- Email: [email protected]
Mercury isotope signatures of seawater discharged from a coal-fired power plant equipped with a seawater flue gas desulfurization system
Environmental Pollution 214 (2016) 822e830
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Mercury isotope signatures of seawater discharged from a coal-fired power plant equipped with a seawater flue gas desulfurization system* Haiying Lin a, b, Jingji Peng a, Dongxing Yuan a, *, Bingyan Lu a, Kunning Lin a, Shuyuan Huang a a b
State Key Laboratory of Marine Environmental Science, College of the Environment and Ecology, Xiamen University, Xiamen 361102, China Guangxi Colleges and Universities Key Laboratory of Environmental Protection, School of Environment, Guangxi University, Nanning 530004, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 January 2016 Received in revised form 10 April 2016 Accepted 16 April 2016 Available online 5 May 2016
Seawater flue gas desulfurization (SFGD) systems are commonly used to remove acidic SO2 from the flue gas with alkaline seawater in many coastal coal-fired power plants in China. However, large amount of mercury (Hg) originated from coal is also transferred into seawater during the desulfurization (De-SO2) process. This research investigated Hg isotopes in seawater discharged from a coastal plant equipped with a SFGD system for the first time. Suspended particles of inorganic minerals, carbon residuals and sulfides are enriched in heavy Hg isotopes during the De-SO2 process. d202Hg of particulate mercury (PHg) gradually decreased from 0.30‰ to 1.53‰ in study sea area as the distance from the point of discharge increased. The results revealed that physical mixing of contaminated De-SO2 seawater and uncontaminated fresh seawater caused a change in isotopic composition of PHg isotopes in the discharging area; and suggested that both De-SO2 seawater and local background contributed to PHg. The impacted sea area predicted with isotopic tracing technique was much larger than that resulted from a simple comparison of pollutant concentration. It was the first attempt to apply mercury isotopic composition signatures with two-component mixing model to trace the mercury pollution and its influence in seawater. The results could be beneficial to the coal-fired plants with SFGD systems to assess and control Hg pollution in sea area. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Mercury isotope signatures Fractionation Desulfurization seawater Coal-fired power plant Tracing source
1. Introduction Human activities, such as coal combustion and other industrial emission sources (Foucher and Hintelmann, 2006, 2009; Sherman et al., 2012; Sun et al., 2013a, 2013b, 2013c), have accelerated the emission and geochemical cycling of mercury (Hg) into the environment (Choe and Gill, 2003; Morel et al., 1998). For instance, coal combustion as one of the largest emitters of Hg in the world, has contributed more than 35,000 tons into the atmosphere (Sun et al., 2016). Coal usually contains high levels of sulfur and trace amount of Hg. In China and other countries in the world, many coal-fired power plants along/near the coast are equipped with the seawater flue gas desulfurization (SFGD) system, which use the
*
This paper has been recommended for acceptance by Prof. W. Wen-Xiong. * Corresponding author. E-mail address: [email protected] (D. Yuan).
http://dx.doi.org/10.1016/j.envpol.2016.04.059 0269-7491/© 2016 Elsevier Ltd. All rights reserved.
alkaline seawater to neutralize acidic sulfur dioxide (SO2) in the flue gas (Feng et al., 2014; Yu et al., 2011). During desulfurization (DeSO2), the flush of sea water also washes Hg off the flue gas, resulting in high Hg content in the seawater (Liang et al., 2010; Liu et al., 2011). A schematic diagram showing the process and Hg transfer from the coal combustion and the SFGD system into the estuary is provided in Fig. S1 of Supplementary. The De-SO2 seawater is mixed with fresh seawater and treated with simple aeration to further neutralize and oxidize SO2 3 , increase pH and dissolved oxygen (DO) (Liu et al., 2011), and then discharged into the adjacent sea. The technique utilizes local seawater resource and avoids producing huge amount of solid waste that is usually generated with common dry desulfurization techniques. The De-SO2 seawater mainly contains relatively high levels of suspended particles and SO2 4 , which are considered not to be the pollutants for coastal environment. However, heavy metals, especially Hg, at high concentration in the De-SO2 seawater could become a potential threat to the ecosystem. A typical air purification system of coal-fired flue gas is
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
composed of a selective catalytic reduction unit, a particulate control unit and a flue gas desulfurization system in series (Hower et al., 2010), installing after the boiler (Fig. S1 in the Supplementary). During the coal combustion (>1200 C) in the boiler, Hg in coal is released as Hg0 into the flue gas (Galbreath and Zygarlicke, 2000b). The selective catalytic reduction unit is used for removal of nitrogen oxides, and at the same time it transfers 30e98% Hg0 into Hg2þ, which is easily to adsorb onto particles and form PHg in the flue gas (Hower et al., 2010). Part of particles and PHg are removed in the particulate control unit. When the flue gas flows through the flue gas desulfurization system, 30e50% PHg, most of Hg2þ and part of Hg0 are transferred into by-products such as gypsum or seawater (Hower et al., 2010). The SY power plant, located in the Xiamen city (southeast coast of China, Fig. 1), is one typically example of coal-fired power plant equipped with SFDG that becomes a Hg emission source (Liang et al., 2010). For instance, the De-SO2 seawater at the outlet of the De-SO2 tower contains total Hg (THg) up to 385.0 ng/L, much higher than China's Class Ⅱ seawater quality standard (200.0 ng/L). According to our previous studies, the De-SO2 seawater can bring 1.03 kg/day of Hg into the adjacent sea (Liu et al., 2011). The result of regular monitoring indicated seawater within approximately 300 m of the SY power plant has shown slightly higher Hg content than that of that in other coastal areas in Xiamen. However, the environmental impact of Hg from the De-SO2 seawater is still a question because of the mismatch of the Hg in the De-SO2 seawater and the discharging sea area; based on the simply Hg mass balance calculations. It is thought that the mixing with local Hg background can result in decreased THg in seawaters of the discharging area, whereas there is a lack of effective tracer to quantify the contribution of Hg from De-SO2 seawater and background. It is not yet clear how large a sea area is influenced by the Hg-contaminated DeSO2 seawater. During the past 15 years, mercury isotope geochemistry has been proved to be a powerful tool to trace sources and geochemical processes of Hg in the environment (Bergquist and Blum, 2009;
823
Blum et al., 2014; Wiederhold et al., 2015). Previous studies have investigated Hg isotopic compositions of coal from different regions of the world (Biswas et al., 2008; Lefticariu et al., 2011; Sun et al., 2014, 2016; Yin et al., 2014), Hg isotope fractionation during coal combustion process (Sun et al., 2013c), and isotopic signature of Hg in atmospheric samples near a coal power plant (Sherman et al., 2012). These studies have facilitated our understanding on Hg transformation, transportation, and environmental impacts during Hg emission by coal combustion. However, Hg isotope fractionation during Hg emission by SFGD is still not well understood. A main technological gap for this is that Hg in seawater is generally too low for Hg isotope analysis. Analytical methods for water Hg isotope determination have been developed recently (Lin et al., 2015; Strok et al., 2014, 2015). To date, only a few studies have reported Hg isotopic compositions in seawaters (Strok et al., 2015); whereas isotopic compositions in the De-SO2 seawater have not been studied. Particulate Hg (PHg) is the main species in water systems and its concentration in natural water varies in different regions around the world. Concentrations of PHg in fresh water of the Isonzo River mouth range from 9.0 to 25.0 ng/L, contributing 51e84% to THg (Stefano et al., 2006). In the seawater of discharging sea area near the SY power plant, 50e97% of THg is in PHg form (Liang et al., 2010; Lin et al., 2015) due to the fact that De-SO2 seawater contains large amount suspended particles, mainly inorganic minerals, carbon residuals and sulfides, generated during the coal combustion process and not removed by the particulate control unit. The particles containing carbon, acid and sulfides have a high affinity to bond Hg (Galbreath and Zygarlicke, 2000a). Therefore, high concentration of PHg was observed in the De-SO2 seawater. This study investigated the variations of Hg isotope compositions in feed coal used in the plant, PHg and DHg in De-SO2 seawater discharged; discussed the transfer and transformation of PHg and DHg in the seawater, and evaluated the source impact of Hg by SFGD. The results of this study could provide information for better understanding environmental behavior of Hg discharged
Fig. 1. Location of the sampling sites.
824
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
from the coal-fired plants with SFGD systems, and could be benefit the evaluation of Hg environment impacts, and to guide future Hg remediation in the studied area.
2. Material and method 2.1. Sampling sites The Xiamen SY power plant is located in Xiamen Bay, in the Jiulongjiang Estuary of Fujian province, China (Fig. 1). Jiulongjiang is the second largest river of Fujian province, running through Xiamen Bay into the Taiwan Strait. Seawater from Xiamen Bay is introduced into the plant for SFGD; the resulted De-SO2 seawater is then mixed with one third fresh seawater, and aerated in two parallel aeration pools to neutralize the SO2 3 concentration, increase the pH and DO level (Liu et al., 2011). After that, the De-SO2 seawater is discharged into the adjacent sea. Four sites within the SY power plant were chosen to collected the samples, which include the fresh seawater inlet (P1), the desulfurization tower outlet (P2), and the two aeration pools (P3, P4). Seawater from additional 15 sites (S1~S15) within 2.0 km of the outlets of the plant were also sampled (Fig. 1). These sites include two sites (S1, S2) at the discharging outlets, and sites (S3~S15) with increasing distances (50, 100, 200, 300, 400, 600, 1500, and 2000 m) off the outlets. The site R1 (Fig. 1) was selected 3.2 km away from the outlets to assess the background. Another site R2 located near the Huangcuo (Xiamen seashore), 12.0 km away from the outlets, was also chosen as a reference site. Surface seawater samples of each site were collected at March 28 and June 18, 2014. The samples were collected at low tide in order to better avoid the influence of fresh seawater and gain access the actual impact sea area, as well as to collect particles at relatively higher concentration and lower the difficulty in analysis of the PHg isotopic composition. Triplicate samples were collected at sites P1~P4, S3, R1~R2; one sample was collected in the other sites. Triplicate samples of feed coal were also collected.
2.2. Sampling and sample pretreatment Sampling and sample preparation were performed following U.S. EPA methods 1669 (U.S. Environmental Protection Agency, 1996). Seawater samples were collected and then pre-oxidized with BrCl solution in the field, which were used for THg concentration analysis. To measure DHg and PHg, seawater samples of 5e10 L were collected and filtered through a 0.45 mm acetate cellulose membrane. The filtrate was used to determine the DHg concentration of the sample following a method developed by Liang et al. (2010). The suspended particles on the membrane were defined as the PHg samples, and the sample together with its filter membrane was immediately soaked in a 5 mL BrCl (5%, v/v) solution after filtration for further treatment. PHg (together with the membrane) samples and coal samples were digested in aqua regia (HCl:HNO3, v/v, 3:1) at 95 C for 3 h, and then diluted with BrCl solution (1%, v/v) after being cooled to room temperature. The samples were diluted to ensure the acid concentrations below 25%(v/v) (Foucher et al., 2013). The digest solution was centrifuged, and the supernatant was used to determine Hg concentration and isotopic composition. For DHg samples with low Hg concentrations, however, Hg was pre-concentrated into a KMnO4 solution using a purge and trap method (Lin et al., 2015). The pre-concentration recoveries of samples and spike standards were in the range of 94.3e105.5%.
2.3. Instrumentation and method A double amalgamation combined with a cold vapor atomic fluorescence spectrometer (Rayleigh Analytical Instrument Corp., China), was used to analyze THg, PHg, and DHg samples (Liang et al., 2010; Liu et al., 2011). A water quality monitoring device (JAP60M/U52-10M, Japan) was used to measure water temperature, pH, salinity, and DO. Mercury isotope compositions were determined using a cold vapor generation (CVG) device combined with a desolvating nebulizer (DSN-100, Nu Instruments, UK), and a multi-collector inductively coupled plasma mass spectrometer (MC-ICP-MS, Nu Plasma HR, Nu Instruments, UK) in State Key Laboratory of Marine Environmental Science of Xiamen University following our previous method (Lin et al., 2015). Instrumental mass bias was corrected using Tl internal standard (NIST SRM 997) and standard-sample bracketing method (Blum and Bergquist, 2007; Sonke, 2011). MDF of Hg isotopes was expressed in d notation as per mil deviations (‰) from NIST SRM 3133 Hg standard (Bergquist and Blum, 2009):
.
h
d Hgð‰Þ ¼ ð Hg=198 HgÞsample ð Hg=198 HgÞnist 31331
i
1000 (1) xxx
199
200
201
202
In this expression, Hg is Hg, Hg, Hg and Hg. MIF of Hg isotopes was expressed as DxxxHg, calculated using the following equation (Kritee et al., 2013; Strok et al., 2014):
Dxxx Hg ¼ d Hg ðd202 Hg bÞ where b is 0.2520, 0.5024 and 0.7520 for respectively.
(2) 199
Hg,
200
Hg,
201
Hg,
2.4. Quality control Quality assurance (QA) blanks indicated that samples were uncontaminated during the sampling and pretreatment processes. The Hg concentration in the procedural blank digestions (0.02e0.05 ng/mL, mean value 0.03 ng/mL) was 4% of the sample digestion solution (0.70e3.00 ng/mL). The Hg concentration in the sediment standard reference material (GBW07304a, GSD-4a) was 0.075 ± 0.004 mg/g, well agreed with the certified value (0.078 ± 0.006 mg/g). The internal precision for 202Hg/198Hg was better than 0.04‰ (1 standard error, 1SE) during Hg isotope measurement. The external precision (uncertainty, 2 standard deviation, 2SD) was assessed by measuring triplicate samples with the same sampling and pretreatment processing (Foucher et al., 2013), or using the UMn standard at the same Hg concentration when no parallel Almade samples were collected (Sun et al., 2013c). The d202Hg, D199Hg and D200Hg for the NIST 3133 and UM-Almaden were 0.00 ± 0.08‰ (2SD, n ¼ 36) and 0.51 ± 0.10‰ (2SD, n ¼ 16); 0.00 ± 0.04‰ (2SD, n ¼ 36) and 0.02 ± 0.04‰ (2SD, n ¼ 16); 0.01 ± 0.04‰ (2SD, n ¼ 36) and 0.02 ± 0.02‰ (2SD, n ¼ 16), respectively. The internal and external precisions were similar to previous studies (Foucher and Hintelmann, 2009; Foucher et al., 2013; Yin et al., 2013). 3. Results and discussion 3.1. Distribution of THg and PHg THg, PHg, and DHg in the seawater sampled for this study ranged in 3.3e389.5 ng/L, 1.0e335.4 ng/L, and 0.6e54.0 ng/L, respectively. Mass balance shown that the sum of PHg and DHg
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
concentration correspond to 93e104% of the THg concentration. Before the De-SO2 process, The THg concentration (3.3 ng/L) in the Xiamen SY plant seawater inlet (P1) was relatively low, similar to that at reference site R1 (3.0 ng/L, Fig. 2). Seawater collected at the De-SO2 tower outlet (P2) has the highest THg concentration (389.5 ng/L), which is 130 times higher than that from inlet seawater (P1). PHg is the predominate speciation of Hg in the seawater, representing approximately 86% of THg in the outlet (P2). When the fresh seawater was sprayed onto the flue gas in the DeSO2 tower, PHg and Hg2þ in the flue gas were washed into the seawater, causing significant increase of THg and PHg. In the aeration pools (P3, P4), the mixing of De-SO2 seawater with 1/3 fresh seawater resulted in the decreasing of Hg concentrations. PHg and THg concentrations dropped to 245.5 ng/L and 266.4 ng/L, respectively. Meanwhile, DHg in seawater decreased from 54.0 ng/L to 20.9 ng/L, possibly due to the mixing effect of the fresh seawater combined the evasion of DHg from seawater to the atmosphere (Sun et al., 2013a). THg concentrations in pools P3, P4 were 262.0 ng/L and 271.0 ng/L, respectively, higher than the threshold value (200.0 ng/L) regulated by China's Class Ⅱ seawater quality, suggesting that the De-SO2 seawater may be a Hg contamination source. THg and PHg concentrations in seawater decreased rapidly as increased distance from the outlets. THg in surface seawater at outlets S1 and S2 (60.0e90.0 ng/L) dropped to ~10.0 ng/L at sites S3~S8 (50e300 m away from the outlets). This level is still higher than that of the reference site R1 (3.0 ng/L), indicating being influenced by the De-SO2 seawater. At sites 400e2000 m away from the outlets, THg concentrations in seawater were low and constant (3.7 ng/L), and similar to that at reference site R1 (3.3 ng/ L). The differences in THg concentrations between seawater at 400e2000 m away and reference site R1 were statistically insignificant (within standard error range). As such, there was no conclusive evidence to prove that seawater 400e2000 m away from the outlets was influenced by De-SO2 seawater. Ratios of PHg to THg (PHg/THg) were 86e94% at P2~P4, which was agreed with the previous study (Liang et al., 2010; Liu et al., 2011). Most of Hg in the coal is transferred into Hg0 during the combustion, and when the flue gas cools down and passes through the denitration device, large amount of Hg0 is oxidized into Hg2þ. Hg2þ is easily to adsorb or bond onto particles as the PHg (Sun et al., 2013c), which made PHg become the main form of the flue gas.
825
Table 1 Correlation coefficients (R) of ancillary parameters and Hg concentration. Correlation coefficient (R, n ¼ 19) SO2 3 (mg/L) THg (ng/L) PHg (ng/L)
b
0.963 0.831b
Temperature ( C) a
0.658 0.759b
pH b
0.825 0.697a
DO (mg/L)
Salinity
0.378 0.492
0.428 0.438
a
Statistically significantly correlated at confidence level of 0.05. Significantly correlated at confidence level of 0.01. Analyses were conducted using the Spearman analysis method with the SPSS software package. b
Most of PHg, Hg2þ and part of Hg0 could be washed into the seawater during the De-SO2 process, leading to high PHg concentration and higher PHg/THg. Although the PHg/THg decreased from 78% to 51% within 2.0 km from the outlets, PHg was still the main speciation. During discharge and transport into the open seawater, suspended particles in De-SO2 seawater either sunk into the seabed or mixed with fresh seawaters, which may explain the decrease of PHg/THg with increased distance from the outlets. Most of particles in De-SO2 seawater were in small size (micrometer level), therefore the sedimentation should be insignificant in the estuary. 3.2. Relationship between the ancillary parameters and Hg concentration Some estuary parameters, including water temperature, SO2 3 concentration, pH, DO and salinity, were monitored on field to find out their relationship with Hg concentration. A gradient was found for water temperature, SO2 3 concentration, pH, THg and PHg, from the De-SO2 tower outlet P2 to the discharging point, and then to the sea area. Table 1 shows the correlation coefficients between the ancillary parameters and Hg concentration at each sampling site (n ¼ 19); THg and PHg were positively correlated with water temperature and SO2 concentration, and negatively correlated 3 with pH. There were no correlations between Hg and DO and salinity. The correlations among THg, PHg concentration, seawater temperature, SO2 concentration and pH may be explained by 3 similar processes during the De-SO2 and aeration. During the DeSO2 process, SO2 and Hg were removed from the flue gas by seawater, resulting high concentrations of SO2 3 (16 mg/L), THg (389.0 ng/L), and PHg (335.0 ng/L) in seawater at De-SO2 tower
Fig. 2. THg and PHg concentrations at sampling sites. (The plotted dots represent d202Hg of PHg, error bars represent 2SD of d202Hg values, Line 1 and Line 2 represent the decreasing trend of THg concentration and d202Hg).
826
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
outlet P2. The high temperature in the De-SO2 tower boiled the seawater to 100 C. The acid gas (HCl, SO2) in the flue gas were removed by the seawater, resulting in the decrease of the pH to 3. During fresh seawater mixing and aeration, THg and PHg and SO2 3 concentrations, temperature decreased whereas pH increased at sites P3 and P4. Fresh seawater for use in the SFGD was collected from the nearby sea. Seawater salinity remained constant during the De-SO2 process, aeration, and discharge, and there were lower variations in DO than in THg and PHg. As such, no correlations existed between salinity, DO, and Hg concentrations. 3.3. Mercury isotope signature of PHg Fig. 2 and Table S1 in SI show the spatial variations of d202Hg for suspended particles. De-SO2 seawater particles collected at tower outlet P2 show more positive d202Hg value (0.31 ± 0.04‰, 2SD, n ¼ 3) than that of fresh seawater at P1 (d202Hg 1.32 ± 0.02‰, 2SD, n ¼ 3). The higher d202Hg value of PHg at P2 was reflective of oxidized Hg in the flue gas, because most of Hg in the De-SO2 seawater came from the coal-fired flue gas. Much higher d202Hg in the flue gas (0.31 ± 0.04‰, 2SD, n ¼ 3) than the feed coal (1.35 ± 0.10‰, 2SD, n ¼ 3) have been observed, indicating isotope fractionation of Hg during coal combustion and flue gas purification. This was inconsistent with a previous study (Sun et al., 2013c) about a dry desulfurization system, where oxidized Hg species in bottom ash, fly ash (d202Hg: 1.96 to 0.82‰) and by-product gypsum (d202Hg: 0.99 to 0.47‰) were enriched with light Hg isotopes relative to feed coal (d202Hg: 0.67 to 0.18‰). The disagreement might be due to the differences of the feed coals and desulfurization techniques used in the two study cases. The feed coal in previous study has relatively higher d202Hg than that in this study, ranging from 0.67‰ to 0.18‰. The process of dry desulfurization is very different from SFDG. The MDF of PHg in this present case might result from Hg thermal combustion, evaporation, diffusion, adsorption, and redox during coal-combustion and air purification processes. The d202Hg (0.38 ± 0.16‰, 2SD, n ¼ 6) of the PHg in the aeration pool (P3, P4) were statistically insignificant from P2 (0.32 ± 0.04‰, 2SD, n ¼ 3), indicating that aeration may not cause PHg isotope fractionation. PHg at discharging outlets S1 and S2 had a similar d202Hg (0.30 ± 0.10‰, 0.33 ± 0.10‰, 2SD, n ¼ 3) as tower outlet P2 and aeration pools P3 and P4 (0.36 ± 0.14‰, 2SD, n ¼ 9), which is well agreed with the fact that those sites mainly received Hg from the power plant, as proven by Hg concentrations. d202Hg values decreased from 0.68‰ (S3) to 0.85‰ (S8), and d202Hg values (1.35 ± 0.36‰, 2SD, n ¼ 5) at S9~S15 were more negative. The overall variation of d202Hg values (up to 1.23‰) is significant compared to the analytical uncertainty of our method (±0.10‰). In addition, fresh seawater inlet P1 and sea area sites S11~S15 have similar d202Hg values, suggestive of less impact by the power plant. In fact, the fresh seawater for De-SO2 came from an area 1.0 km from the outlets, where was the location of sites S11~S15. Turning to the background data, reference site R1 was located near an uninhabited island, 3.2 km away from S1, where seawater is relatively clean and protected from human activities. The d202Hg of PHg at reference site R1 was 1.77 ± 0.04‰ (2SD, n ¼ 3), similar to seawater at another reference site R2 (1.70 ± 0.08‰, 2SD, n ¼ 3), located 12.0 km away from S1. This is consistent with previous data on coastal background sediments, which show low d202Hg values ranging from 3.09‰ to 0.75‰ (Foucher et al., 2013; Mil-Homens et al., 2013; Sonke et al., 2010; Yin et al., 2015). The d202Hg of the particles increased from 1.32‰ at inlet P1 to 0.31‰ at outlet P2 of the De-SO2 tower, and it did not have significantly variation during aeration process. Once released, the
De-SO2 seawater was mixed with fresh seawater. As distance increased, the d202Hg of PHg became more negative, gradually approaching background (R1). Decreased d202Hg has been observed in coastal sediments and were attributed to such mixing effect. Redox and volatilization of Hg in seawater could also cause small variations in d202Hg. As shown in Fig. 2, d202Hg decreased with the decrease of the PHg concentration. The d202Hg of PHg was 1.11 ± 0.04‰ (2SD, n ¼ 2) at 400 m (S9~S10) away from the outlets, and dropped to 1.44 ± 0.16‰ (2SD, n ¼ 5) beyond 400 m (S11~S15), but was still 0.33‰ higher than the reference site R1 (1.77 ± 0.04‰, 2SD, n ¼ 3). With the absence of other local anthropogenic pollution source, our data suggested that seawater at an area of 2.0 km away from the discharge outlets was may be influenced by the De-SO2 seawater. This is much larger than that revealed by Hg concentrations (within 300 m from the outlets), suggesting that Hg isotopes may be more reliable to source tracing, especially in areas with low Hg concentrations.
3.4. MDF of PHg and DHg in seawater The isotopic composition of DHg in seawater were investigated at the fresh seawater inlet (P1), tower outlet (P2), aeration pools (P3 and P4), the outlets (S1 and S2), 50 m (S3) and 100 m (S4) away from the discharging outlets as shown in Fig. 3 and Table S1 in the Supplementary. From site P2 to site S4, d202Hg variations of 0.55‰ (PHg) and 0.71‰ (DHg) were observed. The d202Hg values of PHg (0.87‰) and DHg (0.59‰) at site S4 were still more positive than those of the background. These results indicated that PHg and DHg in the area within 100 m from the discharging outlet was mainly from the De-SO2 seawater. d202Hg values of DHg ranged from 0.94‰ to 0.12‰, showing a similar trend with PHg, however, in general about 0.12‰ to 0.44‰ higher than that of PHg at each site (Fig. 3). As shown in Fig. 3, It reaches its peak (0.12 ± 0.04‰, 2SD, n ¼ 3) at site P2, and then dropped to 0.20 ± 0.06‰ (2SD, n ¼ 3) at P3, 0.18 ± 0.06‰ (2SD, n ¼ 3) at P4, and finally dropped to 0.59 ± 0.08‰ at S4 (2SD, n ¼ 1). Light isotopes prefer to be enriched in solid Hg phase, which has been proven by several studies. It is found that precipitation and adsorption reactions fractionate light Hg isotopes from heavy ones in solution (Foucher et al., 2013). Fractionation of 0.63‰ and 0.32‰ in d202Hg values have been observed during the
Fig. 3. The d202Hg of PHg and DHg at sites P1~P4, S1~S4. (Square symbols represent the d202Hg of PHg, empty triangle symbols represent the d202Hg of DHg, plotted error bars represent the 2SD of d202Hg).
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
precipitation of b-HgS and HgO (Smith et al., 2015). Hg2þ sorption to goethite result in 0.44‰ to 0.30‰ of the absorbent (Jiskra et al., 2012). Thiol-bond Hg have d202Hg values 0.53‰ to 0.62‰ smaller than that of the DHg (HgCl2 and Hg(NO3)2) (Wiederhold et al., 2010). In this study, isotopic variation between dissolved and PHg ranged from 0.12‰ to 0.44‰, indicating that the reactions of adsorption-desorption and precipitation might occur between dissolved and PHg. The variation could also be caused by the Hg0 volatilization of DHg, as its transport from seawater to the air has been seen in the aeration pool and seawater discharge process (Sun et al., 2013a, 2013b).
3.5. MIF of the PHg in seawater The D199Hg/D201Hg ratio is an important parameter revealing the MIF mechanism (Sonke, 2011; Zheng and Hintelmann, 2010). The photo-reduction reactions of Hg2þ and MeHg have D199Hg/ D201Hg ratios of 1.00 (Chen et al., 2012; Das et al., 2013) and 1.36 (Blum and Bergquist, 2007), because of magnetic isotope effect (MIE). In previous studies, D199Hg/D201Hg ratios of 0.98e1.05 are reported for Hg2þ photo-reduction in the samples containing predominantly inorganic Hg2þ (Sherman et al., 2010; Sonke, 2011). In some reactions of MIF resulting from nuclear volume effect (NVE), D199Hg/D201Hg values range from 1.55 to 1.65 (Estrade et al., 2009; Ghosh et al., 2013; Sonke, 2011; Wiederhold et al., 2010; Zheng and Hintelmann, 2010). A small MIF was found in the PHg at all sampling sites except site R1 and R2. Fig. S2 shows a plot of D201Hg versus D199Hg for PHg; D201Hg is highly correlated with D199Hg (R2 ¼ 0.818). This study showed a similar D199Hg/D201Hg of 1.01 with the photo-reduction of Hg2þ, demonstrating that the MIF in PHg could be caused by MIE during Hg2þ photo-reduction in seawater. Comparing with PHg isotopic composition of seawater in the discharging area, relatively large MIF was seen in PHg at the tower outlet (P2) and at the pools (P3 and P4) with D199Hg 0.29 ± 0.04‰ (2SD, n ¼ 9). The D199Hg of Xiamen SY plant feed coal was 0.20 ± 0.06‰ (2SD, n ¼ 3), statistically the same as samples from sites P2~P4 and sites S1~S15 (one way ANOVA, p > 0.05). Hg in three kinds of samples (feed coal, PHg in De-SO2 seawater, and study area seawater), shared a similar MIF signature, suggesting that the signature of MIF was not affected by De-SO2, aeration, and discharging processes. The Hg in the three kinds of samples might come from the same source, with the negative MIF signature resulting from the coal-formation process (Biswas et al., 2008; Yin et al., 2014). A previous study investigated Hg isotopic compositions of differently aged coal deposits (Biswas et al., 2008), revealing that Hg isotopic fingerprints in coal deposits depend on Hg input from the atmosphere and waters or losses during coal formation. MIF signatures can be fractionated by the following reactions: precipitation between Hg and sulfide; redox reaction and adsorption between Hg and organic matters in coal; Hg loss induced by heating; or the scavenging of Hg from hydrothermal solutions during diagenesis processes. The D199Hg/D201Hg ratio in coal samples from China is 1.03 ± 0.07 (2SD, n ¼ 61), mainly due to Hg photochemical reactions prior to combination with coal (Yin et al., 2014). Indian peat is a primary input for the Xiamen SY power plant; the peat consists of partially decomposed plant residues and fully decomposed humus and minerals. Hg in peat may come from surface water, volcanic geothermal activity, and humus decomposition (Lefticariu et al., 2011). With a D199Hg/D201Hg ratio of 1.01, plant coal MIF might have been induced by Hg photochemical reactions, and reactions among Hg, sulfide, and organic species. As such, negative MIF signatures of PHg could result from Hg2þ photoreduction in the sea or Hg reactions during coal formation.
827
3.6. MDF and MIF signature of PHg and DHg in seawater The combination of both Hg-MDF and -MIF signatures provides multidimensional information in discriminating sources of Hg (Blum et al., 2014; Estrade et al., 2010; Wiederhold, 2015). As shown in Fig. 4, plotting of d202Hg versus D199Hg for S1~S15 indicates that the PHg samples could be divided as four groups: S1~S2 (d202Hg: 0.33~0.30‰; D199Hg: 0.18~0.13‰), S3~S8 (d202Hg: 0.85~0.75‰; D199Hg: 0.20~0.11‰), S9~S11 (d202Hg: 1.45~1.09‰; D199Hg: 0.15~0.13‰), and S12~S15 (d202Hg: 1.53~1.12‰; D199Hg: 0.13~0.12‰). The four groups had similar D199Hg but obviously different d202Hg, the significant variation of d202Hg was proved to be caused by the mixing with fresh seawater (section 3.4). Therefore, the Hg in the study sea area came from the same source. In Fig. 4, the data can be grouped into three regions. PHg of P2~P4 and R1 mark the two ends; PHg of S1~S15 scatter in the middle. The variation of d202Hg in PHg at sites S1~S15 was the result of mixing with De-SO2 seawater (with higher d202Hg of 0.32‰ and negative D199Hg of 0.28‰) by background fresh seawater (with lower d202Hg of 1.77‰ and D199Hg of 0.01‰). Such mixing has been reflected by the spatial variations of THg and other environmental parameters (such as SO2 3 concentration and temperature) as mentioned in Sections 3.2 and 3.3. The different isotopic signatures of PHg and DHg at P1~P4 and S1~S4 scattered in different regions in Fig. 4, suggesting that PHg and DHg isotopes behaved differently in the De-SO2 and discharging process. The denitration, electrostatic precipitation and De-SO2 process could fractionate the Hg isotopes for different species (Sun et al., 2013c), resulting in major difference of isotopic signatures between PHg and DHg. The scatter of DHg was similar to that of PHg in the sea area, with the same reasons; the DHg input would have the same source: De-SO2 seawater and fresh seawater.
3.7. Quantification of Hg source contribution using Hg isotope binary mixing model Based on well-estimated Hg isotope signatures for different Hg sources, a Hg isotope based mixing model, has been successfully developed to trace and quantify Hg sources in different environmental systems (e.g. soils, sediments, lichens, etc) by numerous studies (Das et al., 2013; Estrade et al., 2010; Feng et al., 2010; Foucher and Hintelmann, 2009; Gratz et al., 2010; Perrot et al., 2010).
Fig. 4. D199Hg versus d202Hg for PHg and DHg in seawater samples. (Square symbols represent the isotopic data of PHg, empty triangle symbols represent the isotopic data of DHg, and plotted error bars represent the 2SD of d202Hg or D199Hg).
828
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
d202 HgM ¼ XA d202 HgA þ XB d202 HgB
(4)
100% ¼ XA þ XB
(5)
The relevant data for the De-SO2 seawater (component A,
d202HgA 0.32) and the local background (component B, d202HgB
Fig. 5. d202Hg versus 1/HgPHg for PHg samples. (plotted error bars represent the 2SD of d202Hg).
A previous study by Sun et al. (2013b) revealed that only 2.2% of the THg from the SY power plant was volatized as Hg0. Such small amount of Hg0 losses may cause very limited changes in Hg isotopes although Hg-MDF has been reported during Hg(0) volatilization process (Zheng et al., 2007). Therefore, mixing of Hg from De-SO2 seawaters and fresh open seawater could be the main reason for the variations of concentrations and isotopic compositions of PHg, thus the two-component mixing model (see Supplementary) could be used in this study. Plotting of d202Hg versus 1/HgPHg in Fig. 5 shows a fitting equation of d202Hg ¼ 1.64/HgPHg0.38 (R2 ¼ 0.947, P < 0.001). Two end-members, marked as a and b, can be found in Fig. 5, corresponding to the seawaters of the power plant (e.g., P2~P4, S1~S2) and reference R1, respectively. Data of sites S3~S15 located between the two end-members. The fitted curve with a high coefficient further suggested that the PHg in the discharge area came from both De-SO2 seawater and local background. Using traditional isotopic mass balance equations, the contribution of the two sources was quantified.
-1.77, HgB 1.10) were taken into consideration in equation 4 and 5, where XA and XB represent the fractions of seawater from the DeSO2 and the background seawater. Fig. 6 shows the model output results, which shows that Hg originate from the De-SO2 seawater at outlet of the De-SO2 tower (P2) and in the aeration pools (P3, P4) was 99.6e99.7%, 40% of PHg in seawater at site P1 is from De-SO2 seawater. De-SO2 seawater accounts 33.3e98.4% of the PHg in fresh seawater within 2.0 km from the outlets (S1~S15). The fraction of Hg from De-SO2 seawater declined to approximately 33% within 2.0 km from the outlets, where background contribution increased to 67%. The model output in general well agrees with the spatial pattern of THg in Fig. 2, however, it indicates much larger areas of Hg contamination by the power plant, compared to previous estimation that Hg pollution by De-SO2 seawater were only pronounced within ~300 m from the outlets. Our Hg isotope binary mixing model suggests that 30e50% of PHg within 2.0e3.2 km away from the outlets may be originate from De-SO2 seawater. This highlighted that Hg isotopes may be more reliable to source tracing, especially in areas with low Hg concentrations.
4. Conclusion This study for the first time investigated Hg isotope composition in seawater from and near a SFGD-driven power plant. We found that PHg in De-SO2 seawater have relative higher d202Hg values, whereas mixing with fresh open seawater significantly leads to the decrease of d202Hg in the surrounding seawaters of the power plant. By applying a binary mixing model, we successfully evaluated the fractions of two Hg sources from both De-SO2 seawater and local background. The fractions of De-SO2 Hg in the discharging area progressively decreased from 80% (50 m away, S3) to 40% (2.0 km away, S15) as distance from the outlets increased; De-SO2
Fig. 6. Contribution fraction of PHg in seawater from De-SO2 seawater and background seawater.
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
seawater ultimately impacted an area of 2.0e3.2 km away from the discharging outlets. Compared to previous estimated results based on THg mass balance, our results indicate much larger sea area was impacted by Hg contamination from De-SO2 seawater. The study presented a practical technique for assessing mercury pollution of seawater discharging from the coal-fired power plants with typical SFGD systems, and explored biogeochemical Hg cycling in seawater environments. More attention should be paid to Hg pollution discharging from such power plants. Acknowledgements This research was financed by the Natural Science Foundation of China (21277112). The authors would like to thank Joel D. Blum for n in-house secondary standard. providing the UM-Almade Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.envpol.2016.04.059. References Bergquist, B.A., Blum, J.D., 2009. The odds and evens of mercury isotopes: applications of mass-dependent and mass-independent isotope fractionation. Elements 5, 353e357. Biswas, A., Blum, J.D., Bergquist, B.A., Keeler, G.J., Xie, Z., 2008. Natural mercury isotope variation in coal deposits and organic soils. Environ. Sci. Technol. 42, 8303e8309. Blum, J.D., Bergquist, B.A., 2007. Reporting of variations in the natural isotopic composition of mercury. Anal. Bioanal. Chem. 388, 353e359. Blum, J.D., Sherman, L.S., Johnson, M.W., 2014. Mercury isotopes in earth and environmental sciences. Annu. Rev. Earth Planet. Sci. 42, 249e269. Chen, J., Hintelmann, H., Feng, X., Dimock, B., 2012. Unusual fractionation of both odd and even mercury isotopes in precipitation from Peterborough, ON, Canada. Geochim. Cosmochim. Acta 90, 33e46. Choe, K.Y., Gill, G.A., 2003. Distribution of particulate, colloidal, and dissolved mercury in San Francisco Bay estuary. 2. Monomethyl mercury. Limnol. Oceanogr. 48, 1547e1556. Das, R., Bizimis, M., Wilson, A.M., 2013. Tracing mercury seawater vs. atmospheric inputs in a pristine SE USA salt marsh system: mercury isotope evidence. Chem. Geol. 336, 50e61. Estrade, N., Carignan, J., Donard, O.F., 2010. Isotope tracing of atmospheric mercury sources in an urban area of northeastern France. Environ. Sci. Technol. 44, 6062e6067. Estrade, N., Carignan, J., Sonke, J.E., Donard, O.F., 2009. Mercury isotope fractionation during liquidevapor evaporation experiments. Geochim. Cosmochim. Acta 73, 2693e2711. Feng, C., Gao, X., Tang, Y., Zhang, Y., 2014. Comparative life cycle environmental assessment of flue gas desulphurization technologies in China. J. Clean. Prod. 68, 81e92. Feng, X., Foucher, D., Hintelmann, H., Yan, H., He, T., Qiu, G., 2010. Tracing mercury contamination sources in sediments using mercury isotope compositions. Environ. Sci. Technol. 44, 3363e3368. Foucher, D., Hintelmann, H., 2006. High-precision measurement of mercury isotope ratios in sediments using cold-vapor generation multi-collector inductively coupled plasma mass spectrometry. Anal. Bioanal. Chem. 384, 1470e1478. Foucher, D., Hintelmann, H., 2009. Tracing mercury contamination from the Idrija mining region (Slovenia) to the Gulf of Trieste using Hg isotope ratio measurements. Environ. Sci. Technol. 43, 33e39. Foucher, D., Hintelmann, H., Al, T.A., MacQuarrie, K.T., 2013. Mercury isotope fractionation in waters and sediments of the Murray Brook mine watershed (New Brunswick, Canada): tracing mercury contamination and transformation. Chem. Geol. 336, 87e95. Galbreath, K.C., Zygarlicke, C.J., 2000a. Mercury transformations in coal combustion flue gas. Fuel Process. Technol. 65e66, 289e310. Galbreath, K.C., Zygarlicke, C.J., 2000b. Mercury transformations in coal combustion flue gas. Fuel Process. Technol. 65, 289e310. Ghosh, S., Schauble, E.A., Couloume, G.L., Blum, J.D., Bergquist, B.A., 2013. Estimation of nuclear volume dependent fractionation of mercury isotopes in equilibrium liquidevapor evaporation experiments. Chem. Geol. 336, 5e12. Gratz, L.E., Keeler, G.J., Blum, J.D., Sherman, L.S., 2010. Isotopic composition and fractionation of mercury in Great Lakes precipitation and ambient air. Environ. Sci. Technol. 44, 7764e7770. Hower, J.C., Senior, C.L., Suuberg, E.M., Hurt, R.H., Wilcox, J.L., Olson, E.S., 2010. Mercury capture by native fly ash carbons in coal-fired power plants. Prog. Energy Combust. Sci. 36, 510e529. Jiskra, M., Wiederhold, J.G., Bourdon, B., Kretzschmar, R., 2012. Solution speciation
829
controls mercury isotope fractionation of Hg (II) sorption to goethite. Environ. Sci. Technol. 46, 6654e6662. Kritee, K., Blum, J.D., Reinfelder, J.R., Barkay, T., 2013. Microbial stable isotope fractionation of mercury: a synthesis of present understanding and future directions. Chem. Geol. 336, 13e25. Lefticariu, L., Blum, J.D., Gleason, J.D., 2011. Mercury isotopic evidence for multiple mercury sources in coal from the Illinois Basin. Environ. Sci. Technol. 45, 1724e1729. Liang, Y., Liu, X., Yuan, D., Gong, Z., Zhang, Z., 2010. Mercury species in seawater and sediment of Xiamen western sea area adjacent to a coal-fired power plant. Water Environ. Res. 82, 335e341. Lin, H., Yuan, D., Lu, B., Huang, S., Sun, L., Zhang, F., Gao, Y., 2015. Isotopic composition analysis of dissolved mercury in seawater with purge and trap preconcentration and a modified Hg introduction device for MC-ICP-MS. J. Anal. At. Spectrom. 30, 353e359. Liu, X., Sun, L., Yuan, D., Yin, L., Chen, J., Liu, Y., Liu, C., Liang, Y., Lin, F., 2011. Mercury distribution in seawater discharged from a coal-fired power plant equipped with a seawater flue gas desulfurization system. Environ. Sci. Pollut. Res. 18, 1324e1332. rio, J., Caetano, M., Costa, A.M., Lebreiro, S.M., Mil-Homens, M., Blum, J., Cana Trancoso, M., Richter, T., Stigter, H.D., Johnson, M., 2013. Tracing anthropogenic Hg and Pb input using stable Hg and Pb isotope ratios in sediments of the central Portuguese Margin. Chem. Geol. 336, 62e71. Morel, F.M., Kraepiel, A.M., Amyot, M., 1998. The chemical cycle and bioaccumulation of mercury. Annu. Rev. Ecol. Syst. 543e566. Perrot, V., Epov, V.N., Pastukhov, M.V., Grebenshchikova, V.I., Zouiten, C., Sonke, J.E., Husted, S., Donard, O.F., Amouroux, D., 2010. Tracing sources and bioaccumulation of mercury in fish of Lake Baikal-Angara River using Hg isotopic composition. Environ. Sci. Technol. 44, 8030e8037. Sherman, L.S., Blum, J.D., Johnson, K.P., Keeler, G.J., Barres, J.A., Douglas, T.A., 2010. Mass-independent fractionation of mercury isotopes in Arctic snow driven by sunlight. Nat. Geosci. 3, 173e177. Sherman, L.S., Blum, J.D., Keeler, G.J., Demers, J.D., Dvonch, J.T., 2012. Investigation of local mercury deposition from a coal-fired power plant using mercury isotopes. Environ. Sci. Technol. 46, 382e390. Smith, R.S., Wiederhold, J.G., Kretzschmar, R., 2015. Mercury isotope fractionation during precipitation of metacinnabar (b-HgS) and montroydite (HgO). Environ. Sci. Technol. 49, 4325e4334. Sonke, J.E., 2011. A global model of mass independent mercury stable isotope fractionation. Geochim. Cosmochim. Acta 75, 4577e4590. €fer, J., Chmeleff, J., Audry, S., Blanc, G., Dupre , B., 2010. Sedimentary Sonke, J.E., Scha mercury stable isotope records of atmospheric and riverine pollution from two major European heavy metal refineries. Chem. Geol. 279, 90e100. Stefano, C., Raffaella, P., Joze, K., Milena, H., Jadran, F., Antonio, B., 2006. Behaviour of Hg species in a microtidal deltaic system: the Isonzo River mouth (northern Adriatic Sea). Sci. Total Environ. 368, 210e223. Strok, M., Baya, P.A., Hintelmann, H., 2015. The mercury isotope composition of Arctic coastal seawater. Comptes Rendus Geosci. 347, 368e376. Strok, M., Hintelmann, H., Dimock, B., 2014. Development of pre-concentration procedure for the determination of Hg isotope ratios in seawater samples. Anal. Chim. acta 851, 57e63. Sun, L., Feng, L., Yuan, D., Lin, S., Huang, S., Gao, L., Zhu, Y., 2013a. The extent of the influence and flux estimation of volatile mercury from the aeration pool in a typical coal-fired power plant equipped with a seawater flue gas desulfurization system. Sci. Total Environ. 444, 559e564. Sun, L., Lin, S., Feng, L., Huang, S., Yuan, D., 2013b. The distribution and seaeair transfer of volatile mercury in waste post-desulfurization seawater discharged from a coal-fired power plant. Environ. Sci. Pollut. Res. 20, 6191e6200. Sun, R., Heimbürger, L.-E., Sonke, J.E., Liu, G., Amouroux, D., Berail, S., 2013c. Mercury stable isotope fractionation in six utility boilers of two large coal-fired power plants. Chem. Geol. 336, 103e111. Sun, R., Sonke, J.E., Heimbürger, L.E., Belkin, H.E., Liu, G., Shome, D., Cukrowska, E., Liousse, C., Pokrovsky, O.S., Streets, D.G., 2014. Mercury stable isotope signatures of world coal deposits and historical coal combustion emissions. Environ. Sci. Technol. 48, 7660e7668. Sun, R., Sonke, J.E., Liu, G., 2016. Biogeochemical controls on mercury stable isotope compositions of world coal deposits: a review. Earth Sci. Rev. 152, 1e13. U.S. Environmental Protection Agency, 1996. Method 1669: Sampling Ambient Water for Trace Metals at EPA Water Quality Criteria Levels. Office of Water Washington, DC. Wiederhold, J.G., 2015. Metal stable isotope signatures as tracers in environmental geochemistry. Environ. Sci. Technol. 49, 2606e2624. Wiederhold, J.G., Cramer, C.J., Daniel, K., Infante, I., Bourdon, B., Kretzschmar, R., 2010. Equilibrium mercury isotope fractionation between dissolved Hg (II) species and thiol-bound Hg. Environ. Sci. Technol. 44, 4191e4197. € rn, E., Bourdon, B., Wiederhold, J.G., Skyllberg, U., Drott, A., Jiskra, M., Jonsson, S., Bjo Kretzschmar, R., 2015. Mercury isotope signatures in contaminated sediments as tracer for local industrial pollution sources. Environ. Sci. Technol. 49, 177e185. Yin, R., Feng, X., Chen, B., Zhang, J., Wang, W.-X., Li, X., 2015. Identifying the sources and processes of mercury in subtropical estuarine and ocean sediments using Hg isotopic composition. Environ. Sci. Technol. 49, 1347e1355. Yin, R., Feng, X., Chen, J., 2014. Mercury stable isotopic compositions in coals from major coal producing fields in China and their geochemical and environmental implications. Environ. Sci. Technol. 48, 5565e5574.
830
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
Yin, R., Feng, X., Wang, J., Bao, Z., Yu, B., Chen, J., 2013. Mercury isotope variations between bioavailable mercury fractions and total mercury in mercury contaminated soil in Wanshan Mercury Mine, SW China. Chem. Geol. 336, 80e86. Yu, F., Chen, J., Sun, F., Zeng, S., Wang, C., 2011. Trend of technology innovation in China's coal-fired electricity industry under resource and environmental constraints. Energy Policy 39, 1586e1599.
Zheng, W., Foucher, D., Hintelmann, H., 2007. Mercury isotope fractionation during volatilization of Hg(0) from solution into the gas phase. J. Anal. At. Spectrom. 22, 1097e1104. Zheng, W., Hintelmann, H., 2010. Nuclear field shift effect in isotope fractionation of mercury during abiotic reduction in the absence of light. J. Phys. Chem. A 114, 4238e4245.
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Mercury isotope signatures of seawater discharged from a coal-fired power plant equipped with a seawater flue gas desulfurization system* Haiying Lin a, b, Jingji Peng a, Dongxing Yuan a, *, Bingyan Lu a, Kunning Lin a, Shuyuan Huang a a b
State Key Laboratory of Marine Environmental Science, College of the Environment and Ecology, Xiamen University, Xiamen 361102, China Guangxi Colleges and Universities Key Laboratory of Environmental Protection, School of Environment, Guangxi University, Nanning 530004, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 January 2016 Received in revised form 10 April 2016 Accepted 16 April 2016 Available online 5 May 2016
Seawater flue gas desulfurization (SFGD) systems are commonly used to remove acidic SO2 from the flue gas with alkaline seawater in many coastal coal-fired power plants in China. However, large amount of mercury (Hg) originated from coal is also transferred into seawater during the desulfurization (De-SO2) process. This research investigated Hg isotopes in seawater discharged from a coastal plant equipped with a SFGD system for the first time. Suspended particles of inorganic minerals, carbon residuals and sulfides are enriched in heavy Hg isotopes during the De-SO2 process. d202Hg of particulate mercury (PHg) gradually decreased from 0.30‰ to 1.53‰ in study sea area as the distance from the point of discharge increased. The results revealed that physical mixing of contaminated De-SO2 seawater and uncontaminated fresh seawater caused a change in isotopic composition of PHg isotopes in the discharging area; and suggested that both De-SO2 seawater and local background contributed to PHg. The impacted sea area predicted with isotopic tracing technique was much larger than that resulted from a simple comparison of pollutant concentration. It was the first attempt to apply mercury isotopic composition signatures with two-component mixing model to trace the mercury pollution and its influence in seawater. The results could be beneficial to the coal-fired plants with SFGD systems to assess and control Hg pollution in sea area. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Mercury isotope signatures Fractionation Desulfurization seawater Coal-fired power plant Tracing source
1. Introduction Human activities, such as coal combustion and other industrial emission sources (Foucher and Hintelmann, 2006, 2009; Sherman et al., 2012; Sun et al., 2013a, 2013b, 2013c), have accelerated the emission and geochemical cycling of mercury (Hg) into the environment (Choe and Gill, 2003; Morel et al., 1998). For instance, coal combustion as one of the largest emitters of Hg in the world, has contributed more than 35,000 tons into the atmosphere (Sun et al., 2016). Coal usually contains high levels of sulfur and trace amount of Hg. In China and other countries in the world, many coal-fired power plants along/near the coast are equipped with the seawater flue gas desulfurization (SFGD) system, which use the
*
This paper has been recommended for acceptance by Prof. W. Wen-Xiong. * Corresponding author. E-mail address: [email protected] (D. Yuan).
http://dx.doi.org/10.1016/j.envpol.2016.04.059 0269-7491/© 2016 Elsevier Ltd. All rights reserved.
alkaline seawater to neutralize acidic sulfur dioxide (SO2) in the flue gas (Feng et al., 2014; Yu et al., 2011). During desulfurization (DeSO2), the flush of sea water also washes Hg off the flue gas, resulting in high Hg content in the seawater (Liang et al., 2010; Liu et al., 2011). A schematic diagram showing the process and Hg transfer from the coal combustion and the SFGD system into the estuary is provided in Fig. S1 of Supplementary. The De-SO2 seawater is mixed with fresh seawater and treated with simple aeration to further neutralize and oxidize SO2 3 , increase pH and dissolved oxygen (DO) (Liu et al., 2011), and then discharged into the adjacent sea. The technique utilizes local seawater resource and avoids producing huge amount of solid waste that is usually generated with common dry desulfurization techniques. The De-SO2 seawater mainly contains relatively high levels of suspended particles and SO2 4 , which are considered not to be the pollutants for coastal environment. However, heavy metals, especially Hg, at high concentration in the De-SO2 seawater could become a potential threat to the ecosystem. A typical air purification system of coal-fired flue gas is
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
composed of a selective catalytic reduction unit, a particulate control unit and a flue gas desulfurization system in series (Hower et al., 2010), installing after the boiler (Fig. S1 in the Supplementary). During the coal combustion (>1200 C) in the boiler, Hg in coal is released as Hg0 into the flue gas (Galbreath and Zygarlicke, 2000b). The selective catalytic reduction unit is used for removal of nitrogen oxides, and at the same time it transfers 30e98% Hg0 into Hg2þ, which is easily to adsorb onto particles and form PHg in the flue gas (Hower et al., 2010). Part of particles and PHg are removed in the particulate control unit. When the flue gas flows through the flue gas desulfurization system, 30e50% PHg, most of Hg2þ and part of Hg0 are transferred into by-products such as gypsum or seawater (Hower et al., 2010). The SY power plant, located in the Xiamen city (southeast coast of China, Fig. 1), is one typically example of coal-fired power plant equipped with SFDG that becomes a Hg emission source (Liang et al., 2010). For instance, the De-SO2 seawater at the outlet of the De-SO2 tower contains total Hg (THg) up to 385.0 ng/L, much higher than China's Class Ⅱ seawater quality standard (200.0 ng/L). According to our previous studies, the De-SO2 seawater can bring 1.03 kg/day of Hg into the adjacent sea (Liu et al., 2011). The result of regular monitoring indicated seawater within approximately 300 m of the SY power plant has shown slightly higher Hg content than that of that in other coastal areas in Xiamen. However, the environmental impact of Hg from the De-SO2 seawater is still a question because of the mismatch of the Hg in the De-SO2 seawater and the discharging sea area; based on the simply Hg mass balance calculations. It is thought that the mixing with local Hg background can result in decreased THg in seawaters of the discharging area, whereas there is a lack of effective tracer to quantify the contribution of Hg from De-SO2 seawater and background. It is not yet clear how large a sea area is influenced by the Hg-contaminated DeSO2 seawater. During the past 15 years, mercury isotope geochemistry has been proved to be a powerful tool to trace sources and geochemical processes of Hg in the environment (Bergquist and Blum, 2009;
823
Blum et al., 2014; Wiederhold et al., 2015). Previous studies have investigated Hg isotopic compositions of coal from different regions of the world (Biswas et al., 2008; Lefticariu et al., 2011; Sun et al., 2014, 2016; Yin et al., 2014), Hg isotope fractionation during coal combustion process (Sun et al., 2013c), and isotopic signature of Hg in atmospheric samples near a coal power plant (Sherman et al., 2012). These studies have facilitated our understanding on Hg transformation, transportation, and environmental impacts during Hg emission by coal combustion. However, Hg isotope fractionation during Hg emission by SFGD is still not well understood. A main technological gap for this is that Hg in seawater is generally too low for Hg isotope analysis. Analytical methods for water Hg isotope determination have been developed recently (Lin et al., 2015; Strok et al., 2014, 2015). To date, only a few studies have reported Hg isotopic compositions in seawaters (Strok et al., 2015); whereas isotopic compositions in the De-SO2 seawater have not been studied. Particulate Hg (PHg) is the main species in water systems and its concentration in natural water varies in different regions around the world. Concentrations of PHg in fresh water of the Isonzo River mouth range from 9.0 to 25.0 ng/L, contributing 51e84% to THg (Stefano et al., 2006). In the seawater of discharging sea area near the SY power plant, 50e97% of THg is in PHg form (Liang et al., 2010; Lin et al., 2015) due to the fact that De-SO2 seawater contains large amount suspended particles, mainly inorganic minerals, carbon residuals and sulfides, generated during the coal combustion process and not removed by the particulate control unit. The particles containing carbon, acid and sulfides have a high affinity to bond Hg (Galbreath and Zygarlicke, 2000a). Therefore, high concentration of PHg was observed in the De-SO2 seawater. This study investigated the variations of Hg isotope compositions in feed coal used in the plant, PHg and DHg in De-SO2 seawater discharged; discussed the transfer and transformation of PHg and DHg in the seawater, and evaluated the source impact of Hg by SFGD. The results of this study could provide information for better understanding environmental behavior of Hg discharged
Fig. 1. Location of the sampling sites.
824
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
from the coal-fired plants with SFGD systems, and could be benefit the evaluation of Hg environment impacts, and to guide future Hg remediation in the studied area.
2. Material and method 2.1. Sampling sites The Xiamen SY power plant is located in Xiamen Bay, in the Jiulongjiang Estuary of Fujian province, China (Fig. 1). Jiulongjiang is the second largest river of Fujian province, running through Xiamen Bay into the Taiwan Strait. Seawater from Xiamen Bay is introduced into the plant for SFGD; the resulted De-SO2 seawater is then mixed with one third fresh seawater, and aerated in two parallel aeration pools to neutralize the SO2 3 concentration, increase the pH and DO level (Liu et al., 2011). After that, the De-SO2 seawater is discharged into the adjacent sea. Four sites within the SY power plant were chosen to collected the samples, which include the fresh seawater inlet (P1), the desulfurization tower outlet (P2), and the two aeration pools (P3, P4). Seawater from additional 15 sites (S1~S15) within 2.0 km of the outlets of the plant were also sampled (Fig. 1). These sites include two sites (S1, S2) at the discharging outlets, and sites (S3~S15) with increasing distances (50, 100, 200, 300, 400, 600, 1500, and 2000 m) off the outlets. The site R1 (Fig. 1) was selected 3.2 km away from the outlets to assess the background. Another site R2 located near the Huangcuo (Xiamen seashore), 12.0 km away from the outlets, was also chosen as a reference site. Surface seawater samples of each site were collected at March 28 and June 18, 2014. The samples were collected at low tide in order to better avoid the influence of fresh seawater and gain access the actual impact sea area, as well as to collect particles at relatively higher concentration and lower the difficulty in analysis of the PHg isotopic composition. Triplicate samples were collected at sites P1~P4, S3, R1~R2; one sample was collected in the other sites. Triplicate samples of feed coal were also collected.
2.2. Sampling and sample pretreatment Sampling and sample preparation were performed following U.S. EPA methods 1669 (U.S. Environmental Protection Agency, 1996). Seawater samples were collected and then pre-oxidized with BrCl solution in the field, which were used for THg concentration analysis. To measure DHg and PHg, seawater samples of 5e10 L were collected and filtered through a 0.45 mm acetate cellulose membrane. The filtrate was used to determine the DHg concentration of the sample following a method developed by Liang et al. (2010). The suspended particles on the membrane were defined as the PHg samples, and the sample together with its filter membrane was immediately soaked in a 5 mL BrCl (5%, v/v) solution after filtration for further treatment. PHg (together with the membrane) samples and coal samples were digested in aqua regia (HCl:HNO3, v/v, 3:1) at 95 C for 3 h, and then diluted with BrCl solution (1%, v/v) after being cooled to room temperature. The samples were diluted to ensure the acid concentrations below 25%(v/v) (Foucher et al., 2013). The digest solution was centrifuged, and the supernatant was used to determine Hg concentration and isotopic composition. For DHg samples with low Hg concentrations, however, Hg was pre-concentrated into a KMnO4 solution using a purge and trap method (Lin et al., 2015). The pre-concentration recoveries of samples and spike standards were in the range of 94.3e105.5%.
2.3. Instrumentation and method A double amalgamation combined with a cold vapor atomic fluorescence spectrometer (Rayleigh Analytical Instrument Corp., China), was used to analyze THg, PHg, and DHg samples (Liang et al., 2010; Liu et al., 2011). A water quality monitoring device (JAP60M/U52-10M, Japan) was used to measure water temperature, pH, salinity, and DO. Mercury isotope compositions were determined using a cold vapor generation (CVG) device combined with a desolvating nebulizer (DSN-100, Nu Instruments, UK), and a multi-collector inductively coupled plasma mass spectrometer (MC-ICP-MS, Nu Plasma HR, Nu Instruments, UK) in State Key Laboratory of Marine Environmental Science of Xiamen University following our previous method (Lin et al., 2015). Instrumental mass bias was corrected using Tl internal standard (NIST SRM 997) and standard-sample bracketing method (Blum and Bergquist, 2007; Sonke, 2011). MDF of Hg isotopes was expressed in d notation as per mil deviations (‰) from NIST SRM 3133 Hg standard (Bergquist and Blum, 2009):
.
h
d Hgð‰Þ ¼ ð Hg=198 HgÞsample ð Hg=198 HgÞnist 31331
i
1000 (1) xxx
199
200
201
202
In this expression, Hg is Hg, Hg, Hg and Hg. MIF of Hg isotopes was expressed as DxxxHg, calculated using the following equation (Kritee et al., 2013; Strok et al., 2014):
Dxxx Hg ¼ d Hg ðd202 Hg bÞ where b is 0.2520, 0.5024 and 0.7520 for respectively.
(2) 199
Hg,
200
Hg,
201
Hg,
2.4. Quality control Quality assurance (QA) blanks indicated that samples were uncontaminated during the sampling and pretreatment processes. The Hg concentration in the procedural blank digestions (0.02e0.05 ng/mL, mean value 0.03 ng/mL) was 4% of the sample digestion solution (0.70e3.00 ng/mL). The Hg concentration in the sediment standard reference material (GBW07304a, GSD-4a) was 0.075 ± 0.004 mg/g, well agreed with the certified value (0.078 ± 0.006 mg/g). The internal precision for 202Hg/198Hg was better than 0.04‰ (1 standard error, 1SE) during Hg isotope measurement. The external precision (uncertainty, 2 standard deviation, 2SD) was assessed by measuring triplicate samples with the same sampling and pretreatment processing (Foucher et al., 2013), or using the UMn standard at the same Hg concentration when no parallel Almade samples were collected (Sun et al., 2013c). The d202Hg, D199Hg and D200Hg for the NIST 3133 and UM-Almaden were 0.00 ± 0.08‰ (2SD, n ¼ 36) and 0.51 ± 0.10‰ (2SD, n ¼ 16); 0.00 ± 0.04‰ (2SD, n ¼ 36) and 0.02 ± 0.04‰ (2SD, n ¼ 16); 0.01 ± 0.04‰ (2SD, n ¼ 36) and 0.02 ± 0.02‰ (2SD, n ¼ 16), respectively. The internal and external precisions were similar to previous studies (Foucher and Hintelmann, 2009; Foucher et al., 2013; Yin et al., 2013). 3. Results and discussion 3.1. Distribution of THg and PHg THg, PHg, and DHg in the seawater sampled for this study ranged in 3.3e389.5 ng/L, 1.0e335.4 ng/L, and 0.6e54.0 ng/L, respectively. Mass balance shown that the sum of PHg and DHg
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
concentration correspond to 93e104% of the THg concentration. Before the De-SO2 process, The THg concentration (3.3 ng/L) in the Xiamen SY plant seawater inlet (P1) was relatively low, similar to that at reference site R1 (3.0 ng/L, Fig. 2). Seawater collected at the De-SO2 tower outlet (P2) has the highest THg concentration (389.5 ng/L), which is 130 times higher than that from inlet seawater (P1). PHg is the predominate speciation of Hg in the seawater, representing approximately 86% of THg in the outlet (P2). When the fresh seawater was sprayed onto the flue gas in the DeSO2 tower, PHg and Hg2þ in the flue gas were washed into the seawater, causing significant increase of THg and PHg. In the aeration pools (P3, P4), the mixing of De-SO2 seawater with 1/3 fresh seawater resulted in the decreasing of Hg concentrations. PHg and THg concentrations dropped to 245.5 ng/L and 266.4 ng/L, respectively. Meanwhile, DHg in seawater decreased from 54.0 ng/L to 20.9 ng/L, possibly due to the mixing effect of the fresh seawater combined the evasion of DHg from seawater to the atmosphere (Sun et al., 2013a). THg concentrations in pools P3, P4 were 262.0 ng/L and 271.0 ng/L, respectively, higher than the threshold value (200.0 ng/L) regulated by China's Class Ⅱ seawater quality, suggesting that the De-SO2 seawater may be a Hg contamination source. THg and PHg concentrations in seawater decreased rapidly as increased distance from the outlets. THg in surface seawater at outlets S1 and S2 (60.0e90.0 ng/L) dropped to ~10.0 ng/L at sites S3~S8 (50e300 m away from the outlets). This level is still higher than that of the reference site R1 (3.0 ng/L), indicating being influenced by the De-SO2 seawater. At sites 400e2000 m away from the outlets, THg concentrations in seawater were low and constant (3.7 ng/L), and similar to that at reference site R1 (3.3 ng/ L). The differences in THg concentrations between seawater at 400e2000 m away and reference site R1 were statistically insignificant (within standard error range). As such, there was no conclusive evidence to prove that seawater 400e2000 m away from the outlets was influenced by De-SO2 seawater. Ratios of PHg to THg (PHg/THg) were 86e94% at P2~P4, which was agreed with the previous study (Liang et al., 2010; Liu et al., 2011). Most of Hg in the coal is transferred into Hg0 during the combustion, and when the flue gas cools down and passes through the denitration device, large amount of Hg0 is oxidized into Hg2þ. Hg2þ is easily to adsorb or bond onto particles as the PHg (Sun et al., 2013c), which made PHg become the main form of the flue gas.
825
Table 1 Correlation coefficients (R) of ancillary parameters and Hg concentration. Correlation coefficient (R, n ¼ 19) SO2 3 (mg/L) THg (ng/L) PHg (ng/L)
b
0.963 0.831b
Temperature ( C) a
0.658 0.759b
pH b
0.825 0.697a
DO (mg/L)
Salinity
0.378 0.492
0.428 0.438
a
Statistically significantly correlated at confidence level of 0.05. Significantly correlated at confidence level of 0.01. Analyses were conducted using the Spearman analysis method with the SPSS software package. b
Most of PHg, Hg2þ and part of Hg0 could be washed into the seawater during the De-SO2 process, leading to high PHg concentration and higher PHg/THg. Although the PHg/THg decreased from 78% to 51% within 2.0 km from the outlets, PHg was still the main speciation. During discharge and transport into the open seawater, suspended particles in De-SO2 seawater either sunk into the seabed or mixed with fresh seawaters, which may explain the decrease of PHg/THg with increased distance from the outlets. Most of particles in De-SO2 seawater were in small size (micrometer level), therefore the sedimentation should be insignificant in the estuary. 3.2. Relationship between the ancillary parameters and Hg concentration Some estuary parameters, including water temperature, SO2 3 concentration, pH, DO and salinity, were monitored on field to find out their relationship with Hg concentration. A gradient was found for water temperature, SO2 3 concentration, pH, THg and PHg, from the De-SO2 tower outlet P2 to the discharging point, and then to the sea area. Table 1 shows the correlation coefficients between the ancillary parameters and Hg concentration at each sampling site (n ¼ 19); THg and PHg were positively correlated with water temperature and SO2 concentration, and negatively correlated 3 with pH. There were no correlations between Hg and DO and salinity. The correlations among THg, PHg concentration, seawater temperature, SO2 concentration and pH may be explained by 3 similar processes during the De-SO2 and aeration. During the DeSO2 process, SO2 and Hg were removed from the flue gas by seawater, resulting high concentrations of SO2 3 (16 mg/L), THg (389.0 ng/L), and PHg (335.0 ng/L) in seawater at De-SO2 tower
Fig. 2. THg and PHg concentrations at sampling sites. (The plotted dots represent d202Hg of PHg, error bars represent 2SD of d202Hg values, Line 1 and Line 2 represent the decreasing trend of THg concentration and d202Hg).
826
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
outlet P2. The high temperature in the De-SO2 tower boiled the seawater to 100 C. The acid gas (HCl, SO2) in the flue gas were removed by the seawater, resulting in the decrease of the pH to 3. During fresh seawater mixing and aeration, THg and PHg and SO2 3 concentrations, temperature decreased whereas pH increased at sites P3 and P4. Fresh seawater for use in the SFGD was collected from the nearby sea. Seawater salinity remained constant during the De-SO2 process, aeration, and discharge, and there were lower variations in DO than in THg and PHg. As such, no correlations existed between salinity, DO, and Hg concentrations. 3.3. Mercury isotope signature of PHg Fig. 2 and Table S1 in SI show the spatial variations of d202Hg for suspended particles. De-SO2 seawater particles collected at tower outlet P2 show more positive d202Hg value (0.31 ± 0.04‰, 2SD, n ¼ 3) than that of fresh seawater at P1 (d202Hg 1.32 ± 0.02‰, 2SD, n ¼ 3). The higher d202Hg value of PHg at P2 was reflective of oxidized Hg in the flue gas, because most of Hg in the De-SO2 seawater came from the coal-fired flue gas. Much higher d202Hg in the flue gas (0.31 ± 0.04‰, 2SD, n ¼ 3) than the feed coal (1.35 ± 0.10‰, 2SD, n ¼ 3) have been observed, indicating isotope fractionation of Hg during coal combustion and flue gas purification. This was inconsistent with a previous study (Sun et al., 2013c) about a dry desulfurization system, where oxidized Hg species in bottom ash, fly ash (d202Hg: 1.96 to 0.82‰) and by-product gypsum (d202Hg: 0.99 to 0.47‰) were enriched with light Hg isotopes relative to feed coal (d202Hg: 0.67 to 0.18‰). The disagreement might be due to the differences of the feed coals and desulfurization techniques used in the two study cases. The feed coal in previous study has relatively higher d202Hg than that in this study, ranging from 0.67‰ to 0.18‰. The process of dry desulfurization is very different from SFDG. The MDF of PHg in this present case might result from Hg thermal combustion, evaporation, diffusion, adsorption, and redox during coal-combustion and air purification processes. The d202Hg (0.38 ± 0.16‰, 2SD, n ¼ 6) of the PHg in the aeration pool (P3, P4) were statistically insignificant from P2 (0.32 ± 0.04‰, 2SD, n ¼ 3), indicating that aeration may not cause PHg isotope fractionation. PHg at discharging outlets S1 and S2 had a similar d202Hg (0.30 ± 0.10‰, 0.33 ± 0.10‰, 2SD, n ¼ 3) as tower outlet P2 and aeration pools P3 and P4 (0.36 ± 0.14‰, 2SD, n ¼ 9), which is well agreed with the fact that those sites mainly received Hg from the power plant, as proven by Hg concentrations. d202Hg values decreased from 0.68‰ (S3) to 0.85‰ (S8), and d202Hg values (1.35 ± 0.36‰, 2SD, n ¼ 5) at S9~S15 were more negative. The overall variation of d202Hg values (up to 1.23‰) is significant compared to the analytical uncertainty of our method (±0.10‰). In addition, fresh seawater inlet P1 and sea area sites S11~S15 have similar d202Hg values, suggestive of less impact by the power plant. In fact, the fresh seawater for De-SO2 came from an area 1.0 km from the outlets, where was the location of sites S11~S15. Turning to the background data, reference site R1 was located near an uninhabited island, 3.2 km away from S1, where seawater is relatively clean and protected from human activities. The d202Hg of PHg at reference site R1 was 1.77 ± 0.04‰ (2SD, n ¼ 3), similar to seawater at another reference site R2 (1.70 ± 0.08‰, 2SD, n ¼ 3), located 12.0 km away from S1. This is consistent with previous data on coastal background sediments, which show low d202Hg values ranging from 3.09‰ to 0.75‰ (Foucher et al., 2013; Mil-Homens et al., 2013; Sonke et al., 2010; Yin et al., 2015). The d202Hg of the particles increased from 1.32‰ at inlet P1 to 0.31‰ at outlet P2 of the De-SO2 tower, and it did not have significantly variation during aeration process. Once released, the
De-SO2 seawater was mixed with fresh seawater. As distance increased, the d202Hg of PHg became more negative, gradually approaching background (R1). Decreased d202Hg has been observed in coastal sediments and were attributed to such mixing effect. Redox and volatilization of Hg in seawater could also cause small variations in d202Hg. As shown in Fig. 2, d202Hg decreased with the decrease of the PHg concentration. The d202Hg of PHg was 1.11 ± 0.04‰ (2SD, n ¼ 2) at 400 m (S9~S10) away from the outlets, and dropped to 1.44 ± 0.16‰ (2SD, n ¼ 5) beyond 400 m (S11~S15), but was still 0.33‰ higher than the reference site R1 (1.77 ± 0.04‰, 2SD, n ¼ 3). With the absence of other local anthropogenic pollution source, our data suggested that seawater at an area of 2.0 km away from the discharge outlets was may be influenced by the De-SO2 seawater. This is much larger than that revealed by Hg concentrations (within 300 m from the outlets), suggesting that Hg isotopes may be more reliable to source tracing, especially in areas with low Hg concentrations.
3.4. MDF of PHg and DHg in seawater The isotopic composition of DHg in seawater were investigated at the fresh seawater inlet (P1), tower outlet (P2), aeration pools (P3 and P4), the outlets (S1 and S2), 50 m (S3) and 100 m (S4) away from the discharging outlets as shown in Fig. 3 and Table S1 in the Supplementary. From site P2 to site S4, d202Hg variations of 0.55‰ (PHg) and 0.71‰ (DHg) were observed. The d202Hg values of PHg (0.87‰) and DHg (0.59‰) at site S4 were still more positive than those of the background. These results indicated that PHg and DHg in the area within 100 m from the discharging outlet was mainly from the De-SO2 seawater. d202Hg values of DHg ranged from 0.94‰ to 0.12‰, showing a similar trend with PHg, however, in general about 0.12‰ to 0.44‰ higher than that of PHg at each site (Fig. 3). As shown in Fig. 3, It reaches its peak (0.12 ± 0.04‰, 2SD, n ¼ 3) at site P2, and then dropped to 0.20 ± 0.06‰ (2SD, n ¼ 3) at P3, 0.18 ± 0.06‰ (2SD, n ¼ 3) at P4, and finally dropped to 0.59 ± 0.08‰ at S4 (2SD, n ¼ 1). Light isotopes prefer to be enriched in solid Hg phase, which has been proven by several studies. It is found that precipitation and adsorption reactions fractionate light Hg isotopes from heavy ones in solution (Foucher et al., 2013). Fractionation of 0.63‰ and 0.32‰ in d202Hg values have been observed during the
Fig. 3. The d202Hg of PHg and DHg at sites P1~P4, S1~S4. (Square symbols represent the d202Hg of PHg, empty triangle symbols represent the d202Hg of DHg, plotted error bars represent the 2SD of d202Hg).
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
precipitation of b-HgS and HgO (Smith et al., 2015). Hg2þ sorption to goethite result in 0.44‰ to 0.30‰ of the absorbent (Jiskra et al., 2012). Thiol-bond Hg have d202Hg values 0.53‰ to 0.62‰ smaller than that of the DHg (HgCl2 and Hg(NO3)2) (Wiederhold et al., 2010). In this study, isotopic variation between dissolved and PHg ranged from 0.12‰ to 0.44‰, indicating that the reactions of adsorption-desorption and precipitation might occur between dissolved and PHg. The variation could also be caused by the Hg0 volatilization of DHg, as its transport from seawater to the air has been seen in the aeration pool and seawater discharge process (Sun et al., 2013a, 2013b).
3.5. MIF of the PHg in seawater The D199Hg/D201Hg ratio is an important parameter revealing the MIF mechanism (Sonke, 2011; Zheng and Hintelmann, 2010). The photo-reduction reactions of Hg2þ and MeHg have D199Hg/ D201Hg ratios of 1.00 (Chen et al., 2012; Das et al., 2013) and 1.36 (Blum and Bergquist, 2007), because of magnetic isotope effect (MIE). In previous studies, D199Hg/D201Hg ratios of 0.98e1.05 are reported for Hg2þ photo-reduction in the samples containing predominantly inorganic Hg2þ (Sherman et al., 2010; Sonke, 2011). In some reactions of MIF resulting from nuclear volume effect (NVE), D199Hg/D201Hg values range from 1.55 to 1.65 (Estrade et al., 2009; Ghosh et al., 2013; Sonke, 2011; Wiederhold et al., 2010; Zheng and Hintelmann, 2010). A small MIF was found in the PHg at all sampling sites except site R1 and R2. Fig. S2 shows a plot of D201Hg versus D199Hg for PHg; D201Hg is highly correlated with D199Hg (R2 ¼ 0.818). This study showed a similar D199Hg/D201Hg of 1.01 with the photo-reduction of Hg2þ, demonstrating that the MIF in PHg could be caused by MIE during Hg2þ photo-reduction in seawater. Comparing with PHg isotopic composition of seawater in the discharging area, relatively large MIF was seen in PHg at the tower outlet (P2) and at the pools (P3 and P4) with D199Hg 0.29 ± 0.04‰ (2SD, n ¼ 9). The D199Hg of Xiamen SY plant feed coal was 0.20 ± 0.06‰ (2SD, n ¼ 3), statistically the same as samples from sites P2~P4 and sites S1~S15 (one way ANOVA, p > 0.05). Hg in three kinds of samples (feed coal, PHg in De-SO2 seawater, and study area seawater), shared a similar MIF signature, suggesting that the signature of MIF was not affected by De-SO2, aeration, and discharging processes. The Hg in the three kinds of samples might come from the same source, with the negative MIF signature resulting from the coal-formation process (Biswas et al., 2008; Yin et al., 2014). A previous study investigated Hg isotopic compositions of differently aged coal deposits (Biswas et al., 2008), revealing that Hg isotopic fingerprints in coal deposits depend on Hg input from the atmosphere and waters or losses during coal formation. MIF signatures can be fractionated by the following reactions: precipitation between Hg and sulfide; redox reaction and adsorption between Hg and organic matters in coal; Hg loss induced by heating; or the scavenging of Hg from hydrothermal solutions during diagenesis processes. The D199Hg/D201Hg ratio in coal samples from China is 1.03 ± 0.07 (2SD, n ¼ 61), mainly due to Hg photochemical reactions prior to combination with coal (Yin et al., 2014). Indian peat is a primary input for the Xiamen SY power plant; the peat consists of partially decomposed plant residues and fully decomposed humus and minerals. Hg in peat may come from surface water, volcanic geothermal activity, and humus decomposition (Lefticariu et al., 2011). With a D199Hg/D201Hg ratio of 1.01, plant coal MIF might have been induced by Hg photochemical reactions, and reactions among Hg, sulfide, and organic species. As such, negative MIF signatures of PHg could result from Hg2þ photoreduction in the sea or Hg reactions during coal formation.
827
3.6. MDF and MIF signature of PHg and DHg in seawater The combination of both Hg-MDF and -MIF signatures provides multidimensional information in discriminating sources of Hg (Blum et al., 2014; Estrade et al., 2010; Wiederhold, 2015). As shown in Fig. 4, plotting of d202Hg versus D199Hg for S1~S15 indicates that the PHg samples could be divided as four groups: S1~S2 (d202Hg: 0.33~0.30‰; D199Hg: 0.18~0.13‰), S3~S8 (d202Hg: 0.85~0.75‰; D199Hg: 0.20~0.11‰), S9~S11 (d202Hg: 1.45~1.09‰; D199Hg: 0.15~0.13‰), and S12~S15 (d202Hg: 1.53~1.12‰; D199Hg: 0.13~0.12‰). The four groups had similar D199Hg but obviously different d202Hg, the significant variation of d202Hg was proved to be caused by the mixing with fresh seawater (section 3.4). Therefore, the Hg in the study sea area came from the same source. In Fig. 4, the data can be grouped into three regions. PHg of P2~P4 and R1 mark the two ends; PHg of S1~S15 scatter in the middle. The variation of d202Hg in PHg at sites S1~S15 was the result of mixing with De-SO2 seawater (with higher d202Hg of 0.32‰ and negative D199Hg of 0.28‰) by background fresh seawater (with lower d202Hg of 1.77‰ and D199Hg of 0.01‰). Such mixing has been reflected by the spatial variations of THg and other environmental parameters (such as SO2 3 concentration and temperature) as mentioned in Sections 3.2 and 3.3. The different isotopic signatures of PHg and DHg at P1~P4 and S1~S4 scattered in different regions in Fig. 4, suggesting that PHg and DHg isotopes behaved differently in the De-SO2 and discharging process. The denitration, electrostatic precipitation and De-SO2 process could fractionate the Hg isotopes for different species (Sun et al., 2013c), resulting in major difference of isotopic signatures between PHg and DHg. The scatter of DHg was similar to that of PHg in the sea area, with the same reasons; the DHg input would have the same source: De-SO2 seawater and fresh seawater.
3.7. Quantification of Hg source contribution using Hg isotope binary mixing model Based on well-estimated Hg isotope signatures for different Hg sources, a Hg isotope based mixing model, has been successfully developed to trace and quantify Hg sources in different environmental systems (e.g. soils, sediments, lichens, etc) by numerous studies (Das et al., 2013; Estrade et al., 2010; Feng et al., 2010; Foucher and Hintelmann, 2009; Gratz et al., 2010; Perrot et al., 2010).
Fig. 4. D199Hg versus d202Hg for PHg and DHg in seawater samples. (Square symbols represent the isotopic data of PHg, empty triangle symbols represent the isotopic data of DHg, and plotted error bars represent the 2SD of d202Hg or D199Hg).
828
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
d202 HgM ¼ XA d202 HgA þ XB d202 HgB
(4)
100% ¼ XA þ XB
(5)
The relevant data for the De-SO2 seawater (component A,
d202HgA 0.32) and the local background (component B, d202HgB
Fig. 5. d202Hg versus 1/HgPHg for PHg samples. (plotted error bars represent the 2SD of d202Hg).
A previous study by Sun et al. (2013b) revealed that only 2.2% of the THg from the SY power plant was volatized as Hg0. Such small amount of Hg0 losses may cause very limited changes in Hg isotopes although Hg-MDF has been reported during Hg(0) volatilization process (Zheng et al., 2007). Therefore, mixing of Hg from De-SO2 seawaters and fresh open seawater could be the main reason for the variations of concentrations and isotopic compositions of PHg, thus the two-component mixing model (see Supplementary) could be used in this study. Plotting of d202Hg versus 1/HgPHg in Fig. 5 shows a fitting equation of d202Hg ¼ 1.64/HgPHg0.38 (R2 ¼ 0.947, P < 0.001). Two end-members, marked as a and b, can be found in Fig. 5, corresponding to the seawaters of the power plant (e.g., P2~P4, S1~S2) and reference R1, respectively. Data of sites S3~S15 located between the two end-members. The fitted curve with a high coefficient further suggested that the PHg in the discharge area came from both De-SO2 seawater and local background. Using traditional isotopic mass balance equations, the contribution of the two sources was quantified.
-1.77, HgB 1.10) were taken into consideration in equation 4 and 5, where XA and XB represent the fractions of seawater from the DeSO2 and the background seawater. Fig. 6 shows the model output results, which shows that Hg originate from the De-SO2 seawater at outlet of the De-SO2 tower (P2) and in the aeration pools (P3, P4) was 99.6e99.7%, 40% of PHg in seawater at site P1 is from De-SO2 seawater. De-SO2 seawater accounts 33.3e98.4% of the PHg in fresh seawater within 2.0 km from the outlets (S1~S15). The fraction of Hg from De-SO2 seawater declined to approximately 33% within 2.0 km from the outlets, where background contribution increased to 67%. The model output in general well agrees with the spatial pattern of THg in Fig. 2, however, it indicates much larger areas of Hg contamination by the power plant, compared to previous estimation that Hg pollution by De-SO2 seawater were only pronounced within ~300 m from the outlets. Our Hg isotope binary mixing model suggests that 30e50% of PHg within 2.0e3.2 km away from the outlets may be originate from De-SO2 seawater. This highlighted that Hg isotopes may be more reliable to source tracing, especially in areas with low Hg concentrations.
4. Conclusion This study for the first time investigated Hg isotope composition in seawater from and near a SFGD-driven power plant. We found that PHg in De-SO2 seawater have relative higher d202Hg values, whereas mixing with fresh open seawater significantly leads to the decrease of d202Hg in the surrounding seawaters of the power plant. By applying a binary mixing model, we successfully evaluated the fractions of two Hg sources from both De-SO2 seawater and local background. The fractions of De-SO2 Hg in the discharging area progressively decreased from 80% (50 m away, S3) to 40% (2.0 km away, S15) as distance from the outlets increased; De-SO2
Fig. 6. Contribution fraction of PHg in seawater from De-SO2 seawater and background seawater.
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
seawater ultimately impacted an area of 2.0e3.2 km away from the discharging outlets. Compared to previous estimated results based on THg mass balance, our results indicate much larger sea area was impacted by Hg contamination from De-SO2 seawater. The study presented a practical technique for assessing mercury pollution of seawater discharging from the coal-fired power plants with typical SFGD systems, and explored biogeochemical Hg cycling in seawater environments. More attention should be paid to Hg pollution discharging from such power plants. Acknowledgements This research was financed by the Natural Science Foundation of China (21277112). The authors would like to thank Joel D. Blum for n in-house secondary standard. providing the UM-Almade Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.envpol.2016.04.059. References Bergquist, B.A., Blum, J.D., 2009. The odds and evens of mercury isotopes: applications of mass-dependent and mass-independent isotope fractionation. Elements 5, 353e357. Biswas, A., Blum, J.D., Bergquist, B.A., Keeler, G.J., Xie, Z., 2008. Natural mercury isotope variation in coal deposits and organic soils. Environ. Sci. Technol. 42, 8303e8309. Blum, J.D., Bergquist, B.A., 2007. Reporting of variations in the natural isotopic composition of mercury. Anal. Bioanal. Chem. 388, 353e359. Blum, J.D., Sherman, L.S., Johnson, M.W., 2014. Mercury isotopes in earth and environmental sciences. Annu. Rev. Earth Planet. Sci. 42, 249e269. Chen, J., Hintelmann, H., Feng, X., Dimock, B., 2012. Unusual fractionation of both odd and even mercury isotopes in precipitation from Peterborough, ON, Canada. Geochim. Cosmochim. Acta 90, 33e46. Choe, K.Y., Gill, G.A., 2003. Distribution of particulate, colloidal, and dissolved mercury in San Francisco Bay estuary. 2. Monomethyl mercury. Limnol. Oceanogr. 48, 1547e1556. Das, R., Bizimis, M., Wilson, A.M., 2013. Tracing mercury seawater vs. atmospheric inputs in a pristine SE USA salt marsh system: mercury isotope evidence. Chem. Geol. 336, 50e61. Estrade, N., Carignan, J., Donard, O.F., 2010. Isotope tracing of atmospheric mercury sources in an urban area of northeastern France. Environ. Sci. Technol. 44, 6062e6067. Estrade, N., Carignan, J., Sonke, J.E., Donard, O.F., 2009. Mercury isotope fractionation during liquidevapor evaporation experiments. Geochim. Cosmochim. Acta 73, 2693e2711. Feng, C., Gao, X., Tang, Y., Zhang, Y., 2014. Comparative life cycle environmental assessment of flue gas desulphurization technologies in China. J. Clean. Prod. 68, 81e92. Feng, X., Foucher, D., Hintelmann, H., Yan, H., He, T., Qiu, G., 2010. Tracing mercury contamination sources in sediments using mercury isotope compositions. Environ. Sci. Technol. 44, 3363e3368. Foucher, D., Hintelmann, H., 2006. High-precision measurement of mercury isotope ratios in sediments using cold-vapor generation multi-collector inductively coupled plasma mass spectrometry. Anal. Bioanal. Chem. 384, 1470e1478. Foucher, D., Hintelmann, H., 2009. Tracing mercury contamination from the Idrija mining region (Slovenia) to the Gulf of Trieste using Hg isotope ratio measurements. Environ. Sci. Technol. 43, 33e39. Foucher, D., Hintelmann, H., Al, T.A., MacQuarrie, K.T., 2013. Mercury isotope fractionation in waters and sediments of the Murray Brook mine watershed (New Brunswick, Canada): tracing mercury contamination and transformation. Chem. Geol. 336, 87e95. Galbreath, K.C., Zygarlicke, C.J., 2000a. Mercury transformations in coal combustion flue gas. Fuel Process. Technol. 65e66, 289e310. Galbreath, K.C., Zygarlicke, C.J., 2000b. Mercury transformations in coal combustion flue gas. Fuel Process. Technol. 65, 289e310. Ghosh, S., Schauble, E.A., Couloume, G.L., Blum, J.D., Bergquist, B.A., 2013. Estimation of nuclear volume dependent fractionation of mercury isotopes in equilibrium liquidevapor evaporation experiments. Chem. Geol. 336, 5e12. Gratz, L.E., Keeler, G.J., Blum, J.D., Sherman, L.S., 2010. Isotopic composition and fractionation of mercury in Great Lakes precipitation and ambient air. Environ. Sci. Technol. 44, 7764e7770. Hower, J.C., Senior, C.L., Suuberg, E.M., Hurt, R.H., Wilcox, J.L., Olson, E.S., 2010. Mercury capture by native fly ash carbons in coal-fired power plants. Prog. Energy Combust. Sci. 36, 510e529. Jiskra, M., Wiederhold, J.G., Bourdon, B., Kretzschmar, R., 2012. Solution speciation
829
controls mercury isotope fractionation of Hg (II) sorption to goethite. Environ. Sci. Technol. 46, 6654e6662. Kritee, K., Blum, J.D., Reinfelder, J.R., Barkay, T., 2013. Microbial stable isotope fractionation of mercury: a synthesis of present understanding and future directions. Chem. Geol. 336, 13e25. Lefticariu, L., Blum, J.D., Gleason, J.D., 2011. Mercury isotopic evidence for multiple mercury sources in coal from the Illinois Basin. Environ. Sci. Technol. 45, 1724e1729. Liang, Y., Liu, X., Yuan, D., Gong, Z., Zhang, Z., 2010. Mercury species in seawater and sediment of Xiamen western sea area adjacent to a coal-fired power plant. Water Environ. Res. 82, 335e341. Lin, H., Yuan, D., Lu, B., Huang, S., Sun, L., Zhang, F., Gao, Y., 2015. Isotopic composition analysis of dissolved mercury in seawater with purge and trap preconcentration and a modified Hg introduction device for MC-ICP-MS. J. Anal. At. Spectrom. 30, 353e359. Liu, X., Sun, L., Yuan, D., Yin, L., Chen, J., Liu, Y., Liu, C., Liang, Y., Lin, F., 2011. Mercury distribution in seawater discharged from a coal-fired power plant equipped with a seawater flue gas desulfurization system. Environ. Sci. Pollut. Res. 18, 1324e1332. rio, J., Caetano, M., Costa, A.M., Lebreiro, S.M., Mil-Homens, M., Blum, J., Cana Trancoso, M., Richter, T., Stigter, H.D., Johnson, M., 2013. Tracing anthropogenic Hg and Pb input using stable Hg and Pb isotope ratios in sediments of the central Portuguese Margin. Chem. Geol. 336, 62e71. Morel, F.M., Kraepiel, A.M., Amyot, M., 1998. The chemical cycle and bioaccumulation of mercury. Annu. Rev. Ecol. Syst. 543e566. Perrot, V., Epov, V.N., Pastukhov, M.V., Grebenshchikova, V.I., Zouiten, C., Sonke, J.E., Husted, S., Donard, O.F., Amouroux, D., 2010. Tracing sources and bioaccumulation of mercury in fish of Lake Baikal-Angara River using Hg isotopic composition. Environ. Sci. Technol. 44, 8030e8037. Sherman, L.S., Blum, J.D., Johnson, K.P., Keeler, G.J., Barres, J.A., Douglas, T.A., 2010. Mass-independent fractionation of mercury isotopes in Arctic snow driven by sunlight. Nat. Geosci. 3, 173e177. Sherman, L.S., Blum, J.D., Keeler, G.J., Demers, J.D., Dvonch, J.T., 2012. Investigation of local mercury deposition from a coal-fired power plant using mercury isotopes. Environ. Sci. Technol. 46, 382e390. Smith, R.S., Wiederhold, J.G., Kretzschmar, R., 2015. Mercury isotope fractionation during precipitation of metacinnabar (b-HgS) and montroydite (HgO). Environ. Sci. Technol. 49, 4325e4334. Sonke, J.E., 2011. A global model of mass independent mercury stable isotope fractionation. Geochim. Cosmochim. Acta 75, 4577e4590. €fer, J., Chmeleff, J., Audry, S., Blanc, G., Dupre , B., 2010. Sedimentary Sonke, J.E., Scha mercury stable isotope records of atmospheric and riverine pollution from two major European heavy metal refineries. Chem. Geol. 279, 90e100. Stefano, C., Raffaella, P., Joze, K., Milena, H., Jadran, F., Antonio, B., 2006. Behaviour of Hg species in a microtidal deltaic system: the Isonzo River mouth (northern Adriatic Sea). Sci. Total Environ. 368, 210e223. Strok, M., Baya, P.A., Hintelmann, H., 2015. The mercury isotope composition of Arctic coastal seawater. Comptes Rendus Geosci. 347, 368e376. Strok, M., Hintelmann, H., Dimock, B., 2014. Development of pre-concentration procedure for the determination of Hg isotope ratios in seawater samples. Anal. Chim. acta 851, 57e63. Sun, L., Feng, L., Yuan, D., Lin, S., Huang, S., Gao, L., Zhu, Y., 2013a. The extent of the influence and flux estimation of volatile mercury from the aeration pool in a typical coal-fired power plant equipped with a seawater flue gas desulfurization system. Sci. Total Environ. 444, 559e564. Sun, L., Lin, S., Feng, L., Huang, S., Yuan, D., 2013b. The distribution and seaeair transfer of volatile mercury in waste post-desulfurization seawater discharged from a coal-fired power plant. Environ. Sci. Pollut. Res. 20, 6191e6200. Sun, R., Heimbürger, L.-E., Sonke, J.E., Liu, G., Amouroux, D., Berail, S., 2013c. Mercury stable isotope fractionation in six utility boilers of two large coal-fired power plants. Chem. Geol. 336, 103e111. Sun, R., Sonke, J.E., Heimbürger, L.E., Belkin, H.E., Liu, G., Shome, D., Cukrowska, E., Liousse, C., Pokrovsky, O.S., Streets, D.G., 2014. Mercury stable isotope signatures of world coal deposits and historical coal combustion emissions. Environ. Sci. Technol. 48, 7660e7668. Sun, R., Sonke, J.E., Liu, G., 2016. Biogeochemical controls on mercury stable isotope compositions of world coal deposits: a review. Earth Sci. Rev. 152, 1e13. U.S. Environmental Protection Agency, 1996. Method 1669: Sampling Ambient Water for Trace Metals at EPA Water Quality Criteria Levels. Office of Water Washington, DC. Wiederhold, J.G., 2015. Metal stable isotope signatures as tracers in environmental geochemistry. Environ. Sci. Technol. 49, 2606e2624. Wiederhold, J.G., Cramer, C.J., Daniel, K., Infante, I., Bourdon, B., Kretzschmar, R., 2010. Equilibrium mercury isotope fractionation between dissolved Hg (II) species and thiol-bound Hg. Environ. Sci. Technol. 44, 4191e4197. € rn, E., Bourdon, B., Wiederhold, J.G., Skyllberg, U., Drott, A., Jiskra, M., Jonsson, S., Bjo Kretzschmar, R., 2015. Mercury isotope signatures in contaminated sediments as tracer for local industrial pollution sources. Environ. Sci. Technol. 49, 177e185. Yin, R., Feng, X., Chen, B., Zhang, J., Wang, W.-X., Li, X., 2015. Identifying the sources and processes of mercury in subtropical estuarine and ocean sediments using Hg isotopic composition. Environ. Sci. Technol. 49, 1347e1355. Yin, R., Feng, X., Chen, J., 2014. Mercury stable isotopic compositions in coals from major coal producing fields in China and their geochemical and environmental implications. Environ. Sci. Technol. 48, 5565e5574.
830
H. Lin et al. / Environmental Pollution 214 (2016) 822e830
Yin, R., Feng, X., Wang, J., Bao, Z., Yu, B., Chen, J., 2013. Mercury isotope variations between bioavailable mercury fractions and total mercury in mercury contaminated soil in Wanshan Mercury Mine, SW China. Chem. Geol. 336, 80e86. Yu, F., Chen, J., Sun, F., Zeng, S., Wang, C., 2011. Trend of technology innovation in China's coal-fired electricity industry under resource and environmental constraints. Energy Policy 39, 1586e1599.
Zheng, W., Foucher, D., Hintelmann, H., 2007. Mercury isotope fractionation during volatilization of Hg(0) from solution into the gas phase. J. Anal. At. Spectrom. 22, 1097e1104. Zheng, W., Hintelmann, H., 2010. Nuclear field shift effect in isotope fractionation of mercury during abiotic reduction in the absence of light. J. Phys. Chem. A 114, 4238e4245.