Effects of Ethanol on Vehicle Energy Efficiency and ... - ACS Publications

Apr 29, 2013 - ... though changes to the vehicle fleets and fuel infrastructure would be required. .... and Iso-Butanol Blends from a Fleet of Nine PF...
0 downloads 0 Views 1MB Size
Download PDF
Policy Analysis pubs.acs.org/est

Effects of Ethanol on Vehicle Energy Efficiency and Implications on Ethanol Life-Cycle Greenhouse Gas Analysis Xiaoyu Yan,*,†,‡ Oliver R. Inderwildi,§ David A. King,§ and Adam M. Boies† †

Department of Engineering, University of Cambridge, Trumpington Street, Cambridge, CB2 1PZ, U.K. Environment and Sustainability Institute, University of Exeter, Penryn, Cornwall, TR10 9EZ, U.K. § Smith School of Enterprise and the Environment, University of Oxford, Hayes House, 75 George Street, Oxford, OX1 2BQ, U.K. ‡

S Supporting Information *

ABSTRACT: Bioethanol is the world’s largest-produced alternative to petroleum-derived transportation fuels due to its compatibility within existing spark-ignition engines and its relatively mature production technology. Despite its success, questions remain over the greenhouse gas (GHG) implications of fuel ethanol use with many studies showing significant impacts of differences in land use, feedstock, and refinery operation. While most efforts to quantify life-cycle GHG impacts have focused on the production stage, a few recent studies have acknowledged the effect of ethanol on engine performance and incorporated these effects into the fuel life cycle. These studies have broadly asserted that vehicle efficiency increases with ethanol use to justify reducing the GHG impact of ethanol. These results seem to conflict with the general notion that ethanol decreases the fuel efficiency (or increases the fuel consumption) of vehicles due to the lower volumetric energy content of ethanol when compared to gasoline. Here we argue that due to the increased emphasis on alternative fuels with drastically differing energy densities, vehicle efficiency should be evaluated based on energy rather than volume. When done so, we show that efficiency of existing vehicles can be affected by ethanol content, but these impacts can serve to have both positive and negative effects and are highly uncertain (ranging from −15% to +24%). As a result, uncertainties in the net GHG effect of ethanol, particularly when used in a low-level blend with gasoline, are considerably larger than previously estimated (standard deviations increase by >10% and >200% when used in high and low blends, respectively). Technical options exist to improve vehicle efficiency through smarter use of ethanol though changes to the vehicle fleets and fuel infrastructure would be required. Future biofuel policies should promote synergies between the vehicle and fuel industries in order to maximize the society-wise benefits or minimize the risks of adverse impacts of ethanol.

1. INTRODUCTION Bioethanol is increasingly promoted as an alternative transport fuel worldwide and global production rapidly increased from 17 to 86 billion liters between 2000 and 2011 with government support such as mandates, subsidies and tax benefits.1 Ethanol is commonly used as a blend with gasoline (see Supporting Information Table S1) and the most typical blends are low blends such as E10 (E represents ethanol and the number represents the volume percentage of ethanol) for regular gasoline vehicles. High blends such as E85 are used in flexiblefuel vehicles (FFVs) that can run on any mixture between pure gasoline and E85. Brazil is unique in that ethanol accounts for a sizable portion of its vehicle fuel use and Brazilian gasoline contains 20−25% ethanol. Hydrous ethanol, containing 5−6% of water by volume, is also used and Brazilian FFVs can run on any mixture between E20 and hydrous ethanol. The rapid growth of ethanol use has sparked extensive research activities exploring the environmental impact of ethanol, greenhouse gas (GHG) emissions in particular, over the entire fuel life cycle, covering biomass feedstock production, © 2013 American Chemical Society

ethanol conversion and the combustion of ethanol in vehicle engines. However, most life cycle analyses (LCAs) to date have mainly focused on the production stage while assuming end-use efficiencies are the same for ethanol and gasoline2−7 despite the fact that ethanol can, in theory, affect engine efficiency and performance due to different thermo-chemical properties compared to gasoline. The resulting estimates of the net GHG impacts of ethanol could therefore be incomplete. Several recent highly cited LCAs8−11 that have incorporated vehicle efficiency differences between gasoline and ethanol/gasoline blends could potentially be misleading. This is because they have relied on rather differing results from a limited number of sources while ignoring the vehicle-to-vehicle variations and the inherent uncertainties in vehicle efficiency measurements (see Section 4.1.5 for details). Received: Revised: Accepted: Published: 5535

December 20, 2012 April 4, 2013 April 29, 2013 April 29, 2013 dx.doi.org/10.1021/es305209a | Environ. Sci. Technol. 2013, 47, 5535−5544

Environmental Science & Technology

Policy Analysis

The objectives of this paper therefore include (1) to examine the effect of ethanol blending on the energy efficiency of in-use vehicle models using a common metric and available empirical evidence; (2) to explore the impact of incorporating ethanol’s effect on vehicle efficiency into its LCA; (3) to determine whether potential exists for increased vehicle efficiency through ethanol use; (4) to explore whether there is an optimal blending level for a given quantity of ethanol, for example, large quantities of low blends versus small quantities of high blends.

Ethanol’s higher volume (molar) ratio of product gas to reactants per unit energy could theoretically improve engine efficiency as more work is performed during the expansion stroke.18 A recent study21 also found that the molar expansion ratio could be an important factor in explaining the differences in exergy-to-LHV ratio and hence First- and Second-Law efficiency of different fuels. For instance, the exergy-to-LHV ratio of ethanol was found to be higher than those of hydrocarbons, suggesting that ethanol has a higher potential to do work than implied by its LHV.21 In practice, however, whether and to what extent an engine can take advantage of the favorable properties of ethanol depend on engine design parameters such as CR, knock sensor and engine control unit logic, operating conditions such as speed and load, ignition timing and fuel/air equivalence ratio, and the blends used. These can be examined through carefully controlled engine tests. Experimental work on engines in test cells suggests that efficiency and power for carburettor engines at various operating conditions increases with increasing ethanol content for low to medium blends by up to 9% mainly because of higher volumetric efficiency and leaner operation.22−24 The observed efficiency effects for ethanol/gasoline blends on indirect fuel injection (FI) engines, including single- and multipoint injection, and direct injection (DI) engines, were largely inconsistent. Efficiency was found to increase with increasing ethanol content for low to medium blends by up to 9% at knock-limited conditions such as full engine load25−29 while at knock-free conditions such as part load, efficiency remained relatively unchanged25,30−32 or even reduced in some cases.29,33 Efficiency increases of up to 10% at part load30,31 and up to 25% at full load28,29 were observed for high blends. The main reason for these observations is that the control systems of most modern engines can regulate ignition timing at various operating conditions based on the level of knock detected by knock sensors.19,34 At knock-limited conditions, the better antiknock properties of ethanol/gasoline blends allows a more optimal ignition timing in terms of efficiency.25,26,31 The much higher efficiencies for high blends were also believed to be attributed to the charge cooling in the intake system and the combustion characteristics of ethanol.28,30,31 The above engine results are based on gasoline and ethanol/ gasoline blends operating at the same CR. When CR is allowed to increase with increasing ethanol content, even higher efficiency gains can be realized.24,28,33 Although these studies provide insights in the potential effects of ethanol on engine efficiency at controlled operating conditions, they might not reflect the blends currently in use, behavior of production engines (as research engines were used in many studies) and practical operating conditions. In order to quantify the actual effects of ethanol on vehicle efficiency, vehicle tests are examined in detail in the next section.

2. BACKGROUND Life-cycle GHG emissions per MJ fuel energy (GHG intensity) for ethanol via different production pathways have been subject to extensive study.2−7 Although these earlier efforts have highlighted how differences in land use, feedstock and refinery operation can alter the GHG intensity of the fuel, they were all conducted by deterministic LCA. U.S. corn ethanol, for instance, has been shown to have a GHG intensity of 79−85 gCO2e/MJ (compared with 96 gCO2e/MJ for gasoline).2,3 This intensity could increase to 177 gCO2e/MJ if carbon flux from land-use change (LUC) is considered.5 Some more recent studies have incorporated the inherent uncertainties of input parameters using probabilistic modeling methods and found GHG intensity of ethanol to be highly uncertain.12−16 For instance, GHG intensity for U.S. corn ethanol could range from 50 to over 200 gCO2e/MJ when the uncertainties of key parameters such as LUC emissions and soil nitrous oxide emissions are incorporated.12,14 However, most LCAs to date, both deterministic and probabilistic, compare emissions on a fuel energy content or lower heating value (LHV) basis, that is, assuming 1 MJ ethanol displaces 1 MJ gasoline regardless of how the ethanol is used. A reason for such comparison could be that there is a long-standing dispute over the actual effect of ethanol on vehicle efficiency with many conflicting claims. The effect is further complicated due to the different volumetric energy densities of gasoline and ethanol and the fact that vehicle efficiency is usually expressed in volumetric units such as miles per gallon (MPG) and kilometres per liter (km/L). Ethanol has rather different physical and chemical characteristics compared with gasoline, which could potentially affect the performance and efficiency of spark-ignition (SI) engines (see Supporting Information Table S2). Engine efficiency hereafter refers to fuel conversion efficiency as defined by Heywood,17 that is, the ratio of the work produced to the amount of fuel energy supplied. As SI engine efficiency is generally limited by the combustion phenomenon “knock”, the main benefits of ethanol stems from its higher octane rating and hence higher resistance to knock. This allows dedicated ethanol engines to use much higher compression ratios (CRs, 13−16) than those used in regular gasoline engines (CR 8−11), resulting in considerable increases in engine efficiency and power for a given engine size.17,18 The much higher heat of vaporisation of ethanol and the resulting cooling effect increases volumetric efficiency and contributes further to knock resistance.18 The presence of oxygen in the fuel molecule of ethanol enables higher combustion efficiency (i.e., more complete combustion) and hence higher engine efficiency. However, water solubility is very low in gasoline but high in ethanol. Even small amounts of ethanol can considerably increase the water solubility of ethanol/gasoline blends and hence the tendency to absorb water vapor from the atmosphere. If sufficient water is present, phase separation of the blend will occur and can cause serious engine problems.19,20

3. MATERIALS AND METHODS We determine the vehicle efficiency effect of ethanol by compiling paired energy efficiency data from all known studies of existing SI-engine vehicles operating on ethanol/gasoline blends and pure gasoline.35−85 The statistical analysis is based on those studies that tested both ethanol/gasoline blends (with nominal ethanol volume no less than 5%) and gasoline on the same vehicle models under comparable real-world driving conditions or over the same standard drive cycles on a dynamometer, excluding tests on fuel/vehicle combinations 5536

dx.doi.org/10.1021/es305209a | Environ. Sci. Technol. 2013, 47, 5535−5544

Environmental Science & Technology

Policy Analysis

VEE on ethanol/gasoline blends and gasoline are strongly correlated (see Supporting Information Figure S1). Figure 1a

that are not in actual use (e.g., regular gasoline vehicles on blends higher than E20). Tests under different controlled conditions (e.g., different ambient temperatures and accumulated vehicle mileages) and on fuels of different targeted properties (e.g., distillation characteristics and vapor pressures) were considered individually. For studies that only reported emission measurements, we have calculated vehicle efficiencies using the carbon balance method and have excluded studies without CO2 emission measurements. Moreover, the following were excluded unless the authors of these studies were able to supply additional information and data: (1) poorly documented studies with little information available on test conditions and procedures and specifications of test vehicles and fuels; (2) studies that report qualitative results and quantitative data were not accessible; and (3) studies that report aggregated results for a fleet of vehicles rather than for individual vehicle models. The screening process reduces the number of studies for analysis to twenty six (see Supporting Information Table S3 for reasons for excluding the other studies).35−60 These include published scientific articles and reports by various organisations, covering most regions where fuel ethanol is currently in use, various types of vehicles and different ethanol blending levels (see Supporting Information Table S4 for details of these studies). No study was found that tested the efficiency of Brazilian vehicles on both ethanol/gasoline blends and pure gasoline, not surprising given that gasoline containing 20−25% ethanol has been in use for many years in Brazil. In most studies included for analysis, vehicle efficiency is expressed in various volume-based units. To allow meaningful comparisons, we convert all vehicle efficiency measurements into vehicle energy efficiency (VEE, km/MJ) based on fuel energy densities reported in each study. For studies in which fuel energy density was not specified, 32 and 21.1 MJ/l were used for gasoline and ethanol respectively and those for ethanol/gasoline blends were calculated based on the volume shares of the ethanol components. Nominal values were used if the volume share of ethanol in a blend was not measured. The associated uncertainties for these assumptions were assessed in Section 4.1.4. While VEE is important, it does not directly reveal the amount of gasoline displaced by a given amount of ethanol. This is because the ethanol component may affect the conversion efficiency of the gasoline component in a blend, that is, make an engine more or less efficient in its use of the gasoline component. We therefore employ a factor called Effective Substitution Ratio (ESR) to represent the underlying substitution effects between ethanol and gasoline. For instance, the ESR is 1.5 if the use of 1 MJ ethanol leads to a savings of 1.5 MJ gasoline. The ESR is therefore a crucial scaling factor when evaluating the net effects of substituting gasoline with ethanol in terms of energy, GHG emissions and costs. The ESR for all the paired VEE values is calculated by the expression below based on vehicle energy consumption (MJ/ km, the inverse of VEE), ESR = (EG − EGB)/E EB

Figure 1. (a) Relative changes in vehicle energy efficiency (km/MJ) on ethanol/gasoline blends over those on gasoline for different blending levels; (b) Effective substitution ratios for different blending levels. Calculated based on results from refs 35−60.

shows the relative changes in VEE for different blends over gasoline. VEE is on average 2.7% higher for ethanol/gasoline blends than for gasoline but varies within a wide range from 14.9% lower to 23.6% higher. While the mean VEE increases with increasing ethanol content (from 0.3% higher for E5 to 3.3% higher for E85 with an R2 of 0.06), the ranges of VEE changes are quite large for all blends including both positive and negative changes. The results for ESR show that it can range from −1.42 to 4.36 and the overall average for all blends is 1.08 (see Figure 1b). Vehicle-to-vehicle variations, differences in test conditions and the uncertainties in ESR calculations (see Section 4.1.4) all could have contributed to this large range. A negative ESR implies that the use of ethanol/gasoline blends considerably reduces VEE and this can be caused by poor engine tolerance of ethanol. There is a generally decreasing trend for the mean ESR with increasing ethanol content, contrary to the increasing trend for mean VEE discussed earlier. Moreover, the variations for ESR are much higher for low blends than for high blends. These are mainly because the lower the blends, the more sensitive ESR is to changes in VEE, reflecting the leveraging effects of the ethanol component on the gasoline component (see Supporting Information Figure S2). 4.1.2. Results for Different Engine/Fuel Combinations. As the efficiency effects of ethanol are dependent on both engine design and blending levels, we show the observed frequencies

(1)

where, EG is energy consumption on gasoline (MJ/km) and EGB and EEB are consumption of the gasoline and ethanol components for a blend (MJ/km), respectively.

4. RESULTS AND DISCUSSION 4.1. End-Use Stage. 4.1.1. Overall Results. When all vehicle types and ethanol blending levels are considered, paired 5537

dx.doi.org/10.1021/es305209a | Environ. Sci. Technol. 2013, 47, 5535−5544

Environmental Science & Technology

Policy Analysis

Figure 2. Normalized frequency distributions (the number of observations in each bin divided by the total number of observations) of ESR for different engine-fuel combinations: (a) carburettor-engine vehicles on E5-E20; (b) FI-engine vehicles on E5-E20; (c) DI-engine vehicles on E5-E20; (d) FFVs on E85. Orginial data sources: (a) refs 35,37,38,42; (b) refs 37,38,40−46,48,50−53,55−57,59; (c) refs 45,49,50,53−56,58; (d) refs 36,46,47,50,55−57,59,60.

of ESR for combinations of engine technologies and blending levels commercially available (excluding Brazil) in Figure 2. Carburettor- and FI-engine vehicles on E5-E20 achieve mean ESR of 1.25 and 1.17 respectively, both with median slightly lower than the mean. However, the ESR spans over wide ranges for both combinations, with a 95% confidence interval (95%CI) of 0.57:2.36 and 0.50:2.15 for carburettor- and FI-engine vehicles, respectively. In particular, five of the six ESR values higher than 3 observed for FI-engine vehicles are derived from one study,41 and the three highest values are from one vehicle with a knock sensor. The mean and median ESR for DI-engine vehicles on E5E20 are 0.85 and 0.98 respectively (see Figure 2c). However, the five negative ESR values are derived from one vehicle on E5.45 In another study, a vehicle of the same make and model also performed unfavorably on E5,55 achieving ESR of 0.36 and 0.59. These results suggest that this vehicle model is probably not tuned for ethanol tolerance. On the other hand, the two ESR values higher than 2 are derived from “on-road” tests,49 which is expected to be less reliable than dynamometer tests under controlled laboratory conditions.34 The mean and median ESR for DI-engine vehicles on E5-E20 would be 1.03 and 1.04 respectively if these less-reliable values were excluded. Figure 2d shows that FFVs (including both FI and DI engines as the difference between the two was negligible) are most likely to achieve a mean ESR slightly higher than unity with a much tighter probability distribution (a 95%-CI of 0.92:1.16). As there were extreme values for all low-blend combinations, the medians were considered to be more reliable estimates of the true values than the arithmetic means. In general, carburettor-engine vehicles appear to be able to better take advantage of low blends than FI-engine vehicles while DI-

engine vehicles do not seem to benefit. Nevertheless, with many fewer observations and the fact that DI engines are just beginning to enter the market, more tests are needed for DIengine vehicles on low blends. 4.1.3. Effects of Compression Ratios and Drive Cycles. As CR is expected to have a major impact on the engine efficiency effects of ethanol, we examine whether this is observable from the vehicle tests. As many studies did not report CR of the engines, we obtain CR values from public Web sites86−88 based on vehicle details reported in the studies (e.g., make and model, model year, and engine size). The relation between the efficiency effects of ethanol/gasoline blends and engine CR appears to be very weak for carburettor-engine vehicles and insignificant for FI-engine vehicles (see Supporting Information Figure S3). This suggests that while CR plays a role, other engine design and operational factors contribute to VEE changes when ethanol/gasoline blends were used. As some studies conducted vehicle tests on different drive cycles and/or both urban and highway (or extra urban) parts of the same combined cycle, we were able to examine the impact of drive cycles on the VEE effects of ethanol/gasoline blends. However, there were no consistent differences in ESR between urban and highway cycles as well as urban and aggressive cycles in general given variations by vehicle models (see Supporting Information Figure S4). This is somewhat unexpected as VEE is expected to benefit more from ethanol/gasoline blends under aggressive cycles where knock-limited operation is more frequent. 4.1.4. Uncertainties in ESR Calculations. There are many known uncertainties involved in calculating the ESR. For instances, although the LHV of ethanol is constant (as it is a single-molecule fuel), the LHV of gasoline can vary by a few percent by grade, by season and among refiners.34 Vehicle 5538

dx.doi.org/10.1021/es305209a | Environ. Sci. Technol. 2013, 47, 5535−5544

Environmental Science & Technology

Policy Analysis

Figure 3. Probability density functions of Life-cycle GHG emission changes relative to gasoline resulted from the use of 1 MJ corn and switchgrass ethanol with parameters included for uncertainty analysis shown in brackets (default value for corn ethanol does not include LUC emissions).

overall uncertainty would be greatly reduced if LHV is measured (Scenario 2). The contributions of fuel efficiency variations to overall uncertainty are notable for low blends even though they can be quite different (e.g., see results for vehicle A and B on E10 in Supporting Information Table S6). A larger number of repeated tests could help to reduce the uncertainty range of fuel efficiency. Small deviations of ethanol blending level contributed very little to overall uncertainty in all cases. 4.1.5. Revisiting Values Cited in Previous LCAs. It was acknowledged long ago that ethanol’s effect on VEE could greatly influence its energy balance, though very limited vehicle data were available at that time for in-depth analysis.91 Several recent deterministic LCAs8−11 have attempted to incorporate the VEE effects of ethanol (see Supporting Information Table S7). However, they have relied on rather differing results from a limited number of sources and could potentially be misleading. For instance, two of these studies that examined both E10 and E85 argue that a given amount of ethanol would achieve much larger GHG reductions if used in the form of E10 rather than E85.8,11 Although their implied ESR values for E85 are close to the mean and median observed in our analysis, those for E10 (1.52 and 2.18) are much higher. Use of these values could therefore overestimate the benefits of ethanol in the form of E10, especially given the large uncertainty ranges for the ESR for low blends. The ESR values used for Brazilian vehicles10 are likely to be more realistic as these vehicles employ increased CR.92 4.2. Life-Cycle Impacts. We use the GREET model93 to assess the effect of ESR on the net GHG impact of substituting conventional gasoline with corn or switchgrass ethanol, represented by the life-cycle GHG emission changes resulted from the use of 1 MJ ethanol. Stochastic simulations were conducted using default input data in GREET (data for key parameters are presented in SI Table S8). The model was modified to allow incorporation of the uncertainties in ESR for low blends in regular vehicles and high blends in FFVs.

efficiency/CO2 emission measurements from repeated tests on the same vehicle (or model) and the same fuel over the same drive cycle also vary by a few percent. This could be caused by driver-to-driver variability, vehicle preparation and the setup, calibration, and control of the test equipments. We use results from two well-documented vehicle studies39,51 to illustrate the effects of uncertainties in key parameters such as LHV, ethanol blend level and fuel efficiency measurements on ESR calculations using Monte Carlo simulations89 which take as inputs distribution functions for these parameters. Study One51 tested 14 regular FI-engine vehicles of different models on gasoline, E10, E15, and E20, whereas Study Two39 tested several vehicles of two FI-engine FFV models on gasoline and E85. Moreover, Study Two conducted two rounds of tests with the Round2 tests done one year after Round1 (mileage accumulation). Fuel efficiency results for two vehicles on both gasoline and E10 in Study One (as they differ in the number of repeated tests on each fuel) and the two rounds of results for one FFV on gasoline and E85 in Study Two were used. The uncertain parameters examined and the assumptions of their probability distributions are shown in Supporting Information Table S5. Two scenarios are examined for gasoline LHV: the first scenario assumes uniform distributions with a ± 3% range and is intended to represent the case when LHV is not measured or reported; the second assumes normal distributions with standard deviation (S.D.) of 0.10 MJ/l and is intended to represent the case when the LHV is measured (95%-CI range of ∼1.2%90 of the measured value). Ethanol blending level is included to explore the effects of small deviations from the nominal values if it is not measured. While the uncertainty ranges for ESR as a result of these parameters examined here are large they are lower than those observed for different vehicles under various test conditions (see Supporting Information Table S6 and Section 4.1.2). Uncertainty due to gasoline LHV variations can greatly affect the calculation of ESR (Scenario 1) and its contribution to 5539

dx.doi.org/10.1021/es305209a | Environ. Sci. Technol. 2013, 47, 5535−5544

Environmental Science & Technology

Policy Analysis

Probabilistic distributions were fitted89 based on ESR values observed for FI vehicles on E20 or lower blends and FFVs on E85 and used as inputs in the model. Furthermore, uncertainties in LUC emissions for corn ethanol were considered following Boies et al.12 The results are plotted in Figure 3 as probability distribution functions. They indicate that the inclusion of ESR within the stochastic LCA model can greatly affect the net GHG impact of both corn and switchgrass ethanol, especially when ethanol is used in low blends. For example, incorporating ESR slightly reduces the modes of the distributions for high blends (from −31 to −34 gCO2e/MJ for corn ethanol and from −92 to −95 gCO2e/MJ for switchgrass ethanol) and increases the SD (from 10 to 11 gCO2e/MJ for corn ethanol and from 8 to 9 gCO2e/ MJ for switchgrass ethanol). The modes of the distributions reduce modestly when ESR is included for low blends (from −31 to −43 gCO2e/MJ for corn ethanol and from −92 to −104 gCO2e/MJ for switchgrass ethanol) while the SD increase substantially (from 10 to 31 gCO2e/MJ for corn ethanol and from 8 to 31 gCO2e/MJ for switchgrass ethanol). The effect of ESR on the uncertainty of life-cycle GHG change for corn ethanol used as low blends appears to be higher than that of LUC (SD increases from 10 to 24 gCO2e/MJ due to the inclusion of LUC emissions). This uncertainty increases further for corn ethanol used as low blends when both ESR and LUC are included (SD increases to 46 gCO2e/MJ). These results highlight that even though the net GHG impacts of ethanol used in low blends are likely to be modestly better than the previously estimated, they are also much more uncertain with a finite probability of being worse than previous estimates. Given these substantial effects of ESR on the net GHG impact of ethanol observed herein, we assert that it is important and feasible to reduce the uncertainties in ESR and to take ESR into account in future stochastic LCAs to capture these potential risks that have been largely overlooked. 4.3. Future Potentials. Vehicle tests data analyzed in this paper can only provide an overview of the efficiency effects of ethanol on past and current engine technologies. However, much research and development work is ongoing to better utilize the favorable properties of ethanol in order to increase future engine efficiency. Here we summarize the most promising technology pathways and discuss the barriers to their development. Flexible-fuel engines could better take advantage of the beneficial properties of ethanol/gasoline blends while maintaining compatibility with gasoline through better control of various operating parameters such as ignition and valve timing and through increased geometric CR with variable valve timing to reduce effective CR for gasoline or low blends.94−96 Although these techniques could notably increase the engine efficiency on ethanol/gasoline blends over gasoline, the torque and power output will also be increased.94−96 Therefore, if a flexible-fuel engine employs the same displacement as a regular gasoline engine, its efficiency gains on ethanol/gasoline blends would be partly offset by better performance. For instance, a production FFV optimized on E85 produces 20% higher power on E85 than on gasoline while offering 4−7% higher VEE, only slightly higher than the U.S. FFV fleet average of 3%.47 The simulations on standard drive cycles for another flexible-fuel engine being developed (claimed to be optimized on any blends from E0 to E85) show that it still suffers 24−27% lower MPG on E85 than on gasoline,94 implying again 4−7% higher VEE. U.S. national average prices of E85 in the last five years

are 18−38% higher than those of gasoline per unit energy97 despite considerable government support,98 and only ∼5% of the 10 million in-use FFVs were believed to actually use E85 with less than 1% of U.S. fuel ethanol use in the form of E85.99 If current pricing continues, potential users of these “optimised” FFVs would still have little economic incentives to choose E85 over gasoline. Medium blends such as E20 might be a better choice as they enable most of the performance and engine efficiency gains of a high blend96 while potentially offering higher ESR (see Supporting Information Figure S2). In addition, several studies have shown that some existing FFVs might achieve higher VEE with medium blends than with high blends.46,69,80 On the other hand, if a flexible-fuel engine is downsized to match its performance on ethanol/gasoline blends with that of a regular gasoline engine, this could potentially offer much higher energy efficiency. However, its performance on gasoline will be degraded, thus losing the flexibility to some extent. Variable geometric CR engines also have great potential for realizing the antiknock benefits of different ethanol/gasoline blends, especially for FFV. However, their commercial prospect remains challenging due to their incompatibility with major components in current production engines.100 Dedicated ethanol (or E85) engines have the promise to improve efficiency substantially over comparable regular gasoline engines through direct-injection, increased geometric CR, variable valve timing, exhaust gas recirculation, aggressive turbocharging, and downsizing and/or downspeeding.101−107 It has also been demonstrated that lean boosted engines with ethanol can produce efficiency higher than that offered by diesel engines.108 The increased engine efficiency at high load for ethanol/ gasoline blends (as a result of their antiknock properties) could also be explored in hybrid-electric powertrains (with a much narrower and optimized engine operating regime),105 as shown in a simulation study on a series plug-in hybrid vehicle.109 Some innovative engine concepts and designs such as the Direct Injection Ethanol Boosted Gasoline Engine (DIEBGE) show great promise of improving the efficiency of gasoline use through the leverage effect of ethanol and thereby maximize the potential benefits from a limited amount of ethanol. The DIEBGE concept is to use a downsized port-injection gasoline engine with aggressive turbocharging, increased CR and ethanol as a knock suppressant to match the performance of a much larger engine, resulting in an increase in engine efficiency of 30% or more.110 During this type of operation, ethanol is stored in a separate fuel tank and is directly injected only when knock suppression is required. This allows the knock suppression benefit of the ethanol to be utilized much more efficiently than a regular engine or a flexible-fuel engine, where the ethanol and gasoline are premixed by a fixed ratio prior to use. Ford is developing an early prototype engine based on the DIEBGE concept and their initial engine efficiency tests and vehicle simulations suggest that 1 L of E85 could replace 5 L of gasoline on the EPA urban/highway combined cycle.111 This would translate into an astonishing ESR of 9. Others are also working on this promising dual-injection concept.112 4.4. Policy Implications. Current biofuel policies have focused on maximizing ethanol production and consumption (through, e.g., production subsidies, blending mandates, and FFV promotion) and reducing GHG intensity while ignoring the potential to improve vehicle efficiency through ethanol use. Consequently vehicle manufacturers have to date mainly been 5540

dx.doi.org/10.1021/es305209a | Environ. Sci. Technol. 2013, 47, 5535−5544

Environmental Science & Technology

Policy Analysis

optimize the fuel-energy nexus and consumers to make informed decisions. Furthermore, the actual substitution effects between different fuels and energy carriers, which have been broadly neglected, should be clearly communicated based on available evidence. 4.5. Suggested Future Studies. Deriving an ESR that is representative of the substitution effects between ethanol and gasoline across a very wide range of engine technologies and ethanol blending levels is extremely challenging, if not impossible, based on existing experimental results. Future studies should be designed to systematically explore the likely impact on ESR for parameters including test conditions such as ambient temperature and drive cycles, engine designs such as CR, knock-sensing, variable valve timing and turbocharging, and fuel properties such as blending levels and octane ratings. In particular, factors that can lead to vehicle efficiency decrease with ethanol/gasoline blends need to be identified. Fuel properties should always be measured rather than using typical values and more repeated vehicle tests should be conducted when possible. This could help to better quantify ESR and reduce the associated uncertainties. The performance of medium blends in existing FFVs merits further study given that blender pump dispensers able to offer a range of medium blends are emerging.118 The trade-offs between performance and energy efficiency for various ethanol/ gasoline blends when optimizing an FFV also bring about the question as to whether it is appropriate to compare the efficiency of an FFV on ethanol/gasoline blends with that on gasoline (lower power/performance) rather than the efficiency of an equivalent regular gasoline vehicle. Furthermore, as most of the FFVs in the U.S. are running on gasoline rather than E85, efforts should be made to compare the efficiencies of FFVs running on gasoline with those of equivalent regular gasoline vehicles. In addition to ethanol, other promising alternative fuels such as biodiesel,119 Fischer−Tropsch liquids,6,120 and methanol120 that have different thermo-chemical properties compared with their conventional counterparts will need to be assessed with respect to their effects on engine efficiency and the implications on their net GHG impacts.

(passively) producing vehicles that are ethanol-tolerant rather than ethanol-efficient. Future policies should be designed to promote appropriate engine technologies that correspond to the quantity of ethanol available in order to improve vehicle efficiency and perhaps more importantly, to maximize the ESR. This could potentially increase the amount of gasoline displaced by a given amount of ethanol and improve the cost-competitiveness of ethanol compared with gasoline. Larger GHG emission reductions will also be realized regardless of the feedstock used. This is highly beneficial as relatively highcarbon ethanol such as corn ethanol are likely to be used in large amounts in the foreseeable future and the development in low-carbon ethanol has been slower than expected.113 In general, the potential to increase ESR is much more limited for high blends than that for low blends mainly due to the diminishing leveraging effect of the ethanol component on the gasoline component. However, tailored strategies are needed for different regions depending on the current and future level of their ethanol use and changes to the vehicle fleets and fuel infrastructure may be necessary. For instance, ethanol in the U.S. today represents ∼10% by volume of SI-engine vehicle fuel use and this is projected to be in the range of 10− 29% by 2035 depending on future oil prices and policies.114 This level of ethanol use and the fact that less than 2% of the fuel stations currently offer E85115 indicate a rather limited scope for dedicated ethanol vehicles. If the limited amount of ethanol is used in the form of low to medium blends, especially when used to increase the octane rating of fuels, efficiency can be increased slightly for in-use vehicles and modestly for future vehicles through corresponding increases in CR and/or further turbocharging and downsizing.114 This could potentially lead to much higher ESRs and thus higher GHG emission reductions for the same quantity of ethanol than if the ethanol is used in the form of high blends in FFVs. Engine efficiency is expected to increase by 2−5% even if only the effect of increased CR enabled by ethanol blending levels in the range of E10−E30 is considered.114 The ESR would be 1.24−1.30 assuming engine efficiency is increased by 2% for E10 and 5% for E30 while it would be 1.26 even for a dedicated E85 engine that increases efficiency by 20% (see Supporting Information Figure S1). Currently, however, the potential to increase fuel octane rating in the U.S. through the use of ethanol is not realized as the fuel industry has adopted practices to reduce the octane ratings of the gasoline blendstocks with the final fuel blends meeting minimum octane ratings.114 Whereas for Brazil, dedicated ethanol vehicle could be feasible and beneficial given the share of ethanol in its transport fuel pool, offering potentials to increase ESR compared with FFVs running on ethanol. Our analysis also highlights that comparing vehicle efficiency for different fuels is far from straightforward as fuel efficiency is usually reported in volumetric units such as MPG and km/l and volumetric energy content of fuels can vary considerably (and usually not measured). Recent studies have shown that even different volumetric units of fuel economy can be quite misleading116 and better public understanding in energy use and savings is required to realize the benefits of energy strategies.117 A common metric for vehicle efficiency expressed in energy and distance such as km/MJ or MJ/km could therefore be highly beneficial for both policy makers and consumers, especially with the increasingly diversifying types of fuels and vehicles. Vehicle energy efficiency on all commercially available fuels (particularly for FFVs), when made available, could help policy makers to develop appropriate policies that

■

ASSOCIATED CONTENT

S Supporting Information *

Additional information as described in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +44 1326 259485; e-mail: [email protected]. net. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Roy Crookes of Queen Mary, University of London, John Heywood of MIT and Dave Richardson of Jaguar Land Rover for discussions and the following individuals and institutions for supplying additional data in their studies: Keith Knoll of NREL, Brian Hilton of Rochester Institute of Technology, Australian Government Department of Sustainability, Environment, Water, Population and Communities, Luc Pelkmans of VITO, Georgios Karavalakis and Thomas Durbin 5541

dx.doi.org/10.1021/es305209a | Environ. Sci. Technol. 2013, 47, 5535−5544

Environmental Science & Technology

Policy Analysis

(21) Szybist, J. P.; Chakravathy, K.; Daw, C. S. Analysis of the impact of selected fuel thermochemical properties on internal combustion engine efficiency. Energy Fuels 2012, 26, 2798−2810. (22) Al-Hasan, M. Effect of ethanol−unleaded gasoline blends on engine performance and exhaust emission. Energy Convers. Manage. 2003, 44, 1547−1561. (23) Bayraktar, H. Experimental and theoretical investigation of using gasoline−ethanol blends in spark-ignition engines. Renewable Energy 2005, 30, 1733−1747. (24) Al-Baghdadi, M. A. R. S. Measurement and prediction study of the effect of ethanol blending on the performance and pollutants emission of a four-stroke spark ignition engine. Proc. Inst. Mech. Eng., Part D 2008, 222, 859−873. (25) Al-Farayedhi, A. A.; Al-Dawood, A. M.; Gandhidasan, P. Experimental investigation of SI Engine Performance Using Oxygenated Fuel. J. Eng. Gas Turbines Power 2004, 126, 178−191. (26) Najafi, G.; Ghobadian, B.; Tavakoli, T.; Buttsworth, D.; Yusaf, T.; Faizollahnejad, M. Performance and exhaust emissions of a gasoline engine with ethanol blended gasoline fuels using artificial neural network. Appl. Energy 2009, 86, 630−639. (27) Schifter, I.; Diaz, L.; Rodriguez, R.; Gómez, J. P.; Gonzalez, U. Combustion and emissions behavior for ethanol−gasoline blends in a single cylinder engine. Fuel 2011, 90, 3586−3592. (28) Caton, P. A.; Hamilton, L. J.; Cowart, J. S. An Experimental and Modeling Investigation into the Comparative Knock and Performance Characteristics of E85, Gasohol [E10] and Regular Unleaded Gasoline [87 (R+M)/2], SAE Technical Paper 2007-01-0473; SAE: West Conshohocken, PA, 2007; DOI: 10.4271/2007-01-0473. (29) Kar, K.; Cheng, W.; Ishii, K. Effects of ethanol content on gasohol PFI engine wide-open-throttle operation. SAE Int. J. Fuels Lubr. 2009, 2, 895−901. (30) Varde, K.; Jones, A.; Knutsen, A.; Mertz, D.; Yu, P. Exhaust emissions and energy release rates from a controlled spark ignition engine using ethanol blends. Proc. Inst. Mech. Eng., Part D 2007, 221, 933−941. (31) Wallner, T.; Miers, S. A. Combustion Behavior of Gasoline and Gasoline/Ethanol Blends in a Modern Direct-Injection 4-Cylinder Engine, SAE Technical Paper 2008-01-0077; SAE: West Conshohocken, PA, 2008; DOI: 10.4271/2008-01-0077. (32) Wallner, T.; Miers, S. A.; McConnell, S. A comparison of ethanol and butanol as oxygenates using a direct-injection, sparkignition engine. J. Eng. Gas Turbines Power 2009, 131, Article Number: 032802. (33) Nakama, K.; Kusaka, J.; Daisho, Y. Effect of ethanol on knock in spark ignition gasoline engines. SAE Int. J. Eng. 2009, 1, 1366−1380. (34) ChevronTexaco. Motor Gasolines Technical Review; Chevron Products Company: San Ramon, 2004. (35) Lawrence, R. Gasohol Test Program; Technology Assessment and Evaluation Branch, U.S. Environmental Protection Agency, Motor Vehicle Emission Laboratory, 1978. (36) Kelly, K. J.; Bailey, B. K.; Coburn, T.; Clark, W.; Lissiuk, P. Federal Test Procedure Emissions Test Results from Ethanol Variable-Fuel Vehicle Chevrolet Luminas, SAE Technical Paper 961092; SAE: West Conshohocken, PA, 1996; DOI: 10.4271/961092. (37) Apace Research Ltd. Intensive Field Trial of Ethanol/Petrol Blend in Vehicles; ERDC Project No. 2511; Apace Research Ltd., 1998. (38) Knapp, K.; Stump, F.; Tejada, S. The effect of ethanol fuel on the emissions of vehicles over a wide range of temperatures. J. Air Waste Manage. Assoc. 1998, 48, 646−653. (39) Kelly, K.; Eudy, L.; Coburn, T. Light-Duty Alternative Fuel Vehicles: Federal Test Procedure Emissions Results; National Renewable Energy Laboratory, 1999. (40) Ragazzi, R.; Nelson, K. The Impact of a 10% Ethanol Blended Fuel on the Exhaust Emissions for Tier 0 and Tier 1 Light Duty Gasoline Vehicles at 35 Degrees F; Colorado Department of Public Health and Environment: CO, 1999. (41) Reading, A.; Norris, J.; Feest, E.; Payne, E. Ethanol Emissions Testing; AEA Technology, 2002.

of University of California, JEC-Joint Research CentreEUCAR-CONCAWE collaboration, Toshiyuki Hirose of Japan Petroleum Energy Center, Mahmoud Yassine of Chrysler Group, and Phil Stansfield of MAHLE Powertrain.

■

REFERENCES

(1) REN21. Renewables 2012 Global Status Report; REN21 Secretariat: Paris, 2012. (2) Farrell, A. E.; Plevin, R. J.; Turner, B. T.; Jones, A. D.; O’Hare, M.; Kammen, D. M. Ethanol can contribute to energy and environmental goals. Science 2006, 311, 506−508. (3) Hill, J.; Nelson, E.; Tilman, D.; Polasky, S.; Tiffany, D. Environmental, economic, and energetic costs and benefits of biodiesel and ethanol biofuels. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 11206− 11210. (4) Fargione, J.; Hill, J.; Tilman, D.; Polasky, S.; Hawthorne, P. Land clearing and the biofuel carbon debt. Science 2008, 319, 1235−1238. (5) Searchinger, T.; Heimlich, R.; Houghton, R. A.; Dong, F.; Elobeid, A.; Fabiosa, J.; Tokgoz, S.; Hayes, D.; Yu, T.-H. Use of U.S. croplands for biofuels increases greenhouse gases through emissions from land-use change. Science 2008, 319, 1238−1240. (6) Yan, X.; Inderwildi, O. R.; King, D. A. Biofuels and synthetic fuels in the U.S. and China: A review of well-to-wheel energy use and greenhouse gas emissions with the impact of land-use change. Energy Environ. Sci. 2010, 3, 190. (7) Yan, X.; Tan, D. K. Y.; Inderwildi, O. R.; Smith, J. a. C.; King, D. A. Life cycle energy and greenhouse gas analysis for agave-derived bioethanol. Energy Environ. Sci. 2011, 4, 3110−3121. (8) Kim, S.; Dale, B. Ethanol fuels: E10 or E85Life cycle perspectives. Int. J. Life Cycle Assess. 2006, 11, 117−121. (9) Nguyen, T. L. T.; Gheewala, S. H.; Garivait, S. Energy balance and GHG-abatement cost of cassava utilization for fuel ethanol in Thailand. Energy Policy 2007, 35, 4585−4596. (10) Macedo, I. C.; Seabra, J. E. A.; Silva, J. E. A. R. Green house gases emissions in the production and use of ethanol from sugarcane in Brazil: The 2005/2006 averages and a prediction for 2020. Biomass Bioenergy 2008, 32, 582−595. (11) Gnansounou, E.; Dauriat, A.; Villegas, J.; Panichelli, L. Life cycle assessment of biofuels: Energy and greenhouse gas balances. Bioresour. Technol. 2009, 100, 4919−4930. (12) Boies, A. M.; McFarlane, D.; Taff, S.; Watts, W. F.; Kittelson, D. B. Implications of local lifecycle analyses and low carbon fuel standard design on gasohol transportation fuels. Energy Policy 2011, 39, 7191− 7201. (13) Yan, X.; Boies, A. M. Quantifying the uncertainties in life cycle greenhouse gas emissions for UK wheat ethanol. Environ. Res. Lett. 2013, 8, 015024. (14) Mullins, K. A.; Griffin, W. M.; Matthews, H. S. Policy Implications of uncertainty in modeled life-cycle greenhouse gas emissions of biofuels. Environ. Sci. Technol. 2011, 45, 132−138. (15) Spatari, S.; MacLean, H. L. Characterizing model uncertainties in the life cycle of lignocellulose-based ethanol fuels. Environ. Sci. Technol. 2010, 44, 8773−8780. (16) Plevin, R. J.; O’Hare, M.; Jones, A. D.; Torn, M. S.; Gibbs, H. K. Greenhouse gas emissions from biofuels’ indirect land use change are uncertain but may be much greater than previously estimated. Environ. Sci. Technol. 2010, 44, 8015−8021. (17) Heywood, J. B. Internal Combustion Engine Fundamentals; McGraw-Hill: New York, 1988. (18) Bailey, B. K. Performance of ethanol as a transportation fuel. In Handbook on Bioethanol: Production and Utilization; Taylor & Francis: Washington DC, 1996. (19) Larsen, U.; Johansen, T.; Schramm, J. Ethanol as a Fuel for Road Transportation; Technical University of Denmark, 2009. (20) Yanowitz, J.; Christensen, E.; McCormick, R. L. Utilization of Renewable Oxygenates As Gasoline Blending Components; National Renewable Energy Laboratory, 2011. 5542

dx.doi.org/10.1021/es305209a | Environ. Sci. Technol. 2013, 47, 5535−5544

Environmental Science & Technology

Policy Analysis

(42) Orbital Engine Company. Market Barriers to the Uptake of Biofuels Study: A Testing Based Assessment to Determine Impacts of a 20% Ethanol Gasoline Fuel Blend on the Australian Passenger Vehicle Fleet; Orbital Engine Company, 2003. (43) Orbital Engine Company. Market Barriers to the Uptake of Biofuels Study: Testing Gasoline containing 20% Ethanol (E20)Phase 2B Final Report; Orbital Engine Company, 2004. (44) Durbin, T. D.; Miller, J. W.; Younglove, T.; Huai, T.; Cocker, K. Effects of Ethanol and Volatility Parameters on Exhaust Emissions; Coordinating Research Council: Alpharetta, 2006. (45) JECJoint Research Centre-EUCAR-CONCAWE collaboration. Joint EUCAR/JRC/CONCAWE Study on: Effects of Gasoline Vapour Pressure and Ethanol Content on Evaporative Emissions from Modern Cars; Institute for Environment and Sustainability, 2007. (46) Shockey, R. E.; Aulich, T. R.; Jones, B.; Mead, G.; Steevens, P. Optimal Ethanol Blend-Level Investigation; Energy & Environmental Research Center, University of North Dakota: Grand Forks, 2007. (47) West, B. H.; López, A. J.; Theiss, T. J.; Graves, R. L.; Storey, J. M.; Lewis, S. A. Fuel Economy and Emissions of the Ethanol-Optimized Saab 9−5 Biopower, SAE Technical Paper 2007-01-3994; SAE: West Conshohocken, PA, 2007; DOI: 10.4271/2007-01-3994. (48) Orbital. Evaluating the Health Impacts of Ethanol Blend Petrol, eport to Department of the Environment, Water, Heritage and the Arts; Orbital, 2008. (49) Tóth, D.; Cachón, L.; Pucher, E.; Raetzsch, T.; Weissenberger, D. Real World and Chassis Dynamometer Emission Measurement of a Turbocharged Gasoline Vehicle with increased Bio Fuel Blend, SAE Technical Paper 2008-01-1768; SAE: West Conshohocken, PA, 2008; DOI: 10.4271/2008-01-1768. (50) Graham, L. A.; Belisle, S. L.; Baas, C.-L. Emissions from light duty gasoline vehicles operating on low blend ethanol gasoline and E85. Atmos. Environ. 2008, 42, 4498−4516. (51) Knoll, K.; West, B.; Clark, W.; Graves, R.; Orban, J.; Przesmitzki, S.; Theiss, T. Effects of Intermediate Ethanol Blends on Legacy Vehicles and Small Non-Road Engines, Report 1Updated; National Renewable Energy Laboratory, 2009. (52) Hilton, B.; Duddy, B. The effect of E20 ethanol fuel on vehicle emissions. Proc. Inst. Mech. Eng., Part D 2009, 223, 1577−1586. (53) Nobuhiro, O.; Yukihiro, T.; Keiichi, K.; Hiroshi, T.; Masashi, I.; Tadahide, S.; Hideaki, A.; Atsushi, K.; Hideki, K.; Makoto, H.; Tatsuya, M.; Osamu, N.; Toshiyuki, H.; Mamoru, M. Impact of Biogasoline on Peformance of Vehicles in Japanese Market; Japan Petroleum Energy Center, 2010. (54) Storey, J. M.; Barone, T.; Norman, K.; Lewis, S. Ethanol blend effects on direct injection spark-ignition gasoline vehicle particulate matter emissions. SAE Int. J. Fuels Lubr. 2010, 3, 650−659. (55) Pelkmans, L.; Lenaers, G.; Bruyninx, J.; Scheepers, K.; De Vlieger, I. Impact of biofuel blends on the emissions of modern vehicles. Proc. Inst. Mech. Eng., Part D 2011, 225, 1−17. (56) Aakko-Saksa, P.; Rantanen-Kolehmainen, L.; Koponen, P.; Engman, A.; Kihlman, J. Biogasoline Options - Possibilities for Achieving High Bio-share and Compatibility with Conventional Cars. SAE Int. J. Fuels Lubr. 2011, 4, 298−317. (57) Karavalakis, G.; Durbin, T. D.; Shrivastava, M.; Zheng, Z.; Villela, M.; Jung, H. Impacts of ethanol fuel level on emissions of regulated and unregulated pollutants from a fleet of gasoline light-duty vehicles. Fuel 2012, 93, 549−558. (58) Stansfield, P. A.; Bisordi, A.; OudeNijeweme, D.; Williams, J.; Gold, M.; Ali, R. The Performance of a modern vehicle on a variety of alcohol-gasoline fuel blends. SAE Int. J. Fuels Lubr. 2012, 5, 813−822. (59) Yassine, M. K.; La Pan, M. Impact of Ethanol Fuels on Regulated Tailpipe Emissions, SAE Technical Paper 2012-01-0872; SAE: West Conshohocken, PA, 2012; DOI: 10.4271/2012-01-0872. (60) U.S. Environmental Protection Agency Fuel Economy Guide Data 2002−2013. http://www.epa.gov/otaq/fedata.htm, 2012. (61) Stump, F. D.; Knapp, K. T.; Ray, W. D. Influence of ethanolblended fuels on the emissions from three pre-1985 light-duty passenger vehicles. J. Air Waste Manage. Assoc. 1996, 46, 1149−1161.

(62) Chandler, K.; Whalen, M.; Westhoven, J. Final Results From The State Of Ohio Ethanol-Fueled Light-Duty Fleet Deployment Project, SAE Technical Paper 982531; SAE: West Conshohocken, PA, 1998; DOI: 10.4271/982531. (63) Bonnema, G.; Guse, G.; Senecal, N.; Gupta, R.; Jones, B.; Ready, K. L. Use of Mid-Range Ethanol-Gasoline Blends in Unmodified Passenger Cars and Light Duty Trucks; Minnesota Center for Automotive Research, Minnesota State University, 1999. (64) Schifter, I.; Vera, M.; Díaz, L.; Guzmán, E.; Ramos, F.; LópezSalinas, E. Environmental implications on the oxygenation of gasoline with ethanol in the metropolitan area of Mexico City. Environ. Sci. Technol. 2001, 35, 1893−1901. (65) Tantithumpoosit, W. History, roadmap and success of using ethanol blended gasoline in Thailand. In 2nd Asian Petroleum Technology Symposium; 2004. (66) American Coalition for Ethanol. Fuel Economy Study: Comparing Performance and Cost of Various Ethanol Blends and Standard Unleaded Gasoline; American Coalition for Ethanol, 2005. (67) De Serves, C. Emissions from Flexible Fuel Vehicles with Different Ethanol Blends; AVL MTC AB, 2005. (68) Subramanian, M.; Setia, A. K.; Kanal, P. C.; Pal, N. K.; Nandi, S.; Malhotra, R. K. Effect of Alcohol Blended Fuels on the Emissions and Field Performance of Two-Stroke and Four-Stroke Engine Powered Two Wheelers, SAE Technical Paper 2005-26-034; SAE: West Conshohocken, PA, 2005; DOI: 10.4271/2005-26-034. (69) Vicentini, P. C.; Kronberger, S. Rating the Performance of Brazilian Flex Fuel Vehicles, SAE Technical Paper 2005-01-2206; SAE: West Conshohocken, PA, 2005; DOI: 10.4271/2005-01-2206. (70) Karlsson, H. L. Emissions from Conventional Gasoline Vehicles Driven with Ethanol Blend Fuels; AVL MTC AB: Haninge, 2006. (71) Jaroonjitsathian, S.; Akarapanjavit, N.; Sa-norh, S. S.; Chanchaona, S. Investigation of 2-Wheeler Performance, Emissions, Driveability and Durability: Effect of Ethanol-blended Gasoline, SAE Technical Paper 2007-01-2034; SAE: West Conshohocken, PA, 2007; DOI: 10.4271/2007-01-2034. (72) Karlsson, H.; Gåsste, J.; Åsman, P. Regulated and Non-regulated Emissions from Euro 4 Alternative Fuel Vehicles, SAE Technical Paper 2008-01-1770; SAE: West Conshohocken, PA, 2008; DOI: 10.4271/ 2008-01-1770. (73) Tseng, M.-C.; Lin, R.-R.; Liu, F.-L. Assessment on the Impacts of the 3% Ethanol Gasoline Fuel Blend on Passenger Cars and Motorcycles in Taiwan, SAE Technical Paper 2008-32-0019; SAE: West Conshohocken, PA, 2008; DOI: 10.4271/2008-32-0019. (74) Gogos, M.; Savvidis, D.; Triandafyllis, J. Study of the effects of ethanol use on a Ford Escort fitted with an old technology engine. SAE Int. J. Commer. Veh. 2009, 1, 254−259. (75) Martini, G.; Astorga, C.; Adam, T.; Farfaletti, A.; Manfredi, U.; Montero, L.; Krasenbrink, A.; Larsen, B.; Santi, G. D. Effect of Fuel Ethanol Content on Exhaust Emissions of a Flexible Fuel Vehicle; Institute for Environment and Sustainability, 2009. (76) Milford, H.; Loren, I.; Robert, W. Mid-level Ethanol Blend Study: Chassis Dynamometer Study of Flex Fuel Vehicles; University of Nebraska: Lincoln, NE, 2009. (77) Anoop, B.; Vashisth Ajay, K.; Gokula Krishnan, M. Investigation of Performance and Emissions with Low Level Ethanol-Gasoline Blends on a Passenger Car, SAE Technical Paper 2009-26-0034; SAE: West Conshohocken, PA, 2009; DOI: 10.4271/2009-26-0034. (78) Roayaei, E.; Taheri, K. Test run evaluation of a blend of fuelgrade ethanol and regular commercial gasoline: its effect on engine efficiency and exhaust gas composition. Clean Technol. Environ. Policy 2009, 11, 385−389. (79) Yao, Y.-C.; Tsai, J.-H.; Chiang, H.-L. Effects of ethanol-blended gasoline on air pollutant emissions from motorcycle. Sci. Total Environ. 2009, 407, 5257−5262. (80) De Villela, A. C. S.; Carvalho, R. N. Fuel Consumption Technical and Economical Study for Flexible Fuel Vehicles, SAE Technical Paper 2009-36-102; SAE: West Conshohocken, PA, 2009; DOI: 10.4271/ 2009-36-0102. 5543

dx.doi.org/10.1021/es305209a | Environ. Sci. Technol. 2013, 47, 5535−5544

Environmental Science & Technology

Policy Analysis

(81) Colnago, K.; Melo, T.; Fadel, F. Hybrid Electric Vehicles Technology Evolution and Experience with Brazilian Gasolines E25; SAE Technical Paper 2010-01-1442; SAE: West Conshohocken, PA, 2010; DOI: 10.4271/2010-01-1442. (82) Thummadetsak, T.; Tipdecho, C.; Wongjareonpanit, U.; Monnum, P. Thailand Fuel Performance and Emissions in Flex Fuel Vehicles, SAE Technical Paper 2010-01-2132; SAE: West Conshohocken, PA, 2010; DOI: 10.4271/2010-01-2132. (83) Schulz, M.; Clark, S.; Honary, L.; Conconi, C.; Dean, S. W. Vehicle emissions and fuel economy effects of 16% butanol and various ethanol blended fuels (E10, E20, and E85). J. ASTM Int. 2011, 8, Paper ID: JAI103068. (84) Bielaczyc, P.; Szczotka, A.; Woodburn, J. The Effect of Various Petrol-Ethanol Blends on Exhaust Emissions and Fuel Consumption of an Unmodified Light-Duty SI Vehicle, SAE Technical Paper 2011-24-0177; SAE: West Conshohocken, PA, 2011; DOI: 10.4271/2011-24-0177. (85) Yang, H.-H.; Liu, T.-C.; Chang, C.-F.; Lee, E. Effects of ethanolblended gasoline on emissions of regulated air pollutants and carbonyls from motorcycles. Appl. Energy 2012, 89, 281−286. (86) MSN Autos. http://home.autos.msn.com/. http://home.autos. msn.com/ (accessed June 5, 2012). (87) carsales.com.au. http://www.carsales.com.au/ http://www. carsales.com.au/ (accessed June 5, 2012). (88) The Car Specifications Site. http://www.carfolio.com/. http:// www.carfolio.com/ (accessed September 6, 2012). (89) Oracle Oracle Crystal Ball, 2012. http://www.oracle.com/us/ products/applications/crystalball/index.html. (90) ASTM D240-02 (Reapproved 2007) Standard Test Method for Heat of Combustion of Liquid Hydrocarbon Fuels by Bomb Calorimeter; ASTM International, 2007. (91) Chambers, R. S.; Herendeen, R. A.; Joyce, J. J.; Penner, P. S. Gasohol: Does it or doesn’t it produce positive net energy? Science 1979, 206, 789−795. (92) The Royal Society. Sustainable Biofuels: Prospects and Challenges; The Royal Society, 2008. (93) Wang, M. The Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (GREET) Model, Version 1_2011; Argonne National Laboratory, 2011. (94) Cruff, L.; Kaiser, M.; Krause, S.; Harris, R.; Krueger, U.; Williams, M. EBDI® - Application of a Fully Flexible High BMEP Downsized Spark Ignited Engine, SAE Technical Paper 2010-01-0587; SAE: West Conshohocken, PA, 2010; DOI: 10.4271/2010-01-0587. (95) Szybist, J.; Foster, M.; Moore, W. R.; Confer, K.; Youngquist, A.; Wagner, R. Investigation of Knock Limited Compression Ratio of Ethanol Gasoline Blends, SAE Technical Paper 2010-01-0619; SAE: West Conshohocken, PA, 2010; DOI: 10.4271/2010-01-0619. (96) Moore, W.; Foster, M.; Hoyer, K. Engine Efficiency Improvements Enabled by Ethanol Fuel Blends in a GDi VVA Flex Fuel Engine, SAE Technical Paper 2011-01-0900; SAE: West Conshohocken, PA, 2011; DOI: 10.4271/2011-01-0900. (97) U.S. Department of Energy Clean Cities Alternative Fuel Price Report (March 2007 - April 2012); http://www.afdc.energy.gov/ publications/search/keyword/?q= alternative%20fuel%20price%20report. (98) Koplow, D. State and federal subsidies to biofuels: Magnitude and options for redirection. Int. J. Biotechnol. 2009, 11, 92−126. (99) U.S. Energy Information Administration. Alternatives to Traditional Transportation Fuels, 2009; U.S. Energy Information Administration, 2011. (100) Shaik, A.; Moorthi, N. S. V.; Rudramoorthy, R. Variable compression ratio engine: a future power plant for automobiles - an overview. Proc. Inst. Mech. Eng., Part D 2007, 221, 1159−1168. (101) Boretti, A. Towards 40% efficiency with BMEP exceeding 30 bar in directly injected, turbocharged, spark ignition ethanol engines. Energy Convers. Manag. 2012, 57, 154−166. (102) Gardiner, D. P.; Mallory, R. W.; Pucher, G. R.; Todesco, M. K.; Bardon, M. F.; Markel, T. J.; Ohi, J. M. Improving the Fuel Efficiency of Light-Duty Ethanol Vehicles - An Engine Dynamometer Study of

Dedicated Engine Strategies, SAE Technical Paper 1999-01-3568; SAE: West Conshohocken, PA, 1999; DOI: 10.4271/1999-01-3568. (103) Brewster, S. Initial Development of a Turbo-charged Direct Injection E100 Combustion System, SAE Technical Paper 2007-01-3625; SAE: West Conshohocken, PA, 2007; DOI: 10.4271/2007-01-3625. (104) Brusstar, M.; Stuhldreher, M.; Swain, D.; Pidgeon, W. High Efficiency and Low Emissions from a Port-Injected Engine with Neat Alcohol Fuels, SAE Technical Paper 2002-01-2743; SAE: West Conshohocken, PA, 2002; DOI: 10.4271/2002-01-2743. (105) Brusstar, M. J.; Gray, C. L. High Efficiency with Future Alcohol Fuels in a Stoichiometric Medium Duty Spark Ignition Engine, SAE Technical Paper 2007-01-3993; SAE: West Conshohocken, PA, 2007; DOI: 10.4271/2007-01-3993. (106) Cairns, A.; Stansfield, P.; Fraser, N.; Blaxill, H.; Gold, M.; Rogerson, J.; Goodfellow, C. A study of gasoline-alcohol blended fuels in an advanced turbocharged DISI engine. SAE Int. J. Fuels Lubr. 2009, 2, 41−57. (107) Thewes, M.; M\axther, M.; Brassat, A.; Pischinger, S.; Sehr, A. Analysis of the effect of bio-fuels on the combustion in a downsized DISI engine. SAE Int. J. Fuels Lubr. 2012, 5, 274−288. (108) Nakata, K.; Sasaki, N.; Ota, A.; Kawatake, K. The effect of fuel properties on thermal efficiency of advanced spark-ignition engines. Int. J. Eng. Res. 2011, 12, 274−281. (109) Shidore, N.; Ickes, A.; Wallner, T.; Rousseau, A.; Ehsani, M. Evaluation of ethanol blends for PHEVs using engine-in-the-loop. In Vehicle Power and Propulsion Conference (VPPC), 2011 IEEE, Chicago, 2011. (110) Cohn, D. R.; Bromberg, L.; Heywood, J. B. Direct Injection Ethanol Boosted Gasoline Engines: Biofuel Leveraging for Cost Effective Reduction of Oil Dependencies and CO2 Emissions; Massachusetts Institute of Technology, 2005. (111) Stein, R. A.; House, C. J.; Leone, T. G. Optimal use of E85 in a turbocharged direct injection engine. SAE Int. J. Fuels Lubr. 2009, 2, 670−682. (112) Wu, X.; Daniel, R.; Tian, G.; Xu, H.; Huang, Z.; Richardson, D. Dual-injection: The flexible, bi-fuel concept for spark-ignition engines fuelled with various gasoline and biofuel blends. Appl. Energy 2011, 88, 2305−2314. (113) Service, R. F. Is There a Road Ahead For Cellulosic Ethanol? Science 2010, 329, 784−785. (114) Anderson, J. E.; DiCicco, D. M.; Ginder, J. M.; Kramer, U.; Leone, T. G.; Raney-Pablo, H. E.; Wallington, T. J. High octane number ethanol−gasoline blends: Quantifying the potential benefits in the United States. Fuel 2012, 97, 585−594. (115) McAulay, J.; Heywood, J. B. Coordinated strategies for ethanol and flex fuel vehicle deployment: A quantitative assessment of the feasibility of biofuel targets. SAE Int. J. Fuels Lubr. 2010, 3, 303−312. (116) Larrick, R. P.; Soll, J. B. EconomicsThe MPG illusion. Science 2008, 320, 1593−1594. (117) Attari, S. Z.; DeKay, M. L.; Davidson, C. I.; De Bruin, W. B. Public perceptions of energy consumption and savings. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 16054−16059. (118) U.S. Department of Energy Blender Pump Dispensers. http:// www.afdc.energy.gov/afdc/technology_bulletin_0210.html. (119) Shirvani, T.; Yan, X.; Inderwildi, O. R.; Edwards, P. P.; King, D. A. Life cycle energy and greenhouse gas analysis for algae-derived biodiesel. Energy Environ. Sci. 2011, 4, 3773−3778. (120) Yan, X.; Crookes, R. J. Energy demand and emissions from road transportation vehicles in China. Prog. Energy Combust. Sci. 2010, 36, 651−676.

5544

dx.doi.org/10.1021/es305209a | Environ. Sci. Technol. 2013, 47, 5535−5544