Forging the Anthropogenic Iron Cycle ... - ACS Publications

Jun 14, 2007 - Gabrielle Gaustad , Elsa Olivetti , and Randolph Kirchain ..... Waste Production, and Recycling in the European Union and the World in ...
0 downloads 0 Views 828KB Size
Download PDF
Environ. Sci. Technol. 2007, 41, 5120-5129

Forging the Anthropogenic Iron Cycle TAO WANG,* DANIEL B. MU ¨ LLER, AND T. E. GRAEDEL Center for Industrial Ecology, School of Forestry and Environmental Studies, Yale University, New Haven, Connecticut

Metallurgical iron cycles are characterized for four anthropogenic life stages: production, fabrication and manufacturing, use, and waste management and recycling. This analysis is conducted for year 2000 and at three spatial levels: 68 countries and territories, nine world regions, and the planet. Findings include the following: (1) contemporary iron cycles are basically open and substantially dependent on environmental sources and sinks; (2) Asia leads the world regions in iron production and use; Oceania, Latin America and the Caribbean, Africa, and the Commonwealth of Independent States present a highly production-biased iron cycle; (3) purchased scrap contributes a quarter of the global iron and steel production; (4) iron exiting use is three times less than that entering use; (5) about 45% of global iron entering use is devoted to construction, 24% is devoted to transport equipment, and 20% goes to industrial machinery; (6) with respect to international trade of iron ore, iron and steel products, and scrap, 54 out of the 68 countries are net iron importers, while only 14 are net exporters; (7) global iron discharges in tailings, slag, and landfill approximate onethird of the iron mined. Overall, these results provide a foundation for studies of iron-related resource policy, industrial development, and waste and environmental management.

Introduction Anthropogenic cycles of materials are interconnected with the natural biogeochemical cycles in what Baccini and Brunner defined as the “metabolism of the anthroposphere” (1). For iron, the most extensively and diversely used metal, Klee and Graedel suggested that approximately 90% of total mobilized iron flow on Earth surface is driven by human actions (2). The combustion of fossil fuels and the removal of rock by construction all inevitably mobilize this widespread element from Earth’s crust; nevertheless, the most direct and evident human-induced effect is the metallurgical production and use of iron and steel goods. This metallurgical perturbation to the natural iron cycle is mostly brought by the mineral production and by iron release to the environment during production and consumption. The study of the anthropogenic iron cycle has more than scientific incentives. Iron and steel comprise a raw material market of hundreds of billions dollars per year, second only to oil (3). The extraction and production of iron and steel impose considerable energy and environmental consequences: the iron and steel industry (excluding iron ore mining) is responsible for about 13% of the energy con* Corresponding author phone: 1 (203) 675-8378; fax: 1 (203) 432-5556; e-mail: [email protected]. 5120

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 14, 2007

sumption (4-5) and about 20% of the industrial waste emission (6-8) of manufacturing as a whole. For the study of all these topics, the magnitude of mass is vital. This facilitates the development and application of material flow analysis (MFA)sa systematic assessment of the material flows, stocks, and temporal and spatial complexity of a chosen system, whether anthroposphere, nature, or both (9). An initial MFA model for iron flows connecting the United States steel production, manufacturing, and use processes was pioneered in the early 1970s (10). At the same time, an independent study examined the different sources where iron scrap originated (11). More specific and consistent analyses emerged later with the emphasis on the industrial input-output of iron (12) or on the technological performances of steel production in the United States (13), Germany (14), and the European Union (15). In addition to iron, these models broadly explored production-related raw materials (e.g., limestone), energy (e.g., coke and electricity), water, and byproducts (e.g., slag and CO2). Michaelis and Jackson modeled material and energy flow through the United Kingdom’s iron and steel sector and introduced exergy (available energy) as an analytical tool for the cycle performance (16-17). The U.K. historical iron cycles were reviewed and several future scenarios were laid out. Thus, although there is an extensive history of studies of iron and related materials and energy flows, most efforts have been focused on iron production and recycling. To understand and improve the efficiency of the anthropogenic iron system, the iron use (not only flows, but also stocksin-use that may last for decades) and its secular effect on recycling and resource provisioning need to be more comprehensively investigated. In addition, more effort needs to be paid to the emerging economies, e.g., China and India. To respond to these challenges, this article presents a global perspective on anthropogenic iron, with particular emphasis on the in-use life stage and on developing countries.

Methodology The Stocks and Flows (STAF) project, initiated by the Center for Industrial Ecology at Yale University, developed a generic MFA framework (18-20) in which the anthropogenic metal cycles for a system defined by certain temporal and spatial boundaries are characterized on its four life phases: Production, Fabrication and Manufacturing (F&M), Use, and Waste Management and Recycling (WM&R). Most data are available at a country level and on an annual basissthe fundamental unit of STAF analysis. The schematic STAF framework is adapted for iron in Figure 1a. Disaggregated system boundary diagrams are displayed in Figure 1b and c to differentiate transformation and market processes and to represent the links among technologies and economic activities. Production. The Production stage includes mining and processing of iron ore as well as mill operations in the iron and steel industry. Crude ore is extracted from the lithosphere. The ore of low iron grades cannot be directly shipped and must be beneficiated by removing extra phosphorus, alumina, silica, and other problematic elements (3). The waste generated during beneficiation is mainly tailingssthe slurry made up by gangue particles, flotation agents, and water. Dried tailings of relatively rich iron grades are reprocessed; the residuals are stored in tailings ponds (21). The discharge rates of tailings are discussed in the Supporting Information. Steel production consists of three steps. First, a blast furnace (BF) burns coke and reduces useable iron ore (containing ∼60% iron) to molten pig iron (containing ∼94% 10.1021/es062761t CCC: $37.00

 2007 American Chemical Society Published on Web 06/14/2007

FIGURE 1. (a) Simplified schematic diagram of a technological resource cycle, with successive life stages plotted from left to right. Detailed diagrams define the system boundary for (b) iron Production and (c) iron F&M, Use, and WM&R. The colors of flows indicate the data sources: literature (in brown color), empirical model computation (in blue), mass balance computation (in green), or unavailable (in gray). IGS ) industrial and governmental stock; M ) market; BF ) blast furnace; OHF ) open hearth furnace; BOF ) basic oxygen furnace; EAF ) electric arc furnace; IUS ) in-use stock; OPS ) stock of obsolete products; MSW ) municipal solid waste; C&D ) construction and demolition debris; ELV ) end-of-life vehicles; WEEE ) waste from electrical and electronic equipment. VOL. 41, NO. 14, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5121

iron). The impurities react with limestone to form slag (3, 22). Iron ore powder can also be directly reduced to raw iron through reactions with natural gas. Steel-making, the second step, eliminates the remaining impurities and upgrades raw iron to steel (usually containing >98% iron) (22). Today’s prevailing steel-making technologies are the basic oxygen furnace (BOF) and the electric arc furnace (EAF). BOF and the old-fashioned open hearth furnace (OHF) utilize pig iron as the major material input. EAF, on the other hand, can rely on up to a 100% charge of iron scrap (22). In the third and final step, the molten steel output from the furnaces is rolled and fabricated into desired shapes (e.g., sheets and bars). Another iron product is castings, which are made in foundries by remelting pig iron, often along with scrap. Castings and finished steel mill products leave the production stage for subsequent manufacturing. The ferroalloy production, due to its very small contribution to total iron flows, is not included in the system boundary as shown in Figure 1b. In steel mills and foundries, considerable amounts of edge trimmings and defective products, known together as “home scrap”, are generated and sent directly back to the furnaces (23). Other byproducts include slag, dust, sludge, and mill scale. Some of these byproducts can be recycled within the steel mills or foundries. Others, e.g., blast furnace and steel furnace slag, are partially reused as construction aggregate (23-24). Fabrication and Manufacturing. The F&M process involves industries that transform iron castings and steel mill products into final goods. Seven iron and steel product groups are distinguished and explored in the present work: castings; steel angles, shapes, and sections; steel bars and wire rods; steel plates, sheets, and strips, excluding tin-coated plates; tinplate; steel rails; and steel tubes and pipes. They are used in five end use categories: construction (buildings and infrastructure); consumer durables (electronics, appliances, etc.); industrial machinery (electric power machinery, metalworking machinery, etc.); transport equipment (automobiles, railway vehicles, ships, and aircraft); and others (packaging and other miscellaneous uses). In reality, certain iron and steel products serve specific functions. For example, tinplate is exclusively used as food packaging. Data for product groups therefore harness the estimate of iron amounts entering different end use categories (as detailed in the Supporting Information). The iron content embodied in a country’s import/export of semi-manufactured parts and final products (also known as indirect trade in iron (25)) cannot be overlooked. For this purpose, the present study explored the United Nations Commodity Trade (Comtrade) Database (26). Under the Standard International Trade Classification (SITC) code system with 3- to 5-digit resolution, nearly 220 categories of commodities were found to contain iron. These trade flows were computed and aggregated into the iron cycles. Cuttings and defective products, i.e., industrial (or new, prompt) scrap, arise during manufacturing, typically at rates of 10-15% relative to finished iron castings and steel mill products (27). The industrial scrap is sent to the scrap processors and markets in WM&R, in which most is recovered. Use. The iron flow entering Use is computed according to the mass balance of F&M. The other flows relevant to the Use phase are: changes of stocks-in-use and stocks of obsolete products, trade of used and obsolete products, discards of obsolete products and post-use wastes, and inuse dissipation (mainly corrosion loss). The absence of plausible information precludes us from characterizing the trade of used and obsolete products and the change in stocks of obsolete products in the work so far. For corrosive dissipation, the corrosion rates of carbon steel (28) indicate that in-use corrosion loss is quite small. The iron discards are retired from use in several forms: municipal solid waste 5122

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 14, 2007

(MSW), construction and demolition debris (C&D), end-oflife vehicles (ELV), waste from electrical and electronic equipment (WEEE), and “others” (including obsolete machines, ships, trains, and planes). These flows are computed as the product of the bulk discard streams and their iron concentrations (see details in the Supporting Information). The net addition to in-use stocks is then calculated according to the mass balance of the Use phase (see Figure 1a):

NASU ) Ui - Wi - LU Waste Management & Recycling. Both the iron discards from Use and the industrial scrap from F&M are processed in WM&R. The recovered portion of discards is designated as obsolete (or old) scrap (23, 27). Obsolete scrap together with the recovered industrial scrap is termed purchased (or external) scrap, in contrast to home scrap circulated within foundries or steel mills. What often appears in steel and other industrial statistics is the sum of purchased (obsolete + industrial) and home scrap. Because home scrap and industrial scrap are related to crude and finished steel throughput, their quantities can be estimated. By subtracting the net scrap import (available from the Comtrade Database) and home and industrial scrap, the recovered obsolete scrap is ascertained. The non-recovered portion is assumed to be landfilled or dispersed into the environment to close the balance of WM&R.

Results and Analysis The anthropogenic iron cycle ca. 2000 is characterized at three spatial levels: 68 countries and territories (which cover more than 99% of the reported world steel production and use (29)) are selected, grouped into nine regions, and finally aggregated into a global result. Country-Level Iron Cycles. The magnitude of national level iron flows has great variance. Some of this diversity is shown in Figure 2 for four countries. The Supporting Information includes the cycles of all 68 countries and territories. Figure 2a pictures the contemporary iron cycle for the United States, the third biggest iron producer (behind China and Japan) and the second biggest user (behind China). To support the U.S. need, heavy imports of iron ore, steel mill products, and manufactured goods occur. The large discard rate of in-use stocks enables a substantial reuse of iron and steel scrap in the United States. As a result, secondary resources contribute some 60% of the input to domestic crude steel production. For the biggest iron producer and user, Mainland China, the snapshot in Figure 2b is no longer representative, because iron and steel output has almost tripled as of 2005 (29). However, the qualitative features observed from the 2000 cycle are still true: in-use stocks grow rapidly and the domestic scrap is insufficient to meet the need, so China mainly relies on virgin minerals, which are largely imported. Even more dependence upon foreign minerals is seen in Poland, an important steel producing country in Eastern Europe. The iron ore is exclusively imported, and steel products are exported. During the transition toward a market economy in the 1990s, many old facilities and industrial machines became obsolete. The lack of enough domestic capacity to utilize so much scrap makes Poland a scrap exporter. Venezuela is characterized in Latin America for both its iron mining and its steel production. It exports primary iron products, including iron ore, direct reduced iron, steel ingots and plates, but imports sophisticated final goods. Regional-Level Iron Cycles. The contemporary regionallevel iron cycles, the aggregate of the country-level cycles, are illustrated in Figure 3 for the nine regions.

FIGURE 2. Country-level iron cycles, ca. 2000, for (a) United States, (b) China (mainland), (c) Poland, and (d) Venezuela. The values encased in the dashed ovals associated with the production stage represent “phantom flows” required to close the mass balance of production. All values are in Tg (109 kg) Fe per annum. For the trade of iron ore, iron & steel products, parts & final products, and scrap, net import/export, instead of gross import/export, is indicated.

The Antarctic iron cycle is, not surprisingly, the simplest of the regional cycles, and has the smallest flows. This is a consequence of the continent’s tiny population, and of the international treaties that forbid mining and require scientific stations to recover and ship waste and unnecessary materials off the continent (30). The cycles of the other eight regions can be divided into two groups according to their balance of trade of iron ore, steel mill products, and other iron flows. In Asia, Europe, the Middle East, and North America, more iron is imported than exported. Within each of these four regions, iron flows are comparable across the life stages: for every unit of each region’s iron use, the iron flows related to mining, production, trade, and loss to repositories are all within a modest range, from 0.1 to 2.0 units. The Commonwealth of Independent States (CIS), Africa, Latin America & the Caribbean (LAC), and Oceania, on the other hand, are net iron exporters, and possess exportoriented and highly production-biased cycles. Nearly half of the iron and steel semi-products made in CIS are exported. The export rates of iron ore in Oceania and LAC are even higher, 90% and 70%, respectively.

The anthropogenic intervention in the natural iron movement comprises the extraction of iron ore from the lithosphere and the deposition of iron tailings, slag, and other wastes into the environment. Figure 4 illustrates those flows from the regional-level cycles. Antarctica is free from mining virgin iron and depositing waste. In other regions, extraction rates are 2-10 times greater than the rates of iron discharge, except that in Europe emission exceeds extraction. The leading extractor regions are LAC, Asia (even though Asia has to import a lot of minerals), and Oceania. Asia also leads the world by annually losing 64 Tg (109 kg) Fe to repositories, followed by 47 Tg Fe in North America and 32 Tg Fe in CIS. On a global basis, the iron release into the environment and other repositories consists of 43% as tailings, 40% as landfill and dispersion, 13% as slag and other byproducts, and only 4% as in-use corrosion and dissipation. Production emissions, including tailings and slag, dominate iron loss. In the two principal mining regions, LAC and Oceania, tailings constitute 67% and 91% of the iron loss. Use and WM&R are the two life stages least well informed by common data sources. Great endeavor is devoted to estimate the quantities and composition of flows entering VOL. 41, NO. 14, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5123

Use and discards entering WM&R, for which the regionallevel results are portrayed in Figure 5. Asia leads the nine regions in iron use. Its rapid urbanization and industrialization result in the greatest use rates in construction and in machinery, at 51% and 26% of the total iron flow. North America and Europe have large and service-dominated economies, which require less iron than Asia but far more than other regions. Of the global iron entering use, 45% is estimated to be employed in construction, 24% in transport equipment, 20% in industrial machinery, 7% in household and commercial durables, and 4% in miscellaneous other uses. 5124

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 14, 2007

For all the regions and in almost all the countries, the iron input into Use is greater than the output, so the in-use stocks increase. For an emerging economy like Asia, the surging desire for steel to build more houses and to make more vehicles leads to a rate of iron entering Use that is four times as great as that leaving Use, as shown in Figure 3c. For a mature economy, e.g., Europe or the United States, this rate is much lower. In terms of the iron discards and industrial scrap into WM&R, Asia, Europe, and North America again have flows comparable to one another and much larger than other regions. At the global level, industrial scrap accounts for 27%

FIGURE 3. Regional-level iron cycles, ca. 2000, for (a) Africa, (b) Antarctica, (c) Asia, (d) Commonwealth of Independent States, (e) Europe, (f) Latin America & the Caribbean, (g) Middle East, (h) North America, and (i) Oceania. The values encased in the dashed ovals associated with the production stage represent “phantom flows” required to close the mass balance of production. All values are in Tg Fe per annum. For the trade of iron ore, iron & steel products, parts & final products, and scrap, net import/export, instead of gross import/export, is indicated. of the total flow; and post-use discards account for the rest, subdivided as obsolete machinery and other large transport equipment (23%), C&D (17%), ELV (17%), MSW (12%), and WEEE (4%).In industrialized regions, especially North America and Oceania that are highly dependent on private transport, ELV is the largest source of iron discards. In Asia and CIS, the C&D iron is particularly significant. Closing the Iron Cycle. To better assess the iron cycles, we introduce the following performance factors: import ratio, export ratio, extraction ratio, and loss ratio. Their definitions and the values for the nine regions are given in Table 1. For a country or a region, a completely closed cycle will exhibit zero dependence on environmental and external sources and sinks, i.e.,

(Brazil, Australia, Russia) to leading importers at the rightmost (United States, Japan, China). Final products turn out to be of low mass, so are not very significant in this presentation. The United States is the biggest customer of final goods, largely imported from Japan, China, South Korea, and Germany. This behavior mirrors the shift of manufacturing capacity from traditional powers (e.g., United States) to newly industrialized or industrializing countries after World War II. Raw and primary materials (iron ore and steel products) dominate the international trade, together making up threefourths of the iron trading mass. In contrast to virgin minerals, the trade of scrap is minor owing to the low availability and the high possibility of domestic reuse of this secondary resource.

OR ) TS ) LU ) La) Px,y ) Mx,y ) Wx,y ) 0, or

Global-Level Iron Cycle. We forge our best estimate of the global iron cycle (Figure 7) by aggregating all regional cycles and eliminating global trade “remainders” and “phantom flows” required to close the mass balance of Production (details are available in the Supporting Information). The removal of the “remainders” and the “phantom flows” requires no more than 2% fine-tuning of the iron cycle. Approximately 840 Tg Fe was produced for manufacturing in 2000. 746 Tg Fe was transformed into final products, 257 Tg Fe became discards, and there was a significant increase (481 Tg Fe) in the stocks-in-use. About threequarters of the material requirement was met by crude ore and one-quarter was met by purchased scrap recovered from discards and industrial scrap. As with the country- and regional-level cycles, the contemporary global iron cycle

 ) ξ ) µ ) λ ) 0. It is clear that the greater the sum of the four performance factors, the more the cycle deviates from a closed cycle. For example, with Africa’s annual extraction ratio of 2.63 units, only 1 unit of iron is used by Africans, leaving 1.58 units as a trade surplus and 0.33 units lost (Table 1). Oceania’s and LAC’s cycles are even farther from closure, due to their exceptionally high extraction and export ratios. Because of the significance of the international trade flows in the entire cycle, it is necessary to examine them in more detail. Figure 6 shows the breakdown of the iron trade of 68 countries, ranking from leading exporters at the leftmost

VOL. 41, NO. 14, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5125

TABLE 1. Performance Ratios for Regional-Level Iron Cyclesa value ratio

definition

Africa

AQ

Asia

CIS

Europe

LAC

ME

NA

Oceania

import ratio export ratio extraction ratio loss ratio

 ) (Py + My + Wy)/Ui ξ ) (Px + Mx + Wx)/Ui µ ) OR/Ui λ ) (TS + Lu + La)/Ui +ξ+µ+λ

1.09 2.67 2.63 0.33 6.72

1.00 0.67 1.67

0.96 0.48 0.49 0.21 2.14

0.51 2.29 2.83 0.83 6.46

2.12 1.54 0.12 0.20 3.98

0.45 4.90 6.12 0.58 12.1

1.45 0.54 0.44 0.19 2.62

0.91 0.53 0.53 0.29 2.26

0.96 12.5 14.3 1.57 29.3

a

AQ ) Antarctica; ME ) Middle East; NA ) North America. Refer to Figure 1a for the symbols of the flows.

FIGURE 4. Regional flows of iron extraction from the lithosphere and iron deposition into environmental and other repositories, ca. 2000: (a) iron flows, all values are in Tg Fe per annum; (b) composition of iron deposition into repositories, all values are in percentage of iron mass. is open and dependent upon environmental sources and sinks.

Discussion To our knowledge, this paper is the first attempt to comprehensively and coherently characterize the technological multilevel life cycle of iron. In this analysis, the data quality by and large declines from the left (Production and F&M) to the right life stages (Use and WM&R) and from the top (trade) to the bottom (tailings and landfill) on the cycle diagram. Great effort has been paid to quantifying the downstream processes in order to enhance the precision of the resulting iron cycles. Some gaps have been identified but not well resolved. First, while the work centers on flows, the in-use stocks and stocks of obsolete products deserve better 5126

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 14, 2007

characterization in future studies. Second, more research is needed to accurately quantify the trade of used and obsolete products. A previous study for the United States demonstrates that these gaps are not as small as usually assumed, and may be sufficiently large to reshape the iron cycle (31). Finally, the iron loss to repositories (including tailings, slag, and landfill) deserves more accurate quantification as well. The visionary concept of a closed iron cycle is worth considering, as it depicts a steady state of iron use with extreme efficiency. To achieve such a cycle, the secondary iron resource must be fully recovered and completely substituted for iron ore. Additionally and very importantly, iron stocks-in-use must be stabilized, or, in other words, the flow entering Use must equal that leaving Use. The iron input into the Use phase may either replace worn and obsolete

FIGURE 5. Regional flows of iron entering Use and iron discards and scrap entering WM&R, ca. 2000: (a) iron flows, all values are in Tg Fe per annum; (b) distribution of iron entering Use, all values are in percentage of iron mass; (c) composition of discards and industrial scrap entering WM&R, all values are in percentage of iron mass. products or fulfill newly raised needs, but will not exceed output in a closed cycle. This goal can never be completely achieved, of course, but one can envision a cycle much less open than is now the case. How might a closed cycle be approached? With the implementation of resource management measures (e.g., design for recycling) and the decline of iron ore grades, scrap iron will become increasingly competitive in price compared with iron ore. One might hope to improve the recycling

efficiency of iron from obsolete products and discards from the current 50-80% up to 100%. We have found that iron entering and exiting Use is more likely to be balanced in post-industrialized countries. For example, with the plentiful accumulation of iron in the society, the Unite States appears to have saturated its per capita in-use iron stocks (31). The stocks-in-use will thus be able to approach stability in the long-range future when most countries become developed, and the scrap will be sufficiently available to balance the VOL. 41, NO. 14, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5127

FIGURE 6. International trade of iron ore, iron & steel products, parts & final products, and iron & steel scrap, with increasing total net import from left to right. The country codes on x-axis follow the standard of ISO 3166-1 alpha-2. These codes are used for Internet domain names as well. All values are in Tg Fe per annum.

FIGURE 7. Global-level iron cycle, best estimate for ca. 2000. All values are in Tg Fe per annum. global need for iron. At that point, the steel industry will be able to be reconstructed upon the flexible “mini mills” that utilize scrap steel recycled from local obsolete products, and reliance on distant primary materials can be greatly diminished. A closed anthropogenic iron cycle encompasses more than independence from foreign resources. Freed from iron extraction and loss, the cycle is built upon the full recovery and reuse of iron and steel scrap. Secondary steel production avoids the costly and potentially polluting mining and beneficiation operations, and achieves about 70% energy saving in contrast to primary production (32). Thus, a closed cycle is likely to pose much fewer and smaller environmental impacts than an open one. Finally, we note that, unlike many other metals, iron is abundant and environmentally harmless. Characterizing its cycles is therefore not essential to such issues as absolute scarcity or toxicity. The cycles have many other uses, however. 5128

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 14, 2007

For example, an enhanced resource economic analysis can be generated by adding monetary flows. Similarly, by combining iron flow and energy flow analysis, the intensity of energy consumption and greenhouse gas emissions for each individual phase along the iron life cycle can be determined. Last, but not least, the large contemporary production of specialty steels renders iron cycles the controllers of the cycles of many other toxic or strategic metals (e.g., manganese, nickel, chromium, cobalt, molybdenum, niobium, and vanadium). As a consequence, multilevel iron cycles constitute a foundation upon which much of the entire picture of the anthropogenic use of metals can be built.

Acknowledgments We acknowledge the constructive advice of the STAF steel advisory committee: J.-P. Birat, P. Brunner, J. DeFilippi, M. Ericsson, S. Sibley, and H.-J. Weddige. We also thank J. Johnson and B. Reck for the information related to chromium,

nickel, and other ferrous alloying elements. This research was supported by grant BES-0329470 of the United States National Science Foundation.

Supporting Information Available 68 country-level iron cycles and additional details on methodology. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Baccini, P.; Brunner, P. H. Metabolism of the Anthroposphere; Springer-Verlag: Berlin, Germany, 1991. (2) Klee, R. J.; Graedel, T. E. Elemental cycles: A status report on human or natural dominance. Annu. Rev. Environ. Resour. 2004, 29, 69-107. (3) Kesler, S. E. Mineral Resources, Economics and the Environment; Macmillan College: New York, 1994. (4) International Energy Agency (IEA); Organization for Economic Cooperation and Development (OECD). Energy Balances of OECD Countries; IEA, OECD: Paris, 2002. (5) IEA; OECD. Energy Balances of Non-OECD Countries; IEA, OECD: Paris, 2003. (6) Eurostat Environment and Energy Statistics. Generation, Recovery and Disposal of Non-hazardous Industrial Waste in European Union; http://europa.eu.int/comm/eurostat/; accessed in spring 2005. (7) Statistics Bureau of Japan. Statistics of Environment, Disasters and Accidents; http://www.stat.go.jp/english/index.htm; accessed in spring 2006. (8) National Bureau of Statistics of China. Production, Treatment and Utilization of Industrial Solid Wastes by Sector. In China Statistical Yearbook 2005; http://www.stats.gov.cn/tjsj/ndsj/ 2005/indexeh.htm; accessed in spring 2006. (9) Brunner, P. H.; Rechberger, H. Practical Handbook of Material Flow Analysis; Lewis: Boca Raton, FL, 2004. (10) Garrels, R. M.; Mackenzie, F. T.; Hunt, C. Chemical Cycles and the Global Environment: Assessing Human Influences; William Kaufmann: Los Altos, CA, 1975. (11) Sawyer, J. W., Jr. Automotive Scrap Recycling: Processes, Prices, and Prospects; Johns Hopkins University Press: Baltimore, MD, 1974. (12) Konijn, P.; de Boer, S.; van Dalen, J. Input-output analysis of material flows with application to iron, steel and zinc. Struct. Change Econ. Dyn. 1997, 8, 129-153. (13) Andersen, J. P.; Hyman, B. Energy and material flow models for the US steel industry. Energy 2001, 26, 137-159. (14) Schultmann, F.; Engels, B.; Rentz, O. Flowsheeting-based simulation of recycling concepts in the metal industry. J. Clean Prod. 2004, 12, 737-751. (15) Moll, S.; Acosta, J.; Schu ¨ tz, H. Iron & Steel - A Materials System Analysis; Working paper of European Topic Centre on Resource and Waste Management; European Environmental Agency: Copenhagen, Denmark 2005. (16) Michaelis, P.; Jackson, T. Material and energy flow through the UK iron and steel sector. Part 1: 1954-1994. Resour. Conserv.

Recycl. 2000, 29, 131-156. (17) Michaelis, P.; Jackson, T. Material and energy flow through the UK iron and steel sector. Part 2: 1994-2019. Resour. Conserv. Recycl. 2000, 29, 209-230. (18) Graedel, T. E.; van Beers, D.; Bertram, M.; Fuse, K.; Gordon, R. B.; Gritsinin, A.; Kapur, A.; Klee, R. J.; Lifset, R. J.; Memon, L.; Rechberger, H.; Spatari, S.; Vexler, D. Multilevel cycle of anthropogenic copper. Environ. Sci. Technol. 2004, 38, 12421252. (19) Graedel, T. E.; van Beers, D.; Bertram, M.; Fuse, K.; Gordon, R. B.; Gritsinin, A.; Harper, E. M.; Kapur, A.; Klee, R. J.; Lifset, R. J.; Memon, L.; Spatari, S. The multilevel cycle of anthropogenic zinc. J. Ind. Ecol. 2005, 9 (3), 67-90. (20) Johnson, J.; Jirikowic, J.; Bertram, M.; van Beers, D.; Gordon, R. B.; Henderson, K.; Klee, R. J.; Lanzano, T.; Lifset, R.; Oetjen, L.; Graedel, T. E. Contemporary anthropogenic silver cycle: A multilevel analysis. Environ. Sci. Technol. 2005, 39, 4655-4665. (21) U.S. Environmental Protection Agency (EPA). Technical Resource Document, Extraction and Beneficiation of Ores and Minerals, Volume 3. Iron; EPA 530-R-94-030; U.S. EPA: Washington, DC, 1994. (22) Russell, C. S.; Vaughan, W. J. Steel Production: Processes, Products, and Residuals; Johns Hopkins University Press: Baltimore, MD, 1976. (23) American Iron and Steel Institute (AISI). Steel Industrial Technology Roadmap; AISI: Washington, DC, 2001. (24) United Nations Environment Programme (UNEP); International Iron and Steel Institute (IISI). Steel Industry and the Environment: Technical and Management Issues; UNEP: Paris, France; IISI: Brussels, Belgium, 1997. (25) IISI. Indirect Trade in Steel; IISI: Brussels, Belgium, 1996. (26) United Nations Department of Economic and Social Affairs/ Statistics Division. United Nations Commodity Trade Statistics Database; http://comtrade.un.org/db/; accessed from fall 2005 to spring 2006. (27) IISI. Scrap and the Steel Industry; IISI: Brussels, Belgium, 1997. (28) Leygraf, C.; Graedel, T. E. Atmospheric Corrosion; WileyInterscience: New York, 2000. (29) IISI. Steel Statistical Yearbook; IISI: Brussels, Belgium, various years till 2005. (30) Klee, R. J. Industrial metabolism on ice: A case study of industrial materials flows and environmental management alternatives for Scott Base, New Zealand’s Antarctic research station. J. R. Soc. N. Z. 2001, 31, 393-409. (31) Mu ¨ ller, D. B.; Wang, T.; Duval, B.; Graedel, T. E. Exploring the engine of anthropogenic iron cycles. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 16111-16116. (32) Chapman, P. F.; Roberts, F. Metal Resources and Energy; Butterworth: London, UK, 1983.

Received for review November 20, 2006. Revised manuscript received April 12, 2007. Accepted May 1, 2007. ES062761T

VOL. 41, NO. 14, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5129