Investigating the Effect of Varying Ethanol and Aromatic Fuel Blends on Secondary Organic Aerosol (SOA) Forming Potential for a FFV-GDI Vehicle

Investigating the Effect of Varying Ethanol and Aromatic Fuel Blends on Secondary Organic Aerosol (SOA) Forming Potential for a FFV-GDI Vehicle Patrick Roth 1,2 Jiacheng Yang 1,2, Ayla Moretti 1,2, Tom Durbin 1,2, David Cocker 1,2, Georgios Karavalakis 1,2, Akua Asa-Awuku 1,2,3 1. University of California Riverside Chemical and Environmental Engineering 2. College of Engineering-Center for Environmental Research and Technology 3. University of Maryland, College Park Acknowledgments: Support for this research is from ICM Inc., South Coast Air Quality Management District, UC Transportation Graduate Fellowship, and the Ester Hays Fellowship Correspondence: gkaraval@engr.ucr.edu and asaawuku@umd.edu

Secondary Aerosol (SA) SA has been measured to contribute up to ~ 75% of PM 2.5 in anthropogenic regions (Huang, et al. 2014) Inorganic salts & Secondary Organic Aerosol (SOA) Inorganic salt formation in the atmosphere is well understood NO (. OH) x HNO 3 +NH 3 SO x H 2 SO 4 +NH 3 Condensation Secondary Aerosol VOCs (NMHC) oxidation (. OH, O 3,NO 3 ) Semivolatile Compounds Nucleation Growth Secondary Organic Aerosol Organic Aerosol, 20%-90% of submicron particulate mass, is less understood (Jimenez, et al, 2009) Primary OA, primarily combustion of fossil fuels, and biomass burning Secondary OA, is formed through the reaction of volatile organic compounds Gasoline powered motor vehicles have been credited to the majority of SOA mass in large cities (Bahreini, et al, 2012) 2

Ethanol Fuel Effects o o o o Reduction in various gaseous emissions o Variable THC and NMHC Lower tailpipe particle mass concentration Reduction in aromatic content o May directly decrease overall SOA potential of vehicles (Karavalakis, et al, 2014) (Durbin, et al, 2007). Previous work utilizing flow tube reactor displayed decreasing SOA formation from increasing ethanol fuel blends (Timonen, et al, 2016) (Timonen, et al, 2016) C. Giametta, 2006 3

UCR s Mobile Atmospheric Chamber (MACh) Current mobile chambers restricted to a smaller volume (Platt et al. 2013, Presto et al. 2011) Low surface to volume ratio of 2.2:1 Minimize background effects (zero air/non-reactive chamber material/leaks) and wall losses (Aerosols/Semi-volatile aerosol precursors) ~30m 3 Volume makes it the largest mobile chamber UV Lights Single collapsible ~30m 3 2 mil FEP fluoropolymer film reactor (Saint Gobain) Injection Manifold UV Lights Aerosol and Gas Phase Characterization UV lights are a substitute for sunlight Peak UV=365nm Anodized aluminum sheets - 4250E Super UltraBrite 95, ACA Corp. Aerosol: Size, Number, Density, Volatility Hygroscopicity, Black Carbon Gas: NOx, CO, CO 2, H 2 O, and O 3 GC-FID, SIFT 4

Vehicle Used For Study Make & Model Engine Mileage 2017 Chevy Equinox 2.4L FFV 20,308 # Fuels Aromatic Content Low Aromatic 22.9 % High Aromatic 3 % 21.1 % 5.5 % Each fuel was tested over a cold and hot start -92 Driving Cycle in triplicate Dilution for all tests was ~185:1 5

Tailpipe Mass (mg/mi) Aerosol Mass (mg/mi) Results *All composition data was measured with an HR-ToF AMS* 1 1 Composition of Tailpipe Exhaust 75-90% of PM was BC Decreased PM emissions in hot start (average of 50% reduction) Highest emissions from the (10.26 mg/mi) PM was strongly affected by both ethanol and aromatic content 1 1 0 20 40 60 80 Ethanol Content (%V) 6

Aerosol Mass (mg/mi) Results Tailpipe Mass (mg/mi) 1 1 Composition of Tailpipe Exhaust 75-90% of PM was BC Decreased PM emissions in hot start (average of 50% reduction) Highest emissions from the (10.26 mg/mi) PM was strongly affected by both ethanol and aromatic content 65% decrease in PM comparing vs (23%, 21% Aromatic V.) displayed the least amount of PM with 0.49 mg/mi 1 1 0 10 20 20 40 30 60 80 40 Aromatic Ethanol Content (%V) 7

Aerosol Mass (mg/mi) Aerosol Mass (mg/mi) Results Composition of Tailpipe Exhaust Composition after Irradiation Expt 1 1 1 1 After the irradiation experiment, a negligible amount of secondary aerosol was formed all fuels The two fuels formed a small amount of SOA in the hot start The vehicle exhaust when introduced to an ultra clean environment 8 does not form secondary aerosol

Surrogate Purpose Surrogate is a mixture of reactive organic gases (ROG) developed to mimic an urban area Emission data was lumped into categories by reactivity Surrogate s purpose is provide a baseline reactivity Within each category, species were weighted by reactivity and abundance in atmosphere Surrogate on its own forms no appreciable secondary aerosol mass 3.28% 2.35% 20.92% 19.60% This is more consistent with what would occur when vehicle exhaust is emitted in urban areas 24.12% 29.03% Aldehydes Aromatics Alkanes Olefins Ketones Isoprene

Surrogate ROG Mixture Surrogate compound concentration ppb/ppmc Compound 46 Acetaldehyde 3.28% 2.35% 5 m-xylene 5 1,2,4-Trimethylbenzene 90 n-butane 14 trans-2-butene 14 Toluene 22 2-Methylbutane 13 Methylcyclopentane 20.92% 19.60% 24.12% 29.03% Aldehydes Aromatics Alkanes Olefins Ketones Isoprene 16 Ethylene 14 Propylene 3 1-Pentene 17 Methyl Ethyl Ketone Percentage of each category in surrogate (Kacarab, 2016) 2 Isoprene This is the surrogate that was used for the remaining chamber experiments All procedures were the same as summarized earlier, with the addition of 1 ppmc of the surrogate, and ~45ppb NOx 10

Aerosol Mass (mg/mi) Aerosol Mass (mg/mi) Surrogate Results Composition of Tailpipe Exhaust Composition after Irradiation Expt 1 1 7 6 5 4 3 2 1 Results after irradiation displayed the formation of a large amount of Ammonium Nitrate For all fuels excluding, less secondary aerosol formed in the hot start experiments 11

Aerosol Mass (mg/mi) Aerosol Mass (mg/mi) Results After Irradiation Vehicle emitted an average of 2.9 & 2.1 mg/mi of NO x for cold and hot starts respectively 1 1 Resulted in only ~5-10 ppb NO x in chamber Exhaust Only 7 6 5 4 3 2 1 Exhaust with Surrogate & NOx NO x concentrations in surrogate experiments were 45-60 ppb Similar NH 3 concentrations in both 12

Total Carbonaceous (mg/mi) Carbonaceous Aerosol (mg/mi) Total Carbonaceous (mg/mi) Carbonaceous Surrogate Aerosol Results Concentration 10 10 10 10 0 0 0 0 0 Composition after Irradiation Expt 3.28 3.36 2.44 2.44 1.27 2.80 2.82 Increasing aromatic results in increasing carbonaceous aerosol Increasing ethanol results in a decreasing carbonaceous aerosol 2.16 Hot & Cold starts have different trends 1 1 1 1 1 1 0 10 20 30 40 Aromatic Content (%V) 0 20 40 60 80 100 Ethanol Content (%V) 13

Carbonaceous Aerosol (mg/mi) SOA(mg/mi) Carbonaceous Surrogate Aerosol Results Concentration 10 10 10 10 0 0 0 0 0 Composition after Irradiation Expt 3.28 3.36 2.44 2.44 1.27 2.80 2.82 2.16 3.5 3.0 2.5 1.5 1.0 0.5 5.0 1 15.0 2 THC (mg/mi) Largest amount of SOA formed with the cold start (3.36 mg/mi) hot start formed the least amount of SOA (low THC emissions) Overall there was a strong trend relating the SOA formed to the THC emitted Both hot & cold starts fall on similar trendline May allow prediction of SOA potential 14

Aerosol Mass (mg/mi) Conclusions Average of 50% reduction in tailpipe PM in hot start 1 1 15

Tailpipe Mass (mg/mi) Conclusions Tailpipe Mass (mg/mi) Average of 50% reduction in tailpipe PM in hot start Ethanol and Aromatics have inverse effects on tailpipe PM 1 1 1 1 0 20 40 60 80 Ethanol Content (%V) 0 10 20 30 40 Aromatic Content (%V) 16

Aerosol Mass (mg/mi) Conclusions Aerosol Mass (mg/mi) Average of 50% reduction in tailpipe PM in hot start Ethanol and Aromatics have inverse effects on tailpipe PM Negligible SA formed in exhaust only experiments SA increased drastically with addition of surrogate and NOx Consistent reactivity & increased NOx 1 1 Exhaust Only 7 6 5 4 3 2 Exhaust with Surrogate & NOx 1 17

Total Carbonaceous (mg/mi) Conclusions Total Carbonaceous (mg/mi) Average of 50% reduction in tailpipe PM in hot start Ethanol and Aromatics have inverse effects on tailpipe PM Negligible SA formed in exhaust only experiments SA increased drastically with addition of surrogate and NOx Consistent reactivity & increased NOx Total Carbonaceous (BC, POA, and SOA) Increased with increased aromatic Decreased with increasing ethanol Different trends with varying driving conditions 15.0 1 15.0 1 5.0 0 10 20 30 40 Aromatic Content (%V) 5.0 0 20 40 60 80 18 100 Ethanol Content (%V)

Conclusions SOA(mg/mi) Average of 50% reduction in tailpipe PM in hot start Ethanol and Aromatics have inverse effects on tailpipe PM Negligible SA formed in exhaust only experiments SA increased drastically with addition of surrogate and NOx Consistent reactivity & increased NOx Total Carbonaceous (BC, POA, and SOA) Increased with increased aromatic Decreased with increasing ethanol Different trends with varying driving conditions SOA formation vs THC trend least reliant on driving condition 3.0 1.0 5.0 1 15.0 19 2 THC (mg/mi)