Physicochemical Characterization And Oxidative Potential Of HULIS In PM_2.5 Collected From Suburban Shanghai Atmosphere

The vigorous urbanization and incredible economic growth have driven to increase air pollution and haze events in megacities in China.Shanghai,one of the megacities in China has frequently suffered heavy haze weather in recent years and the particulate matters（PMs）with an aerodynamic diameter less than 2.5μm(PM_2.5)has become the main threatening factor for visibility impairment（Haze）not only in Shanghai also in most cities in China.More recently,most of the PM studies have intensively shifted to the aerosol-phase Humic-like substances（HULIS）,the most ubiquitous class of water-soluble organic compounds in the atmosphere and also enhance the generation of reactive oxygen species（ROS）,and thus,play a significant role in impacting the aerosol chemistry and health effects.However,most of the HULIS studies all over the world were emphasized on physicochemical characterization,source identification of HULIS and a few studies have attempted to measure the oxidative potential and ROS-active potential on its surrogate compounds（SRFA）and atmospheric HULIS by using cell-free assays;however,there is still very limited information on cellular toxicity of HULIS.In this study,twenty-three PM_2.5samples were collected in the atmosphere of suburban Shanghai from November 29 to December 17,2015,and March 17 to April30,2016,during haze and non-haze days.Aqueous-phase HULIS-C was isolated by solid-phase extraction（SPE）method and HULIS-C contents were quantified by using of TOC-VCPN analyzer.The carbonaceous contents of PM_2.5 such as WSOC,OC/EC,and ionic species including five anions(F^-,C1^-,NO₃^-,SO₄^2-,C₂O₄^2-)andtotal of twenty-nine elements were detected through an inductively coupled plasma-mass spectrometry（ICP-MS）to extrapolate the interrelations between HULIS and trace metals.The oxidative potentials of HULIS and HULIS or SRFA-Fe（II）binary mixtures were assessed by using a cell-free DTT assay.Moreover,the first time study of cytotoxicity of HULIS and HULIS-Fe（II）or SRFA-Fe（II）binary mixtures to A549 cells was carried out through biotic assays（ROS and MTT）.The potential source regions contributing to the PM_2.5 and HULIS in Shanghai were also computed by 72hr-backwards trajectories,clustering analysis,and two-hybrid receptor models（PSCF and CWT）.The summaries of this study are as follows:1.The average mass concentrations of PM_2.5 during haze and non-haze days were 126.54μg/m³,85.55μg/m³in winter and 93.77μg/m³,63.29μg/m³ in spring which were relatively higher than 24-hr PM_2.5 air quality standard of China except for spring during non-haze days.The distribution trend of PM_2.5 in those seasons was in the order of winter（haze）>spring（haze）>winter（non-haze）>spring（non-haze）,indicating the influence of weather conditions on PM_2.5 pollution in Shanghai.The average mass concentration of PM_2.5during hazy days was relatively 1.5 times higher than that of non-hazy days in both seasons.Besides,winter haze days exhibited～1.5 times of that in winter non-haze and spring haze days and about 2 times of non-haze days in spring.PM_2.5 concentration was highly correlated with visibility（p<0.05）in winter（non-haze）days and spring（haze）days while insignificant positive correlations in winter during hazy days and weakly in spring during non-hazy days.Also,there were negative relationships with mean temperature and also with RH in both seasons while negative correlations between RH and visibility in spring during hazy days.Therefore,the elevation of PM_2.5 and the formation of haze in Shanghai are mainly concerned with their respective meteorological conditions.2.The carbonaceous contents（OC,EC,and TC）showed the seasonal variations which are inconsistent with that of PM_2.5 while the water-soluble inorganic ions had the greater effects of weather conditions i.e.haze and non-haze as in PM_2.5.The average OC contents were 16.6μg/m³,11.76μg/m³,7.79μg/m³ and 7.3μg/m³during haze and non-haze days in winter and spring whereas EC contents were～4.3μg/m³ for both haze and non-haze days in winter and～2.0μg/m³for those days in spring.During hazy days,the contributions of OC to PM_2.5 were in the range of～10-22%in winter and～7-14%in spring while～10-17%and 8-17%of PM_2.5 mass in those seasons during non-hazy days.The OC/EC ratios were extremely high（2-6）in both seasons except for some sampling days while the highest OC/EC ratios observed in spring during non-haze days.Moreover,the OC contents had strong correlations with EC in spring during both haze and non-haze days,suggesting that OC and EC in spring might be from the same pollution sources while no correlation between OC and EC in winter.3.The nitrate,sulfate,and ammonium are the most dominant species in suburban Shanghai in both seasons during haze and non-haze days;however,the order of distribution of all analyzed ions depends on the weather conditions.The accounted for more than 80%of TWSIs except for winter non-haze periods.Among ions,the significant elevation of oxalate concentration was observed in this study when compared with other studies in urban Shanghai and Beijing.During haze days,the oxalate concentration was exceeded nearly 2 times in winter and 1.3 times in spring during non-haze days.Besides,K⁺ion in winter during both haze and non-haze days exceeded 2-3.5 times of spring.Most of the PM_2.5 samples in spring observed in winter（haze and non-haze days）and spring haze PM_2.5samples.In suburban Shanghai,K,Na,Al,and Fe are the most abundant metals whereas K and Fe contributed～20%of total metals（TMs）and the lowest contributions from Na among the high-mass level metals.According to correlation and factor analysis results,biomass burning,secondary aerosols formation,marine sources,and industrial emissions are the principal sources of PM_2.5 formation in winter whereas PM_2.5 in spring had combined effects from above mention sources and also the crustal sources.4.The mass level of HULIS-C during non-hazy days was higher than that in hazy days in those seasons.The mean concentrations of HULIS-C in spring during relatively higher than in that of winter（1.93±0.95μg/m³ and 1.31±0.28μg/m³）Moreover,the study revealed that the contributions of HULIS-C to WSOC,OC,and PM_2.5 in spring（～54%,28%,and 3%）were relatively higher than in winter（27%,11%,2%）although WSOC contents in winter were exceeded 4 times during hazy days and 3 times during non-haze days in spring.These facts suggest that the formation hydrophobic fraction in spring is more important than that of winter during both haze and non-haze days.The correlation results of HULIS-C with chemical components of PM_2.5 were consistent with that of PM_2.5.Moreover,the pattern of HULIS-C contents was similar to that of the K element（biomass burning tracer）and SIA.The HULIS-C formation in winter was associated with primary organic carbon（OC）emissions,biomass burning,secondary inorganic aerosol formation process,and also the industrial emissions sources in winter whereas mixed effects were observed in spring including the crustal sources and marine aerosols.5.The results of DTT showed that the overall DTT consumption rate by individual HULIS or SRFA was the highest and their Fe（II）binary mixtures had lower DTT consumption.However,the HULIS or SRFA-Fe（II）complex had a higher consumption rate than Fe（II）alone,suggests the large effects of organic-metal complexation on the generation of reactive oxygen species.The atmospheric HULIS in both seasons exhibited higher DTT consumption over its surrogate（SRFA）compounds.This is due to the different origins,emissions,aging time,and transport pathways of atmospheric HULIS.In DTT_m,the trend of the oxidative potential of HULIS was winter（non-haze）>winter（haze）>spring（non-haze）>spring（haze）whereas DTT_v activities showed the order of seasonal variation.The DTT_v and DTT_m of HULIS in those seasons was highly correlated with their respective Fe（II） binary mixtures whereas DTT_v of HULIS and its Fe（II）complex had strong associations with carbonaceous components,ions and also metals,especially in spring.6.The results of the ROS assay showed that higher ROS levels were released from individual HULIS or SRFA（Suwannee River Fulvic Acid）compounds than HULIS-Fe（II）complexes and Fe（II）alone.DTT and ROS correlation results were also in line with each other.Incredibly,the obtained data revealed that individual atmospheric HULIS samples have a powerful capacity to produce the oxidative potential in both cell-free DTT assay and cell-based ROS assay.DTT and ROS correlation results were also in line with each other but ROS in winter has anti-correlations with most of the metals.7.Dose-dependent MTT results revealed that the highest cell mortality（44%）observed at high HULIS concentration（50μg/ml）and the lowest for 10μg/ml.Like DTT and ROS assays,individual HULIS and SRFA compounds have high cell mortality when compared with their Fe（II）binary mixtures.However,MTT assay exhibited a reverse trend of DTT and ROS assays.In a reverse trend,the highest cell mortality in HULIS-Fe（II）system might be due to the effects of mixing time（40min）between HULIS-C and Fe（II）,and the participation of Fenton-like reaction.It is quite interesting to note that the organic-metal complexation time and the Fenton reaction induced by Fe（II）play a major role in the cell mortality rate.DTT and ROS assays had strong correlations in those seasons and in turn,MTT and ROS showed high relationships with each other.According to the correlation results,the cell mortality of A549 cells by winter HULIS-Fe（II）complex had strong positive correlations with all carbonaceous species of PM_2.5except EC,biomass burning（K）,secondary inorganic aerosols（nitrate,sulfate,ammonium,and SIA）,sea salt（Na）and some metals（Se,Rb,Pb）whereas the cell mortality by spring HULIS showed negative correlations with WSOC,OC,EC and insignificantly with most of the metals.8.Based on the backward trajectories and hybrid receptor（PSCF and CWT）models,the principal potential source regions of PM_2.5 and HULIS-C were the YRD regions,the inland regions（Jiangsu,Shandong,Hebei,Anhui）and the marine areas（the Yellow Sea,the East China Sea,the Bohai Sea）in winter during both haze and non-haze days.In spring,the YRD regions,the adjacent（Jiangsu）province,and local（urbanized and industrialized areas in Shanghai）and the marine areas had stronger effects while moderate effects from other countries（Japan and North Korea）during haze days.Apart from the above-mentioned areas including the PRD regions showed the formation of these pollutants in spring during non-haze days.Interestingly,the local Shanghai（urbanized and industrialized areas）made a significant contribution to HULIS-C and PM_2.5 in both seasons while little effects from the long-range transports（northwestern regions）.