| Vehicular emissions contributes a significant share of urban atmospheric aerosol pollution in China.Besides primary particle emission,vehicle exhaust also contribute plenty of secondary aerosol through precursors,i.e.,volatile organic compounds(VOCs)and NOx.It has been reported that the vehicle-related secondary aerosol has a large contribution to air fine particles.However,due to the wide variety of organic precursors and complicate driving factors,there is still an uncertainty of knowledge in the quantification of secondary organic aerosol(SOA),leading to a gap between the results of previous studies.In addition,SOA studies has been generally through the ambient research,source information SOA formation is expected to obtain more information.Since the limit of time-resolution,the key factors(e.g.,driving conditions)effecting SOA of vehicle exhaust can not well be quantified.Based on this,this study firstly constructed a high time resolution system in SOA aging.The we conducted a series of in-situ oxidation experiments,including dynamometer tests and tunnel investigation,aim to study the driving factors effecting the SOA formation from vehicle exhaust.(1)A fast-response oxidative flow reactor system(Veh-OFR)is developed and employed to quantify the SOA formation under quasi-transient driving conditions.Compared to the state-of-the-art reactors,e.g.,potential aerosol mass chamber(PAM)and Go:PAM,Veh-OFR presents a higher laminar flow due to the improvements of inlet and outlet types.The flow in the Veh-OFR has a residence time of 45s.The time resolution of primary particle pulse response is 10s,while the time resolution of SOA formation pulse is 15s.Veh-OFR can generate a oxidative extent of~20 equivalent atmospheric aging days.The best ratio of sample flow and sheath flow is 1:2 and the particle loss can be controlled less than 5%during the photo-oxidation experiments.(2)The quasi-transient SOA concentration is calculated through the real-time SA measurement combined with the nitrate generation yield which is obtained in the sensitive experiments.The results show that the SOA production varies by order of magnitude due to the different transient conditions.The highest SOA production peaks occur in the high-speed region,especially during deceleration.In addition,we found that the sensitivity of SOA production to driving conditions tends to weaken with emission standard updates.The SOA yields are found to have a significant decline for the stringent standard vehicle.This underscores the important effects of transient driving condition and emission standard on SOA production.The average SOA production factors based on WLTC cycle tests are 276±131,145±79,and 41±27mg kg-fuel-1,or 18±11,8±4,and 2±1 mg km-veh-1for China IV,China V,and China VI vehicles,respectively.The SOA productions are 17-58 times of primary organic aerosol(POA)emissions,indicating the high level of SOA potential formation from vehicle exhaust.In addition,the reductive extent for SOA is less than that of POA with emission standard updates.This underscores the importance of organic vapor emission control for the future vehicle mitigation.(3)The SOA production are characterized based on the VOCs,SOA/CO,total hydrocarbon(THC)emissions.Then,the VSP determined SOA production model is built through the high time-resolved measurement of SOA.This will benefit and improve the“down-up”way to evaluate the SOA contribution of vehicle exhaust to the atmosphere.The estimated SOA through the VOCs source profiles and reference yields is far from the measured SOA.The estimated SOA based on the measured VOCs concentrations account for 21%,27%,and 34%of the measured SOA for China IV,China V,and China VI vehicles,respectively.This implies that the semi-/intermediate volatile organic compounds(S/IVOCs)and other organic vapors likely contribute a large proportion of SOA.In addition,we found this proportion seems to decrease with the updates of emission standards.The correlation analysis between THC and SOA show that the SOA yield of China V vehicles is highest,while the SOA yield of China VI vehicles is lowest.The SOA production from old emission standard vehicles is driven by both THC concentration and composition,while the SOA production from newest emission standard vehicles is likely to be driven by THC concentration.This indicates that the THC emission is tend to be more representative on SOA production with emission standard updates.The ratio of SOA/CO varied largely by the driving conditions with the higher values always occur in the regime of low speed and the lower values occur in the high speed bins.The measured SOA/CO variations show no statistic significant difference in the emission standards.This may be because the emissions of organic vapors and CO from vehicle exhaust have a synergistic effect on emission reduction.In addition,the ratios obtained by dynamometer tests are~3 times higher than recently reported ambient ratios.Therefore,the losses of organic precursors should be considered in the ambient evaluation using the SOA/CO ratio.(4)The SOA potential formation from the real-world vehicle fleet exhaust is obtained via in-situ photo-oxidative simulation in the tunnel environment.The concentration of SOA firstly increased with the OH exposure,peaking at the~3equivalent atmospheric aging days,and then decreased with the OH exposure.The SOA production obtained under the condition of seed aerosol in the reactor is~20%higher than that of no seed condition.This implies that the importance role of sufficient condensation sink on the SOA formation for the laminar reactor.The fuel-based SOA production factor of the real-world vehicle fleet emission is 639.2 mg kg-fuel-1,which is higher than that obtained through dynamometer tests.This is because that the driving condition in the tunnel is mild,resulting in a decrease of carbon emission,while the THC is not in the decline trend.This reveals that the real-world vehicle emission may lead to a higher potential SOA formation to the atmosphere.The VOCs can explain 37%of measured SOA for the real-world vehicle emission.While IVOCs account for 5%of identified organic vapors,it explains 11%of measured SOA.Overall,the estimate SOA is far from closed with the measured SOA,indicating that the understanding of organic precursors still need to be improved. |