Characteristics of PM 2.5 Pollution with Comparative Analysis of O 3 in Autumn–Winter Seasons of Xingtai, China

: Pollutants emission, meteorological conditions, secondary formation, and pollutants transport are the main reasons for air pollution. A comprehensive air pollution analysis was conducted from the above four aspects in the autumn–winter seasons of 2017–2018 and 2018–2019 at Xingtai, China. In addition, the relationship between PM 2.5 and O 3 was also studied from the aspects of secondary formation and meteorological conditions to ﬁnd the rules of cooperative management of PM 2.5 and O 3 combined pollution. Taking measures of concentrated and clean heating and controlling biomass burning could make the concentrations of EC, K + and SO 42 − decrease. The variation trends of PM 2.5 and O 3 concentration in the autumn–winter season of Xingtai were different, and with the increase in secondary formation effects, the concentration of O 3 decreased. Furthermore, the key meteorological conditions that affected O 3 and PM 2.5 formation were temperature and relative humidity, respectively. The relationships of NOR (nitrate oxidation rate) and SOR (sulfate oxidation rate) against temperature presented a “U” shape, suggesting that gas-phase oxidation and gas–solid-phase oxidation were all suppressed at a temperature of around 4 ◦ C. The cities located in the east had more pollutant transporting effects during the pollution processes of Xingtai, and the main transport routes of O 3 and PM 2.5 were not all the same.


Introduction
The accelerated urbanization, booming economy, increased vehicles, and developing industry has brought urgent environmental problems worldwide [1][2][3]. Air pollutants released from human production and life not only induce severe health problems, from respiratory illnesses to cardiovascular diseases but also have adverse impacts on ecosystems and transportation [4,5]. In the past several years, as one of the fastest developing countries in the world, China has also suffered a series of air pollution issues, such as acid rain, sandstorm, fine particulate matter (PM 2.5 ), and ozone (O 3 ) [6][7][8][9]. The first two issues have been controlled through the unremitting efforts of the government and the people of China. However, the air pollution issues of PM 2.5 and O 3 are still to be settled.
China has issued a series of control measures to mitigate PM 2.5 pollution, such as the Air Pollution Prevention and Control Action Plan (APPCAP) and emergency measures in autumn and winter acts. Moreover, a large amount of human and financial resources has been put into mitigating PM 2.5 pollution. The project for finding out the cause and control methods of heavy air pollution was supported by the prime minister's fund and was conducted from 2017-2019. Moreover, twenty-eight science teams were stationed at Beijing-Tianjin-Hebei and its surrounding areas to carry on the on-site work and help the local government to improve air quality. However, with the sharp decrease in PM 2.5 levels, O 3 concentration had an upward trend. Thus, it is of great importance to analyze Kendall's tau coefficient was used to present the coefficient relationships between factors. In addition, the variation trends of the concentration of PM2.5 and O3 and the coefficient relationships of meteorological conditions with O3 were also illustrated. The cluster of backward flow trajectories and air pollution of O3 and PM2.5 were used to describe the main transmission routes of the pollutants of the pollution processes in the autumn-winter seasons.

Sampling Position, Period, and Samples Collection
Xingtai is a typical industrial city, which has lots of glass, cement, and coking industries. Xingtai is also a northern city dominated by a north wind. Thus, many industrial plants are located in the south county of Shahe. In order to study the effects of transfer, emission, and weather conditions, three sampling sites were located in the main city (Yizhong Station, YZ) and the upwind (Neiqiu Station, NQ) and downwind Shahe Station, SH) direction of the main city during the autumn-winter season (Seen Figure 1). Sampling points all lay within 1.5-15 m from the ground in strict accordance with the provisions of the National Technical Specification for Layout of Ambient Air Quality Monitoring Points HJ 664-2013. The sampling periods were two autumn-winter seasons which were from 15 October 2017 to 30 January 2018 and 15 October 2018 to 30 January 2019. Daily PM2.5 samples were collected continuously from 09:00 to 08:30 of the next day, and the sampling time was 23.5h. Six hundred forty-eight film samples contained quartz and polypropylene were obtained for one autumn-winter season.
Two high-volume samplers (TH-150C, Tianhong Instrument Co. Ltd., Wuhan, China) were used to collect PM2.5 samples. Before the sampling process, the samplers were The sampling periods were two autumn-winter seasons which were from 15 October 2017 to 30 January 2018 and 15 October 2018 to 30 January 2019. Daily PM 2.5 samples were collected continuously from 09:00 to 08:30 of the next day, and the sampling time was 23.5 h. Six hundred forty-eight film samples contained quartz and polypropylene were obtained for one autumn-winter season.
Two high-volume samplers (TH-150C, Tianhong Instrument Co. Ltd., Wuhan, China) were used to collect PM 2.5 samples. Before the sampling process, the samplers were calibrated, the flow-rate range of the samplers was from 60 to 150 L/min (prolongable), with an accuracy of ±2.5%, and the relative error of the flow-rate of less than 2%. In this sampling process, the flow rate was set as 100 L·min −1 ; quartz and polypropylene films were loaded to capture PM 2.5 . After sampling, the filter samples were placed in the refrigerator at −4 • C for preservation.

Quality Assurance and Quality Control of Sampling
The quartz films were first calcined at 450 • C for 5 h in order to remove the organics and other impurities on the films. And then, the films were packed into aluminum foil papers and placed in a constant temperature and humidity chamber of 25 • C and 50 ± 5% relative humidity for 24 h. Then the films were sealed in film boxes at the temperature of −20 • C. Three blank films were reserved as the blank samples to correct the data of each sample in order to guarantee the accuracy and reliability of the analyzed data of the collected samples.

Carbonaceous Species
A circular quartz filter with an area of 0.558 cm 2 was used to determine organic carbon (OC) and elemental carbon (EC) concentrations by a thermal/optical carbon analyzer (Model 2008; Desert Research Institute, Reno, Nevada USA).
The background contamination was regularly monitored by blank tests, which were used to validate and correct the corresponding data. Calibration of the analyzer was done before and after sample analysis every day. The first sample was analyzed every ten samples again, and the precision should be less than 5%.

Inorganic Element
After the digestion process, metal components, such as Li, Be, Mg, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Mo, Cd, Sn, Sb, Ba, Hg, Pb, Bi, Ca, K, Mg, Na, were determined using Inductively Coupled Plasma-Mass Spectrometry (7700 Series ICP-MS, Agilent Technologies Inc., PaloAlto CA, USA). In addition, the concentrations of Al and Si were determined by Inductively Coupled Plasma Atomic Emission Spectrometry (8300 ICP-AES, PerkinElmer company, Boston, Massachusetts, USA)

Online Data Source
Air quality data were obtained from China's air quality online monitoring and analysis platform (http://www/aqistudy.cn/historydata, accessed on 27 February 2021), and meteorological data were obtained from the National Oceanic and Atmospheric Administration (ftp://arlftp.arlhq.noaa.gov/pub/archives/gdas1/, accessed on 27 February 2021) which are all open-source database for research.

Analysis of Secondary Pollution
The secondary formation is an important reason for PM 2.5 pollution. Sulfate and nitrate are the main productions by the secondary formation in PM 2.5 , and the components of which are related to the oxidation efficiency of SO 2 and NO 2. Usually, the secondary formation rates of SO 2 and NO 2 are characterized by SOR (sulfate oxidation rate) and NOR (nitrate oxidation rate), which can be calculated by the following equations: Atmosphere 2021, 12, 569 where (SO 4 2− ) and (NO 3 − ) are concentrations in PM 2.5 , µg/m 3 ; (SO 2 ) and (NO 2 ) are the concentrations of the gas phase, µg/m 3 .
Secondary organic carbon in the atmosphere is formed by photochemical reactions or gas-particle conversion of volatile and semi-volatile organic compounds. The degree of secondary carbon pollution can be characterized by indicators such as OC/EC and SOC/OC. The higher the ratio represents, the more serious secondary pollution. The secondary OC (SOC) concentrations were determined by the EC tracer method following Equations (3) and (4).
where POC, SOC, and OC represent the estimated primary OC, secondary OC, and measured total OC, respectively. (OC/EC) min is the minimum OC/EC ratio in each sampling period. Moreover, the value of the OC/EC ratio can also reflect the different combustion emission sources. It was suggested by the studies, the OC/EC ratio of vehicle exhaust emissions is 2.5-5.0, coal combustion is 5.0-10.5, wood-burning is 16.8-40.0, biomass burning is 7.7 [26][27][28].

Back Trajectory and Clustering Analysis
The 48 h backward trajectories of air mass during the pollution processes were investigated by the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) (http://ready.arl.noaa.gov/HYSPLIT.php, accessed on 27 February 2021) model developed by the National Oceanic and Atmosphere Administration (NOAA). The model started at a height of 100 m above ground level (AGL) with a time interval of one hour. Based on the results of the backward trajectory analysis, the trajectories and pollution concentrations were used to cluster analysis by MeteoInfoMap, which is an open-source geographic information system and scientific computing environment software.
Cluster analysis is an objective classification method to study multiple elements (or variables). It looks for a statistical quantity that can objectively reflect the distance relationship between samples and then divides samples into several categories according to the statistical quantity. The clustering method based on airflow trajectory is to group a large number of air flow tracks according to their spatial similarity (transmission velocity and direction). In this study, the Angle Distance algorithm provided by TrajStat software was used to cluster the airflow trajectory, and total spatial variance (TSV) was used to judge the classification quality. The principle is as follows: the TSV of the first several classification steps increases rapidly and then increases slowly. When the categories are divided to a certain number, TSV increases rapidly again, indicating that the merged categories are different. The classification merger is over, and the categories before the merger are the classification results. The average trajectories of these categories are the main flow trajectories of the target point in the analysis periods. two autumn-winter seasons are shown in Figure 2, which shows that the study periods were dominated by polluted weather, specifically, the weather with AQI above 100 (light pollution) accounted for 65.7% and 64.8% in the two autumn-winter seasons, respectively. presented in Table 1, and the concentrations of other online air pollutants are also shown in this table. The mean concentrations of PM2.5, NO2, SO2, and O3 were 125.0, 60, 32, and 51 μg/m 3 in the autumn-winter season of 2017-2018 and those were 112.1, 64, 31, and 40 μg/m 3 in that of 2018-2019. Compared to the autumn-winter season of 2017-2018, the concentrations of PM2.5 and SO2 in that of 2018-2019 had a slight decrease. However, the concentrations of NO2 and O3 had a little increase. The air quality index (AQI) of the two autumn-winter seasons are shown in Figure 2, which shows that the study periods were dominated by polluted weather, specifically, the weather with AQI above 100 (light pollution) accounted for 65.7% and 64.8% in the two autumn-winter seasons, respectively.  Water-soluble inorganic ions are important components of PM2.5 which mainly include NO3 − , SO4 2− , NH4 + , K + , Ca 2+ , and Cl − [29,30]. These inorganic ions play an important role in the formation of PM2.5. On average, the most abundant water-soluble species in and NH 4 + were all similar, which suggested that (NH 4 ) 2 SO 4 was the main formation of SO 4 2− and NH 4 + in PM 2.5 . Furthermore, it also can be seen that the mass concentration of NO 3 − was more than that of SO 4 2− , which suggests that the pollution style of Xingtai was of nitrate, indicating that the usage control of carbon reduced SO 2 emissions effectively, however, NO x emissions of vehicles and gas-fired boilers should be controlled further.

Results and Discussion
Carbonaceous species, which mainly include EC and OC, were found to contribute significantly to the formation of fine particles. EC is one of the main light-absorbing species in fine particles and also is the gas to particle reaction medium for SO 2 and NO x [31].
Moreover, EC is a good indicator of primary anthropogenic pollutants [28]. The source of OC is more complex. Besides the primary emissions from industrial production, fuel combustion, and natural sources, there is secondary organic carbon (SOC) produced by photochemical reactions of gaseous precursors in the atmosphere [32].  Table 1 shows the mass concentrations of PM 2.5 and its major chemical components in the three monitoring stations. Neiqiu and Shahe are two counties located in the upwind and downwind directions of the main urban city of Xingtai during the autumn-winter season. Among the three sampling sites, Shahe is an important industry county, which had many glass, ceramics, and coking industries. In general, as seen in Table 1, compared to the year 2017-2018, the mass concentration of PM 2.5 in the three sampling sites of YZ, NQ, and SH all decreased by 13.7%, 4.3%, and 12.6%, respectively, in the year 2018-2019. However, both in the two autumn-winter seasons, the mass concentration of PM 2.5 in SH was the highest, which could be attributed to more agminated factories. Table 1 also shows that all the PM 2.5 components have a similar pattern. The average concentrations of the total water-soluble ions at YZ, NQ, and SH were 52.7, 48.3, and 55.6 µg/m 3 in the autumn-winter season of 2017-2018, and those were 42.2, 47.4, and 45.5 µg/m 3 in that of 2018-2019, the decrease ratios were 19.9%, 1.9%, and 18.2%, respectively. Moreover, the decreasing rates of ion concentrations were calculated, as for SNA, and the decrease ratios of YZ and SH were all larger than those of NQ. It can be speculated that NQ was easily affected by the pollution transport. Moreover, it was to note that the concentration of NO 3 − was increased at NQ, which all suggested that nitrate pollution was a severe air pollution issue.
As seen in Table 1 and EC values of YZ and SH all had a slight decrease in that of 2018-2019. However, as for NQ, OC value was increased, and EC value kept steady. It could be concluded that in autumn-winter seasons, the additional fossil fuels clearly lead to an increase in OC emissions, and under the unfavorable meteorological conditions, the higher concentrations of OC may be precursors and attributed to the increase in secondary formations [35,36].
The mass concentrations of Cl − , K + , and Ca 2+ were relatively low in the total watersoluble inorganic ions. K + could be used to trace biomass emissions and fireworks emissions [37]. It was suggested that the mass concentration of K + in the autumn-winter season of 2018-2019 was lower than that of 2017-2018. It was known that the Administrative Measures of Xingtai City on Banning Open Burning (Trial) was released in 2018-2019, and biomass burning has been further controlled effectively so that the mass concentration of K + of the three sampling sites all had decreased. Chloride ions (Cl − ) mainly come from coal combustion [38], and the mass concentrations of Cl − in YZ and SH all decreased in the autumn-winter season of 2017-2018 compared to those of 2018-2019, suggesting that the coal-fired combustion was further controlled in 2018-2019 compared to 2017-2018. Calcium ions (Ca 2+ ) mainly come from crustal-derived ions and are indicators of soil-related sources [39]. As shown in Table 1, the concentration of Ca 2+ in the two autumn-winter seasons were similar, suggesting that those seasons suffered a similar sandstorm situation.

Chemical Components Analysis at Different PM 2.5 Concentration Levels
Among the normal six-item air pollutants, CO is not an active photochemical and has a chemical life of about several months. In a short period, CO can be regarded as an inert pollution tracer, mainly controlled by meteorological parameters [40][41][42][43]. The ratios of PM 2.5 /CO, SO 2 /CO, NO 2 /CO, and O 3 /CO were calculated under different AQI levels to understand the relative contribution of the chemical processes to PM 2.5 .
It can be seen in Table 2, the concentration ratios of PM 2.5 /CO and SO 2 /CO did not change obviously with different PM 2.5 concentration levels in both years. However, the ratio of NO 2 /CO and O 3 /CO all decreased with an increase in PM 2.5 concentration levels, suggesting that on heavy pollution days, the chemical conversion processes of the pollutants happened. NO 2 acted as reactants to produce the chemical reactions and forming more sulfates and nitrates, and O 3 formation process was restrained. It can be seen in Table 2 that the concentration ratio of O 3 /CO had an obvious decrease with the increase in the concentration ratio of PM 2.5 /CO. Thus, it can be speculated that on heavy pollution days, the concentration of PM 2.5 and O 3 had a negative correlation because of the chemical conversion processes, and O 3 could act as an important oxidizing agent. On the other hand, photochemical production of O 3 decreased during the occurrences of haze events which also attribute to the decrease in O 3 /CO.  25 24 In order to analyze the chemical components of PM 2.5 at different pollution levels, in this study, we further split the days into five levels according to the concentration of PM 2.5 as shown in Figure 3. The columns presented the average concentrations of PM 2.5 at their respective pollution levels. It can be seen in Figure 3, the concentration of Ca 2+ and K + are at low levels, suggesting that sand storms and biomass burning had fewer contributions to the concentration of PM 2.5 . The concentration of Cl − stays stable with the increase in pollution level, indicating that the contribution of coal combustion to the air pollution did not further increase with the deepening of the pollution. It is to note in Figure 3, the concentration variation trends of SO 4 2− and NH 4 + were similar, which increase with the increase in PM 2.5 concentration. It was indicated that (NH 4 ) 2 SO 4 was the important component of PM 2.5 . Considering that SO 4 2− also mainly come from coal combustion, the emission level of coal combustion did not increase due to the stable level of Cl − concentration. Thus, it could be concluded that the atmospheric oxidation increased with the deepening of pollution and making the secondary conversion level increase. NO 3 − was the main water-soluble ion of PM 2.5 , and the pollution style of the autumn-winter season in Xingtai was nitrate dominant. Moreover, the concentration of NO 3-increased sharply when the concentration of PM 2.5 was higher than 150 µg/m 3 , which also proved that the secondary conversion process was enhanced at the heavy pollution weather.  25 24 In order to analyze the chemical components of PM2.5 at different pollution levels, in this study, we further split the days into five levels according to the concentration of PM2.5 as shown in Figure 3. The columns presented the average concentrations of PM2.5 at their respective pollution levels. It can be seen in Figure 3, the concentration of Ca 2+ and K + are at low levels, suggesting that sand storms and biomass burning had fewer contributions to the concentration of PM2.5. The concentration of Cl − stays stable with the increase in pollution level, indicating that the contribution of coal combustion to the air pollution did not further increase with the deepening of the pollution. It is to note in Figure 3, the concentration variation trends of SO4 2− and NH4 + were similar, which increase with the increase in PM2.5 concentration. It was indicated that (NH4)2SO4 was the important component of PM2.5. Considering that SO4 2− also mainly come from coal combustion, the emission level of coal combustion did not increase due to the stable level of Cl − concentration. Thus, it could be concluded that the atmospheric oxidation increased with the deepening of pollution and making the secondary conversion level increase. NO3 − was the main watersoluble ion of PM2.5, and the pollution style of the autumn-winter season in Xingtai was nitrate dominant. Moreover, the concentration of NO3-increased sharply when the concentration of PM2.5 was higher than 150 μg/m 3 , which also proved that the secondary conversion process was enhanced at the heavy pollution weather.  creased, the secondary conversion rate increased. OC/EC level was higher at the high concentration level of PM2.5, suggesting that chemical conversion was enhanced during high pollution weather. The concentration ratios of NO3 − /SO4 2− are also shown in Figure 4, which indicates that the concentration ratios of NO3 − /SO4 2− in the autumn-winter season of 2018-2019 were higher than those of 2017-2018, which also proves the importance of NOx pollution control.

Impacts of Meteorological Parameters and Secondary Formation
Fine particulate matter (PM2.5) and O3 are the main substances causing air pollution in China, and it was testified that the combined pollution of PM2.5 and O3 were affected by the meteorological parameters and secondary formation [43][44][45].

Impacts of Meteorological Parameters on the Formation of PM2.5 and O3
The online concentrations of PM2.5 and O3 in the two autumn-winter seasons are shown in Figure 5, the average online concentration of PM2.5 and O3 were 100 and 51 μg/m 3 at the autumn-winter season of 2017-2018, and those were 100 and 40 μg/m 3 in the autumn-winter season of 2018-2019. Thus, it could be concluded that PM2.5 was the main air pollutant in the autumn-winter season in Xingtai. As shown in Figure 5, the concentrations of PM2.5 and O3 had different variation tendencies. Thus, it is speculated that the meteorological parameters and secondary formation had different effects on the formation of PM2.5 and O3.

Impacts of Meteorological Parameters and Secondary Formation
Fine particulate matter (PM 2.5 ) and O 3 are the main substances causing air pollution in China, and it was testified that the combined pollution of PM 2.5 and O 3 were affected by the meteorological parameters and secondary formation [43][44][45] Figure 5, the concentrations of PM 2.5 and O 3 had different variation tendencies. Thus, it is speculated that the meteorological parameters and secondary formation had different effects on the formation of PM 2.5 and O 3 .
The meteorological parameters of the two autumn-winter seasons are shown in Figure 6. The average wind speed, temperature, and relative humidity were 3.0 m/s, 4 • C, and 49.9% in the autumn-winter season of 2017-2018, and those were 2.9 m/s, 4 • C, and 47.7% in the autumn-winter season of 2018-2019. Thus, it could be concluded that the meteorological conditions of the two autumn-winter seasons were similar, which all had low wind speed and high relative humidity. It could be concluded from Figure 6, high humidity and low wind speed could promote the formation of air pollution. It was because in the autumn-winter seasons, low temperature and low wind speed caused the accumulation of pollution. Moreover, the high relative humidity increased the reaction rate of the heterogeneous and liquid phase reactions, which was in favor of the pollution process. In addition, it also can be seen in Figure 6, that wind has obvious effects on removing pollution. Thus, it could be concluded that the meteorological conditions of the two autumn-winter seasons were similar, which all had low wind speed and high relative humidity. It could be concluded from Figure 6, high humidity and low wind speed could promote the formation of air pollution. It was because in the autumn-winter seasons, low temperature and low wind speed caused the accumulation of pollution. Moreover, the high relative humidity increased the reaction rate of the heterogeneous and liquid phase reactions, which was in favor of the pollution process. In addition, it also can be seen in Figure 6, that wind has obvious effects on removing pollution. In order to explore the correlations between the concentrations of PM2.5, O3, and the meteorological parameters, Kendall's Tau coefficients were calculated, and the calculation results are shown in Table 3. It can be seen in Table 3, both PM2.5 and O3 had a weak correlation with the average air pressure. It can be seen that wind speed had a positive correlation with the concentration of O3 and had a negative correlation with the concentration of PM2.5. Thus, it is suggested that wind had the opposite effects on the formation of PM2.5 In order to explore the correlations between the concentrations of PM 2.5 , O 3 , and the meteorological parameters, Kendall's Tau coefficients were calculated, and the calculation results are shown in Table 3. It can be seen in Table 3, both PM 2.5 and O 3 had a weak correlation with the average air pressure. It can be seen that wind speed had a positive correlation with the concentration of O 3 and had a negative correlation with the concentration of PM 2.5 . Thus, it is suggested that wind had the opposite effects on the formation of PM 2.5 and O 3 . The formation of O 3 mainly depended on the secondary formation of the emission pollutants, and it was speculated that wind could increase the disturbance of the chemical matters in the atmosphere, further increase the reactants contacting probability of O 3. As for PM 2.5 , nucleation is the main step for the formation of PM 2.5 . However, high wind speed increases the disturbance of airflow, inhibiting the forming of the nucleus. It also can be seen in Table 3, the temperature had a higher Kendall's Tau coefficient with the concentration of O 3 than that of PM 2.5 , suggesting that O 3 is affected more easily by temperature. Higher temperatures were beneficial for the photochemical reaction to form O 3 , and went against the coagulating effects for forming PM 2.5 . The average relative humidity also had a positive correlation with the concentration of PM 2.5 , which demonstrated that water played an important role in the formation of PM 2.5 , because water molecules are an important component in the nucleation process.  Table 3, the temperature and relative humidity have more effects on the formation of PM 2.5 and O 3 , thus in order to study the relationships between the secondary aerosols and PM 2.5 concentration, O 3 concentration, relative humidity, and temperature, the correlations of them at the two autumn-winter seasons are shown in Figures 7 and 8, respectively. Kendall's tau coefficient between NOR, SOR, and the temperature, relative humidity concentration of PM 2.5 , O 3 are shown in Table 4.  Table 3, the temperature and relative humidity have more effects on the formation of PM2.5 and O3, thus in order to study the relationships between the secondary aerosols and PM2.5 concentration, O3 concentration, relative humidity, and temperature, the correlations of them at the two autumn-winter seasons are shown in Figure 7 and Figure 8, respectively. Kendall's tau coefficient between NOR, SOR, and the temperature, relative humidity concentration of PM2.5, O3 are shown in Table 4.   As shown in Figures 7 and 8, the relationships of NOR and SOR against the temperature present a "U" shape. The values of NOR and SOR show small values at a temperature around 4 °C. It was speculated that the oxidation processes of SO2 and NO2 could be conducted either in the gaseous phase of the atmosphere or the interface between the gas phase and particle phase of the nuclei, and higher temperature benefited oxidation reaction in the gas phase. However, lower temperature benefited the nucleation process of the pollutants in the atmosphere, which provided larger surface areas for the inter-phase oxidation reactions. Thus, when the temperature was around 4 °C, gas-phase oxidation and gas-solid-phase oxidation were also not promoted, and NOR and SOR were relatively low. As shown in Figures 7 and 8, the relationships of NOR and SOR against the temperature present a "U" shape. The values of NOR and SOR show small values at a temperature around 4 • C. It was speculated that the oxidation processes of SO 2 and NO 2 could be conducted either in the gaseous phase of the atmosphere or the interface between the gas phase and particle phase of the nuclei, and higher temperature benefited oxidation reaction in the gas phase. However, lower temperature benefited the nucleation process of the pollutants in the atmosphere, which provided larger surface areas for the inter-phase oxidation reactions. Thus, when the temperature was around 4 • C, gas-phase oxidation and gas-solid-phase oxidation were also not promoted, and NOR and SOR were relatively low.
It can be seen in Figures 7 and 8, the relative humidity (RH) plays an important role in the formation of nitrates and sulfates. Thus, NOR and SOR increase with the increase in RH. It was suggested by the previous studies, RH could promote the formation of NO 3-and SO 4 2− in two ways: one way was that original sulfate and nitrate could adsorb water to enlarge the particles' surfaces, and lager particle surfaces also promote the heterogeneous reaction to form NO 3 − and SO 4 2− ; the other way was that higher RH could decrease the viscosity of the surface of the particles, which increased the conversion rate of gas precursors to particulate matter kinetically, resulting in more pollutant converting to particulate state, which promotes heterogeneous reactions [46].
By calculation, Kendall's tau coefficients were 0.440 and 0.295 for NOR and SOR to the concentration of PM 2.5 , respectively in the autumn-winter season of 2017-2018, and those were 0.252 and 0.205 in that of 2018-2019 (As seen in Table 4). NOR and SOR showed a positive correlation with the concentration of PM 2.5 , indicating that the secondary formation contributes to the formation of PM 2.5 , and the promotion effects were not very strong, which suggested that except for the secondary formation, many other factors all affected the formation of PM 2.5 pollution. The Kendall's tau coefficients were −0.

Identification of Potential Transported Sources
Regional transport is another important reason for the formation of air pollution. Thus, the meteorological conditions and transport of PM 2.5 and O 3 during the pollution processes were studied. As shown in Table 5, there were six pollution processes during the sampling periods in 2017-2018, and the information on the pollution processes is shown in Table 5. The wind roses of the six pollution periods and the whole sampling period at the autumn-winter season of 2017-2018 are shown in Figure 9. It can be seen in Figure 9A, the predominant wind direction in the autumn-winter season was northwest. However, the predominant wind directions during the pollution periods were southeast and north-east directions, as shown in Figure 9B. Thus, it can be suggested that the pollution of Xingtai is mainly affected by the east cities at Shandong and Hebei province. The clustering analysis of the backward flow trajectory of the six typical pollution processes was conducted, and the clustering results are shown in Figure 10. It can be seen that all the pollution processes were affected by the airflow coming from the east. In order to make clear the pollution concentrations in each airflow trajectory, three pollution processes were selected, and the pollutants (O3 and PM2.5) contribution ratio is shown in Figure 11. As shown in Figure 11, O3 and PM2.5 usually have different main transport paths. The main PM2.5 transport paths of the pollution processes I, III, and VI were path 2, path 2, and path 3, respectively; however, the main O3 transport paths of those pollution processes were path 1, path 4, and path 3, respectively. It could be concluded that the main pollution transport paths of O3 in Xingtai were all from the east. Thus, O3 pollution of Xingtai in winter was easily affected by Shandong province except for the Beijing-Tianjin-Hebei area. As seen in Figure 9B, the predominant wind direction in Xingtai is northwest. However, among the main pollution transport paths of the pollutants, only in pollution process I, the main PM2.5 pollution was transported from the northwest. Thus, it was suggested that the pollution emissions of the cities located in the east of Xingtai had more effects on the pollution processes of Xingtai. It can be considered that more focus should be paid to the industrial restructuring and energy structure optimization of the cities at the east of Beijing-Tianjin-Hebei area to control PM2.5 and O3 pollution synergistically. Figure 10. The clustering results of the six pollution processes, path 1, path 2, path 3 (1,2,6); path 1, path 2, path 3, path 4, path 5 (3,5); path 1, path 2, path 3, path 4, path 5, path 6 (4). The clustering analysis of the backward flow trajectory of the six typical pollution processes was conducted, and the clustering results are shown in Figure 10. It can be seen that all the pollution processes were affected by the airflow coming from the east. In order to make clear the pollution concentrations in each airflow trajectory, three pollution processes were selected, and the pollutants (O 3 and PM 2.5 ) contribution ratio is shown in Figure 11. As shown in Figure 11, O 3 and PM 2.5 usually have different main transport paths. The main PM 2.5 transport paths of the pollution processes I, III, and VI were path 2, path 2, and path 3, respectively; however, the main O 3 transport paths of those pollution processes were path 1, path 4, and path 3, respectively. It could be concluded that the main pollution transport paths of O 3 in Xingtai were all from the east. Thus, O 3 pollution of Xingtai in winter was easily affected by Shandong province except for the Beijing-Tianjin-Hebei area. As seen in Figure 9B, the predominant wind direction in Xingtai is northwest. However, among the main pollution transport paths of the pollutants, only in pollution process I, the main PM 2.5 pollution was transported from the northwest. Thus, it was suggested that the pollution emissions of the cities located in the east of Xingtai had more effects on the pollution processes of Xingtai. It can be considered that more focus should be paid to the industrial restructuring and energy structure optimization of the cities at the east of Beijing-Tianjin-Hebei area to control PM 2.5 and O 3 pollution synergistically. The clustering analysis of the backward flow trajectory of the six typical pollution processes was conducted, and the clustering results are shown in Figure 10. It can be seen that all the pollution processes were affected by the airflow coming from the east. In order to make clear the pollution concentrations in each airflow trajectory, three pollution processes were selected, and the pollutants (O3 and PM2.5) contribution ratio is shown in Figure 11. As shown in Figure 11, O3 and PM2.5 usually have different main transport paths. The main PM2.5 transport paths of the pollution processes I, III, and VI were path 2, path 2, and path 3, respectively; however, the main O3 transport paths of those pollution processes were path 1, path 4, and path 3, respectively. It could be concluded that the main pollution transport paths of O3 in Xingtai were all from the east. Thus, O3 pollution of Xingtai in winter was easily affected by Shandong province except for the Beijing-Tianjin-Hebei area. As seen in Figure 9B, the predominant wind direction in Xingtai is northwest. However, among the main pollution transport paths of the pollutants, only in pollution process I, the main PM2.5 pollution was transported from the northwest. Thus, it was suggested that the pollution emissions of the cities located in the east of Xingtai had more effects on the pollution processes of Xingtai. It can be considered that more focus should be paid to the industrial restructuring and energy structure optimization of the cities at the east of Beijing-Tianjin-Hebei area to control PM2.5 and O3 pollution synergistically. Figure 10. The clustering results of the six pollution processes, path 1, path 2, path 3 (1,2,6); path 1, path 2, path 3, path 4, path 5 (3,5); path 1, path 2, path 3, path 4, path 5, path 6 (4). Figure 10. The clustering results of the six pollution processes, path 1, path 2, path 3 (1,2,6); path 1, path 2, path 3, path 4, path 5 (3,5); path 1, path 2, path 3, path 4, path 5, path 6 (4).