Emissions coefficients for production
Emissions coefficients and estimates abound, especially in the "grey literature." Published numbers disagree spectacularly in some instances. Some of the early EPA-sponsored studies (such as NRC, 1977b; GCA, 1973; Davis, 1972, 1980) were admittedly crude, but being readily available - have been widely quoted, especially in surveys such as those carried out by the National Academy of Sciences/ National Research Council in the mid-1970s (NAS, 1980; NRC, 1977a; NRC, 1977c, 1981), and also in the more recent series of symposium volumes edited by Nriagu (Nriagu, 1978, 1980a, 1980b, 1980d). Reasons for disagreement among authors include use of questionable data in some early estimates, aggregation from nonrepresentative samples, failure to distinguish between different process stages (for example in the primary copper industry), failure to distinguish between land-destined and other wastes, and failure to distinguish between gross (uncontrolled) emissions and net emissions after the implementation of controls.
The only truly satisfactory approach for a complex industry is a combination of materials-balance and plant-specific emissions data, as exemplified by Survey of Cadmium Emission Sources (GCA, 1981) in the case of cadmium. In the case of copper, the best source we have uncovered is the second edition of Metallurgy of Copper, published in 1924 (Hofman and Hayward, 1924). It is regrettable that the few such studies tended to focus on only a single pollutant and neglected to estimate the overall effectiveness of controls in effect at the time of the study. This makes it very difficult to assess the average level of emissions control actually in effect at a given time. We have made our own rough estimates, however, as will be discussed later in more detail.
Tables 1 and 2 summarize our assumed emissions coefficients for metallurgical activities and fossil-fuel combustion, respectively. Table 3 shows our assumed particulate emission-control efficiencies since 1880.
Table 1 Uncontrolled emissions from metallurgical operations (ppm)
Table 2 Uncontrolled emissions from fossil-fuel combustion (ppm)
Table 3 Estimated particulate emission control efficiencies over time (percentage of particulates removed, by weight)
Dust- and smoke-control technologies were first applied to some nonferrous smelters and refineries early in this century, partly because of the inherent value of the flue dust (e.g. to recover arsenic and precious metals). Cottrell precipitators were used, in addition to dust chambers, in the copper industry prior to 1914 (Hofman and Hayward, 1924). In 1915, the Balbach lead refinery in Newark achieved 90 per cent particulate recovery using a Cottrell precipitator (Hofman, 1918); by 1933, virtually all copper smelters included Cottrell treaters obtaining better than 90 per cent dust-collection efficiency. Efficiencies cited in Newton and Wilson (1942, table 5) included Garfield, Utah (90 per cent), Anaconda, Montana (90 per cent), Noranda, Quebec (95 per cent), and Cerro de Pasco, Peru (97 per cent). However, dust recovery efficiency from ferrous melting furnaces was much lower because of the low value of the recovered materials.
Table 4 Average annual US metal production (1,000 metric tons)
All data are five-year averages centred on year shown, except:
a. 1880, 1900 estimated at 90 per cent smelter output (remainder = old scrap).
b. Assumes secondary production is 10 per cent of smelter output in 1880 and 1900 (probably too high).
c. 1900 estimated from 1899 US Census data (lead refined from imported bullion).
It must be pointed out that pollutants (e.g. fly ash) removed from waste streams by "end-of-pipe" control technologies, like Cottrell precipitators, are usually disposed of in landfills. Such disposal is not necessarily permanent. Landfills can (and often do) leak. However, we have not considered the problem of toxic landfill leachates in this paper. To this extent, we have underestimated total emissions.
Table 4 gives the average national production by decade for seven categories of metals: steel, blister copper, primary refined copper, secondary refined copper (from old scrap), lead bullion, secondary refined lead, and secondary slab zinc.
Table 5 gives the average annual national combustion of fossil fuels by decade and by category.
The computation of heavy metal emissions from coal and fuel oil combustion now proceeds in a straightforward fashion by using the gross emissions coefficients (tables 1 and 2), assumed control efficiencies (table 3), national average annual metals production data (table 4), and national average fossil fuel consumption data (table 5). The results are given in tables 6 and 7.
Table 5 Average annual US combustion of fossil fuels (million metric tons)
a. Includes cement, steel, rolling mills and other.
b. Includes gas manufacture.
c. Residential and commercial establishments (retail distributors).
d. Lead content of tetra-ethyl lead (TEL) in gasoline, not gasoline consumed (see Ayres et al., 1988, vol. II, chap. 8).
Table 6 Average annual US metallic emissions due to metallurgical operations (metric )
Table 7 Average annual US metallic emissions - fossil fuel combustion (metric tons)
[Ukrainian] [English] [Russian]