E85 / FFV cars
- Flexible-fuel vehicles (FFV)
- Legislation, standards, and typical fuel properties
- Evaporative emissions, CO, and HC
- Aldehydes and ethanol
- Methane, 1,3-butadiene, benzene, and toluene
- Nitrogen oxides and ammonia
- Ozone forming potential
- Particles, PAH, and mutagenicity
- Intermediate blends (E30-E60)
High-oxygen-containing fuels, can be used in special flexible-fuel vehicles (FFV). Fuel containing ethanol up to 85% (E85) is used, for example, in Brazil, North America and in many European countries. In Brazil, FFVs are also designed for the use of hydrous E100 fuel. In the US, so called P-Series fuel consisting of butane, pentanes, ethanol and the biomass-derived co-solvent methyltetrahydrofuran (MTHF) is permitted for use in FFVs. From the beginning, FFVs were designed for 85% methanol blend (M85), which is used today for example in China.
FFVs are basically spark-ignition gasoline cars with some modifications. For example, all materials in are compatible with ethanol, which is more aggressive towards materials than gasoline. Due to E85 fuel’s low heating value, fuel injectors in FFVs are designed for higher fuel flows than in conventional gasoline cars leading to higher volumetric fuel consumption. Feedback control in FFVs adjusts fuel delivery and ignition timing. Ethanol’s higher octane rating would enable an increased compression ratio to achieve better energy efficiency. However, even modern FFVs still represent a compromise as compared to dedicated ethanol cars.
The ignition of ethanol is poor. When using E85 fuel, excess fuel is injected during cold starts to achieve performance similar to gasoline cars. Therefore exhaust emissions tend to be high until the three-way catalyst warms up (Lupescu 2009). Improved engine- and emissions-control technology is expected to reduce the exhaust emissions of FFVs in cold starts. Catalyzed hydrocarbon traps have been developed to store organic gases in cold starts until they can be removed when the TWC warms up (Lupescu 2009). Also intake port heating to reduce non-methane organic gas emissions has been studied (Chiba et al. 2010) and heated fuel injectors (Kabasin et al. 2009). All in all, the development of FFV cars is continuing in many areas.
Volumetric fuel consumption is higher for E85 fuel than for gasoline. The manufacturer’s figures for one FFV car equates to 33% higher volumetric fuel consumption for E85 than for gasoline, even though energy consumption as MJ/km is lower for E85 than for gasoline: 6.7 l (220 MJ) vs 8.9 l (205 MJ) per 100 km.
FFVs could in principle use many type of oxygenates, but E85 is the most commonly used today. Properties of E85 are close to ethanol properties. Ignition of ethanol as such is poor and therefore at least 15 vol-% of gasoline is needed in the E85 blend, and even more in the winter-quality fuel. The low energy content of ethanol leads to high volumetric fuel consumption. The octane numbers of E85 are higher than those of gasoline. E85 is not as sensitive towards water as low level ethanol/gasoline blends, however, phase separation may occur particularly with intermediate blends in the case of mixed refueling of E85 and gasoline and at low temperatures.
A number of issues must be taken into account when infrastructure and safety aspects are considered for E85 fuel. Special vehicles, materials and distribution systems are needed. Ethanol is flammable over a wide temperature range, which requires special safety measures. Guidance on materials, electrical conductivity, safety issues and other E85 aspects are presented in the E85 Handbook (2013) by the U.S. Department of Energy. These aspects are discussed also in Chapter ethanol properties.
In the U.S., E85 is specified by ASTM D5798 “Ethanol Fuel Blends for Flexible-Fuel Automotive Spark-Ignition Engines”. The properties of ethanol for E85 blending should meet ASTM D4806. Hydrocarbon blendstock may be unleaded gasoline, natural gasoline or other hydrocarbons, which meet the requirements given by ASTM D5798.
In Europe, E85 is specified by CEN/TS 15293 “Automotive fuels – Ethanol (E85) automotive fuel – Requirements and test methods” (Table 1). The EN 15293 defines four volatility classes. The ethanol content in Class d (winter grade) is 50 - 85%, and in Class b (summer grade) 70 - 85%. Ethanol for blending should comply with EN 15376.
In Canada, E85 is specified by CAN/CGSB-3.512-2013 “Automotive ethanol fuel (E50-E85). &rdquo Four volatility classes are defined by this standard each with a specific 10% low-end design temperature for ethanol. The volatility class is based on the expected minimum temperature of use. Class I (summer) is greater than 5 degrees Celsius and Class IV (winter) is less than -20 degrees Celsius. The ethanol component shall meet the requirements of CAN/CGSB-3.516, Type 2 and the gasoline component shall meet the requirements of CAN/CGSB-3.5 or blendstock for oxygenate blending (BOB), that when blended with up to 10% by volume denatured fuel ethanol produces an oxygenated gasoline that complies with CAN/CGSB-3.511.
Examples of standards for E85 fuel in Europe, in the U.S and Canada are given in Table 1.
Table 1. Selected requirements for E85 fuel in Europe, the U.S and Canada. Complete standards are available from respective organizations.
Fuel evaporative emissions are lower for E85 than for gasoline due to the E85 fuel’s low vapor pressure, potential reduction being some 30% or more (Yanowitz and McCormic 2009, Westerholm 2008, CRFA 2003). Nearly all FFVs have onboard vapor recovery system (E85 Handbook 2013, Martini et al. 2012). Haskew and Liberty (2006) observed lower permeation emissions for E85 fuel tested in the flexible-fuel vehicle system than for the low ethanol-content fuel. This is also seen in Figure 1.
Figure 1. The effect of ethanol on permeation (Stahl et al. 1992, Kassel 2006).
Carbon monoxide (CO) and non-methane hydrocarbon emissions (NMHC) have been generally lower, or not significantly changed when E85 is compared with gasoline at normal test temperature (Yanowitz and McCormic 2009, Graham et al. 2008, West et al. 2007). However, at -7 °C higher CO and HC have been observed for E85 than for gasoline (De Serves 2005, Westerholm et al. 2008, Aakko and Nylund 2003).
The calculation methodology has a significant impact on the HC results with the E85 fuel. Firstly, the flame ionization detector (FID) detects all carbon-containing compounds, also oxygenates, and not only hydrocarbons (Sandström-Dahl et al. 2010). In the US, this is taken into account in calculation methods of non-methane hydrocarbons (NMHC = HCFID – 1.04 x CH4 – 0.66 x ROH) and non-methane organic gases (NMOG = ΣNMHC + ΣROH + ΣRHO), but not in the European emissions regulations. In European regulations, a higher exhaust gas density of 0.932 g/dm3 is used for for E85 (C1H2.74O0.385) than for gasoline (0.619 g/dm3 C1H1.85,), whereas in the US calculation method density of 0.619 g/dm3 is used for both fuels. An example of the effect of calculation methods on the HC results is given in Table 2 for an FFV using the E85 fuel. The results obtained by the European calculation method for E85 are close to the NMOG results obtained by the US calculation method.
Table 2. HC emissions from FFV car using E85 fuel with different calculation methods (Aakko-Saksa et al. 2011).
Chiba et al. (2010) measured ethanol, formaldehyde, and acetaldehyde emissions during engine cold starts (Figures 2 and 3). The sum of these constituents represents a major part of NMOG emissions with E85. The latent heat of vaporization of ethanol is higher than that of gasoline, leading to poor cold-startability and high organic gas emissions during cold starts. NMOG emissions increased by about 50% in cold starts when E85 fuel was compared with gasoline, and ethanol represented the highest share of NMOG. The evaporative characteristics of E85 fuel leads to the condensation of unburnt alcohol in the combustion chamber, and condensed alcohol is released in the cold starts (Chiba et al. 2010).
Figure 2. Ethanol, formaldehyde, acetaldehyde, and total hydrocarbons in cold starts with E85 fuel (Chiba et al. 2010).
Figure 3. The effect of fuel ethanol content on NMOG emissions (Chiba et al. 2010).
Elevated acetaldehyde and formaldehyde emissions have been observed when E85 fuel is compared with gasoline. (Yanowitz and McCormic 2009, Graham 2008 and West et al. 2007). The differences in emissions have been greatest at low test temperatures: acetaldehyde emissions 8–15 times higher for E85 than for E5 at normal temperature, and even more than 100 times higher at a test temperature of -7 °C (Westerholm et al. 2008). Also a recent study with Euro 4 and Euro 5a modern passenger cars found increase in acetaldehyde and formaldehyde emissions when moving from E75-E85 to E5. Acetaldehyde emission increased up to 932% at -7 °C, for example, and formaldehyde emission on average 139% and 205% at 22 °C and -7 °C, respectively (Clairotte et al. 2013). Over the hot-start test acetaldehyde emissions were low with both E85 and E5 regardless of the test temperature (De Serves 2005, West et al. 2007). In the hot-start test when engine is warmed-up, the catalyst effectively reduces emissions. However, formaldehyde emission may stay at relatively high level even with warmed-up engine (Aakko-Saksa et al. 2014).
Ethanol emissions are substantially higher for E85 fuel than for gasoline, particularly at low test temperatures, but not necessarily in the hot-start test (Yanowitz and McCormic 2009, Westerholm et al. 2008 and West et al. 2007). At cold temperatures, even 2.5% of the ethanol that is fed into the engine may released unburned (Laurikko et al. 2013). Ethanol may also lead to further increased acetaldehyde emissions through atmospheric transformation (Clairotte et al. 2013).
Higher methane emissions have been observed for E85 fuel than for gasoline at normal and at -7 °C. Also a recent study found elevated methane emissions for E75-E85 when compared with E5 by using Euro 4 and Euro 5a modern passenger cars at -7 °C. 1,3-butadiene, benzene and toluene emissions are generally lower for E85 than for gasoline (Clairotte et al. 2013, Yanowitz and McCormic 2009, Westerholm et al. 2008). Exhaust gas contains more water when using E85 instead of E5, which can disturb reactions in three-way catalyst leading to decreased conversion of hydrocarbons (Clairotte et al. 2013).
NOx emissions from FFVs running on E85 fuel are generally lower than or at the same level as those from gasoline-fuelled cars (Yanowitz and McCormic 2009, Graham et al. 2008 and Westerholm et al. 2008). Yanowitz et al. (2013) notised with nine in-use FFVs (Tier 1 and Tier 2) a reduction in the NOx emission on average −25%, with a range of −54% to +10% between E76 and E10. De Serves (2005) observed that NOx emissions were significantly lower for E85 than for E5, both in the cold-start and hot-start tests. NOx consisted almost totally of NO, indicating that NO2 emissions are low from FFVs. At -7 °C, the effect of E85 and E5 on NOx was not consistent in a study by Westerholm et al. (2008). Ammonia formation with three-way catalyst (TWC) equipped cars have been observed at normal and at -7 °C (Westerholm et al. 2008, Aakko-Saksa et al. 2014). Ammonia is formed in the chemical reactions of TWC, and it is not primarily fuel-related emission, however, for example low-sulfur fuels have been reported to enhance ammonia formation (Mejia-Centeno et al. 2007). Clairotte et al. (2013) observed lower ammonia emission associated to E75–E85 than to E5 regardless of temperature. This could be due to leaner air-fuel ratio for E85 than for gasoline, or due to high water content of exhaust gas with E85. In the latter case H2 required for the NH3 formation is limited or water may absorb the hygroscopic NH3. (Clairotte et al. 2013).p>
Yanowitz and McCormick (2009) reviewed studies of the ozone-forming potential (OFP) of E85 fuel. With Tier 1 vehicles, the ozone reactivity of the exhaust gases was lower for E85 fuel than for reformulated gasoline in some studies, although the ozone-forming potential was higher. Cold-start emissions seem to dominate the result. Studies did not consider atmospheric chemistry at individual sites, nor the effect of E85 fuel on NOx emissions. In the study by Graham et al. (2008), OFP was lower for E85 than for gasoline-fuelled FFV cars. Aakko-Saksa et al. (2011) observed higher OFP for E85 than for gasoline due to increased ethanol, ethene and acetaldehyde emissions. A recent study found higher OFP for E75-E85 than for E5 with Euro 4 and Euro 5a modern cars at -7 °C temperature when the cold start excess emissions were included (Clairotte et al. 2013). Jacobson (2007) studied the effect of E85 fuel on cancer and mortality in the US, Los Angeles basin. He concluded that E85 fuel may increase ozone-related mortality, hospitalization, and asthma when compared to gasoline. Modeling was based on emissions inventories from 11 studies. Four studies were from the 1990s, and the others were based on tests with cars up to model year 2007. The ratio between VOC and NOx is critical in estimating OFP. Ozone may increase for example in locations where the baseline ratio of reactive organic gases to NOx is below 8:1, and either NOx decreases or reactive organic gases increase. Generally, both NOx and VOC emissions from vehicles are decreasing with tightening emission limits. Millet et al. (2012) pointed out that the atmospheric impacts of increased fuel ethanol use will be minimal, because significant sources of atmospheric acetaldehyde already exist. In addition, the potency-weighted toxicity will be reduced with E85 use.
An FFV using E85 fuel generally has a low level of particulate matter (PM) emissions. For example, Yanowitz and McCormic (2009) reported lower PM emissions from an FFV using E85 fuel than from a gasoline-fuelled car. De Serves (2005) also observed lower PM emissions with E85 than with E5 fuel, but not consistently. PM emissions are generally low for the MPFI cars at normal temperature, for example below 1 mg/km in a study by Westerholm et al. (2008). In a same study, at -7 °C, PM was higher for E85 fuel than for gasoline probably due to the cold-start behavior of the E85 fuel.
Direct-injection (DI) fueling can improve fuel economy and vehicle power when compared with port fuel injection (PFI), but drawback of DI engine concerns increased particle emissions. Under the IEA AMF Annex 35-2 researchers investigated the particle size distributions and particle number (PN) emission rates of gasoline direct injection vehicles using E85 fuel and operated over different drive cycles and at different ambient temperatures (Rosenblatt et al. 2014). The rate of PN emissions was reduced by 70-90% between E85 and E0, and the distribution peak occurred at a smaller particle size (Figure 4). Consistent results were observed from different labs indicating the potential of E85 to mitigate particle emissions from gasoline direct injection engines under a variety of operating scenarios. Also Szybist et al. (2011) observed low PN emissions from DI car when using E85 fuel, comparable to emissions from PFI. E85 enabled achieving the efficiency and power advantages of DI without generating the increase in PN emissions.
Figure 4. Average particle number size distributions for a FFV operated using E0, E10 and E85 fuels over the FTP-75 (EPA Federal Test Procedure) and NEDC (New European Driving Cycle) at different ambient temperatures. (Rosenblatt et al. 2014, IEA-AMF Annex 35-2).
Particulate and semi volatile-associated PAHs together with cancer potency have been reported to be at the same level or lower for E85 fuel than for E5 at normal test temperature, however, at -7 °C also opposite results have been seen (Westerholm et al. 2008, Aakko-Saksa et al. 2014).
Intermediate-level ethanol fuels are common as a result of mixed re-fuelling of E10 and E85. In addition, blender pumps are available in the US. Haskew and Liberty (2011) studied emissions with E32 and E59 fuels in comparison to E10 and E85 over three driving cycles. NMHC emission decreased with increasing ethanol content over the US06 high speed/load test, whereas such trend was not seen over the cold start FTP test. CO and NOx did not indicate a trend with ethanol level. Fuel economy decreased with increasing ethanol content. The higher ethanol blends (E59 and E85) resulted in higher diurnal emission levels (different conclusion from CRC E-65-3), which was not in line with expectation of lower permeation of ethanol molecules for E85 than for E10. Acetaldehyde emission increased with increasing ethanol content with all three driving cycles. Formaldehyde emission increased with increasing ethanol content over two cycles, but not over the US06. The average Carter Reactivity of the exhaust decreased with increasing ethanol content of the fuels on the cold start FTP, but the results were not consistent for the US06 and Unified Cycle tests.
Yanowitz et al. (2013) determined the fuel economy and tailpipe emissions impact of operation on E40 on nine in-use FFVs (Tier 1 and Tier 2). Testing was conducted after a fuel change to study how the engine adapts the new ethanol concentration. The intermediate blends were not commonly available when older FFVs were built, and failure to rapidly adapt to a new fuel results in a non-optimal operation. Evidence was found of incomplete adaptation during the hot test immediately after refueling with E40, however, on average adaption to midrange blends was successful with average emissions falling between those of E10 and E76. Generally, the increase in emissions between E10, E40, and E76 was small compared to the measured differences between the different vehicle models.
In a study with low-oxygen fuels, E30 and E85 at -7 °C a reduction in NOx emission was observed with increasing fuel oxygen content for the Euro 4 emission level FFV (Aakko-Saksa et al. 2014). E30 and lower ethanol concentrations resulted in lower acetaldehyde, formaldehyde, ethanol, methane, ethene, and acetylene emissions when compared to E85. The emission level of 1,3-butadiene was very low in all cases. Acetaldehyde and ethanol emissions increased with increasing ethanol content of the fuel non-linearly: when changing from E30 to E85, acetaldehyde emission increased by 7.6 times and ethanol emission by 27 times. PM and PAH emissions were low and changes seemed not to be fuel-related with FFV. The indirect mutagenicity of PM extracts were lower for FFV than for the DISI car, but higher than for the MPFI car.
High-concentration ethanol fuels can be used in special flexible-fuel vehicles (FFV). Special infrastructure and safety measures are also needed for high concentration ethanol fuels. E85 is the most common fuel for FFVs today, but also methanol blends are used for example in China. Fuel injectors of FFVs are designed for higher fuel flows than in conventional gasoline cars compensating the low heating value of E85. Volumetric fuel consumption is higher for E85 than for gasoline despite of better energy consumption when using E85. The energy efficiency of an FFV engine could be improved by using a higher compression ratio as ethanol’s octane rating is high, but engines still today represent a compromise as compared to dedicated ethanol cars.
Fuel evaporative emissions are lower for E85 fuel than for gasoline. E85 fuel generally reduces CO, NOx, benzene, toluene and 1,3-butadiene emissions compared with gasoline. Acetaldehyde and ethanol emissions increase substantially with E85 fuel, and formaldehyde emissions to some extent, particularly at low temperatures. Methane emission increases when E85 is compared with gasoline. E85 fuel generally reduces particulate matter, PAH emissions, and cancer potency compared with gasoline at normal temperature, but reverse may be seen at low temperatures. The ozone-forming potential of E85 fuel tends to be higher than that of gasoline. However, this issue is complex and involves VOC and NOx emissions from cars as well as regionally varying atmospheric emissions.
Many emission species from FFV cars are high after cold-start, particularly at low ambient temperatures, as excess ethanol injection is needed before the car warms up. When the engine and catalyst are fully warmed-up, the differences in exhaust emissions between fuels are small.
The results of Graham (2008) are summarized in Figure 4, and other studies in Table 3.
Figure 5. The effect of E85 on exhaust emissions (Graham 2008).
Table 3. Examples of changes in emissions when E85 is compared to gasoline (negative values = reduction in emissions, positive values = increase in emissions).
Aakko, P. and Nylund, N.-O. (2003) Particle emissions at moderate and cold temperatures using different fuels. Warrendale: Society of Automotive Engineers. SAE Paper 2003-01-3285.
Work within IEA-AMF Annex 22, Final report
Aakko-Saksa, P., Rantanen-Kolehmainen, L., Koponen, P., Engman, A. and Kihlman, J. (2011) Biogasoline options – Possibilities for achieving high bio-share and compatibility with conventional cars. SAE International Journal of Fuels and Lubricants, 4:298–317 (also SAE Technical Paper 2011 24-0111). Full technical report: http://www.vtt.fi/inf/pdf/workingpapers/2011/W187.pdf
Chiba, F., Ichinose, H., Morita, K., Yoshioka, M., Noguchi, Y. and Tsugagoshi, T. High Concentration Ethanol Effect on SI Engine Emissions (2010) SAE Technical Paper 2010-01-1268.
CRFA (2003) Canadian Renewable Fuels Association Website. http://www.greenfuels.org.
De Serves, C. (2005) Emissions from Flexible Fuel Vehicles with different ethanol blends. AVL MTC AB, Sweden. Report No AVL MTC 5509. Online http://www.senternovem.nl/mmfiles/Emissions%20from%20flexible%20fuel%20vehicles_tcm24 280163.pdf.
E85 Handbook (2010) Handbook for Handling, Storing, and Dispensing E85. US Department of Energy. DOE/GO-102010-3073.
Graham, L., Belisle, S. and Baas, C.-L. (2008) Emissions from light duty gasoline vehicles oper-ating on low blend ethanol gasoline and E85. Atmospheric Environment 42(2008) 4498–4516.
Haskew, H. and Liberty, T. (2006) Fuel permeation from automotive systems: E0, E6, E10, E20 and E85. Final report. CRC Project No. E-65-3.
Jacobson, M. (2007) Effects of Ethanol (E85) versus Gasoline Vehicles on Cancer and Mortality in the United States. Environ. Sci. Technol. 2007, 41, 4150–4157.
Karlsson, H., Gåste, J. and Åsman, P. (2008) Regulated and non-regulated emissions from Euro 4 alternative fuel vehicles. Society of Automotive Engineers. Warrendale. Tech-nical Paper 2008-01-1770.
Kassel, R. (2006) An Environmental Perspective: EPA’s RFS Proposal. Presentation in the meet-ing of Mobile Sources Technical Review Subcommittee (MSTRS), October 4, 2006.
Rosenblatt, D., Morgan, C., McConnell, S., and Nuottimäki, J. Particulate Measurements: Ethanol and Isobutanol in Direct Injection Spark Ignited Engines. IEA-AMF Annex 35-2, Final report
Stahl, W. and Stevens, R. (1992) Fuel-alcohol permeation rates of fluoroelastomers fluoroplastics, and other fuel resistant materials. SAE Technical Paper 920163. Referred to in Kassel (2006) An Environmental Perspective: EPA’s RFS Proposal.
Westerholm, R., Ahlvik, P. and Karlsson, H.L. (2008) An exhaust characterization study based on regulated and unregulated tailpipe and evaporative emissions from bi-fuel and flexi-fuel light-duty passenger cars fuelled by petrol (E5), bioethanol (E70, E85) and biogas tested at ambient temperatures of +22 °C and 7 °C. Final Report to the Swedish Road Administration, March 2008.
West, B., López, A., Theiss, T., Graves, R., Storey, J. and Lewis, S. (2007) Fuel economy and emissions of the ethanol-optimized Saab 9-5 biopower. SAE Technical Paper 2007-01-3994.
Yanowitz, J. and McCormic, R. (2009) Effect of E85 on tailpipe emissions from light-duty vehicles. J. Air & Waste Manage. Assoc. 59(2009)172–182.