Pollution Characteristics of PM 2 . 5 Aerosol during Haze Periods in Changchun , China

To study the pollution characteristics of PM2.5 during the haze period in October 2014, the hourly automatic monitoring data and 22-h atmospheric PM2.5 samples were collected in Changchun city. According to the statistical results of the hourly average PM2.5 concentration, heavy pollution occurred from October 13 to November 1 in 2014 in Changchun city. The daily concentration of PM2.5 during this period was 53.18 μg m to 450.69 μg m. In addition, the PM2.5 concentration was higher at night and lower during the daytime on haze days, which is affected by pollutant emission and meteorological conditions. The Pearson correlation coefficient between PM2.5 and CO was high (r = 0.745), which implies that the concentration suddenly increased, probably because of the combustion of fossil fuels and other organic matter. The meteorological condition investigation shows that the wind speed, temperature and pressure are low and that the RH value is relatively high on haze days. In addition, stable weather during haze days makes the pollution heavier. The analysis of water-soluble inorganic ions (WSIIs) suggests that WSIIs constituted up to 21–56% of PM2.5, and the total ions, anions and cations showed a significant correlation. During the haze days, the NO3/SO4 ratio was substantial (3.7–4.6), except at the CP and JZP stations, which suggests that the pollutants in the atmosphere are from a mixture of stationary emission and mobile emission. The NOR and SOR values were more than 0.1 at most sampling stations, so serious secondary pollution occurs in the atmosphere on haze days, and secondary ion formation is enhanced by the atmospheric conditions and emissions of gaseous SO2 and NO2.


INTRODUCTION
Haze weather has increased in recent years; fine particulate matter (PM 2.5 ) and inhalable particulate matter (PM 10 ) have received increasingly more attention.Particulate matter (PM) significantly affects visibility, human health, climate change and the ecosystem.In particular, fine particulate matter can penetrate people's respiratory systems and cause many fatal effects on the human body.
PM 2.5 observations and research have been performed in many countries worldwide.PM 2.5 is composed of various chemical components such as carbonaceous components (organic carbon and elemental carbon), water-soluble ions, metal elements and organic matter (Kulkarni and Venkataraman, 2000;Park et al., 2002).Many scholars have analyzed the composition of PM 2.5 to perform source apportionments such as biomass burning, traffic emission, coal combustion, soil dust, and industrial emissions.
Changchun is an industrial city in the northeastern plain of China, and it has a temperate continental monsoon climate.The temperature increases and precipitation decreases from east to west.The climate is dry and windy in spring, hot and humid in summer, invigorating in autumn, and cold in winter.The annual average temperature in Changchun is 4.8°C, the highest temperature is 39.5°C, and the lowest temperature is -39.8°C.The annual average precipitation ranges from 522 mm to 615 mm.
Water-soluble inorganic ions (WSIIs) account for at least one-third of the fine particulate matter.The WSII concentrations distinctly vary in different seasons.Typically, SO 4 2-, NO 3 -and NH 4 + are the major WSII species in PM 2.5 and the most abundant component in secondary ion species (Yin et al., 2012).The particulate compositions during hazy weather and normal weather were analyzed in 2009 (Tan et al., 2009).The SO 4 2-, NH 4 + and NO 3 -concentrations significantly increased on haze days.Some studies have investigated the sources of PM 2.5 in many cities of China such as the Beijing-Tianjin-Hebei region, the Yangtze River delta and the Pearl River Delta (Wang et al., 2008;Gu et al., 2011;Wang et al., 2012;Cheng et al., 2013;Dai et al., 2013;Huang et al., 2013) However, fewer studies on PM 2.5 in northeast China have been performed.To better understand the temporal and spatial distribution and the component characteristics of PM 2.5 , the hourly automatic monitoring data in 2014 and 22-h atmospheric PM 2.5 samples were collected in Changchun during the haze period in October 2014.The PM 2.5 concentration and its WSII components during haze days were analyzed in this paper.

Sites and Sampling
Changchun The PM 2.5 samples in this study were collected at nine Changchun environmental monitoring stations, excluding SWZ.A medium-volume TSP/PM 10 /PM 2.5 sampler with Teflon filters (KC-120 Particulate Sampler, China) was used to collect particulate samples for the WSII analysis on October 21, 2014.There were 9 samples with a 22-h sampling duration.Before and after sampling, the filters were conditioned at a temperature of (25 ± 1)°C and a relative humidity of (35% ± 2%) for 24 h and weighed using an electronic balance with a detection limit of 0.01 mg.The PM 2.5 mass concentrations for each sampling were calculated from the weight and corresponding air volume.

Water-soluble Inorganic Ions Analysis
The PM 2.5 sample filters were submerged in a 50-mL colorimetric tube with 15 mL of pure water; the WSIIs from the samples were extracted by ultrasound for 10 min (KQ-700E Ultrasonic, China).The extract solutions were filtered with 0.45-μm filters and analyzed using an ion chromatograph.According to the peak retention time and peak height (or peak area) of ions in the mixed standard solution, the WSIIs in the samples were identified and quantified.Cations (Na + , NH 4 + , K + , Mg 2+ , and Ca 2+ ) in the extract solution were analyzed using a DX-120 Ion Chromatograph (Dionex Corporation, America); the anions (F -, Cl -, NO 3 -, and SO 4 2-) were analyzed using Metrohm-MIC-7 (Metrohm AG, Switzerland).

Mass Concentration of PM 2.5
The contour map produced using Surfer software shows the hourly variation in PM 2.5 concentrations during each week in 2014.The average PM 2.5 concentrations of 10 environmental monitoring stations are presented in Fig. 1.The horizontal coordinate is time, and the vertical coordinate represents 52 weeks.In 2014, each data point indicates the weekly average PM 2.5 concentration and is denoted with different colors.
The result in Fig. 1 shows that heavy pollution occurred from October 13 to November 1 in 2014 in Changchun city.The PM 2.5 concentration was high from 16:00 to 8:00 and had a peak value at 19:00.The air quality in May, June and September was relatively low.The haze period and normal-weather period are compared in Fig. 2. The haze period was from October 13 to November 1, and the normalweather period was from November 2 to November 10.The PM 2.5 concentrations on October 14,18,22,25,28,29,30,and 31 and November 1 were obviously higher than those on the other days and show an increasing trend.Hazy weather initially occurred every 3 days; then, a high-PM 2.5 pollution occurred for up to 5 days because of the concentration of pollutant; the PM 2.5 concentration reached a peak on October 30.The daily average concentration of PM 2.5 in this period ranged from 53.18 µg m -3 to 450.69 µg m -3 .According to the statistical results, 11 daily average values exceeded 150 µg m -3 , and 6 exceeded 250 µg m -3 .The PM 2.5 -to-PM 10 ratio shows that PM 2.5 is the main component of PM 10 , and the changing trend of the PM 2.5 concentration is similar to that of the PM 2.5 /PM 10 ratios.The PM 2.5 concentration and PM 2.5 /PM 10 decreased from November 2 to November 10, whereas PM 2.5-10 had no obvious variation, which indicates that the increase in the particulate matter concentration in air is directly related to PM 2.5 pollution, and a high PM 2.5 concentration can aggravate the haze.Further study is currently being performed to observe the hourly variation in PM 2.5 concentrations from October 13 to November 1.In Fig. 3, the horizontal coordinate is time, which ranges from 0:00 to 23:00, and each column represents the 10-90% quantile of concentration values in

Correlations between PM 2.5 and Gaseous Pollutants (SO 2 , NO X , CO and O 3 )
The Pearson correlation coefficient (SPSS, bivariate correlation) was used to investigate the relationship among SO 2 , NO x , CO, O 3 and PM 2.5 .This analysis was based on the hourly data of ten automatic monitoring sites in Changchun city from October 28 to November 1.We also performed two-sided tests, as shown in Table 1.
The analysis results show a strong correlation between PM 2.5 and CO (r = 0.745), which implies that the CO emission process is accompanied by the emission of fine particles.The sudden increase in concentration is probably because of the combustion of fossil fuels and other organic matter.However, the correlations among SO 2 , NO x , O 3 and PM 2.5 are slightly lower (PM 2.5 and SO 2 : r = -0.190;PM 2.5 and NO x : r = 0.214; PM 2.5 and O 3 : r = -0.280).Previous studies have shown that the oxidation of SO 2 and NO x forms sulfuric acid, nitric acid, and sulfate and nitrate aerosols.These species chemically react with one another in the presence of sunlight and produce toxic secondary pollutants such as O 3 and aerosols, which are major concerns for urban air quality (Xie et al., 2015).

Analysis of Meteorological Conditions
The mass concentration of PM 2.5 is closely related to the metrological conditions in the atmosphere during the heavy haze period.Investigating the meteorological factors can provide insight into the variations in the PM 2.5 mass concentrations.The major meteorological conditions in the haze period are shown in Fig. 4.
In Fig. 4, the variations in temperature and wind speed are similar.On October 13, 17, 21, 25, 28, and 31, the temperature and wind speed were notably low; on October 15, 19, 24 and 26, they were high.According to the PM 2.5 variation in Fig. 2, haze pollution occurred on October 14, 18, 22 and 25.Thus, low wind speed and temperature are not conducive to pollutant dispersion.Moreover, the high RH and low pressure on October 18, 25 and 31 indicate that high humidity and low air pressure can accelerate the development of hazy weather conditions.
To study the effect of atmospheric activity on the formation and development of hazy weather in Changchun, the 72-h backward and forward trajectories that began from Changchun (43.84N, 125.29E) at the arrival level of 500 m and mixed layer depth (Oct 29-31) were generated with the NOAA HYSPLIT model (http://ready.arl.noaa.gov/HYSPLIT_traj.php) in Figs. 5 and 6.The trajectories were calculated at the starting time of 00:00 on October 30.Based on the HYSPLIT model, the air flow ran at a notably low height from 18 pm October 28 to a relatively high height on October 31.The mixed layer depth in Changchun remained notably low (even less than 80 m) and caused the high PM 2.5 concentration on the 29, 30 and 31.Thus, stable weather on these days made the PM 2.5 pollution heavier.

Water-soluble Inorganic Ion (WSII) Compositions in PM 2.5
WSIIs account for 21-56% of PM 2.5 .The order is NO 3 The correlation analysis results of WSIIs are shown in Table 2. Na + , NH 4 + , K + , Mg 2+ and Ca 2+ are highly homologous with a significant correlation; F -, SO 4 2-and Cl -also have a high homology.The total ions, anions and cations show significant correlations and homology.The good correlation between K + and Cl -indicates that straw burning contributes to pollution in Changchun during haze days.

Ratio of NO 3 -/SO 4 2-
The mass concentrations of NH 4 + , SO 4 2-and NO 3 -, which were collected at nine sampling stations on haze days, and the NO 3 -/SO 4 2-ratios are shown in Fig. 7. Previous studies show that the major WSIIs in PM 2.5 are NH 4 + , SO 4 2-and NO 3 -, and these are also the most abundant components in the secondary ion species (Yue et al., 2015).The estimated NO x -to-SO x ratios from the emission of gasoline and diesel fuel burning are 13:1 and 8:1, respectively.The sulfur content in coal is approximately 1%, and the estimated emission NO x -to-SO x ratio is 1:2 from coal burning (Wang et 2005).Based on the characteristics of these two ratios, the values of NO 3 -/SO 4 2-and the stationary/mobile emission ratio were calculated and are shown in Table 3.
As shown in Fig. 4, the NO 3 -/SO 4 2-ratio in Changchun    0.5 2-3 3.5-5.5 5-8 6.5-10.5 8-13 was 3.7-4.6except at stations CP and JZP, which implies that pollutants are from the mixture of stationary emission and mobile emission, and stationary emission is the main pollution source; the ratio of stationary to mobile is approximately 6:4.This result occurred partly because of the beginning of coal combustion heating activities during October 2014 in Changchun.

Values of NOR and SOR
The nitrogen oxidation ratio (NOR) and sulfur oxidation ratio (SOR) values can reflect the degree of atmospheric conversion of NO 2 and SO 2 to NO 3 -and SO 4 2-, respectively.Higher SOR and NOR values indicate that more secondary aerosol particulates are converted from SO 2 and NO 2 , which are generated in gas or liquid phase reactions.In the primary pollutant, SOR is typically below 0.1, so it can be considered a cutoff value for the SO 2 secondary conversion.In the atmosphere, SO 2 is photochemically oxidized when SOR is above 0.1 (Ohta and Okita, 1990;Kaneyasu et al., 1995).
The SOR and NOR are defined as follows: The NOR and SOR in each sampling station are shown in Fig. 8.All SOR and NOR values were more than 0.1.The JYP sampling station had the highest SOR and NOR of 0.30 and 0.36, respectively.Thus, there is serious secondary pollution in the atmosphere during hazy weather.Moreover, the secondary ion formation progress is enhanced by the atmospheric condition and gaseous SO 2 and NO 2 emissions.

CONCLUSIONS
In this paper, the hourly automatic monitoring data and the 22-h atmospheric receptor PM 2.5 samples were collected in Changchun city to investigate the pollutant characteristics of PM 2.5 during the haze period in October 2014.
According to the statistical results of the hourly average PM concentration, heavy pollution occurred from October 13 to November 1 in 2014 in Changchun city.The daily average concentration of PM 2.5 during this period was 53.18 µg m -3 to 450.69 µg m -3 , and the PM 2.5 concentration was higher on October 14,18,22,25,28,29,30,and 31 and November 1 than on other days.The PM 2.5 -to-PM 10 ratio shows that the change in PM 2.5 concentration is similar to the change in PM 2.5 /PM 10 , which indicates that the increase in the mass concentration of particulate matter in air is directly related to the PM 2.5 pollution.The hourly variation in PM 2.5 concentration from October 13 to November 1 has a U-shaped trend and is higher at night and lower in the day time during haze days, which indicates that straw burning and coal combustion at night make an enormous contribution to pollution in Changchun city.In addition, the meteorological conditions are an additional factor that affects the PM 2.5 concentration during haze days.At night, the low temperature and unobvious pressure gradient are not conducive to the spread of pollutants to high altitudes.In addition, the inversion layer promotes the accumulation of pollutants.
The Pearson correlation coefficient is used to investigate the relationship among SO 2 , NO x , CO, O 3 and PM 2.5 .The analysis results show a strong correlation between PM 2.5 and CO (r = 0.745), which implies that the sudden increase in concentration is probably because of the combustion of fossil fuels and other organic matter.However, the correlations among SO 2 , NO x , O 3 and PM 2.5 are slightly lower (PM 2.5 and SO 2 : r = -0.190;PM 2.5 and NO x : r = 0.214; PM 2.5 and O 3 : r = -0.280).
Meteorological conditions were investigated from October 13 to November 10 in 2014 in Changchun.The results show that the low wind speed, temperature, and pressure are not conducive to pollutant dispersion.Meanwhile, a highhumidity environment promotes haze.The low mixed layer depth in Changchun caused the high PM 2.5 concentration on October 29, 30 and 31.Thus, stable weather during these days made the pollution heavier.
According to the WSII analysis of PM 2.5 , WSIIs constituted up to 21-56% of the PM 2.5 .The correlation analysis results of the nine WSIIs show that the total ions, anions and cations have significant correlations.Meanwhile, the NO 3 -/SO 4 2-ratio was substantial (3.7-4.6), except at stations CP and JZP.Pollutants are from the mixture of stationary emission and mobile emission, and the stationary emission is the main pollution source.NOR and SOR at each sampling station were more than 0.1.The JYP sampling station had high SOR and NOR values of 0.30 and 0.36, respectively, which indicates that there was serious secondary pollution in the atmosphere during the haze.

Fig. 1 .
Fig. 1.Hourly variation of PM 2.5 concentrations in each week in 2014.

Fig. 8 .
Fig. 8. Values of NOR and SOR at nine sampling stations.

Table 3 .
Diagnostic values for NO 3