Assessing the Impacts of Ethanol and Isobutanol on Gaseous and Particulate Emissions from Flexible Fuel Vehicles
Total Page:16
File Type:pdf, Size:1020Kb
Article pubs.acs.org/est Assessing the Impacts of Ethanol and Isobutanol on Gaseous and Particulate Emissions from Flexible Fuel Vehicles † ‡ † ‡ † † § † ‡ Georgios Karavalakis,*, , Daniel Short, , Robert L. Russell, Heejung Jung, , Kent C. Johnson, , † ‡ † ‡ Akua Asa-Awuku, , and Thomas D. Durbin , † Bourns College of Engineering, Center for Environmental Research and Technology (CE-CERT), University of California, 1084 Columbia Avenue, Riverside, California 92507, United States ‡ § Department of Chemical and Environmental Engineering and Department of Mechanical Engineering, Bourns College of Engineering, University of California, Riverside, California 92521, United States *S Supporting Information ABSTRACT: This study investigated the effects of higher ethanol blends and an isobutanol blend on the criteria emissions, fuel economy, gaseous toxic pollutants, and particulate emissions from two flexible-fuel vehicles equipped with spark ignition engines, with one wall-guided direct injection and one port fuel injection configuration. Both vehicles were tested over triplicate Federal Test Procedure (FTP) and Unified Cycles (UC) using a chassis dynamometer. Emissions of nonmethane hydrocarbons (NMHC) and carbon monoxide (CO) showed some statistically significant reduc- tions with higher alcohol fuels, while total hydrocarbons (THC) and nitrogen oxides (NOx) did not show strong fuel effects. Acetaldehyde emissions exhibited sharp increases with higher ethanol blends for both vehicles, whereas butyraldehyde emissions showed higher emissions for the butanol blend relative to the ethanol blends at a statistically significant level. Particulate matter (PM) mass, number, and soot mass emissions showed strong reductions with increasing alcohol content in gasoline. Particulate emissions were found to be clearly influenced by certain fuel parameters including oxygen content, hydrogen content, and aromatics content. ■ INTRODUCTION offer advantages over their PFI counterparts, there can be issues Ethanol is the world’s most popular biofuel for use with existing relating to fuel preparation, including fuel contact with the spark-ignition (SI) engines.1 In the U.S., the increase in ethanol cylinder wall surfaces during combustion, which leads to the formation of soot because the wall quenches the flame and Downloaded via UNIV OF CALIFORNIA RIVERSIDE on July 10, 2019 at 22:29:20 (UTC). use has been promoted by several legislative measures, including the Energy Independence and Security Act (EISA 2007) and the prevents complete combustion of the fuel, especially during the See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. 4,5 Renewable Fuel Standard (RFS), which was initiated in 2005 and cold-start. 2 Analogous to ethanol, higher alcohols have been the subject of expanded in 2007. The latter mandates the use of 36 billion 6,7 gallons of renewable fuels in the transportation fuel pool by 2022. increased interest as potential fuels in SI engines. Particular Commercial U.S. gasoline contains ethanol at a concentration of emphasis has been given to butanol isomers, which combine the advantages of an energy density closer to that of gasoline with the 10% by volume (E10). Ethanol is also available as E85, which 8 after a recent change in specifications is allowed to contain as oxygen content and renewability of ethanol. Butanol isomers much as 83% v/v and as little as 51% v/v ethanol. Vehicles can be produced from biochemical pathways via fermentation using biomass-derived feedstocks, including corn, sugar cane, designed to use higher levels of alcohol fuels in gasoline, 8,9 especially ethanol, are known as flexible fuel vehicles (FFVs) and and cellulosic biomass. Compared to ethanol, butanol isomers are currently being offered by many manufacturers. exhibit lower vapor pressure, are less corrosive, and less Currently, there is a widespread growth of gasoline direct hydrophilic. However, the octane number and latent heat of injection (GDI) vehicles aiming at improving fuel economy over vaporization of butanol isomers are lower than ethanol. Butanol conventional port fuel injection (PFI) engines and reducing carbon dioxide (CO2) emissions. Most GDI engines employ Received: July 17, 2014 wall-guided designs, in which the fuel spray is directed from a Revised: October 24, 2014 side-mounted fuel injector toward a contoured piston and then Accepted: November 6, 2014 upward toward the spark plug.3 While wall-guided GDI engines Published: November 6, 2014 © 2014 American Chemical Society 14016 dx.doi.org/10.1021/es5034316 | Environ. Sci. Technol. 2014, 48, 14016−14024 Environmental Science & Technology Article isomers contain less oxygen per unit volume compared to ethanol and 90% gasoline), which served as the baseline fuel for ethanol. This gives butanol isomers a higher lower heating value this study, and two more ethanol blends, namely E51 and E83. of 33.1 MJ/kg compared to 26.8 MJ/kg for ethanol, which more The ethanol fuels were blended by Haltermann Solutions, closely resembles the heating value of gasoline (42.7 MJ/kg).10 Channelview, TX, to represent ethanol fuels that would be These factors offer the potential for butanol isomers to be utilized in California in terms of properties such as aromatic blended at higher blending levels in gasoline than ethanol under content, Reid vapor pressure (RVP), and other properties. The existing regulations and with current technology engine control higher ethanol blends represent the upper and lower blend limits systems, providing additional pathways to satisfy the volume of the current E85 specification. Isobutanol was blended with targets set by the RFS. gasoline at a proportion of 55% (Bu55) by volume. This was the The use of midlevel and higher level blends of ethanol has been highest volume of isobutanol that could be blended while still − widely investigated in older and modern technology FFVs,11 14 meeting the California summer gasoline specifications. The main while data on emissions from higher level butanol blends is more physicochemical properties of the test fuels are provided in the sparse.15,16 Most chassis dynamometer studies on the effects of Supporting Information, Table S1. higher ethanol blends on tailpipe emissions have been conducted Testing was conducted on two late-model FFV pickup trucks on FFVs equipped with PFI engines and generally focused on the with similar horsepower ratings and certification levels. This regulated emissions and gaseous air toxic pollutants.17,18 Haskew included a 2013 model year (MY) Ford F150 (PFI-FFV) with a and Liberty14 tested a fleet of seven 2006−2007 model year FFVs 3.7 L V6 engine and PFI fueling having a rated horsepower of 302 and did not find statistically significant fuel effects between E6 hp at 6500 rpm and a 2014 MY Chevrolet Silverado (GDI-FFV) and E85 for most of the regulated emissions on most of the test with a 5.3 L V8 engine and wall-guided direct injection fueling cycles. Hubbard et al.19 reported increases in acetaldehyde, having a rated horsepower of 355 hp at 5600 rpm. Both vehicles fl formaldehyde, methane (CH4), and ammonia with increasing were equipped with three-way catalysts (TWCs), were exible- ethanol content in the fuel when they tested a 2006 model year fuel capable, and were certified under California ULEV II/Tier 2 FFV over the Federal Test Procedure (FTP) cycle. They also Bin 4 emission standards. The vehicles had accumulated mileages found lower nitrogen oxide (NOx) and nonmethane hydro- of 13700 for the Ford F150 and 2649 for the Chevrolet Silverado carbon (NMHC) emissions compared to E0. Similar findings at the beginning of the test campaign. were seen by Yanowitz and co-workers20 when they tested nine Driving Cycles and Measurement Protocol. Each vehicle FFVs over the LA92 test procedure. They found reductions in was tested on each fuel over three FTPs and three UC tests. The NOx, carbon monoxide (CO), and carbon dioxide (CO2) as well six tests on a particular fuel were conducted sequentially once the as increases in emissions of ethanol, acetaldehyde, and vehicle was changed to operate on that fuel, and the fuel was not formaldehyde with increasing ethanol concentration. Graham changed to another fuel during this time. The preconditioning et al.21 showed that the use of E85 resulted in statistically procedure was similar to that specified in the Code of Federal fi signi cant decreases in NOx, NMHC, 1,3-butadiene, and Regulations (40 CFR 86.132-00). For each fuel change there benzene emissions relative to E0 over the FTP and statistically were multiple drain and fills and 2 LA4s along with idle periods significant increases in formaldehyde and acetaldehyde. To date, between the testing on each fuel to condition the vehicle and there is limited data on PM emissions from FFVs equipped either ensure no carryover effects. Detailed information on the driving with PFI or GDI engines. Recently, Mamakos and colleagues22 cycles employed in this study and the testing protocol can be found large reductions in particle number and PM mass found elsewhere.25 emissions from a Euro 5 GDI-FFV with the use of 75−85% Emissions Testing and Analysis. All tests were conducted ethanol/gasoline blends over the New European Driving Cycle in CE-CERT’s Vehicle Emissions Research Laboratory (VERL), (NEDC) and the Artemis cycles. Reductions in particle number which is equipped with a Burke E. Porter 48-in. single-roll electric concentrations with E85 compared to gasoline were also seen in a dynamometer. A Pierburg Positive Displacement Pump- study conducted by Lee et al.23 when they tested a 2005 model Constant Volume Sampling (PDP-CVS) system was used to year PFI-FFV over the NEDC. Magara-Gomez and co-workers24 obtain certification-quality emissions measurements. For all tests, showed reductions in particle number emissions with E85 and standard bag measurements were obtained for THC, CO, NOx, E65 compared to E35 and E6 when they tested a 2007 model NMHC, and CO2. NMHC was determined from the combined year PFI-FFV over the LA92 cycle.