Analysis of Reduction Potential of Primary Air Pollutant Emissions from Coking Industry in China

Air pollutant emissions from the coking industry in China, the world’s largest coke producer, are a major societal concern. This study employed a bottom-up emission-factor methodology to estimate the country’s coking-industry airpollutant emissions. Total suspended particulate matter (TSP), particulate matter with an aerodynamic diameter less than or equal to 2.5 μm (PM2.5), sulfur dioxide (SO2), nitrogen oxides (NOx), volatile organic compounds (VOCs), polycyclic aromatic hydrocarbons (PAHs), carbon monoxide (CO), and methane (CH4) were considered. The emissions in 2015 were estimated to be 368.36, 23.28, 402.54, 174.43, 1325.42, 28.24, 2036.43, and 71.68 kt for TSP, PM2.5, SO2, NOx, VOCs, PAHs, CO, and CH4, respectively, with an annual average growth rate of 1.1% during the 12th five-year-plan (2011– 2015). A comparative analysis was performed on emission contributions from air pollutants produced by various types of coke ovens and using various coking procedures. The objectives of the 13th five-year-plan (2016–2020) were used to predict China’s production of coke and methods of control in 2020 and to analyze the potential reduction in typical air pollutants. The results show that emissions of TSP, SO2, and NOx will be reduced by 82.9%, 94.4%, and 6.9%, respectively. Moreover, this study proposes a series of feasible control measures for air-pollutant emissions from the coking industry in China.


INTRODUCTION
Coke is a crucial raw material and a fuel used in blast furnace ironmaking.The rapid development of China's steel industry has increased the status of the country's coking industry and emphasized its role in China's economic development (Hu et al., 2010;CAE, 2011).The world's largest coke producer, China generated a total coke output of 447.78 Mt in 2015, which accounted for 68.6% of the world's total output (NBSC, 2016).However, the coking industry with coal as the primary raw material is characterized by a relatively low energy utilization rate, high energy consumption per unit, complex process flows, abundant pollution-producing links, and it's being a large source of highly polluting matters (Zhang, 2000;Li, 2009;Tian et al., 2010;Huo et al., 2012;Mu, 2013).
Therefore, the coking industry is considered a highly polluting industry.The exhaust gas pollutants generated during coking include total suspended particulate matter (TSP), particulate matter with aerodynamic diameters less than or equal to 2.5 µm (PM 2.5 ), sulfur dioxide, nitrogen oxides, volatile organic compounds (VOCs), and polycyclic aromatic hydrocarbons (PAHs).The coking industry is also a substantial contributor to greenhouse gas emissions (Zhu et al., 2001;He et al., 2006;Liu et al., 2009;Mu, 2013;Ge et al., 2016;Wu et al., 2017).Long-term exposure to coke oven emissions results in inflammation of the skin and inner eyelids and lesions in the lung and stomach (Lloyd, 1971;IARC, 1984IARC, , 2002)).The United States Environmental Protection Agency (US EPA) classifies coke oven emissions as Group A carcinogens and links them to lung cancer (US EPA, 1999).
In March 2014, China's Ministry of Industry and Information Technology announced its latest revision of the coking industry admission conditions, which proposed restrictions for coking companies through concrete rules for the production process and equipment used and the energy consumption per unit product of coke, in addition to formulating supplementary environmental protection measures (MIIT, 2014).Data from the China Coking Industry Association (CCIA) indicated that, of the 420 companies that applied for coking company admission qualification before 2016, 367 were admitted, and these companies combined for an actual coke output of 377.66 Mt, or approximately 50.0% of the country's total output (CCIA, 2016a).Since January 1st, 2015, a new version of the Emission Standard of Pollutants for the Coking Chemical Industry has been take effect (MEP, 2012).The part of China's 13th five-yearplan (2016-2020) that concerns the development of the coking industry suggests that the coking industry should meet the following conditions to ensure that all companies achieve their respective emission targets (NDRC and NEA, 2016): (1) the coking industry must eliminate coking companies that fail to meet the standard output; (2) 70.0% of the output standard for those admitted into the industry access conditions.
China's coking industry is, however, characterized by low production concentration, generally less advanced production technologies and lower management quality, abundant small enterprises loosely distributed across the nation, and a high percentage (> 80.0%) of coking companies being independent small-and medium-sized enterprises (CCIA, 2016b).For small-and medium-sized coking companies in deficit that do not use advanced technologies, meeting the standard environmental protection targets is challenging due to the high cost of energy reservation and emission reduction.
Several studies have estimated the emissions of typical air pollutants from China's coking industry, such as PAHs, greenhouse gases, and SO 2 , etc. (Zhu et al., 2001;He et al., 2006;Liu et al., 2009;Mu, 2013;Ge et al., 2016;Wu et al., 2017).However, studies analyzing the emission characteristics of the coking industry have been mainly performed for the steel industry (Wu et al., 2015a).Coking companies serving as auxiliary enterprises in the steel industry only account for 13.3% of the total number of coking companies in China, implying that few studies have collected and analyzed data of coking-industry-only companies (CCIA, 2016c).
Presently, coke used in metallurgical industries (including the steel industry) is mainly produced through the coking of coal.The ability to reduce emission for air pollutants is a crucial factor for overall pollution prevention and control in the coking industry.Reducing and better monitoring the emissions of air pollutants from coke ovens is required to improve ambient air quality, maintain a sufficient level of coke production, and meet ongoing new regulations.Therefore, the emissions of typical primary air pollutants from the coking industry in China were estimated in this study.The air pollutants considered included TSP, PM 2.5 , SO 2 , NO x , VOCs, PAHs, CO, and CH 4 .The future emissions reduction potential was projected based on scenario analysis to provide a theoretical basis and technical support for emission controls and the implementation of emission standards in heavily polluting industries.

Methodology
The "bottom-up" emission factors methodology was used to estimate the atmospheric emissions of air pollutants typically emitted by the coking industry (TSP, PM 2.5 , SO 2 , NO x , VOCs, PAHs, CO, and CH 4 ).This methodology demonstrated a diverse approach for estimating the air pollution emissions of different industries and included production processes, enterprise scales, air pollution control devices, and geographical distribution.The historical trends in air pollutant emissions during the 12th five-yearplan (2011)(2012)(2013)(2014)(2015) were summarized.The emissions were calculated by combining the best available activity data and the specific emission factors of different air pollutants.The emission inventory can be compiled as 3 , , , , , , , , where E is the emission of air pollutants (unit kt); A is the coke produced (kt); EF is the air pollutant emission factors during coking process (kg t -1 ); i denotes the air pollutant type; j denotes the coke oven types (including topcharging, stamp-charging, and heat recovery coke ovens); k denotes the heights of the coke oven's carbonization chamber, classified as less than 4.3 m, 4.3-6.0m, or more than 6.0 m; l denotes the Chinese provinces in which the oven is located; p denotes different coking processes including coal charging, coke pushing, coke transportation, coke quenching, and chimney emissions; and y denotes the year.The potential emission reduction at the end of the 13th five-year-plan (2016-2020) was projected using the following equations: where P is the potential emission reduction (kt); FV is the flue gas volume in the coking process (m 3 t -1 ), the FV parameters for different coking processes are presented in SI Table S5); EL is the air pollutant emission limitation defined by the emission standard of the coking industry (mg m -3 ); T is the target year; and C is the base year.

Activity Data Collection and Analysis
Activity data statistics including the application ratio of different coke oven types, carbonization chamber types, and specific coke production data were collected from the CCIA (CCIA, 2016d).In 2015, the worldwide coke production output was 652.41 Mt, of which China produced 68.6%.China's coke production was 13.6-times that of Japan (33 Mt), which was the second largest producer.The yearly variations in coke production output among the world's primary coke-producing countries are illustrated in Fig. 1.China produced 447.78 Mt of coke in 2015, with a coke production capacity of 686.76 Mt and an energy utilization rate of 65.2%, suggesting a surplus coke production capacity.Data on the coke output per production unit in China's 29 administrative regions in 2015 are presented in SI Table S1 and reveal that the most productive regions were Shanxi, Hebei, Shandong, Shaanxi, and Inner Mongolia.Together, these five regions accounted for 54.9% of China's total coke output.
The number of coking companies in China decreased from more than 730 in 2010 to 602 at the end of 2015; of these, 50 were heat recovery coke oven plants, 72 were semi-coke plants, and 480 were traditional coke plants (80 plants in the steel industry and 400 independent plants).The annual average coke production output of coking enterprises increased from 0.68 Mt per year in 2010 to 1.14 Mt per year in 2015; 367 companies had been admitted into the coking industry by the end of 2015 and had an output of 377.66 Mt, of which 328.73 Mt was generated from conventional coke ovens (55.4%).The respective percentages of all coke ovens that used various technologies in these admitted companies are presented in SI Fig. S1.

Emission Factors
The coke production techniques used in the coking industry are generally divided into top charging, stamp-charging, and clean heat recovery.In particular, top-charging coke ovens are the most complex and most frequently applied.China's relevant policies (of gradually eliminating outdated and highly polluting coke ovens with a height of less than 4.3 m and encouraging large-scale coke ovens), because of the distinct characteristics of pollutants emission from coke chambers with various heights, divide top-charging coke ovens into three height sizes: < 4.3 m, 4.3-6.0m (4.3 m included), and > 6.0 m.The sizes of stamp-charging and heat recovery coke ovens are not categorized in detail because they differ slightly in specifications, have only been in use for a relatively shorter period of time, feature coke chambers of similar heights, and emit similar types of pollution.Recent domestic studies were used to obtain the air pollutant emission factors for the various coke production procedures (coal charging, coke pushing, transport, quenching and emissions through chimneys) of different types of coke ovens (see Tables 1 and 2) (Li, 2009;FNPSCD, 2010;Mu, 2013;MEP, 2014a, b).

Potential Emission Reduction Projection
In 1996, China stipulated the Emission Standard of Air Pollutants for Coke Ovens (GB 16171-1996), which only specified tier 1-3 emission limits for TSP, benzene-soluble organics, and benzo (a) pyrene that were released from coke ovens (SEPA, 1996).Because the present scale and quality of China's coking industry have undergone substantial upgrades, this standard is limited in scope and thus cannot meet current requirements regarding to pollution prevention and control, environmental management, and environmental quality improvement, resulting in certain difficulties in the legal enforcement of environmental protection policy.
On January 6th, 2016, the CCIA announced the outline of the 13th five-year-plan concerning the development of the coking industry, which specified more stringent requirements for coking-industry development than those in the previous plan (NDRC and NEA, 2016).The objectives of the 13th five-year-plan and the status quo of China's coking industry were used to conduct a scenario analysis of various types of coke oven.In addition, the emission limits proposed  ) were adopted to estimate the emission factors for air pollutants that will be released from these coke ovens in 2020 (MEP, 2012;NDRC and NEA, 2016).

Temporal and Spatial Distribution Characteristics
The historical trends and spatial distribution of the atmospheric emissions of typical air pollutants from the coking industry during 2011-2015 are illustrated in Fig. 2. The 2015 emissions of typical air pollutants were estimated to be 368.36kt for TSP, 23.28 kt for PM 2.5 , 402.54 kt for SO 2 , 174.43 kt for NO x , 1325.42 kt for VOCs, 28.24 kt for PAHs, 2036.43 kt for CO, and 71.68 kt for CH 4 , respectively, and the average annual growth rate was 1.1% during the 12th five-year-plan.In the implementation of the Access Conditions of the coking industry (revised in 2014), backward production capacity was banned and optimized, which caused a substantial moderation of the rate of increase, from 7.1% in 2013 to 0.8% in 2014 (MIIT, 2014).Moreover, emission limits for most coking enterprises that were stipulated by the emission standard caused a decrease in emissions in 2015 of 6.1% (MEP, 2012).
As illustrated in Fig. 2, air pollutant emissions were concentrated in Shanxi, Hebei, Shandong, Shaanxi, and Inner Mongolia, which are located in Northern China, and these regions accounted for 54.8% of total emissions.This suggests that these five regions, in which independent coking companies are highly concentrated, accounted for more than half of China's total coke output.

Emission Characteristics of Various Coking Procedures and Types of Coke Oven
Due to the lack of data on some emission factors, the present study estimated the emissions of only three air pollutants (TSP, SO 2 , and NO x ) by comparing emissions from various types of coke oven and using various coking procedures, namely coal charging, pushing, transport, and quenching, in addition to emissions through chimneys (Fig. 3) (Li, 2009;FNPSCD, 2010;Mu, 2013;MEP, 2014a, b).The estimation results demonstrated that top-charging coke ovens with heights 4.3-6.0m and stamp-charging coke ovens were the major sources of air pollutant emissions because they were highly productive and exhibited relatively high emission factors for the three air pollutants.TSP was mainly emitted from coal charging and coke pushing and transport, whereas coke quenching and coal charging were the main contributors of CO and CH 4 , respectively.
Coke is typically produced in China using conventional mechanical coke ovens with horizontal chambers or using heat recovery coke ovens incorporating stamp-charging technology, which produce air pollutants primarily from coal charging, coke pushing, quenching, and desulphurization (Wang, 2015;CCIA, 2016d).
Coal charging produces three types of exhaust gas: (1) hot air excluded from coking chambers during coal  (2) raw gas generated from the volatilization and decomposition reactions of coal in contact with highly heated coking chamber walls; and (3) water vapor produced from the vaporization of moisture in the raw coke.Ascending hot air in the oven, raw gas produced from coal decomposition, and water vapor leaking through the coal charging hole or oven doors can release substantial smoke and dust that severely pollute the environment.Research indicates that, on a short-term basis, coal charging alone produces a TSP amount seven times that produced from coking procedures, and accounts for 50.0%-60.0%of the total pollutant emission from the overall coking process (Wang, 2015).
During coke pushing, TSP can leak during repeated oven-door opening and the pushing of coke from the coke oven.The emission of pollutants from the coke oven during this process is characterized by multiple leak points, wide dissemination, dispersed distribution, multiple pollutant types, high degrees of hazard, and difficulty in hazard prevention and control.
Coke quenching is mostly performed using two methods: wet and dry quenching.Wet quenching involves spraying quenching water onto red-hot coke delivered into the quenching tower, during which considerable water evaporation and pollutant effusion occur.This method is currently employed worldwide because of its simplicity and low cost.Dry quenching can be used to recover the sensible heat of coke and produces less pollutant than wet quenching.In China, dry quenching is used to produce only 32.0% of the total domestic coke production.

Emission Comparison of Heavy Polluting Industries and Uncertainty Analysis
Data from several highly polluting industries in China that are the target industries for national pollution control (i.e., the thermal power, ceramics manufacturing, steel smelting, and coking industries) were collected and their air pollutant emissions compared (Table 3) (Chen et al., 2014;Wu et al., 2015a, b;Wang et al., 2016;Hua et al., 2016).Of these industries, the coking industry emitted the highest amount of VOCs.The emissions estimation of the study of Wu et al. (2015a) contained the emission processes of sintering, pelleting, iron-making, steel-making, and steel rolling processes of iron and steel manufacturing industry.In China, approximately 35% coke production capacity concentrates in iron and steel manufacturing enterprises in 2015.However, the emissions of coking process which belongs to iron and steel manufacturing industry were not considered.
Monte Carlo simulations were employed to quantify the uncertainties in emission estimates (Zheng et al., 2009;Zhao et al., 2011;Cheng et al., 2015).A normal distribution with a coefficient of variation (CV, the standard deviation divided by the mean) of 30.0% was assumed for coke production.With regard to the uncertainties in the estimates, we performed a sensitivity analysis of the total emissions of various pollutants.Given that uncertainty values were not available for some of the inputs, assumed values are assigned to each input to test the sensitivity of the results to each.The uncertainty of different model inputs (SI Table S2) was assigned randomly.The overall uncertainties were in the approximate range -110.3% to +90.8%.The large uncertainties were mainly because of incomplete investigations and inadequate field test data for the activity data and coking industry emission factors.Field tests of the characteristics of air pollutants emitted by the coking industry are urgently needed for future research.

Emission Reduction Potential Analysis and Relevant Pollution Control Proposals
The objectives of the 13th five-year-plan and the current status quo in China's coking industry (see SI Table S3) were used to estimate the production output, percentage of various coke ovens used, and air pollutant emission factors in 2020 (SI Fig. S2 and SI Table S4) (CCIA, 2016c;NDRC and NEA, 2016;MEP, 2016).The estimation results revealed that the number of top-charging coke ovens with heights < 4.3 m will decrease from 3.0% 2015 to becoming completely banned in 2020.By contrast, the number of top-charging coke ovens with heights > 6 m will increase from 6.0% in 2015 to 35.0% in 2020.The percentages of other oven types used will also change.Fig. 4 presents the estimated potential emission reduction of typical air pollutants generated from the coking industry that will occur under the 13th five-year-plan.The scenario analysis revealed that, by the end of the plan, the emission of TSP, SO 2 , and NO x from China's coking industry will be reduced by 82.9%, 94.4%, and 6.9%, respectively, down to 62. 84, 22.72, and 162.32 kt.To realize the objectives of the 13th five-year-plan concerning the coking industry and reduce the emission of air pollutants, clean production technologies must be further promoted to minimize pollution at its source, increase resource utilization rates, reduce or prevent the generation and emission of pollutants, and thereby reduce environmental hazards and the risks to human health.Because coke production involves multiple pollutionproducing steps and its air pollutant emissions are severe, highly economical technologies must be employed for all potential pollution-producing steps.A number of pollution control technologies and methods can be used in the coking industry (MEP, 2010;CCIA, 2016d).The specific control technologies for different coking procedures and their corresponding air pollution control effects are illustrated in SI Table S7.Approaches for coal storage and blending include large coal storage silos, pneumatic selective crushing, coal moisture control (CMC), CMC integrated with pneumatic classification and separation, and the coal blending coking technique.For coking procedures, large coke ovens (height of coking chamber ≥ 6 m, volume ≥ 38.5 m 3 , top-charging) and the stamp-charging coking technique (height of coking chamber ≥ 5.5 m) are preferable.Coke quenching techniques that facilitate the control of pollutant emissions include dry quenching, low-moisture quenching, and conventional wet quenching.Control methods feasible for use in coal-gas purification include large coal storage silos, pneumatic selective crushing, CMC, CMC integrated with pneumatic classification and separation, and the coal blending coking technique.Pollution can be controlled during coal-gas purification by using desulfurization and decyanation techniques relying on the vacuum carbonate, Sulfiban, and HPF processes.Urban air particulate matter can be mitigated using dust-removal filter bags, wind-proof dust nets, large ground stations employing dry dust-removal and purification technologies, gas collection using side-mounted suction tubes, and dust-removal technologies using fog-removing devices and baffle plates, all of which have a dust-removal efficiency of more than 99.0%.To meet the coking emission standard, VOCs emissions must be controlled by using absorption technologies to collect organic solvents or employing catalytic and thermal combustion technologies.

CONCLUSIONS
In the present study, a comprehensive emission inventory of typical primary air pollutants (TSP, PM 2.5 , SO 2 , NO x , VOCs, PAHs, CO, and CH 4 ) from the coking industry was established by using the bottom-up emission-factors methodology for the period 2011-2015.The emissions were estimated at 368.36 kt for TSP, 23.28 kt for PM 2.5 , 402.54 kt for SO 2 , 174.43 kt for NO x , 1325.42 kt for VOCs, 28.24 kt for PAHs, 2036.43 kt for CO, and 71.68 kt for CH 4 in 2015 at an annual average growth rate of 1.1% during the 12th five-year-plan.These air pollutants were mainly concentrated in northern areas of China such as Shanxi and Hebei, where coke is produced in considerable volume and coking companies independently operate plants that use relatively less advanced technology.Coal charging, coke pushing, and transport contributed to the primary TSP emissions, while coke quenching and pushing were the main contributors of CO and CH 4 , respectively.Moreover, future emissions and the potential reduction in typical air pollutants during the 13th fiv-year-plan (2016-2020) were projected in the scenarios for coke production and control technologies by considering the development plan and emission restrictions of the coking industry.The reduction potential of the emissions of TSP, SO 2 , and NO x will reach 82.9%, 94.4%, and 6.9%, respectively, by the end of 2020.In addition, to support the emission-reduction potential and enhance air quality, a series of feasible control proposals for each pollutant-discharge procedure has been illustrated.

Fig. 1 .
Fig. 1.Historical trend of coke production in major countries of the world.

Fig. 2 .
Fig. 2. Temporal and spatial distribution of air pollutant emissions from coking industry.

Fig. 3 .
Fig. 3.The emission of typical air pollutants of different coking procedures and types of coke oven.

Fig. 4 .
Fig. 4. Projection of emission reduction potential of typical air pollutants until 2020.

Table 1 .
Emission factors of TSP, SO 2 , and NO x from different coking technology and production processes, kg t -1 .

Table 2 .
Emission factors of VOCs, PAHs, CO, and CH 4 from different coking technology and production processes, kg t -1 .

Table 3 .
The comparison of the emissions of typical air pollutants of heavy polluting industries.