The IGT gasifier (Figure 2.2) is directly heated, which implies that some char and/or biomass are burned to provide the necessary heat for gasification. Direct heating is also the basic principle applied in pressurised reactors for gasifying coal. The higher reactivity of biomass compared to coal permits the use of air instead of pure oxygen.

Steam + oxygen

This could be fortuitous at modest scales because oxygen is relatively costly (Con — sonni and Larson 1994a). However, for the production of methanol from biomass, the use of air increases the volume of inert (N2) gas that would have to be carried through all the downstream reactors. Therefore, the use of oxygen thus improves the economics of synthesis gas processing. Air-fired, directly heated gasifiers are considered not to be suitable before methanol production.

This gasifier produces a CO2 rich gas. The CH4 fraction could be reformed to hydrogen, or be used in a gas turbine. The H2:CO ratio (1.4:1) is attractive to produce methanol, although the large CO2 content lowers the overall yield of methanol. The pressurized gasification allows a large throughput per reactor volume and diminishes the need for pressurization downstream, so less overall power is needed.

The bed is in a fluidized state by injection of steam and oxygen from below, allowing a high degree of mixing. Near the oxidant entrance is a combustion zone with a higher operation temperature, but gasification reactions take place over the whole bed, and the temperature in the bed is relatively uniform (800-1000 °C). The gas exits essentially at bed temperature. Ash, unreacted char, and particulates are entrained within the product gas and are largely removed using a cyclone.

An important characteristic of the IGT synthesis gas is the relatively large CO2 and CH4 fractions. The high methane content is a result of the nonequilibrium nature of biomass gasification and of pressurized operation. Relatively large amounts of CO2 are produced by the direct heating, high pressure, and the high overall O:C ratio (2:1). With conventional gas processing technology, a large CO2 content would mean that overall yields of fluid fuels would be relatively low. The synthesis gas has an attractive H2:CO ratio for methanol production, which

reduces the need for a shift reactor. Since gasification takes place under pressure, less downstream compression is needed.

When operated with higher steam input the IGT gasifier produces a product gas with a higher hydrogen content. This maximum hydrogen mode is especially useful if hydrogen would be the desired product, but the H2:CO ratio is also better for methanol production. However, the gasifier efficiency is lower and much more steam is needed.

Wet-Milling and Dry-Grind Corn Processes for Ethanol Fermentation

Corn is prepared for ethanol fermentation by either wet milling [20] or dry grinding [16] (Figure 4.4). One quarter of the ethanol produced in the United States comes from large-capacity wet-milling plants, which produce ethanol along with a variety of valuable coproducts such as pharmaceuticals, nutriceuticals, organic acids, and solvents. Dry-grind facilities, which account for the remainder of domestic ethanol production, are designed specifically for production of eth­anol and animal feed coproducts. Due to the relatively lower capital cost of dry — grind plants and the spread of ethanol plants out of the heart of the U. S. cornbelt, new plants under development and construction are dry-grind facilities.

Although both the dry-grind and wet-mill processes produce ethanol, they are very different processes. In dry grinding, dry corn is ground whole and fermented straight through to ethanol. The only coproduct, distillers dry grains with solubles (DDGS), is sold as animal feed. DDGS, which consists of the dried residual materials from the fermentation, contains the nonfermentable parts of the corn and the yeast produced during the fermentation. CO2 can also be collected and sold to soft-drink producers, but represents a low-profit and limited market.

In wet milling, by contrast, corn kernels are fractionated into each of their major individual components: starch, gluten, germ, and fiber. This imparts two very important advantages compared to dry grinding. First, the parts of the corn can each be marketed separately. So, the germ is used to produce corn oil, the gluten is sold as a high-protein feed to the poultry industry, and the fiber is combined with liquid streams, dried, and sold as a low-protein animal feed. Second, the wet mill produces a pure starch steam, which allows for the starch to be made into numerous different products. In addition to being fermented to ethanol, the starch can be modified for use in textiles, paper, adhesives, or food. Maltodextrins and high-fructose corn syrup, the major sweetener used by the U. S. food industry, are made enzymatically from starch. The starch can also be con­verted enzymatically to a fairly pure glucose stream and then fermented to any number of products. A partial list includes amino acids, vitamins, artificial sweet­eners, citric acid, and lactic acid, in addition to ethanol. If ethanol is produced, the yeast can be spray-dried and marketed as distillers yeast, a high-protein, low — fiber product suitable for feeding animals and fishes. Although no wet mill makes all of these products, it is not unusual for large facilities to have multiple starch product streams.

Dry-grind plants do not have the capability to ferment corn starch to these products in part because the additional products are nonvolatile and, therefore, cannot be simply separated by distillation from all of the other material in the fermentation. In summary, a wet mill that converts all its starch to ethanol produces at least two or three additional high-value products compared to a dry-grind facility. Of course, these additional products are realized only with much higher capital expenses. As discussed later in this chapter, there are several efforts under­way to develop less capital-intensive processes for either totally or partial frac­tionating corn that would be suitable for implementation at dry-grind facilities.

Batch fermentation is the most commonly studied process for butanol production. In the batch process the substrate (feed) and nutrients are charged into the reactor that can be used by the culture. In a batch process, a usual substrate concentration of 60-80 gL-1 is used as higher concentration results in residual substrate being in the reactor. The reaction mixture is then autoclaved at 121°C for 15 minutes followed by cooling to 35-37°C and inoculation with the seed culture. During cooling, nitrogen, or carbon dioxide is swept across the surface to keep the medium anaerobic. After inoculation, the medium is sparged with these gases to mix the inoculum. Details of seed development and inoculation have been pub­lished elsewhere (Formanek et al., 1997; Qureshi and Blaschek, 1999a). Depend­ing on the size of the final fermentor, the seed may have to be transferred several times before it is ready for the production fermentor.

Various substrates can be used to produce butanol including corn, molasses, whey permeate, or glucose derived from corn (Qureshi and Blaschek, 2005). However, some substrates may require processing prior to fermentation, known as “upstream processing,” such as dilution, concentration, centrifugation, filtra­tion, hydrolysis, etc. The usual batch fermentation time lasts from 48 to 72 h after which butanol is recovered, usually by distillation. During this fermentation period, ABE up to 33 gL-1 is produced using hyperbutanol producing C. beijer­inckii BA101 (Chen and Blaschek, 1999; Formanek et al., 1997). This culture results in a solvent yield of 0.40-0.42 (Formanek et al., 1997). The ABE con­centration in the fermentation broth is limited due to butanol inhibition to the cell. At a butanol concentration of approximately 20 gL-1, strong cell growth inhibition occurs that kills the cells and stops the fermentation. Butanol produc­tion is a biphasic fermentation where acetic and butyric acids are produced during acidogenic phase followed by their conversion into acetone and butanol (solven — togenic phase). During the acidogenic phase, the pH drops due to acid production and subsequently rises during solvent production. At the end of fermentation, cell mass and other suspended solids (if any) are removed by centrifugation and sold as cattle feed. Figure 6.2 shows fermentation profile of butanol production in a typical batch fermentation process from cornstarch using C. beijerinckii BA101.

Butanol can be produced both by using corn coproduct from i) corn dry-grind and ii) wet-milling processes. During the dry-grind process corn fiber and germ are not removed prior to fermentation. At the end of fermentation (after starch utilization during fermentation), corn fiber and other insoluble solids are removed by centrifugation, dried, and sold as cattle feed. The dried solids are known as “Distillers Dry Grain Solids” or DDGS. On the contrary, during the wet-milling

process, corn fiber and germ are removed prior to fermentation. In this process, cornstarch can be converted to any of the three products (liquefied cornstarch, glucose syrup, or glucose) each of which is fermentable by C. beijerinckii to produce butanol. It should be noted that often corn refineries add sodium met­abisulfite during the wet-milling operation as a corn kernel softening agent and preservative to the liquefied cornstarch. The presence of sodium metabisulfite may interfere with the direct fermentation of the liquefied cornstarch. However, glucose syrup or glucose does not contain any such fermentation inhibitors. The unit operations that are applicable to the corn dry-grind and wet-milling fermen­tation of butanol are given in Table 6.2.

During the 1940s and 1950s, production of butanol on an industrial scale (Terre Haute, IN, and Peoria, IL) was carried out using large fermenters ranging in capacity from 200,000 to 800,000 L. The industrial process used 8-10% corn mash, which was cooked for 90 min at 130-133°C. Corn contains approximately 70% (dry weight basis) starch. The use of molasses offers many advantages over using corn, including the presence of essential vitamins and micronutrients (Paturau, 1989). In industrial processes, beet and invert and blackstrap molasses were diluted to give a fermentation sugar concentration of 50 to 75 gL-1, most commonly 60 gL-1. The molasses solution was sterilized at 107 to 120°C for 15 to 60 min followed by adding organic and inorganic nitrogen, phosphorus, and buffering chemicals. The yield of solvent using C. acetobutylicum was usually low at 0.29-0.33. Distillation has been the method of choice to recover butanol; however, during the last two decades a number of alternative techniques have been investigated for the economical recovery of butanol, which will be discussed in the recovery section.

Discussions of performance targets and efficiencies for DMFCs are complicated due to the wide-ranging conditions, fuel and oxidant sources, and intended appli­cations for DMFCs. In this section, a survey of performance data and examples of target system requirements listed by government agencies are used to give a sense of the state of the art. Also, targets set by researchers in the literature are discussed.

In 2002, Jorissen et al. suggested DMFC performance targets to compete in terms of efficiency with reformate fed PEFCs [24]. The target they set for a DMFC is a power density of 250 mW cm-2 at a cell voltage of 500 mV and that furthermore, parasitic power loss due to methanol crossover should be no more than 50 mA cm-2 at a power density of 250 mW cm-2.

In a 1999 review of advanced electrode materials for use in DMFCs, Lamy and Leger discussed the suitability of a number of energy systems in relation to DMFCs for use in automobiles [17]. Secondary batteries (e. g., Li-ion) are limited by recharge time and power density (100-150 Wh kg-1 at maximum). PEFCs are attractive with specific power densities on the order of 1000 W kg-1 and specific energy density >500 Wh. Energy density of pure H2 is 33 kWh kg-1 but storage concerns make it less attractive and less efficient. Performance characteristics of DMFCs circa 1999 is 200 mA cm-2 at 0.5 V, or 100 mW cm-2 with electrocatalyst loadings under 1 mg cm-2.

Performance targets for a complete DMFC power system were posted in the Spring of 2005 by the U. S. Army Operational Test Command (OTC). The spec­ifications are target requirements for a ruggedized DMFC power plant for use in the field on armored and other military vehicles [25]. The specifications outline threshold requirements and objective targets for the power system. A summary of the requirements are listed in Table 9.1. In an effort to meet the objectives listed in Table 9.1, a 300-W prototype DMFC power plant was developed by T. Valdez and his team at the Jet Propulsion Laboratory [26]. The demonstration

power plant was designed for 100 hours of continuous operation and used 80 cells with active areas of 80 cm2. The electrocatalyst was PtRu at the anode and cathode. The plant generated 370 W during bench testing and had a start-up time of 18 minutes. The plant was operated continuously for 8 hours, generating a lower than expected power of 50 W. The continuous operation test was ended due to water accumulation in the stack exhaust manifold.

Subsequent to testing of the prototype power plant, the stack was torn down and components evaluated. The wettability of the cathodes of the MEAs had increased and evidence of the ruthenium migration was observed. These obser­vations were the impetus to study of the long-term stability of DMFC MEAs. The team at JPL individually ran four MEAs on a single-cell test stand for 250 hours. All of the MEAs showed irreversible voltage decay ranging from 0.2 to 0.6 mV hr-1 at a current density of 100 mA cm-2 that resulted in an average decline in power of 20%. However, unlike when the MEAs were run as compo­nents of the stack in the prototype power plant, the individually run MEAs showed no evidence of electrocatalyst migration. The important issue of electrocatalyst migration will be addressed again in the final section of this chapter.

According to Knights et al. at Ballard Power Systems, fuel cell power plants used in automobile, bus, and stationary applications require operational lifetimes on the order of 4000, 20,000, and 40,000 hours, respectively [27]. The degradation rate of the power supply is set by the beginning-of-life (BOL) and end-of-life (EOL) performances; a degradation rate on the order of 10 to 25 p V hr-1 is common for DMFCs. The group studied the strategy of load cycling in DMFCs to reduce performance degradation caused by water build-up at the cathode with time.

Ball Aerospace is developing a personal DMFC power system to meet the needs of the U. S. foot soldier [28]. It was developed under the Defense Advanced Research Projects Agency (DARPA) Palm Power program and produces average power of 20 W at 12 V and has a 30-W peak power. The unit operates for 50 hours on the fuel provided by one fuel cartridge, and is ten times lighter than the equivalent battery power plant; weighing in at three pounds with full fuel complement.

Yi et al. characterize the changes in MEA morphology of a single-cell DMFC run for a little longer than three days [29]. Long-term stability of the cell and electrocatalyst are important questions. The cell was run at 100 mA cm-2 and suffered from irrecoverable performance degradation, degrading at the rate of 1.0 to 1.5 mV hr-1. Following the run, Yi and his group found signs of delamination between the layers of the MEA and that both of the carbon-supported electrocat­alysts, PtRu/C on the anode and Pt/C on the cathode, had undergone a particle size redistribution resulting in larger particle sizes on average. The redistribution for the PtRu electrocatalyst was more pronounced than for Pt and more severe in the anode.

An assessment of the state of the art in DMFC performance can be made from relevant data from references in this chapter; data are listed in Tables 9.2 and 9.3. Where possible, the data listed from a particular reference includes data for the “best” test cell and the associated control cell. The best test cell is considered the one with highest maximum power density. The control cell is usually of a typical Nafion MEA construction consisting of carbon-supported PtRu on the anode, car­bon-supported Pt on the cathode and a Nafion 115 membrane as the separator. Efforts have been made to include operating conditions and loadings. Where an entry is listed as “n/a” the value for that parameter is not available. That is, the reference does not explicitly state the value of that parameter.

Robert Haber

One Accord Food Pantry, Inc.

Troy, New York

CONTENTS

Introduction………………………………………………………………………………………………….. 250

The Project — Phase 1………………………………………………………………………………….. 251

Alcohol Fuel………………………………………………………………………………………. 251

Municipal Solid Waste (MSW) Fuel — Wood and Cardboard…………. 251

Carbon-Cycle Neutral……………………………………………………………… 251

Procedure…………………………………………………………………………………. 252

Potential Savings……………………………………………………………………… 252

Aquaponics……………………………………………………………………………………….. 252

Fish Produced………………………………………………………………………….. 253

Auto-Feeders……………………………………………………………………………. 253

Roof of the Structure………………………………………………………………. 254

Floor of the Structure………………………………………………………………. 254

Balanced Diet………………………………………………………………………….. 255

Efficient Use of Water…………………………………………………………….. 255

Power Generation…………………………………………………………………….. 255

Potential Yield…………………………………………………………………………. 255

Smaller Family-Sized Unit……………………………………………………….. 256

Bacteria Production…………………………………………………………………. 256

Fish Feed Formulation………………………………………………………………………. 256

Fish Hatchery and Seedling Greenhouse…………………………………………… 257

Hatchery………………………………………………………………………………….. 257

Greenwater System………………………………………………………………….. 258

Seedling Greenhouse……………………………………………………………….. 258

Energy Plantation……………………………………………………………………………… 258

Compost……………………………………………………………………………………………. 259

Processing…………………………………………………………………………………………. 259

Technology Transfer — Website………………………………………………………. 260

The Project — Phase 2………………………………………………………………………………….. 260

Abundance of Biomass……………………………………………………………………. 260

Saving Family Farms………………………………………………………………………… 261

Food Imported into the Northeast……………………………………………………. 261

Reserve Food Supply………………………………………………………………………… 261

Energy Plantation……………………………………………………………………………… 262

Growth of Fish Feed from Plant Sources………………………………………….. 262

Compost-Based Aquaponic Greenhouses………………………………………… 263

Using Vertical Space — Potatoes in Scrap Tires and

Strawberries…………………………………………………………………………….. 263

References…………………………………………………………………………………………………….. 264

INTRODUCTION

The evolution of this project took a period of over 20 years. Originating in the pre-Reagan era of what we once thought were high gasoline prices, the concept was to simply make ethanol for electric or transportation use and feed the by­products to pigs and chickens. This still left remaining waste to manage, however. With the change of politics and policies, all federal grants for alcohol research were cancelled. As a result, the concept went unfulfilled for 20 years. But the world is a different place now. Today, the concept has evolved to design and implement a zero-discharge, closed, recirculating, environmentally isolated sys­tem, which produces microelement-enhanced, high-quality protein food using municipal solid waste as a source for nonpetroleum power generation. After having been the 13-year director of a rural food pantry, which met the emergency food needs of over 40,000 rural needy a year (half of whom were children), I enjoyed a unique perspective of the massive amounts of food waste that are discarded daily, especially breads and bakery sweets. As a resource for this project, the huge quantity of useable bakery waste was staggering and dictated the type of fuel to be made. At issue for the project were not only the need to generate heat and electricity, but also the need to have an ingredient base for on­site manufactured fish feed. These three expenses (heat, electric, and feed) com­prise the bulk of all operating expenses associated with the long-term success or failure of the project. Reducing or eliminating these expenses would then enhance the economic viability and potential success of the project. Of critical importance were the ingredients for the feed, since it was the only nutrient input into the system for both fish and plants. Only one type of fuel met all three needs — ethanol — and in particular, ethanol from bakery waste. Additionally, and for the purposes of this project, were the by-products of fermentation (carbon dioxide and DDGS (distillers dried grains and solubles)) and combustion (carbon dioxide and water vapor). The following is an in-depth description of the project.

THE PROJECT — PHASE 1

The first phase of this project is composed of the following subsystems:

1. Alcohol fuel.

2. Solid MSW fuel — wood and cardboard.

3. Aquaponics.

4. Fish feed formulation.

5. Fish hatchery.

6. Energy plantation.

7. Compost.

8. Processing.

9. Technology transfer — website.

Two reactor types predominate in plants built after 1970 (Cybulski 1994; Kirk — Othmer 1995). The ICI low-pressure process is an adiabatic reactor with cold unreacted gas injected between the catalyst beds (Figure 2.5, left). The subsequent heating and cooling leads to an inherent inefficiency, but the reactor is very reliable and therefore still predominant. The Lurgi system (Figure 2.5, right), with the catalyst loaded into tubes and a cooling medium circulating on the outside of the tubes, allows near-isothermal operation. Conversion to methanol is limited by equilibrium considerations and the high temperature sensitivity of the catalyst. Temperature moderation is achieved by recycling large amounts of hydrogen-rich gas, utilizing the higher heat capacity of H2 gas and the higher gas velocities to enhance the heat transfer. Typically a gas phase reactor is limited to about 16% CO gas in the inlet to the reactor, in order to limit the conversion per pass to avoid excess heating.

The methanol synthesis temperature is typically between 230 and 270°C. The pressure is between 50 and 150 bar. Higher pressures give an economic benefit,

FIGURE 2.5 Methanol reactor types: adiabatic quench (left) and isothermal steam raising (right).

since the equilibrium then favors methanol. Only a part of the CO in the feed gas is converted to methanol in one pass through the reactor, due to the low temperature at which the catalyst operates. The unreacted gas is recycled at a ratio typically between 2.3 and 6.

The copper catalyst is poisoned by both sulfur and chlorine, but the presence of free zinc oxides does help prevent poisoning.

As mentioned earlier in the discussion of dry-grind technology, the dry-grind process is being adapted to capture some of the advantages of the wet-milling system. The Quick Germ [34] and Quick Fiber [35] fractionation methods increase processing efficiency as well as the value of feed coproducts. These methods add technology to the beginning of the dry-grind process, removing the germ and fiber fractions of the corn kernel prior to starch processing. First, the corn is soaked in water for 3-12 hours to hydrate the germ, which is recovered by density separation. Next, the specific gravity of the mash is adjusted and pericarp fiber is recovered using hydrocyclones. The starch is then fermented in the traditional dry-grind process, and the germ and fiber can be processed sepa­rately into other value-added products: corn oil (from germ) and corn fiber oil and corn fiber gum (from corn fiber). Corn fiber oil, which is distinct from traditional corn germ oil, is potentially a valuable coproduct because it contains phytosterols known to have cholesterol-lowering properties. Another process, known as enzymatic milling or the E-mill process, refers to the addition of proteases and amylases prior to germ separation [36]. E-milling allows recovery of the gluten, while avoiding a full steeping process and the health and environ­mental concerns associated with sulfite.

Because the germ and fiber do not enter the fermenter, the fermentation residuals have lower fiber content and correspondingly higher protein content, resulting in a higher-value feed product. On the process side, these modified milling technologies increase the effective capacity of fermentation tanks, because removal of the germ and fiber frees reactor volume for additional starch. Bringing these modified processes to the dry-grind plant would therefore increase the facility’s ethanol production by 8 to 27% [37].

Over the last 30 years, ethanol has been used widely to blend with gasoline in the United States, Brazil, and other countries. In the United States, ethanol is usually blended in a mix of 10% ethanol and 90% gasoline. Early in the use of ethanol blends, this blend was referred to as gasohol, but it is now commonly referred to as E10. The purpose of blending a small percentage of ethanol into gasoline is to oxygenate the fuel for cleaner combustion and fewer carbon mon­oxide and hydrocarbon exhaust emissions. The most common additive to gasoline to improve oxygen is methyl tertiary butyl ether (MTBE), but it is an extremely toxic chemical that has been found to contaminate groundwater. A comparison of the emission of sulphur, olefins, carbon dioxide, aromatics, and NOx from MTBE oxygenated gasoline and ethanol oxygenated gasoline is shown in Table 7.1. The U. S. Environmental Protection Agency is beginning the process of eliminating MTBE from gasoline (6). Iowa and South Dakota have already phased out MTBE (2). If MTBE were completely replaced with ethanol, it would produce a 12-billion-gallon market for ethanol each year (2), which is considerably more

ethanol than is currently produced in the United States. To be considered an oxygenated gasoline, the fuel must contain at least 2.7% oxygen by weight. This can be obtained by blending 15% by volume MTBE or 7.5% by volume ethanol, but there is a difference in emissions between the two as shown below (2). Ethanol produces dramatically fewer sulfur and olefin emissions, but comparable emis­sions of other environmental hazards.

It is important to note most countries do not blend the minimum amount (7.5%) of ethanol for use as an oxygenate. Each country has its own concentration of ethanol to blend with gasoline. E10 has been the choice in the United States. It has also been the choice in areas of Canada. From 1929 to 1957, E10 was the only type of gasoline sold in Queensland. In 2001, E10 was reintroduced to Queensland by the government (7). On the other hand, all gasoline in Brazil is 22% ethanol (E22) (7). Finland has shown that E15 (15% ethanol/85% gasoline) vehicles can operate with stock engines (8). Other countries have considered or employed variations in ethanol concentration from E10 to E25. Table 7.2 below shows how relative emissions change as a function of ethanol concentration. It is important to note that carbon monoxide, hydrocarbon, and NOx emissions

decrease with increasing ethanol concentration, but aldehyde emissions increase with ethanol concentrations. Also, Thailand has shown that the emission rates of benzene, toluene, and xylene are decreased in cars using E10 and E15 fuels (9). This decrease in emissions is important due to the major health effects (including leukemia) of long-term inhalation of benzene and toluene (10). However, E10 and E15 fuels show an increase in formaldehyde and acetaldehyde emissions and exposure to formaldehyde and acetaldehyde has been shown to cause eye irrita­tion, respiratory problems, and nervous disorders (9).

It is also important to consider that E10 is considered an oxygenated fuel, but not an alternative fuel. E85, E95, and biodiesel have large enough biofuel concentrations to be considered alternative fuels, but E10 is simply considered an oxygenated fuel. From 1992 to 1998, the U. S. consumption of vehicle fuel increased by 14.5% (11). However, the U. S. consumption of alternative fuels increased 49.1% and the U. S. consumption of oxygenated fuels has increased 96.9% (11). This shows that more consumers are using alternative and oxygenated fuels today than in 1992. However, there has only been a 21.6% increase in the use of ethanol as an oxygenate (12). This will likely increase as MTBE is phased out due to environmental issues.

Ethanol is an easy fuel to work with because it is liquid at room temperature, can be stored in conventional fuel tanks, is less toxic than many fuels, and is easy to splash blend with gasoline at any stage of the production/distribution process.

Until the hydrogen economy is well established, it is more sensible to generate electricity directly from alcohols or hydrocarbons. SOFCs may become very attractive for portable, transportation, and stationary applications if alcohols and hydrocarbons can be utilized directly without applying any fuel pretreatments. The main advantage of liquid hydrocarbons is their relatively higher energy density compared to alcohols. However, most hydrocarbon fuels such as natural gas, bioderived gases, diesel, and gasoline contain impurities such as hydrogen sulfide and halogens, which may lead to poisoning of the SOFC electrode mate­rials. Particularly, sulfur content in such fuels should be reduced through pre­treatments to prevent the fuel cell electrodes from poisoning. Alternatively, progress is being made toward development of sulfur-resistant electrode materials for long-lasting operation of SOFCs using hydrocarbon fuels, which generally contain sulfide compounds in relatively high concentrations. For instance, a highly sulfur-tolerant anode composed of Cu, CeO2, and YSZ was developed to operate a SOFC using hydrogen with H2S levels up to 450 ppm at 1073 K [6]. Another study based on LaxSr1-xVO3- as anode material for SOFC showed a maximum power density of 135 mW/cm2 at 280 mA/cm2 when the fuel was a 5% H2S-95% H2 mixture at 1273K [7].

The advantage of liquid oxygenated hydrocarbons, such as alcohols, in com­parison to gasoline is that they are cleaner (low sulfur content) and can be derived from agricultural by-products and biomass as a renewable energy source. Alcohol is an ideal fuel for the fuel cells because of ease of transportation, storage, and handling, as well as their high energy density. Partially oxidized (hydrated) fuels may be easily reformed, such as alcohols, as they contain oxygen, in a liquid form. Since water is often used for internal reforming of the fuel, water solubility of alcohols (especially methanol, ethanol, and propanol) offers the advantage that additional fuel processing may not be necessary for operation of the fuel cell.

An anode-supported SOFC utilizing direct alcohol was reported by Jiang and Virkar [8]. A thin-film YSZ electrolyte was deposited on a Ni-YSZ anode with a composite of Sr-doped LaMnO3 and YSZ as a cathode. Pure methanol and an equivolume mixture of ethanol and water were used as fuels to operate the cells over a range of temperatures. Power densities achieved with ethanol and water mixtures were between 0.3 W/cm2 at 650°C and 0.8 W/cm2 at 800°C, and with methanol between 0.6 W/cm2 at 650°C and 1.3 W/cm2 at 800°C as shown in Figures 11.2 and 11.3. Carbon deposition on the electrodes was not observed when methanol was used as fuel. On the other hand, maximum power density using humidified H2 was 1.7 W/cm2 at 800°C. This indicates that a lack of H2 in the fuel may substantially increase concentration polarization thus limiting the performance of the cell.

Another study on direct-alcohol SOFCs reported a comparison of methanol, ethanol, propanol, and butanol as fuel sources [9]. With an increasing carbon number of the alcohol, a decrease in cell voltage was observed, which was attributed to slower decomposition and/or reforming kinetics of alcohols. Decreasing operational temperatures led to an increase of unreacted alcohols, aldehydes, and aromatic compounds. Thermochemical calculations were used to reveal the equilibrium amounts of reaction products of fuels during fuel cell operation [10,11]. Figure 11.4 shows the limit lines of carbon deposition as a function of temperature in the C-H-O diagram. No carbon deposition is expected if the carbon-to-oxygen ratio is less than unity. It is shown that the addition of

C

Further development of electrode materials that do not require introduction of water, will lead to better performance of the SOFCs, provided carbon deposi­tion can be suppressed. Recent studies showed that coking issues can be resolved through selection of appropriate catalysts and anode materials in fuel cell devel­opment [4,5]. Because nickel is an excellent catalyst for hydrocarbon cracking, Ni/ZrO2 cermets are used as anode materials for YSZ-based SOFCs. As mentioned earlier, these cermets can only be used in hydrocarbon or alcohol fuels if excess water is present to ensure complete fuel reforming. Mixing /Vo-octane with water, alcohol, and surfactant to produce an oil in water microemulsion was successful in reducing the carbon formation significantly, while retaining a high octane number [12]. It has been shown that the problem of carbon deposition may be avoided by using a copper-ceria anode [13] or applying an yttria-ceria interface between YSZ and Ni-YSZ cermet anode [3]. A nickel-free SOFC anode, La0.75Sr0 25Cr0.5Mn0 5O3 with comparable electrochemical performance to Ni/YSZ cermets was developed for methane oxidation without using excess steam [5]. A recent study showed that a Ru-CeO2 catalyst layer with a conventional anode allows internal reforming of /Vo-octane without cocking and yields stable power densities of 0.3 to 0.6 W/cm2 in a SOFC design operating at intermediate tem­peratures [14].

The seedling greenhouse is a smaller structure designed to provide replacement plants for those harvested in the aquaponics greenhouse. Very shortly after sprout­ing, the young plants are switched to being irrigated with fish tank effluent. This provides the maximum growth potential in the shortest amount of time. This part of the project will be an extremely rich learning environment.

Energy Plantation

The majority of the energy crops grown will be hybrid poplar and the project will utilize whole tree technology in the harvest and use of the trees. This means the entire tree is cut during the winter and processed into chips. The stump does not die — the root system is well established by this time — but rather the stump sprouts new shoots, which are then trimmed to one central leader. Because of the extensive root system, the tree reaches its original size again in 3 years rather than 5 and is ready for harvest once more. At least one year will be needed to assess the potential of the property, but in particular to plant cover crops and build soil fertility before planting energy crops. Given these restrictions, it will take 5 years from planting before the project is able to harvest a first crop of trees. Also, the project will be experimenting with salix (willow), alfalfa, and rapeseed (canola) crops during this 5-6 year lag in time. The project will also establish fish grow-out tanks on this land. However, the effluent will be used to irrigate outdoor energy crops during the summer and only produce vegetables during the fall, winter, and spring. The heart of this phase of the project is the gasification unit, which converts crop residues, stems from hay, and wood chips into synthesis gas. Feedstock is placed in the unit, which heats it to about 1500 degrees. The heat chamber is without oxygen so the feedstock will not burn, but rather give off a gas that is then filtered, cooled, and stored. This fuel cannot only provide heat and electricity, but also used directly in an internal combustion engine. The unit can also produce alcohol fuel and diesel fuel, #2 home heating fuel, and accommodate rubber tires as a feedstock. Excess heat from the operation of the gasifier will be used to dehydrate certain components for fish feed including the protein remains (DDGS — distillers dried grains and solubles) from the alcohol fuel project, vegetable waste and fish meal from the aquaponics project and other experimental feeds such as duckweed.