Smog Chamber Studies on SOA Formation from Gasoline Exhaust and Pure Precursors E. Z. Nordin 1, A. C. Eriksson 1,2, J. E. Carlsson 1, P. T. Nilsson 1, M. K. Kajos 3, P. Roldin 2, J. Rissler 1, M. Hallquist 4, M. Kulmala 2,3, B. Svenningssson 2, M. Bohgard 1, E. Swietlicki 2 and Joakim Pagels 1 1 Ergonomics & Aerosol Technology, Lund University, Lund, Sweden 2 Nuclear Physics, Lund University, Lund, Sweden 3 Dept. Physics, Division of Atmospheric Sciences, Helsinki University, Helsinki, Finland 4 Dept. Chemistry, Göteborg University, Gothenburg, Sweden
Outline/Abstract Background: Idling gasoline exhaust emits SOA precursors Method: Smog chamber ageing experiments, cold idling vehicles Results: Apparent mass yield (C6-C10 aromatics) and chemical composition Conclusions: Differences between SOA from gasoline exhaust and pure precursors
Background and Scientific Questions VOC-emissions from a single cold start in modern gasoline cars can correspond to thousands of driven kilometers (Weilenmann et al. 2009). Gasoline vehicles are major contributors of Ammonia emissions in urban environments. Can the magnitude of SOA-formation from gasoline exhaust be explained from light aromatics only? Can the composition of SOA from gasoline exhaust be simulated using simple precursor mixtures? What is the relation between SOA and Ammonium Nitrate formation in photo-oxidized gasoline exhaust?
Light aromatic VOCs Light aromatics: Benzene, Toulene, Xylenes, C9 and C10 aromatics Benzene ring with 0,1,2,3 functional groups (methyl, ethyl, propyl) Known SOA precursor Idling Gasoline exhaust consists of 10-15 % light aromatics
Method: Experimental setup 6 m 3 FEP teflon chamber Black lights (350 nm), 20*100 W NO 2 photolysis rate: 0.2 min -1 Exhaust sampled with heated inlet <10 % RH 22 ± 1.5 o C
Gasoline exhaust Cold Idling Volvo V40 (98) Engine temp ~50 C Exhaust injection 5-15 min Primary dilution ratio: 5
List of experiments Exp Type Initial C6-C10 Formed Mass Yield ID to NO ratio OA (µg/m 3 ) G1 Cold idling 8.8 26 0.18 G2 Cold idling 6.4 46 0.23 G3 Cold idling 21.7 29 0.26 G5 Cold idling 1.6 60 0.26 G6 Cold idling 8.0 7 0.10 G7 Cold idling 5.9 6 0.08 G4 Cold start 4.1 16 0.09 P2 Precursor 5.8 52 0.12 (T, mx, TMB) P3 Precursor (mx) 6.2 73 0.19 15-20 ug/m 3 Ammonium Sulphate seed added 10-120 ppb NO added Total C6-10 light aromatics ~150-300 ppb DR~50-100
Particle and Gas-Phase Characterisation HR-TOF-Aerosol MS (Aerodyne): On-line Particle Composition PTR-MS (Ionicon): Time resolved concentration of selected VOC s C6-C10 light aromatics etc. SMPS (Hauke DMA) : Mobility size distribution (10-700 nm) O 3, NO/NO 2, GC-MS-samples, Undiluted exhaust gas analysis (TVOC, CO, NO x )
Mass concentration (µg/m 3 ) PPB Results: Gas, AMS (not wall loss corrected) and SMPS data 120 100 5 4 3 2 1 80 60 40 20 NO NO2 O3 0 0 20 40 60 80 100 120 140 160 Time (min) SMPS Cl NH4 org NO3 SO4 30 25 20 15 10 5 Volume concentration (µm 3 /cm 3 ) SOA and NH 4 NO 3 formation starts when NO approaches 0. Organic aerosol and ammonium nitrate cocondenses on the seed particles. The sulfate concentration is seemingly increased due to increased collection efficiency. For more information see poster: 8P270 by Eriksson et al. 0 0 0 20 40 60 80 100 Time (min) 120 140 160
ppb ppb Light Aromatics PTR-MS measurements of the most abundant aromatic precursors in gasoline exhaust. Results consistent with GC-MS sampling. ΔVOC = Total reacted C6- C10 aromatics due to photochemical reactions: 120 100 80 60 40 20 120 100 80 60 40 20 Gasoline exhaust experiment G2 0 0 60 120 Time(min) Precursor experiment P2 (m/z) 79 (m/z) 93 (m/z) 107 (m/z) 121 (m/z) 135 (m/z) 93 (m/z) 107 (m/z) 121 0 0 60 120 180 240 Time (min)
Gasoline Exhasut vs Pure Precursors Gasoline Exhaust COA Pure Precursor Toluene-, m-xylene- & 1-2-4TMB- mixture Wall loss correction: Hildebrandt et al. 2009
Apparent Mass Yield The apparent Mass Yield ΔVOC= ΣC6-C10 The Apparent Yield from gasoline exhaust is higher than for pure precursors The Yield is dependent on the initial light aromatics to NO ratio 0.35 0.3 0.25 0.2 0.15 0.1 Gasoline Exh., VOC/NO 6-9 Gasoline Exh., VOC/NO 22 0.05 Gasoline Exh., VOC/NO 2 Precursor, VOC/NO 6-7 0 0 20 40 60 80 100 OA mass µg/m 3 The black line is fitted to the function : A*X^B and represents VOC/NO ~7
AMS Mass Spectra Aged Gasoline Exhaust m/z 44 CO 2 + m/z 43 C 3 H 7 + C 2 H 3 O + Organics Nitrate Ammonium
m/z 43 m/z 44 Gasoline Exhaust Organics Nitrate Ammonium m/z 43 m/z 44 Precursor Experiment: (Toluene, m-xylene, 1,2,4-TMB)
Comparison with atmospheric observations The Ng-triangle (Ng et al. 2010) f44 0.25 0.2 0.15 G1 G2 G3 G4 G5 G6 G7 P2 P3 0.1 0.05 0 0 0.05 0.1 0.15 0.2 f43
Conclusions SOA emission factors from gasoline exhaust cannot entirely be explained from classical precursors (light aromatics). Additional precursors in the gasoline exhaust might be one explanation. SOA from gasoline exhaust is closer to atmospheric SOA observations compared to pure precursor mixtures. SOA from gasoline co-condenses with ammonium nitrate, the formation of ammonium nitrate in high NO x experiments is of the same magnitude as the SOA formation.
Photograph : Kenneth Ruona Thank you for your attention! References: Hildebrandt et al. Atmos. Chem. Phys., 9, 2973 2986, 2009 Ng et al. Atmos. Chem. Phys., 10, 4625 4641, 2010 Weilenmann et al. Atmospheric Environment 43 2419 2429, 2009
This work was supported by the Swedish research council FORMAS through projects 2007-1205, 2008-1467, 2010-1678 and METALUND Oral presentation about SP-AMS measurements by: Axel Eriksson: #6E2 Oral presentation about atmospheric processing of diesel soot by: Joakim Pagels, #6B1, (Carlsson et al 2011)