delignification) and moderate hemicellulose solubilization (above 50% xylan solubilization at residence times of 20 min or more), although hemicellulosic sugars are generally recovered in oligomeric form and would require additional processing to liberate monomeric sugars. Enzymatic digestibility of ARP — pretreated corn stover is also high (about 90% conversion of residual cellulose to glucose and about 70% conversion of residual xylan to xylose) using a commercial cellulase preparation (19). However, economic analysis has revealed that the high liquid volumes and subsequent dilute process streams does not allow the ARP process to economically compete with other pretreatments, even when efficient ammonia recovery and recycle is assumed (21). In this respect, ARP is similar to liquid hot water and dilute acid percolation pretreatments in that such processes may not be economically competitive, but they are of value in research appli­cations to generate pretreated solids with a wide range of hemicellulose and lignin removal extents for enzymatic hydrolysis and related compositional and ultrastructure studies.

Although, as stated earlier, there are a number of gas-to-liquid processes capable of produc­ing liquid transportation fuels from biomass syngas, a mixed-alcohol synthesis process is specifically discussed here because of the synergies with the biochemical approach in pro­ducing ethanol as the primary product. The commercial success of mixed-alcohol synthesis has been limited by poor selectivity and low product yields. Single-pass yields are on the order of 10% syngas conversion (38.5% carbon monoxide conversion) to alcohols, with methanol typically being the most abundant alcohol produced (41, 42). For mixed-alcohol synthesis to become an economical commercial process, improved catalysts are needed (43). Improvements in mixed-alcohol synthesis catalysts could increase alcohol yields and the se­lectivity of ethanol production from clean syngas, as well as improve the overall economics of the process through better heat integration and control and fewer syngas recycling loops. Specific research needed to accomplish these objectives is as follows:

• Develop improved mixed-alcohol catalysts that increase the single-pass carbon monoxide conversion from 38.5 to 50% (and potentially higher) and improve the carbon monoxide selectivity to alcohols from 80 to 90%.

• Develop improved mixed-alcohol catalysts with higher activity that require a lower oper­ating pressure (1000 psia compared with 2000 psia) to decrease process-operating costs. The combination of lower syngas pressure for alcohol synthesis and less unconverted syn­gas to recompress and recycle has the added benefit of lowering the energy requirement for the improved synthesis loop.

• Explore alternative mixed-alcohol synthesis reactors and catalysts. Greatly improved tem­perature control of the exothermic synthesis reaction has been demonstrated to improve yields and product selectivity. Precise temperature control reactor designs need to be de­veloped for the mixed-alcohol synthesis reaction to improve the yields and economics of the process.

2.3.1.3 Integration/demonstration needs

For any sophisticated conversion process, combining individual unit operations into a com­plete, integrated, systematic process is a challenge. To demonstrate economic competitive­ness, individual pilot-scale operations and complete integrated pilot development runs will be required. A specific challenge is to continue to demonstrate process intensification and higher yields at pilot scale to reduce capital costs.

 

Biomass Recalcitrance: Deconstructing the Plant Cell Wall for Bioenergy. Edited by M. E. Himmel © 2008 Blackwell Publishing Ltd. ISBN: 978-1-405-16360-6

The chapters in this book describe the state of the art, as well as promising new approaches, to overcome the critical science and engineering barriers to enabling modern biorefineries. We have assembled chapters that focus on topics extending from the highest levels of biorefinery design and biomass life-cycle analysis, to detailed aspects of plant cell wall structure, chemical treatments, enzymatic hydrolysis, and product fermentation options. Such compendia are often mere signposts in time. However, we hope that our unique assembly of carefully integrated topics, presented with reviews of background science, will remain relevant for those working in the biomass conversion field.

Acknowledgments

The authors wish to thank Todd Vinzant for the scanning electron micrograph image used on the cover of this book.

The U. S. Department of Energy Office of the Biomass Program is acknowledged for supporting the compilation of this book as well as its almost three decades of support for the biomass conversion sciences. Without this visionary programmatic perspective, much of the work reported here would not have been possible.

Plant stem tissues can be generally categorized as dermal, fundamental, and vascular tissues. The epidermis comprises the surface protection layer, which includes three types of cells: epidermal cells, which have a fairly thick, cuticle-like wall, and the guard and subsidiary cells, which are specialized cells that form the stomata.

The fundamental tissue of maize stem contains three types of cells. Sclerenchyma cells are found immediately beneath the epidermis. Typically, there are 1-3 layers of collenchyma cells, which are non-lignified, and lie below the epidermis. These cells are elongated axially with irregularly thickened walls. There are usually no sclerenchyma cells directly under the stomata. The parenchyma is the most numerous cell type in maize, forming the bulk of the stem. Parenchyma cells often have thin, non-lignified primary walls. The vascular bundles are surrounded by 2-11 layers of bundle sheath (fiber). The fibers are slender elongated cells
with multi-layered secondary walls. As the stem matures, the cells in the outer part of this zone become lignified.

Maize stem has scatted vascular bundles that are distributed throughout section, which is the common arrangement in monocots. These vascular bundles are scattered rather evenly through the stem pith but more numerous near the periphery. Each vascular bundle contains xylem, phloem, and other types of cells. Xylem, which is the tissue that conducts water through the plant, is composed of several different cell types. Protoxylem consists of one or two medium-diameter vessels and surrounding parenchyma. In some bundles, a cavity will be observed in the protoxylem, sometimes referred to as a protoxylem canal or protoxylem lacuna, which represents the position previously occupied by the first-formed protoxylem elements and is essentially an empty space. Metaxylem consists of wide vessels with a few narrow tracheids between them and also some surrounding non-lignified parenchyma. Tracheids are very elongated cells with bordered pit-pairs present along the walls of two adjacent cells. They differ from vessel elements in the imperforate end walls. Phloem, which is the tissue that moves sugars and other products through the plant, is also composed of several specialized cell types. Sieve tubes (wide in cross section) and companion cells (narrow and often darkly stained) form a regular crisscrossed pattern, which is typical of monocots. Most of the phloem is the latter-differentiated metaphloem, but one may also see the remnants of protophloem occurring as an irregular green line toward the outer face of the phloem and beneath the bundle sheath cells.

Significant advances have been made to transform the feedstock supply process from tra­ditional technologies historically used in smaller distributed livestock, forage, and wood product industries to an assembly system specifically designed for the biorefinery industry.

Feedstock infrastructure development is difficult because equipment, methods, and lo­gistics vary not only among resources (e. g., agricultural residues versus forest residues), but also among geographic regions (e. g., dry agricultural residues in dryer regions versus wet agricultural residues in wetter regions). Consequently, the feedstock supply infrastructure must be developed for each class of biomass resource. Three general classes that cover the range of feedstocks are as follows:

• Dry herbaceous (examples: stover, straw, and switchgrass that are harvested at <15% moisture dry basis by weight). Dry herbaceous feedstocks present the fewest logistical challenges for use as biorefinery feedstocks. Although limited in overall volume, they pro­vide a good opportunity for near-term utilization for establishing Phase II biorefineries.

• Wet herbaceous (examples: stover and switchgrass that are harvested at >15% moisture dry basis by weight). The use of wet herbaceous feedstocks is limited by a host of infras­tructure barriers. Because wet herbaceous feedstocks represent a significant portion of the overall feedstock resource, overcoming these barriers provides the greatest potential to achieve the projected tonnage targets.

• Woody (example: logging residues). Logging residues have been used for energy in Europe and the United States for nearly 30 years. As a result, the logging residue supply system is quite mature, and systems and methods are already developed to support this industry (30). Because near-term woody feedstock will consist largely of logging residues, the infrastructure for this feedstock can be readily adapted and validated against resource environment, resource policy, and other regional factors.

The R&D activity plan for developing and validating a feedstock supply infrastructure capable of producing large tonnages of biorefinery feedstocks at the lowest possible cost addresses one of three key factors — equipment capacity, equipment efficiency, or feedstock quality — affecting feedstock supply system costs. The specific research plan that focuses on the application of these cost factors to each of the supply system elements for achieving the feedstock R&D targets is described below.

2.2.1.2.1 PRODUCTION

Production is a critical component of the feedstock supply system, and it is a key component to ensuring an adequate and sustainable feedstock supply. Specific research needed to address production issues includes:

• Assessing the cost and availability of the feedstock resource on a local basis to define production costs (e. g., grower payments) and identify regional tonnages available within each feedstock type or classification at or under the feedstock threshold costs

• Identifying and validating sustainable agronomic and silviculture practices specific to feedstock types and regional variables to ensure sustainable production

• Investigating crop production improvements (e. g., increased yields, decreased yield vari­ability, and consistent quality) through genetic modification

• Developing a perennial crop program that includes matching varieties to site conditions, establishing optimum agronomic and silviculture practices, and developing a seed pro­duction program.

AFM measures attractive or repulsive forces between a probe or “tip” and the sample. The height imaging (z-axis) measures the topography and the phase imaging detects varia­tions in composition, adhesion, friction, viscoelasticity, and perhaps other properties. Phase imaging is particularly useful for mapping variations in sample properties at very high res­olution, often with superior image detail. AFM is also able to image samples under fluids that could directly visualize biomacromolecules under physiological conditions (37). AFM has been used increasingly to map cell wall surface structure. Cell walls of various plant species, both from the Monocotyledonae and the Dicotyledonae, have been imaged using AFM (9, 38-42). Early AFM studies of cell walls mostly relied on extensive sample prepara­tion processes using grinding and/or chemical extraction (37-40, 43), which could disrupt the native structure of cell wall fibrils (34). Recently, fresh cell walls from intact tissues have been imaged in water or in a partially hydrated state resulting in varied observations. In one case (42), celery epidermal peels were imaged under water with sequential dehy­dration conditions (e. g., using various concentrations of ethanol). The microfibrils were found to be smaller, uniformly distributed, and highly parallel in the never-dried speci­men. However, the microfibrils were found to be distinctly larger and more disorganized after dehydration treatment with ethanol or air. These investigators then concluded that dehydration processes could significantly affect the structure and arrangement of primary cell wall microfibrils. Our previous study (9) revealed an accurate measurement of maize cell wall dimensions with nanoscale resolution. Figure 3.9 shows a high-resolution im­age of maize parenchyma cell wall taken by AFM phase imaging. For example, a single microfibril can be measured in maize primary walls and it is approximately 3-5 nm in diameter and tens to hundreds of microns in length (supporting our previously proposed model).

Achieving near-term economic competitiveness with gasoline and starch-based ethanol will enable a viable lignocellulosic ethanol industry. Ethanol from lignocellulosic feedstocks will then be able to join starch-based ethanol in providing a sustainable, renewable resource for the world’s transportation needs. However, market analysis (4) indicates that supplying a significant fraction of transportation fuel needs with ethanol for the long-term will require additional technology advancements in all areas of the biorefinery. This is predominately driven by the need to capture higher cost feedstocks to maximize the overall impact of the biorefinery.

Future R&D efforts will need to focus on four complementary approaches. Indepen­dently, the approaches will not be sufficient to meet the long-term goals of the biorefinery, but taken collectively they will combine revolutionary scientific breakthroughs with evolu­tionary process developments to maximize the potential of the biorefinery concept to supply a significant fraction of transportation fuel needs.

Some cost reductions will be achieved by continuous process improvements to near-term technologies. For example, the construction and operation of full-scale biorefineries will highlight opportunities for unit operation optimization and provide operational experience for process optimization and cost reductions. The accumulation of operating experience and engineering data will lead to larger-scale biorefineries, which will further reduce bio­fuels production costs by leveraging economies of scale. These are the evolutionary cost reductions. However, more dramatic cost reductions will be required from scientific break­throughs for biorefineries to reach their ultimate potential.

Earlier sections of this chapter described technologies for feedstock supply systems, bio­chemical conversion, and thermochemical conversion to accomplish near-term economic competitiveness. In the future, advances will be made in all three areas, and there will be opportunities for cost savings through integrating biochemical and thermochemical con­version technologies into larger facilities. The four areas of future technology advancement needed to accomplish the ultimate potential of biorefineries are as follows:

1 Advanced, large-tonnage feedstock supply systems

2 Systems biology to improve biochemical processing

3 Selective thermal transformation to improve thermochemical processing

4 Technology integration, economies of scale, and evolutionary process optimization.

Michael E. Himmel and Stephen K. Picataggio

1.1 The modern lignocellulose biorefinery

Alternative and renewable fuels derived from lignocellulosic biomass offer the potential to reduce our dependence on imported oil, support national economic growth, and miti­gate global climate change (1, 2). However, breakthrough technologies are still needed to overcome barriers to developing cost-effective processes for converting biomass to fuels and chemicals. These needed breakthroughs include improved pretreatment processes that boost the yield of fermentable sugars while minimizing the formation and release of toxic byprod­ucts; low-cost cellulases that hydrolyze crystalline cellulose; and microbial biocatalysts that enable rapid and efficient fermentation of the mixed sugars in cellulosic hydrolysates (3).

We also understand that feedstock costs will be a major component of the commodity end-product cost of biomass-derived liquid fuel products, such as ethanol and butanol. Therefore, perhaps the highest near-term priority is boosting the yield of lignocellulose — derived sugars. Yield issues touch many other critical biorefinery operations, including particle size reduction, pretreatment and detoxification, solids/liquid separation, enzyme hydrolysis, and fermentation of sugars to products.

It is now apparent that new process scenarios are also important for ensuring the success of future energy biorefineries. For example, the consolidation of existing process schemes may deliver significant economic and technical advantages. The well-known direct micro­bial conversion process proposed that a single microorganism could produce the cellulase enzymes and ferment sugars released from biomass to ethanol in high concentrations. Such a strain does not exist today, but could be constructed with suitable acquisition of a new, deeper understanding of various critical metabolic and enzymatic processes occurring in selected bacteria and yeast. Novel microbes may also allow a staged process to optimize these steps separately.

Plant cell walls are complex and dynamic structures composed of a large degree of cross — linked polysaccharide networks, glycosylated proteins, and lignins. Primary cell walls are formed during early cell growth. The primary cell walls vary in thickness among different cell types and are usually not lignified. However, in older tissue, parenchyma cell walls can show various degrees of lignification. The secondary cell wall is deposited when cell growth has ceased and they are often highly lignified. The tracheary elements (tracheids, vessel elements) and fibers are cells that develop rather thick secondary cell walls. Primary cell walls often become lignified when secondary wall formation begins. The cell wall is generally comprised of long cellulose microfibrils interconnected by hemicellulose polysaccharides, such as xylan and xyloglucan. These amorphous hemicelluloses are generally hydrophilic, thus retaining water in the presence of the hydrophobic and highly crystalline cellulose chains and fibers. Additional polymers, such as pectins and lignins, fill in spaces in structures, such as secondary cell walls and the middle lamella.

Harvest and collection advances are required in three key areas: 1) selective harvest (including forest thinning operations), 2) single-pass or minimum-impact harvest, and 3) harvest and collection efficiencies. The primary drivers for improved harvest technologies are reduced costs and access to larger tonnages of biomass through increased producer participation. For example, improved harvest technologies that address soil quality concerns — such as carbon sequestration, nutrient/water retention, erosion, and compaction will become increasingly important for enticing grower participation and accessing biomass resources.

Performance metrics for new harvest and collection systems include: 1) efficiency, 2) equipment capacity (an element of efficiency that includes technologies that reduce capital and improve throughput of equipment), and 3) quality. Without these improvements, the accessible biomass tonnage remains restricted.

Needed research in this area includes:

• Developing innovative harvest and collection methods for all resource types to eliminate or reduce unit operation costs and agronomic silviculture operational impacts

• Understanding, quantifying, and validating harvesting-specific quality related to com­positional effects, pretreatment effects, contaminant reductions, and bulk handling im­provements

• Developing and testing innovative equipment specific to woody feedstocks for which existing equipment is too costly and inefficient.