Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment

Environmental Pollution xxx (2017) 1e9

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment* Lian Duan a, Na Cheng a, Guangli Xiu a, *, Fujiang Wang b, Ying Chen b a

State Environmental Protection Key Lab of Environmental Risk Assessment and Control on Chemical Processes, East China University of Science and Technology, Shanghai 200237, China Shanghai Key Laboratory of Atmospheric Particle Pollution Prevention, Department of Environmental Science & Engineering, Fudan University, Shanghai 200433, China b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 20 July 2016 Received in revised form 11 October 2016 Accepted 12 October 2016 Available online xxx

Total Suspended Particulate (TSP) samples were collected at Huaniao Island in northern East China Sea (ECS) from March 2012 to January 2013. Chemical analysis were conducted to measure the concentration of total particulate mercury (TPM) and speciated particulate mercury including HCl-soluble particulate mercury (HPM), elemental particulate mercury (EPM) and residual particulate mercury (RPM). The bromine (Br) and iodine (I) on particles were also detected. The mean concentration of TPM during the study period was 0.23 ± 0.15 ng m
3, while the obviously seasonal variation was found that the concentrations of TPM in spring, summer, fall and winter were 0.34 ± 0.20 ng m
3, 0.15 ± 0.03 ng m
3, 0.15 ± 0.05 ng m
3 and 0.27 ± 0.26 ng m
3, respectively. The statistically strong correlation of bromine and iodine to HPM was only found in spring with r ¼ 0.81 and 0.77 (p < 0.01), respectively. While the strongest correlations between EPM and bromine and iodine were found in winter with r ¼ 0.92 (Br) and 0.96 (I) (p < 0.01), respectively. The clustered 72-h backward trajectories of different seasons and the whole sampling period were categorized into 4 groups. In spring, the clusters passed a long distance across the East China Sea and brought about low concentration of mercury due to the deposition of mercury over the sea. The cluster of air mass across the sea had low concentration of HPM in winter, which suggested that the oxidation of mercury in winter might be related to other oxidants. During the whole sampling period, the air mass from the north of China contributed to the higher concentration of TPM in Huaniao Island. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Speciated mercury Bromine Iodine Long-range transportation

1. Introduction Mercury, with high toxicity and strong bioaccumulation, poses a great threat on human health and ecosystem (Lindberg and Stratton, 1998). Mercury is commonly emitted into the atmosphere from a variety of natural (e.g., surface waters, biomass burning and volcanoes) and anthropogenic sources (e.g. fossil-fuel combustion, metals manufacturing and incinerators) (Pirrone et al., 2010). Generally, mercury exists primarily in the atmosphere as three forms: gaseous elemental mercury (GEM, Hg0), gaseous oxidized mercury (GOM, Hg2þ) and total particulate mercury (TPM)

*

This paper has been recommended for acceptance by Yuan Wang. * Corresponding author. E-mail address: [email protected] (G. Xiu).

(Lindberg and Stratton, 1998). As the predominant form of mercury in the atmosphere, GEM has a relatively long atmospheric lifetime (0.5e2 years) due to its high volatility and stability and can be transported on the global scale (thousands of kilometers). While GOM and TPM, typically released from the combustion processes (i.e., coal-fired energy generating units), usually transport across much shorter distances (tens to hundreds of kilometers) due to their physical and chemical properties (Choi et al., 2013; Kim et al., 2009; Lindberg and Stratton, 1998; Schroeder and Munthe, 1998). Although atmospheric TPM accounts for less than 10% of the total atmospheric Hg (Xiu et al., 2005), it plays an important role in mercury deposition. Based on the previous studies (Duan et al., 2016; Xiu et al., 2005, 2009), the sequential extraction methods for mercury species in particles was adopted to define the three important forms including HPM (HCl-soluble particle-phase

http://dx.doi.org/10.1016/j.envpol.2016.10.103 0269-7491/© 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Duan, L., et al., Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment, Environmental Pollution (2017), http:// dx.doi.org/10.1016/j.envpol.2016.10.103

2

L. Duan et al. / Environmental Pollution xxx (2017) 1e9

mercury), EPM (elemental particle-phase mercury) and RPM (residual particle-phase mercury), in order to reveal different deposition mechanisms during transportation of elemental mercury in the atmosphere. HPM is supposed to consist of such oxidized mercury as HgO, HgSO4, HgCl2, HgBr2, HgI2, while EPM were usually formed through condensation of elemental mercury on particles. RPM might contain HgS, HgSe, organic mercury, and some other species in the core of the particles. It has been reported that halogen species play significant role in mercury deposition and were responsible for Atmospheric Mercury Depletion Events (AMDEs) (Lindberg et al., 2002; Schroeder et al., 1998). During AMDEs, the concentration of GEM rapidly decreased to a low enough level which is even below global background concentration, while the concentrations of GOM and TPM increased initially (Lindberg et al., 2002; Moore et al., 2013; Schroeder et al., 1998). AMDEs were originally thought to primarily occur at high-latitude regions or low temperature regions in the presence of reactive halogen species and sunlight. Recent studies (Duan et al., 2016; Gao et al., 2010; Xu et al., 2015) suggested that high level of halogens play the disproportionately large role in conversion from Hg0 to Hg2þ in the marine boundary layer (MBL) in coastal cities, thereby leading to rapid deposition of gaseous elemental mercury to the ocean body. Atmospheric mercury pollution must be one of the important problems in China because more than 50% of coal in the world was consumed there. Higher atmospheric mercury emission may result in elevated atmospheric Hg concentration and deposition flux in the neighbor regions of sources in China. In addition to the continental sources, the ocean was also recognized to be a special source of GEM and sink for oxidized mercury (i.e., gaseous oxidized mercury (GOM) and TPM) (Cheng et al., 2013, 2014). There are rich reactive bromine and iodine in ocean atmosphere which can potentially enhancing the oxidization of GEM to GOM, while the affinity of GOM with NaCl concentrated on sea salt aerosols over the ocean is also conductive to the partition of reactive mercury on particle surface (Rutter and Schauer, 2007). It therefore is important to study the potential impacts of marine atmosphere on the environmental behavior of atmospheric mercury. However, studies on the atmospheric TPM in the eastern coastal island remain limited. The objectives of this study were to investigate the levels, spatial distribution and seasonal variation of TPM in the atmosphere of East China Sea, and to ascertain the relationships between TPM and halogens (Br and I). Findings from this investigation are discussed in terms of better understanding the behavior and cycling of atmospheric TPM in the marine environment.

2. Materials and methods 2.1. Description of sampling site Sampling was conducted on Huaniao Island (122.67  E, 30.86 which is located in the northern East China Sea (ECS). Fig. 1 shows the location of sampling site which is located approximately 100 km to continental area in the Yangtze River estuary. The sampling site was not influenced by the local emission since only about 200 fishermen households live there. This site was representative of the frontline of ECS influenced by the airflow transported from eastern China, the details are given by previous publication (Wang et al., 2016). In order to compare the different characteristics of mercury between oceanic aerosols and continental aerosols, previous data collected in Xuhui District of Shanghai (121.26  E, 31.09  N) by our group was also adopted (Duan et al., 2016).  N)

2.2. Sample collection Total Suspended Particles (TSP) samples were collected by a high volume sampler (Thermo Scientific) loaded with acid-cleaned Whatman Grade 41 cellulose filter (20.3  25.4 cm). The flow - rate was controlled at 1130 L min
1 and lasted for 24 h. All filters were conditioned under constant temperature (20 ± 1  C) and a relative humidity (40± 1%) before and after sampling. The samples were packed and sealed in polyethylene plastic bags, and stored in a refrigerator at about
20  C until analysis.

2.3. Analytical method and QA/QC The total aerosol mass was determined gravimetrically through weighing the filters before and after sampling by an analytical balance (Sartotius, 2004, 0.01 mg, MP). One thirty-second (1/32) of each TSP sample or operational blank was digested with 5 mL of ultrapure HNO3 and 2 mL H2O2, 0.5 mL HF in CEM Mars Xpress microwave digestion system (PyNN Corporation,USA). The solutions were then transferred to a 50 mL flask and diluted to 50 mL with milli-Q water (18.2 MU cm
1), 10 mL of the solutions were used for Hg detection. The rest were then added with 1 mL per-sulfate (20% Na2S2O8) and 1 drop (about 0.05 mL) of silver nitrate (0.5% AgNO3 solution). The samples were heated at 45  C by a water bath for 10 min, and were cooled naturally down to ambient temperature, then stored for analysis of Br and I analysis. One thirty-second (1/32) was cut and extracted by 10 mL of 1 mol/L HCl and 0.5 mL 1% CuSO4, the solutions were identified as HCl-soluble particulate mercury (HPM); then 10 mL 2 mol/L HNO3 was added, the solutions were identified as elemental particulate mercury (EPM); finally the filters were digested following the above way, the solutions were then transferred to 25 mL flask and diluted to 25 mL with milli-Q water (18.2 MU cm
1), identified as residual particulate mercury (RPM). All the Hg samples were detected by cold-vapor atomic fluorescence spectrometry (CVAFS) (AFS-9130, China). The Br and I were determined using an Inductively Coupled Plasma Mass Spectroscopy (ICP-MS PerkinElmer, USA). Detailed procedures for analysis of Hg, Br and I were given in previous studies (Duan et al., 2016; Gao et al., 2010).

2.4. Trajectories cluster analysis 72-h backward trajectories of air mass to Huaniao Island (our observation site) were calculated during sampling period by using HYSPLIT-4 model (Draxler and Hess, 1998). Data came from NCEP/ NCAR Reanalysis meteorological database (http://arlftp.arlhq.noaa. gov/pub/archives/gdas1). The model was run four times per day at starting times of 00:00, 06:00, 12:00 and 18:00 UTC (08:00, 14:00, 20:00 and 02:00 LT-local time, respectively), and the starting height was set at 500 m. The method used in trajectory clustering was based on the GIS-based software TrajStat (http://www. meteothinker.com/TrajStatProduct.aspx) (Wang et al., 2009; Zhao et al., 2015).

2.5. Statistical analysis The correlation analysis was carried out by SPSS 17.0. The oneway ANOVA test was used to determine the difference in TPM concentrations among different seasons.

Please cite this article in press as: Duan, L., et al., Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment, Environmental Pollution (2017), http:// dx.doi.org/10.1016/j.envpol.2016.10.103

L. Duan et al. / Environmental Pollution xxx (2017) 1e9

3

Fig. 1. Location of sampling site on Huaniao Island, East China Sea.

some extent, the concentration was much higher than those in remote areas and urban sites in Europe and America. In general, the average concentration of TPM was lower than those in urban sites such as Shanghai, Beijing, Qingdao and Changchun because TPM was mostly affected by anthropogenic sources (e.g. automobile exhaust, coal combustion and exhaust from chemical factories) in urban areas while it was mainly attributed to sea salt and secondary aerosol over Huaniao Island (Wang et al., 2016).

3. Results and discussion 3.1. TPM concentrations The concentrations of particulate mercury were compared with those in other cities and summarized in Table 1. The average concentration of TPM was 0.23 ± 0.15 ng m
3 with a range of 0.09e0.99 ng m
3 which was much higher than the background value in northern hemisphere (<1.0e5.0 pg m
3). To

Table 1 Comparison of TPM concentration in Huaniao Island with those in other sites. Location

Time

Type

TPM(ng m
3)

Reference

Huaniao Island CN Shanghai CN Beijing CN Beijing CN Qingdao CN

Ocean Island Urban Urban Urban Urban

Waliguan CN Mt Changbai Mt Gongga Wisconsin USA Florida USA CHO Okinawa

Sep. 2007eAug.2008 Aug. 2005eJul. 2006 May. 2005eJul.2007 Summer, 1993e1995 Aug.eSep. 1995 Mar.eMay. 2004

0.23 ± 0.15 0.56 ± 0.22 0.57 1.18 ± 0.82 0.304 ± 0.229 0.194 ± 0.125 0.276 0.109 0.019 ± 0.018 0.077 ± 0.0136 0.023e0.032 0.0021e0.028 0.001e0.029 0.003

This study (Xiu et al., 2009) (Schleicher et al., 2015) (Wang et al., 2006) (Zhang et al., 2015)

Changchun CN

Mar. 2012eJan. 2013 Jul. 2004eApr. 2006 Jan.eApr. 2006 Jan.eDec. 2003 2008e2011 (Dust day) 2008e2011 (non-dust day) Jul. 1999eJan. 2000

Urban Suburban Remote Remote Remote Remote Urban/Rural Ocean Island

(Fang et al., 2004) (Fu et al., 2012) (Wan et al., 2009) (Fu et al., 2008) (Lamborg et al., 2000) (Graney et al., 2004) (Chand et al., 2008)

Please cite this article in press as: Duan, L., et al., Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment, Environmental Pollution (2017), http:// dx.doi.org/10.1016/j.envpol.2016.10.103

4

L. Duan et al. / Environmental Pollution xxx (2017) 1e9

Table 2 The correlation between mercury and halogens in different seasons. TPM

Spring Summer Fall Winter

HPM

EPM

RPM

Br

I

Br

I

Br

I

Br

I

0.95** 0.42 0.93** 0.90**

0.97** 0.07 0.94** 0.89**

0.81** 0.28 0.53
0.63

0.77** 0.21 0.49
0.54

0.84** 0.77* 0.39 0.92**

0.86** 0.56* 0.41 0.96**

0.90** 0.30 0.14 0.91**

0.98** 0.03 0.11 0.97**

Note: **: Correlation is significant at the 0.01 confidence level (2-tailed); *: Correlation is significant at the 0.05 confidence level (2-tailed).

Atmospheric particulate mercury is commonly emitted from local pollution sources or other sources through long range transportation, and rarely origins from natural emissions. However, it can be formed by reactive gaseous (oxidized) Hg through bounding with particles, particularly in the marine boundary layer (Feddersen et al., 2012). Several studies has found that 50%e60% of particulate mercury was bounded with coarse aerosols (1.1e5.8 mm) at both the coastal sites (Mao and Talbot, 2012) and marine sites (Feddersen et al., 2012), the sea salt particles must play the significant role for formation of TPM. Apart from direct emission, atmospheric particulate mercury might have two formation mechanisms: adsorption of gaseous mercury on particles and chemical gas - particle transformation (Xiu et al., 2005). Mercury in coarse particles might be generally associated with adsorption of gaseous element mercury in coarse particles, while mercury in fine particles were determined by the two mechanisms (Xiu et al., 2005). In this study, the average TPM/TSP is 4.94 mg g
1 with a range of 0.86e9.96 mg g
1, which is lower than the mass ratio in PM2.5 at Shanghai (7.74e11.28 mg g
1) (Bo et al., 2016), but higher than that in TSP at Shanghai (1.05e2.85 mg g
1) (Xiu et al., 2005). Compared with the mercury content in some materials like 0.046e4.8 mg g
1 in Chinese coal (Zheng et al., 2007) and
pez-Anto
n et al., 2007), the mercury 1.00e4.90 mg g
1 in fly ash (Lo in ambient airborne particles of this study is relatively higher than the source proflle. Therefore, the mercury must be accumulated during atmospheric transportation process. 3.2. Seasonal variability The seasonal concentrations of TPM and speciated particulate mercury were shown in Fig. 2a. There was a statistically significant difference in mercury concentration between different seasons by ANOVA analysis (p < 0.01). TPM varied obviously seasonally with 0.44 ± 0.25 ng m
3 in spring, 0.27 ± 0.21 ng m
3 in winter, 0.16 ± 0.04 ng m
3 in summer and fall. The phenomena that the concentration of TPM was lower in summer and fall might result from the following reasons:(1) the predominant wind was from northeast in summer, which was also seen from the cluster analysis in Section 3.4; (2) the wet precipitation of TPM was stronger due to more raining in summer; (3) the contribution of anthropogenic source to the aerosol over Huaniao Island was the lowest (18%) in fall (Wang et al., 2016). On the other hand, the higher concentration of TPM in winter could be explained by the strongest anthropogenic contribution from continental aerosols with the highest value of 76.6%, strongly influenced by the prevailing winds of East Asian Monsoon (Wang et al., 2016). However, it is surprising that the highest concentration of TPM was found in spring because the common dust pollution usually has lower mercury content. The possible cause is the geographic characteristics of Shanghai because some studies have illustrated that it could be associated with the atmospheric mercury depletion events which occurred in marine and coastal environment during springtime (Ebinghaus et al., 2002; Ferrari et al., 2008; Moore et al.,

2013). Another possible explanation is the result of the aging process of aerosol which is suspected to input mercury through dusts from western China though there was not enough evidence. It should not be neglected that the contribution of ship exhaust emission was suspected to be higher in spring and summer than that in fall and winter (Wang et al., 2016), the ships burn heavy oils and might release more mercury to bring higher concentration of atmospheric mercury in nearby cities. Whereas, the super typhoon Haikui passed through Huaniao Island from Aug. 5 to Aug. 10 in 2012, consequently, the typhoon diluted the particle bounded mercury to decrease significantly the contribution of shipping activity to atmospheric mercury. Different mercury species in particles usually depend on different formation mechanisms in theory, so the different seasonal variation patterns were found in this study. The concentration of HPM was the highest in winter, followed by spring, and the lowest concentration occurred in fall. The concentration of EPM varied little with the seasons. As for the seasonal variation of RPM, it was generally, similar to that of HPM except that in summer. RPM, different from other seasons, actually took the higher fractions in all seasons. The results of RPM were not consistent with the previous study in Shanghai (Duan et al., 2016; Xiu et al., 2009) because of the different particle size like TSP in Huaniao Island and PM2.5 in Shanghai. It should be noted that at the marine site like Huaniao Island, TPM might be dominated by larger sea salt, while it was influenced by smaller size particles in the coastal site (Feddersen et al., 2012). The seasonal variation of mass ratio of TPM/TSP, HPM/TSP, EPM/ TSP and RPM/TSP were presented in Fig. 2b. The ANOVA test also supported statistically the significant differences of mercury contents among different seasons (p < 0.001), and the seasonal patterns were quietly different from the concentrations of mercury species. Fig. 2b shows that the mass ratio of TPM/TSP was the highest in fall followed by summer, so were the mass ratio of EPM and RPM except HPM. The ratio of HPM/TSP was higher in summer, the possible explanation is that the photochemical oxidation of Hg0 can be enhanced to result in more HPM in particles to some extent. In fall, the higher EPM/TSP might be attributed to the biomass combustion, as suggested by some studies (Skodras et al., 2007), because biomass sorbent could greatly enhance the absorption of Hg0. The variability of mercury species in the typical climate conditions was also observed and illustrated in Fig. 2c, while the mass ratios of mercury species showed the opposite variation in Fig. 2d. In foggy days, the concentrations of all species increased greatly, which was consistent with previous studies (Xiu et al., 2009). The possible explanation is that the relative humidity was usually high in foggy days, the aqueous or heterogenic reaction in droplet accelerated mercury transformation. The aerosols during rainy days was slightly acidic, which makes it possible for gaseous elemental
2
mercury to react with such inorganic ions as SO2
and 4 , Cl , S reach equilibrium. These complex reactions might contribute to any forms of mercury in particles. The opposite variation of those mass ratios in Fig. 3d showed that mercury species accounted for the higher proportion in sunny and cloudy days than those in rainy and foggy days due to the different concentration of particles (Duan et al., 2016). 3.3. Relationships between speciated mercury and halogens (Br and I) on particles Halogen species are commonly responsible for the atmospheric mercury depletion events and believed to play a major role in the atmospheric oxidation of elemental mercury (Hg0) to divalent mercury (Hg2þ) (Itsaso et al., 2014; Seigneur and Lohman, 2008).

Please cite this article in press as: Duan, L., et al., Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment, Environmental Pollution (2017), http:// dx.doi.org/10.1016/j.envpol.2016.10.103

L. Duan et al. / Environmental Pollution xxx (2017) 1e9

5

Fig. 2. The distribution of mercury species among different seasons and climate conditions.

Significant correlations between TPM and halogens had been found to be quite different from the previous results in continental sampling site that there were no correlations between TPM and halogens in Shanghai (Duan et al., 2016). Those differences might result from the different samples and different source profiles of the two sites. PM2.5 samples were collected in Shanghai while TSP samples were collected in Huaniao Island. The source profiles were also different in these two sites because Xuhui Site in Shanghai and Huaninao Island represents typically urban environment and remote marine environment, respectively. The average concentrations of Br and I over Huaniao Island were 94.9 ± 148.0 ng m
3 and 23.1 ± 47.4 ng m
3, respectively, which were much higher than those at the urban site in Shanghai. The highest correlations between TPM and halogens were found in spring (Br, r ¼ 0.95, p < 0.0001; I, r ¼ 0.97, p < 0.0001), which was consistent with the highest concentration of TPM in spring (Section 3.2). The results also showed that the concentration of TPM in marine site was greatly affected by halogens. Moore et al. (2013) reported that the halogen radicals can oxidize Hg0 to Hg2þ, producing an Hg2þ halogen complex (such as HgCl2, HgBr2 and HgI2) through the intermediate products of Hgþ1 halogen complex that is unstable at high temperatures.

Br2 þ hv/2Br

Hg0 þ Br4HgBr HgBr þ X/HgBrX Another pathway may include reaction of Hg with BrO_, though those reactions have been considered endoergic.

Br2 þ hv/2Br Br þ O3 /BrO þ O2 BrO þ Hg 0 /HgO þ Br or

BrO þ Hg 0 4HgBr þ O or

BrO þ Hg 0 /HgBrO Thus, the halogens could promote the conversion of Hg0 to Hg2þ. The Hg2þ halogen complex produced by the two ways can be captured by the particles and turned to particulate mercury. To further study the influence of halogen on Hg, the correlations between speciated mercury and halogens were also investigated

Please cite this article in press as: Duan, L., et al., Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment, Environmental Pollution (2017), http:// dx.doi.org/10.1016/j.envpol.2016.10.103

6

L. Duan et al. / Environmental Pollution xxx (2017) 1e9

Fig. 3. 72-h air particle backward trajectories (every 6 h in sampling days) during the four seasons. The numbers in brackets indicate the percentage of allocation to a cluster.

among the four seasons. The statistically significant correlations of HPM and halogens were only found in spring which suggested that halogens might accelerate the oxidation of Hg0, producing HgBr2 and HgI2. It was consistent with both the high concentration and highest percentage of HPM in spring (seen in Section 3.2) and the highest concentration of TPM in spring. Nevertheless, both EPM and RPM showed the highest coefficient with halogens in winter. Since EPM were supposed to be formed by adsorption of gaseous elemental mercury and the adsorbent process is exothermic, the lower temperature can enhance the adsorption of Hg0 on particles. However, few research has been published to reveal the exact components of RPM, the formation mechanism of RPM in TSP might be more complex in marine site. In addition, negative correlation between halogens and HPM were found in winter. The possible explanation is that halogens were not the main oxidants for speciation conversion of mercury in winter (Nerentorp Mastromonaco et al., 2016). In fall, although halogens showed high correlations with TPM, while the TPM showed low concentration in fall which may be due to the low concentration of halogens in fall (seen in Section 2.2). 3.4. Backward trajectory analysis The inter-exchange of continental and marine aerosols is an important pathway for atmospheric mercury transportation and will impact the concentration of TPM. The external source and

geographic characteristic of regions during transportation must be the two important considerations. Air mass coming from different regions or transporting different regions might bring about different chemical species. If the air mass from northwestern China passes the polluted metropolitan areas, such as Beijing, Shandong, TPM content might increase in Shanghai. Therefore the HYPLIST model was used to trace the source of air mass in different seasons to identify the contribution of long-range transportation to TPM in Shanghai. The clustered 72-h backward trajectories of different seasons were shown in Fig. 3. In spring, cluster 1 and cluster 3 were the main transportation routes of particles to Huaniao Island and accounted for 33.3% and 41.7%, respectively. When air mass followed the route of cluster 3, the average concentration of TSP was the highest and reached 179.9 ± 104.1 mg m
3, but the TPM/TSP (2.09 ± 0.96 mg/g) wasn't the lowest. While the concentration of TPM (0.44 ± 0.25 ng m
3) and the ratio of TPM/TSP (3.29 ± 2.96 mg/g) were the highest when air mass followed the route of cluster 1 which came from Mongolia, Xinjiang and Inner Mongolia of China, the arid and semiarid area. Compared with the concentrations of mercury species of cluster 1, the concentrations associated with the other clusters were relatively low. The obvious difference is because that the other clusters passed long distance over East China Sea and the deposition over the sea could result in the low concentrations. Both TSP and mercury associated with all clusters in summer were relatively low because of the supper typhoon, except the higher TPM/TSP

Please cite this article in press as: Duan, L., et al., Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment, Environmental Pollution (2017), http:// dx.doi.org/10.1016/j.envpol.2016.10.103

L. Duan et al. / Environmental Pollution xxx (2017) 1e9

(4.05 ± 1.96 mg/g) of cluster 1 which accounted for 30.9%. There was no obvious difference of TPM concentration among the four clusters in fall, but the different TSP concentrations led to the different TPM/TSP ratios. The concentration of TSP (110.5 ± 26.2 mg m
3) from Cluster 1 was the highest, and the content of TPM/TSP (1.48 ± 0.47 mg/g) was the lowest but the concentration of HPM (159.1 ± 68.1 pg m
3) was the highest probably due to the oxidation of Hg0 over the sea. In winter, the categorized 4 clusters of air masses were all originated from Mongolia continent. Though with the low ratio, cluster 3 had the highest concentration of mercury and TSP (163.9 ± 60.2 ng m
3), TPM/TSP (2.95 ± 0.93 mg/g), except the lowest content of HPM (54.4 ± 55.5 pg m
3), contributing to the main source of mercury in winter. Cluster 1, without passing any major water body except for a short distance over East China sea, account for the highest concentration of HPM (175.8 ± 34.2 pg m
3), although all of the other clusters traveled a

7

long distance over the sea, this result also corresponded with the negative correlation between HPM and halogens (in Section 2.3). The air mass backward trajectories during the whole sampling period were calculated by the HYSPLIT model and summarized in Table 3. The clustered 72-h backward trajectories during the whole sampling period were shown in Fig. 4. The trajectories were categorized into 4 groups: cluster 1, accounting for 22.9%, represented air mass originated from Mongolia through long-range transportation. Cluster 2, accounting for 13.7%, represented air mass coming from the south of pacific ocean through Taiwan strait. Cluster 3, accounting for 22.9%, represented air mass coming across pacific ocean over East China Sea. Cluster 4, accounting for 40.5%, was the main route of dust from Inner Mongol, Liaoning province to Huaniao Island. The concentrations of TSP and TPM from cluster 1 and cluster 4 were both higher than the average concentration in Huaniao Island.

Table 3 TSP and mercuric species from different clusters during the whole sampling period. Clusters

TSP(mg m
3)

TPM(ng m
3)

1 2 3 4

131.2 ± 46.7 86.4 ± 57.6 62.8 ± 48.7 120 ± 79.9

0.28 0.16 0.16 0.26

± ± ± ±

0.19 0.06 0.04 0.17

HPM(pg m
3)

EPM(pg m
3)

125.9 ± 66.2 53.1 ± 43.4 73.6 ± 22.9 70.3 ± 51.0

73.1 37.5 38.1 62.8

± ± ± ±

34.8 8.4 11.6 32.5

RPM(pg.m
3) 235.4 148.1 158.0 255.6

± ± ± ±

148.7 37.7 95.0 166.9

TPM/TSP(mg/g) 2.26 2.22 4.08 2.65

± ± ± ±

1.76 0.89 2.56 1.58

Fig. 4. 72-h air particle backward trajectories (every 6 h in sampling days) during the whole sampling period. The numbers in brackets indicate the percentage of allocation to a cluster.

Please cite this article in press as: Duan, L., et al., Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment, Environmental Pollution (2017), http:// dx.doi.org/10.1016/j.envpol.2016.10.103

8

L. Duan et al. / Environmental Pollution xxx (2017) 1e9

Both of these two trajectories mainly came from the arid and semiarid area such as Mongolia, Xinjiang and Inner Mongolia of China. The coarser particles could be transported here. As a result, there was higher TSP, and lower TPM/TSP ratio of cluster 1 and 4 compared with cluster 2 and 3. In addition, the cluster 2 and 3 have similar mercury profiles, both of which may result from the ocean aerosol. The concentration of TSP and TPM from Cluster 3 was the lowest, while the TPM/TSP ratio was the highest. Compared with EPM and RPM, HPM associated with cluster 3 was slightly higher. The oxidation of Hg0 over the sea might enhance the concentration of HPM, leading to the highest TPM/TSP. 4. Conclusions In this study, we analyzed the characteristics and potential sources of the total particulate mercury over Huaniao Island in northern ECS based on a one-year measurement of speciated mercury and halogens in particles. The mean level of TPM was considerably lower than those at urban sites in China, but the mass ratio was higher than in Shanghai. TPM showed obviously seasonal patterns with the highest average concentration in spring, not in winter and the lowest average concentration in fall and summer. The mass ratio represented the opposite seasonal variations, the higher HPM/TSP in summer and the higher EPM/TSP in fall represented the photochemical reaction and absorption of gaseous mercury, respectively. The results of correlations between mercury and halogens indicated that halogens accelerate the oxidation of Hg0 especially in spring, leading to the higher concentration of HPM in spring. Besides, halogens can affect the gas-particle transformation through improving the adsorption process, particularly in cool days. During the whole sampling period, the trajectories analysis demonstrated that the air mass from the north of China contribute to the higher concentration of TPM in Huaniao Island. But it shouldn't be neglected that the marine airflow can increase the content of HPM due to the oxidation of Hg0. Besides, there were obvious differences among the four seasons. The air mass over the sea carried with low concentration of mercury in spring mainly due to the deposition over the long distance across the sea. While, in winter, the clusters across the sea showed low concentration of HPM, which suggests that the oxidation of mercury in winter might be more strongly related to other oxidants. This study on the relationship between mercury species and halogens and the source of particulate mercury in marine environment will provide useful information to evaluate the influence of marine aerosols on the atmospheric transport and deposition of atmospheric mercury to the North Pacific Region. Acknowledgment This research was financially supported by the National Natural Science Foundation of China (NO. 21277044; No. 41375141). Many thanks for the assistance in the measurement of Br/I by Dr. Bianfang Shi from School of Chemical Engineer, East China University of Science and Technology. References Bo, D., Cheng, J., Xie, H., Zhao, W., Wei, Y., Chen, X., 2016. Mercury concentration in fine atmospheric particles during haze and non-haze days in Shanghai, China. Atmos. Pollut. Res. 7, 348e354. Chand, D., Jaffe, D., Prestbo, E., Swartzendruber, P.C., Hafner, W., Weiss-Penzias, P., Kato, S., Takami, A., Hatakeyama, S., Kajii, Y., 2008. Reactive and particulate mercury in the Asian marine boundary layer. Atmos. Environ. 42, 7988e7996. Cheng, I., Zhang, L., Blanchard, P., Dalziel, J., Tordon, R., Huang, J., Holsen, T.M., 2013. Comparisons of mercury sources and atmospheric mercury processes between a coastal and inland site. J. Geophys. Res. 118, 2434e2443. Cheng, I., Zhang, L., Mao, H., Blanchard, P., Tordon, R., Dalziel, J., 2014. Seasonal and

diurnal patterns of speciated atmospheric mercury at a coastal-rural and a coastal-urban site. Atmos. Environ. 82, 193e205. Choi, H.D., Huang, J., Mondal, S., Holsen, T.M., 2013. Variation in concentrations of three mercury (Hg) forms at a rural and a suburban site in New York State. Sci. Total Environ. 448, 96e106. Draxler, R.R., Hess, G.D., 1998. An overview of the HYSPLIT_4 modelling system for trajectories, dispersion and deposition. Aust. Meteor. Mag. 47, 295e308. Duan, L., Xiu, G., Feng, L., Cheng, N., Wang, C., 2016. The mercury species and their association with carbonaceous compositions, bromine and iodine in PM2.5 in Shanghai. Chemosphere 146, 263e271. € we, G.A., Richer, A., Burrows, J.P., Ebinghaus, R., Kock, H.H., Temme, C., Einax, J.W., Lo 2002. Antarctic springtime depletion of atmospheric mercury. Environ. Sci. Technol. 36, 1238. Fang, F., Wang, Q., Li, J., 2004. Urban environmental mercury in Changchun, a metropolitan city in Northeastern China: source, cycle, and fate. Sci. Total Environ. 330, 159e170. Feddersen, D.M., Talbot, R., Mao, H., Sive, B.C., 2012. Size distribution of particulate mercury in marine and coastal atmospheres. Atmos. Chem. Phys. 12, 10899e10909. Ferrari, C.P., Padova, C., Fain, X., Gauchard, P.A., Dommergue, A., Aspmo, K., Berg, T., Cairns, W., Barbante, C., Cescon, P., Kaleschke, L., Richter, A., Wittrock, F., Boutron, C., 2008. Atmospheric mercury depletion event study in Ny-Alesund (Svalbard) in spring 2005. Deposition and transformation of Hg in surface snow during springtime. Sci. Total Environ. 397, 167e177. Fu, X., Feng, X., Zhu, W., Zheng, W., Wang, S., Lu, J.Y., 2008. Total particulate and reactive gaseous mercury in ambient air on the eastern slope of the Mt. Gongga area, China. Appl. Geochem. 23, 408e418. Fu, X.W., Feng, X., Liang, P., Deliger, Zhang, H., Ji, J., Liu, P., 2012. Temporal trend and sources of speciated atmospheric mercury at Waliguan GAW station, Northwestern China. Atmos. Chem. Phys. 12, 1951e1964. Gao, Y., Sun, M., Wu, X., Liu, Y., Guo, Y., Wu, J., 2010. Concentration characteristics of bromine and iodine in aerosols in Shanghai, China. Atmos. Environ. 44, 4298e4302. Graney, J.R., Dvonch, J.T., Keeler, G.J., 2004. Use of multi-element tracers to source apportion mercury in south Florida aerosols. Atmos. Environ. 38, 1715e1726.
Castillo, Bozzelli, J.W., 2014. Mercury oxidation via chlorine, Itsaso, A.-M., Alvaro, bromine, and iodine under atmospheric conditions: thermochemistry and kinetics. J. Phys. Chem. 118, 2959, 2575. Kim, S.H., Han, Y.J., Holsen, T.M., Yi, S.M., 2009. Characteristics of atmospheric speciated mercury concentrations (TGM, Hg(II) and Hg(p)) in Seoul, Korea. Atmos. Environ. 43, 3267e3274.
pez-Anto
n, M.A., Díaz-Somoano, M., Abad-Valle, P., Martínez-Tarazona, M.R., Lo 2007. Mercury and selenium retention in fly ashes: influence of unburned particle content. Fuel 86, 2064e2070. Lamborg, C.H., Fitzgerald, W.F., Graustein, W.C., Turekian, K.K., 2000. An examination of the atmospheric chemistry of mercury using 210Pb and 7Be. J. Atmos. Chem. 36, 325e338. Lindberg, S.E., Brooks, S., Lin, C.J., Scott, K.J., Landis, M.S., Stevens, R.K., Goodsite, M., Richer, A., 2002. Dynamic oxidation of gaseous mercury in the Artic troposphere at polar sunrise. Environ. Sci. Technol. 36, 1245. Lindberg, S.E., Stratton, W.J., 1998. Atmospheric mercury speciation: concentrations and behavior of reactive gaseous mercury in ambient air. Environ. Sci. Technol. 32, 49e57. Mao, H., Talbot, R., 2012. Speciated mercury at marine, coastal, and inland sites in \newline New England e Part 1: temporal variability. Atmos. Chem. Phys. 12, 5099e5112. Moore, C.W., Obrist, D., Luria, M., 2013. Atmospheric mercury depletion events at the Dead Sea: spatial and temporal aspects. Atmos. Environ. 69, 231e239. Nerentorp Mastromonaco, M., Gårdfeldt, K., Jourdain, B., Abrahamsson, K.,
jean, G., Jacobi, H.W., 2016. AntGranfors, A., Ahnoff, M., Dommergue, A., Me arctic winter mercury and ozone depletion events over sea ice. Atmos. Environ. 129, 125e132. Pirrone, N., Cinnirella, S., Feng, X., Finkelman, R.B., Friedli, H.R., Leaner, J., Mason, R., Mukherjee, A.B., Stracher, G.B., Streets, D.G., Telmer, K., 2010. Global mercury emissions to the atmosphere from anthropogenic and natural sources. Atmos. Chem. Phys. 10, 5951e5964. Rutter, A., Schauer, J., 2007. The Impact of Aerosol composition on the particle to gas partitioning of reactive mercury. Environ. Sci. Technol. 41, 3934e3939. €fer, J., Blanc, G., Chen, Y., Chai, F., Cen, K., Norra, S., 2015. AtSchleicher, N.J., Scha mospheric particulate mercury in the megacity Beijing: spatio-temporal variations and source apportionment. Atmos. Environ. 109, 251e261. Schroeder, W.H., Anlauf, K.G., Barrie, L.A., Lu, J.Y., Steffen, A., Schneeberger, D.R., Berg, T., 1998. Arctic springtime depletion of mercury. Nature 394, 331e332. Schroeder, W.H., Munthe, J., 1998. Atmospheric mercurydan overview. Atmos. Environ. 32, 809e822. Seigneur, C., Lohman, K., 2008. Effect of bromine chemistry on the atmospheric mercury cycle. J. Geophys. Res. 113. Skodras, G., Diamantopoulou, I., Zabaniotou, A., Stavropoulos, G., Sakellaropoulos, G.P., 2007. Enhanced mercury adsorption in activated carbons from biomass materials and waste tires. Fuel Process. Technol. 88, 749e758. Wan, Q., Feng, X.B., Lu, J., Zheng, W., Song, X.J., Li, P., Han, S.J., Xu, H., 2009. Atmospheric mercury in Changbai Mountain area, northeastern China II. The distribution of reactive gaseous mercury and particulate mercury and mercury deposition fluxes. Environ. Res. 109, 721e727. Wang, F., Chen, Y., Meng, X., Fu, J., Wang, B., 2016. The contribution of anthropogenic

Please cite this article in press as: Duan, L., et al., Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment, Environmental Pollution (2017), http:// dx.doi.org/10.1016/j.envpol.2016.10.103

L. Duan et al. / Environmental Pollution xxx (2017) 1e9 sources to the aerosols over East China Sea. Atmos. Environ. 127, 22e33. Wang, Y.Q., Zhang, X.Y., Draxler, R.R., 2009. TrajStat: GIS-based software that uses various trajectory statistical analysis methods to identify potential sources from long-term air pollution measurement data. Environ. Model. Softw. 24, 938e939. Wang, Z., Zhang, X., Chen, Z., Zhang, Y., 2006. Mercury concentrations in sizefractionated airborne particles at urban and suburban sites in Beijing, China. Atmos. Environ. 40, 2194e2201. Xiu, G., Cai, J., Zhang, W., Zhang, D., Büeler, A., Lee, S., Shen, Y., Xu, L., Huang, X., Zhang, P., 2009. Speciated mercury in size-fractionated particles in Shanghai ambient air. Atmos. Environ. 43, 3145e3154. Xiu, G.L., Jin, Q., Zhang, D., Shi, S., Huang, X., Zhang, W., Bao, L., Gao, P., Chen, B., 2005. Characterization of size-fractionated particulate mercury in Shanghai

9

ambient air. Atmos. Environ. 39, 419e427. Xu, L., Chen, J., Yang, L., Niu, Z., Tong, L., Yin, L., Chen, Y., 2015. Characteristics and sources of atmospheric mercury speciation in a coastal city, Xiamen, China. Chemosphere 119, 530e539. Zhang, Y., Liu, R., Wang, Y., Cui, X., Qi, J., 2015. Change characteristic of atmospheric particulate mercury during dust weather of spring in Qingdao, China. Atmos. Environ. 102, 376e383. Zhao, M., Huang, Z., Qiao, T., Zhang, Y., Xiu, G., Yu, J., 2015. Chemical characterization, the transport pathways and potential sources of PM2.5 in Shanghai: seasonal variations. Atmos. Res. 158e159, 66e78. Zheng, L., Liu, G., Chou, C.-L., 2007. The distribution, occurrence and environmental effect of mercury in Chinese coals. Sci. Total Environ. 384, 374e383.

Please cite this article in press as: Duan, L., et al., Characteristics and source appointment of atmospheric particulate mercury over East China Sea: Implication on the deposition of atmospheric particulate mercury in marine environment, Environmental Pollution (2017), http:// dx.doi.org/10.1016/j.envpol.2016.10.103