3 Ethanol Production
The preparation of ethanol from cellulose-, starch-, and sugar-containing raw materials involves the following general steps:
1. Pretreatment-the physical or chemical conversion of the raw material to a hydrolyzable substrate.
2. Hydrolysis-the enzymatic reaction that converts the starch or cellulose to sugars.
3. Yeast fermentation-the conversion of sugars to ethanol and carbon dioxide.
4. Purification-the separation of ethanol from the by-products and wastes.
Pretreatment for cellulose-containing raw materials will be considered first, followed by the simpler preparation of starch-containing raw materials. For sugar-containing raw materials, only extraction or dilution is required. Fermentation and purification (steps 3 and 4) are identical for all three types of source materials.
PRETREATMENT OF CELLULOSE FOR ENZYMATIC HYDROLYSIS
For acid hydrolysis, wood chips are adequate, but for enzymatic hydrolysis, some chemical or physical pretreatment will usually be required to achieve a reasonable rate and extent of hydrolysis. The objectives of pretreatment are to reduce crystallinity and to increase available surface by maximum destruction of fiber structure and interaction between the cellulose molecules. This can be accomplished by dissolving cellulose in cadoxen, cuprammonium, or strong acid and then reprecipitating and washing. However, because of economic and pollution control constraints, the solvent must be recovered (4-5 grams of solvent are required to dissolve 1 gram of cellulose), and this is a formidable operation. Furthermore, it is much easier to dissolve pure cellulose than to extract cellulose from lignocellulose. Extraction by dilute acid, with some removal of hemicellulose or swelling in dilute alkali' is more economical, but the product then requires washing or pH adjustment before exposure to enzymes. Partial removal of lignin by white rot fungi is a slow process and would entail some loss or degradation of the original carbohydrate. Steam explosion by the Iotech process is effective for hardwoods and agricultural residues, but less so for softwoods and municipal wastes. Ball or attrition milling and two-roll compression milling are effective for many substrates and provide a product of high bulk density, permitting use of 20-30 percent slurries in the saccharification reactor. This is important if concentrated sugar solutions are to be produced.
Considerable attention has been given to agents that will cause swelling of the cellulose and disrupt the crystalline structure. There are two ways in which this occurs:
1. Intercrystalline swelling, because of uptake of water between the crystal units, which causes a reversible volume change of up to about 30 percent.
2. Intracrystalline swelling, which involves penetration of the crystalline structure and can lead to unlimited swelling or complete solution of the cellulose.
Sodium hydroxide, amines, and anhydrous ammonia have been used for intercrystalline swelling; in Europe, during World War II, high concentrations (70-75 percent) of sulfuric acid or fuming hydrochloric acid and metal chelate solvents were used for intracrystalline swelling. Current chemical processes include alkali treatment and treatment with sulfur dioxide.
This has been used for many years as a means of improving the texture of cellulose textiles (mercerization) and to improve the nutritive value of forage and forest residues for feeding ruminants. The treatment of cellulose-containing residues with low concentrations of alkali makes them considerably more susceptible to enzymatic and microbiological conversion and is particularly important to the alcoholic fermentation of these materials.
Steeping various straws in 1.5 percent sodium hydroxide for 24 hours can increase their ruminant digestibility from an initial 30-40 percent to 60-70 percent. This process was patented by Beckmann and used extensively in Europe during World Wars I and II. To avoid loss of soluble sugars during the washing stage of this process, in which the excess alkali is removed, a "dry" process was developed by Wilson and Pigden. The straw is treated with 20 percent alkali and, after standing, is neutralized with acetic acid or simply by mixing with silage (which contains lactic acid). This technique has also been applied to wood residues, resulting in increased rumen digestibility for some woods but not for others. The increase in digestibility is related to the lignin content. Straws and hardwood residues, with lignin contents generally less than 26 percent, respond to the treatment, while softwood residues, with lignin contents higher than 26 percent, do not.
Aqueous and anhydrous ammonia have also been used to increase digestibility, though in general the enhancement is less than that of the sodium hydroxide treatment.
Commercial procedures are based on lignin removal through the selective action of chemical pulping and bleaching agents. However, these processes are not sufficiently selective, and much of the carbohydrate is lost along with the lignin, reducing the yield and decreasing the amount of substrate for enzymatic conversion.
Sulfur Dioxide Treatment
Disruption of lignin-cellulose bonds by treating moist wood with gaseous sulfur dioxide under pressure at 120° C for 2-3 hours appears to offer potential for large-scale processing, though the economics are as yet uncertain. This method increases digestibility to around 60 percent.
Irradiation, milling, and simple heating have also been used to break down lignocellulose. Electron radiation and ball milling are particularly effective in increasing the rate of hydrolysis and yield of sugar under dilute acid saccharification conditions. They have the advantage of yielding a high bulk density of cellulose without the need for washing. However, they are both relatively high technologies that require considerable energy and have not yet been developed commercially.
The best pretreatments currently available may be:
· High pressure and high temperature steam. Stake Technology and Iotech, both in Canada, have pretreatment processes that produce animal feed from straws or hardwoods by steam treatment.
· Dilute alkali treatment (the Beckmann process). The material is treated with 1 percent sodium hydroxide at 45° C for 3 hours - very effective for straw or bagasse.
· Concentrated alkali treatment. Straw is treated with 20 percent sodium hydroxide and the residual caustic neutralized rather than washed out.
Any procedure that increases the digestibility of cellulose for animals is also a good pretreatment for enzymic hydrolysis. This may also be a stepping-stone to the use of treated cellulose for ethanol production - it can be used for animal feeding until such time as conversion of cellulose to ethanol is feasible under local conditions. Then, if large quantities of ethanol are being produced by fermentation processes, the stillage residues can be dewatered and used as animal feed.
SACCHARIFICATION OF CELLULOSE
Saccharification is the process by which the pretreated cellulosic substrate is converted to a sugar solution, which in turn can be used as a substrate for a yeast alcoholic fermentation.
The saccharification process can be carried out chemically (by dilute acid hydrolysis) or by enzymic hydrolysis. As explained earlier, dilute acid hydrolysis is very effective in breaking the glycosidic linkages between component hexoses, but it also breaks down the sugar hexose units. As a result, the product's acidity must be neutralized, and the amount of sugar is less than quantitative because of sugar degradation.
The yield of sugar depends on the relative rates of two reactions that occur when cellulosic materials are treated with dilute acid:
cellulosic material --k1-->sugar
In simple batch processes, the rates of reactions k1 and and k2 are approximately equal, so that maximum yields are limited and the hydrolyzate contains as much breakdown products as contaminants. For starch, which is amorphous, the rate of hydrolysis is much faster than degradation, and sugar yields approach the theoretical level. Lignin has apparently little effect on the rate, as most woods hydrolyze faster than cotton or ramie. Crystallinity of the cellulose is thus the governing factor in dilute acid hydrolysis.
Pretreated substrate, neutralized to approximately pH 4.8, is mixed with enzymes at the required level of activity (Figure 9). In commercial fermentation, the highest substrate concentrations that can be stirred (10 percent or more) are used, and the mixture is incubated at 4550° C and pH 4.5-4.8. The enzymes are destroyed by even brief exposure to high temperatures (60° C or greater), or by a pH below 3.0 or above 8.0.
The costs of both pretreatment and saccharification are functions of scale; for systems in which mechanical or chemical pretreatment is followed by fermentation, large-volume operation would be attractive in industrialized countries as a means of reducing unit costs. It may, however, be possible to have comparatively cheap methods at the other end of the scale. One such simple process has been developed by Toyama et al. (Figure 10).
Mixed chopped, pretreated substrate, such as alkali-treated straw, bagasse, or sawdust (adjusted to pH 4.0) is mixed with either crumbled Trichoderma koji (solid culture), preferably made from the same substrate as the saccharification substrate, or with commercial cellulase in a Shocho jar - a large ceramic jar with a narrow neck to keep out air and prevent contamination. (Shocho is a sweet potato liquor.) The substrate can be in a very thick slurry, since once it is mixed well it will not be stirred. Citric acid (0.5 percent) is added as a preservative and the jar stored at 45° C. In 5-6 days this should yield 15-20 percent sugar syrup.
FUTURE OF CELLULOSE FERMENTATION TECHNOLOGY
In its current state, cellulose fermentation technology - except perhaps for limited, small-scale applications such as the Toyamakoji method of saccharification -has little likelihood of contributing to the production of alcohol fuels in the immediate future. The processes are either too complex or expensive, or require too much acid and alkali or energy for pretreatment, to be able to compete with other potential sources. However, given the increasing need for liquid fuels other than petroleum, the development of cheap and reliable saccharification technologies is necessary and will undoubtedly be achieved. The research and development will be better done in the industrialized countries, and the developing world should monitor progress and take advantage of improvements.
In the meantime, it would be wise for developing country governments and technical assistance agencies to anticipate the development of cheap and practical saccharification, and perhaps direct fermentation, of lignocelluloses to ethanol. The availability of this technology to produce relatively cheap ethanol will have profound economic and social implications for many countries. Operating a pilot plant based on current technology and planting fast-growing species of trees for biomass (such as leucaena or eucalyptus) as potential sources of renewable lignocellulose feedstock will enable organizational and management requirements to be identified in practice. The result will be smoother adoption of the new technology as it becomes available.
SMALL-SCALE ETHANOL PRODUCTION
Despite much interest and discussion, there is very little information on small-scale alcohol production. Although various universities, government and private research institutes, and entrepreneurs have designed, built, and tested (and in some cases sold) small units, unbiased information on technical or economic performance is rare.
For some rural areas, small-scale ethanol production units may be viable if local demands for fuel can be balanced against raw material availability. With smaller units, problems of growing, harvesting, and transporting raw materials and product distribution are minimized. Waste utilization or disposal may also be simplified (Figure 11).
Some small-scale units, discussed below, have been proposed or developed in Brazil, Australia, and the United States.
Brazil: Farm-Size Units
In Brazil, the Instituto de Pesquisas Tecnologicas (Sao Paulo) has developed an alcohol facility for the production of 25,000-50,000 liters per year from sugarcane. With this volume, the fuel needs of a small farm with an electrical generator, a truck, and a tractor could be met. The plant is designed to be operated by farm labor and to produce alcohol at an acceptable cost with a low investment. To meet these requirements, a simple design using construction materials found in rural areas was employed. Wood stave distillation columns packed with pieces of bamboo and ferrocement storage tanks were developed. A prototype plant was built and is shown in Figure 12.
The use of cassava is also being evaluated in an intermediate-sized distillery in northeast Brazil (Figure 13).
Australia: Village-Level Units
In Australia, APACE Research, Ltd., has designed two types of villagelevel alcohol plants-one using conventional distillation and the other solar distillation.
The first design is suggested for use with cassava or with other starchy crops. This wood-fired unit uses a single 5,000-liter stainless steel tank for hydrolysis, fermentation, and distillation. The tank is mounted in firebricks or stones over an enclosed, flued firebox (Figure 14).
During operation, the tank is charged with 2,500 liters of water and brought to a boil. About 2.5 tons of 5-10 mm cassava chips are added, along with 700 cc of bacterial amylase, and the mix simmered for 2 hours. The temperature is then allowed to drop to 70° C, after which 1.6 liters of amyloglucosidase and 6 kg of sulfuric acid are added and the mixture well stirred and allowed to cool to 30° C. At this point, yeast and nutrients are added, and the mash is allowed to ferment for about 2 days. When fermentation is complete, the fire is restarted and the mixture brought to a boil.
The mixture may be distilled to provide solely fuel-grade ethanol (95 percent) for engine use or to provide partly fuel-grade and partly crude ethanol (75 percent) by adjusting the distillation conditions. The crude ethanol would be used for heating and lighting. Either 450 liters of 95 percent alcohol or 225 liters of 95 percent alcohol and 285 liters of 75 percent alcohol may be produced.
The second Australian design incorporates a solar distillation unit and a fiberglass fermentation tank. It is proposed for use with sugar based raw materials, which do not require the heating necessary to saccharify cassava starch. With this unit, sugarcane juice is fermented in a 5,000-liter fiberglass tank and the alcohol recovered in a solar still.
The still resembles a flat plate solar collector and uses a chromium black copper surface to absorb solar energy into the alcohol solution. When the sun heats the unit to a predetermined temperature, a small pump starts and draws a vacuum of 300 mm in the vessel. Feed is then automatically admitted through a solenoid valve, which opens and closes according to the temperature in the still. The vapors are condensed in a modified truck radiator. At night the unit closes down automatically and restarts on the next sunny day.
Prototype units have been built in modules of 3 m2; each module can produce about 10 liters of alcohol per day. All controls are preset and require no intervention.
This solar unit is more complex than the wood-fired unit, but easier to operate. Once fermentation is complete, the filtered product is transferred to a storage vessel for feeding to the still. Using four 3-m2 distillation units, this storage tank would be automatically emptied over about 10 days. The labor required to produce 450 liters of alcohol from sugar fermentation is estimated to be only about 2 manhours.
U.S. Small-Scale Units
The U.S. National Alcohol Fuels Commission has published information on small-scale units using maize or milo as starting materials (Figure 15).
One producer, Apple Agri-Sales (Indiana), uses ground maize. For a typical batch, 360 kg of ground maize are added to 780 liters of 65° C water containing bacterial amylase. The mix is then boiled and stirred for 30-40 minutes to begin starch hydrolysis. The mash is then cooled to 60° C by adding water, and the pH is adjusted to 4.3. The saccharification enzyme is added, and the mash is mixed for 30 minutes. Additional water is added, to bring the total to about 4.5 liters per kilogram of maize and the mix brought to 32° C using internal cooling coils.
About 0.5 kg of yeast is added and mixed in for an hour, after which the mash is left to ferment. Temperature is maintained between 30° C and 36° C using internal cooling coils and intermittent mixing. After about 60 hours, the alcohol concentration reaches 8-10 percent.
When fermentation is complete, the batch is distilled through a 3-m column (14 cm internal diameter) packed with metal turnings. About 135-140 liters of 180, proof alcohol are obtained over a 7-8 hour distillation.
The stillage is filtered with the solids retained for animal feed and the liquid discharged to a drain field. About 28-32 kg of solids at about 28 percent protein is obtained per 100 kg of maize.
Developing technologies that may decrease the cost of ethanol production can be considered in terms of pretreatment, fermentation, alcohol recovery, by-product recovery, and waste treatment.
Much of the research on ethanol processes is aimed at improving pretreatment for lignocellulose feedstocks to enhance the efficiency and reduce the cost of their hydrolysis to sugars.
Some of the processes currently being examined are discussed below.
Tsao has developed a unique process for hydrolyzing crop residues and wood. Hemicellulose is first removed with dilute acid and then the cellulose and lignin are dissolved in concentrated sulfuric acid. The cellulose and lignin are then precipitated from the acid by addition of methanol. Since the precipitated cellulose is in an amorphous form, it is readily hydrolyzed by the appropriate enzymes.
University of Pennsilvania - General Electic Process
In this process, wood chips are first heated in alkaline aqueous butanol to separate the hemicellulose, cellulose, and lignin. The hemicellulose dissolves in the aqueous phase, the lignin dissolves in the butanol, and the cellulose remains undissolved.
The degraded hemicellulose can be fermented to additional butanol or converted to the sweetener xylitol. The lignin-butanol fraction can be cooled to separate the lignin or used as a fuel. The cellulose can be washed and hydrolyzed to glucose for fermentation to ethanol.
The Natick process, outlined in Figure 16, consists of five steps:
1. Selection of an abundant and inexpensive cellulosic substrate such as municipal waste or aspen chips.
2. Pretreatment of this substrate to enhance its enzyme susceptibility, preferably by ball-milling or two-roll compression milling.
3. Production of active cellulase. As a result of screening thousands of organisms over the past 40 years, the Natick group has selected Trichoderma reesei as the best source of active cellulase.
4. Utilization of the cellulase and ß-glucosidase to saccharify cellulose yields 5-15 percent glucose syrups in continuous hydrolysis or 10-30 percent glucose syrups in batch hydrolysis.
5. Fermentation of the resulting glucose syrups to ethanol with Candida or Saccharomyces yeasts.
All steps of the Natick process have been carried out at 200-400 liters, pilot-plant scale (Figure 17), and a complete description and economic analysis is available from the Natick laboratory.
In this process, wood chips are exposed to high-pressure steam for several seconds, followed by explosive decompression.
At the Georgia Institute of Technology, samples of untreated and steam-exploded poplar chips were extracted with water and solvents. Results indicated a fivefold increase in ethanol extractables in the steam-exploded chips over the untreated samples - from about 5 percent to about 25 percent. Since lignin is the major component in the extractables, this steam treatment may facilitate degradation of the remaining cellulose.
Researchers at Battelle-Geneva Laboratories are constructing a pilot plant to test phenols as delignification solvents. The pilot unit will be installed at the San Marco Distilleries in Ferrara, Italy, and will use straw as a raw material. About 1 ton of straw per hour will be separated into cellulose, hemicellulose (recovered as a pentose solution), and lignin. The cellulose will be hydrolyzed into glucose for fermentation to ethanol; the pentose solution will be tested as a source for the production of single-cell protein; and the lignin will be burned for energy. The lignin will also be evaluated as a raw material for adhesives and as a feedstock for phenol production.
The production of more digestible animal feeds from straw through biological conversion has been accomplished by preferential modification of lignin. Biodelignification studies of wheat straw using the white-rot fungus Pleurotus ostreatus have been reported by Detroy et al. As part of this work, the cellulose content of wheat straw and the susceptibility of this cellulose to enzymatic hydrolysis was tracked during a 50-day fermentation with P. ostreatus.
During the first 20 days, the fungus utilized soluble nutrients, lignin, cellulose, and hemicellulose and yielded a residue with only slightly greater susceptibility to cellulose attack than untreated wheat straw. After 30 days, however, the available cellulose doubled, and, after 50 days, the conversion of cellulose to glucose in the residue increased fivefold. Biological modification of wheat straw, coupled with cellulose hydrolysis, resulted in a 72-percent conversion of the cellulose component to glucose without milling or other physical preparation.
Fermentation by Nonconventional Organisms
Various nonconventional organisms have been tested to reduce costs or improve yields in the fermentation step. Processing changes have included solid-phase fermentation, continuous fermentation, vacuum fermentation, and extractive fermentation.
Considerable attention is being given to the possibility of selecting, manipulating, or modifying genes by recombinant DNA techniques to produce organisms or enzymes that will convert cellulose or starch directly to ethanol or other alcohols - for example, Zymomonas strains - so that the process can be carried out in one step. Another example is Clostridium thermocellulum, a thermophilic anaerobe that can utilize cellulose (MIT process). Other clostridia have the capability to ferment sugars to produce ethanol, butanol, isopropanol, acetic acid, acetone, and similar products, and perhaps the genes responsible for these properties can be incorporated into organisms that can produce alcohols directly from cellulose. Mutation to increase yields, alcohol tolerance, and thermotolerance (which would allow the alcohol to be removed as vapor at, say, 70° C) would be an important part of this development.
Economics could also be improved through use of microorganisms with competitive advantages over Saccharornyces spp. Cheaper substrates such as waste cellulose could become major sources of ethanol if satisfactory cellulolytic microorganisms can be developed. As bacteria generally have shorter doubling times and may be easier than yeasts to manipulate genetically, new candidates for use may arise from this class.
Probably the most interesting bacteria for this application are Zymomonas mobilis and Clostridium thermocellum. Rogers et al. at the University of New South Wales have examined the kinetics of alcohol production by Z. mobilis and found that in comparison with Saccaromyces carlsbergensis, Z. mobilis had specific ethanol productivities several times greater than the yeast. Moreover, Zymomonas grows anaerobically and may provide less susceptibility to contamination.
Work with Clostridium thermocellum by Wang et al. at the Massachusetts Institute of Technology indicates that the direct production of ethanol and acetic acid from cellulosic biomass is possible. A strain capable of tolerating 5 percent ethanol has been isolated. Ljungdahl and Wiegel at the University of Georgia have isolated new strains of anaerobic bacteria from hot springs. These organisms have optimum growth at about 70° C with doubling times of 2-3 hours at this temperature. Because they can be used to convert glucose to ethanol at high temperatures, they may allow new methods of continuous fermentation.
At Cornell University, Wilson and Bellamy have genetically modified Escherichia cold by implanting a gene - taken from a thermophilic bacterium - that codes for cellulase production. Although other researchers have implanted a cellulase-producing gene in E.coli the gene used at Cornell operates at 65° C, almost twice the temperature of previous work.
Fermentation Processing Changes
In this type of fermentation the starch or sugar crop is pulped and fermented without prior extraction. This process is exemplified by sake production from rice, kao-liang from sorghum, and, more recently, ethanol from pulped sugar beet.
For sake production, steamed polished rice is successively saccharified and fermented in the mash state. The 20-25 day process results in a dense mass with up to 20 percent ethanol (see Rose, p. 424-425).
In kao-liang production, sorghum grains are steamed, inoculated, and fermented at about a 50 percent moisture content. When conversion is complete, the mass is steam distilled to produce the liquor.
From work on sugar beet, Kirby and Mardon at CSIRO (Australia) suggest that solid-phase fermentation may be the most economical approach for 1-5, million-liter-per-year ethanol plants. In the CSIRO process, sugar beets are chopped into 3-mm pieces, adjusted to pH 4.5 with sulfuric acid, and inoculated with Saccharomyces cerevisiae. The pulp is allowed to ferment anaerobically at 25°-30° C for 10-16 hours and then pressed to remove the fiber. The press juice is centrifuged to separate and recycle the yeast. The juice, which contains about 9.5 percent ethanol, is distilled. Ethanol yields of up to 92 percent of theoretical yields have been obtained.
The Australian wine industry has also used solid-phase fermentation for ethanol production. Tarac Industries ferments grape mare (the residue of skins, seeds, and stems after pressing, containing about 5 percent residual sugar) in open heaps for 3-4 days. The fermented mare is then distilled to recover ethanol.
Barford et al. at the University of Sydney (Australia) have examined tower fermentation for converting sugars to ethanol. In this process, the fermentable substrate is pumped upward through a column containing high concentrations (70-90 g per liter) of flocculated yeast (Figure 18). Residence times of 3-6 hours are required for 15 percent sugar solutions; for 20 percent sugar solutions, residence times of 812 hours are needed for complete conversion to alcohol. Continuous runs of up to 14 weeks have been made without deterioration in performance.
The advantages of tower fermenters lie in their simplicity of construction and operation. No agitation is needed to keep yeast cells in suspension, nor are external yeast separation and recycling facilities needed to achieve the high cell populations necessary for high fermentation rates.
Wick has tested a new vessel design for continuous fermentation of grape juice and glucose solutions. With this vessel, shaped like an inverted right triangle, the substrate enters at the bottom and, through force of flow and gas evolution, creates a rolling turbulence that disperses and suspends the yeast (Figure 19). With a retention time of 4 hours, no loss of yeast activity or contamination problems were encountered during a continuous 52-day run.
Rosen has described a commercial continuous fermentation unit (Figure 20) used in Denmark for the conversion of molasses to alcohol. Two stirred cylindrical fermenters are used in series, each with a capacity of 170 m³. Once started, 6,000 kg per hour of molasses (diluted to 22,O00 liters) is charged to the system. Residence time in each fermenter is 10,5 hours.
Chibata, at the Tanabe Seiyaku Co. in Osaka, Japan, has described a system in which yeast cells are immobilized in a carrageenan gel in a packed column. A 20-percent glucose solution is fed continuously to the base of the column, and as the solution moves upward, the glucose is converted to ethanol and carbon dioxide. With a 2.5-hour residence time, the effluent from the top of the column contains about 10 percent ethanol. Continuous runs have been made for 3 months without loss of the yeast's activity.
Studies on vacuum fermentation have been conducted in Great Britain, the United States, and elsewhere. Work by the W. S. Atkins Group at Manchester University has centered on reducing energy use and fermentation time through vacuum fermentation of concentrated sub strafes with high yeast levels. Substrates such as molasses and sac charified cassava are fed directly to the fermenter, and a portion of the fermenting mass is recirculated through a flash separator to remove alcohol as it is formed.
Work on vacuum fermentation by Ramalingham and Finn at Cornell University indicates that a threefold higher sugar concentration can be fermented in one-third of the time needed in a conventional process.
Preliminary process studies by Cysewski and Wilke at the University of California indicate that ethanol plant capital costs for vacuum operation may be reduced up to 71 percent over batch processing.
Rolz at ICAITI (Guatemala) has developed an extractive fermentation system based on sugarcane. In this process, whole cane from the fields is cleaned and cut, with the foliage retained for fuel use. The cane is then chipped and placed in a fermenter, where water and yeast are added. The sucrose from the cane is extracted and fermented simultaneously in the same equipment. When the fermentation is finished, the extracted chips are separated and the solution is charged to a batch of fresh chips. After this second fermentation, the solution (containing about 5 percent ethanol) is distilled to recover the alcohol.
In examining the energy consumption of alternative recovery systems, Vergara has summarized the status and efficiency of several approaches as shown in Table 8.
Other possibilities for low-energy alcohol recovery include the use of reverse osmosis, molecular sieves, solvent extraction, and supercritical carbon dioxide.
In studies comparing these techniques at Battelle Pacific Northwest Laboratories (USA), a solvent extraction method looks most promising. In this process, a proprietary solvent mixture is contacted with the ethanol-water mix derived from fermentation and extracts the ethanol. This is done under slight pressure, so that when the pressure is subsequently reduced, the bulk of the solvent flashes off and is recovered for reuse.
Battelle also examined a carbon dioxide extraction process developed by Arthur D. Little (USA) in which liquid CO2 is used to extract ethanol and then depressurized to flash off the CO2. Results of these tests are compared with other conventional and nonconventional techniques in Table 9.
TABLE 8 Energy Efficiency of Alternative Distillation Systems
An alternative to the final distillation step used to produce anhydrous alcohol has been examined by Maiorella at the University of California. Using the process developed by the Mobil Oil Corporation for the conversion of methanol to gasoline, 85 percent ethanol is first partially converted to ethylene and ethyl ether and then catatyfically converted to a synthetic gasoline. The reactions and conceptual design are shown in Figures 21 and 22.
By-Product Recovery and Waste Treatment
The principal by-products in ethanol production are carbon dioxide, fusel oils, and stillage.
This gas is produced during the fermentation step and is usually vented to the atmosphere. In most cases, recovery of carbon dioxide and its compression for use as a solid refrigerant ("dry ice") or use in promoting growth in controlled environment agriculture (greenhouses) is uneconomic: about 575 kg of carbon dioxide are produced for each 1,000 liters of ethanol.
TABLE 9 Energy Requirements of Ethanol Separation Processes
a Sum of distillation to azeotrope and azeotropic distillation
These are higher boiling coproducts of ethanol produced during fermentation. They consist mainly of 5-carbon alcohols and glycerin and may be recovered during distillation. If the ethanol is destined for fuel use, however, fusel oils may be left in the mixture to act as a denaturant. About 4 liters of fusel oil are produced for each 1,000 liters of ethanol.
This is the liquid residue remaining after the ethanol is distilled from the fermentation mix, which can represent a major disposal problem. Stillage is produced at about 10-15 times the volume of alcohol produced and contains about 10 percent solids. Potential uses for stillage include crop application, animal or fish feed, and biogas generation.
Since the nutrients in stillage are at a fairly low level (1 percent N, 0.2 percent P, 1.5 percent K), land application is justifiable only in areas within a few kilometers of the distillery. In Brazil, for example, there are many distilleries close to plantations, and land use of stillage in these situations seems effective. The stillage is applied from fiberglass tank trucks to fields that have been recently harvested; it is used in gravity irrigation systems and applied undiluted to sugarcane by means of giant sprinklers (Figure 23). There are no long-term studies on possible adverse effects of salt accumulation from continued use of stillage.
For use in animal feeding, stillage can be evaporated to about 50 percent solids and mixed with feed concentrates; this partially dried residue is currently selling for $50 a ton in the United States and Europe. Unless solar drying is used, however, evaporation can be costly, and the value of stillage as feed depends on the availability of alternative feeds. In many developing countries there is little demand for compounded feeds.
The Bacardi Corporation has developed an anaerobic treatment system for the stillage from their 389,000-liter-per-day rum distillery in Catano, Puerto Rico. This facility includes a 2.7-million-liter holding tank for the stillage and a 12.6-million-liter anaerobic reactor. This anaerobic tank is packed with plastic sheeting, which provides a surface area of hundreds of thousands of square meters for the active microorganisms to attach themselves to and grow on. The unit treats about 1.2 million liters per day of stillage and yields about 20,600 cubic meters per day of biogas. The treated stillage is released to the ocean, and the biogas is burned to heat the boilers in the distillery.
The Thailand Institute of Scientific and Technological Research has also developed an anaerobic stillage treatment plant. Scaled to process 100 m³ per day of stillage, the design was based on pollution problems experienced by small distilleries in Thailand (Figure 24).
REFERENCES AND SELECTED READINGS
Anonymous. 1982. Catalyst converts methanol to dry ethanol. Chemical and Engineering News September 20:41-42.
Anonymous. 1982. Simple system converts ethanol to gasoline. Chemical and Engineering News August 9:6.
Anonymous. 1983. E. cold engineered to digest cellulose. Chemical and Engineering News March 14:25.
Anonymous. 1983. Pilot plant to upgrade biomass. Chemical and Engineering News January 17:30.
Arcuri, EJ. 1982. Continuous ethanol production and cell growth in an immobilized-cell bioreactor employing Zymomonas mobilis. Biotechnology and Bioengineering 24:595:604.
Atkins Planning. 1980. Market Opportunities in the Exploitation of Fermentation Alcohol as a Liquid Fuel and Chemical Feedstock. A joint study for Rolls-Royce, Ltd., and W. S. Atkins Group, Ltd. Epsom, Surrey, England.
Black, C. 1980. Distillation modeling of ethanol recovery and dehydration processes for ethanol and gasohol. Chemical Engineering Progress September:78-85.
Card, J.C., and Farrell, L.M. 1982. Separation of Alcohol-Water Mixures Using Salts.
Order Number DE 82013346. NTIS, Springfield, Virginia, USA.
Chibata, I., and Tosa, T. 1981. Use of immobilized cells. Annual Review of Biophysics and Bioengineering 10:197-216.
Cysewski, G.R., and Wilke, C.R. 1978. Process design and economic studies of alternative fermentation methods for the production of ethanol. Biotechnology and Bioengineering 20:1421-1444.
DeYoung, H.G. 1982. Ethanol from biomass: the quest for efficiency. High Technology 2(1):61-67.
Durand, G., and Navarro, J.M. 1978. Immobilized microbial cells. Process Biochemistry 13(9): 14-23.
Eakin, D.E., Donovan, J.M., Cysewsky, G.R., Petty, S.E., and Maxham, J.V. 1981. Preliminary Evaluation of Alternative Ethanol/Water Separation Processes. Prepared for the U.S. Department of Energy under Contract DE-AC06-76RLO 1830. Pacific Northwest Laboratory, Richland, Washington, USA.
Emert, G. H , Katzen, R., Fredrickson, R. E., Kaupisch, K.F. 1 980 . Gasohol/biomass developments: economic update of the Gulf cellulose alcohol process. Chemical Engineering Progress September:47-52.
Glassett, J., and Rankin, J. 1981. Ethanol dehydration. Gasohol USA 3(4): 13-14.
Hsie, M.C., and Su, Y.C. 1980. Improvement of alcohol fermentation. In: lV International Symposium on Alcohol Fuels Technology. Guaruja, Sao Paulo, Brazil.
Kirby, K.D., and Mardon, CJ. 1980. Solid phase fermentation for intermediate scale ethanol production. In: IV International Symposium on Alcohol Fuels Technology. Guaruja, Sao Paulo, Brazil.
Ladisch, M.R., and Dyck, K. 1979. Dehydration of ethanol: new approach gives positive energy balance. Science 205:898-900.
Lee, K.J., Skotnicki, M.L., Tribe, D.E.,and Rogers, P.L.1981. The effect of temperature on the kinetics of ethanol production by strains of Zymomonas mobilis. Biotechnology Letters 3(6):291-296.
Ljungdahl, L.G., and Wiegel, J. 1979. Ethanol Fermentation Using Anaerobic Thermophilic Bacteria. Department of Biochemistry, University of Georgia, Athens, Georgia, USA.
Maiorella, B.L. 1982. Fermentation alcohol: better to convert to fuel. Hydrocarbon Processing August 1982:95-97.
Minier, M., and Goma, G. 1980. Ethanol production by extractive fermentation. In: International Conference on Energy from Biomass, 4-7 November 1980. Brighton, England.
Priegnitz, J.W. 1982. Process for the separation of water from ethanol (absorptive separation on corn meal). June 8, 1982. U.S. Patent 4,333,740.
Prince, I.G., and McCann, DJ. 1978. The continuous fermentation of starches and sugars to ethyl alcohol. In: Alcohol Fuels. A conference held at the Sebel Town House, Sydney, Australia. Institution of Chemical Engineers, N.S.W. Group, University of Sydney, Australia.
Prouty, J.L. 1980. The impact of differing raw materials and alternate fermentation methods on process designs for the production of fuel grade ethanol. In: IV International Symposium on Alcohol Fuels Technology. Guaruja, Sao Paulo, Brazil.
Ramalingham, A., and Finn, R.K. 1977. The Vacuferm process: a new approach to fermentation alcohol. Biotechnology and Bioengineering 19:583-589.
Rogers, P.L., Lee, KJ., and Tribe, D.E. 1980. The Zymomonas process for fermentation alcohol. Fuel Ethanol Research and Development Workshop, February 4-5, 1980, Canberra, Australia.
Rogers, P.L., Lee, KJ., Skotnick, M.L., and Tribe, D.E. 1981. Ethanol fermentation by highly productive strains of Zymomonas mobilis Pp. 189-194 in: Advances in Biotechnology, Vol. II, M. Moo-Young and C.W. Robinson. Pergamon Press, New York, New York, USA.
Rolz, C. 1980. New technology to ferment sugar cane directly: the ex-ferm process. Process Biochemistry 15(6):2-6.
Rose, A.H. 1977. Alcoholic Beverages. Academic Press, New York, New York, USA.
Rosen, K. 1978. Continuous production of alcohol. Process Biochemistry May:25-26.
Schwartz, D.M. 1980. Direct application of solar energy to alcohol fuel production. In: IV International Symposium on Alcohol Fuels Technology. Guaruja, Sao Brazil.
Thomas, D.M.S., Barford,J.P., Prince, R.G.H. 1980.Pilot Plant Design for the Assessment of Alcohol Production by Continuous Tower Fermentation. Department of Chemical Engineering, University of Sydney, Sydney, Australia.
Tsao, G.T. 1979. Selective solvent extraction in utilization of stored solar energy in cellulosic biomass. Third Annual Biomass Energy Systems Conference, June 5-7, 1979, Golden, Colorado, USA.
Ueda, S., and Koba, Y. 1980. Alcohol fermentation of raw starch without cooking by using black-koji amylase. Journal of Fermentation Technoloy 58(3):237-242.
Vergara, W. 1979. Continuous Vaccuum Thermophilic Fermentation for the production of Alcohol Using Clostridium Thermocellum. Department of Agricultural Engineering, Cornell University, Ithaca, New York, USA.
Vergara, W. 1980. Improving the scenario for ethanol production: the new ethanol producers. In: IV International Symposium on Alcohol Fuels Technology. Guaruja, Sao Paulo, Brazil.
Wada, M., Kato, J., and Chibata, I. 1980. Continuous production of ethanol using immobilized growing yeast cells. European Jurnal of Applied.Microbiology and Biotechnology 10(4):275-287.
Wang, D.I.C., Biocic, I., Fang, H.Y., and Want, S.D. 1979. Direact microbial conversion of cellulosic biomass to ethanol. Third Annual Biomass Energy Systems Conference, June 5-7, 1979, Golden, Colorado, USA.
Wick, E. 1980. New Vessel Design for Rapid, Continuous Fermentation. Science and Education Administration, U.S. Department of Agricubure, Washington, D.C., USA.
Anonymous. 1980. Small-scale ethanol production. Energy Fact Sheet 8. Volunteers in Technical Assistance, Mt. Rainier, Maryland, USA.
Anonymous. 1981. Brazil plans to include micro distillers in national alcohol program. Synfuels December 4:5.
Badger, P.C., Pile, R.S., and Waddell, E.L.1981. Farmstead production of fuel alcohol. Proceedings of the Annual Meeting of the American Section of the International Solar Energy Society 4(1):40-44.
Berglund, G.R., and Richardson, J.G. 1982. Design for a small-scale fuel alcohol plant. Chemical Engineering Progress August 1982:60-67.
Bruschke, H. 1979. Plant for ethanol production. In: Chemical Technology for Developing Countries, edited by A. A. C. M. Beenackers. Nigerian Society of Chemical Engineers, Zaria, Nigeria.
Buchele, W.F. 1977. Design of a Farm Distillery. Iowa State University, Ames, Iowa, USA.
Chambers, R.S. 1979. The Small Fuel-Alcohol Distillery: General Description and Economic Feasibility Workbook. ACR Process Corporation, Westfield, New Jersey, USA.
Chen, T.H., and Day, D.L. 1980. Feedlot Biogas as Process Fuel for Farm Alcohol Plants. Paper prepared for the International Symposium on Biogas, Microalgae and Livestock Wastes, Taipei, Taiwan.
Hollingan, PJ. 1980. On Farm Fuel Production. Department of Agriculture, New South Wales, Australia.
Lightsey, G.R., and Kadir, O. n.d. Use of Plastic Construction Materials in Small-Scale Fuel Alcohol Production. Department of Chemical Engineering, Mississippi State University, Mississippi State, Mississippi, USA.
Lockeretz, W. 1982. On-farm alcohol production: economic considerations and implications for farm management. Energy in Agriculture 1:171-184.
McCann, D.J. 1980. Village-Based Fuel Ethanol Plants. APACE Research Centre, Rotherwood, Australia.
Nellis, M. 1979. Making It on the Farm: Alcohol Fuel Is the Road to Independence. American Agriculture News. Iredell, Texas, USA.
Ozkan, H.E., Chaplin,J., and Marley, SJ. 1981. Farm scale alcohol production. The Iowa State University ethanol distillery. Order number DE 82903916. NTIS, Springfield, Virginia, USA.
Sklar, S. 1980. Small-Scale Ethanol Production and Economics. Paper prepared for the Social Economic Impact Analysis Seminar, BioEnergy '80, April 24, 1980, Atlanta, Georgia. National Center for Appropriate Technology, Washington, D.C., USA.
Solar Energy Information Data Bank. 1980. Fuel from Farms. A Guide to Small-Scale Ethanol Production. Solar Energy Research Institute, Golden, Colorado, USA.
Tentscher, W. 1982. Ethanol Production by Direct Fermentation of Sugar Cane in Small-Scale Plants with Respect to Appropriate Technology. Paper prepared for the V International Alcohol Fuel Symposium, May 13-18,1982, Auckland, New Zealand. Zealand.
Thimsen, D. 1982. Building a farm-scale fuel ethanol plant. Minnesota Agricultural Experiment Station Misc. Report 181: 1 -4.
U.S. National Alcohol Fuels Commission 1980. Fuel Alcohol on the Farm: A Primer on Production and Use. U.S. National Alcohol Fuels Commission, Washington, D.C., USA.
Villen, R.A. 1980. Microusinas pare producao de alcool etilico. In: TV Internaiional Symposium on Alcohol Fuels Technology. Guaruja, Sao Paulo, Brazil.
Anonymous. 1979. Continuous cellulose-to-glucose process. Chemical and Engineering News 57(41):19-24.
Anonymous. 1981. New Zealand government approves syn-gasoline plant using Mobil process. Synfuels October 2:8.
Anonymous. 1982. HF key to improved cellulose-cracking process. Chemical and Engineering News February 1.22-23.
Anonymous.1982. Wood-to-alcohol costs pegged. Chemical and Engineering News March 1:19.
Beckmann, E. 1922. British Patent 151,229.
Buchholz, K., Puls, J., Godelmann, B., and Dietrichs, H.H. 1980/81. Hydrolysis of cellulosic wastes. Process Biochemistry 16(1):37-43.
Detroy, R.W., Cunningham, R.L., Bothast, R.J., Bagby, M.O., and Herman, A. 1982. Bioconversion of wheat-straw cellulose/hemicellulose to ethanol by Saccharomyces uva rum and Pachysolen tannophilus. Biotechnology and Bioengineering 24:1105-1113.
Gaden, E.L., Mandels, M.H., Reese, E.T., and Spano, L.A. 1976. Enzymatic Conuersion of Cellulosic Materials: Technology and Applications. Symposium No. 6 of Biotechnology and Bioengineering. John Wiley & Sons, New York, New York, USA.
Ghose, T.K. 1981. Cellulosic conversion. In: Advances in Food Producing Systems for Arid and Semiarid Lands, edited by J.T. Manassah and E.J. Briskey. Academic Press, Inc., New York, New York, USA.
Ghose, T.K., and Ghose, P. 1978. Bioconversion of cellulosic substances. Journal of Applied Chemical Biotechnology 28.309-320.
Guha, B.K., and Titchener, A.L. 1981. Acid Hydrolysis of Wood. New Zealand Energy Research and Development Committee, University of Auckland, Private Bag, Auckland, New Zealand.
Higgins, FJ., and Ho, G.E. 1982. Hydrolysis of cellulose using HCI: a comparison between liquid phase and gaseous phase processes. Agricultural Wastes 4:97-116.
Higginson, B., and Thornton, R.H. 1980. Automotive Fuels from Cellulose Materials. New Zealand Energy Research and Development Committee, University of Auckland, Private Bag, Auckland, New Zealand.
Jenkins, D.M., and Reddy, T.S. 1979. Report to U.S. Department of Energy on Economic Evaluation of the MIT Process for Manviacture of Ethanol. Battelle, Columbus Laboratories, Columbus, Ohio, USA.
Klimoua, Z.K. et al. 1977. Hydrolysis of the hemicelluloses of plant raw material for the production of yeast. Gidroliznaya: Lesolhimichcskaya Promyshlennost 6:26-28.
Krieger, J. 1982. Process yields marketable biomass fractions. Chemical and Engineering News News 31-28.
Ladisch, M.R. 1979. Fermentable sugars from cellulosic residues. Process Biochemisty 14(1)21-25.
Lamport, D.T.A., Smith, G., Mohrick, S., Hawley, M.C., Chapman, R., and Selke, S. 1980. HF Saccharification: The Key to Ethanol from Wood. Paper prepared for the International Conference on Energy from Biomass, Brighton, England.
Lyons, T.P. 1981. Cellulose as a raw material. Gasohol USA 3(2):14-28.
Mandels, M. 1981. Cellulases. In: Annual Reports on Fermentation Processes, Vol. 5. Academic Press, New York, New York, USA.
Mandels, M., and Andreotti, R.E. 1978. Problems and challenges in the cellulose to cellulase fermentation. Process Biochemistry 13(5):6-14.
Mednick, R.L., Weiss, L.H., and Xippolitas, E.G. 1982. Ethanol from Cellulosics. Chemical Engineering Progress August 1982:68-73.
Mendelsohn, H..R., and Wettstein, P. 1981. Ethanol from wood. Chemical Engineering June 15:62-65.
Mort, AJ., and Parker, S.F. 1980. Use of Ankydrous HF Solvolysis in Conversion of Biomass to Glucose. Paper prepared for the First SERI International Workshop on Biotechnology for the Production of Chemicals/Fuels from Biomass, Vail, Colorado. Solar Energy Research Institute, Golden, Colorado, USA.
National Research Council. 1976. The Potential of Lignocellulosic Materials for the Production of Chemicals, Fuels, and Energy. A Panel Report for the Committee on Renewable Resources for Industrial Materials, Board on Agriculture and Renewable Resources, Commission on Natural Resources, National Research Council. National Academy of Sciences, Washington, D.C., USA.
Nuclear Assurance Corporation. 1981. Feasibility Study of the Commercial Production of Ethanol from Wood. U.S. Department of Energy, Office of Alcohol Fuels, Washington, D.C., USA.
O'Neil, DJ., Colcord, A.R., Bery, M.K., Roberts, R.S., and Sondhi, D.K. 1979. Alcohol Fermentation Process Development Unit. Paper prepared for the 72nd Annual Meeting, American Institute of Chemical Engineers, 25-29 November 1979, San Francisco, California, USA.
Roberts, R.S., Sondhi, D.K., Bery, M.K., Colcord, A.R., and O'Neil, DJ. 1980. Process optimization for saccharification of cellulose by acid hydrolysis. In: Second Symposium on Biotechnology in Energy Production and Conservation, edited by C.D. Scott. John Wiley & Sons, New York, New York, USA.
Saeman, J.F. 1981. Key factors in the hydrolysis of cellulose. In: Biomass as a Nonfossil Fuel Source, edited by D.L. Klass. ACS Symposium Series 144. American Chemical Society, Washington, D.C., USA.
Spano, L., Tassinari, T., Ryu, D.D.Y., Allen, A., and Mandels, M. 1979. Enzymatic Hydrolysis of Cellulose to Fermentable Sugar for Production of Ethanol. Presented at the 29th Canadian Chemical Engineering Conference, Sarnia, Ontario, Canada.
Taylor, John D. 1980. Continuous Autohydrolysis, a Key Step in the Economic Conversion of Forest and Crop Residues into Ethanol. Paper prepared for the Energy from Biomass Conference, Brighton, England.
Toyama, N., Ogawa, K., and Toyama, H. 1979. Production of Sugar from Cellulose with Trichoderma Cellulase. Applied Microbiology Laboratory, Department of Agricultural Chemistry, Faculty of Agriculture, Miyazaki University, Miyasaki 880, Japan.
U.S. Army.1981. Enzymatic Hydrolysis of Cellulose to Glucose: A Report on the Natick Program. United States Army, Natick Research and Development Command, Natick, Massachusetts, USA.
Wayman, M. 1980. Alcohol from cellulosics: the autohydrolysis-extraction process. In: IV International Symposium on Alcohol Fuels Technology. Guaruja, Sao Paulo, Brazil.
Wilke, C.R., Yang, R.D., Sciamanna, A.S., Freitas, R.P. 1978. Raw Materials Evaluation and Process Development Studies for Conversion of Biomass to Sugars and Ethanol. Paper prepared for the Second Annual Symposium on Fuels from Biomass, June 20-22, 1978. Rensselaer Polytechnic Institute, Troy, New York, USA.
Wilson, R.K., and Pigden, WJ. 1964. Effect of a sodium hydroxide treatment on the utilization of wheat straw and poplar wood by rumen microorganisms. Canadian Journal of Animal Science 44:122.
Anonymous. 1976. Stillage treatment with by-products recovery. Centro de Tecnologia Promon Newsletter 1(2):3.
Bruce,J., and Hawrylyshyn, G. 1981. A bonanza for Brazil's fish-alcohol swill is fine food. Work Environment Report 7(14): 1-2.
Fridman, H.L. 1982. Bacardi makes the most of 'molto,' turns waste to fuel. San Juan Star Thursday, April 22, 1982.
Kohler, A.A., and Pagan-Font, F.A. 1978. Evaluation of rum distillation wastes, pharmaceutical wastes and chicken feed for rearing Tilapia aurea in Puerto Rico. Aquaculture 14:339-347.
National Research Council. 1981. Feeding Value of Ethanol Production By-Products. National Academy Press, Washington, D.C., USA.
Sang, S.L., et al. 1980. Some notes on studies of alcohol slops treatment in Taiwan. In: IV International Symposium on Alcohol Fuels Technology. Guaruja, Sao Paulo, Brazil.
Yoshizawa, K. 1978. Treatment of waste-water discharges from a Sake brewery using yeast. Journal of Fermentation Technology 56(4):389-395.
[Ukrainian] [English] [Russian]