The major problem associated with using ethanol as a fuel is the low reaction kinetics of ethanol oxidation versus methanol oxidation [9]. Traditional hydro- gen/oxygen PEM fuel cells and DMFCs typically employ Pt-based catalysts for oxidation of fuel, but pure Pt catalysts have lower catalytic activity toward ethanol. Researchers have shown that ethanol oxidation at polycrystalline platinum sur­faces showed carbon dioxide, acetaldehyde, and acetic acid as products [10], but at high concentration the major products are carbon dioxide and acetaldehyde [11-12]. This means that a portion of ethanol is completely oxidized to carbon dioxide (12 electron process) via the reaction above and a portion of ethanol is partially oxidized through the following 2-electron process:

CH3CH2OH ^ CH3CHO + 2H+ + 2e-

However, in the absence of water, the ethanol reacts with 2 ethanol molecules to form ethanol diethylacetal [13]. It is important to note that the efficiency of the system is quite different for ethanol than methanol. Methanol oxidation shows approximately 90% of products are carbon dioxide, whereas ethanol oxidation varies between 20 and 40% depending on the catalyst [13]. Even though methanol oxidation has higher conversion efficiency, the methanol by-product of methanol oxidation has much higher toxicity than ethanol (OSHA exposure limits are 1 ppm for methanol and 200 ppm for ethanol and the LD50 during inhalation for rats or mice: 203 mg/m3 for methanol and 24,000 mg/m3 for ethanol [13]).

The project will generate a continuous supply of nitrifying bacteria for addition to the system on a weekly basis. Bacteria, which are an integral and critical part of the system, cannot be shipped during the winter since cold temperatures will kill them. The pantry has grown bacteria previously for its aquaculture project. Weekly additions of balanced bacteria to the system will guarantee a large and healthy population of nitrifying bacteria within the system. Lack of bacteria will result in the accumulation of lethal concentrations of ammonia and nitrites, which can kill fish in a matter of hours. The bacteria must be in balance, i. e., equally strong populations. Since the two types of nitrifying bacteria grow at extremely different rates, it is essential to allow enough time for the two populations to equalize to similar population density. Determining population density is through default. Monitoring culture medium for ammonia, nitrite, and nitrate gives a very accurate indication of bacterial growth. All that is required for monitoring is a water quality test kit.

Fish Feed Formulation

An important secondary benefit of generating alcohol fuel from bakery waste is that the residual solids remaining after processing are largely the protein content of the bakery waste in a concentrated form. Also, waste from fish processing can be dehydrated and substituted for marine, or salt water fish meal, since the protein signature is identical to the fish being raised. Marine fish meal is becoming scarce and expensive due to the overharvesting of native fisheries. Feed components are not a minor consideration in any type of aquaculture project, since feed costs can constitute 35-90% of all production costs. When each tank is at full capacity, feed requirements will be about 900 pounds per day. Waste not appropriate for feed will be composted. Some of this compost will be used during the summer for the outdoor cultivation of common yet specific garden weeds that are rich in nutrients for feeding fish. The carbon dioxide captured during fermentation will be used in a “photobioreactor” for the growth of blue-green algae, which will also be incorporated into fish feed. This algae will also be used to seed the Greenwater System for specific species of fish we intend to raise. Equipment for the processing of fish feed will be situated in about 1/4 of the alcohol fuel greenhouse. Drying will be accomplished through the use of commercial dehy­drators and formulation through the use of a hammermill, mixer, and pelletizer. It is the goal of the project to produce all the components of fish feed from plant sources. The only feed component input outside the project is kelp meal, which supplies vital microelements.

The selected systems were modeled in Aspen Plus, a widely used process simu­lation program. In this flowsheeting program, chemical reactors, pumps, turbines, heat exchanging apparatuses, etc. are virtually connected by pipes. Every com­ponent can be specified in detail: reactions taking place, efficiencies, dimensions of heating surfaces, and so on. For given inputs, product streams can be calculated, or one can evaluate the influence of apparatus adjustments on electrical output. The plant efficiency can be optimized by matching the heat supply and demand. The resulting dimensions of streams and units and the energy balances can subsequently be used for economic analyses.

The pretreatment and gasification sections are not modelled, their energy use and conversion efficiencies are included in the energy balances, though. The models start with the synthesis gas composition from the gasifiers as given in Table 2.1

The heat supply and demand within the plant is carefully matched and aimed at maximizing the production of superheated steam for the steam turbine. The intention was to keep the integration simple by placing few heat exchangers per gas/water/steam stream. Of course, concepts with more process units demanding more temperature altering are more complex than concepts consisting of few units. First, an inventory of heat supply and demand was made. Streams matching in temperature range and heat demand/supply were combined: e. g., heating before the reformer by using the cooling after the reformer. When the heat demand is met, steam can be raised for power generation. Depending on the amount and ratio of high and low heat, process steam is raised in heat exchangers or drawn from the steam turbine: if there is enough energy in the plant to raise steam of 300°C, but barely superheating capacity, then process steam of 300°C is raised directly in the plant. If there is more superheating than steam-raising capacity, then process steam is drawn from the steam cycle. Steam for gasification and drying is almost always drawn from the steam cycle, unless a perfect match is possible with a heat-supplying stream. The steam entering the steam turbine is set at 86 bar and 510°C.

Table 2.4 summarizes the outcomes of the flowsheet models. In some concepts still significant variations can be made. In concept 4, the reformer needs gas for firing. The reformer can either be entirely fired by purge gas (thus restricting the recycle volume) or by part of the gasifier gas. The first option gives a somewhat higher methanol production and overall plant efficiency. In concept 5, one can choose between a larger recycle and more steam production in the boiler. A recycle of five times the feed volume, instead of four, gives a much higher

TABLE 2.4

Results of the Aspen Plus Performance Calculations for 430-MWth Input HHV Systems (equivalent to 380 MWth LHV for biomass with 30% moisture) of the Methanol Production Concepts Considered

HHV Output (MW)

1 IGT — Max H2, Scrubber, Liquid-Phase Methanol

Reactor, Combined Cycle

2 IGT, Hot Gas Cleaning, Autothermal Reformer,

Liquid-Phase Methanol Reactor with Steam Addition, Combined Cycle

3 IGT, Scrubber, Liquid Phase Methanol Reactor

with Steam Addition, Combined Cycle

4 BCL, Scrubber, Steam Reformer, Liquid-Phase

Methanol Reactor with Steam Addition and Recycle, Steam Cycle

5 IGT, Hot Gas Cleaning, Autothermal Reformer,

Partial Shift, Conventional Methanol Reactor with Recycle, Steam Turbine

6 BCL, Scrubber, Steam Reforming, Partial Shift,

Conventional Methanol Reactor with Recycle, Steam Turbine

1 Net electrical output is gross output minus internal use. Gross electricity is produced by gas turbine and/or steam turbine. The internal electricity use stems from pumps, compressors, oxygen separator, etc.

2 The overall energy efficiency is expressed as the net overall fuel + electricity efficiency on an HHV basis. This definition gives a distorted view, since the quality of energy in fuel and electricity is considered equal, while in reality it is not. Alternatively, one could calculate a fuel only efficiency, assuming that the electricity part could be produced from biomass at, e. g., 45% HHV in an advanced BIG/CC (Faaij et al. 1998), this definition would compensate for the inequality of electricity and fuel in the most justified way, but the referenced electric efficiency is of decisive importance. Another qualification for the performance of the system could use exergy: the amount of work that could be delivered by the material streams.

methanol production and plant efficiency. Per concept, only the most efficient variation is reported in Table 2.4.

Based on experiences with low calorific combustion elsewhere (Consonni et al. 1994; van Ree et al. 1995), the gas flows in the configurations presented here are expected to give stable combustion in a gas turbine. Table 2.4 only includes the advanced turbines. Advanced turbine configurations, with set high compressor and turbine efficiencies and no dimension restrictions, give gas turbine efficiencies of 41-52% and 1-2% point higher overall plant efficiency than conventional configurations. Based on the overall plant efficiency, the methanol concepts lie in a close range of 50-57%. Liquid-phase methanol production preceded by
reforming (concepts 2 and 4) results in somewhat higher overall efficiencies. After the pressurized IGT gasifier hot gas cleaning leads to higher efficiencies than wet gas cleaning, although not better than concepts with wet gas cleaning after a BCL gasifier.

Several units may be realized with higher efficiencies than considered here. For example, new catalysts and carrier liquids could improve liquid-phase meth­anol single-pass efficiency up to 95% (Hagihara et al. 1995). The electrical efficiency of gas turbines will increase by 2-3% points when going to larger scale

(Gas Turbine World 1997).

The utility of any biomass crop as a feedstock for ethanol production will depend in large part on its chemical composition, both in terms of the amount of potentially fermentable carbohydrates and the presence of compounds that may limit the yield of these carbohydrates. Current commercial yeast strains only utilize glucose as a substrate for ethanol production. Glucose can be derived from cellulose in the cell walls of biomass species. Therefore cellulose is of greater value than hemicellulose or pectin, polysaccharides composed of numerous sugars other than glucose. However, genetically modified yeast strains and other microorganisms are under study and under development that will use a wider diversity of hexose and pentose sugars. Reduced concentrations of hemicellulose and lignin, a phenolic polymer in the cell wall, would provide benefits to an ethanol conversion system by reducing pretreatment process inputs of heat and acid prior to cellulose addition. Also, reduced lignin content of biomass should result in high concentrations of the cell wall polysaccharides, thereby increasing the potential amount of fermentable sugars. Unfortunately, composition of biomass crops is very diverse and varies due to species, genetics, maturity, and growth environment.

A survey of 190 alfalfa plant introductions in the U. S. germplasm collection found that leaves averaged 283 g crude protein (CP) kg-1 dry matter (DM) compared to only 93 g CP kg-1 DM in stem material (Jung et al., 1997). In contrast, the neutral detergent fiber (NDF) concentration of stems far exceeded that of leaves (658 and 235 g NDF kg-1 DM, respectively). These differences are reflective of the role of stems in providing an upright growth form and supporting the leaf mass. Stems of alfalfa develop extensive xylem tissue (wood) with thick

cell walls comprised of cellulose, hemicellulose, pectin, and lignin (Theander and Westerlund, 1993; Wilson, 1993). Because leaves are the site of most pho­tosynthetic activity in alfalfa, the leaves have high concentrations of enzymes and thin cell walls to facilitate light absorption and gas exchange. Representative composition of alfalfa stem material is shown in Table 5.1. Both leaves and stems have low concentrations of simple sugars and starch (Raguse and Smith, 1966), although alfalfa roots store substantial quantities of starch (150 to 350 g kg-1 DM) (Dhont et al., 2002). Lipid content of alfalfa is quite low (~20 g kg-1 DM) (Hatfield et al., 2005).

Because alfalfa is indeterminate in its growth habit, the plants increase in size and mass until harvested or a killing frost occurs. Alfalfa leaf mass increases during maturation, but at a lower rate than the increase in stem mass (Sheaffer et al., 2000). This results in a decline in leaf percentage in the total herbage harvested that can range from more than 70% leaf during early vegetative stages to less than 20% leaf when ripe seed is present (Nordkvist and Aman, 1986). During plant maturation, alfalfa leaves change very little in CP or NDF concen­tration whereas stem CP declines and NDF content increases dramatically (Sheaf­fer et al., 2000). The reason for the increase in NDF content of alfalfa stems during maturation is the addition of xylem tissue due to cambial activity (Jung and Engels, 2002). This xylem tissue has thick secondary walls and stem xylem accounts for most cell wall material when the crop is harvested.

Cell walls of alfalfa differ from grass cell wall material because of the greater pectin content of alfalfa cell walls. In very immature alfalfa stem internodes that are growing in size, pectins can account for up to 450 g kg-1 of the cell wall. Cellulose and hemicellulose contribute 340 and 120 g kg-1, respectively, to the total cell wall, with lignin accounting for the remaining wall material, in such young internodes (Jung and Engels, 2002). At this developmental stage, all of the lignin is localized in the protoxylem vessel cells and no other tissues are lignified. Once alfalfa internodes complete their growth in length, cambium meristematic activity begins to add new xylem fiber and vessel cells that lignify almost immediately. The predominant cell wall component in these tissues is cellulose (400 g kg-1 cell wall) with the rest of the cell wall material being equally divided among hemicellulose, pectin, and lignin (Jung and Engels, 2002). Phloem fiber cells also develop thickened secondary cell walls as the plant matures; however, this secondary wall is especially rich in cellulose and does not contain lignin (Engels and Jung, 1998). Lignin is deposited in a unique ring structure in the primary wall region of phloem fiber cells. With the exception of pith paren­chyma cells, all of the other tissues in alfalfa (chlorenchyma, collenchyma, epidermis, cambium, secondary phloem, and protoxylem parenchyma) do not lignify no matter how mature the stem becomes (Engels and Jung, 1998). These tissues retain only primary cell walls that are rich in pectin. The pith parenchyma will ultimately lignify, although with only marginal secondary wall development, but usually pith parenchyma cells senesce, leaving a hollow stem cavity (Jung and Engels, 2002).

The composition of the major cell wall polysaccharides and lignin also change during maturation. Hemicellulose composition shifts from slightly more than 50% xylose residues, with the remainder being primarily to mannose, in very immature elongating stem internodes to 80% xylose residues in very mature internodes (Jung and Engels, 2002). The composition of the pectin fraction shifts less dramatically, with uronic acids increasing from 60% of the pectin to 67% with decreases in galactose and arabinose content, but no change in rhamnose con­centration. The largest shift in cell wall composition due to maturity is in mono — lignol components of lignin. The syringyl-to-guaiacyl ratio increases from 0.29 to 1.01 as alfalfa stem internodes mature (Jung and Engels, 2002).

While maturity is the single most important factor that impacts composition of alfalfa, growth environment causes some additional shifts in composition. Unfortunately these environmental impacts are complex and difficult to predict. In a study by Sanderson and Wedin (1988), alfalfa herbage from a summer regrowth harvest in one year had a substantially higher NDF concentration than observed for that year’s spring harvest (538 and 476 g NDF kg-1 DM, respec­tively); however, the same plots harvested in the following year showed a small difference between summer and spring harvests (588 and 546 g NDF kg-1 DM, respectively). Acid detergent lignin (ADL) concentration of the NDF fraction was greater for summer-harvested alfalfa in both years. During the spring growth period of the second year, air temperatures were warmer and there was less rainfall than in the first year of the study (Sanderson and Wedin, 1988). Vegetatively propagated clones of individual alfalfa plants divergently selected for stem cell wall quality traits showed environmental variability when evaluated over twelve cuttings (two locations, over two years, with three harvests per year). One clone averaged 233 g kg-1 for stem Klason lignin concentration but varied in response from 198 to 261 g kg-1 over the environments tested. Another clone selected for stem cellulose concentration ranged from 396 to 467 g kg-1 for the twelve samples (Lamb and Jung, unpublished data).

In the previous study, the impacts of temperature and moisture cannot be evaluated separately. When these two environmental factors have been evaluated independently, the major effect of moisture stress alone appeared to be on amount of cell wall accumulated by alfalfa plants as opposed to changes in cell wall composition. When rainfall was eliminated using a moveable shelter and alfalfa plots were irrigated to three field capacities (65, 88, and 112% saturation), stem cell wall concentration was reduced when the alfalfa was grown under water — deficit conditions (Deetz et al., 1994). Klason lignin concentration of the cell walls was not altered due to water-deficit and concentrations of xylose, galactose, and rhamnose in the cell wall were marginally increased and glucose was decreased, under the 65% field capacity treatment. In contrast to the impact of moisture, temperature was found not to alter cell wall concentration, but did apparently influence cell wall composition. A greenhouse study where alfalfa was grown under adequate moisture conditions indicated that higher temperatures (32°C and 26°C, day and night respectively) resulted in no changes in leaf or stem NDF concentration compared to cooler growth conditions (22°C and 16°C, day and night respectively), but ADL content of the NDF was increased by the higher temperatures (Wilson et al., 1991). However, these temperature effects should be viewed with some caution because both the NDF and ADL concentra­tions observed for the greenhouse-grown alfalfa in this study were much lower than normally observed for field grown plants.

Metallic substances that are degraded by E85 include: zinc, brass, aluminum, and lead-plated steel. Alloys containing these metals must be individually investigated to determine their E85 compatibility. For example, lead-tin alloy is not E85 compatible. Unfortunately, many vehicles use aluminum in the fuel delivery systems to save weight, including in the fuel pump, lines, fuel rail, and fuel pressure regulator. Fuel also often is allowed to contact the aluminum block of many two-stroke engines. Furthermore, older vehicles often use lead-plated steel for the vehicle fuel storage tanks. These materials will react with E85, partially dissolving in the fuel. This can contaminate the fuel system, leading to clogged fuel filters and injectors, which in turn, cause poor vehicle drivability. Aluminum can be safely used if it is hard anodized or nickel plated. Most FFVs use hard anodized aluminum for the fuel delivery systems. Also, most modern vehicles use fuel storage tanks that are made of polymer compounds (which are resistant to E85) instead of lead-plated steel; thus, this problem is only a factor in older vehicles.

Other metallic compounds that are resistant to E85 include: unplated steel, stainless steel, black iron, and bronze. These materials can be substituted for the other compounds as required.

The “wired” technique is most commonly used today for enzyme immobilization and can be employed with Nafion® polymers, as shown in Figure 12.3. However, this approach decreases the activity of the enzyme due to the change in the three­dimensional configuration of the enzyme that results from covalent bonding between the enzyme and the polymer. Another problem associated with this technique is that the enzyme is still subjected to the chemical environment of the

FIGURE 12.3 Enzyme immobilization by “wiring” technique.

matrix and not protected from its surroundings. Therefore, the enzyme can be easily denatured, and this limits the lifetime of the enzymatic catalytic activity.

Sandwich Technique

A second type of enzyme immobilization employing Nafion® is the sandwich technique in which the enzyme is trapped in between the polymer and the electrode surface, as shown in Figure 12.4. This is accomplished by simply casting the enzyme solution onto an electrode surface before casting the Nafion® suspen­sion. Sandwich techniques are powerful and successful for enzyme immobiliza­tion; however, the enzyme’s optimal activity is not retained due to the physical distress applied by the polymer. In addition to this, the diffusion of analyte through the polymer is slowed limiting its applications.

The plantation will consist of hybrid poplar trees (at 600 trees per acre) planted in a “lawn” of Dutch white clover. During the first year of growth, the young trees can tolerate no competition from weeds and require irrigation in order to become established. The project will mow the “lawn” with a bagging commercial mower to remove the clippings, which will then be dehydrated and used as a base for fish feed, or used to produce compost. Colonies of honeybees will be estab­lished to forage on the clover, since these plants produce an excellent water-white honey. As a result, the project will harvest 3 different crops from the same piece of land. The trees are ready to harvest the 5th year and will regenerate to harvestable size every 3 years thereafter. One pound of hardwood chips generates approximately 7500 BTUs of energy and poplar contains about 60% of this value — or about 4500 BTUs per pound. This reduction in heat value is more than offset by the rapid growth and regeneration. Poplar also absorbs and stores more carbon dioxide during growth than the wood gives off when used as a fuel source. Harvest will commence after leaf drop and when the ground is frozen to reduce turf damage. The project will use a feller/buncher to cut the trees, which are then placed in windrows. A self-propelled chipper reduces the entire tree into 2» or less sized chips — this size being optimal for the gasifier. Chips can then be stored in a silo that self-feeds directly into the gasifier. After this point the entire system is automatic and needs no operator — all that is required is a continuous supply of feedstock (chips). At the heart of this phase is a gasification unit, which heats and thermally degrades the biomass in a chamber devoid of oxygen. The gas generated from this heating (pyrolysis) is extracted, filtered, cooled, and stored. It contains approximately 50% of the BTUs in natural gas and can be used for electric, heat generation, or transportation. The difference here is that gasification can utilize so many other different feedstocks that are unsuitable for the first phase. There is no exhaust from this process, since there is no active combustion, as a result, there is no pollution. The “exhaust” is actually the gas produced. The gasification unit can also be configured to produce methanol (wood alcohol), which is used in the conversion (transesterification) of raw and used vegetable oil into biodiesel. It can also produce dimethylether (“DME”), a clean­burning replacement for diesel fuel, or #2 home heating fuel — depending on the configuration. Solids and nutrient ash from wood biomass pyrolysis can be incorporated into compost as a bulking/nutrient agent. Inert solids remaining from plastic and tire pyrolysis can be used for making asphalt paving products or concrete blocks. The only feedstock under consideration that is not carbon-cycle neutral is plastic.

The wet milling and dry grind fermentation processes share the same biological basis for conversion of corn starch to ethanol: starch is converted by the combined actions of heat and enzymes to glucose and maltose, which are fermented by yeast to ethanol. Starch is a mixture of two glucose polymers: amylose, a linear molecule with a-1-4 linkages, and amylopectin, a branched molecule which has the same a-1-4 linkages and also contains a-1-6 branch points. Starch forms crystalline granules in the seed [15]. The granules (Figure 4.2) are insoluble in water and, in fact, have hydrophobic interiors. Pores extend from the surface into the hollow core of the granule. Heating an aqueous starch suspension weakens the hydrogen bonds within and between starch molecules, causing swelling of the starch granules due to absorption of water. The swelling process is called gelatinization [16]. Gelatinized starch is converted to glucose in the industrial process primarily by two enzymes, alpha-amylase and glucoamylase. First, the starch polymer is hydrolyzed by alpha-amylase to shorter chains called dextrins in a process known as liquefaction because the breakdown of polymers yields a thinner solution. Finally, the dextrins are degraded to glucose and maltose (a glucose dimer) by glucoamylase. The release of simple sugars from a polymer is called saccharification [17].

Corn fiber is a coproduct of the corn wet-milling industry. It is a mixture of corn kernel hulls and residual starch not extracted during the wet-milling process. Corn fiber is composed of approximately 40% hemicellulose, 12% cellulose, 25% starch, 10% protein, 3% oil, and 10% other substances such as ash and lignin (Singh et al., 2003). Corn fiber represents a renewable resource that is available in significant quantities from the corn dry — and wet-milling industries. Approxi­mately 6.3 x 106 dry tons of corn fiber is produced annually in the United States. Typically 4.5 lb of corn fiber is obtained from a bushel (56 lb) of corn, which can be converted to about 3.0 lb of fermentable sugars. The major fermentable sugars from hydrolysis of lignocellulosic biomass, such as softwood, hardwood and grasses, rice and wheat straw, sugarcane bagasse, corn stover and corn fiber, are D-glucose and D-xylose (except that softwood also contains substantial amounts of mannose) (Sedlak and Ho, 2004). Industrial Saccharomyces yeast strains used for fermenting sugars to ethanol lack the ability to utilize xylose, one of the major end products of hemicellulose hydrolysis. This is a major obstacle for the utilization of corn fiber or other forms of lignocellulosic-based biomass.

Economically, it is important that both xylose and glucose present in corn fiber be fermented to butanol in order for this renewable biomass to be used as feedstock for butanol production. Solventogenic clostridia have an added advan­tage over many other cultures as they can utilize both hexose and pentose sugars (Singh and Mishra, 1995) released from lignocellulosic biomass upon hydrolysis to produce butanol. Fond and Engasser (1986), during their evaluation of the fermentation of lignocellulosic hydrolysates to butanol by C. acetobutylicum ATCC 824, demonstrated that the culture utilized both xylose and glucose, although xylose was utilized more slowly than glucose and also supported lower butanol production. However, C. beijerinckii BA101 has been shown to utilize xylose and can effectively coferment xylose and glucose to produce butanol (Ebener et al., 2003). Parekh et al. (1988) produced acetone-butanol from hydroly­sates of pine, aspen, and corn stover using C. acetobutylicum P262. Similarly Marchal et al. (1984) used wheat straw hydrolysate and C. acetobutylicum, while Soni et al. (1982) used bagasse and rice straw hydrolysates and C. saccharoper — butylacetonicum to convert these agricultural wastes into butanol.

An important limitation of corn fiber utilization comes from the pretreatment and hydrolysis of corn fiber to glucose and xylose. Saccharification of corn fiber can readily be achieved by treatment with dilute H2SO4. However, this acid — catalyzed reaction leads to the degradation of glucose to hydroxy methyl furfural (HMF) and xylose to furfural at the temperatures of hydrolysis, resulting in inhibition of fermentation by these degradation products. Other degradation prod­ucts include syringaldehyde, acetic, ferulic, and glucuronic acids. The formation of these degradation products lowers the yield of fermentable sugars obtained from the corn fiber and the degradation products are inhibitory to yeast and bacterial fermentations. C. beijerinckii BA101 is able to completely utilize enzyme-hydrolyzed corn fiber to produce acetone-butanol, but performed poorly in the bioconversion of acid-hydrolyzed corn fiber to acetone-butanol due to the presence of inhibitory compounds generated during hydrolysis (Ebener et al., 2003). Therefore, the development of strains that can tolerate the inhibitory compounds generated during acid pretreatment and hydrolysis of corn fiber remains a priority.

Pioneering work in DMFC technology was undertaken by Shell, Exxon-Alsthom, Allis Chalmers, and Hitachi during the 1960s and 1970s [1,6]. Research focused on developing noble metal catalysts in liquid acid and alkaline electrolytes [1]. During this period, the mechanistics of methanol oxidation at Pt-based catalysts were studied [1,7,8]. While fundamental understanding of methanol oxidation became more clear, maximum current densities remained low. It was thought the limitation on current density was largely due to inadequate ionic conduction and stability of the PEMs employed in the fuel cells.

Membrane Development

In the mid 1960s, DuPont introduced the perfluorinated superacid membrane Nafion®. It was considered a major advancement in PEM materials [3]. Earlier, less effective PEM materials included polystyrene-based ionomers and heteroge­neous sulfonated divinylbenzene cross-linked polystyrene [2,3]. These early PEMs had poor long-term chemical stability and low proton conductivity. Nafion performance was considerably better than these other materials, however, it was quickly recognized that methanol crossover through Nafion would limit its use­fulness as a separator in DMFCs [9]. Crossover diminishes cell efficiency and occurs when fuel that is fed to the anode crosses through the membrane to the cathode and reacts directly with the oxidant. The process also poisons the cathode electrocatalyst with methanol oxidation products.

In the mid-1980s, Nafion membranes became more widely available and solubilized Nafion was introduced to the market. As Nafion became more widely available, PEFC research began in earnest and has since continued. The most notable improvement is a dramatic reduction in catalyst loading at ever-increasing power outputs [6]. Another important discovery was that the stability of the Pt electrocatalyst is greatly enhanced when Nafion is added to the electrocatalyst later. These developments set the stage for a revival of interest in low-temperature DMFCs during the late 1990s [6,10,11,12]. Figure 9.3 demonstrates the increase in DMFC research activity over this period.

As will be seen in later sections of this chapter, current efforts in DMFC research include minimizing methanol crossover through the separator of DMFCs while maintaining high proton conductivity, developing methanol tolerant oxygen reduc­tion catalysts, and identifying more cost-effective methanol oxidation catalysts [13].