Research On The Applicability And Screening Of Organic Markers In PM_2.5 Refined Source Apportionment

With the implementation of the clean air policy throughout the world,declines in particulate matter concentrations has been very effective.However,due to the optimization of energy structure and changes in people’s lifestyles,the sources of particulate matter have become more complex.Most elements are emitted from various types of sources,so it is difficult to identify the source based on the element data.In order to better control the source of particulate matter,refined source apportionment need to be done.The advantage of using organic compounds in source apportionment is that they have the identification function for specific sources.Thus,they can be used to reduce the co-linearity of sources during source apportionment and improve the accuracy of particle source apportionment.However,PMF is unable to apportion particulate matter sources with large fractions of organic components and few elemental compositions.in our analysis is to perform an initial screening to identify and exclude compounds that do appear to be significantly modified during transport from source to receptor.Therefore,it is necessary to carry out screening research on organic and inorganic markers before they were performed in the PMF model.This study focuses on solving the above issues,the PM_2.5 samples in Chengdu in 2018 were analyzed for conventional inorganic components such as elements,ions,OC and EC,and for organic components such as PAHs,hopanes,steranes,and n-alkane.The analysis of the indicators is combined with the characteristics of the organic and inorganic components,a statistical analysis method with a model verification method for screening organic and inorganic markers are constructed.At the same time,the source apportionment results of PM_2.5 in Chengdu in 2018 were evaluated.The main conclusions are listed as follows:（1）Significant seasonal differences were found in the concentration of PM2.5and its components during the monitoring period,and the main sources of organic components in PM2.5 are predicted based on the indicators.The primary chemical composition of PM_2.5during the monitoring period were water-soluble ions（31%）and OC（18%）.The concentration and percentage of n-alkanes and PAHs in organic components are higher than those of hopanes and sterane.The average mass concentrations of PM_2.5,inorganic elements,water-soluble ions,OC,EC,PAHs,hopanes,sterane,and low-carbon n-alkanes（C14-C22）were higher in winter than in summer.The high-carbon n-alkanes（C22-C33）is much higher in summer than in autumn,spring and winter,indicating that the main source of n-alkanes in summer is the emission of biological sources such as plant waxes.Through the analysis of organic and inorganic markers in this study,PAHs can be used as marker for motor vehicles and coal-fired sources.Hopanes can be used as the marker for motor vehicles and coals of different maturity.N-alkanes can be used as markers of fossil fuels with different maturities and biological sources.The results of the diagnostic ratio of PAHs indicate that the PAHs in PM_2.5 mainly come from vehicle emission,also emitted by coal emissions in autumn and winter.The ratio of the hopanes configuration indicates that the hopanes in PM_2.5 mainly comes from the motor vehicle and coal combustion in spring,autumn and winter,while is mainly emitted by motor vehicles in summer.The indicators of CPI in n-alkanes also show that n-alkanes in PM_2.5 are mainly from anthropogenic sources in the autumn and winter,while are emitted by biological sources in summer.（2）In this study,the organic and inorganic markers included in the PMF model were screened by statistical methods and model screening verification methods,and the applicability of the screened organic components in the particle source apportionment was verified by the model.When a large number of organic components and a small amount of inorganic components are directly performed in the PMF model for source apportionment without screening,the co-linearity of the model increases due to the large correlation between some organic components,that is,the model extracts organic components uniformly to one or two factors,thereby masking the identity of the organic components,making it impossible to distinguish the sources by its identity,increasing the difficulty of source apportionment.In this study,1.through the analysis of the characteristics of the source emissions and the time difference of the acceptor components of the conventional components and organic components of the particulate pollutants,using the information of the role of organic markers from different sources,2.Analysis of the characteristics of the emission sources of organic markers and the changes in the receptors.3.By evaluating the results of the model,the stability and reliability of the screening results are determined,and the effectiveness of the organic components in the refined source apportionment after screening is further verified.（3）The organic and the inorganic markers used in the PMF model to realize the refined source apportionment.In this study,the screened organic and inorganic markers are jointly incorporated into the PMF model and the source apportionment of urban particulate matter samples was performed.Through comparison with the source apportionment results of conventional PMF that include only the inorganic markers,it is found that the source apportionment result obtained from the screened organic and inorganic markers are basically the same as the conventional PMF source apportionment result.Refined source apportionment can be achieved more accurately by adding organic markers.It has a finer effect on the apportionment of motor vehicle sources and coal-fired sources.It can refine the motor vehicle sources（22%）into gasoline vehicles（14%）and diesel oil vehicles（8%）.Coal-fired sources（14%）were subdivided into mature coal-fired sources（10%）and immature coal-fired sources（4%）.