Mercury Transformation and Distribution Across a Polyvinyl

Jan 15, 2014 - The production of polyvinyl chloride (PVC) via the calcium carbide process utilizes a catalyst containing large amounts of mercury (Hg)...
0 downloads 0 Views 662KB Size
Article pubs.acs.org/est

Mercury Transformation and Distribution Across a Polyvinyl Chloride (PVC) Production Line in China Wen Ren,† Lei Duan,† Zhenwu Zhu,‡ Wen Du,‡ Zhongyi An,‡ Lingjun Xu,‡ Chi Zhang,‡ Yuqun Zhuo,*,‡ and Changhe Chen‡ †

School of Environment, State Environmental Protection Key Laboratory of Sources and Control of Air Pollution Complex, Tsinghua University, Beijing 100084, P. R. China ‡ Department of Thermal Engineering, Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Tsinghua University, Beijing 100084, P. R. China ABSTRACT: The production of polyvinyl chloride (PVC) via the calcium carbide process utilizes a catalyst containing large amounts of mercury (Hg) and is therefore one of the most important sources of anthropogenic Hg in China. To measure the emission of Hg from PVC production, we established a flowchart for the calcium carbide process, for which we quantified the Hg content of the material/product at each step. Results indicated that 71.5% of the total Hg (HgT) was lost from the catalyst, most of which was recovered by the Hg remover, accounting for 46.0% of the total Hg (HgT). We determined that 3.7% of the HgT was released into the environment, mostly in solid wastes and byproducts such as hydrochloric acid. Furthermore, no Hg has been detected in the PVC end product. However, we were only able to account for 78.1% of the Hg across the whole system, leaving 21.7% unaccounted for in the mass balance. A rough estimation indicates that most of the “missing” Hg had accumulated in deposits on the inner surface of converters and downstream pipelines; however, the emission to the atmosphere was ≤1% of the HgT. For a PVC production line equipped with a Hg remover, emissions of Hg to the atmosphere have been estimated to be 4.9 g per tonne PVC. Currently, almost all calcium carbide facilities have been equipped with a Hg remover, which may reduce the release of Hg in China by ∼500 t/year.



INTRODUCTION In recent years, atmospheric emissions of mercury (Hg) have become a global concern,1,2 and a new international protocol for reducing the use and release of Hg was to be signed in 2013.3 Historically, Europe, followed by North America, has been the major region for anthropogenic Hg emissions. However, after substantial reductions in Hg emissions from these regions over the past three decades, China is now the largest source of Hg emissions to the atmosphere.4,5 China emits at least 700 t (metric ton) of Hg per year, mainly from coal combustion, metal smelting, and cement production, which is ∼1/4 of the total global emissions.6−8 Currently, China is also the world’s largest consumer of Hg; annual use of Hg is >1000 t/year, which accounts for ∼50% of the global use.8 Therefore, heavy pressure is being placed on China to reduce the use and release of Hg into the environment. The process for producing polyvinyl chloride (PVC) utilizes the largest amount of Hg in China.9 PVC is a plastic material that is commonly used in sewage pipe, window/door frames, and as a cable insulator due to its low cost and strong resistance to corrosion. By the end of 2010, global production of PVC was ∼49 Mt/year.9 Most of the producers are distributed throughout Asia, North America, and Western Europe. China’s PVC production capability is 20.43 Mt/year, which represents 41.7% of the global production. Due to the economic downturn © 2014 American Chemical Society

in recent years, China produced only 11.3 Mt of PVC in 2010, but remained the world’s largest PVC producer. Unlike other countries, most of the PVC (80.9%) produced in China is made through the calcium carbide process9 because China has larger sources of coal than oil. The catalyst used in this process is activated carbon impregnated with Hg chloride (HgCl2). The catalyst, which normally contains 8−15% HgCl2, is used to ensure the reactivity and selectivity of vinyl chloride monomer (VCM) formation. It has been estimated that to produce 1 t of PVC through the calcium carbide process used in China, 1−2 kg of Hg-impregnated catalyst is consumed.10 Therefore, for the annual production of 8 Mt of PVC through the calcium carbide process, >800 t of Hg is required. No doubt exists that the production of PVC from the calcium carbide process is one of the most important sources of anthropogenic Hg in China. Although many attempts have been made to recover Hg during the process, a considerable amount of Hg may be lost in byproducts and emissions to the environment. Very few reports have investigated the transformation and distribution of Hg in this process. Therefore, the Received: Revised: Accepted: Published: 2321

September 22, 2013 January 9, 2014 January 15, 2014 January 15, 2014 dx.doi.org/10.1021/es404147c | Environ. Sci. Technol. 2014, 48, 2321−2327

Environmental Science & Technology

Article

Figure 1. Simplified diagram for PVC production process.

goals of this research were to establish a flowchart for the transformation and distribution of Hg in the calcium carbide process and to conduct a mass balance of Hg across the whole system. The results of this research will provide useful information on Hg recovery and catalyst development in the chloro-alkali industry.

excessive HCl, C2H4Cl2, and other impurities from the gas stream. The major process parameters have been summarized in Table 1. Figure 3 can also be referenced for the detailed process of the PVC production line investigated.



Table 1. Major Process Parameters of a 200 000 t per Year PVC Production Line

MATERIALS AND METHODS PVC Production Line. The PVC production line we investigated is located in central China and is capable of producing 200 000 t of PVC per year. It is a typical, large-scale calcium carbide facility and accounts for ∼2% of the PVC production in China. A simplified diagram of the PVC production process is displayed in Figure 1. The chemical reactions related to PVC production in each major piece of equipment include the following.

parameter PVC production per year annual operation hours C2H2 temperature C2H2 pressure C2H2 flow rate HCl temperature HCl pressure HCl flow rate

Electrolyzer: 2NaCl + 2H 2O → 2NaOH + Cl 2 + 2H 2 (1)

Combustor: Cl 2 + H 2 → 2HCl

(2)

Hydrolyzer: CaC2 + 2H 2O → C2H 2 + Ca(OH)2

(3)

Converter: C2H 2 + HCl → C2H3Cl

(4)

Polymerizer: nC2H3Cl → − [CH 2 − CHCl]n −

(5)

Stage I VCM conversion rate Stage II VCM conversion rate temperature at converter exit pressure at converter exit

It is worthwhile to point out that, in the above reactions, only Reaction 4 was catalyzed by HgCl2. Converters containing HgCl2 catalyst (HgCl2-impregnated activated carbon in pellet form) were arranged in 2 stages in series. Catalyst in converters was updated periodically. Fresh catalyst was always installed into Stage II (the latter stage) to ensure the maximum C2H2 conversion, while the used catalyst, usually the catalyst in its second half-life, was installed into Stage I. Some catalyst sludge (cracked activated carbon containing large amount of moisture from upstream) might be collected from the bottom of converters when catalyst was transferred from Stage II to Stage I, and when withdrawn from Stage I in the end. Apart from C2H3Cl (VCM), the byproduct of C2H4Cl2 was inevitably produced in converters via the following reaction: Converter side reaction: C2H 2 + 2HCl → C2H4Cl 2

remarks 200 kt/y 8000 h 10 °C 150 kPa 6500 m3/h 10 °C 150 kPa 7150 m3/h

full load operation hour absolute pressure operating conditions absolute pressure Operating conditions, 110% of C2H2 flow rate

71.5% 96% 65 °C 114 kPa

Absolute pressure

Sampling and Analysis. While the production line was operating at full capacity, samples of raw materials, and intermediate and final products were collected for Hg analysis. The sampling points are presented in Figure 3. Solid samples included fresh catalyst, catalyst sludge, waste catalyst, calcium carbide, Ca(OH)2 sludge, and PVC. Liquid samples included freshwater, condensed water, wastewater, sodium chloride (NaCl) solution, HCl, dichloroethane (C2H4Cl2), and waste alkali solutions. All liquid samples were stored in glass bottles that had been previously cleaned with 0.1 N HCl and rinsed with deionized water. The Hg concentration in all intermediate and final products was determined onsite. Hg content in liquid and solid samples, except PVC, was analyzed using a Lumex RA-915+ Portable Zeeman Mercury Analyzer assisted by a Pyro-915+ Atomizer.11 The maximum Hg detection limit for various kinds of samples was 5 ng/g. Hg content in PVC was analyzed by ICP-AES after digestion. Gaseous samples included C2H2, VCM (Stage I and II exit gas, and distilled VCM), distillation exhaust, and dryer exhaust. To differentiate Hg species in gas samples, the US ASTM

(6)

To ensure the quality of the final product of PVC, a series of process gas cleaning equipment, including Hg remover, acid/ alkali scrubbers, and distillation column, was installed between Stage II converters and polymerizer to remove Hg, water, 2322

dx.doi.org/10.1021/es404147c | Environ. Sci. Technol. 2014, 48, 2321−2327

Environmental Science & Technology

Article

Figure 2. Apparatus for sampling Hg2+ (left) and HgT (right).

Figure 3. Flowchart for the PVC production line and the Hg concentrations (in red).

D6784 method was applied.12 As shown in Figure 2, two scrubbers containing 100 mL of 1 N potassium chloride (KCl) were connected in series to capture oxidized Hg (Hg2+). The amount of gas passing through the scrubbers was recorded with a precalibrated volumetric flowmeter. For quantifying total Hg (HgT) in the gas stream, we used U.S. EPA method 30B13 to develop an appropriate sampling method for gases containing high quantities of Hg. The sorbent trap was a glass tube (length 15 cm; i.d. 6 mm) packed with activated carbon that was separated into two sections, the measurement section and verification section, by quartz wool. Gas was drawn into the trap by a syringe connected to the verification section. After several trials, we determined that only 20 mL of gas could be analyzed due to the very high concentrations of Hg to avoid penetration of the measurement section by the sample. The temperature and pressure of the sample environment were also recorded for further calculations. If the gas stream contained relatively low concentrations of elemental Hg (Hg0;