Fuel Etanol from Cellulosic Biomass

Fuel Etanol from Cellulosic Biomass

(Parte 2 de 3)

Ethanol Production from Celiulosic Materials

Processing options. Several approaches have evolved for the conversion ofcellulosic materials to ethanol; these differ primarily in the method ofhydrolysis and the fermentation system used. Hydrolysis of cellulosic materials can be accomplished with acids or cellulase enzymes. Projected selling prices for ethanol produced from cellulose by acid hydrolysis are currently comparable to those for enzyme-based processes (37). Enzymatic processes are at a much earlier state of technological maturity; however, in the absence of unforeseen breakthroughs for acid-based processes, research is likely to result in enzyme-based processes that are significantly cheaper than acid-based processes. Steps in conversion ofcellulosic biomass into ethanol by enzymatic processes are depicted in Fig. 1.

Energy balance. The ratio of energy output to energy input, R, may be defined for cellulose-based processes with reference to Fig. 1 as

Fig. 1. Production of ethanol from cellulosic materials by means of enzymatic hydrolysis.

where E is exported electricity, A is agricultural inputs, T is raw material transport, C is chemical inputs, D is distribution, P is plant amortization, and all energy flows are expressed as fractions of the lower heating value of ethanol. The 1 in the numerator represents ethanol and the multiplier ofE reflects the displacement of thermal energy for conventional power generation. Estimated parameter values are as follows: E = 0.08 (38),A = 0.15 (Table 2), T = 0.04

(39, 40), C = 0.01 (41, 42), D = 0.01 (43), and P = 0.04 (4, 45). Thus, current understanding ofethanol production from cellulose is consistent with a value of5 for R. In contrast, R is generally less than 1 for corn-based processes without coproduct credits and is approximately 1 if coproducts are considered (45, 46).

A key factor in considering the energetics of ethanol production from cellulose is the energy available from residues remaining after fermentation. It is thought that unfermentable raw material components, in particular lignin, can be mechanically dewatered and burned to provide 30,0 to 40,0 Btu per gallon of ethanol, an amount in excess of processing energy requirements for current designs with a wood feedstock (38). This excess energy can be used

SCIENCE, VOL. 251 on May 13, 2009 w.sciencemag.org Downloaded from

- CarbonConversion --- Energy

Fig. 2. Carbon and energy flows for production and utilization of fuel alcohol from biomass. [Adapted from (53) with permission of Humana Press, copyright 1989] to produce electricity in a cogenerative fashion. The thermal effi- ciency (heat ofcombustion ofethanol plus three times the electricity production relative to the heat ofcombustion ofthe raw material) of ethanol production from cellulosic materials for a process with high yields is in the range of 45 to 70%, depending on the feedstock composition and process configuration.

Global climate change implications. Carbon dioxide production accompanies fermentation of the carbohydrate fraction of biomass to ethanol, combustion of unfermentable biomass fractions to provide process energy, and combustion of fuel ethanol to provide useful work. The quantity ofCO2 released, however, is precisely that which was previously removed from the atmospheric pool by photosynthesis in the course offeedstock production. The cellulose ethanol fuel cycle thus involves cyclic carbon flow (Fig. 2).

Energy inputs are required at several points to drive the cyde depicted in Fig. 2. Agricultural inputs can be satisfied by either fossil fuels or fuels that do not contribute to CO2 accumulation in the atmosphere, such as ethanol in mobile applications and wood or lignin for stationary boilers. The same is true for smaller energy and material inputs associated with equipment depreciation, fertilizer production, and fuel distribution. An indication ofthe contribution offuel options to CO2 accumulation is the net carbon produced per unit energy N. For cellulose ethanol, this parameter may be estimated from wherefis the fraction of energy inputs met by fossil fuels and Cf represents CO2 produced per unit energy for fossil energy inputs. Although Cfwill vary for different scenarios, a reasonable value is 80 mg of CO2 per kilojoule (47), which is representative for gasoline. With R = 5.0 (see above), Eq. 2 indicates that N is 16 mg of CO2 per kilojoule if only fossil fuels are used for energy inputs, corre- sponding tof = 1. N is 0 for the case in which energy inputs are provided by sources that do not contribute to CO2 accumulation, however, corresponding tof = 0. Thus, current understanding of cellulose ethanol technology is consistent with a best case scenario involving no contribution to CO2 accumulation and a worst case scenario resulting in a CO2 contribution about one-fifth that of gasoline.

Environmental impacts. Airborne emissions, liquid effluents, and solid wastes from ethanol production processes appear to pose no problems that cannot be addressed by conventional waste-treatment technology (31, 48). Ethanol is substantially less toxic than methanol and gasoline at the same dosage levels (49). The predominant

Fig. 3. Past and project- = ed costs (1988 basis) for " 6 ethanol and gasoline. * Past gasoline prices are from (2); the range of q,.54 future gasoline prices is *s 'S based on DOE oil gaso- 'ea Ethanol line price projections ; = 2

(4). For ethanol, prices a Gasoline are estimated from pastI research and an aggres- - 0 sive program for future 1980 1990 2000 2010 research. The range X Year shown arises from assumed capital recovery, with the higher values being for a capital recovery factor typical of private financing and the lower values being for a capital recovery factor more likely for municipal or utility-like finance structures.

toxicity issue associated with ethanol use is intentional consumption as an intoxicant. Additives such as 3% gasoline, used in Brazil, probably will be added to discourage such consumption.

Conversion economics. As shown in Fig. 3, progress in cost reductions has been substantial over the last 10 years, resulting in an approximately threefold reduction in the projected selling price (37) to $1.35 per gallon in 1988 for technology proven on a laboratory scale (42). Cost reductions to date stem from minimizing endproduct inhibition of cellulase, improved cellulase enzymes and fermentative microorganisms, and improved systems for xylose fermentation. The current cost ofproducing ethanol from cellulose is the major impediment to utilization of this technology.

Given the cost of representative cellulosic feedstocks (Table 2) and the wholesale selling price required for ethanol to be competitive (see above), operating costs, capital recovery, and secondary raw materials will have to cost in the range of $0.30 to $0.40 per gallon to be competitive with gasoline prices anticipated in the year 2000. A cost ratio of selling price to primary raw material cost of a factor of 2 is unusually large for a commodity chemical (50), which supports our conviction that an economic process is realistic. This conviction is further supported by considerations addressed below.

Ofthe ethanol production steps (Fig. 1), only utilities and residue processing are well developed in the context of cellulose-based processes. Thus, the current technology for conversion ofcellulosic biomass to ethanol has potential for significant improvements in the areas of pretreatment, biologically mediated process steps, and product recovery. Biologic process steps are the most costly by a factor of more than 2 in process designs to date, are the least well developed, and have the greatest potential for improvement (37,

51). Currently, the cost of enzyme constrains the reaction time to values far above the limit imposed by substrate reactivity (Fig. 4).

Various improvements are being investigated that would lower the effective enzyme cost, including increasing the reactivity of the substrate after pretreatment, improving enzyme production systems, improving enzyme activity, and recycling enzyme. Both naturally occurring (52, 53) and genetically engineered (54) systems, wherein the fermenting organism produces its own cellulase, have been described. Such systems have the potential to produce enzyme at little or no incremental cost, in which case the cost of the biologic process steps becomes that ofhydrolysis and fermentation only, and radical cost reductions can be anticipated. Reactor design for high-productivity solids conversion is another area with great potential for lowering the cost of biologic steps.

Reducing enzyme cost from that shown in Fig. 4 changes the minimum on the total cost curve to a reaction time at which the cost of hydrolysis and fermentation is also lower. Such coupled benefits are the rule rather than the exception when improvements in cellulose ethanol processes are considered. In general, improvements on May 13, 2009 w.sciencemag.org Downloaded from

Fig. 4. Trade-off be- - tween the cost for enzyme production and ° 0.8 hydrolysis - fermentation Total (biologic steps) for the SSF process de- * sign being developed by E 0.4 Hydrolysis and

SERI. Costs are estimat- Enzyme fermentation ed from (37). The sub- : Substrate strate reactivity limit reactivity limit (time for essentially o 2 4 6 8 10 complete reaction in a Time for hydrolysis (days) well-mixed system with excess enzyme) is different for different systems; the range indicated is from (53, 65).

in a given process step or parameter have pervasive impacts on several other process steps that may appear to be unrelated (37, 51). For example, an improved fermentation system that achieves a higher product yield will decrease the cost of every process step shown in Fig. 1 with the exception of product recovery. Further, improved ethanol tolerance can lower the costs offermentation and also ofproduct recovery and utilities. Coupled benefits such as these make process economics sensitive to improvements.

The goal ofthe DOE Ethanol from Biomass Program is to reduce the wholesale selling price of ethanol to $0.60 (mid-1980s, basis) per gallon. On the basis ofthe significant opportunities for improvement associated with approaches such as those described above, the

Solar Energy Research Institute (SERI) estimates that technology meeting this goal could be available by the year 2000, given aggressive R&D.

Concluding Remarks

The impacts of alternative fuel use on CO2 accumulation, energy security, balance of payments, sustainable supply, and conversionrelated environmental effects are determined primarily by the energy source rather than the fuel per se. The significant potential benefit of fuel derived from cellulosic biomass with respect to these issues results from the use of renewable nonfood feedstocks. Ethanol appears to score well in terms of fuel-determined impacts as well, which include engine performance, infrastructure compatibility, and utilization-related environmental effects.

Production ofethanol from cellulosic biomass is believed to be an emerging energy technology with particularly great potential for the

U.S. transportation sector. Research to improve conversion processes and to develop cellulosic energy crops is necessary to reduce costs and to increase production potential. Success can reasonably be expected in both these areas in light ofthe immature state ofcurrent technology and the powerful approaches available.

1. "Greater reliance on foreign oil feared as U.S. output tumbles," New York Times, 18 January 1990, p. Al. 2. Mon. Energy Rev., DOE/EIA-0035(89/12) (DOE, Washington, DC, 1990).

3. Econ. Indic. (Wash.) (December 1989). Based on data for the first three quarters of 1989 and conservatively valuing imported petroleum products (about 10% oftotal petroleum consumption) at the cost ofcrude oil. 4. Annual Enery Outlook (Energy Information Administration, DOE, Washington, DC, 1990). 5. "Assessment of costs and benefits of flexible and alternative fuel use in the U.S.

transportation sector, Progress Report One: Context and Analytical Framework," Rep. DOEIPE 0080 DE88 006893 (1989).

6. EPA Lists Places Failing to Meet Ozone or Carbon Monoxide Standards (EPA,

Washington, DC, 1989). 7. A. G. Russell, D. St. Pierre, J. B. Milford, Science 247, 201 (1990). 8. Policy Optionsfor Stabilizing Global Climate Change (EPA, Washington, DC, 1989) {reported by E. Marshall [Science 243, 1544 (1989)] and P. Zurer [Chem. Eng. News 67 (no. 13), 23 (1989)]}; Policymakers Summary ofthe ScientjficAssessment of Global Climate Change (third draft report to the Intergovernmental Panel on

Climate Change (IPPC), IPPC Group 1, Meteorological Office, Bracknell, U.K., May 1990). 9. J. A. Edmonds, W. B. Ashton, H. C. Cheng, M. Steinberg, Rep. DOE/NBB-0085

(1989). 10. Energy Balances of OECD and Other Countries 1971-1987 (International Energy

Agency, Paris, 1989).

1. H. Ahmad, N. Rask, E. D. Baldwin, Biomass 19, 215 (1989). 12. New Fuels Rep. 1 (no. 13), 13 (1990) (available from J. E. Sinor Consultants,

Inc., Niwot, CO); Oxy-Fuel News 1 (no. 50), 9 (1990). 13. 'Fuel ethanol and agriculture: An economic assessment" [Rep. 562, U.S. Department of Agriculture (USDA), Washington, DC, 1986]; J. E. Murtagh, Process Biochem. 21 (no. 2), 61 (1986). 14. Clean Fuels Rep. 1 (no. 2), 82 (1990). 15. E. Anderson, Chem. Eng. News 65 (no. 43), 1 (1988). 16. H. Rothman, R. Greenshields, F. R. Calle, TheAlcohol Economy (Pinter, London, 1983); H. Bernton, W. Kovarik, S. Sklar, The Forbidden Fuel (Boyd Griffin, New York, 1982); D. Houghton-Ailco, Akohol Fuels-Policies, Production and Potential

(Westview, Boulder, CO, 1982). 17. J. D. Ferchak and E. K. Pye, Sol. Energy 26, 17 (1981).

18. Air Quality Benefits ofAlternative Fuels (EPA, Washington, DC, 1987). 19. L. H. Browning, R. K. Pefley, T. P. Suga, Proceedings ofthe 16th Intersociety Energy

Conversion Engineering Conference (American Society of Mechanical Engineers, New York, 1981).

20. C. Risch, paper presented to the California Air Resources Board Public Meeting on

Clean Fuels, Sacramento, CA, 9 September 1988 (available from Ford Motor Company, Dearborn, MI). 21. "Comparative automotive engine operation when fueled with ethanol and methanol" (Rep. HCP/W1737-01 UC-96, DOE, Washington, DC, 1978). 2. R. D. Wilson, Statement before the Consumer Subcommittee on Commerce,

Science, and Transportation, U.S. Senate, November 1987. 23. W. Bernhardt, Proceedings ofthe International Symposium on Akohol Fuel Technology-Methanol and Ethanol (National Technical Information Service, Springfield, VA, 1978); W. H. Kampen, Hydrocarb. Process. 59 (no. 2), 72 (1980).

24. Analysis ofthe Economic and Environmental Effects ofEthanol as an Automotive Fuel (EPA, Washington, DC, 1990).

25. Supported by both historical precedent (in 1983, oil was $28.9 per barrel and gas sold for $0.882 per gallon) and by the approximately $9 per barrel increment reflecting the added value ofgasoline relative to crude oil during the 1980s; see (2).

26. Air Quality Management Plan (South Coast Air Quality Management District, El

(Parte 2 de 3)