Article pubs.acs.org/est
Metal Dissipation and Inefficient Recycling Intensify Climate Forcing Luca Ciacci,*,†,‡ E. M. Harper,† N. T. Nassar,†,§ Barbara K. Reck,† and T. E. Graedel† †
Center for Industrial Ecology, School of Forestry & Environmental Studies, Yale University, 195 Prospect Street, New Haven, Connecticut 06520, United States ‡ Interdepartmental Centre for Industrial Research “Energy & Environment”, University of Bologna, Via Angherà 22, Rimini, Italy § U.S. Geological Survey, Reston, Virginia 20192, United States S Supporting Information *
ABSTRACT: In the metals industry, recycling is commonly included among the most viable options for climate change mitigation, because using secondary (recycled) instead of primary sources in metal production carries both the potential for significant energy savings and for greenhouse gas emissions reduction. Secondary metal production is, however, limited by the relative quantity of scrap available at end-of-life for two reasons: long product lifespans during use delay the availability of the material for reuse and recycling; and end-of-life recycling rates are low, a result of inefficient collection, separation, and processing. For a few metals, additional losses exist in the form of in-use dissipation. The sum of these lost material flows forms the theoretical maximum potential for future efficiency improvements. Based on a dynamic material flow analysis, we have evaluated these factors from an energy perspective for 50 metals and calculated the corresponding greenhouse gas emissions associated with the supply of lost material from primary sources that would otherwise be used to satisfy demand. A use-by-use examination demonstrates the potential emission gains associated with major application sectors. The results show that minimizing in-use dissipation and constraints to metal recycling have the potential to reduce greenhouse gas emissions from the metal industry by about 13−23%, corresponding to 1% of global anthropogenic greenhouse gas emissions.
1. INTRODUCTION Technological progress has provided many benefits for human wellbeing, often in the form of material goods. An unintended consequence of this progress has been the significant increase in the quantity and complexity of materials in everyday use, and in the large greenhouse gas (GHG) emissions associated with them, with the metal industry representing about 8% of the global energy demand1 and 6% of the global anthropogenic GHG emissions.2 End-of-life recycling rates are currently very low for many metals,3 with implications for energy use and resource availability: using secondary (recycled) instead of primary sources in metal production carries both the potential for significant energy savings4 and for mitigating future supply concerns related to these nonrenewable resources. The dynamics of metal supply and demand and product lifespans determine the accumulation of material stocks in-use and future scrap generation at end-of-life. Employing metals in dissipative uses, in which scattering and dispersion in the environment occurs with virtually no chance for recovery (e.g., fertilizers, metal powder used in pyrotechnics), or in ways for which no viable recycling options currently exist (e.g., the deoxidizing aluminum used in steelmaking) reduces the quantity of a metal that could potentially be recycled.5 Further © XXXX American Chemical Society
losses can occur during scrap collection and recovery processes such as when a metal becomes an impurity and undergoes recycling in which its functionality is lost (i.e., nonfunctional recycling).6 The inefficiency of current metal recoveries can be expressed in terms of energy embodied in the metal losses: the sum of these losses significantly reduces the amount of material available to recyclers and also reduces the potential for significant energy savings and GHG emissions abatement. In this study, we estimated the absolute amount of losses during use (in-use dissipation) and after use (constraints to metal recycling) for 50 selected metals and metalloids (referred to as “metals” hereafter) as a function of the contemporary anthropogenic stocks in-use. The sum of these losses forms the theoretical maximum potential for future efficiency improvements and enabled us to calculate the corresponding potential for GHG emissions gains. Received: June 1, 2016 Revised: September 23, 2016 Accepted: September 23, 2016
A
DOI: 10.1021/acs.est.6b02714 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Figure 1. Examples of direct and indirect energy inputs and energy-related greenhouse gas emissions associated with a generic metal’s life cycle. Modified from UNEP (2013).4
2. MATERIALS AND METHODS A material flow analysis was conducted to quantify the global material flows at the various life cycle stages (i.e., ore mining and processing, smelting and refining, fabrication and manufacturing, use, and end-of-life) for each metal up to year 2008. The “use” stage of a metal’s life cycle accounts for its utilization in finished goods that are used by consumers or industry and that accumulate in society as products in use (or in-use stocks) as determined by the lifespans of these finished goods and consumer behaviors. Specifically, given historical material flows into use and lifespans distributions of major enduses for each metal, a “top-down” approach was applied to determine global metal flows out of use as a delayed inflow distributed over time.7 When discarded, metals contained in obsolete products can be collected for functional recycling and return to a material’s life cycle or cannot be recovered (including nonfunctional recycling). Inflows and outflows were calculated for the various life cycle stages, and additional data needed for the analysis such as losses to tailings and slag, fabrication and manufacturing losses, and end-of-life recycling rates were estimated where possible.8−15 Losses during use (i.e., in-use dissipation) were calculated based on work by Ciacci et al. (2015).5 Annual metal outflows from use are utilized to estimate the recovery efficiencies at end-of-life, because they represent the theoretical maximum amount of metal available for recycling in a given year. From annual metal outflows from use we subtracted the quantity that was functionally recycled3 to calculate the total quantity that is not being recovered for reuse (see Table S1−S4 in the Supporting Information). For each metal, the sum of flows that are dissipated during use, not recovered, and not functionally recycled at end-of-life
constitutes the basis used to quantify the respective potentials for climate change mitigation. Cradle-to-gate (i.e., from ore mining through the production of the refined metal or of the most common form employed) gross energy requirement (GER) and global warming potential (GWP) data on a per unit mass basis (see Table S5 in the Supporting Information), were utilized to derive the maximum potential for energy and GHG emissions gains associated with dissipative losses and constraints to metal recycling. GER accounts for the cumulative amount of direct and indirect primary energy inputs required in all stages of a (primary or secondary) metal’s production life cycle. Examples of direct energy inputs include fossil fuel combustion for blasting or crushing virgin ores and concentrates during mining and processing, electricity and heat inputs to smelting and refining in primary metal production, or heat for remelting during metal recycling. Indirect energy inputs refer, for instance, to the use of fuels or electricity for the production of floatation agents and solvents employed in ore mining or for the recycling infrastructure (e.g, construction of scrap sorting facilities). GWP is defined as the ratio of the time-integrated radiative forcing from the instantaneous release of a greenhouse gas relative to that of carbon dioxide, with results displayed in units of carbon dioxide equivalent (CO2e).16 Thus, energy-related GWP quantifies potential global warming effect due to energy inputs required in all stages of a (primary or secondary) metal’s production life cycle: GHG emissions are released, for instance, through direct use of fossil fuels, fugitive emissions from metal processing, or indirectly through the consumption of electricity, the use of infrastructure and other material inputs (Figure 1).17 This analysis addresses 50 metals and 15 principal end-use sectors: agriculture, batteries, building and infrastructure, catalysts, coatings and solder alloys, general metal goods, B
DOI: 10.1021/acs.est.6b02714 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Figure 2. Per-kilogram gross energy requirement (GER) data for metals as a function of absolute losses during and after use in year 2008. Energy data are weighted on the basis of GER data for principal physical forms of elements in common uses (see Supporting Information for more detail). End-of-life (functional) recycling rates were estimated by Graedel et al. (2011).3 The area of the data points is proportional to the 2008 losses as a share of metal production in 2008. Metal loss can exceed 100% in a year in which production is low and losses are high.
electrical and electronic equipment, electrochemical, glass or ceramic manufacturing, machinery and equipment, magnets, phosphors, pigments and chemicals, transportation, and a final category for all other miscellaneous uses. The analysis estimates the GER and GWP assuming that the supply of a certain form of metal lost during or after use is entirely derived from primary sources following the main production routes. The calculation is based on the assumption of 1:1 substitution for primary material and a credit is given for the avoided production of primary metal by recycling. In some circumstances, for instance when a material undergoes multiple cycles of recycling, the recycled material may replace both secondary and primary material.18 However, the crediting of recycling with avoided primary production is an approach widely used in life cycle assessment and agrees with the ISO 14044:2006 guidelines for systems expansion to account for both close loop recycling (i.e., discarded materials are recycled into the same types of products as before) and, of more interest here, “open loop product systems where no changes occur in the inherent properties of the recycled material. In such cases the need for allocation is avoided since the use of secondary material displaces the use of virgin (primary) material”.19 As demonstrated by Frees (2007), such a crediting approach should be also preferred for reasons of price inelasticity of metal scrap supply.20 For each application, the principal form in which metals are employed was determined and the main production routes investigated with the purpose of computing energy requirements and GHG emissions consistently associated with each metal form. Metallic forms are commonly used to produce semifinished and finished goods, as well as to obtain commercial compounds (e.g., metal oxide). Life cycle inventories for the production of one kilogram of refined metal were used as a proxy for common forms that use the metallic form as an intermediate product, if more specific data were not available (see Table S5 in the Supporting Information). This assumption likely underestimates energy inputs and GHG emissions associated with the further processing of metallic forms, but the contribution to the total
impact of a given material production can be considered marginal because ore concentration, smelting, and refining are typically the most energy intensive production stages. 4 Whenever appropriate, life cycle inventories are estimated by modeling direct production from mineral forms. We acknowledge that, over the past few years, technological improvements in mining and refining have often reduced the energy use per kilogram of ore (and, consequently, GHG emissions). Increasing shares of renewable energy sources in electricity production have in many world regions further reduced the GHG emissions associated with electricity consumption.21 At the same time, a decline in ore grades has likely led to greater per-kilogram GER values.22 In this study, we assumed that these changes compensated for each other over time. This is a first approximation and we cannot exclude that major results of this study might partially change, should more historic data on technological evolution become available. It is worth noting that values reported in existing life cycle inventories often refer to averages of some years to decade. Despite this feature limits the representativeness of the results, such databases represent the state of the art, especially, at global level. We decided to refer to year 2008 because GER and GWP values for the 50 target metals were available and, consistently, we quantified absolute material losses during use and after use at the same year of reference. We have since extended this information to 2012, but find very little change over that four year period. We comment that in spite of technology evolution, as exemplified in enhanced natural gas recovery and increased use of renewable energy, the transformations over this time period were not great enough to cause significant revisions in the results. This is largely because the momentum of material use in current technology is very large, and transformations tend to evolve over long time periods. End-of-life recycling rates used to calculate the fraction of metal outflows that is functionally recycled are based on the work in Graedel et al. (2011),3 which has provided the first comprehensive overview on metal recycling efficiency at endof-life. For some metals, end-of-life recycling rates were roughly C
DOI: 10.1021/acs.est.6b02714 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Figure 3. Disaggregation of Figure 2 results by major types of losses during and after use, utilizing the methodology of Ciacci et al. (2015).5 Losses after use are generally most relevant, with many elements having both recycling inefficiencies and currently unrecyclable uses. In a few cases, in-use dissipation significantly reduces the amount of discarded products generated at end-of-life. GER is gross energy requirement.
estimated and have been updated in following studies.8−15 The confidence of our estimates was validated by conducting a sensitivity analysis for the influence of the main parameters on GHG emissions associated with metal losses. For each metal of focus, lower and upper bound values for uncertainty were employed for absolute flows into use, in-use dissipation rates, end-of-life recycling rates, and GWP of principal metal forms (see Table S6 and Figure S1 in Supporting Information).
intensified substantially due to overall increase in demand. Energy losses arising from the inefficiency of current metal recoveries are of particular significance for applications that use elements in their metallic form or as an intermediate material for generating commercial compounds. As an example, cobalt oxides used as catalysts in petroleum refining are obtained by the controlled oxidation of cobalt metal, followed by cooling in a protective atmosphere.23 Similarly, zinc oxide, the most utilized chemical form of zinc, can be obtained by different routes, most of which use pure zinc as a starting material.24 In other cases, the production of compounds bypasses the metal form and follows routes that have mineral forms as the direct starting material, with cadmium oxides and arsenic trioxide being representative examples.25 Figure 2 plots the per-kilogram GER values2 needed to produce the most common forms in which metals are used against the absolute quantity of metal losses that are estimated to have occurred during and after use. The results converge
3. RESULTS AND DISCUSSION Losses are directly related to two factors: the overall use of the metal in question and the sectors of employment. For some metals, overall production has declined and remained relatively constant (e.g., thorium and mercury); for other metals, reductions in material dissipation via the minimization of uses of toxic substances has led to a decrease in losses (e.g., thallium in pesticides, arsenic in pressure treated wood, tin and cadmium in metal coating), but for most metals absolute losses have D
DOI: 10.1021/acs.est.6b02714 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Figure 4. Estimates of greenhouse gas (GHG) emissions embodied due to metal losses in major end-uses, and metal breakdown for 2008 potentials (bar charts) following the grouping of Graedel et al. (2015).8
around a best-fit line shown in the figure. Platinum-group metals and other precious metals have low absolute losses and large per-kilogram GER values. Aluminum is somewhat off the
overall trend due to the high energy requirement for electrolytically extracting the metal from its oxide. Similar considerations apply to gallium, which is mainly obtained as a E
DOI: 10.1021/acs.est.6b02714 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology byproduct of aluminum ores.26 To provide a measure of how losses during and after use compare to primary material supply, the area of the data points in Figure 2 is made proportional to the ratio of the mass loss of a metal to its primary production input. For most metals, avoiding these losses has the potential to reduce annual primary production material input by about 60%. In some cases, such as tantalum, the potential gains are greater still. Figure 3 disaggregates metal losses noted in the results of Figure 2 as due to in-use dissipation, currently unrecyclable uses, or recycling inefficiency based on the analysis in Ciacci et al. (2015)5 and on estimates of end-of-life recycling rates.3,8−15 The potential for reducing annual primary supply by eliminating losses due to in-use dissipation is relevant for only a few elements, including zinc, arsenic, mercury, and the platinum-group metals. Losses due to recycling inefficiencies (including nonfunctional recycling) and to a historical metal’s use in currently unrecyclable applications provide greater potential to reduce global reliance on primary sources for many metals. For example, most rare earth elements and some metals employed in specialized applications (e.g., thallium used in medical imaging) are almost entirely lost because recycling at end-of-life is minimal or nonexistent. In this perspective, because the accumulation of alloying elements from diverse scrap sources is a barrier to recycling, the most promising techniques are those that would enable recyclers to sort scrap by alloy type.27−29 For instance, should scrap sorting by alloy type be widely implemented, the end-of-life recycling rate of aluminum alloys employed in the transportation sector could be substantially increased.30 However, improvements will be difficult to achieve if recycling considerations are not part of the design of materials and goods.31 The results demonstrate that current absolute losses are significant with respect to global production, putting severe limitations to an efficient closure of material cycles. Depending on the form and concentration of metals in a complex product, a waste flow, or loss to an environmental compartment (e.g., soil, water, sediment), the energy (and the cost) required to collect and recover a metal can vary greatly. This is often a major hindrance to recycling.32 However, in a scenario in which ore grades are decreasing and energy required in primary metal processing thereby tends to increase, recovery of metals from material discard flows could become attractive if financially profitable on a local basis. Compared to primary metal production, recycling generally takes place on a smaller industrial scale and considerations of economies of scale are important: metal prices, capital investments, cost of labor, low metal concentrations and small recoverable amounts often compromise the economics of material recovery, thereby limiting an extensive application of advanced recycling technology.33 Losses of platinum-group metals are often target flows because of their market value, but it can be expected that interest may shift toward other metals should conditions of economic feasibility emerge.34 From the perspective of a circular economy,35 metal losses reduce the quantity of material available to secondary supply and require deriving this amount of material from ores to satisfy metal demand. Figure 4 provides information on where the greatest contribution to energy-related GHG emissions associated with metal losses are embodied: such potential for GHG mitigation is computed as a result of per-kilogram GWP of primary in-use forms of the metal2 multiplied by annual amounts of losses in those forms. Lost metals are grouped
according to the methodology described in Graedel et al. (2015),8 and major end-uses are sorted into 15 principal enduse sectors. The bar charts in Figure 4 identify for the 15 sectors the metals to which the respective potential GHG emissions can be assigned in 2008. The variety of elements used in modern products, together with the increase in the complexity of materials assembly are reflected in the estimated number of elements (more than 30), provides a significant potential for climate change mitigation. Because the mitigation challenge cannot be addressed by one sector alone,36 these results suggest that a complementary strategy to prioritize effective measures be implemented in parallel and/or across more than one end-use, with targets for recycling improvement that can be identified by matching the principal end-uses of the metals to the industrial sectors in which they are employed. Metal losses from traditional end-uses such as machinery and equipment, transportation, building and infrastructure, and general metal goods are responsible for the largest absolute quantities of embodied GHG emissions. Common commercial sectors such as pigments and chemicals, coatings, and solder alloys are responsible for a similar potential GHG emission embodied in metal losses as the electrical and electronic equipment sector. This result may be surprising, because the elements contained in pigments, coatings, and chemical applications have relatively low energy-related GWPs on a per unit mass basis. However, they combine fairly large production volumes with little or no recycling. For these sectors, a lack of collection prevents metal recycling.5,33 The remaining sectors displayed in Figure 4 are mostly niche end-uses in which desired properties of specific elements are exploited, such as those of rare earths in phosphors and magnets. These elements are often “critical” due to the absence of substitutes.37 Design for environment (DfE) actions aimed at increasing end-of-life sorting and processing (as mentioned above), take-back schemes for end-of-life products, fixed recycling targets on a single metal basis are often pointed as examples of measures that will likely have the greatest impact for securing access to secondary sources of essential materials38,39 and products in greener energy systems, and for reducing global GHG emissions associated with their production. Redesigning the application to avoid dissipation or increasing material resource efficiency to reduce losses at end-of-life would likely lead to decrease in energy requirementand GHG emissions accordinglyfor secondary materials. Energy consumption may also be decreased by reusing components40 or by light-weighting structural members,41,42 but such approaches are outside the scope of the present work. Overall, from 1989 to 2008, energy-related GHG emissions due to metal losses have increased in all sectors, rising approximately from 250 to 450 Tg CO2e per year, and thus representing about one-fifth of global GHG emissions from the primary metals industry at current levels. Exploiting such a potential for GHG emission gains depends on trade-off between the capacity of eliminating or reducing losses due to in-use dissipation, currently unrecyclable uses, and recycling inefficiencies on one side, and fundamental recycling limits (i.e., no process will ever be 100% efficient, and recycling requires energy and produces emissions) on the other. From one perspective, energy requirements for primary production provide the upper limit to permissible energy use by recyclers;38 ultimately, even if the energy needed for recycling equals that of primary metal production, recycling can still be F
DOI: 10.1021/acs.est.6b02714 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Figure 5. Pareto plots showing absolute amount and cumulative percentage of material losses, gross energy requirement (GER), and greenhouse gas (GHG) emission associated with principal metals. Colors follow the grouping of Graedel et al. (2015).8
justified from the perspective of resource preservation and reduction of other environmental burdens.4 However, because a portion of global metal demand would then be met by recycling of those metals, energy would need to be provided to reprocess the recycled metal. In practice, absolute energy demand and related GHG emissions for metal reprocessing would be almost entirely related to those for iron, steel, and aluminum (Figure 5): given their very high rates of use relative to other metals,43 GHG emission gains associated with lost iron and aluminum would represent more than 70% of the total theoretical savings, with the remaining 48 metals analyzed that would collectively account for only about 25%. Ashby (2009) notes that the embodied energy requirement for recycled steel is about onethird that for virgin steel,44 while this ratio is much smaller for recycled aluminum compared to that of primary metal production (∼6%).4 Thus, assuming that the energy required to process and remelt lost iron and aluminum equals that for the current processing of secondary sources of these metals at end-of-life, the net potential (i.e., as resulting from primary GWP less secondary GWP) for corresponding GHG emissions gains in the metal industry results at 18% (varying at range 13− 23%, as estimated by the sensitivity analysis), representing about 1% of global anthropogenic GHG emissions. If secondary GWPs for the metals displayed in Figure 5 other than iron and aluminum (i.e., zinc, copper, titanium, chromium, lead, silver, and manganese) are also included in the calculation, the potential gains are little reduced. This result is not intended to quantify the exact emissions savings resulting from increased metal recycling, but it rather gives a measure of the potential for GHG emissions gains should in-use dissipation and constraints to metal recycling be minimized. It is worth noting that GER and GWP values for secondary metals reported in literature commonly cover processing and remelting stages, but excludes the energy required for the collection of discarded products and the separation of scrap. Because lost materials are generally dispersed in the environment at diluted concentrations, chemically reacted (e.g., in the case of corrosion), and/or physically bound to other materials in a way that prevents recycling, accounting for the effort to
recover and reprocess those material streams might even exceed the resulting potential benefits. Even in the case that limitations associated with metal collection and preprocessing be reduced (theoretically) to the level needed, secondary metal production can be still very energy intensive due to thermodynamic limits. During metallurgical reprocessing, elements with similar properties tend to behave similarly and, in many cases, the separation of metals that concentrate in the melt alloy mix is very challenging.45 Per se, this would not be a problem if the alloy mix is utilized for creating a new alloy of the same, or a similar, composition and quality. However, when impurities cannot be sufficiently removed and the alloy mix is not functionally recycled−as this occurs frequently for many alloying elements− the addition of primary material to dilute the contamination levels of the melt (also known as “sweetening”) is often the only practical approach. Such “sweetening” carries with it the energy required to produce the added primary metal, which, in turn, adds to the overall energy requirement for secondary metal production. Haupt et al. (2016) have estimated that low quality steel scrap has ∼1.4 times higher energy demand than secondary steel production from high quality internal scrap;46 for aluminum, Cullen and Allwood (2013) have quantified that an energy penalty of up to ∼20% is incurred due to the need for primary metal necessary for blending or diluting the alloy mix.39 There is another important consideration in regard to the degree to which limit recycling can be forced. Despite 100% recycling rate being practically impossible to achieve, recycling can be pushed with the cost of higher energy inputs, determining a possible discrepancy between the greatest realizable energy savings (i.e., “optimum recycling rate”) and the greatest amount of metal recycled (i.e., “maximum recycling rate”).38 For instance, for nickel−cadmium batteries and aluminum packaging, optimum recycling rates based on energy requirements have been quantified at 80−90%.47,48 These considerations are very dependent on the recycling scenario and vary from metal to metal, but it is clear that such limitations would imply further energy requirements and related emissions, thereby eroding some of the theoretical G
DOI: 10.1021/acs.est.6b02714 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
(4) UNEP (United Nations Environment Programme). Environmental Risks and Challenges of Anthropogenic Metals Flows and Cycles, A Report of the Working Group on the Global Metal Flows to the International Resource Panel. van der Voet, E.; Salminen, R.; Eckelman, M.; Mudd, G.; Norgate, T.; Hischier, R., Eds.; UNEP DTIE, Sustainable Consumption and Production Branch: Paris. 2013. (5) Ciacci, L.; Reck, B. K.; Nassar, N. T.; Graedel, T. E. Lost by design. Environ. Sci. Technol. 2015, 49, 9443−9451. (6) Reuter, M.; van Schaik, A. Thermodynamic metrics for measuring the “sustainability” of design for recycling. JOM 2008, 60, 39−46. (7) Elshkaki, A.; van der Voet, E.; Timmermans, V.; Van Holderbeke, M. Dynamic stock modelling: A method for the identification and estimation of future waste streams and emissions based on past production and product stock characteristics. Energy 2005, 30, 1353− 1363. (8) Graedel, T. E.; Harper, E. M.; Nassar, N. T.; Nuss, P.; Reck, B. K. Criticality of metals and metalloids. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 4257−4262. (9) Nassar, N. T.; et al. Criticality of the geological copper family. Environ. Sci. Technol. 2012, 46, 1071−1078. (10) Nuss, P.; Harper, E. M.; Nassar, N. T.; Reck, B. K.; Graedel, T. E. Criticality of iron and its principal alloying elements. Environ. Sci. Technol. 2014, 48, 4171−4177. (11) Nassar, N. T.; Du, X.; Graedel, T. E. Criticality of the rare earth elements. J. Ind. Ecol. 2015, 19, 1044−1054. (12) Panousi, S.; et al. Criticality of seven specialty metals. J. Ind. Ecol. 2016, 20, 837−853. (13) Harper, E. M.; et al. The criticality of four nuclear energy metals. Resour. Conserv. Recy. 2015, 95, 193−201. (14) Harper, E. M.; et al. Criticality of the geological zinc, tin, and lead family. J. Ind. Ecol. 2015, 19 (4), 628−644. (15) Nassar, N. T. Global stocks and flows, losses, and recoveries of platinum-group elements (Order No. 3663642). Available from Dissertations & Theses @ Yale University; ProQuest Dissertations & Theses Global. (1701991638). Retrieved from http://search. proquest.com/docview/1701991638?accountid=15172, 2015. (16) IPCC. Climate Change 2001: The Scientific Basis. In Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change; Houghton, J. T., Ding, Y.; Griggs, D.J.; Noguer, M.; P. J., van der Linden, Dai, X.; Maskell, K.; Johnson, C.A., Eds.; Cambridge University Press: Cambridge, United Kingdom and New York, NY, USA, 2001. (17) Eckelman, M. J.; Ciacci, L.; Kavlak, G.; Nuss, P.; Reck, B. K.; Graedel, T. E. Life cycle carbon benefits of aerospace alloy recycling. J. Cleaner Prod. 2014, 80, 38−45. (18) Ekvall, T. A market-based approach to allocation at open-loop recycling. Resour. Conserv. Recy. 2000, 29, 91−109. (19) ISO. ISO 14044 Environmental Management − Life Cycle Assessment − Requirements and Guidelines, 2006. (20) Frees, N. Crediting aluminium recycling in LCA by demand or by disposal. Int. J. Life Cycle Assess. 2008, 13 (3), 212−218. (21) Ciacci, L.; et al. Historical evolution of greenhouse gas emissions from aluminum production at a country level. J. Cleaner Prod. 2014, 84, 540−549. (22) Northey, S.; Mohr, S.; Mudd, G. M.; Weng, Z.; Giurco, D. Modelling future copper ore grade decline based on a detailed assessment of copper resources and mining. Resour. Conserv. Recy. 2014, 83, 190−201. (23) Donaldson, J. D.; Beyersmann, D. Cobalt and Cobalt Compounds. In Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-VCH Verlag GmbH & Co. KGaA, 2000. (24) Graf, G. G. Zinc. In Ullmann’s Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA, 2000. (25) Anger, G. et al. Chomium Compounds. In Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-VCH Verlag GmbH & Co. KGaA, 2000. (26) Nassar, N. T.; Graedel, T. E.; Harper, E. M. By-product metals are technologically essential but have problematic supply. Science Advances 2015, 1, 1−10.
GHG emissions gains estimated. Although the benefits of recycling are not questioned, the feasibility of converting these theoretical GHG emissions reductions into actual savings requires further research into recycling technologies and policy development. The results of this analysis present the first quantitative assessment of the theoretical recycling prospects for metals used in today’s materials and products, and support intra- and cross sectoral approaches36,49 to deal with dissipation, constraints to metal recycling, and GHG emissions savings. A use-by-use examination has demonstrated which application sectors embody the greatest energy-related GHG emissions due to lost metal flows. Metal losses are a valuable resource in terms of natural ores conservation, embodied energy, and potential GHG emission savings. For many metals, the recovery of losses occurring during and after use would reduce our reliance on primary sources of elements that are scarce in nature, benefiting the climate and securing access to secondary sources needed to satisfy metal demand.50 Economic unfeasibility and thermodynamic limits risk, however, to compromise the achievement of the potential environmental gains associated with the recovery of metal losses. Pretending to recover all these losses might seem rather utopian, but margins for mitigating GHG emissions from the metal industry are substantial if a more extensive deployment of measures to increase the recovery of metal losses are promoted and pursued worldwide.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.6b02714. A detailed description of the data employed in the calculation and of the sensitivity analysis results (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*Phone: +39-0541-434483; fax: +39-0541-434480; e-mail: luca.
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research was funded by the United States National Science Foundation (Award Number: 1336121) and by the “Wealth from Waste Cluster”, a research collaboration between the Australian Commonwealth Scientific and Industrial Research Organisation (CSIRO), University of Technology Sydney (UTS), The University of Queensland, Swinburne University of Technology, Monash University, and Yale University. We gratefully acknowledge the contribution of each partner and the CSIRO Flagship Collaboration Fund. The Wealth from Waste Cluster is part of the Minerals Resources Flagship and is supported by the Manufacturing Flagship.
■
REFERENCES
(1) Electris, C., Raskin, P., Rosen, R., Stutz, J. The century ahead: Four global scenarios, Technical documentation. Tellus Institute. Boston, USA, 2009. (2) Nuss, P.; Eckelman, M. J. Life cycle assessment of metals: a scientific synthesis. PLoS One 2014, 9, e101298. (3) Graedel, T. E.; et al. What do we know about metal recycling rates? J. Ind. Ecol. 2011, 15, 355−366. H
DOI: 10.1021/acs.est.6b02714 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
(49) Baron, R.; Buchner, B.; Ellis, J. Sectoral Approaches and the Carbon Market. OECD-IEA Climate Change Expert Group, Paper No. 2009 (3), 2009. (50) Liu, G.; Bangs, C. E.; Muller, D. B. Stock dynamics and emission pathways of the global aluminium cycle. Nat. Clim. Change 2013, 3, 338−342.
(27) Reuter, M. A.; van Schaik, A.; Ignatenko, O.; de Haan, G. J. Fundamental limits for the recycling of end-of-life vehicles. Miner. Eng. 2006, 19, 433−449. (28) Modaresi, R.; Müller, D. B. The role of automobiles for the future of aluminum recycling. Environ. Sci. Technol. 2012, 46, 8587− 8594. (29) Gaustad, G.; Olivetti, E.; Kirchain, R. Design for recycling. J. Ind. Ecol. 2010, 14, 286−308. (30) Hatayama, H.; Daigo, I.; Matsuno, Y.; Adachi, Y. Evolution of aluminum recycling initiated by the introduction of next-generation vehicles and scrap sorting technology. Resour. Conserv. Recy. 2012, 66, 8−14. (31) Allwood, J. M.; Ashby, M. F.; Gutowski, T. G.; Worrell, E. Material efficiency: A white paper. Resour. Conserv. Recy. 2011, 55, 362−381. (32) Dahmus, J. B.; Gutowski, T. G. What gets recycled: An information theory based model for product recycling. Environ. Sci. Technol. 2007, 41, 7543−7550. (33) Reck, B. K.; Graedel, T. E. Challenges in metal recycling. Science 2012, 337, 690−695. (34) Westerhoff, P.; et al. Characterization, recovery opportunities, and valuation of metals in municipal sludges from U.S. wastewater treatment plants nationwide. Environ. Sci. Technol. 2015, 49, 9479− 9488. (35) Ellen MacArthur Foundation. Towards the circular economy. Economic and Business Rationale for an Accelerated Transition.; Ellen MacArthur Foundation: Cowes, UK, 2013. (36) Barker, T., et al. Mitigation From A Cross-Sectoral Perspective. In Climate Change 2007: Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change; Metz, B.; Davidson, O.R.; Bosch, P.R.; Dave, R.; Meyer, L.A., Eds.; Cambridge University Press: Cambridge, United Kingdom and New York, NY, USA, 2007. (37) Graedel, T. E.; Harper, E. M.; Nassar, N. T.; Reck, B. K. On the materials basis of modern society. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 6295−6300. (38) Norgate, T. E. Metal Recycling: An Assessment Using Life Cycle Energy Consumption As a Sustainability Indicator, CSIRO Minerals Report DMR-2616, 2004. (39) Cullen, J. M.; Allwood, J. M. Mapping the global flow of aluminum: from liquid aluminum to end-use goods. Environ. Sci. Technol. 2013, 47, 3057−3064. (40) Cooper, D. R.; Allwood, J. M. Reusing steel and aluminum components at end of product life. Environ. Sci. Technol. 2012, 46, 10334−10340. (41) Allwood, J. M.; Cullen, J. M.; Milford, R. L. Options for achieving a 50% cut in industrial carbon emissions by 2050. Environ. Sci. Technol. 2010, 44, 1888−1894. (42) Allwood, J. M.; Cullen, J. M. Sustainable Materials With Both Eyes Open; UIT Cambridge Ltd: Cambridge, UK, 2012. (43) Gutowski, T. G.; Sahni, S.; Allwood, J. M.; Ashby, M. F.; Worrell, E. The energy required to produce materials: Constraints on energy-intensity improvements, parameters of demand. Philos. Trans. R. Soc., A 2012, A371, 1−14. (44) Ashby, M. F. Materials and the Environment: Eco-Informed Material Choice; Butterworth-Heinemann: Burlington, MA, 2009. (45) Hiraki, T.; et al. Thermodynamic criteria for the removal of impurities from end-of-life magnesium alloys by evaporation and flux treatment. Sci. Technol. Adv. Mater. 2011, 12, 1−10. (46) Haupt, M.; Vadenbo, C.; Zeltner, C.; Hellweg, S. Influence of Input-Scrap Quality on the Environmental Impact of Secondary Steel Production. J. Ind. Ecol. 2016, DOI: 10.1111/jiec.12439. (47) Quinkertz, R.; Rombach, G.; Liebig, D. A scenario to optimize the energy demand of aluminium production depending on the recycling quota. Resour. Conserv. Recy. 2001, 33, 217−234. (48) Rydh, C.; Karlstrom, M. Life cycle inventory of recycling portable nickel-cadmium batteries. Resour. Conserv. Recy. 2002, 34, 289−309. I
DOI: 10.1021/acs.est.6b02714 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
