**3. Alternatives for biofuel production**

high protein fraction is preserved and the high fiber fraction is processed in the field requiring

(A) (B)

**Figure 3.** Prototype alfalfa leaf stripper. A. Process of stripping the leaf fraction from alfalfa plants. In this prototype machine, harvesting stems was a separate activity from harvesting of the leaf fraction. The stem fraction was left stand‐ ing in the field until leaves had been removed and then stems were cut and chopped for ensiling. Next generation har‐ vesters would combine these two operations into a single pass over the field. B. Alfalfa stems with 80-90% of the leaves

The genetic make up of alfalfa has been studied over the past 20 years to maximize quality and digestibility. A key component of this research in the past has been genetic selection for alfalfa germplasm that can withstand frequent cuttings as opposed to the accumulation of large amounts of biomass. Now there is interest to exploit the genetic potential to increase more biomass then is currently available for alfalfa. Efforts to genetically select for a biomass-type alfalfa that produces larger stems and more branching with greater total yields has been successful[13, 24, 27]. According to Lamb et.al.,[24, 27] alfalfa genetically selected for increased biomass production and managed to maximize yields resulted in a 40% increase in tons per acre. Revised management techniques amounted to decreased stand density providing more space for individual plant growth and development coupled with a delayed harvest i.e., switching from early bud stage to plants at 50% bloom or later. This provides the biomass alfalfa plant to accumulate higher amounts of total plant material, both leaves and stems. With the larger more robust stems lodging is minimized compared to the typical hay type alfalfa [13]. Coupled with a new harvesting technique of in-field fractionation, this could improve the amount of biomass for biofuels while still producing a high-protein fraction for value-added products. The theoretical ethanol yield for alfalfa stems would be 137 gal/acre compared to 174 gal/acre for corn stover assuming only half of the stover is removed to maintain soil health and long term productivity[13]. Including the grain for ethanol production (473 gal/acre), corn far outpaces the amount of ethanol potential from alfalfa. However, the estimated protein yield per acre would be 0.49 tons/acre for alfalfa leaves, zero for the corn stover and 0.34 tons/acre for corn grain [13]. In the face of growing world populations protein production will be of increasing concern. In terms of outright biomass production, the system of crop rotations between corn and alfalfa lags behind year after year of corn production. From an economic

less post harvest processing at the biofuel production sites.

removed.

554 Biofuels - Status and Perspective

Current technologies rely primarily on the yeast-ethanol platform to create liquid fuels. The process has been well studied and continues to undergo development to utilize more of the cell wall sugars in addition to the cellulosic glucose. Much of the current biofuel industry is based on yeast fermentation of glucose that is derived from starch primarily from corn grain, although any cereal grain could be used. Brazil has adopted a slightly different approach and has based much of its ethanol production on sugarcane using yeast fermentation. These systems are not sustainable in the long run due to ever increasing populations with increasing demands for food. Capturing biomass for conversion to biofuels is a big part of the vision for decreasing dependence upon fossil fuels. Biomass to biofuels does not directly compete with production needs for food and feed and provides opportunities to maximize utilization of our landscape in ways that are sustainable and improves productivity. However, converting biomass to biofuels efficiently is a critical part of the story.

At this time ethanol production is the main form of biofuel product proposed for biomass[1-2]. This system utilizes yeast-based fermentation using primarily glucose as the substrate for ethanol production. The challenge in using corn stover or any other source of biomass in this process is the complexity of the plant cell wall. Cell walls are complex matrices composed of largely of cellulose microfibrils embedded in a matrix of structural polysaccharides. Once cell walls have reach their maximum size lignification occurs producing a hydrophobic polymer that drives the water from free spaces within the wall as it fills in these open areas (Figure 4) imparting additional strengthen to the wall. This process creates regions within the cell wall that are difficult to hydrolyze especially once the wall has been dried. A comparison of alfalfa stem cell wall composition with that of corn stover provides similar proportions of glucose on a kilogram of dry matter basis (Table 1). To render the glucose available for fermentation current technologies for ethanol production rely heavily on pretreatments to release sufficient amounts of the cellulosic portion of the wall for enzymatic conversion to glucose [28]. Pre‐ treatments are designed to disrupt the cell wall matrix allowing cellulytic enzymes access to the cellulosic components while minimizing the formation of degradation products. Typically dilute acids combined with high temperatures are the most common form of biomass pre‐ treatment [28]. In the case of grasses pretreatments effectively disrupt cross-linking of cell wall

**Figure 4.** Cell wall model showing formation of lignin in grass wall matrix. Lignin in grasses is attached to ferulates that are shuttled out into the wall attached to arabinosyl side chains of arabinoxylans. This creates a tightly integrated wall matrix of lignin with wall structural polysaccharides. Similar cross-linking most likely occurs in dicot walls except the ferulates are not likely to be the most prominent anchor points to the wall carbohydrates. Treatment of walls with hot dilute acid solutions removes most of the non-cellulosic polysaccharides opening up the matrix to be more easily degraded by the addition of cellulosic enzyme cocktails.

arabinoxylans via ferulate dimers and to lignin via ferulate bridges (Figure 4) [29]. Acid treatments easily hydrolyzed arabinofuranose side chains of arabinoxylans, including those with attached ferulates allowing the wall to relax and expand for easier access by wall hydrolyzing enzymes. Treatment of alfalfa stems with low levels of acid during ensiling increased the amount of ethanol that could be produced [30]. However, best ethanol produc‐ tion was obtained after washing stem material after the acid treatment to remove degradation products that would interfere with yeast fermentation. A problem with acid hydrolysis of cell


**Table 1.** Cell wall composition of alfalfa stems compared to corn stover and corncobs. Other hexoses include the C6 sugars galactose and mannose and other pentoses refers primarily to the C5 sugar arabinose. Data from [13] and [55].

walls especially at high temperatures is the production of furfurals that inhibit yeast. The advantage of coupling dilute acid with ensiling is avoiding the need for high temperatures. Instead utilizing the longer-term storage of the biomass to allow limited degradation of the polysaccharides while minimizing the formation furfurals and other degradation prod‐ ucts[30]. There may be highly effective means of solubilizing the cell wall (e.g., complete acid hydrolysis of all cell wall polysaccharides to monomeric sugars), but such methods are prohibitively expensive or make it difficult to remove byproducts. To prevent unwanted microbial fermentation of the released sugar, yeast-based fermentation must be maintained in a sterile environment. Providing and maintaining a sterile environment must be factored into the sequence of events from pretreatment to fermentation; it can be achieved, but at an additional cost to the overall process. From a utilization of the total biomass standpoint yeast fermentation leaves a 20 to 40% of potentially fermentable carbohydrates behind (Table 1) simply because yeast cannot deal effectively with them. This leaves a good deal of potential energy forming material off the table.

Ethanol is not the only biofuel under consideration as a product for biomass. Alternative systems for the conversion of biomass to biofuel are the syngas platform (details of this system can be found on the National Renewable Energy Laboratory website: www.nrel.gov/biomass/ biorefinery.html) and the carboxylate platform. The syngas platform requires large inputs of energy to produce effective amounts of a useful biofuel. The carboxylate platform requires undefined mixed bacterial cultures under anaerobic conditions [31] (Figure 5). One of the big advantages of this system is the flexibility of the undefined mixed bacterial cultures to handle a wide range of substrates going into the system. More importantly they do not require a sterile environment in which to function. Popular sources of mixed anaerobic cultures are sewage sludge digesters and marine sediments[31-32]. The carboxylate platform works by the process of anaerobic degradation of carbohydrates to produce volatile fatty acids primarily acetic (C2), propionic (C3), and butyric (C4) acids although other VFAs can be produced.

arabinoxylans via ferulate dimers and to lignin via ferulate bridges (Figure 4) [29]. Acid treatments easily hydrolyzed arabinofuranose side chains of arabinoxylans, including those with attached ferulates allowing the wall to relax and expand for easier access by wall hydrolyzing enzymes. Treatment of alfalfa stems with low levels of acid during ensiling increased the amount of ethanol that could be produced [30]. However, best ethanol produc‐ tion was obtained after washing stem material after the acid treatment to remove degradation products that would interfere with yeast fermentation. A problem with acid hydrolysis of cell

degraded by the addition of cellulosic enzyme cocktails.

556 Biofuels - Status and Perspective

**Figure 4.** Cell wall model showing formation of lignin in grass wall matrix. Lignin in grasses is attached to ferulates that are shuttled out into the wall attached to arabinosyl side chains of arabinoxylans. This creates a tightly integrated wall matrix of lignin with wall structural polysaccharides. Similar cross-linking most likely occurs in dicot walls except the ferulates are not likely to be the most prominent anchor points to the wall carbohydrates. Treatment of walls with hot dilute acid solutions removes most of the non-cellulosic polysaccharides opening up the matrix to be more easily

**Figure 5.** Schematic of biofuel production systems. Ethanol platform is the typical process proposed for conversion of plant biomass to liquid fuels. The carboxylate platform is an alternative method of producing liquid fuels using mixed anaerobic bacteria (ruminal microbes have advantages over typical systems utilizing sewage sludge or marine sedi‐ ments) to produce VFAs for conversion to volatile fuel components. Cultures can be manipulated to produce higher ratios of C5, C6 and C7 VFAs for more energy dense molecules.

An advantage of the carboxylate platform is the general low inputs needed to obtain materials that can be modified to produce biofuels or bio-refinery products. Pre-treatments are mini‐ mized and may be confined to particle size reduction or mild chemical treatments providing the greatest advantages[31]. Most importantly the carboxylate platform does not require an antiseptic environment in which to operate, greatly simplifying handling of raw materials going into digesters. Significant work has been done on carboxylate platforms utilizing mixed cultures from sewage sludge treatments [31, 33]. Such systems have a great deal of flexibility when it comes to handling a wide range and complexity of crop residues or other carbon based materials from agricultural practices. These organic materials may be relatively abundant and of relative low value in their present form before fermentation to VFAs. A disadvantage of the sewage sludge inoculum is the general slow conversion rate and methanogens producing large amounts of methane[31]. In the case of manure or other organic waste digesters where time is not a limiting factor this is quite acceptable and the methane can be easily captured and used as an energy source. With the right type of microbial mix, it is possible to produce longer-chain carboxylates caproate (C6) and caprylate (C8) from acetate in addition to the typical acetate, propionate, and butyrate through a process referred to as reverse β-oxidation[34]. The potential down side of this approach is the process tends to be slow and requires inhibition of methanogens to force the system to produce larger quantities of the longer-chain VFAs, e.g., n-caproate (C6) and n-caprylate (C8). Inhibition of methanogens can be efficiently achieved with compounds like bromoethane sulfonic acid, but this is relatively expensive and would be prohibitive on a large scale[31].

An alternative source of anaerobic microbes for the carboxylate platform for the conversion of plant biomass would be the cow's rumen. In comparison to waste stream anaerobic microbes, the rumen is a more specialized system having evolved to extract nutrient value out of a wide range of plant materials [35]. Although cell wall degradation and total feed utilization by dairy and beef cows can be improved, the microbial community in these ruminants has evolved to degrade fibrous plant material relatively quickly to supply needed nutrients to the animal [36]. The rumen is a mixed culture of anaerobic organisms effectively degrades carbohydrates, proteins, and fats present in feed mixtures to produce short-chain VFAs. The efficiency of this ruminal system appears to be much greater than what is in the typical waste stream sys‐ tems[37]. The advantage of a ruminant-based carboxylate platform is the ability to degrade all the organic materials (polysaccharides, proteins, fats, and oils) with the exception of the lignin within short time periods of 24-72 hours. High producing ruminants like the dairy cow must be able to extract sufficient energy from feed materials within 48 hours to support her main‐ tenance and milk production. Cow ruminant microbial communities have evolved over time to handle a diversity of substrates (i.e., easily degraded starch to more recalcitrant fiber materials). Ruminal microbial communities are quite complex with redundancy in the types of hydrolytic abilities that may come into play as substrates change coming into the cow [36]. Due to the relatively short incubation times slower growing acetogens (convert C3-C6 VFAs to acetate) and the methanogens (convert acetate to methane) do not have a chance to become well established. This in turn restricts methane production (8-15% of total energy) in this type of carboxylate platform avoiding the need to add specific methane inhibitors [36]. The small amount of methane that is produced could be captured and utilized as an energy input to maintain incubation temperatures.

**Figure 5.** Schematic of biofuel production systems. Ethanol platform is the typical process proposed for conversion of plant biomass to liquid fuels. The carboxylate platform is an alternative method of producing liquid fuels using mixed anaerobic bacteria (ruminal microbes have advantages over typical systems utilizing sewage sludge or marine sedi‐ ments) to produce VFAs for conversion to volatile fuel components. Cultures can be manipulated to produce higher

An advantage of the carboxylate platform is the general low inputs needed to obtain materials that can be modified to produce biofuels or bio-refinery products. Pre-treatments are mini‐ mized and may be confined to particle size reduction or mild chemical treatments providing the greatest advantages[31]. Most importantly the carboxylate platform does not require an antiseptic environment in which to operate, greatly simplifying handling of raw materials going into digesters. Significant work has been done on carboxylate platforms utilizing mixed cultures from sewage sludge treatments [31, 33]. Such systems have a great deal of flexibility when it comes to handling a wide range and complexity of crop residues or other carbon based materials from agricultural practices. These organic materials may be relatively abundant and of relative low value in their present form before fermentation to VFAs. A disadvantage of the sewage sludge inoculum is the general slow conversion rate and methanogens producing large amounts of methane[31]. In the case of manure or other organic waste digesters where time is

ratios of C5, C6 and C7 VFAs for more energy dense molecules.

558 Biofuels - Status and Perspective

Recently Weimer et.al., 2014 [38] demonstrated the ability of rumen microbial cultures to produce large amounts of valeric and caproic in short time periods of 48-72 hour incubations. It has been demonstrated that the addition of dilute amounts of ethanol to mixed culture fermentations in the carboxylate platform results in the extension of the short chain VFAs to medium length molecules thus capturing the fuel value of ethanol in a form that could be more easily recovered [34, 39]. What is unique and promising about the work of Weimer et.al., is the ability to speed up this process using ruminal mixed culture fermentations as opposed to the typical source of sewage digesters [38]. In addition they found that supplementing the mixture with ruminal derived *Clostridiumum kluyveri* an ethanol-utilizing bacteria resulted in produc‐ tion levels of 4.9-6.1 g/L of caproate in 48-72 hours using either switchgrass or alfalfa stems as the substrate. The level of caproate production seen by the Weimer group is similar to what

others have achieved [34, 40], but in a 10 to 30 times less time frame for incubation. Being able to generate longer VFAs increases the energy density in each molecule increasing the value of the material for liquid fuels. In addition, the longer chain VFAs are easier to extract from the fermentation media decreasing recovery costs[38-39]. For any biomass to biofuel production process a key element is being able to produce sufficient amounts of fuel molecules in short periods of time and with limited inputs. The carboxylate platform based on ruminal microbes supplemented with additional strains of more specialized bacteria (e.g., *Clostridiumum kluyveri*) appears to hold a great deal of promise for biomass conversion. Little sample preparation was needed to treat the switchgrass and alfalfa stems for biofuel production using the ruminal microbial system. The fermentation process described here could be combined with other platforms that produce ethanol. For example concept of consolidated bioprocessing (CBP) [36, 41] is considered as a possible avenue for the production of ethanol from biomass to avoid the need for the addition of expensive hydrolytic enzymes. In most cases the CBP system does not produce sufficient ethanol to be cost effective [41]. However, coupled with a ruminal microbi‐ al based carboxylate platform the limited ethanol production could be effectively utilized to produce longer chain VFAs increasing energy density of each molecule [38].

**Figure 6.** Multiple pathways for converting VFAs to volatile compounds that can serve as biofuels or as intermediates for the formation of additional organic compounds.

Volatile fatty acids must be converted to a form that increases their volatility to be good energy molecules. The medium length VFAs can be recovered by extraction [42] to allow additional modifications. Conversion of VFAs can be accomplished in different ways depending upon the tis desirable end product and its potential use. Possible conversion practices could utilize pure cultures of specific bacteria, electrochemical and thermochemical process. Useful end products that could be used for energy, solvents, or other biorefinery intermediates include ketones, aldehydes, alcohols, and alkanes (Figure 6). Due to the flexibility in the type of end product there are several avenues available to reach the desired outcome. Conversion process can be accomplished in a multitude of different ways using a single or multiple steps to reach desired products. Products such as ketones from VFAs using catalytic coupling [43] or ketones and secondary alcohols as produced in the MixAlco process [33]. The formation of volatile esters can be formed as demonstrated by Lange et.al., [44],Levy et.al., [45] or using microbial systems [46]. Production of alkanes can be achieved by decarboxylation of using pure cultures of microbes [47] or the use of electrochemical process using the Kolbe and/or the Hoefer-Moest processes [48]. The conversion of VFAs especially the medium length (C4-C6) increases volatility and at the same time decreases miscibility with water improving extraction process to isolate the biofuel molecules. The added advantage of VFA production (C2-C6 or longer) coupled with conversion technologies is the flexibility to produce a wide range of molecules that can be used for higher energy density fuel molecules or as starting molecules for other organic materials.

others have achieved [34, 40], but in a 10 to 30 times less time frame for incubation. Being able to generate longer VFAs increases the energy density in each molecule increasing the value of the material for liquid fuels. In addition, the longer chain VFAs are easier to extract from the fermentation media decreasing recovery costs[38-39]. For any biomass to biofuel production process a key element is being able to produce sufficient amounts of fuel molecules in short periods of time and with limited inputs. The carboxylate platform based on ruminal microbes supplemented with additional strains of more specialized bacteria (e.g., *Clostridiumum kluyveri*) appears to hold a great deal of promise for biomass conversion. Little sample preparation was needed to treat the switchgrass and alfalfa stems for biofuel production using the ruminal microbial system. The fermentation process described here could be combined with other platforms that produce ethanol. For example concept of consolidated bioprocessing (CBP) [36, 41] is considered as a possible avenue for the production of ethanol from biomass to avoid the need for the addition of expensive hydrolytic enzymes. In most cases the CBP system does not produce sufficient ethanol to be cost effective [41]. However, coupled with a ruminal microbi‐ al based carboxylate platform the limited ethanol production could be effectively utilized to

produce longer chain VFAs increasing energy density of each molecule [38].

**Figure 6.** Multiple pathways for converting VFAs to volatile compounds that can serve as biofuels or as intermediates

for the formation of additional organic compounds.

560 Biofuels - Status and Perspective

Typically biomass to biofuel systems are envisioned with a centrally located processing plant to handle large amounts of biomass. Unlike the grain ethanol production systems in which the grain is of relatively high density in terms of potential energy per volume, biomass tends to be much bulkier unless it is pelletized to increase bulk density [49]. When one is considering the utilization of corn stover and/or alfalfa stems these materials can be field processed into relatively high-density bales to improve the efficiency of shipping [50]. This is just one step in the complete process of collecting and moving biomass to centralized points for conversion to biofuels [51]. The challenge is keeping the collection, improving bulk density, and transpor‐ tation costs to minimal levels to help final economic returns and the minimizing the carbon footprint associated with biomass to biofuels[50]. Perhaps it would be feasible to consider on farm conversion at least for the initial steps of the conversion process. In this scenario the harvested plant material (corn stover, alfalfa stems, switchgrass, etc) would be stored on the farm more with an ensiling process compared to dry storage. This provides an opportunity to add enzymes or dilute chemicals to enhance the subsequent digestion of the materials. Size reduction could also be incorporated into the process and storing materials wet eliminates the need for rehydration for fermentation. It could be envisioned that small on farm digesters could be used to process the biomass materials to produce VFAs (select additions of pure cultures and ethanol to create products for special uses) that would be recovered and transported to conversion sites. Processing on farm eliminates the need for consolidating biomass for shipment to centralized processing plants and open opportunities for other types of storage that could enhance conversion efficiency. Recovery of the VFAs or conversion on site to intermediates followed by extraction results in a improvements in energy density and allows materials to be shipped greater distances for further processing into molecules that provide the greatest benefit either as biofuels or as precursors for other organic based materials.

One of the challenges of any biomass conversion platform is dealing with the fermentation residual materials. Lignin is a primary component of the fermentation waste and in many schemes it is recovered and burned to supply energy for other steps in the complete process. With the carboxylate platform based upon mixed ruminal microbes, one of the by products could be the microbial protein as a value-added material. In the normal rumination process, formation of microbial protein is an important component to supply needed protein to the animal. In dairy production, microbial protein helps supply critical amino acids required for milk production, especially methionine and lysine that are often low or lacking in many foragebased diets [52]. Harvesting the microbial protein after biomass conversion to biofuels could provide an important protein supplement for dairy cow diets that is enriched in methionine and lysine. The microbial proteins would be insoluble along with the typical insoluble materials, i.e., lignin and other cell wall components. Recovery of these insoluble materials would be relatively straightforward. As an alternative the lignin-microbial-carbohydrate residue from the fermentation process could be used to replace phenolic-formaldehyde based adhesives[53]. Many of the ruminal microbes contain glycocalyx materials surrounding the individual cells that help them adhere to plant materials during digestion. The glycocalyx is a glycoprotein-polysaccharide complex that surrounds the cell membrane of some bacteria[54]. It has also been demonstrated that the lignin-microbial residues from ruminal fermentations, as proposed for the carboxylate platform, could be used to replace phenol-formaldehyde compounds as adhesives in the production of plywood composites[53]. Up to 70% of the typical phenol-formaldehyde formulation could be replaced by the more environmentally friendly residues that are byproducts of ruminal-based fermentations. Even if it would not be possible to replace all of the phenol-formaldehyde adhesive, decreasing significant amounts of this material would provide for healthier composites by decreasing the amount of formaldehyde outgassing that are a human health concern[53]. Key to the effectiveness of fermentation residues is creating the correct balance of lignin, the blend of rumen microbes and the types of glycocalyx material, and other minor phenolic materials in the plant materials.
