Strategic Issues Facing Transportation, Volume 5: Preparing State Transportation Agencies for an Uncertain Energy Future (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

136 The displacement of petroleum with liquid fuels produced from biomass, commonly referred to as “biofuels,” promises several potential advantages. Estimates suggest that domesti- cally produced biofuels could further the objective of energy security by meeting roughly a third of U.S. transportation fuel needs (Parker et al. 2011). Also, depending on their feed- stocks and production methods, biofuels could help reduce greenhouse gas emissions from the transportation sector. Biofuels can be classified by their technological and economic maturity. First-generation biofuels include ethanol from the fermentation of corn, sugar cane, and other sugary or starchy food crops along with biodiesel derived from oil seeds such as soy and canola through a chemical process known as trans- esterification. (Waste fats and cooking oil can also be used to produce biodiesel, but these feedstocks are more limited in quantity.) These first-generation fuels are widely in use today, and most people use the term “biofuels” to mean ethanol and biodiesel. There are a wide variety of potential feedstocks and produc- tion pathways for more advanced, second-generation biofuels. Most prominent among these is the production of cellulosic ethanol or butanol through biochemical processes using wood, grasses, or crop wastes (e.g., corn stover) as feedstocks. Thermo- chemical processes can also be used with biomass feedstocks to create drop-in gasoline and diesel replacements (often described as green gasoline and green diesel or renewable diesel) that work well with existing vehicles and distribution infrastruc- ture without the need for blending. These thermochemical processes include biomass gasification followed by Fischer- Tropsch catalytic processing, gasification to produce methanol followed by the conversion of methanol to gasoline, pyrolysis followed by hydroprocessing, and other hybrid approaches (NRC 2009, 2013). Algae, which offers the advantage of requir- ing less land and water to produce, is also being explored as a potential feedstock for biofuels (NPC 2012, NRC 2013). There are a number of challenges and uncertainties related to the broader use of biofuels. These include issues associated with distribution logistics, biomass feedstock availability, and the environmental effects of large-scale domestic produc- tion and consumption of biofuels. It remains unclear which approaches for producing biofuels will be most effective in the long run at scales and costs needed to offset a significant portion of U.S. petroleum consumption. This appendix dis- cusses recent experience with biofuels in the United States along with issues and factors that will influence the future penetration of biofuels in the surface transportation sector. Unlike other alternative fuels such as natural gas or hydro- gen, first-generation biofuels are generally not used alone in vehicles but rather are blended with gasoline or diesel fuel. Ethanol is routinely blended with gasoline in mixtures of 10% or less (E10) but can be blended all the way up to 85% (E85) in the United States. One advantage of the E10 blend is that conventional engines can burn this mix of ethanol efficiently without any modifications. Biodiesel can likewise be blended with conventional diesel fuel in a mix known as BD20. As with E10 for gasoline engines, conventional diesel engines can use BD20 efficiently without any modifications. Ethanol is the dominant biofuel for light-duty vehicles in the United States today (the majority of which use gasoline engines), with biodiesel used in some light-, medium-, and heavy-duty vehicles with diesel engines. Much of the material in this appendix focuses on ethanol, but the discussion encom- passes biodiesel and other biofuels as well. C.1 Production, Distribution, and Refueling Biofuel production can be broken up into several steps. First, biomass must be grown and harvested. After being har- vested, the biomass must be stored and transported to a bio- refinery, where it is converted to biofuel. After the biofuel is produced, it must be transported to blending stations before it is distributed to retail stations for sale and consumption. The following discusses these steps in more detail. A p p e n d i x C Liquid Biofuels

137 C.1.1 Production For ethanol currently refined in the United States, the dom- inant production supply chain begins with corn production, mostly in the Midwest. The corn is typically delivered by truck or rail to an ethanol plant or biorefinery, where it is then processed into cornstarch and fermented. In 2011, etha- nol supplied about 10% by volume of total U.S. demand for gasoline (NRC 2013). An alternative to corn-based ethanol is cellulosic ethanol. Cellulosic ethanol can be made from breaking down the woody fibers in trees, grasses, and crop wastes. The process used to make cellulosic ethanol requires less energy per unit of fuel and generally produces lower emissions than the corn ethanol pro- cess. During 2011, three cellulosic biorefineries were in opera- tion worldwide, collectively producing an estimated 3 million gallons of ethanol (Parker et al. 2011). However, the commer- cial viability of cellulosic ethanol conversion technologies at much larger scales has yet to be demonstrated. Biodiesel is produced by processing vegetable oils or ani- mal fats and can be made from a wide variety of feedstocks. The biodiesel fuels produced by different feedstocks vary in certain characteristics, which can limit their suitability for commercial use. One of the most important properties is known as the cloud point, the temperature at which the fuel begins to gel and solidify. Higher cloud points are problem- atic for operations in cold-weather climates during winter months. The domestic production of biodiesel is consider- ably less than ethanol. Total U.S. biodiesel production was 343 million gallons in 2010, 967 million gallons in 2011, and 969 million gallons in 2012 (EIA 2013c). In 2011, bio- diesel supplied less than 1% of total U.S. transportation fuel demand (NRC 2013). Renewable diesel, a second-generation biofuel, is the product of fats or vegetable oils that are refined in a process that involves hydrogenating triglycerides to remove metals and compounds with oxygen and nitrogen. A key benefit of renewable biodiesel in comparison to biodiesel is that it meets quality standards that allow it to be blended in any combination with petroleum- based diesel fuel without modifying vehicle engines or fueling infrastructure (EERE 2012b). Another pathway is algae-based biofuels. Algae cultivation suitable for biofuel production requires sunlight and a source of carbon dioxide. To enhance productivity, most algae culti- vation schemes for biofuels involve using carbon dioxide con- centrations well above the atmospheric level—for example, a stream of captured CO2 emissions from a fossil-fueled power production plant. Due to the early stage of development of algal oil processes, the net greenhouse gas emissions perfor- mance of algal biofuels remains uncertain, as does the ability of algae to provide sufficient feedstock for large-scale produc- tion at reasonable cost (Bartis and Van Bibber 2011, NPC 2012, NRC 2013). Algae does, however, have the advantage of requir- ing much less land and water for production than herbaceous energy crops, nearly eliminating concerns about emissions from indirect land-use change (Williams et al. 2009). C.1.2 Distribution The most economical method of transporting liquid fuels is via pipelines. Gasoline, for example, is transported in large quantities from refineries to distribution centers via pipeline. In the United States to date, however, about 60% of ethanol is delivered to fuel blending facilities by rail, with another 30% by truck and 10% by barge (NPC 2012). Pipelines are not generally used to transport ethanol for a variety of reasons. First, water and other impurities that normally reside in fuel pipelines can be absorbed by ethanol. The water and impuri- ties can then damage engines or degrade performance. Sec- ond, much of the existing pipeline infrastructure flows from the Gulf of Mexico north and east, while biofuels produced in the Midwest need to travel in the opposite direction and to the coasts (NPC 2012). While ethanol is produced in more than 20 states, 90% of the production capacity is clustered in just eight Midwest states—Iowa, Nebraska, Illinois, Min- nesota, South Dakota, Indiana, Kansas, and Wisconsin. Since roughly 80% of the U.S. population lives near the coasts, the distribution of biofuels at much greater volumes would pose significant challenges (Denicoff 2007). Transporting bio- diesel via pipeline is also problematic; due to certain proper- ties, it can sometimes cling to pipeline walls and get picked up by other fuels that flow through the pipeline subsequently (NPC 2012). C.1.3 Consumption and Refueling Once ethanol is transported to blending facilities, it is mixed with gasoline into the final fuel blend and then trucked to retail outlets. Consumption of ethanol in the United States has grown from 1.7 billion gallons in 2000 to 12.9 billion gal- lons in 2012 (EIA 2013c). Nearly all of the ethanol currently consumed in the United States is blended with gasoline in volumes containing up to 10% ethanol (E10) for use in con- ventional vehicles. The EPA issued waivers allowing the sale of 15% ethanol blends (E15) for use in any light-duty vehicles with a model year of 2001 or later. According to Koenig (2012), E15 was first offered for sale by a station in Kansas in 2012. While the retail price of E15 is less than that of lower ethanol blends, Koenig notes that the price differential is somewhat mislead- ing given that the higher ethanol concentration reduces the energy content of the fuel, resulting in a reduction in miles per gallon and vehicle range. Auto manufacturers have also been resistant to endorse the use of E15, even in new vehicles,

138 Legislation Tax Credit ($/gallon) Expiration Date Volumetric Ethanol Excise Tax 0.45 12/31/2011 Small Ethanol Producer Credit 0.10 12/31/2011 Biodiesel Tax Credit 1.00 12/31/2013 Small Agri-Biodiesel Producer Credit 0.10 12/31/2013 Renewable Diesel Tax Credit 1.00 12/31/2013 Credit for Production of Cellusoic and Algae-Based Biofuel 1.01 12/31/2013 Source: Schnepf (2013). Table C.1. Federal tax credits available for qualifying biofuels. claiming that the vehicles are not optimized to use blends of ethanol higher than E10 (Koenig 2012). Many automakers have produced specially configured flex- fuel vehicles to accommodate ethanol in blends of up to 85% (E85). The usage of E85 is currently concentrated in the Midwest, where the majority of American corn is produced. The availability of E85 fueling stations in the United States remains limited to date, with about 2,300 stations that are predominately concentrated in the Midwest (EERE 2013). The weather in many states dictates that the ethanol content of E85 be reduced to a 70% blend during colder months to prevent cold starting problems. As with ethanol, biodiesels can be blended with conven- tional diesel at manufacturing facilities before being distrib- uted by tanker truck, avoiding the need for separate storage infrastructure at gas stations. U.S. consumption of biodiesel, most of which is currently produced from soybean oil, has grown from 10 million gallons in 2001 to 870 million gallons in 2012, but continued growth may be hindered by compet- ing land uses (EIA 2013c). C.1.4 Government Policies to Support Biofuels While biofuels offer energy security benefits, they face impor- tant limitations in terms of cost and performance. Reductions in greenhouse gas emissions are modest with current feedstocks and production processes, and there are additional environ- mental concerns related to land use, water consumption, and local air pollution emissions. Even with these limitations, how- ever, use of biofuels has grown rapidly over the past two decades. Strong federal and state regulations and incentives have played a major role in spurring this growth. The federal and state governments have supported biofuels through a broad array of policy interventions. Without this support, biofuels would not be cost-competitive with con- ventional gasoline and diesel fuel, even at the higher petro- leum prices experienced over the past few years. In response to the 1970s oil embargo, the Energy Tax Act of 1978 removed a $0.04 per-gallon excise tax on gasoline for “gasohol” blends containing at least 10% ethanol (Koplow 2006). This was fol- lowed by a series of federal tax credits to further support bio- fuels, as summarized in Table C.1. The United States had also imposed a tariff of $0.54 per gallon on ethanol from Brazil for many years to help protect U.S. ethanol producers, though this expired in 2011 (Schnepf 2013). Beyond tax policy, the federal government and some states have further supported growth in biofuels through renewable fuel standards that require fuel vendors to provide an escalat- ing supply of ethanol, biodiesel, and more advanced biofuels. At the federal level, the Energy Independence and Security Act of 2007 set a renewable fuel standard (RFS2) that began with 13 billion gallons in 2010 and increases each year to reach a total of 36 billion gallons by 2022. RFS2 was finalized in Febru- ary 2010 and defines four categories of fuels: cellulosic biofuel, biomass-based diesel, advanced biofuels, and other renewable fuels. Fuels from each category must meet different standards for greenhouse gas reductions to qualify—for instance, cellu- losic biofuels must meet a 60% reduction over the petroleum baseline—and each category has separate volumetric targets for each year (EPA 2010). Schnepf (2013) estimated that U.S. government outlays for biofuels peaked at $7.7 billion in 2011 and then declined to $1.3 billion in 2012 with the expiration of the ethanol blend- er’s tax credit. Since current biofuel production processes, if stripped of tax incentives, are not yet cost-competitive with petroleum-based fuels for the most part, future increases in biofuel use in the United States will likely require continued subsidies along with the implementation of progressively more-stringent renewable fuel standards under RFS2. C.2 Vehicle Technologies Biomass-derived ethanol can be combusted efficiently in conventional vehicles using E10, but low-alcohol blends such as E10 do not qualify as alternative fuels under the Energy Policy Act of 1992. FFVs sold in the United States are able to use blends that range from 100% gasoline to 85% ethanol;

139 selling for $4.96 per gge (EERE 2012a). Biodiesel also tends to be more expensive than diesel fuel on an energy-equivalent basis, though the margin is much smaller than that observed for E85 and gasoline. In January of 2012, the national average price of diesel was $3.86, while BD20 sold for $4.02 per gallon of diesel equivalent (EERE 2012a). C.3.3 Operational Performance Ethanol contains less energy than gasoline on a volumetric basis, which is why the respective costs of gasoline and E85 are typically compared on an energy-equivalent basis. A gallon of ethanol contains the energy content of 0.67 gallons of gaso- line. This means that an FFV running on E85 will have only 72% of the range provided by pure gasoline, or about 74% of the range provided by E10. Correspondingly, the fuel economy for FFVs using E85 is about 25% to 30% lower, measured on a per-gallon basis rather than on an energy-equivalent basis, than when using conventional fuel (West et al. 2007). Beyond cost considerations, then, the use of E85 is also less convenient since it requires more frequent refueling stops. Note that vehicles optimized specifically for ethanol blends of E85 through E100 (100% ethanol) can in theory operate at equal or greater efficiency than when running on pure gaso- line. This is due to ethanol’s higher octane rating, allowing its use in higher-compression engines. Sensors monitoring the alcohol content can work in tandem with a turbocharger to push extra air into the engine cylinders when running on high-ethanol blends, producing extra power without losing fuel economy. This type of technology, however, is currently only in use in a few high-end supercars and concept cars (West et al. 2007). C.3.4 Greenhouse Gas Emissions Given the energy- and fertilizer-intensive nature of U.S. corn production and the potential for associated land-use changes to affect emissions, several studies (e.g., Mullins, Griffin, and Mathews 2010) of the well-to-wheels greenhouse gas effects of corn ethanol showed negligible reductions or even net increases in GHGs. (“Well-to-wheels” refers to the full fuel cycle, including producing, distributing, and com- busting the fuel.) Hence, the well-to-wheels GHG emissions associated with biofuels are subject to uncertainty and are a continued focus of research. In an earlier study, Delucchi (2006) compared well-to- wheels GHGs for gasoline, corn-based E90, and cellulosic E90 based on hypothetical specifications for an FFV in both the 2010 and 2040 time frames. The results are shown in Table C.2. Conducted before the recent advent of more-stringent federal fuel economy standards, the analysis assumes only a modest improvement in the fuel economy for a conventionally fueled blends above E85 are not used because higher ethanol con- centrations can make it difficult for vehicles to start in cold weather. FFVs include a sensor that automatically detects the amount of ethanol in the fuel, and the vehicle computer then modifies the fuel injection and spark timing accordingly. This involves a modest incremental cost, although automakers generally offer FFVs at the same price as comparable conventional vehicles. The impressive sales of FFVs in the United States to date may reflect the concerted marketing efforts of auto manufac- turers motivated by the opportunity to gain credits under the CAFE mandate rather than a response to strong consumer demand for the ability to fuel with E85. Between 1998 and 2009, the number of FFVs in the United States increased from a very small number to nearly 10 million (EIA 2013b). Yet available estimates indicate that only about 860,000 of these are fueled primarily with E85 (EIA 2013a). C.3 Cost and Performance Configuring conventional vehicles to run on biofuels involves minimal additional cost. Absent subsidies, however, ethanol and biodiesel remain more expensive than gasoline and diesel on an energy-equivalent basis. Additionally, bio- fuels face certain power and range limitations. Finally, though advanced biofuels promise potentially impressive reductions in GHG emissions, they may exacerbate some air quality challenges, especially for PM and NOx. C.3.1 Vehicle Cost In comparison to other alternative-fuel vehicle technolo- gies, such as electric vehicles or hydrogen fuel-cell vehicles, FFVs are relatively easy to produce, requiring only modest changes to conventional vehicle design. Corts (2010) estimates that the additional cost for producing an FFV in comparison to an otherwise identical conventional vehicle is only around $100 in most cases. Assuming that the other challenges asso- ciated with biofuels—such as higher energy-equivalent fuel cost and relative paucity of refueling infrastructure—could be addressed, vehicle cost would not be an impediment to much broader adoption. C.3.2 Energy Cost of Travel While gasoline and ethanol prices are both prone to vola- tility, ethanol tends to be more expensive than gasoline on a gge basis. In October of 2010, for example, the national aver- age price of gasoline was $2.78 per gallon, while E85 sold for $3.45 per gge (EERE 2010). In January of 2012, the average cost of gasoline had risen to $3.76 per gallon, while E85 was

140 biomass are compared on a well-to-wheels basis. For cur- rent biofuels, corn-based E85 estimates from GREET are used. For the 2050 ethanol baseline scenario, baseline 2050 fuel economy estimates from NRC (2013) and GREET esti- mates of carbon intensity for switchgrass-based cellulosic E85 are used. For the 2050 ethanol optimistic scenario, the fuel economy is increased to reflect NRC’s (2013) optimistic estimates. ANL’s GREET model (ANL 2012) estimates that a cur- rent FFV running on corn-based E85 would produce about 25% less GHGs on a well-to-wheels basis than an other- wise similar conventional vehicle; with cellulosic E85, the well-to-wheels reduction could be greater than 65%. It is important to note that GREET does not include emissions from indirect land-use changes associated with growing bio- mass, which are uncertain but potentially substantial (NPC 2012, NRC 2013). As noted, under some assumptions of emissions from land use change, current corn ethanol can have higher GHGs than conventional gasoline. Because the GREET model is used to compare GHGs across fuels, in Figure C.1 a case is not shown where corn ethanol has higher GHGs than gasoline, but it is important to continue to moni- tor the research in this area. Sources of uncertainty regarding future emissions for bio fuels include the effects of land use changes, potential process improvements, changes in emis- sions estimates associated with coproducts, and whether car- bon capture and storage is used during production. To bound future estimates, the researchers also looked at future cases of thermochemical drop-in gasoline with and without CCS, both of which include the indirect effects of land use changes as estimated by NRC (2013). Table C.3 provides a summary of the assumptions used in the GHG comparison test cases, including fuel economy for the vehicles in miles per gallon of gasoline equivalent (mpgge) and carbon intensity of the fuels in kilograms of carbon dioxide equivalent (CO2e) per gallon of gasoline equivalent (kg CO2e/gge). Based on these assumed specifications and GREET’s emis- sions factors, Figure C.1 graphs the greenhouse gas emissions performance for current and future conventional and biofuel vehicle, and in turn for FFVs, between 2010 and 2040. Even so, the potential GHG emissions reductions for cellulosic ethanol in the 2040 time frame, and even in the 2010 time frame, are significant. The reductions for corn-based ethanol, in contrast, are modest. Cellulosic ethanol, examined on a well-to-wheels basis, is not generally expected to eliminate all GHG emissions because there are still emissions from the production of feedstocks and fertilizers and from distribution and land use changes. Indeed, estimates of the full fuel-cycle GHG emissions for eth- anol vary substantially depending on the method of analysis, with assumptions about the proper allocation of emissions to coproducts and about the estimation and inclusion of indirect emissions stemming from land use changes responsible for much of the uncertainty (e.g., Farrell et al. 2006, Searchinger et al. 2008, Williams et al. 2009). Additionally, cellulosic etha- nol has yet to be produced at commercial scale, adding further uncertainty to the estimates. To facilitate a consistently framed examination of the GHG emissions reduction potential for the different fuels and vehi- cle technologies considered in this study, the research team conducted an exercise relying on emissions factors from ANL’s GREET model (ANL 2012). The team used the GREET speci- fications for a mid-size passenger vehicle in 2010 along with two hypothetical configurations for similar vehicles in the 2050 time frame. The two future cases are broadly consistent, with expectations in the literature regarding possible advances in fuels and vehicle technologies, but one embeds moderate baseline assumptions while the other is more optimistic. In the following, the GHG emissions modeling results are presented for conventional vehicles and biofuel vehicles for 2010 and 2050. The intent of including both current and future conventional vehicles is to help clarify not only how biofuels might perform in relation to the current light-duty fleet but also how they might perform against conventional fuels and vehicles with much-improved fuel economy in the future. On-road (as opposed to EPA-rated) fuel economy is used for all vehicles, and corn-based E85, switchgrass-based cellulosic E85, and gasoline derived thermochemically from Fuel 2010 2040 GHG Emissions (grams/mile CO2 equiv.) Percent Reduction vs. 2010 Gasoline GHG Emissions (grams/mile CO2 equiv.) Percent Reduction vs. 2010 Gasoline Percent Reduction vs. 2040 Gasoline Gasoline 550.7 — 504.1 8% — E90 corn 539.7 2% 458.7 17% 9% E90 cellulosic 308.4 44% 146.2 73% 71% Source: Delucchi (2006). Table C.2. Well-to-wheels GHG reductions for E90 vehicles in 2010 and 2040.

141 corn-based E85 could lead to increased emissions for vola- tile organic compounds, nitrogen oxides, and fine particulate matter in comparison to standard gasoline. Future reductions in air pollutants from biofuels could be achieved, however, as biofuel feedstocks and production processes are improved and rely on electricity with fewer emissions. C.4 Market Prospects Assisted by government subsidies and RFS2 requirements, the production of biofuels has scaled up rapidly in recent years, and the number of FFVs has increased as well. Most projections indicate increased production over the next sev- eral decades, though there is greater uncertainty regarding their ultimate market potential. vehicles in grams CO2-equivalent per mile of travel. All of the estimates are for well-to-wheels emissions. Note that no value is listed for drop-in green gasoline in 2010 because this prod- uct is still in the research and development phase. Additionally, given that the optimistic case for green gasoline includes car- bon capture and sequestration in the production process, net GHG emissions are actually negative, consistent with plausible future outcomes from NRC (2013). C.3.5 Local Air Pollutant Emissions Although advanced liquid biofuels promise significant GHG reduction benefits in future decades, their expected effects with respect to air quality are less certain. The GREET model (ANL 2012) estimates, for example, that the use of Source: Computations by authors based on data from ANL (2012) and NRC (2013). -200 -100 0 100 200 300 400 500 Green Gasoline ICE E85 FFV Gasoline ICE Grams CO2-Equivalent per Mile of Travel (well-to-wheels) 2010 2050 Conserva ve 2050 Op mistic Figure C.1. GHG reduction prospects for liquid biofuels in 2050. Assumptions 2010 2050 Base 2050 Optimistic Assumed fuel economy for gasoline ICE and FFV mpgge 24.8 72 91 Life-Cycle Fuel Carbon Intensity Assumptions for FFV running on E85 Pathway Corn ethanol Cellulosic ethanol from switchgrass Cellulosic ethanol from switchgrass kg CO2e/gge 8.3 3.2 3.2 Life-Cycle Fuel Carbon Intensity Assumptions for gasoline ICE running on drop-in green gasoline Pathway N/A Thermochemical with indirect land-use change Thermochemical with indirect land-use change and carbon capture and sequestration kg CO2e/gge N/A 5.0 -9.0 Source: Computations by authors based on data from ANL (2012) and NRC (2013). Table C.3. Assumptions for biofuels emissions comparisons.

142 than $60 per ton. They found that the potential resource is more than adequate to produce 20 billion gallons of cellulosic ethanol (DOE 2011). C.4.2 Factors Affecting Market Prospects While the projections and studies just discussed suggest the possibility for continued growth in biofuels, there are several remaining issues and challenges that could, if not successfully addressed, act to limit their ultimate market potential. Economics. The primary barrier for biofuels to displace large amounts of petroleum is economic; unless the cost to produce cellulosic biofuels declines, such fuel will require subsidies or mandates to be competitive with oil at any price less than $190 per barrel (NRC 2013). While drop-in fuels such as green gasoline and green diesel appear to hold prom- ise over the longer term, they are still at a relatively early stage in the research and development cycle, and their economic viability remains uncertain. Another concern is the potential for higher food prices if food crops are diverted for fuel pro- duction or if the production of energy crops drives up the price per acre of arable land that could otherwise have been used to produce food. Uncertain technical prospects for cellulosic ethanol. Presently in the United States almost all ethanol is produced from corn. While cellulosic ethanol promises higher yields and greater greenhouse gas emissions reductions, it is pro- duced in only limited quantity—mainly in demonstration projects—given its high cost. Many technologies for biologi- cal and thermochemical conversion of cellulosic feedstocks to ethanol are being explored, but to date, a low-cost, efficient, and commercially scalable solution has yet to emerge (Yang and Wyman 2008, Laser et al. 2009). Farmers will need to have some certainty that the demand for cellulosic feedstocks will rise significantly before they are willing to invest in the production of cellulosic feedstocks. Air quality effects. Emissions for certain criteria pollutants can be worse for corn-based ethanol than for gasoline-fueled vehicles. If criteria pollutant emissions continue to play an important role in government regulations, corn-based bio- fuels may not qualify for subsidies or other benefits. Bio- fuels perform better with respect to greenhouse gases, but the benefits from corn-based biofuels are modest at best. This could lead to more pressure to move from corn to cellulosic feedstocks for greater emissions reductions. Blending and storage capacity. The present number of blending facilities is relatively small and would need to be greatly increased to accommodate a significant role for ethanol and biodiesel within the transportation sector. These facilities have relatively high capital costs, and producers may be unwill- ing to make the necessary investments until there is greater certainty regarding future market share for biofuels. Ethanol C.4.1 Future Market Projections In its most recent Annual Energy Outlook, the EIA (2013b) projects that the share of new light-duty cars and trucks sold with flex-fuel capability will hold steady at approximately 9% between 2012 and 2040. However, usage of E85 in the light- duty fleet over this same time period is expected to acceler- ate, growing from about 117 million gallons in 2012 to about 2 billion gallons in 2040. EIA’s forecast for E85 is highly sensi- tive to assumptions about the price of oil. Under EIA’s high oil price scenario, for example, use of E85 in 2040 increases to 7.4 billion gallons. The U.S. Department of Agriculture (USDA) issues 10-year projections for corn production, including the share that will be devoted to corn ethanol, with the most recent release offer- ing estimates through 2023 (USDA 2013). About 5 billion bushels of U.S. corn were diverted to make ethanol and by- products in 2011-2012, representing about 40% of the total U.S. corn crop. In 2022–2023, the USDA expects that the amount of corn used for ethanol will be about the same, at 5.3 billion bushels. Because total production is expected to rise, though, corn for ethanol will only account for about 34% of total corn production. The farm price per bushel is forecasted to drop from $6.22 in 2011–2012 to $4.85 in 2022– 2023, which could translate into lower costs for corn ethanol (USDA 2013). Several additional studies have examined long-range prospects for biofuel production in the United States (e.g., Perlack et al. 2005, DOE 2011, NRC 2011). Perlack et al. (2005) examined the question of whether biofuel production could displace 30% of current U.S. petroleum consumption by 2030, which would translate to 90 billion gallons of biofuels per year. The authors concluded that biofuels could achieve that goal under aggressive assumptions about improved grain yields, expansion of production to idle cropland and pastures, and greater utilization of residues and manures. Several reviews of the Perlack study, however, have cau- tioned that the assumptions employed might be overly aggres- sive. West et al. (2009) used a sensitivity analysis to understand the likelihood of meeting the goal of 90 billion gallons under various conditions. They concluded that oil and feedstock prices, large-scale development of energy crops, and improve- ments in cellulosic conversion yields would all be important factors in whether the goal could be met. For example, if short-rotation woody crops were not available, the expected yield would be about 78% of the 90 billion gallon per year tar- get. If no additional energy crops became available, the yield might be only 50% of the target value. The DOE updated the Perlack study in 2011. The updated study forecasted that 767 to 1,305 million tons of additional biomass could be available in 2030, depending on energy crop productivity assumptions, at a farm gate price of less

143 EERE. 2010. Clean Cities Alternative Fuel Price Report. EERE. 2012a. Clean Cities Alternative Fuel Price Report. EERE. 2012b. Hydrogenation-Derived Renewable Diesel. Alternative Fuels Data Center. http://www.afdc.energy.gov/fuels/emerging_ green.html (accessed October 17, 2013). EERE. 2013. Alternative Fueling Station Locator. Alternative Fuels Data Center. http://www.afdc.energy.gov/locator/stations/ (accessed October 17, 2013). EIA. 2013a. Renewable and Alternative Fuels: Alternative Fuel Vehi- cle Data. http://www.eia.gov/renewable/afv/index.cfm (accessed October 17, 2013). EIA. 2013b. Annual Energy Outlook 2013. EIA. 2013c. Monthly Energy Review, September 2013. EPA. 2010. Renewable Fuel Standard Program (RFS2) Regulatory Impact Analysis. Farrell, A., R. Plevin, B. Turner, A. Jones, M. O’Hare, and D. Kammen. 2006. “Ethanol Can Contribute to Energy and Environmental Goals.” Science, 311 (5760): 506–508. Koenig, B. 2012 (July 12). “Kansas Gas Station First to Offer E15 Ethanol- Blend Fuel.” The New American. Koplow, D. 2006. Biofuels – At What Cost? Government Support for Ethanol and Biodiesel in the United States. The Global Subsidies Initiative, International Institute for Sustainable Development, Geneva. Laser, M., H. Jin, K. Jayawardhana, B. Dale, and L. Lynd. 2009. “Pro- jected Mature Technology Scenarios for Conversion of Cellulosic Biomass to Ethanol with Coproduction Thermochemical Fuels, Power, and/or Animal Feed Protein.” Biofuels, Bioproducts and Bio- refining, 3 (2): 231–246. Moreira, J. R. and J. Goldemberg. 1999 (April). “The Alcohol Program.” Energy Policy, 27 (4): 229–245. Mullins, K. A., W. M. Griffin, and H. S. Matthews. 2010. “Policy Impli- cations of Uncertainty in Modeled Life-Cycle Greenhouse Gas Emissions of Biofuels.” Environmental Science & Technology, 45 (1): 132–138. NPC. 2012. Future Transportation Fuels Study: Advancing Technology for America’s Transportation Future. NRC. 2009. Liquid Transportation Fuels from Coal and Biomass: Tech- nological Status, Costs, and Environmental Impacts. The National Academies Press, Washington, D.C. NRC. 2011. Renewable Fuel Standard: Potential Economic and Environ- mental Effects of U.S. Biofuel Policy. The National Academies Press, Washington, D.C. NRC. 2013. Transitions to Alternative Vehicles and Fuels. The National Academies Press, Washington, D.C. Parker, N., B. Jenkins, P. Dempster, B. Higgins, and J. Ogden. 2011. Chapter 1: The Biofuels Pathway. In J. Ogden and L. Anderson (eds.), Sustainable Transportation Energy Pathways: A Research Summary for Decision Makers. Institute of Transportation Studies, University of California, Davis. Perlack, R., L. Wright, A. Turhollow, R. Graham, B. Stokes, and D. Erbach. 2005. Biomass as Feedstock for a Bioenergy and Bioprod- ucts Industry: The Technical Feasibility of a Billion-Ton Annual Supply. U.S. Department of Agriculture and U.S. Department of Energy. Searchinger, T., R. Heimlich, R. Houghton, F. Dong, A. Elobeid, J. Fabiosa, S. Tokgoz, D. Hayes, and T.-H. Yu. 2008. “Use of U.S. Croplands for Biofuels Increases Greenhouse Gases through Emissions from Land-Use Change.” Science, 319 (5867): 1238–1240. Schnepf, R. 2013. Agriculture-Based Biofuels: Overview and Emerging Issues. Congressional Research Service. is commonly blended into conventional gasoline in concen- trations of up to E10, and this may rise over time to E15. As the total volumetric requirements of the RFS rise annually, however, and as a progressively more efficient vehicle fleet uses less gasoline, higher blending percentages will be required to meet biofuel targets. In contrast, this blend-wall challenge is not an issue for drop-in biofuels, which will likely hasten their development. Transport capacity. The biomass ethanol supply chain uses trucks, rail, and, in some cases, barges to distribute feedstocks, ethanol, and the final blended product. As discussed earlier, nearly 60% of ethanol currently travels by rail and another 30% by truck. Significant increase in the production and transporta- tion of feedstocks and biofuels, without the wider use of pipe- lines, would add further stress to the national freight system. Vehicle stock and refueling infrastructure. Tyner, Dooley, and Viteri (2011) provide further analysis of the blend wall that caps the amount of ethanol usable by the transportation sector as a result of overall fuel demand and limits on blend- ing ratios. Expanding consumption to meet a mandate of 22 billion gallons per year of renewable fuel by 2022 using only ethanol would, in the authors’ estimation, require over $30 billion in infrastructure and manufacturing investments to expand usage of E85 from 30 million gallons in 2010 to 23.5 billion gallons in 2022. This is compatible with projections of growth in flex-fuel vehicles but would require massive expansion in the number of gas stations offering E85 dispens- ers for those FFVs to actually use E85. Fuel price volatility. If the cost of feedstocks and the price of gasoline fluctuate significantly, farmers and biofuel pro- ducers may not have confidence in their ability to produce biofuels profitably over the longer time frames required to justify significant capital investments. As an example, the high price of sugar combined with the low price of petroleum contributed to the short-term crash of biofuels in Brazil dur- ing the early 1990s (Moreira and Goldemberg 1999). References ANL. 2012. The Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (GREET), Fuel Cycle Model version 1 2012. U.S. Department of Energy. Bartis, J. T. and L. Van Bibber. 2011. Alternative Fuels for Military Appli- cations. RAND Corporation, Santa Monica. Corts, K. 2010. “Building Out Alternative Fuel Retail Infrastructure: Government Fleet Spillovers in E85.” Journal of Environmental Economics and Management. 59: 219–234. Delucchi, M. 2006. Lifecycle Analysis of Biofuels. Institute of Transporta- tion Studies, University of California, Davis. Denicoff, M. R. 2007. Ethanol Transportation Backgrounder: Expansion of U.S. Corn-Based Ethanol from the Agricultural Transportation Perspective. U.S. Department of Agriculture. DOE. 2011. U.S. Billion-Ton Update: Biomass Supply for a Bioenergy and Bioproducts Industry.

144 A. Lutz, C. Shaddix, N. Brinkman, C. Wheeler, and D. O’Toole. 2009. Feasibility, Economics, and Environmental Impact of Produc- ing 90 Billion Gallons of Ethanol per Year by 2030. Sandia National Laboratory. Williams, P., D. Inman, A. Aden, and G. A. Heath. 2009. “Environmental and Sustainability Factors Associated with Next-Generation Bio- fuels in the U.S.: What Do We Really Know?” Environmental Science & Technology, 43 (13): 4763–4775. Yang, B. and C. Wyman. 2008. “Pretreatment: The Key to Unlocking Low-Cost Cellulosic Ethanol.” Biofuels, Bioproducts and Biorefining. 2 (1): 26–40. Tyner, W., F. Dooley, and D. Viteri. 2011. “Alternative Pathways for Ful- filling the RFS Mandate.” American Journal of Agricultural Econom- ics, 93 (2): 465–472. USDA. 2013. Agricultural Projections to 2023. http://www.usda. gov/wps/portal/usda/usdahome?contentid=2013/02/0021. xml&contentidonly=true (accessed October 17, 2013). West, B., A. López, T. Theiss, R. Graves, J. Storey, and S. Lewis. 2007. “Fuel Economy and Emissions of the Ethanol-Optimized Saab 9-5 Biopower.” SAE Technical Paper, SAE 2007-01-3994. West, T., K. Dunphy-Guzman, A. Sun, L. Malczunski, D. Reichmuth, R. Larson, J. Ellison, R. Taylor, V. Tidwell, L. Klebanoff, P. Hough,