The inbound logistics and handling of feedstocks at the plant gate is one of the important logistics operations in biomass utilization for food, feed, fuel, and bioproducts. Fuel produc­tion or power plants using biomass feedstocks need to have enough inventory to ensure at least 10 days of production, and supply needs to be such that this inventory level is always maintained at the plant gate. Inventory can result in significant costs and operating inefficien­cies (Dilworth 1992) and thus careful planning at the production plant must be made to secure the required weekly inventory year round. This planning will include an evaluation of the feedstock material flow layout for receiving operations, storage space, and inventory manage­ment to ensure that the quality of stored feedstock is maintained until consumed. Fire hazard prevention is especially important for highly combustible and reactive feedstocks like plant biomass.

The current biorefineries using grains for the production of fuel ethanol were designed to receive the majority of their feedstocks by trucks. Unlike for some other manufacturing operations, feedstocks to be delivered to a processing plant come from hundreds of producers in a region having different travel distances to the plant. The inbound logistics (material flow) of feedstocks delivered at the plant gate should be coordinated such that delivery trucks are able to spend the minimum amount of time waiting to process and off-load their cargo. This will prevent having long lines of waiting trucks at the plant waiting to off-load. Traffic con­gestion by waiting trucks can pose traffic hazard and also make communities antagonistic to having biorefineries or similar operations in their communities. For example, a 110 million gallon per year (MMGY; 416.4 mil. L per year) corn ethanol plant has about 110 truck deliv­eries of corn per day with each truck carrying about 25,500kg (1000bu) of corn. With a conversion rate of 10.2 L of ethanol per 25.5 kg (1 bushel) of corn, a truck load of 25,500 kg will produce about 10,200 L of ethanol. For comparison with lignocellulosic biomass like corn stover, a truck load of corn stover will hold about 17.5 dry tons of feedstock (39 rect­angular bales of 8′ length x 4′ width x 3′ height; Mukunda 2007) that can be converted to about 4769L of ethanol at a conversion rate of 272.5L of ethanol per dry ton of feedstock. This means that about twice the number of trucks of corn stover (220 trucks) to the plant to keep the same level of production with corn grain are needed in this scenario. Traffic conges­tion possibilities become a big issue and thus inbound logistics of delivering feedstock to the plant need to be carefully designed to prevent traffic congestion.

This section will discuss the inbound logistics of delivering lignocellulosic biomass to a biorefinery. This will be presented in three sections, namely: (1) a comparison of the inbound logistics of three feedstock types, corn grain, and corn cobs that would give us an idea of what could be designed for lignocellulosic biomass; (2) components of the inbound logistics of feedstock delivery at the plant gate; and (3) analysis of inbound logistics of biomass delivered to a biorefinery.

Agricultural residues can provide an attractive feedstock for cellulosic ethanol production in the near term because of their current availability, and many with high cellulose and hemicel — lulose content are amenable to conversion to ethanol with high yields. However, in estimating potential contribution to large — scale fuel production, consideration must be given to how much can be removed without problematic environmental consequences such as depletion of soil carbon and soil erosion. In addition, the cost of gathering and transport must be factored in, and collection strategies should be employed that minimize collection of dirt and stones. Storage techniques must also be developed that are low in cost but result in little degradation of the feedstock. And we must be sure that the sugars can be extracted from the carbohydrate fractions with high yield.

Because of the large impact feedstocks have on overall costs, selection of low-cost residues can be particularly important for overcoming the many obstacles to implementa­tion of cellulosic ethanol technology for the first time in the near term including perceived risk and the associated high rates of return on capital, overdesign to compensate for risk, suboptimal facility sizes that keep investment costs lower but fail to capitalize on econo­mies of scale, and other disadvantageous burdens. For example, cutting the cost by $30/ dry ton can reduce cash costs by over $0.35/gal of ethanol produced. However, even then taking advantage of other economic levers such as integration into an existing fermentation or power facility to reduce capital costs; production of valuable coproducts from lignin, minerals, or other components; and use of low — cost debt financing through partnerships with municipalities or others can have a tremendous impact on commercial success (Wyman and Goodman 1993) . Unfortunately, given the great fluctuations in petroleum prices, all of these factors may not be enough to overcome the huge obstacles facing first-time implementation, and government policy would make a major impact on bringing the tech­nology into play before dire economic situations make it sufficiently profitable for the first projects to be successful (Wyman 2007). Although such assistance could take many forms, it must be structured in a manner that will not strand huge investments by the private sector while still ensuring rigorous due diligence that will result in economically viable projects. For example, government investments as an equity partner would buy down the high capital costs of first projects, show the private sector of the government’s seriousness, and provide a payback to the government. Once in place, significant learning curve improve­ments and technology advances will lead to lower costs that can compete without govern­ment support. However, because huge lead times are needed to build up meaningful capacity, such a commitment must be made sooner rather than later if we really hope to reduce our mounting dependence on oil imports and have any hope of affecting buildup of CO2 and other GHGs. Otherwise, we will continue to twiddle our thumbs while glaciers melt and coral reefs die.

As most recombinant strains cannot tolerate high concentrations of ethanol, an alternative solution to this problem could be the simultaneous removal of ethanol as it is produced. As ethanol is removed from the fermentation broth, the culture is relieved of the ethanol toxicity effects and as a result produces more ethanol. As ethanol is produced, the level of sugar present in the reactor is reduced. To replace sugar, the reactor is fed with a concentrated sugar solution at a rate that is compatible to sugar utilization by the microbe for ethanol production. This process is called fed-batch and is applicable to systems where the use of a concentrated sugar solution is toxic to the culture. It should be noted that most lignocellulosic hydrolysates contain low sugar levels, hence a combination of lignocellulosic hydrolysate supplemented with glucose to raise the sugar level is an approach to consider. Alternately, cultures that are not inhibited by high sugar concentrations can be used in batch systems with simultaneous product recovery. In such systems all the sugars present in the bioreactor are converted to a final product. Application of these simultaneous product recovery techniques would reduce reactor size, process stream volume, and result in energy-efficient removal of final product. Use of these techniques for butanol fermentation has been successfully demonstrated in labo­ratory scale reactors (Qureshi 2009). Two of the most common technologies that can be used for product recovery include gas stripping and pervaporation. Gas stripping can be performed using fermentation gases (CO2 in the case of ethanol production, and CO2 plus H2 in the case of the butanol fermentation). The use of pervaporation requires a selective membrane that allows diffusion of biofuel only and restricts water transport across the membrane.

Thus far, we have mostly discussed anaerobic digestion of diluted farm-based wastes (~3%- 8% TS in the reactor), such as slurries of swine waste and dairy manure. However, “dry fermentation” has found a niche in the bioenergy industry as well (~ 18 %—3 5 % TS in the reactor), especially when energy crops are used as feedstock. These systems are most often operated at thermophilic conditions to take advantage of the superior hydrolysis rates (Richards et al. 1991; De Baere 2000). The advantage of high-solids digestion compared with low-solids digestion is that high-solids digestion requires smaller reactor volumes than low — solids digestion due to high VMPRs. However, possible toxicity of metals and ammonia must be taken into consideration (Jewell et al. 1993). Dry fermentation has also been referred to as anaerobic composting, dry digestion, or high-solids digestion (Jewell et al. 1993; Chyi and Dague 1994; De Baere 2000). Systems that are operated with a TS content between ~10% and 18% have been referred to as semidry digestion (Mata-Alvarez et al. 1993). Numerous pilot — scale and full — scale plants with different designs have been built and operated, and excellent performances have been reported. For example, a volumetric biogas production rate of 9.2L/L/day (this means that 10 times as much biogas is produced than the volume of the digester itself) was reported for a full-scale dry anaerobic composting (DRANCO) process for which the TS in the system averaged 31.3% (De Baere 2000).

Polyhydroxyalcanoates (PHA) are a class of naturally occurring polyesters that are synthe­sized by various microorganisms as a reserve material for carbon and energy. When the bacteria are grown under low nutrition conditions, PHAs accumulate in the cytoplasm of the bacteria as hydrophobic granules and act as a carbon reservoir as well as an electron sink. Since these polymers are biodegradable and also biocompatible, they are of high demand in the pharmaceutical, fiber, and horticulture industries. In Ralstonia eutropha and most other PHA accumulating bacteria, polyhydroxybutyrate (PHB) is synthesized from acetyl — CoA through three enzymes: (1) P-ketothiolase, which condenses two molecules of acetyl-CoA to P-acetoacetyl-CoA, (2) an NADPH-dependent acetoacetyl-CoA reductase that catalyzes the formation of D-(2)-3-hydroxybutyryl-CoA (3HB-CoA), and (3) PHA synthase, which polym­erizes 3HB-CoA to PHB (Schubert et al. 1988; Slater et al. 19882 . Currently, industrial synthesis of PHAs involves aerobic processes that are energy intensive. There is a need, therefore, to develop newer fermentative ways of synthesizing these compounds. When Methylobacterium rhodesianum MJ 126-J and R. eutropha DSM11348 were cultured in a medium containing crude glycerol and casein hydrolysates, the amount of PHB produced was 50g/L and the net conversion of glycerol to PHB was 17% (Bormann and Roth 1999). E. coli has also been engineered to produce PHBs by mutating the arcA2 locus. The ArcAB system of E. coli represses genes that encode the enzymes involved in aerobic respiration, such as those of the tricarboxylic acid cycle, under anaerobic and microaerobic conditions (Salmon et al. 2005). The mutation in arcA region would elevate the tricarboxylic acid cycle activity, which supplies large amounts of reducing equivalents such as NADH and NADPH. Excess reducing equivalents favor the formation of an electron sink like PHBs (Nikel et al. 2006). The recombinant E. coli with mutation in arcA produced 1.44g/L ofPHBs in glucose media while the corresponding non-mutant produced only 0.07 g/L of PHBs. When recom­binant E. coli was used for the fermentation of glycerol, the amount of PHBs accumulated in the fermentation reactor was 1.47 times higher than that produced when glucose was the carbon source (Nikel et al. 2008). The difference in PHB concentrations was attributed to the difference in redox characteristics of glucose and glycerol carbon sources (Nikel et al. 2008).

Sugarcane is a high-yield (up to 155 t/ha), high-moisture content (80+%) crop. It is grown to collect the sugar (sucrose) produced in the stalk. The stalks are crushed and the juice col­lected. This juice is then concentrated into molasses that is subsequently centrifuged to produce crystalline sugar. The raw sugar is washed to produce the pure white product we see in the sugar bowl on our breakfast table.

Harvesting

A sugarcane harvester cuts the stalk at the base, and then cuts it into 30-cm-long billets. Air is blown through the billets as they are cut to blow away as much leaf as possible and leave it in the field. The billets are conveyed into a side-dump wagon that travels with the harvester. There is no onboard storage on the sugarcane harvester, as with the cotton harvester, thus the side-dump wagon must be in place for the harvester to operate.

The side-dump wagons proceed to the edge of the field where an elevated loading ramp is prepared (Figure 7.14). The wagons dump directly into bins on trucks, and these trucks deliver the bins directly to the sugar mill.

Logistics

Cycling of the side-dump wagons and the highway trucks must be well coordinated for the harvest to proceed with maximum efficiency. The cut stalks spoil so quickly that there is no storage between field and mill. This same constraint will apply to any option that proposes to ferment a “sugar biomass " directly to produce ethanol.

Truck trailers hauling the bins are equipped with a pivot point such that the bins can be dumped directly onto the conveyor feeding material into the mill. Time to dump a load (trailer with two bins) is 3 minutes. Some truck tractors pull a double trailer (four bins), and the time to dump this load is 7 minutes. If a given load is not needed to fill the conveyor, the loaded bins are removed and stacked two high on a graveled storage yard. Empty bins are removed from storage and placed on the truck to be returned to the field. This operation takes 3-4 minutes. One sugar mill in South Florida unloads 1000 trucks per day, a total delivery of about 24,500t/day.

Harvesting and hauling is done only during daylight hours, thus the at-plant storage is essential to operate the plant 24/7. Each bin has a pivot point so it can be dumped during the evening when placed in position with the forklift. The sugarcane system is unique in that harvesting and hauling does occur 7 days per week.

One sugar company in South Florida harvests and processes 3.3 million t of sugarcane in a 140-day season. The industry is a unique example of plantation agriculture. The company owns the production fields, the harvest equipment, the service roads through the fields, and the sugar mill. It owns the trailers that haul the bins, but typically contracts with trucking companies for the truck tractors that pull the trailers.

When the truck tractors pull a double trailer, the second trailer is hooked behind the main trailer. Each trailer hauls two bins for a total of four bins, thus the total length of the vehicle is almost 35 m. A vehicle like this would not be legal on most public roads across the United States. The labor productivity (t/hour) for truck drivers in the sugar industry cannot be equaled for any other biomass logistics.

Trucks typically haul 10 loads per day. With no delays, they could theoretically haul 13 loads per day, thus the truck productivity factor is 10/13 = 0.77, or 77%. This exceptional performance is possible because one entity, the sugar company, has control of all segments of the harvest logistics system. The harvest equipment is managed and trucks are scheduled to minimize the load and unload times. This management capability is key to the efficiency of a short-haul operation. It is also significant that the trucks travel on company-owned roads, thus they do not encounter the traffic seen on public roads. It is unlikely that any other biomass system that collects material from fields and delivers it to a central plant can equal the truck productivity, 77%, of the sugar industry in South Florida. Also, it is unlikely that the level of coordination between infield hauling and over-the — road hauling can be duplicated any­where else. It certainly cannot be duplicated if farmers haul in the raw biomass and make deliveries on their own schedule.

Another sugar mill in South Florida uses a “mobile accumulator” concept to supply the mill 24/7. This mill has railroad tracks extending from the mill out through the production fields. There are loading platforms at periodic intervals along these tracks. Side-dump wagons fill the railcars at the loading platforms. When the cars are filled, the train is pulled onto a siding at the mill. As needed for 24/7 operations, the cars are moved into position and dumped.

Again, the plantation model offers an advantage. The sugar company owns the railroad tracks through their production fields, thus they can use railcars as a mobile storage for the sugar mill. Once the investment in tracks and other infrastructure is recovered, the train can be more economical than trucks hauling bins.

Toxic compounds in lignocellulosic hydrolysates can be divided into four major groups: sugar degradation products, lignin degradation products, compounds derived from lignocellulose structure, and heavy metallic ions (Mussatto and Robert 2004). Typical inhibitory compounds include carboxylic acids, aldehydes, furans, and phenolics (Zaldivar and Ingram 1999; Palmqvist, and Hahn-Hagerdal 2000b; Oliva et al. 2003; Varga et al. 2004). The first step in the conversion of biomass to fermentable sugars is pretreatment. A pretreatment process uses mainly physical and chemical methods to increase surface area; decrystallize cellulose;

Biofuels from Agricultural Wastes and Byproducts Edited by Hans P. Blaschek, Thaddeus C. Ezeji and Ju rgen Scheffran 233 © 2010 Blackwell Publishing. ISBN: 978-0-813-80252-7

remove sheathing on cellulose by hemicellulose and lignin; alter lignin structures; and some­times partially remove lignin to improve both the rate of enzymatic hydrolysis and the yield of monosaccharides (Mosier et al. 2005a). Different pretreatment methods, including steam explosion (Ohgren et al. 2005) , dilute sulfuric acid (Lloyd and Wyman 2005) , hot water (controlled pH) (Mosier et al. 2005b), ammonia fiber/freeze explosion (AFEX; Teymouri et al. 2005), ammonia recycle percolation (ARP; Kim and Lee 2005), lime (Kim and Holtzapple 2005), ionic liquid (Dadi et al. 2006, 2007), and peroxide (Saha and Cotta 2006), among others, have been proposed and tested for deconstruction of a variety of lignocellulosic biomass. The pretreatment of lignocellulosic biomass by acids or water often results in the degradation of cellulose, hemicellulose, and lignin (Figures 11.1-11.3). These degradation compounds exert inhibitory effects on fermentation microorganisms.

Enzymatic hydrolysis of lignocellulosic biomass may also release inhibitors from biomass

OH

Figure 11.4. Inhibitors from linkages between hemicellulose and lignin (Klinke et al. 2004).

arabinoxylan, coumaric acid, and ferulic acid are significant components of solubilized corn stover hemicellulose produced during simultaneous saccharification and fermentation (SSF). These acids from biomass structure are toxic to fermentation microorganisms. When 0.3 g/L of p-coumaric and ferulic acids were incorporated into a fermentation medium, the acetone — butanol-ethanol (ABE) production by Clostridium beijerinckii BA101 decreased significantly (Ezeji et al. 2007).

Heavy metal ions (iron, chromium, nickel, and copper) can originate from corrosion of hydrolysis equipment (Figure 11.5). Although they are not always produced in large quanti­ties, they can have some toxic effect on the alcoholic fermentation microorganisms (Mussatto and Robert 2004).

The toxic effects of various inhibitory compounds on the growth response of ethanolo — genic Escherichia coli were found to be in the following order: aldehydes > organic acids > alcohols (Zaldivar and Ingram, 1999; Zaldivar et al. 1999, 2000). When plotting series of separate functional groups of phenol, aldehydes, ketones, and acids, a correlation between the hydrophobicity and inhibition of volumetric ethanol productivity was reported.

Figure 11.5. Inorganic salts and heavy metal ions (Klinke et al. 2004; Mussatto and Roberto 2004).

The more hydrophobic the compound was, the more the inhibition was evident. The phe — nolics were reported to be among the most toxic compounds to fermentation microorganisms (Larsson et al. 2000; Klinke et al. 2004).

Detoxification Methods

To overcome the toxic effects of inhibitory degradation products of lignocellulosic biomass, many detoxification methods have been investigated. The focus has been placed on removing the inhibitory compounds from the hydrolysates, modifying the inhibitory compounds, or improving the resistance of the fermenting microorganisms to the toxic effects of the inhibi­tory compounds (Palmqvist and Hahn-Hagerdal 2000a; Pienkos and Zhang 2009). The detox­ification methods can thus be generally divided into three categories: chemical, physical, and biological methods (Table 11.1).

Electrolyzed water is a technique first developed in Japan in the 1990s. The concept involves electrolysis of water containing a small amount of sodium chloride (0.1%) in an electrolysis chamber where anode and cathode are separated by a bipolar membrane imparting unique characteristics to the water collected from the two electrodes. The water from the anode normally has a pH of <2.7 and an oxidation reduction potential (ORP) of >1,100 mV The

water produced from the cathode side has a pH of >11.4 and ORP of <-795 mV The sche­matic diagram depicting the process is included in Figure 3.5. The low pH acidic electrolyzed water (AEW) and the high pH alkaline electrolyzed water have been used recently to pretreat DDGS, and subsequently for fermentation into ABE (Wang et al. 2009a). Scanning electron micrograph (SEM) images of DDGS pretreated with AEW showed the crystalline structure of the DDGS was disrupted by the pretreatments with AEW at pH 2.7 (Wang et al. 2009a). A major advantage of using electrolyzed water is that it is produced using water with no added chemicals other than very dilute sodium chloride.

In the enzymatic hydrolysis of plant cell walls, cellulose digestion is highly dependent on hemicellulose digestion. Although other factors such as crystallinity and lignin content have been suggested as barriers to the enzymatic attack on lignocellulose (Kong et al. 1993), the key to increasing lignocellulose digestibility depends on the increase of the cellulose surface that is accessible to the enzymes. Hemicellulose surrounds the cellulose fibrils, protecting them from any biological attack. This makes it a necessity to hydrolyze hemicellulose first. It is now thought that digestion of hemicellulose loosens the rigid, complex structures cover­ing the microfibrils of the cell wall and exposes the cellulose surface to cellulase attack (Ding and Himmel 2006) . Indeed, recent studies of the augmentation of cellulase systems with xylanases and carbohydrate esterases demonstrate clearly that cellulose digestibility is linked to a synergistic relationship between these enzymes (Selig et al. 2008).

The last rows in flow diagrams in Figures 7.2 and 7.3 show harvesting and collecting the entire crop that includes straw and grain in a single operation. The entire material is trans­ferred to a central location where the crop is fractionated into grain and biomass. The whole crop harvesting and fractionation concept has been researched for many years (Buchele 1976). A whole crop wheat harvester was developed in Sweden in the early 1980s (Lucas 1982) at a cost of more than $5 million. The self-propelled machine was able to harvest the entire crop, thresh and clean the grain, and bale the straw, all in one step. Recent efforts by Quick and Tuetken (2001) have been reported to develop a whole crop harvester and transporter for corn.

The McLeod Harvester (St. George 2000) developed in Canada fractionates the harvested crop into straw and graff (graff is a mixture of grain and chaff). The straw is left on the field. Grain separation from chaff and other impurities take place in a stationary system at the farmyard. The new machine is credited with higher capacity and efficiency than current grain combines. PAMI (1998) conducted an economic analysis to show that whole crop baling resulted in the highest net return among six different systems including McLeod harvester. For the whole crop baling, the crop (wheat) was cut and placed in a windrow for field drying. The entire crop was then baled and transported to the processing yard. The bales were unwrapped and fed through a stationary processor that performed all the functions of a normal combine. The straw was then rebaled.