Environ. Sci. Technol. 2008, 42, 5971–5977
Global Emission and Production of Mercury during the Pyrometallurgical Extraction of Nonferrous Sulfide Ores LARS D. HYLANDER* AND ROGER B. HERBERT Department of Earth Sciences, Uppsala University, Villava¨gen 16, S-752 36 Uppsala, Sweden
Received February 18, 2008. Revised manuscript received May 20, 2008. Accepted May 26, 2008.
The contribution of the milling, smelting, and refining of sulfide ores to Hg emissions and to Hg byproduction is not adequately quantified in a global context. In this study, we estimate Hg emissions from the pyrometallurgical treatment of Cu, Pb, and Zn sulfide ores. We base our calculations on quantities processed and Hg content in Cu, Pb, and Zn concentrates, derived from unique global databases on smelter feed and production. In 2005, about 275 tons of Hg were emitted globally to the atmosphere from Cu, Pb, and Zn smelters. Nearly one-half was emitted from Zn smelters and the other half equally divided between Cu and Pb smelters. Most Hg was emitted in China, followed by the Russian Federation, India, and South Korea. Global emission factors were 5.81, 15.71, and 12.09 g of Hg ton-1 of metal for Cu, Pb, and Zn smelters, respectively. Calculations indicate that Hg abatement technologies applied to flue gases may have recovered 8.8 tons and 228 tons Hg from Pb and Zn smelters, respectively, most of which was probably sold as a byproduct. In conclusion, Hg emitted from processing copper, lead, and zinc ores has been largely underestimated in Hg emission inventories. Reducing these emissions may be one of the most economical measures to reduce global Hg emissions.
Introduction Mercury (Hg) is a heavy metal of worldwide concern because of its toxicity to living cells (1). Methyl Hg is of special concern because of bioaccumulation, notably in birds, fish, and mammals, where biomagnification results in hazardous exposure for top predators and humans. Elemental Hg is transformed to methyl Hg in nature by processes difficult to control; at steady state, an increased concentration of inorganic Hg will result in increased methyl Hg formation (2). Restricting anthropogenic emissions of Hg is thus a priority for counteracting increasing methyl Hg levels in biota (1, 3). Many anthropogenic sources of Hg vapor sources have not been well-studied, where the combustion of coal is a relatively well-known exception. Potentially significant sources of Hg to the environment include small-scale gold mining, chemical production, especially chlorine and polyvinyl chloride (PVC), and the production of metals on an industrial scale. This paper will provide, through a mass balance * Corresponding author phone: +46-(0)18-471 22 65; fax: +46-(0)18-55 11 24; e-mail:
[email protected]. 10.1021/es800495g CCC: $40.75
Published on Web 07/11/2008
2008 American Chemical Society
approach, an estimate of potential air emissions from one subsector within the universe of metals-producing industries: the smelting of Cu, Pb, and Zn sulfide ores. The smelting of sulfide ores is also a potential source of virgin Hg. It is essential to consider this source along with the recorded production of primary Hg from dedicated Hg mines, especially since Hg production from Hg mines has been reduced by an order of magnitude since its peak around 1970 to presently less than 1500 tons annually (4, 5). The production of primary Hg is expected to continue to decrease as an effect of increased awareness about its environmental and health effects. Presently, the demand for Hg is increasingly met by recycled Hg, especially from Hg cells in chloralkali plants that are converting to Hg-free membrane technology. This will result in a large excess of marketable Hg, which emphasizes the need to address the fate of byproduct Hg from sulfide ores. Sulfide ores as well as fossil fuels have comparatively high Hg levels, which is an effect of mercury’s affinity for sulfur and carbon in certain compounds. Most of this Hg is released during combustion or roasting/smelting as C-Hg and S-Hg bonds are broken. Mercury is subsequently either emitted to the atmosphere as a result of the high vapor pressure of Hg or partially recovered by a retort or flue gas cleaning device. Recovered Hg may later be used in products or processes and will result in emissions from dissipative uses in society, from waste handling, and from landfill sites. Large efforts have recently been made to globally quantify historic and present anthropogenic Hg emissions and to predict potential future emissions from Hg used by industry and in consumer products, such as the thousands of tons of Hg used globally in Hg cells by the chlor-alkali industry, batteries, dental fillings, and measurement and control instruments (1, 4, 6–10). These emission inventories are based on measured Hg emissions from large point sources such as chlor-alkali plants, waste incineration plants, metallurgical plants, and coal combustion utilities, or based on estimated regional combustion rates of fossil fuels and various types of waste and their measured or estimated Hg content. Diffuse emissions of Hg to air and water are often missing or are roughly estimated from population densities (11). No complete inventory has been performed concerning global Hg emissions from nonferrous metal production and the potential recovery of byproduct Hg. Existing, partial inventories could largely underestimate Hg emissions from metal production (6, 7, 10), considering that Habashi (12) demonstrated that approximately 20 000 tons of Hg occur in the volume of nonferrous sulfide ores (cinnabar, HgS, excluded) that are processed annually. Lower-grade ores are being extracted nowadays versus those from three decades ago, but the quantities of Hg handled yearly during sulfide ore processing may be the same if considering the affinity of Hg for sulfide rather than for the main metal in the ores. Related Hg emissions could be larger than generally considered. Outside of Scandinavia, information about Hg emissions from nonferrous metal smelters is limited (1, 10), which complicates policy decisions on appropriate actions regarding future Hg strategies (3). The objective of this study is to determine the quantities of Hg handled yearly on a global scale when processing sulfide ores, and to estimate the global emission of Hg from sulfide ore processing. An additional goal is to acquire relevant information on the byproduction of Hg, which can then serve as a basis for companies and governmental authorities in developing strategies for pollution reduction as well as pollution abatement equipment and methodologies. VOL. 42, NO. 16, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5971
FIGURE 1. Schematic design of SO2 abatement for flue gases at a smelter combined with a sulfuric acid plant. Equipment for the removal of Hg can be inserted at the numbers (1 indicates thiocyanate-sulfide; 2, Outokumpu; 3, Boliden-Norzink and Bolchem processes as well as activated carbon) (14).
Emission Abatement Techniques Broadly speaking, metal extraction from sulfide ores can be divided into pyrometallurgical and hydrometallurgical techniques. With pyrometallurgy, high temperatures are used to drive off sulfur from the metal sulfides, leaving a metal oxide residue. Mercury is volatized at these high temperatures. Hydrometallurgical extraction is potentially coupled with the bacterial oxidation of sulfide minerals (“bioleaching”) and is exploited to dissolve sulfide minerals, releasing both sulfur and metals into the aqueous phase. Since there are very limited atmospheric releases associated with direct leaching (13), hydrometallurgical extraction will not be discussed further. Before pyrometallurgical metal extraction, the ore is crushed and milled to a fine sand, after which the ore grains are separated from other minerals, generally in a flotation tank with water and chemicals that are added to promote flotation and the formation of foam when pressing air bubbles through the bath. The product of flotation is a concentrate or dressed ore; concentrates are commercial products which have a global market. The extraction of nonferrous metals from sulfide ores requires roasting, sintering, and/or smelting of the ore at high temperatures (900-1000 °C). At these high temperatures, Hg impurities in sulfide ores are converted to gaseous Hg0. An example can be taken from the extraction of Zn sulfide, where Hg is primarily released to the gaseous phase during roasting, when the Zn sulfide ore is converted to an oxide form (see eqs 1 and 2). In the roasting furnace, oxygen is reacted with sulfides (e.g., sphalerite and ZnS) at a temperature of about 900-1000 °C, producing metal oxides or metal ferrites (calcine, e.g., ZnOFe2O3) and releasing sulfur dioxide (SO2) as well as other volatile compounds (e.g., Hg0) (13). ZnS (s) + 1.5O2 (g) f ZnO (s) + SO2 (g)
(1)
ZnS (s) + 2FeS (s) + 5O2 (g) f ZnOFe2O3 (s) + 3SO2 (g) (2) Gaseous Hg will primarily follow SO2 in the flue gases, although minor amounts of Hg may also adsorb to particulates produced during heating (7), which are captured in flue gas particle filters (e.g., electrostatic precipitators, cyclones, wet scrubbers, or fabric filters) (14). Environmental legislation in many countries requires the removal of SO2 from flue gases prior to emission. Gaseous components such as SO2 are removed using gas scrubbing systems that employ wet and dry scrubbers. Sulfur in SO2, which has economic value as a product, may also be recovered from the gas stream using sulfur capture techniques, where sulfur is recovered as elemental sulfur, liquid SO2, gypsum, or sulfuric acid. For gases with a SO2 concentration in excess of 1% (14), sulfur is often recovered in a sulfuric acid plant (SAP). For the production of high-quality sulfuric acid, impurities such as Hg must be removed, so SAPs will be equipped for Hg removal and in some cases byproduct Hg recovery for resale (see below). In the acid plant, smelter gases are cooled, dried, and passed over a vanadium pentoxide catalyst bed for oxidation and conversion to sulfur trioxide (SO3). In a single-contact plant, the gases are passed through a series of four or more catalyst beds for conversion. The SO3 formed is then absorbed into 98% sulfuric acid, which is subsequently diluted. The 5972
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 16, 2008
single-contact process has generally an efficiency of 99.5%. As in the single-contact process, sulfuric acid is produced. A diagram of the gas cleaning and conversion process in a sulfuric acid plant is depicted in Figure 1. The acid generated during gas cleaning by electrostatic precipitation and scrubbing, but before the SAP (Figure 1), contains up to 50% sulfuric acid, lesser amounts of HCl and HF, and metal and nonmetal impurities including Hg (up to 1900 ppm), Pb (up to 50 ppm), and As (up to 10000 ppm) (14). Since sulfuric acid is a product of commercial value, undesirable impurities must be removed by gas cleaning prior to conversion in the acid plant. Sulfuric acid is typically sold with a product specification of “
