Prodigiosin is a red pigment produced by many strains of Serratia marcescens and other microorganisms such as Pseudomonas magneslorubra and Vibrio psychoerythrous. The pro — digiosin group of natural products belongs to the Psychroerythrus family of tripyrrole red pigments that contain a common 4-methoxy, 2-2 bipyrrole ring system. The biosynthesis of the pigment is by a bifurcated process in which mono — and bipyrrole precursors are synthe­sized separately and then assembled to form prodigiosin (Boger and Patel 1988). This pigment is capable of inducing apoptosis in several cancer cell lines, except nonmalignant cells. It can be used, therefore, as a potential antineoplastic candidate. Serratia marcesens has been reported to use crude glycerol as a carbon source to produce up to 583mg/L of prodigiosin (Tao et al. 2005).

Astaxanthin

Astaxanthin is a red or orange pigment usually included in marine feed to improve fish color and visual appeal. In their natural habitat, the food chain provides chemicals and nutrients that give some species of marine fish their characteristic pink color. To retain this pink color under artificial environment, astaxanthin is often incorporated into fish feed. Astaxanthin is naturally produced by the yeast Phaffia rhodozyma from P-carotene via a two-step reaction catalyzed by P-carotene ketolase and P-carotene 3,3’-hydroxylase. This organism has been reported to ferment crude or industrial glycerol to produce 34 mg/L of astaxanthin (Kusdiyantini et al. 1998).

There are two principles common to the logistic system for the herbaceous biomass examples (cotton, sugarcane), and the woody biomass example (fuel chips).

1. Hauling efficiency (t-per-day-per-truck) is maximized by procedures that minimize the truck load time and unload time. This principle is the key to any short-haul logistics system.

2. Some at-plant storage is unavoidable, thus a system must be in place to facilitate the flow of material into and out of at-plant storage. The sugarcane harvest operates closer to just-in-time delivery than the other two harvests, because it is an integrated system under one management. Even the sugarcane system, however, requires some minimum at-plant storage to provide for uninterrupted operation during the night hours.

An advantage is gained when the infield hauling and over-the-road hauling operations are uncoupled, as is done with the cotton system (Figure 7.15). An additional advantage is gained when the harvesting operation is uncoupled from the infield hauling. Two new cotton harvesters have recently been introduced. One bales the seed cotton in a block about half the size of a conventional module, a concept that emulates the big square bale in hay har­vesting. The other concept bales the seed cotton into a 2.4-m diameter round bale and wraps it with a solid plastic sheet. Both these concepts were developed to uncouple the harvesting operation from the infield hauling. With this uncoupling, the harvester can proceed without having to wait for the infield hauling. In fact, the hauling can be done several days, or perhaps several weeks, later.

Gin

(Some at-gin storage on gravel yard)

Fuel chips are a flowable material. Size reduction is done in the forest, and the flowability advantage is then realized throughout the remainder of the logistics system. The truck can be unloaded by flowing the material out of the back. When there is no queue, unloading (total time at the plant) is 10 minutes. The use of the bulldozers to flow the fuel chips into and out of at-plant storage is expensive, but study has shown that it is a least cost option; the estimated cost to operate the at-plant storage is less than $2/t. This cost is increasing as diesel fuel costs increase, thus other options may become competitive in the future.

One option for a herbaceous biomass system is to emulate the fuel chip system and do the size reduction in the field. Another option is to remove biomass from storage (woody biomass can be harvested year-round but herbaceous has a limited harvest season, thus some storage is unavoidable) and chop the material and blow it into a van trailer, as is done with the fuel chips. Can chopped grass be dumped out of the back of a van trailer after it has been vibrated during transport? Also, a plan must be put in place to flow the chopped material into and out of at-plant storage, and the question remains—can this be done in a cost-effective manner?

The need for some minimum at-plant storage is acknowledged by most bioenergy plant design teams. The alternative is just-in-time delivery from field/forest for 24/7 operation, and this is not judged to be a practical option.

It is appropriate to review the two at-plant storage options currently in place. In the cotton system the modules are stored on the ground in a graveled area at the gin. When a module is needed, the “yard” module hauler picks up a module and places it in the conveyor that feeds the module into the module feeder, which meters a continuous flow of seed cotton into the gin. The sugarcane system is similar to the cotton system—the raw biomass remains in the hauling “package” until it is processed. In this case, the hauling bins are stacked two high in a graveled lot at the mill and dumped as needed throughout the night shift to maintain a continuous supply of material into the mill.

Solvent extraction utilizes the selective dissolving of one or more constituents of a solution into a suitable immiscible liquid solvent. It has been widely used for refining petroleum products, chemicals, vegetable oils, and vitamins. When applying solvent extraction to remove inhibitors, Wilson et al. (1989) found that ethyl acetate extraction was more effective than roto-evaporation in removing the inhibitors. The roto-evaporation removed furfural and most of the acetic acid but did not reduce lignin-derivative levels. The ethyl acetate extraction removed all the inhibitory compounds, except acetic acid, which was not completely removed by the ethyl acetate extraction process.

Aqueous Two-Phase Extraction

Aqueous two-phase systems (ATPS) are clean alternatives for traditional solvent extraction systems. ATPS are formed when two polymers, or one polymer and one salt are mixed together at appropriate concentrations and at a particular temperature. The two phases are mostly composed of water and nonvolatile polymers, thus eliminating the use of volatile organic solvents. ATPS are normally performed under mild conditions, for example, 250C, which do not harm or denature unstable/labile biomolecules or microorganisms. In ATPS, the interfacial stress (at the interface between the two layers) is lesser (400-fold less) than that in water-organic solvent systems used for solvent extraction, causing less damage to the molecules to be extracted. The separation of the phases and the partitioning of the compounds occur rapidly. The ATPS have been tested for a number of years in biotechnological applica­tions as a benign separation method. In addition, ATPS have been investigated for extractive fermentation (HahmHagerdal et al. 1981. Jarzebski et al. 1992. Banik et al. 2003) and removal of inhibitors (Hasmann et al. 2008) from lignocellulosic hydrolysates during biofuel production from biomass.

Major disadvantages of ATPS include the relatively high cost of polymer, recycle of polymer(s), and poor selectivity, although specialized and efficient systems may be devel­oped by varying factors such as temperature, degree of polymerization, and presence of certain ions.

Traditionally, the primary substrates for the industrial production of acetone-butanol were corn and molasses. The use of molasses as a fermentation substrate offers some advantages as compared to corn, such as easy handling and utilization by saccharolytic Clostridium species. Cane molasses was used in the commercial production of butanol in South Africa until about 1980 (Ezeji et al. 2004a). Solventogenic Clostridium species express the genes for potent amylolytic enzymes constitutively and do not need amylases added to substrate to metabolize starchy products. For this reason, other starchy substrates such as millet, wheat, rice, cassava, tapioca, and potatoes have been used successfully for growth and ABE produc­tion by solventogenic Clostridium species. The United States has the capacity to produce 13 billion gallons of biofuel per year from corn alone, and any further increase in biofuel pro­duction will most likely come from utilization of feedstocks other than corn grain because of limitations in supply (Gray et al. 2006). Production of butanol from low-cost lignocellu — losic biomass that does not compete with food crops may be the important aspect to meeting the DOE target (to blend 7.5 billion gallons of renewable fuels into gasoline by 2012), and for biobutanol production to become economically viable as well as sustainable (Ezeji and Blaschek 2008b).

Substrate cost has long been recognized as having the most influence on butanol price and has been identified as a major factor affecting economic viability of butanol production by fermentation (Qureshi and Blaschek 2000). To produce butanol at a competitive price, the use of more economic agricultural residues such as DDGS, corn fiber, corn stover, wheat straw, rice straw, and wood has been evaluated by many investigators including the labora­tories of the authors. Butanol-producing cultures are able to use a wide variety of carbohy­drates such as starch, cellobiose, sucrose, glucose, fructose, mannose, dextrin, galactose, xylose, and arabinose. The use of these carbohydrates by butanol-producing cultures illus­trates the potential to ferment sugars derived from hydrolysates of agricultural residues to butanol (Qureshi and Ezeji 2008).

The primary type of ancillary module, which is common to most glycoside hydrolases, is the CBM (Linder and Teeri 1997; Boraston et al. 2004). Many CBMs serve to target the parent glycoside hydrolase to the substrate. The first CBMs to have been described were initially termed CBDs (cellulose-binding domains), owing to their substrate specificities and binding to crystalline types of cellulose. The CBDs were thus divided into “types” on the basis of amino acid sequence, in a manner similar to the GH families. Further work, however, revealed that some of the CBD types were not specific for crystalline cellulose (such as type 4) or to cellulose at all (some members of type 2 bound to cellulose, whereas others bound to xylan). Moreover, some protein modules were found to exhibit binding specificity to non-cellulosic polysaccharides. Today, the different CBMs are now divided into over 50 different families showing broad specificity patterns, sometimes within a given family and even by a given module. The CBMs exhibit various functions, including targeting of the parent enzyme to the undigested substrate, targeting of given modules to portions (conformations) of the sub­strate during deconstruction, and attachment of the parent enzyme to the microbial surface.

Densification involves compacting loose density biomass in a die into a solid compact such as a briquette or pellet. Biomass arrives at the plant in chops or bales. The bales are cut into short pieces using a hydraulic piston pressing the hay against a grid of knives. The bales can also be shredded using a roller and knife arrangement. If the moisture is more than 15%, the chopped biomass is dried in a drum dryer.

Figure 7.5 shows the relationship between particle size and bulk density of biomass for an industrial grinder. The spread in data can be attributed possibly to variations in actual particle size distribution in various size groups. Note that in this particular example size groups vary from 1 to 3 mm. The bulk density for 2.5 mm is slightly more than 100 kg m-3. Thebulkdensity increases to more than 160 kg m-3 as particle size decreases to less than 1 mm. Table 7.3 indi­cates that bulk density can be increased by almost 25% by tapping (vibrating) the container.

In preparation for pelleting, the dried chops are ground in a hammer mill. For cubing, the chops are not ground. For pelleting, the ground biomass is mixed with saturated steam—in a paddle mixer located on top of the mill. Steam heats and moisturizes grind biomass. For cubing, small quantities of water are added to biomass. The steam or water acts as a lubricant to enhance binding. The moisture content of mash before pelleting is usually in the 10% range and that of chops before cubing is 12%.

Pellet mills are equipped with a large diameter short screw, a die ring, and from one to three press rolls. The feed screw pushes the biomass uniformly toward the openings in the die ring. Press wheel forces the feed through the die openings in the ring. The pressures in the mill range from 24 to 34MPa (Tabil et al. 1997). Pellets and cubes exit the mill warm and moist. They are cooled and dried to a moisture content of roughly 10% for cubes and 8% for pellets. The cooled pellets and cubes are stored under roof in a flat storage or in a hopper bottom silo. Pellets and cubes are loaded into rail cars or trucks using a front-end loader or from self-unloading overhead bins.

In some cases, the preprocessing of biomass may consist only of grinding (Mani et al.

2006) . The grind will have a bulk density of 180 kg/m3 in the truck box. This density is suit­able for short hauls. For longer hauls and long term storage, it is preferred to densify biomass to pellets or cubes.

Densified biomass requires less area and volume for storage and transport than loose biomass. In addition to savings in transportation and storage, densified biomass lends itself to easy and cost — effective handling. Dense cubes and pellets have the flowability charac­teristics similar to those of cereal grains. Bulk handling equipment for granular material is well developed and available commercially (Fasina and Sokhansanj 1996).

Bin Yang, Yanpin Lu, and Charles E. Wyman

Abstract

Cellulosic biomass is inexpensive and abundant and provides a unique resource for large-scale and low-cost solar energy collection and storage. Agricultural residues are particularly promising for initial commercial applications because of their potential low cost and near-term availability. Because the rapidly evolving tools of biotechnology can radically lower conversion costs and enhance yields, biological processing presents a particularly promising approach to converting these solids into liquid fuels that better fit our transportation infrastructure while providing unparalleled environmental, economic, and strategic benefits. Yet breakdown of the cellulose and hemicellulose in these naturally resistant cellulosic materials to release fermentable sugars is projected to be the most expensive processing step. In addition, the pretreatment step needed to realize high yields has pervasive impacts on all other major operations from choice of feedstock through product recovery and residue processing. Thus, knowledge of how agricultural residues respond to pretreatment and integrate with other operations is vital to successful applica­tions. This chapter begins with an overview of biological processing of agricultural resi­dues to ethanol followed by a summary of environmental considerations in their use and some estimates of availability based on these factors. Information is also given on the composition of major agricultural residues and reported yields of sugars from many such materials to provide a perspective on their suitability for ethanol production. Then, approaches and needs for harvesting, transporting, and storing agricultural residues are discussed. The chapter closes with a simplified analysis of the cost of processing cel- lulosic biomass to ethanol to point out key cost factors and the importance of employing low-cost feedstocks and realizing high yields. In addition, opportunities for advanced technologies to lower the cost of biological processing to ethanol and other products are outlined.

Biofuels from Agricultural Wastes and Byproducts Edited by Hans P. Blaschek, Thaddeus C. Ezeji and Ju rgen Scheffran 175 © 2010 Blackwell Publishing. ISBN: 978-0-813-80252-7

Introduction

Cellulosic biomass provides a truly unique resource for large-scale sustainable production of liquid fuels that integrate into our existing transportation infrastructure, and no other raw material can match its potential impact and cost (Lynd et al. 2008). For example, cellulosic biomass costing about $63/dry ton is as inexpensive as petroleum at $20/barrel on an equiva­lent energy content basis (Lynd et al. 1999). Furthermore, the U. S. Departments of Agriculture and Energy estimate that over 1.3 billion dry tons of biomass could be available annually, of which about 1 billion dry tons per year are agricultural resources available on the sustainable basis (Perlack et al. 2005; Long 2008). This quantity is enough to make a major impact on energy supplies (Lynd et al. 2008), with the result that conversion of agricultural biomass to organic liquid fuels (e. g., ethanol) can enhance energy security, reduce trade deficits, enhance global competitiveness, and create rural employment. In addition, processing cellulosic biomass to ethanol can continue to employ the power of biotechnology to simplify technology and realize high yields vital to low costs that address concerns about mounting petroleum prices (Wyman 1993, 1994b, 1999a). Because transportation is the single largest contributor to carbon dioxide (CO2) emissions in the United States (Tyson 1993; Wyman 1994a; Farrell et al. 2006; U. S. DOE 2006), the promise for cellulosic fuels to reduce greenhouse gas (GHG) emissions by about 90% and more compared to gasoline coupled with the low-cost potential and large resource base are vital as we seek avenues to abate increasing temperatures and deterioration in the climate. In fact, the only other potentially cost-effective energy options to power mobility with a low carbon footprint are through energy storage in batteries, hydro­gen, or compressed air, provided the electricity required to power each is derived from sus­tainable technologies at low cost. Even then, liquid fuels will be essential for long distance transport and aircraft. Combining cellulosic fuels with plug-in hybrids, more public transpor­tation, and better fuel efficiency will likely prove the most cost-effective avenue to affordable local and long distance mobility with low carbon emissions.

Despite its great promise and tremendous progress in improving cellulosic conversion technology, no commercial facilities are yet in place, with a vital challenge being to overcome the perceived risk of implementing the technology for the first time (Wyman 1999a). Once commercialized, costs are expected to drop dramatically through the learning curve effect, as clearly demonstrated for cane sugar ethanol in Brazil and corn ethanol in the United States, and projects will become both more profitable and less risky as more capacity comes on line (Wyman 2007). The current situation presents a classical chicken-and-egg challenge of how to overcome the greater risk and lower returns associated with first commercial plants to realize lower costs and higher returns of mature projects.

Because feedstock costs are dominant in processing economics, it is critical to seek those that are low in cost for first applications while being sufficiently abundant. However, high product yields and ease of processing are also vital to minimizing costs, while sufficient amounts must be available to support a large enough facility to achieve reasonable econo­mies of scale. Agricultural residues are expected to serve as a major biofuels feedstock, and their potential low cost and current availability can be particularly important in the near term (Perlack et al. 2005) . Thus, this chapter will summarize estimated amounts of leading agricultural residues and their potential for making ethanol. However, first an over­view will be presented of the biological conversion of these materials to ethanol to provide a context on key feedstock and processing considerations. The economics of converting residues to ethanol will then be outlined to demonstrate the importance of feedstock com­position, availability, and cost to good returns on capital. In addition, some other important considerations in process economics and financing will be summarized. Finally, strategies will be discussed to introduce technologies for biological conversion of agricultural residues to ethanol.

High-rate anaerobic treatment systems rely on biomass settling or membrane technology to achieve longer solids retention times compared with HRTs for satisfying performance under high volumetric loading rates. This results in smaller reactor volumes, and thus lower construction costs than low-rate systems, because the volume of anaerobic digesters is sized based on the HRT, while the performance of all digesters is dependent on the sludge reten­tion time. Full-scale high-rate anaerobic digesters have shown excellent and stable perfor­mance for over 30 years, but have been used mainly for low-solids industrial and domestic

wastewater treatment (Lettinga 1995; Hulshoff Pol et al. 1997; Seghezzo et al. 1998). For high-solids wastewater or slurries, such as diluted swine waste, existing technologies (e. g., up-flow anaerobic sludge blanket and its derivatives) showed problems with solids separa­tion and solids accumulation in the reactor (Zeeman et al. 1997; Elmitwalli et al. 1999; Kalogo and Verstraete 1999). The anaerobic sequencing batch reactor (ASBR) was devel­oped to treat a high — solid influent with a high-rate technology (Figure 4.2 ; Dague et al. 1970; Dague and Pidaparti 1992). Both the ASBR and an alternative technology—the plug — flow anaerobic baffled reactor—have made it possible to treat high-solids waste (e. g., animal waste) by high-rate systems (Zhang et al. 1997; Boopathy 1998).

Once we have selected a set of enzymes to be included into a cellulase system—be it a free cellulase system, designer cellulosomes, native cellulosomes, or any combination thereof—it may be desirable to consider improving the enzymes. In theory, this can be accomplished by rational design or by directed evolution. Moreover, enzyme improvement can assume different forms. We may want to increase the activity of the enzymes. Alternatively, we may want to change the optima of their physical properties, such as temperature and pH. In changing the latter properties, directed evolution can be employed with some level of efficiency, since temperature and pH can be used as a selection pressure, and the stability of the mutated enzymes can be assayed relatively easily. Indeed, some success in the literature has been reported for such endeavors (Murashima et al. 2002; Wang et al. 2005) Choi et al. 2008) . In recent studies (Hughes et al. 2006) , combinatorial and robotic handling methods have recently been instituted for improvement of cellulase activity in individual endoglucanases. However, the methodology employed a soluble chromogenic substrate, and the approach was again taken with the intention of identifying mutants with heightened activities at low pH.

Improvement in cellulase activity per se is another story. It is one thing to isolate endoglucanase mutants using soluble substrates, but it is quite another to isolate mutants of cellulases that work better on insoluble substrates. Moreover, the defining characteristic of cellulase and cellulosome action is not the improvement of a given cellulase, but how the different cellulases will work together to overcome the recalcitrant properties of the substrate. In this context, the rate-limiting step in the hydrolysis of crystalline cellulose is not the cleavage of the glycosidic bond of the cellulose chain, but the detachment of a single chain from the crystalline matrix (Bayer et al. 2007- Himmel et al. 2007, 2008a). Currently, there is really no acceptable assay for this function that can be employed for medium — or high-throughput procedures necessary for screening and selection of potent cellulases. Since there are no clear relationships between cellulase activities on soluble substrates and those on insoluble substrates, soluble substrates should not be used to screen or select for improved cellulases. Some exoglucanases, for example, show little or no activity on any substrate, but contribute substantially to the overall synergistic activity of enzyme mixtures and on insoluble cellulosic substrates. Theoretically, such assays should be based on relevant solid substrates, such as paper or plant cell walls (Zhang et al. 2006- Himmel et al. 2007). However, in practice, this has yet to be achieved in a reliable manner.

Conclusions

The structural polysaccharides in lignocellulosic biomass are a rich and renewable source of fermentable sugars for industrial production of biofuels. It is important to note that in attempting to utilize these carbohydrates at the commodity scale, we must overcome a key principle set forth in the evolutionary development of the cell wall of terrestrial plants: essential recalcitrance to deconstruction. Fortunately, progress is being made in this endeavor. Indeed, although these general process technologies are known, the key cost challenges remain the subject of considerable international research focus today. It is clear that only through dedicated, fundamental science guided by clearly defined applied objectives can such complex processes be made a reality. In this case, new and improved enzyme systems closely coupled to related process technologies, such as biomass pre­treatment, are required to provide cost-effective and large- scale quantities of liquid fuels from biomass.

Acknowledgm ents

The biomass structure, chemistry, and enzyme engineering review presented in this work was supported by the BioEnergy Science Center (a U. S. Department of Energy Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science); the remainder of the review was supported the Israel Science Foundation (Grant Nos. 966/09 and 159/07), by grants from the United States-Israel Binational Science Foundation (BSF), Jerusalem, Israel, and by the National Research Initiative Competitive Grant Nos. 2002-35206- 11634 and 2006-35206- 16652 from the USDA Cooperative State Research, Education, and Extension Service. E. A.B. holds The Maynard I. and Elaine Wishner Chair of Bio-Organic Chemistry.

Most corn ethanol dry-grind plants in the Midwest usually keep an inventory of corn for 10 days of production (Mukunda 2007). Corn is normally delivered to the plant by trucks from within a 60-mile radius. Just like in elevators, delivery trucks queue in line to be sampled, weighed, and their cargo dumped. The rule of thumb is that for every 1MMGY of plant capacity, a truck of 1000bu (25,500 kg) of grain is delivered daily. Thus, for a 40, 60, and 100MMGY, an average of 40, 60, and 100 trucks will deliver corn to the plant daily. With the current capacities in fuel ethanol plants, there has been no report about challenges or difficulties in receiving grain. In fact, some of the ethanol plants have contracted their grain procurement to the major grain logistics operations that have many years of experience and infrastructure handling and shipping grain. If these plants were to be retrofitted to process cellulosic feedstocks, how would their current layout designed for granular feedstocks perform? This could be one of the major challenges faced by existing biorefineries using grain feedstocks that are looking to change to cellulosics in the nearby future. In the next section, a framework for designing and mapping the layout of a biorefinery or preprocessing facility using lignocellulosic feedstock will be discussed.