Control strategies of atmospheric mercury emissions from coal-fired power plants in China

Abstract

Atmospheric mercury (Hg) emission from coal is one of the primary sources of anthropogenic discharge and pollution. China is one of the few countries in the world whose coal consumption constitutes about 70% of total primary energy, and over half of coals are burned directly for electricity generation. Atmospheric emissions of Hg and its speciation from coal-fired power plants are of great concern owing to their negative impacts on regional human health and ecosystem risks, as well as long-distance transport. In this paper, recent trends of atmospheric Hg emissions and its species split from coal-fired power plants in China during the period of 2000–2007 are evaluated, by integrating each plant's coal consumption and emission factors, which are classified by different subcategories of boilers, particulate matter (PM) and sulfur dioxide (SO₂) control devices. Our results show that the total Hg emissions from coal-fired power plants have begun to decrease from the peak value of 139.19 t in 2005 to 134.55 t in 2007, though coal consumption growing steadily from 1213.8 to 1532.4 Mt, which can be mainly attributed to the co-benefit Hg reduction by electrostatic precipitators/fabric filters (ESPs/FFs) and wet flue gas desulfurization (WFGD), especially the sharp growth in installation of WFGD both in the new and existing power plants since 2005. In the coming 12th five-year-plan, more and more plants will be mandated to install De-NO_x (nitrogen oxides) systems (mainly selective catalytic reduction [SCR] and selective noncatalytic reduction [SNCR]) for minimizing NO_x emission, thus the specific Hg emission rate per ton of coal will decline further owing to the much higher co-benefit removal efficiency by the combination of SCR + ESPs/FFs + WFGD systems. Consequently, SCR + ESPs/FFs + WFGD configuration will be the main path to abate Hg discharge from coal-fired power plants in China in the near future. However, advanced specific Hg removal technologies are necessary for further reduction of elemental Hg discharge in the long term.

Implications

Controlling of atmospheric Hg discharge from coal-fired power plants have aroused great concerns for its adverse impacts on regional environment and human health risks, as well as long-distance transportation. It is of great significance for Chinese decision makers to be aware of the current status of Hg emissions from coal-fired power plants, so that the regulations and policies regarding Hg abatement can be made that are cost-effective and feasible implementation. This study provides the recent trend of atmospheric Hg emissions from coal-fired power plants, and accordingly proposes the preliminary comprehensive Hg control strategies suggestion in the future, which will be helpful for relevant policy making to minimize the harmful risks on environment and human health in China.

Introduction

One of the major utilization of coal is to generate power through combustion (Pavageau et al., 2002). During the coal combustion process, significant amounts of volatile toxic trace elements are released and distributed among flue gas and fly ash, such as mercury (Hg), arsenic (As), and selenium (Se), etc. (Al-Abed et al., 2008; Otero-Rey et al., 2003; Pavageau et al., 2002). The dangerous effects and risks of associated with exposure of Hg on the environment and public health have gained growing attention throughout the world (Nelson, 2007; Wiedinmyer and Friedli, 2007; Zhang and Wong, 2007). In the United States, the Environmental Protection Agency (EPA) has begun to control toxic emissions of 11 metals from combustors and incinerators, under the Title III of the 1990 Clean Air Act Amendments, including Hg (Biswas and Wu, 1998). Also, the United Nations Environment Programme (UNEP) Mercury Programme, a process to assess to what extent contamination by mercury released from anthropogenic and natural sources may affect human health and ecosystems, has started since 2002, in view of the significance of environmental issues related to mercury released to the atmosphere by major anthropogenic sources, which certainly include, but are not limited to, power plants for energy production and a variety of industrial plants.

China is one of the few countries in the world whose energy consumption mix is dominated by the highly polluted fossil fuel coals. Coal constitutes about 70% of the total primary energy production and consumption, and over half of coals are burned for electricity generation by coal-fired power plants. By the year of 2007, the total coal consumption has increased to 2727.46 million tons (Mt), and about 1532.4 Mt were burned directly by coal-fired power plants scattering throughout the country, accounting for over 50% of the totals (The National Bureau of Statistics of China [NBS] and the National Development and Reform Commission of China [NDRC], 2010). Thus, coal-fired power plants have become a main source of anthropogenic atmospheric Hg emissions and exposure in China (Jiang et al., 2005; Streets et al., 2005; Wang et al., 2000), and have aroused great concern over the control and reduction of atmospheric Hg emissions for their adverse impacts on regional environment and human health risks, as well as the long-distance transport (Jiang et al., 2005; Streets et al., 2005; Tian et al., 2010; Wang et al., 2000; Wu et al., 2006). For example, Wang et al. (2000) estimated the total mercury emitted into the atmosphere from coal combustion as 213.8 t in 1995, among which 72.86 t were emitted from coal-fired power plants (Wang et al., 2000). Wu et al. (2006) estimated that mercury emission from coal-fired power plants in China increased from 63.4 to 100.1 t, with an annual growth rate of 5.9% during 1995∼2003 (Wu et al., 2006). Tian et al. (2010) presented a comprehensive emission inventories of Hg, As, and Se from coal in China, they estimated that the national total atmospheric emissions of Hg, As, and Se from coal had rapidly increased from 73.59, 635.57, and 639.69 t in 1980 to 305.95, 2205.50, and 2352.97 t, respectively, in 2007 (Tian et al., 2010). With the growing stress on environment and human health effects of Hg reduction both domestically and abroad, it is very necessary to assess the current status of Hg emissions from coal-fired power plants and promulgate integrated Hg control strategies in China. However, there are few reports on the features of Hg emission rate change and comprehensive control strategies till now.

In this paper, by using the best available coal consumption data for each unit and the newly updated Hg emission factors since our previous publication (Tian et al., 2010), which are specified by different patterns of boilers and the particulate matter/sulfur dioxide (PM/SO₂) control devices downstream, recent temporal trends of atmospheric Hg emissions from coal-fired power plants in China for the period of 2000–2007 are evaluated, and the temporal and spatial variations of net combined Hg emission rate from coal-fired power plants are discussed in detail. Further, preliminary control strategies for abating Hg discharge are proposed according to the comprehensive assessment of Hg co-benefit reduction effects by the existing conventional air pollution control device (De-duct, De-SO₂, De-NO_x) configuration installed in coal-fired power plants.

Methodologies and Data Sources

Methodology of Hg emission estimation

A bottom-up emission factor method is adopted in this study to estimate atmospheric Hg emissions from coal-fired boilers of power plants in China. Anthropogenic atmospheric Hg emissions from coal-fired power plants are calculated by combining the detailed coal consumption data, averaged Hg content in coals by province, and the specific Hg emission factors that are classified by different boiler patterns, and PM and SO₂ control device configuration. The basic formulas can be expressed as follows:

(1)

where E is the atmospheric emission of Hg; F is the amount of coal burned by power plant unit; C is the averaged content of Hg in feed coal by province; R is the fraction of Hg released from coal-fired utility boilers; η_PM and η_FGD are the fractions of Hg removed by PM and SO₂ emission control devices, respectively; i is the province; and j is the subcategories for coal-fired power plants that are specified by different patterns of boiler furnaces and PM/SO₂ control device configuration; EF is called the combined Hg emission factor, i.e., the final discharge rate of Hg into the atmosphere by burning per tones of coal in power plants.

Average content of Hg in raw coals as produced by provinces

The content of Hg in coals can provide useful information to the pollution control during coal combustion and utilization from an environmental point of view. Even when being present in only parts per million levels in coal, combustion of coal can result in tons of hazardous air pollutants being discharged into the ambient environment. China is a huge country with 34 provinces and special administrative districts, and the content of Hg in coals mined from different places varies substantially due to the coal-forming plants and the coal-forming geological environments. Previous studies have demonstrated that the content and distribution of Hg differ from provinces, sources, and even the coals of the same seam (Tian et al., 2010). Besides the modes of occurrence, mineral content, and the distribution of Hg in coals, there are also some other sources causing variation of hazardous trace elements contents, such as random sampling errors, measurement errors, or samples are not representative. In this study, we have compiled and summarized the tested content of Hg in Chinese raw coals that are reported in the available published literature.

Figure 1 shows out the average content of Hg in raw coals as produced (Hg_(P)) from 30 provinces (autonomous regions and municipalities) on the Chinese mainland; Taiwan province and Hong Kong and Macau Special Administrative Districts are not included tentatively. Xizang is also not considered because there are no large-scale coal-fired power plants. The values of average Hg content used in our study as shown in Figure 1 are compiled and summarized from both U.S. Geological Survey (USGS) data (Streets et al., 2005; USGS, 2004; Wu et al., 2006) and domestic literature data (Bai, 2003; Chen et al., 2006; Feng et al., 2002; Huang and Yang, 2002; Ren et al., 2006; Tian et al., 2010; Zheng et al., 2007; Zhuang et al., 1999). We calculated the arithmetic mean values of Hg content in coals for each province reported in the references above on accounting of the representativeness of data samples, and applied the mean values to estimate the Hg emissions from coal-fired power plants. Among these 30 provinces, the lowest mean concentration of Hg is 0.025 mg/kg in Xinjiang, whereas the highest concentration is 0.368 mg/kg in Guizhou province. The remarkable variation among provinces is mainly due to the differences of coal-forming plants and coal-forming geological environments in different areas (Bai, 2003; Feng et al., 2002; Ren et al., 2006). As a result, the national average Hg content of raw coals as produced is regarded as 0.180 mg/kg.

Figure 1. Average content of Hg in raw coals as produced and consumed in China (P denotes coal as production; C denotes coal as consumption).

Average content of Hg in raw coals and coal products as consumed

The geographical distribution of coal resources in China is extremely unbalanced, rare in the southern and eastern areas while rich in the northern and western areas in general. As a result, large volumes of coal mined have to be transported long-distance from coal-base areas (such as Shanxi, Shaanxi, Inner Mongolia, Guizhou, etc.) to consumption areas, mainly the eastern and southern provinces where energy-intensive manufacturing industries are concentrated and where it is densely populated, such as Jiangsu, Zhejiang, Guangdong, etc. It implies that there will be significant variations between the average content of toxic elements in coals as produced and consumed within one province. Therefore, it is very essential to know the Hg content of Chinese coals as consumed, not just as produced, to obtain reliable Hg emission estimates from coal-fired power plants in China. According to the statistical data retrieved from China Energy Statistical Yearbooks (1997–2008) (NBS and NDRC, 2010) and China Coal Industry Yearbooks (2000–2008) (The State Administration of Coal Mine Safety [SACMS], 2009), annual coal flow matrixes among 30 provinces are established to quantify in-province coal use and inter-province coal flows as described in the previous studies (Streets et al., 2005; Tian et al., 2010; Wu et al., 2006), then we have calculated out the weighted average content of Hg in raw coals as consumed (Hg_(C)) by provinces and districts, as shown in Figure 1. Notably, though the inter-province coal supply patterns were relatively steady for the majority of provinces during the period of 2000–2007, the calculated weighted average content of Hg in coal as consumed for the studying period are variant somewhat owing to the difference of both the magnitude and proportion of coal import/export among 30 provinces, as can be seen for details in the China Coal Industry Yearbooks (SACMS, 2009). In addition, there is no raw coal production in Hainan, Shanghai, and Tianjin, thus the values of Hg content in raw coals as produced are assumed to be zero.

As shown in Figure 1, China is normally divided into six large regions, namely the Northern Region (N), the Northeastern Region (NE), the Eastern Region (E), the Central and Southern Region (CS), the Southwestern Region (SW), and the Northwestern Region (NW). In the Southwestern Region, there is little change in Hg content between raw coals as produced and consumed. The main reason is that the coal consumed in this district is mainly obtained from within-province supply, such as Chongqing, Sichuan, Gansu, and Guizhou, whereas in other regions, especially the Eastern Region and the Central and Southern Region, a significant difference in Hg content between raw coals as produced and consumed can be seen, because part or even the majority of coals consumed are imported from other coal-export provinces (such as Shanxi, Inner Mongolia, Shaanxi, etc.) by long-distance transport. Interestingly, the average content of Hg in raw coals as consumed for each large region in China is far lower than that as produced (see Figure 1). The main reason is that the average contents of Hg in raw coal as produced in Shanxi, Inner Mongolia, Shaanxi provinces, etc., the major exporters of coal in China, are relatively lower. Thus, trans-province coal transport not only has satisfied the energy demand for the eastern developed provinces, it also adjusted and lowered the average contents of hazardous trace elements in feed coals, including Hg.

Installed capacity and coal consumption of coal-fired power plants

With the rapid growth of the economy and urbanization in China during the past 3 decades, the demand for electricity has increased continuously and the combustion of coal for electricity generation has grown steadily and quickly, especially from the beginning of 21st century. By the end of 2007, the total installed power capacity in China had reached 718.22 GW, an increase of 15.2% over the previous year and 77.4% of which were coal-fired power plants, amounting to 556.07 GW, up by 14.9% over 2006 (Edition Commission of Chinese Power Statistical Yearbook [ECCPSY], 2009). From 2000 to 2007, the total electricity generation by coal-fired power plants in China increased from 1107.9 to 2720.9 TWhr, meanwhile, the volume of coals as consumed increased from 619.49 to about 1532.4 Mt, more than doubled during the past 7 years (ECCPSY, 2009). Figure 2 illustrates provincial coal use and thermal electricity generation by coal-fired power plants in 2007.

Figure 2. Provincial coal use and electricity generation by coal-fired power plants, 2007.

Emission factors of Hg from coal-fired power plants

Hg and its species released from coal burning

Mercury is commonly found in many rocks including coal. When coal is burned, not only carbon dioxide (CO₂), sulfur dioxide (SO₂), nitrogen oxides (NO_x), but also some hazardous trace metals occurred in coals such as mercury compounds are released. Normally, Hg is released from coal-fired furnaces as a vapor in a mixture of three chemical states or species: in the form of elemental mercury (Hg⁰), as gaseous oxidized mercury (Hg²⁺), and some adsorbed to fly ash particles (Hg^P). Coal-fired power plants are the largest human-caused source of mercury emissions to the air in the United States, accounting for over 50% of all domestic human-caused mercury emissions (EPA, 2005).

The element content of coals is a key factor influencing Hg release into the atmosphere. Besides, patterns of combustion facilities and the equipped conventional PM/SO₂/NO_x control devices downstream that are utilized on coal-fired utility boilers for reducing PM, SO₂ and NO_x, will affect the speciation of Hg and its compounds and are effective in reducing the final discharge into the atmosphere. The degree of this co-benefit removal effects can vary substantially depending upon the specific air pollution control device (APCD) configuration.

In recent years, the characteristics of trace elements emitted from coal-fired power plants in China have been investigated by some researchers and institutions, especially for Hg (Wang et al., 2010; Yi et al., 2008; Zhang et al., 2008). For example, Wang et al. (2010) investigated the Hg emission from six coal-fired power plants in China, and they found the release rate and speciation of Hg varied substantially for different boilers and the equipped PM and flue gas desulfurization (FGD) system. In our study, coal-fired power plants in China are classified into different subgroups based on boiler patterns, particulate control devices, as well as FGD systems. By the end of 2007, the number and capacity of coal-fired power plants that had installed selective catalytic reduction (SCR) to reduce further NO_x emission was relatively small and some were believed not in full operation in most times. Thus, the influence of SCR devices on Hg emissions is not considered tentatively, though the SCR catalyst can alter the speciation of Hg and promote some of elemental Hg to oxidize into Hg²⁺, which is much easier to be absorbed by the downstream wet flue gas desulfurization (WFGD) scrubber.

Release rate of Hg for combustion devices

Combustion devices used in coal-fired power plants are divided into three types: pulverized-coal (PC) boilers, fluidized-bed combustion (FBC) furnaces, and stoker-fired boilers. Presently, pulverized-coal boiler represents a large proportion of coal-fired power plants in most of the provinces in China. The remaining share is represented by the other two patterns, which are commonly adopted in small plants. The national average proportion of installed capacity for pulverized-coal boilers, fluidized-bed combustion furnaces, and stoker-fired boilers are approximately 83.3%, 12.9%, and 3.8% by 2007, respectively (ECCPSY, 2009). Some researchers, abroad and domestic (Demir et al., 2001; Lee et al., 2006; Meij et al., 2002; Wang et al., 1996, 2010; Zhang et al., 2008; Zhou et al., 2008; Zhu et al., 2002), have investigated the Hg release rate for different types of boilers. Zhang et al. (2008) and Wang et al. (2010) conducted field tests of Hg emission at 12 coal-fired power plants in China and they found the over 99% of Hg in coals will evaporated into flue gas. Here, we adopt the arithmetic mean value of the Hg release ratio for different types of utility boilers reported in available literature (Demir et al., 2001; Lee et al., 2006; Meij et al., 2002; Wang et al., 1996, 2010; Zhang et al., 2008; Zhou et al., 2008; Zhu et al., 2002), and we assume that the averaged release ratio of Hg is about 99.4% for pulverized-coal boilers, about 83.2% for stoker-fired boilers, and approximately 98.9% for fluidized-bed combustion boilers, respectively.

Removal efficiency of Hg through particulate matter (PM) control devices

Particulate matter control devices are installed to capture fly ash in flue gas from coal-fired utility boilers. When the flue gas temperature drops, part of Hg in the gas phase condenses on or is captured by fly ash, and subsequently can be removed from flue gas by downstream PM control devices. At present, all of coal-fired power plants in China have installed particulate control devices mandated by the central governmental emission regulations and standards, such as cyclones, wet scrubbers, electrostatic precipitators (ESPs), as well as fabric filters (FFs). Therein, ESPs are now the most widely used PM abatement devices in coal-fired power plants and central heating plants in China. By the end of 2001, about 87% of the total installed capacity heat units were equipped with ESPs and the proportion had grown up to about 95% by 2007, the average PM removal efficiencies of ESPs are reported at about 97.5%, and they can normally capture about 28∼50% of the total Hg released in flue gas (Wang et al., 2010; Wang and Zhang, 2009; Zhang et al., 2008). FFs are regarded as more effective for capturing fine particles and in reducing gaseous Hg emissions. In order to comply with the more strict PM emission standards (GB13223-2003) (The State Environmental Protection Administration of China [SPEA], 2003), some power plants (especially those plants burning lignite with high volatiles, leading to high fly ash resistivity, which lowers ESP capture efficiency) have adopted or retrofitted with bag-house filters instead of ESPs, but still not widespread. Bag-house filters were only reported in 9187 MW of the installed power capacity by the end of 2007, representing less than 1.7% of the total thermal power capacity (ECCPSY, 2009). Several studies indicate that Hg removal efficiency of FFs varies from 58% to 90%, whereas that of wet scrubbers and cyclones is very low (Helble, 2000; Meij and te Henk, 2007; Pavlish et al., 2003; Wang and Zhang, 2009). In this study, we adopt the arithmetic mean value of the field test Hg removal efficiency reported in available literature (Afonso and Senior, 2001; Brekke et al., 1995; Chu and Porcella, 1995; Demir et al., 2001; Helble, 2000; Lee et al., 2006; Meij and te Henk, 2007; Meij et al., 2002; Pavlish et al., 2003; Srivastava et al., 2006; Tian et al., 2010; Wang et al., 1996, 2010; Wang and Zhang, 2009; Yi et al., 2008; Zhang et al., 2008; Zhou et al., 2008; Zhu et al., 2002), and the average co-benefit removal efficiency of Hg by ESPs, FFs, wet scrubbers, and cyclones are assumed to be about 32.3%, 67.9%, 15.2%, and 6.0%, respectively.

Removal efficiency of Hg through flue gas desulfurization systems

Flue gas desulfurization (FGD) systems are commonly used for reducing SO₂ emissions in coal-fired power plants throughout the world. The application of FGD has significant co-benefit impacts on removal of hazardous trace elements, especially the high-volatility Hg (Wang et al., 2010; Zhang et al., 2008). Since 2000, the installed capacities of coal-fired power plants with FGD have increased rapidly, as can be seen in Figure 3. According to the statistical data provided by the Ministry of Environmental Protection of the People's Republic of China (MEP), 712 power plants have adopted FGD with an installed capacity of 265.6 GW, representing 47.8% of the total coal-fired power capacity by the end of 2007. Therein, the proportion of units installed with FGD is higher in Beijing, Qinghai, Hunan, Chongqing, and Guangdong provinces across the country. There were no coal-fired power plants installed with FGD in Xinjiang till the end of 2007. Of all the coal-fired units equipped with FGD system, more than 92.3% adopted wet scrubbers, and about 90% of wet FGD were the wet limestone/lime absorption FGD process. The scrubbing process in WFGD system is capable to transfer Hg in flue gas into the slurry, especially for the oxidized Hg and fine particle Hg. In recent years, there have been some investigations on the removal effects of trace elements by different types of SO₂ control device configuration, and the tested removal efficiency varied substantially from 15% to 80% (Afonso and Senior, 2001; Chu and Porcella, 1995; Demir et al., 2001; Díaz-Somoano et al., 2007; Fahlike and Bursik, 1995; Helble, 2000; Lee et al., 2006; Meij and te Henk, 2007; Meij et al., 2002; Nelson et al., 2010; Park et al., 2008; Pavlish et al., 2003; Renninger et al., 2004; Senior et al., 2000; Srivastava et al., 2006; Wang et al., 1996, 2010; Wang and Zhang, 2009; Yi et al., 2008; Zhang et al., 2008; Zhou et al., 2008; Zhu et al., 2002). For example, Wang et al. (2010) report that the Hg removal efficiency of WFGD of six coal-fired power plants in China is in the range of 15.3–79.2%, with an average of about 51.8%. In this study, on accounting the most recent test results form Wang et al. (2010), we have update our former database (Tian et al., 2010) and adopted a new averaged efficiencies as about 56.3% (Afonso and Senior, 2001; Chu and Porcella, 1995; Demir et al., 2001; Fahlike and Bursik, 1995; Helble, 2000; Lee et al., 2006; Meij and te Henk, 2007; Meij et al., 2002; Park et al., 2008; Pavlish et al., 2003; Srivastava et al., 2006;, Tian et al., 2010; Wang and Zhang, 2009; Wang et al., 1996, 2010; Yi et al., 2008; Zhang et al., 2008; Zhou et al., 2008; Zhu et al., 2002). Consequently, the overall removal efficiency of Hg by the combination of ESPs and WFGD system is assumed at about 70.4%, a little lower than the average value (73%) indicated by Wang et al. (2010).

Figure 3. Trend of installation capacity of coal-fired units installed with FGD.

Combined emission factors of Hg

By combining with the averaged content of Hg in coals as consumed, release rates of different combustion facilities, as well as the averaged removal efficiencies of PM and SO₂ control systems, the final combined emission factors of Hg are calculated to reflect the final rate of Hg emissions through burning per Mt coal in power plants in China (see Figure 4). It is obvious that the final combined emission factors are directly proportional to the content of Hg in coal and the release rates while inversely proportional to the co-benefit removal efficiencies by PM and SO₂ control systems. Also, it implies that remarkable differences exist not only in the emission factors of Hg among different provinces because of the large variation of Hg content in coals, but also in the emission factors of Hg among different coal-fired power plants within one province, owing to the different scale of PM and FGD device application. Therein, the higher values of Hg combined emission factor happened in Guizhou, Guangxi, and Anhui provinces are primarily determined by the high Hg content in coals as output and consumption, whereas the lower values occurred in northwestern provinces can be explained by the low Hg content in coals. However, the relatively lower values for some provinces, such as Beijing, Jiangsu, and Hunan provinces, are primarily ascribed to the high proportion of ESP and FGD application.

Figure 4. Combined emission factors of Hg from coal-fired power plants.

Results and Discussion

Trend of Hg emissions from all coal-fired power plants

Based on the detailed coal consumption by each power plant and the specific emission factors assumed above, the atmospheric emissions of Hg from coal-fired power plants in China have been estimated by using eq 1. As can be seen in Figure 5, the final national gross emissions of Hg from coal-fired power plants have experienced a rapid growth until the peak value of 139.19 t in 2005, and then declined gradually to about 134.55 t by 2007, which mainly thanks to the co-benefit reduction effects by the existing and newly installed air pollution control devices (APCDs) in coal-fired power plants since 2005, such as ESPs, FFs, and WFGD. Especially, the widely installation and operation of WFGD system, for complying with the strict SO₂ control policies during the 11th five-year-plan issued by the central government, have contributed much co-benefit Hg reduction. Consequently, the total coal consumed in power plants by the end of 2007 had increased to as high as 1532.4 Mt, but the final Hg emission rate by burning every Mt of coal dropped from 0.113 t in 2003 (Wu et al., 2006) to about 0.086 t in 2007 in our study, leading to the gradually lowering tendency of Hg emissions.

Figure 5. Trend of total Hg emissions from coal-fired power plants.

Several factors influence the estimation of atmospheric Hg emissions, including activity levels and emission factors. Because the data of annual coal use for power plants are compiled from official statistics, the main uncertainty in our estimation will mainly come from the calculated average Hg content in coal, the assumed release rate, as well as the removal efficiency of PM and SO₂ control device configuration, which is similar to the uncertainty analysis conducted by Wu et al. (2010). In this study, we estimated that the total Hg emissions from coal-fired power plants for the year of 2003 is about 109.91t, a little higher than Wu et al.’s estimation (100.1 t) for the same year (Wu et al., 2006), which can be mainly explained by the little higher average Hg content in coal that we compiled with more new available domestic literature. Also, our estimation for the year of 2003 is about 21.4% higher than the best estimate (90.5 t) for total Hg emissions from coal-fired power plants in China in 2003, but it is still within the uncertainty range from 57.1 to 154.6 t indicated by Wu et al. (2010). This implies that our estimation of Hg emissions from coal-fired power plants in China in this study is reasonable and comparable (Streets et al., 2005; Wu et al., 2006).

Provincial emission inventories of Hg from coal-fired power plants

The provincial emission inventories of Hg from coal-fired power plants in 2007 are illustrated in Figure 6. Remarkable unevenness can be observed among provincial-level inventories. Generally speaking, Hg emissions in the eastern and central provinces of China are much higher than those in the west, except for provinces involved in the program of electricity transmission from west to east, such as Sichuan, Guizhou, Yunnan, Shaanxi, etc. Provinces with large installed coal-fired capacities and electricity generation, for example, Shandong, Henan, Jiangsu, and Inner Mongolia, emit much more Hg into the atmosphere. In comparison, the relatively higher Hg emissions in Gizhou, Henan, and Anhui are primarily due to the higher combined Hg emission factor mentioned above, which resulted from the higher Hg content in coal as output and consumption, as well as the lower penetration of emission control devices, as shown in Figures 2, 4, and 6.

Figure 6. Provincial Hg emissions and its speciation from coal-fired power plants in China, 2007.

Speciation of total Hg emissions from coal-fired power plants in China

Normally, mercury (Hg) emitted from coal-fired power plants exists in three primary forms, which are, namely, elemental mercury (Hg⁰), gaseous oxidized mercury (Hg²⁺) and particle-bound mercury (Hg^P). In the combustion zone, most of the Hg in the fuel coals is evaporated and exists as Hg⁰, which is cooled by passing through the downstream of the flue gas and can be oxidized by flue gas components such as HCl, SO₂, H₂O, and fly ash to form Hg²⁺ (Wang et al., 2010; Yi et al., 2008; Zhang et al., 2008; Zhu et al., 2002). Some of the Hg²⁺ can be attached physically or reacts with the particulate in the flue gas and then forming particle-bound Hg^P. The majority of Hg^P can be captured and removed by ESPs or FFs and thus low proportions of Hg^P can be monitored at the outlet of PM control devices (Lee et al., 2006; Wang et al., 2010; Zhang et al., 2008). Hg²⁺ is water-soluble and therefore amenable to be retained in WFGD or wet scrubbers. However, Hg⁰ is insoluble in water and cannot be retained effectively in wet scrubbers, so that the final Hg stack discharge may be mostly Hg⁰ vapor at the presence of ESPs/FFs + FGD systems (Díaz-Somoano et al., 2007; Nelson et al., 2010; Park et al., 2008; Senior et al., 2000; Wang et al., 2010; Zhang et al., 2008). For example, Wang et al. (2010) have conducted comprehensive field measurements in six coal-fired power plants in China, and they indicated that the average share of Hg⁰ to total Hg is about 46% for the boilers installed with ESPs, and the share of Hg⁰ to total Hg in stack gas to the atmosphere is 75–94% for the boilers with ESPs + WFGD system.

To identify the current status of Hg emission and to estimate the fate and behavior of Hg from coal-fired power plants, an estimation of Hg speciation is also presented. As demonstrated by field tests results, Hg emissions from large power plants can be effectively captured and reduced by the installed ESPs/FFs and FGD systems. Here, we adopted the arithmetic mean value of Hg speciation reported in available references, and assume that the average proportion of Hg⁰, Hg²⁺, and Hg^P in total Hg emissions for the boilers installed with ESPs is about 52.4%, 45.8%, and 1.8%; and the average proportion for the boilers installed with FFs is about 27.4%, 61.3%, and 11.3%, whereas the average proportion for the boilers with ESPs and WFGD is about 81.0%, 18.6% and 0.4%, respectively (Brekke et al., 1995; Díaz-Somoano et al., 2007; Nelson et al., 2010; Park et al., 2008; Renninger et al., 2004; Senior et al., 2000; Srivastava et al., 2006; U.S. EPA, 2005; Wang et al., 2010; Yi et al., 2008; Zhang et al., 2008). Thus, based on the percentage of coal consumed in units installed with different patterns of PM control devices and WFGD systems in each province, we estimate a speciation split of total Hg emissions in 2007 are as follows: about 81.94 t of Hg⁰, about 50.70 t of Hg²⁺, and about 1.91 t of Hg^P, representing about 60.90%, about 37.68%, and about 1.42% of the totals, respectively.

Normally, Hg²⁺ and Hg^P exist in the atmosphere for only a few days, whereas Hg⁰ can stay in the atmosphere for more than 1 year. In recent years, with the widely application of ESPs/FFs and wet FGD in coal-fired power plants all over the country, the majority of Hg species emitted from coal-fired power plants are being changed from mainly Hg⁰ + Hg²⁺ to Hg⁰. Hg⁰ stays in the atmosphere longer than Hg²⁺ and Hg^P, and is much easier for long-distance transport and even causing transboundary arguments. Thus, the potential environmental problems due to the changes of speciation of Hg emissions should be highlighted when the rapid growth of total Hg emissions have been restrained effectively in China.

Proposals on Hg emission control strategies

By applying the detailed coal consumption of power plant units, averaged Hg content in feed coals, and the specific emission factors for different boiler types and APCD configuration, we have estimated the atmospheric emissions of Hg from coal-fired power plants for the period of 2000–2007. Our results demonstrate that the overall Hg emissions from coal-fired power plants have begun to decline since 2005, which is favorable for Hg abatement in China. According to Wu et al. (2006), the total coal consumed in power plants in China increased from 505.6 to 886.3 Mt from 1995 to 2003, meanwhile, mercury emission increased from 63.4 t to 100.1 t with an annual growth rate of 5.9%. By the end of 2007, the total coal consumed in power plants had increased to as high as 1532.4 Mt, whereas the final Hg emission rate by burning every Mt of coal had dropped from 0.113 t in 2003 (Wu et al., 2006) to about 0.086 t in 2007 in this study. This decline trend mainly thanks to the more and more installation and operation of WFGD in coal-fired power plants since 2005, for complying with the SO₂ control policies issued during the 11^th five-year-plan issued by the central government. It indicates that the combination of conventional PM and SO₂ control systems can attain well co-benefit Hg reduction effects, which is a cost-effective choice for Hg removal in coal-fired power plants in China.

After passing through ESPs and WFGD, much of the particle-bound and reactive gaseous mercury will be removed from flue gas, and thus the total Hg emission rate decrease substantially. However, Hg⁰ is much more difficult to be removed by ESPs and WFGD, and the main Hg speciation emitted will shift from Hg²⁺ and Hg^P to Hg⁰, which has much longer lifetime in the atmosphere before it settle down through dry and wet deposition. With the growing coal burned by more and more coal-fired power plants, unless other technologies such as activated carbon injection are employed, the absolute amount of Hg⁰ is likely to increase under the current conventional air pollution control device configuration. Thus, some field pilot-experiments and demonstration projects on advanced technologies such as activated carbon injection should be initiated and speeded up in order for further abating Hg emissions in the future.

One of the effective methods for Hg⁰ removal is to convert Hg⁰ to its oxidized form Hg²⁺, which can be readily captured by a WFGD system downstream. Catalytic oxidation of Hg⁰ in the presence of gaseous HCl has been given much attention because HCl is usually present in coal-fired flue gases (Chen et al., 2007; Hu et al., 2009). SCR catalysts, which is equipped with boilers to transform NO_x in the flue gas into inertial N₂, have been proved as an potential elemental mercury oxidation catalyst for coal-fired plants, especially for those firing coals with high Cl content (Lee et al., 2006; Uddin et al., 2009; Wang et al., 2010; Zhou et al., 2008; Zhu et al., 2002).

In order to combat the worsening urban air quality and regional ground ozone and haze pollution, which are closely related with NO_x emission from stationary and mobile sources, NO_x will be added to the list of two restricting indexes for atmospheric environment improvement by the Chinese central government during the coming 12th five-year-plan. New and some existing coal-fired power plants will be requested to install De-NO_x systems to reduce NO_x emissions into the atmosphere, mainly using SCR because of high efficiency and widespread application experiences in developed countries, such as Japan, Germany, and the United States. Just as mentioned formerly, the SCR catalyst can promote the transformation of Hg⁰ to Hg²⁺, which is much easier to be captured and removed by the ESPs/FFs and wet FGD scrubber downstream (Díaz-Somoano et al., 2007; Wang et al., 2010). By now, limited field tests in Chinese power plants indicated that the SCR system can oxidize about 16% of elemental mercury and reduce about 32% of total mercury, and the overall Hg removal efficiency of the combined SCR + ESPs/FFs + WFGD configuration can reach over 80% (Wang et al., 2010; Zhang et al., 2008). Thus, we may anticipate that additional about 10∼20% of Hg emission reduction can be obtained if SCR systems are widely applied in the coal-fired power plants in China.

Activated carbon injection

Activated carbon is an effective sorbent for mercury capture from flue gas. Many years of research, development, and over 50 full-scale demonstrations have shown that activated carbon injection (ACI) can greatly reduce mercury emissions from most configurations in coal-fired power plants in the United States (New Jersey Association of Counties [NJAC], 2010). Commercial ACI systems began to be sold to the power generation industry in 2005 and ACI is now considered the most robust technology for Hg control at coal-fired power plants. By now, ACI systems have been widely installed at coal-fired power plants in the United States, and most of which can achieve over 90% of mercury reduction (NJAC, 2010; Sjostrom et al., 2010). However, the Chinese power industry has obtained few experiences on design and operation of actual ACI system to date. The less stringent emission limit, added capital and operation cost, as well as possible impacts on boilers operation and ash performance, will affect the willingness of electricity enterprises on ACI application. Thus, successful widespread implementation of this technology throughout coal-fired power plants in China will encounter many challenges and require continued secondary development efforts, including (1) understanding the impacts of technologies to control other pollutants, such as SO₃, for the enhancement of particulate control or SCR for NO_x control; (2) some full-scale demonstration projects should be conducted to verify the effectiveness of this technology on different Chinese coals, and possible ways to continue reductions in capital and operation costs; (3) options to continue using ash containing activated carbon in cement and concrete industry; (4) reliable techniques to assure the quality of delivered carbon; (5) techniques to improve the effectiveness of activated carbon; and (6) domestic manufacturing facilities to produce sufficient carbon supply (NJAC, 2010; Sjostrom et al., 2010).

Stop mining and burning of coal with high Hg content

As indicated in our results above, the provincial-level average Hg content in coal is the most important parameter to control the final Hg emissions from coal-fired power plants. Wu et al. (2010) have demonstrated that the majority of uncertainty in their Hg emission estimation results from the mercury content of coal through stochastic simulation. Thus, one may think it is a good option to stop mining and burning coal with high Hg content, especially for provinces such as Guizhou and Guangxi. However, generating electricity from coal-fired power plants by burning locally mined coal and transporting electricity to eastern Guangdong province is a part of China's national strategies of Transforming Electricity from West to East, which is intending to simultaneously promote the economic development in the western provinces and solving the electricity supply shortage in the southern developed province. Therefore, it is not so easy to totally ban mining and burning such coals, because most of coal reserved in these areas is found with high Hg content, local economy and the residential daily life are all highly relying on local coal mining and burning.

Coal washing

Coal washing before burning is regarded as an optional effective way to reduce Hg content in coal, though the main purpose is to wash away part of ash and SO₂ and to improve the heating value of cleaned coal simultaneously. The removability of Hg during coal washing is highly dependent on the modes of occurrence of Hg in coals as well as the washability of inorganic minerals. Normally, Hg demonstrates a strong affinity to inorganic materials and the most likely forms of occurrence are association with sulfides, especially with pyrite. It is reported that sulfur in pyrite can be removed effectively during physical coal washing and the removal efficiency can reach up to 50% (Finkleman, 1994; Song et al., 2006). Consequently, a large proportion of Hg associated with pyrite can be washed away simultaneously, and almost all the physical coal washing processes tested on Chinese coals demonstrate Hg removal efficiency higher than 50% (Song et al., 2006; Wang, 2007). However, only about 20% of total raw coal mined are washed, and are primarily used for coke making in iron and steel industry (SACMS, 2009). The majority of power plants in China are burned with raw coal directly till now. The main reasons include insufficient cleaned coal supply and elevated price, lower price of raw coal, as well as the coal-mining companies are not willing to wash raw coals and supply cleaned coals in case of added capital investment and possible secondary water and soil pollution risks.

SCR + ESPs/FFs + WFGD configuration

Generally speaking, the integrated situation of fossil-fuel resource reserve conditions of relatively rich in coal while lacking oil and natural gas, the complex international market of oil and natural gas supply, as well as the current status of renewable energy technologies have determined that coal will be the dominant energy source in the foreseeable future. It implies that more and more coal output will be used for generating electricity by coal-fired power plants, owing to the increasing demand for clean electricity. Thus, except for rising the ratio of raw coal washing before burning, applying the combination of SCR + ESPs/FFs + WFGD configuration may be the best available choice on accounting of technical feasibility and cost-benefit for coal-fired power plants in the near future, which can achieve effective co-benefit Hg reduction. Also, advanced specific Hg removal technologies with high efficiency and lower cost such as ACI system are especially welcoming for deep Hg abatement in the long term. However, application of these new control devices may bring different burdens and potential side effects to the coal-fired power plant enterprises, such as elevating capital and operation costs, effectiveness for different types of coal, adaptability for local dispatch pattern and electric load variation, and influence on the ash quality for cement and concrete use. All these factors should be paid much attention and taken into consideration comprehensively.

Energy conservation and renewable energy

Finally, even with widespread application of SCR, FGD, and ESPs/FFs, as well as some specific advanced Hg control technologies such as ACI, Hg release from coal-fired power plants cannot be totally eliminated. Thus, energy conservation and substitution of some coal burning with other clean energy sources for electricity generation in China, such as wind power, nuclear power, hydropower, and solar power, may be the best effective options in the medium- and long-term future.

Conclusions and Recommendations

In this paper, we have assessed the temporal trend and characteristics of atmospheric Hg emissions from coal-fired power plants in China, and proposed preliminary control strategies for Hg reduction accordingly. With the continuous economic growth and increasing demand for electricity, much more coals will be used for power generation. It is anticipated that over 2 billion tones of coal will be fired by power plants by 2020, almost all coal-fired power plants will be mandated to be equipped with SCR/SNCR, ESPs/FFs, and FGD to minimize the emissions of NO_x, PM, and SO₂, and the main species of Hg emitted will be the elemental mercury. Atmospheric Hg emissions from coal-fired power plants will be a great concern not only for local and regional environment and human health risks, but also for widespread impacts owing to its long lifetime and long-distance transboundary transportation in the atmosphere.

Thus, integrated control measures should be promulgated and implemented to minimize the final Hg emissions. In case of the current status of technology development and economic cost, the combination of SCR + ESPs/FFs + WFGD will be the best available path both in technical feasibility and cost-benefit for coal-fired power plants in the near future, which can achieve substantial co-benefit Hg reduction in China, whereas the advanced specific Hg emission control technologies such as activated carbon injection should be developed and demonstrated to test the adaptability for different types of Chinese coals, such as bituminous, anthracite, and lignite. Also, energy conservation and substitution of some coal with other clean energy sources for electricity generation may be the best options to eliminate Hg emissions in the long term. Our results and suggestions should be beneficial for better understanding the historical trend and characteristics of Hg emissions from coal-fired power plants, and also helpful for policy making on further Hg abatement in China.

Acknowledgments

This work is funded by the National Natural Science Foundation of China (20677005, 40975061, and 21177012), the Beijing Natural Science Foundation (8113032), and the 2011 key consulting program of Chinese Academy of Engineering (CAE) on Sustainable Clean Coal Exploitation and Utilization in China. The authors thank the editors and anonymous reviewers for their valuable comments and suggestions to improve this paper.