Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target

Applied Energy xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target q Koji Tokimatsu a,b,⇑, Henrik Wachtmeister c, Benjamin McLellan d, Simon Davidsson c, Shinsuke Murakami e, Mikael Höök c, Rieko Yasuoka f, Masahiro Nishio b a

Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku, Yokohama, Kanagawa 226-8503, Japan National Institute of Advanced Industrial Science and Technology, 1-2-1, Namiki, Tsukuba, Ibaraki 305-8564, Japan Global Energy Systems, Department of Earth Sciences, Uppsala University, Villavägen 16, SE-751 21 Uppsala, Sweden d Graduate School of Energy Science, Kyoto University, Yoshida-Honmachi Sakyo-Ku, Kyoto 606-8501, Japan e School of Engineering, The University of Tokyo, 7-3-1 Hongo Bunkyo-ku, Tokyo 113-8656, Japan f Systems Research Center, Co. Ltd, KY Bldg., 3-16-7, Toranomon, Minato, Tokyo 105-0001, Japan b c

h i g h l i g h t s
Nexus approach was applied using an energy model to estimate metal requirements.
Two original energy scenarios were developed: ‘‘Coal & Nuclear” and ‘‘Gas & Renewable”.
CCS was expanded in both scenarios, with either nuclear or PV in the two scenarios.
The metal requirement to meet the 2 °C target in the both scenarios was estimated.
Concerns exist that some metals might not meet requirements for PV in ‘‘Gas & Renewable”.

a r t i c l e

i n f o

Article history: Received 15 January 2017 Received in revised form 11 May 2017 Accepted 22 May 2017 Available online xxxx Keywords: Energy-mineral nexus Energy modeling 2 °C target Zero emissions scenario Metal requirement

a b s t r a c t Stringent GHG emission cuts are required for meeting the so-called Paris Agreement. Due to higher metal intensities of renewable energy, such a transition must also include required amounts of metal. This study estimates the metal requirement for various power generation technology mix scenarios by using a cost-minimizing energy model on the global energy-mineral nexus. Two energy and climate scenarios were developed to represent primarily economic efficiency and environmental performance, respectively, under climate policies with net zero emissions satisfying the 2 °C target, and without any constraints (i.e. Business As Usual). Based on the future additions of various power generation technologies, metal requirements and cumulative production were estimated in zero-order and conservative scenarios, to compare with production levels in 2015 and reserves. The results suggest that there may be cause for concern about metal requirement and/or availability in PV, nuclear, and (Plug-in Hybrid) Electric Vehicles in 2100. For PV in the Gas & Ren scenario, most of the metal usage exceeded their production levels and the reserves. It is concluded that mineral availability and production rates should be given greater attention for planning and modeling of sustainable energy systems. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction q This article is based on the 8th International Conference on Applied Energy (ICAE2016), October 8–11, 2016, Beijing, China (Original paper title: ‘‘Global energy-mineral nexus by systems analysis approaches” and Paper No.:328), which is a short proceedings paper in Energy Procedia. It has been substantially modified and extended, and has been subject to the normal peer review and revision process of the journal. This paper is included in the Special Issue of ICAE2016, edited by Prof. J. Yan and Prof. F. Sun. ⇑ Corresponding author at: Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku, Yokohama, Kanagawa 226-8503, Japan. E-mail address: [email protected] (K. Tokimatsu).

1.1. Background The Paris Agreement entered into force on 4 November 2016 by ratification of nations representing over 55% of total global greenhouse gas (GHG) emissions [1]. The Agreement aims to halt the rise in global mean temperature from global warming at well below 2 °C above pre-industrial levels, and to pursue efforts to limit the temperature increase to 1.5 °C. However, the (Intended) Nationally

http://dx.doi.org/10.1016/j.apenergy.2017.05.151 0306-2619/Ó 2017 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

2

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

Determined Contributions ((I)NDC) submitted by the 197 Parties to the agreement are insufficient to meet the target [2]. More stringent GHG emission cuts are required to meet the target, such as zero cumulative emissions over this century, by large-scale deployment of renewable energy, hydrogen energy, and carbon capture and storage (CCS) in all sectors of industry, as well as residential and transportation sectors [3]. In Japan specifically, since the Great East Japan Earthquake, Japanese energy policy strategies have been directed towards seeking more diversified energy options, especially fuel switching to gas, rapid introduction of renewable energy, and pushing towards a hydrogen economy [4]. Despite such stringent climate and energy policy, little consideration has been given to the mineral resources that would be used as materials for various energy technologies. Thus, while a secure supply of energy is typically argued within the context of energy resources within Japanese energy policy, there has been insufficient incorporation of minerals supply security in such policy. The nexus approach requires understanding of not only the relationships of the resources concerned, but also the complex interactions between the natural environment and human society. It is anticipated that such an approach will lead to better understanding of the relationships between the resources by us. When analyzing one resource, some trade-offs with other resources can be revealed. Holistic understanding of these complex systems can assist in resolving such problems. The nexus between energy and mineral resources can be defined as ‘‘all the relations in supply chains between mineral resources and energy across various aspects including economy, technology, policy, society, geology, and nature” [5]. Nexus approaches have become popular recently in a variety of intersecting sectors with relations to sustainability—for example, the energy-biomass (food)-water nexus has been widely examined [6,7]. Other nexus examples can be found among the Sustainable Development Goals formulated by the UN (2015), with intersecting issues like clean and affordable energy for production, poverty alleviation and other uses in society [8]. The energy-mineral nexus has also recently gained attention, especially following the publication of reports on critical metals by the Department of Energy, United States of America (USDOE) [9], and by the European Commission Joint Research Centre (JRC) [10]. In this sense, there is a significant cross-over with the critical materials literature. Many studies (e.g., [11–15]) have addressed scarce (or critical) metals used in particular energy technologies, such as wind power (WP), automobiles (for rare earth elements (REEs) used in magnets and generators), fuel cells (platinum group metals (PGMs)), thinfilm photovoltaics (PV, using elements such as gallium, and indium), lithium ion batteries (lithium, cobalt, etc.), and other uses. Renewable energy technologies are more metal intensive (per unit of output) than current energy sources, and decarbonization is expected to increase demand for many materials

[16]. At the same time, this restructuring of the energy sector will likely imply that mining, manufacturing, and recycling industries will also become increasingly interdependent with the energy sector as the share of renewable energy increases [15,17]. In addition to academic activities, other government bodies have turned their focus to these minerals, for example, the USDOE launched the Critical Material Institute (CMI) [18], an energy innovation hub consisting of national institutes, universities, and private companies. 1.2. Research objectives and originality The primary originality is in the application of an energy model to metal requirements in energy technologies. Studies applying such an approach through the collaboration of communities on metals and mineral resources with energy modelers are scarce, if available at all. In past studies, most have addressed specific mineral elements in technologies by borrowing energy scenarios from authorities (e.g., Strategic Energy Technologies (SET) plan, International Energy Agency (IEA) Energy Technology Perspectives (ETP), World Energy Council (WEC), or World Wildlife Fund (WWF)) [e.g., 13,19–22]. Some have applied empirical estimation models for their projections of future demand. By comparison, the present study applies models that have been developed specifically for this purpose, incorporating resources of energy (herein fuel minerals, fossil plus uranium), minerals (non-fuel minerals used for materials production), biomass, and food; to illustrate future metal requirements. This approach is a highlight of the model, in that it enables the flexible development of alternative energy and climate policy scenarios. The secondary point of originality arises with the scenarios themselves (see Table 1). Two policy scenarios on energy and climate (two times two combinations) have been developed here, for which the details are described in a later section. The considered climate policies are extremely stringent regarding GHG emissions; ‘‘business as usual (BAU)” without any constraints and ‘‘net zero emission (hereafter net ZERO1)” in which cumulative emissions are zero over the time horizon, allows positive emissions over the coming several decades that would be balanced-out by negative emissions in the latter half of the century [3]. Energy policies are ‘‘Coal & Nuclear” and ‘‘Gas & Renewable”, stressing economic efficiency and environmental compatibility, respectively. These are inspired by Japanese energy policy before and after the Fukushima nuclear accident, as well as the shale revolution in the USA. These two policies are simulated only by changing the amount of resources provided in the model, keeping other settings and constraints identical. Within the authors‘ knowledge, no such modeling exercise has been similarly undertaken in which these two contrasting energy policy scenarios have been run simultaneously to test the effect of this specific parameter (i.e., the amount of the resource provided to the model). 1.3. Two scenarios on energy and climate

Table 1 The four energy technology and climate scenarios used in this study. Metal requirements are analyzed for the two ‘‘net ZERO” scenarios. Climate Energy

BAU

Net ZERO

Coal & Nuclear

Coal & Nuclear under BAU

Gas & Renewables

Gas & Renewables under BAU

Coal & Nuclear under net ZERO Gas & Renewables under net ZERO

Note: Common constraints on share of generation types were provided in the all scenarios; sum of bio + oil power, sum of PV + WP + ocean, coal power, and gas power (allowed as baseload operation) was less than (10, 20, 30, and 40)%, respectively.

Two patterns of energy (especially power) scenarios and two climate policy scenarios were set-up. One energy scenario is dominated by gas and renewables (denoted as Gas & Ren), while in the other, coal and nuclear (Coal & Nuc) can be introduced substantially. The computation of changes in these scenarios was executed by assuming cheap gas and uranium, respectively, in each energy scenario. Common constraints on share of generation types were provided in all the four scenarios; sum of bio + oil power, sum of

1 2 °C is much easier to understand for general readers, our scenario (net ZERO) is not sole scenario to 2 °C, meaning that 2 °C may be attainable other GHG emissions paths. Our result shows 2.0 ± 0.3 °C with/without non GHG emissions.

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

PV + WP + ocean, coal power, and gas power (allowed as baseload operation) was less than (10, 20, 30, and 40)%, respectively. The climate policy scenarios are (1) business as usual (BAU) with no emission control on greenhouse gases (GHGs), and (2) net ZERO, whose cumulative emissions are zero over the time horizon using the cumulative emissions of Wigley Richels Edmonds (WRE) [23] 350 ppm constraints over the computational time horizon (from 2010 to 2150). The four scenarios on energy and climate used in this study. Metal requirements are analyzed for the two net ZERO scenarios. The structure of this article is as follows. Section 2 provides a literature review. Section 3 describes the model utilized and policy scenarios on energy and climate change. Sections 4 and 5 present the results and discussion, respectively; Section 6 describes the conclusions and future work.

2. Metal requirement: a short review 2.1. General Future estimates of metal requirements have been increasingly of interest in recent years, in part due to China‘s restriction of Rare Earth Element exports [24], and global growth having fueled significant price rises in various metals [25]. The consideration of such future estimates generally takes into account a number of the following components: 1. Demand – composed of: a. Material intensity in an application or economy b. Change (typically growth) in the application or economy c. Change (typically reduction) in the material intensity per unit output d. Substitution with alternative materials or technologies e. Competition amongst sectors or applications and alternative materials 2. Supply – composed of: a. Primary production – ore production from mines and reserves/resources b. Supply chain – processing of ore to metal c. Recycling – and potential changes in recycling rates d. Non-conventional resources – e.g. tailings re-mining or deep ocean mining 3. Balancing: a. Physical (tonnages) b. Economic (price-wise balancing) c. Trade-structure or re-structure (e.g. through I-O tables) The ‘‘critical minerals” research has been one segment of this literature, that has focused on global [26], national [9,27–30] and a variety of sectoral assessments, many specifically on energy for example [31–33]. On the supply side, there has been significant work on examining the ‘‘peak minerals” concept [34] that is an expansion or analogue of the Hubbert ‘‘peak oil” modeling – for example, on lithium [26], copper [35], iron ore [36], and fossil fuels [37]. For minerals, the ore grade while not the only factor) has significant implications for whether a deposit is considered economically mineable, as well as for the environmental impacts of extraction, and thus the focus on estimates of reserves, resources and production often has reference to grade [38,39]. There is inherent uncertainty in all estimations of reserves of minerals [40], but these errors become significantly large when attempting to estimate the resources – which are not often reported, contain a wider range of lower grade, complex and deeper deposits. Thus there is a significant uncertainty in the potential available material – although it can confidently be assumed that ultimately recoverable resources

3

will surpass known reserves – unless legislation, loss of social license, or dramatic social or environmental calamities were to restrict extraction. When non-conventional resources, such as deep ocean resources, are considered, the potential for increased error is significant, but restrictions on mineral availability may be vastly overstated if some of these estimated resources are able to be extracted in a commercially and environmentally acceptable manner [41]. A further important concern is the level of recycling – particularly for minor and alloying elements – as current rates of recycling are often quite low overall, which often implies an underestimate of future potential availability, particularly under serious supply restrictions see for example, the increase in interest in recycling rare earths particularly consequent to a period of inflated prices [42–44]. Besides recycling, metal resources differ from energy mineral resources, such as fossil fuels, in that many of the metals used in low carbon energy technology are extracted only as by-products of refining processes for base metals, such as zinc and copper [45]. This makes reserves estimates and future availability assessments of such metals more difficult and complex compared to, for example, oil and gas. Also, since increasing demand for and interest in such metals are comparably new, more advanced resource assessment techniques have not yet been applied and only limited evidence exists to base assessments on [40]. Currently, almost all major material requirement studies rely solely on USGS reserve estimates [46]. The USGS data set is in turn based on different international and commercial sources with different estimation methods which introduce additional uncertainty. For many metals only reserves are reported according to the USGS definition ‘‘part of the reserve base which could be economically extracted or produced at the time of determination”. For some metals resource estimates are also available, according to the definition ‘‘material in or on the Earth’s crust in such form and amount that economic extraction of a commodity from the concentration is currently or potentially feasible” [47]. All in all, current metal reserve, and in particular resource, estimates are very uncertain, and it is reasonable to expect that future reserve growth of known deposits and future discovery of currently unknown deposits will increase total reserve estimates. How much is an open and historically recurring question. As an example, see [48–50] for a discussion and different perspectives on future availability of copper. On the demand-side, studies have typically considered scenarios that predict demand on the basis of past demand trends for individual technologies, or on the basis of correlations to economic (e.g. GDP per capita) or demographic (per capita) indicators. Within the context of energy, there has been particular focus on specific technologies: nuclear [32], wind turbines [51], [22], batteries [52], fuel cells [53], [54] photovoltaics [29,55,56]. There have also been a number of studies examining either the targets of regions or countries [19,30,57], or other scenarios for clean energy expansion [58]. Such studies typically draw a timeline out to 2030 or 2050, and in most cases there are expectations that sufficient resources will be available. 2.2. Technology The commissioning of clean energy technologies always requires a range of different materials. What materials are required, and what quantities, varies between different technologies. For some, the materials required vary between different alternative designs, and can also change with time in line with technological developments. A range of different photovoltaic designs exist, although the global photovoltaics (PV) market is currently completely dominated by crystalline silicon (c-Si) technologies. Availability of silicon is

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

4

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

not often suggested to be a potential issue for large scale deployment of photovoltaics, but the use of silver and the high energy requirement of production of solar grade silicon (SoG-Si) have been suggested to be potential limitations for terawatt scale development of c-Si PV [59]. Although current designs utilize significant amounts of silver, it is frequently suggested that this could be substituted for other materials, such as copper [60]. Thin film (TF) PV technologies, especially Copper indium gallium selenide (CIGS) and cadmium telluride (CdTe), have also had some degree of commercial success and contributes with a few percent of the total market, although the market share for TF appears to have been falling in recent years [61]. Kavlak et al. [62] present up to date estimates of current material intensities for In, Ga, and Se in CIGS PV, and Te and Cd in CdTe PV technology, as well as two alternative material intensities for 2030 if improvements would be made. In the future, a wide range of other PV designs could very well be commercialized, that could utilize other materials than the current ones [63]. For wind energy two main technologies exist, commonly referred to as direct-drive (DD) and gearbox turbines [51]. Some DD designs rely on electrical excitation magnets, but others use permanent magnets requiring the use of the rare earth elements neodymium and of dysprosium, While the designs with gearboxes are currently most common, it is possible that DD designs become increasingly common, but it has also been argued that availability of neodymium is not likely to be a ‘‘show stopper” for large scale deployment of wind energy in total, although it could limit specific designs [14]. On the other hand, all types of wind energy utilizes large quantities of ‘‘bulk materials” such as steel, copper, and concrete [17]. The same applies to most energy conversion technologies, including nuclear, geothermal, hydro, coal, oil, gas, hydrogen, biomass, CCS, and transportation and distribution of energy carriers such as electricity and hydrogen.

3. Methodology 3.1. Outline of the overall model The global model applied here consists of three hard-linked models (Fig. 1) composed of a simplified climate model with the following resources: energy (or fuel minerals, fossil plus uranium), minerals (non-fuel minerals used for materials production), and biomass and food. Unlike the World model using a system dynamic techniques in the Limit to growth study, our model applying linear programing is a type of bottom-up technology model like MARKAL however, distinct feature of ours is including the mineral resource balance model that excluded in other techno-economic types. One of our strengths is to discuss on copper requirement comparing with its demand endogenously obtained from the mineral model, while our weakness is excluding end-use technologies (e.g., lighting) that covers in the MARKAL family when compared with similar study [31]. The models provide a consistent structure for supplying the resources to meet exogenous demand scenarios.2 The left side of Fig. 1 indicates resource supply, while the right side shows demand together with end-use products and waste disposal. The upper section illustrates the mineral and material flows; the lower section shows biomass and food flows; and the middle section shows energy flows. These three sections correspond to the three resource models for the balance of materials, biomass and food via land use, and energy systems, respectively. 2 Energy (electricity, heat, and transportation), materials (electricity, machinery, transportation, construction, and civic infrastructure), food (pork and chicken, lamb and beef, and cereals), and wood (lumber and boards, paper, and fuel).

The blue, red, green, and orange lines indicate flows to meet the demands for electricity, heat, and transportation via hydrogen and liquid fuels, respectively; while the solid and dotted lines indicate flows of energy products and their resources, respectively. The black lines show material flows, with solid lines representing mainstream industry; both dashed and dot-dashed lines represent peripheral or recycling industry, and indicate biomass residues and scrap materials, respectively. The linkages are explained in Section 3.3.2. The left side of this figure indicates resource supply, while the right side shows demand together with end-use products and waste disposal. The upper section illustrates the mineral and material flows; the lower section shows biomass and food flows; and the middle section shows energy flows. 3.2. The objective function The modeling approach use here is predicated on perfect foresight, assuming that costs and expansion rates of technologies3 are known and can be taken into account via linear programing optimization. This idealized approach provides consistent, economically efficient future scenarios of technology deployment and resource allocation to meet the climate target. The global model used here, includes 10 regional areas or groups4 (rg) with time horizons between 2010 and 2150 at 10year intervals5 (yr). Hence, the objective function of the overall model is expressed as the discounted sum of the total supply costs (TC) of energy resources (cost of fuel; fossil plus uranium: FC), non-fuel minerals and materials (non-fuel mineral and materials cost: NFC), and biomass and food (land cost: LC), with a discount rate (q) of 2%/yr and 10-year time step (n), as follows:

TC ¼

10n X 14  X 1  ðFC rg;2010þ10n þ NFC rg;2010þ10n 1þq rg n¼0 þ LC rg;2010þ10n Þ:

ð1Þ

3 Technology options included 28 types of power (8 types of fossil fuel (coal, integrated gasification combined cycle (IGCC), oil, and gas, without and with CO2 capture), four types of biomass (co-firing and integrated gasification (IBGCC), without and with CO2 capture), hydrogen, five types of nuclear energy (light water reactor (LWR), fast breeder reactor (FBR), three types of nuclear fusion), and 10 types of renewables (PV, CSP, SPSS, onshore wind, offshore wind, conventional hydropower and pump, small- and medium-scale hydropower, geothermal power, ocean wave and tidal power, and ocean thermal energy conversion (OTEC)); 15 types of liquids, including refined oil, ethanol (bioethanol by biomass residue fermentation, without and with CO2 capture), methanol (coal, gas, and biomass residue), biodiesel, and FT synfuel (biomass liquefaction, coal, natural gas, and heat utilization of nuclear fusion for biomass residue, without and with CO2 capture); 12 types of hydrogen production, including fossil (coal, oil, and gas, without and with CO2 capture), biomass (gasification, without and with CO2 capture), nuclear (high-temperature gas cooling reactor (HTGR) and heat utilization of nuclear fusion for biomass residue, without and with CO2 capture), and renewable (electrolysis by large deployment of PV); eight types of heat, including biomass (biomass pellet heating, biomass heating with CHP (combined heat and power), biomass anaerobic digestion with CHP, and municipal solid waste with CHP), geothermal (conventional deep geothermal with CHP, advanced deep geothermal with CHP, and shallow geothermal heating and cooling), and solar; 11 types of transportation, including passenger car (internal combustion engine (ICE), plug-in hybrid electric vehicle (PHEV), electric vehicle (EV), and fuel cell vehicles (FCV)), bus (ICE and FCV), truck (ICE and FCV), aviation, marine, and rail; eight types of steel production, including blast furnace with converter and electric furnace with directly reduced iron (DRI) for construction steel and mechanical machinery steel, with and without CO2 capture; five types of non-ferrous metal production, including aluminum, copper (dry and leached), lead, and zinc; four types of cement kilns, including wet, dry, advanced dry, and advanced dry with CO2 capture; and three types of cement mills, including Portland cement, blast furnace cement, and Portland fly-ash cement. 4 North America, Western Europe, Japan, Oceania, China, Southeast Asia (including member countries of the Association of Southeast Asian Nations (ASEAN) and India), the Middle East and North Africa, Sub-Saharan Africa, Latin America, and the former Soviet Union and Eastern Europe. 5 2010, 2020, . . ., 2150.

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

5

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

Mineral resources

Energy resources

Biomass & foods

Fig. 1. the overall model structure. The left side of this figure indicates resource supply, while the right side shows demand together with end-use products and waste disposal. The upper section illustrates the mineral and material flows; the lower section shows biomass and food flows; and the middle section shows energy flows.

The linear programming consists of three sets of equations: an objective function, balance equations, and constraint equations. The objective function (C) is generally expressed by the sum of the products of the cost coefficients (cj) and decision variables (xj), determined via optimization (minimization of cost):

min C ¼

X c j  xj :

ð2Þ

j

One of the terms in the objective function, NFC from Eq. (1), is provided as an example in Eq. (3), below. The cost coefficients, decision variables (beginning with X), and suffixes, respectively,

of this Eq. (3) are listed in Table 2. The equations and tables of this part, energy resources (EC) and land resources (LC) are detailed described in our publications elsewhere [3,64,65]. Some description on cost data, constraints, and parameters setting are given here. Cost data are gathered from numerous publications (e.g., Series of the IPCC special reports, the Projected cost of generating electricity) for energy conversion technologies (including performance data), land cost, and fuel & mineral resources. In the energy modeling and the mineral resource balance modeling, energy flow and material flow/stock are determined by the leastcost algorism on technology choice and resource productions, whereas the land-use model flow of biomass and food resources

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

6

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

is distributed by parameter settings, the decision variable of land area, such as crop land, is determined by the algorism. Regarding example of constraints, power generation technologies were allocated to base, middle, and peak demand, within which common constraints on share of generation types were provided in all the four scenarios in this study (see 1.3). Finally on parameter settings, we also assumed a year of introduction and increase in conversion efficiency corresponding to anticipated energy technological improvements. Unlike to energy, biomass and food, mineral resources are not disappeared after their end of use of products. Although all the end of products are assumed ideally corrected (implying no illegal dumping), in the land use model the parameter settings including recovery rate are given all the flows and their intersections, while in the mineral resource balance model determination of all parts of their balances and intersections including final disposal and recovery to use as old scrap are determined by the least-cost algorism. XX NFC rg;yr ¼ CSSTM mp;st;rg  XPRM2mp;st;rg;yr þ LIMCS  XLIMErg;yr mp

st

þ SLGCS  XSLAGrg;yr þ FSHCS  XFLASrg;yr þ MACCS  XSGMArg;yr X XX þ TRCSM mt;rg  XIMMmt;rg;yr þ REFCSMmp;rg  RFEFM mp;ms mt

mp ms

 XMSP mp;ms;rg;yr þ

yr X

X

STLCSirn;rg  XMPIRirn;rg;yr1

irn yr1¼yrnþ1

þ

X

yr X

CKLCSck  XMPCK ck;rg;yr1 þ

X

CMICScm

cm yr1¼yrnþ1

ck yr1¼yrnþ1

 XMPCM cm;rg;yr1 þ

yr X

yr X

CRCS  XMPCN rg;yr1

yr1¼yrnþ1

þ

XX MPRCSmi;md;rg  XMPDmi;md;rg;yr

þ

X ðREFCSMmp¼mc;rg þ RECCSmc Þ  XRCMmc;rg;yr

þ

X ðREFCSMmp¼ms;rg þ SECCSms Þ  XSCM ms;rg;yr

þ

X STCSIN mc  XMSTIN mc;rg;yr

mi md

mc

ms

ms

þ

yr X X

STCSOT mc  XMSTOT mc;rg;yr;yr1

mc yr1¼yr2

þ CRCS  XCO2LMrg;yr þ ðCRCS þ O2CSÞ 

yr X

XCO2CK ck;rg;yr;yr1

yr1¼yrnþ1

þ MDPCS  XMDISP rg;yr ð3Þ

3.3. The mineral resource balance model 3.3.1. General overview of the mineral resource balance model Fig. 2 shows the processes comprising the material flows and stocks in the supply model. The metals used in the manufacturing process of final products are produced via the smelting and refining process from mined minerals and collected both old and new scrap. After the end of life of the final products, they are separated into three categories: out-of-use stock, old scrap recycled into the smelting and refining process after collection and segregation, and remaining scrap for final disposal. Imports from and exports to other regions in the respective processes are also considered. The constraint conditions are formulated to balance the input and output resources throughout the overall system as well as within individual processes. Mined resources are sent to final products via processing with both new and old scrap, which is a part of the discharged from the end of life of the final products, rest goes to out-of-use stock and disposal.

3.3.2. Components of the mineral balance model (see figure in Appendix) In this section, some components of the resources model are provided. Two aggregated patterns of final steel products for steel making were modeled, namely: relatively low grade for construction steel, and high quality for automobiles, electricity, and machinery. For each product, two patterns of steel making processes were assumed; one was a combination of blast furnace/ basic oxygen furnace, the other an electric arc furnace with direct reduced iron (DRI). Both patterns were assumed to include the potential of being equipped with or without carbon capture and storage (CCS) technologies. Hence two products, times two processes, times two CCS alternatives (with/without) equals eight patterns of aggregated technologies that were expressed. The input and output of the technologies are common for iron ore and the products, respectively; however, the only difference between the two steel making processes is the use of scrap (recycled) steel, which was assumed usable for the construction steel processes while unavailable for high quality steel. The inputs and outputs of iron ore and scrap steel were both balanced globally. In the cement production process, our model includes the following three production processes: clinker from cement kilns, cement production, and concrete production. For clinker production from limestone, the following four processes were treated: wet, dry, advanced dry process, and the advanced dry process equipped with CO2 recovery. In the cement production process, the following three products are modeled: Portland cement, blast furnace cement (to which slag from the blast furnace is fed), and fly ash cement (to which coal ash from coal-fired power plants is supplied). After producing cement, concrete production processes are modeled along with recycling of waste concrete. Only global material balancing is considered in regard to the cement products because these are not considered to be distributed across regions. Two patterns of copper production process are assumed: pyrometallurgical and hydrometallurgical (or leaching process). Both of these patterns are expressed in two aggregated processes namely: (1) the mining and dressing to produce either concentrate (via the pyro process) or copper solution (via leaching), and (2) smelting and refining (solvent extraction and electro winning; SXEW) to produce refined copper product. Regarding the global material balance, the following are considered: concentrate (only for the pyro process), refined copper, and copper new/old scrap (only for the pyro process). Inputs and output are common in both processes for copper ore and copper products (for the above four demand sectors); however, the differences between pyro and hydro are the global material balances in the processes of mining & dressing (i.e., concentrate) and of recycling (old/new scrap input). Similar production processes are modeled for the non-ferrous metals (i.e., aluminum, lead, and zinc) with certain differences. The three metals are common across two processes (mining & dressing, smelting & refining) and the global balance of intermediary materials (e.g., concentrate, alumina, slab, refined metal, and new/old scraps). The zinc process, however, has some distinct features that connect it with the steel making process because zinc is chiefly used for steel coating (galvanization). Steel products may be with or without zinc coating. The zinc that is not used for steel coating is expressed in the same process as other minerals; however, those used for galvanizing are fed to steel scrap production for recycling and use in the low-grade iron/steel making process (i.e., construction steel) by electric furnace & DRI, some of which are captured and recovered from dust in the process into the smelting & refining process for zinc. 3.4. Metal requirement and availability The following factors are the determinants of the metal requirement at a point in time (i.e., tonnes of metal per year) and cumu-

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

7

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx Table 2 Cost coefficients, decision variables, and subscripts in Eq. (3). Items

Symbol

Meaning

Unit

Cost coefficients

CKLCSck CMICScm CRPCN CRCS CSSTMmp,st,rg FSHCS LIMCS MACCS MDPCS MPRCSmi,md,rg O2CS REFCSMmp,rg REFCSMmp=mc,rg + RECCSmc REFCSMmp=ms,rg + SECCSms SLGCS STCSINmc STCSOTmc STLCSirn,rg TRCSMmt,rg XCO2CKck,rg,yr,yr1 XCO2LMrg,yr XFLASrg,yr XIMMmt,rg,yr XLIMErg,yr XMDISPrg,yr XMPCKck,rg,yr1 XMPCMcm,rg,yr1 XMPCNrg,yr1 XMPDmi,md,rg,yr XMPIRirn,rg,yr1 XMSPmp,ms,rg,yr XMSTINmc,rg,yr XMSTOTmc,rg,yr,yr1 XPRM2mp,st,rg,yr XRCMmc,rg,yr XSCMms,rg,yr XSGMArg,yr XSLAGrg,yr ck cm irn mc

Investment cost for cement kiln Investment cost for cement mill Investment cost of concrete finishing Costs of CO2 capture in cement kiln Costs of mining, beneficiation, and leaching Production cost of fly ash Production cost of limestone Production cost of crushed rock Costs of CO2 capture in non-ferrous minerals Production cost of final products Costs of oxygen production facility Smelting, refining, and leaching (SX-EW) costs Cost of Smelting, refining, and reaching (SxEw) and old scrap collection and segregation Cost of Smelting, refining, and reaching (SxEw) and new scrap collection and segregation Production cost of slag from blast furnace Cost of receiving scrap stock Cost of shipping scrap stock Cost of iron and steel production Transportation costs CO2 emissions from fuel combustion (cement kiln) CO2 emissions from limestone (cement kiln) Supply amount of fly ash Import amounts of all kinds of resources in this model Supply amount of limestone CO2 emissions from fuel combustion by processes other than iron- and steel-making Capacity of cement kiln facility Capacity of cement mill facility Capacity of concrete finishing facility Production amount of final products Capacity of iron- and steel-making facility Supply amount of mineral ore Amount of inbound scrap stock Amount of outbound scrap stock Production amount of mineral ore Amount of scrap input in refining process Amount of new scrap input Supply amounts of natural sand, gravel, and crushed rock Supply amount of blast furnace slag Cement kiln Cement mill Iron manufacturing technology Scrap (copper, lead, zinc, alumina, automotive steel, electric machinery steel, construction steel, concrete) Industry (electric machinery, automotive, construction and civil engineering, telecommunication and electric power generation, etc.) Final product (copper, lead, zinc, alumina, steel, concrete) Mineral ore (copper, lead, zinc, alumina, iron ore) Mineral ore (copper, lead, zinc, bauxite, iron ore) Base metal and steel (copper, lead, zinc, alumina, steel) Mineral ore, base metal and steel, product, scrap Life time of facilities for iron- and steel-making and cement production the 10 regional allocations Excavation step the time step

US$/kg US$/kg US$/kg US$/tC US$/kg US$/kg US$/kg US$/kg US$/kg US$/kg US$/tC US$/kg US$/kg US$/kg US$/kg US$/kg US$/kg US$/kg US$/kg Gton C Gton C Mton Mton Mton Gton C Mton Mton Mton Mton Mton Mton Mton Mton Mton Mton Mton Mton Mton

Decision variables

Subscripts

md mi mp mr ms mt n rg st yr, yr1

Fig. 2. The overall structure of the mineral resource balance model. Mined resources are sent to final products via processing with both new and old scrap, which is a part of the discharged from the end of life of the final products, rest goes to out-of-use stock and disposal.

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

8

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

lative metal production over the time horizon of computation (i.e., tonnes of metal): i) additional (newly-built) capacity of energy technologies, ii) intensity of use of metals in the technologies and the rate of reduction or improvement, iii) share of the technologies in the same category, iv) lifetime (or service years) of the technology, and v) end-of-life recycling rate. Table 3 summarizes these determinants. The intensity of use data are sourced from various references [11,17,19,62,66–68] and range widely. From this, the maximum number was picked for this study to provide a conservative assessment. No reduction rate (meaning no progress in reducing the amount of metals used per unit output) is assumed for their intensity of use. Critical appraisal or reviews of these data will require further examination in future, and was not considered to be part of the scope of this study. The requirement is calculated by multiplying additional (newly-built) capacity of energy technologies with intensity of use of metals in the technologies. The share of technology should be taken into consideration. For example, our model does not explicitly identify types of PV, such as crystal silicon (c-Si), amorphous silicon (a-Si), copper-indium-gallium-selenium (CIGS), and cadmium-tellurium (CdTe). In this study, two share patterns are investigated; one is 100% c-Si, the other is 50/50% of CIGS/CdTe over the time horizon. In the calculation of cumulative production, the lifetime of the additional capacity and recycling rate should be considered so that cumulative production from unmined resources changes due to the contribution from recycling. In this calculation, after the lifetime of the plant, all the metals are discharged to discard (i.e., zero recovery rate).

4. Results 4.1. Scenarios on the global carbon balance and power mix structure for energy and climate policies Fig. 3 illustrates global carbon balance and global mean temperature rise under basically the Gas & Ren scenario with Coal & Nuc, by using lines (dashed for BAU, solid for net ZERO, red for carbon emission, blue for temperature rise) and by bar graphs (positive for emissions, negative for deployment of CCS in various sectors). BAU line in Coal & Nuc (red, dashed with dots) in temperature rise is somewhat higher than that of Gas & Ren, apparently shows the latter strategy goes greener pathways. Because of this reason, the amount of net carbon emissions as well as CCS is somewhat different, however, the temperature profile is almost identical in the two scenarios. Fig. 4 shows the basis of the power mix structure in the electricity produced (in exajoules, where 1 EJ = 1018 joules), for the four combinations of energy and climate policy. The results indicate the following three common features for both scenarios, BAU and net ZERO: 1) total power supply and 2) portion of renewable energy supply (highlighted by parenthesis) is almost the same, and 3) all fossil fuels are equipped with CCS in net ZERO. One significant difference is that PV, biomass, and geothermal are promoted in the Gas & Ren scenario compared with Coal & Nuc. These results can be interpreted as follows. As described, the two energy policies are changed only in the amount of resources (uranium and gas) by keeping all the other parameters identical, while the two climate policies just consider constraints on (cumu-

Table 3 data settings for metal requirement applied in this study. Technology

Metal

Intensity of use of metals for installed capacity [t/GW]

Lifetime

Reference

c-Si PV

Si Silver Indium Gallium Selenium Cadmium Tellurium Copper Iron Neodymium Dysprosium Lithium Cobalt Nickel Manganese Lithium Cobalt Nickel Manganese Hafnium Indium Silver Molybdenum Copper Copper Copper Copper Copper Copper Copper Copper Copper Vanadium Niobium Nickel Copper

6630 36 28 9 161 138 156 2000 140,000 186 33 12.7 (kg/unit) 8.8 (kg/unit) 46.5 (kg/unit) 91.5 (kg/unit) 5.1 (kg/unit) 3.5 (kg/unit) 18.6 (kg/unit) 36.6 (kg/unit) 0.48 1.6 8.3 70.8 2500 1300 3050 3160 890 1100 1110 1110 1200 100 100 1145 10,000

30

Kavlak et al. [62] ITRPV [69] Kavlak et al. [62] Kavlak et al. [62] Kavlak et al. [62] Kavlak et al. [62] Kavlak et al. [62] Kleijn and van der Voet [14] Kleijn and van der Voet [14] USDOE [9] Hashimoto and Murakami [70] USDOE [9] USDOE [9] USDOE [9] USDOE [9] USDOE [9] USDOE [9] USDOE [9] USDOE [9] Moss et al. [19] Moss et al. [19] Moss et al. [19] Moss et al. [19] Hernández [68] Hernández [68] Lea [67] Lea [67] Lea [67] Lea [67] Lea [67] Lea [67] Lea [67] Moss et al. [19] Moss et al. [19] Moss et al. [19] Hernández [68]

CIGS PV

CdTe PV Wind

EV

PHEV

Nuclear

Geothermal Hydro Coal Oil Gas Hydrogen Biomass CCS

Trans. & Dist.

20

30

50 30

50

Note: Capacity factors are given as follows; 85% for power generation by fossil fuels, hydrogen, biomass; 75% for nuclear; 70% for Geothermal; 50% for wind; 45% for hydro; and from 17% in 2010 to 40% after 2050 for PV.

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

9

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

Fig. 3. global carbon balance and global mean temperature rise. The global mean temperature rise is almost identical under the two energy scenarios while BAU emission lines are somewhat different because of the difference of the energy policies. This result excluding non CO2 GHGs while the temperature is 2.3 °C when included.

NUC

Power Generation [EJ/yr] 200 400 600 800

600 400

Ocean Wind(offshore)

200

Power Generation [EJ/yr]

800

(a)Coal&Nuc Scenario (BAU) (b)Coal&Nuc Scenario (netZERO)

Ocean Wind(offshore)

NUC

IGCC 0

0

COAL 2020 2040 2060 2080 2100

2020 2040 2060 2080 2100

Power Generation [EJ/yr] 200 400 600 800

600 400 200

GAS

Ocean Wind(offshore) PV Biomass geotherm

GAS+CCS

0

0

Power Generation [EJ/yr]

800

(c) Gas&Ren Scenario (BAU) (d) Gas&Ren Scenario (net ZERO)

Ocean Wind(offshore) PV Biomass geotherm

2020 2040 2060 2080 2100

CHP electricity Ocean (OTEC) Ocean (wave, tidal) Wind (offshore) Wind (onshore) Solar (SPS) Solar (CSP) Solar (PV) Biomass IGCC with CCS Biomass IGCC Biomass with CCS Biomass Geothermal Hydropower (small) Hydropower (large) FBR LWR Hydrogen Methanol Gas with CCS Gas Oil with CCS Oil Coal IGCC with CCS Coal IGCC Coal with CCS Coal

CHP electricity Ocean (OTEC) Ocean (wave, tidal) Wind (offshore) Wind (onshore) Solar (SPS) Solar (CSP) Solar (PV) Biomass IGCC with CCS Biomass IGCC Biomass with CCS Biomass Geothermal Hydropower (small) Hydropower (large) FBR LWR Hydrogen Methanol Gas with CCS Gas Oil with CCS Oil Coal IGCC with CCS Coal IGCC Coal with CCS Coal

2020 2040 2060 2080 2100

Fig. 4. global power supply structure. Either Coal & Nuc or Ren & Gas is obtained by only changing one data for resource supply cost (i.e., uranium, gas). Net ZERO shift power mixture utilizing CCS equipped fossil fired, expanding nuclear (in Coal & Nuc) and renewable (in Ren & Gas).

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

10

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

lative) emissions. In order to find computational solutions (optimal – economically efficient) for all four combinations of the policies, the operation windows are made as wide as possible (conventional, somewhat conservative, but not extreme and radical, e.g., 100% renewable). The resulting solutions seem to be found at the upper bounds. In other words, our exercise as conducted, produced cases at the upper limit, within the supplied constraints. Either Coal & Nuc or Ren & Gas is obtained by only changing one data for resource supply cost (i.e., uranium, gas). Net ZERO shift power mixture utilizing CCS equipped fossil fired, expanding nuclear (in Coal & Nuc) and renewable (in Ren & Gas). Fig. 5 illustrates additional (new) capacity of the power mix in Tera Watt per year (TW/yr), which is converted from Fig. 4 by dividing each plant capacity factor by 8760 h annually. In the growth of total amount of new capacity, the four profiles are similar, while their compositions have some commonalities and some differences. They have in common: 1) being equipped with CCS

instead of without CCS in net ZERO compared with BAU, while 2) expanding nuclear and renewables. The total capacity almost doubled when compared with past experience between 2010 and 2030, and between 2050 and 2100, respectively. 4.2. Metal requirement and availability 4.2.1. Technology type Assessment was undertaken with regards to metal requirement and availability. Annual metal requirement in 2100 is compared with resource production in 2015. Some (Nd, Dy, La, and Ce) are projection by the DOE report [9] from those in 2010 level and others [70], rest are sourced from the USGS report [47]. In terms of availability, cumulative production over the time horizon is compared with reserves of the metal, most sourced from [47] other than some (In, Ga, Cd) referred from [14]. These comparisons are made by technology type and by resource type. Among the com-

Fig. 5. additional capacity increase in BAU and net ZERO and difference between them (Diff.) in 2030, 2050, 2100, and historical increase (1990, 2010) in net capacity. The figure in capacity basis is derived from the former figure in produced electricity basis.

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

11

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

reserve (3 kt). The selenium requirement is 0.7 and 15 kt/yr in 2100 for Coal & Nuc and Gas & Ren respectively. This is 20%, and more than 5-times larger than, current production (3 kt/yr). Cumulative production reaches 280 and 2,200 kt, well exceeding the reserve (120 kt). The tellurium requirement is 0.7 and 14 kt/yr in 2100 for Coal & Nuc and Gas & Ren, respectively. This is identical to, and more than 20 times larger than current production (0.7 kt/yr). Cumulative production reaches 270 and 2,130 kt, well exceeding the reserve (25 kt). In this paragraph, the latter group categorized as (2) above in Fig. 6 is described. Silver in the PV requirement is well below the current production level (0.3 and 6.5 vs 27.3 kt/yr for Coal & Nuc, Gas & Ren, and current, respectively) while cumulative production in Gas & Ren somewhat exceeds the reserve (123, 972, and 570 kt). The gallium requirement is well below what is needed in Coal & Nuc but exceeds what is needed in Gas & Ren, compared with the current production level (0.04, 0.8 vs 0.4 kt/yr, respectively) while cumulative production is more or less than reserves (0.02,

parisons, concerns are identified in annual metal requirement and availability (either one or both) for PV, nuclear, and (PH)EV. The results are presented, first by technology type; then by commonly used metals of the technology types (i.e., silver, indium, copper). Fig. 6 displays various metals required for PV, comparing current production in 2015, annual metal requirement in 2100, cumulative production, and reserves. Metals in PV can be placed in two categories: (1) both requirement and cumulative production exceed current production levels and reserves; (2) either requirement or cumulative production in either of energy policy scenarios is exceeded. Indium, selenium, and tellurium are in the former group, while silver, gallium, and cadmium are in the latter. In this paragraph, the former group categorized as (1) above in Fig. 6 is described. The requirement for indium in PV is 0.13 and 2.55 kt/yr in 2100 for Coal & Nuc and Gas & Ren respectively. This is less than 20%, and more than 3-times larger than current production of indium (0.76 kt/yr), respectively. Cumulative production is projected to reach 48 and 382 kt, well exceeding the known

100000

Reserves COAL&NUC GAS&REN

2015 Producon COAL&NUC GAS&REN

1000 100 10 1 0.1

2100 metal req.

Copper

Tellurium

Cadmium

Selenium

Gallium

Indium

Silver

Silicon

Copper

Tellurium

Cadmium

Selenium

Gallium

Indium

Silver

0.01 Silicon

metal req., prod. in 2015 [kton/yr] cum. metal prod., res. [kton]

10000

Cum. Metal Prod.

Fig. 6. Annual metal requirement, annual metal production in 2015, cumulative production, reserves in PV. Metals in PV can be placed in two categories: (1) both requirement and cumulative production exceed current production levels and reserves; (2) either requirement or cumulative production in either of energy policy scenarios is exceeded.

10000 1000

2015 Producon

Reserves

COAL&NUC GAS&REN

COAL&NUC GAS&REN

100 10 1 0.1

2100 metal req.

Iron

Copper

Molybdenum

Silver

Indium

Hafnium

Iron

Copper

Molybdenum

Silver

Indium

0.01

Hafnium

metal req., prod. in 2015 [kton/yr] cum. metal prod., res. [kton]

100000

Cum. Metal Prod.

Fig. 7. Annual metal requirement, annual metal production in 2015, cumulative production, reserves in nuclear.

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

12

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

0.12, and 0.11 kt). Only cumulative production of cadmium in Gas & Ren well exceeds its reserve (1,900 vs 600 kt), while that in Coal & Nuc (0.2 kt/yr), and requirement and current production (0.6, 13, and 24 kt/yr) are well below the known reserve. The requirement for silicon requirement in PV is (60 and 1,200 kt/yr) in 2100 for Coal & Nuc and Gas & Ren respectively. This is far less than current production (8,100 kt/yr). Cumulative production is (23,000 and 180,000) kt, which cannot be compared in number because of abundancy of silicon reserves. Fig. 7 shows results of nuclear. The indium requirement and cumulative production in nuclear reaches 0.2 kt/yr and 26 kt, respectively, which is less than 30% of current production, but 10-times higher than the known reserve. The requirements for silver, molybdenum, and copper are well below current production, and cumulative production also well below the reserve. Hafnium cannot be assessed because of the lack in current production level and reserve. Fig. 8 indicates results of (PH)EV. Number of vehicles are estimated based on IEA-MoMo [71] to extrapolate over the time

4.2.2. Commonly used metals Silver, indium, and copper are commonly used metals used in nuclear, PV, and other power technologies. The requirement for both silver and indium for nuclear are added to those in PV, as illustrated in Fig. 9. Even though some was added to the Coal & Nuc scenario, the Gas & Ren scenario showed fivefold more in metal requirement in 2100 and cumulative production. The copper requirement and cumulative requirement in both energy scenarios are very small, both sources of the intensity of use, compared with current production level, are shown in Fig. 10. However, the requirement for copper in cables and wires (for transmission lines, distribution networks) are comparatively large. In this analysis, intensity of copper use in cables and wires

Reserves COAL&NUC GAS&REN

2015 Producon COAL&NUC GAS&REN

1,000,000 100,000 10,000 1,000 100

2100 metal req.

Manganese

Nickel

Cobalt

Lithium

Manganese

Nickel

Cobalt

10

Lithium

metal req., prod. in 2015 [kton/yr] cum. metal prod., res. [kton]

10,000,000

horizon. Because of this reason the number is identical to both scenarios. All the metals to be used in batteries for (PH)EV and their cumulative production requires well exceeding the current production and the reserve.

Cum. Metal Prod.

Fig. 8. Annual metal requirement, annual metal production in 2015, cumulative production, reserves in (PH)EV. Note that no difference in levels between the two scenarios.

metal req., prod. in 2015 [kton/yr] cum. metal prod., res. [kton]

10000 1000 100

Nuc in COAL&NUC, Silver PV in COAL&NUC, Silver PV in GAS&REN, Silver Nuc in COAL&NUC, Indium PV in COAL&NUC, Indium PV in GAS&REN, Indium

Resources

2015 Producon 10 1 0.1 Silver

Indium 2100 metal req.

Silver

Indium

Cum. Metal Prod.

Fig. 9. Silver and Indium: Annual metal requirement, annual metal production in 2015, cumulative production, reserves. The Gas & Ren scenario showed fivefold more than Coal & Nuc scenario.

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

13

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

Reserves COAL&NUC GAS&REN

1000 100 2015 Producon 10

COAL&NUC GAS&REN

1

2100 metal req.

Trans.&Dist.

Oth. P. Gen.

ICA P.Gen

Trans.&Dist.

Oth. P. Gen.

0.1

ICA P.Gen

metal req., prod. in 2015 [kton/yr] cum. metal prod., res. [kton]

10000

Cum. Metal Prod.

Fig. 10. Copper: Annual metal requirement, annual metal production in 2015, cumulative production, reserves. The two scenarios show a small difference, caveating indifference of intensity of copper use in cables and wires.

is not distinguished between the two energy policies, which must be listed as one of future research agendas, though Gas & Ren (distributed renewables) would be anticipated to require more copper than Coal & Nuc (centralized) does.

5. Discussion 5.1. Scenarios on power mix As described in the introduction, the power mix structures utilized here for the two energy policies are one aspect of the originality in this article. An extensive body of publications on energy scenarios can be divided into two categories: one shows representative energy perspectives with modifications to some degree by energy policy or deployment of technologies [72–74], the other proposes pathways such as 100% renewables [75–77]. These have in common, support by interest groups, private companies, governmental agencies, and non-governmental organizations (NGO) with some exceptions. It is important to stress that, within the authors‘ knowledge, no previous study provides simultaneously, both policy directions (i.e., Coal & Nuclear, Gas & Renewable); existing studies typically treat only one direction as described above. This is the key point of this modeling exercise. It has been attempted here, to illustrate both directions of energy policy by minimizing the number of changed parameters while maximizing all others. The results presented are almost the sole solution after numerous trials. The derived power mix scenarios were compared against those in Section 5 of the IPCC-AR5 report [78]; however, no suitable analysis was possible that had the same basis (i.e., demographic and economic scenarios, climate policy intervention). While no comparison is available on the same basis, the ranges are able to be compared. For 2050, the total power supply in the net ZERO scenario is comparable for Coal & Nuc and Gas & Ren (375 and 350 EJ/yr). Of this, coal w/CCS (57), gas w/CCS (90 in Gas & Ren), nuclear (120 in Coal & Nuc), biomass w/CCS (0 and 7 in Coal & Nuc and Gas & Ren), solar (10 and 30 in Coal & Nuc and Gas & Ren), wind (25 in both), geothermal (6 and 16 in Coal & Nuc and Gas & Ren), and hydropower 7 in both), respectively. When compared with the IPCC-AR5 report, the levels with ranges assessed Fig. 7.11 in Chapter 7) [79], all except coal w/CCS (>60), gas w/CCS (max 40), and nuclear (>60), were within the ranges in our report.

5.2. Metal requirement and availability The additional capacity of PV in Gas & Ren and that of nuclear in Coal & Nuc are at somewhat similar levels. The metal requirements and cumulative production of commonly used metals in the technologies such as silver and indium, are larger in Gas & Ren than in Coal & Nuc. These results suggested that in some cases, the PV metal requirement and cumulative production may exceed current day production levels and reserves. Although WP (hidden in result) is only assessed for bulk metals and scarce metals such as dysprosium and neodymium for permanent magnets (PM) and electromagnets (EM), dependency on WP might be better than PV for promoting renewable energy in terms of the scarce metal requirement. Copper requirements are somewhat similar in both energy scenarios. The more prominent issue with copper is the requirement for cables and wires for transmission lines, which is higher than the total copper requirement for various generation technologies. This was estimated as 6.7 Mt/yr in 2100 (rather higher than by the demand model (3.8 Mton-Cu/yr in 2100) from the mineral demand and supply model). Although our analysis does not distinguish between the energy policies regarding intensity of copper use, there are some disadvantages for Gas & Ren scenario. One of the biggest differences between BAU and net ZERO in the two energy policies relates to CCS, in which only pipeline transport is considered based on the Strategic Energy Technology (SET) plan [80]. Some metals such as vanadium and niobium (scarce), and iron (bulk) are assessed; however, they are insignificant regarding the overall metal requirement and cumulative production. 5.3. Future work – scenarios, metal requirement and its assessment, uncertainties Through this exercise, it was recognized that the following remaining and challenging issues require addressing. Firstly, the contrasted metal requirements comparing two energy and climate policy scenarios were undertaken, but the work did not address ‘‘radical” or ‘‘extreme” scenarios at the global level, such as 100% renewables (like Denmark) or 70–80% nuclear (like France) which is left as a task for the future. It is however, clear that such a radical capacity expansion may increase metal requirements and cumulative production by several fold; hence, the trends identified in this study would not be greatly changed. Secondly, as pointed out in past

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

14

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

Coal & Nuclear, Gas & Renewable) while existing studies in scenario building community typically treat only one direction. As illustrated in the results, employing the energy models to metal requires enables us to estimate the requirement under scenarios developed by the modelers based on our original energy & climate policy scenarios. The collaboration of the both communities in the energy-mineral nexus bring great benefit to policy makers and research scientists for better understandings and decision makings in policy choice. Based on our first look at estimates from a conservative (or cautions) view, the metal requirement in 2100 and cumulative production in some metals in PV were determined to exceed the 2015 production levels and reserves especially in PV under the Gas & Ren scenario while such excess was little observed under the Coal & Nuc scenario. In order to obtain more solid conclusions about metal requirements, further work is required to reduce uncertainties related to the intensity of use of metals and metal conservation (via technological progress), technology deployment, and recycling rate under more radical energy scenarios.

studies, and recognized in this one, the intensity of use of metals has large uncertainty. In this case, the need to meet metal requirements and availability as well as geographical concentration of a few suppliers might be expected to decrease because the results obtained are extreme examples used to understand more easily when the intensity of use of resources is higher. The share of technology, and recyclability of end-of-use plants, are other issues with great uncertainty that should be addressed. Investigation and critical appraisal of these issues may be listed as another task for the future. Thirdly, the metals untreated in this study and other technologies (i.e., hydrogen related, fuel cells) are also need to be addressed. 6. Summary and conclusions We summarize our originalities and conclude our outcomes. The primary point of originality in this study arises from employing an energy model to approach the energy-mineral nexus. To the authors‘ knowledge, very few energy modelers have previously addressed this topic with the collaboration of communities in materials, metals and mineral resources. Past studies on mineral requirements were addressed by borrowing energy scenarios from authorities, while in this case original energy and climate policy scenarios were developed by addressing the 2 °C target, which is the second point of originality. Our third originality in scenario building is providing simultaneously both policy directions (i.e.,

Acknowledgements The first author gives his deepest thank to National Institute of Advanced Industrial Science and Technology (AIST) and other research funds including Arai Science and Technology Foundation and TEPCO Memorial Foundation for supporting to this study.

Appendix. Detailed structure of overall mineral balance model

New scrap

Copper ore

Mining, milling

Copper Concentrate

Smelting, refining

Anode Refined Copper

Mining, leaching

Copper Solution

SxEw

Refined Cathode Copper

Scrap stock Final product manufacturing

Scrap stock

New scrap mining, alumina production

Bauxite

Alumina

Anode Refined Aluminum

Furnace

Final product manufacturing

Mining, milling

Lead Concentrate

Smelting, refining

Anode Refined Lead

Scrap stock

Mining, milling

Zinc Sulfide Concentrate

Smelting, refining, Converter

Anode Refined Zinc

Final product manufacturing

Lead

Galvanized Zinc in electric furnace

Lead scrap

Scrap stock

New scrap

Zinc sulfide ore

Aluminum scrap

Aluminum

New scrap

Lead ore

Copper scrap

Copper

Final product manufacturing

Zinc

Zinc scrap Zinc scrap on the steel

Zinc recovery New scrap

mining

Iron

Coal

Coke

DRI

Blast furnace

DRI

Pig Iron

electric furnace casting Converter casting

Slab

Slab

Roll mill

Roll mill

Construction steel Construction steel

Scrap stock Final products making

Construction Steel

Steel scrap

New scrap

Oil

Converter slug

Gas

converter

New scrap Pig Iron

Converter casting

Slab

Roll mill

machinery steel

Electricity DRI

DRI

electric furnace casting

Slab

Roll mill

machinery steel

Final products making

Steel for electric and machinery, automobile

New scrap

Limestone

Cement kiln (wet) Cement kiln (dry)

Coal-fired power

granulated slag Clinker

Cement kiln (dry-advanced) Cement kiln (dry-advanced with capture)

Fly-ash

Cement mill (converter cement) Cement mill (Portland cement) Cement mill (flyash cement)

Scrap stock Cement

concrete finalize (mixture)

Concrete

Concrete scrap

concrete aggregate Disposal Natural sands, gravels, stones

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

References [1] United Nations Framework Convention on Climate Change (UNFCCC); 2016. http://unfccc.int/paris_agreement/items/9444.php [accessed December 15, 2016]. [2] Climate Action Tracker (CAT); 2016. http://climateactiontracker.org/ methodology/263/CAT-rating-system-update.html [accessed December 15, 2016]. [3] Tokimatsu K, Yasuoka R, Nishio M. Global zero emissions scenarios: the role of biomass energy with carbon capture and storage by forested land use. Appl Energy 2017; 185(Part 2): 1899–906 (January 1). [4] Agency for Natural Resources and Energy (ANRE). Strategic energy plan; 2014 http://www.enecho.meti.go.jp/en/category/others/basic_plan/pdf/4th_ strategic_energy_plan.pdf [5] Murakami S, McLellan B. Introduction to the Special Issue: Minerals-Energy Nexus. Energy and Resources 2016; 37(3): 11–5 [in Japanese]. [6] Food and Agriculture Organization (FAO). The water-food-energy nexsus: A new approach in support of food security and sustainable agriculture. Rome: Food and Agriculture Organizaion of the United Nations; 2014. [7] Smajgl A, Ward J, Pluschke L. The water–food–energy nexus - realising a new paradigm. J Hydrol 2016;533(2):533–40. [8] Yan J, Shamim T, Chou SK, Desideri U, Li H. Clean, efficient and affordable energy for a sustainable future. Appl Energy 2017;185:953–62. [9] United States Department of Energy (USDOE). Critical Materials Strategy, United States Department of Energy; 2011. [10] Joint Research Centre (JRC). Critical metals in the path towards the decarbonisation of the EU energy sector, European Commission Joint Research Centre, Report EUR 25994; 2013. https://setis.ec.europa.eu/sites/ default/files/reports/JRC-report-Critical-Metals-Energy-Sector.pdf [accessed December 15, 2016]. [11] Vikström H, Davidsson S, Höök M. Lithium availability and future production outlooks. Appl Ener 2013;110:252–66. [12] Giurco D, McLellan B, Franks DM, Nansai K, Prior T. Responsible mineral and energy futures: views at the nexus. J Clean Prod 2014;84:322–38. [13] Murakami S, Kawamoto T, Masuda A, Daigo I. Metal demand to meet SDG energy-related goals. Glob Env Res 2015;19:181–6. [14] Kleijn R, van der Voet E. Resource constraints in a hydrogen economy based on renewable energy sources: An exploration. Renew Sustain Energy Rev 2010;14:2784–95. [15] Elshkaki A, Graedel TE. Dynamic analysis of the global metals flows and stocks in electricity generation technologies. J Clean Prod 2013;59:260–73. [16] Kleijn R, Van der Voet E, Kramer GJ, Van Oers L, Van der Giesen C. Metal requirements of low-carbon power generation. Energy. 2011;36:5640–8. [17] Davidsson S, Grandell L, Wachtmeister H, Höök M. Growth curves and sustained commissioning modelling of renewable energy: Investigating resource constraints for wind energy. Ener Pol 2014;73:767–76. [18] Critical Materials Institute (CMI); 2016. [accessed December 15, 2016]. [19] Moss RL, Tzimas E, Kara H, Willis P, Kooroshy J. The potential risks from metals bottle necks to the deployment of strategic energy technologies. Energy Policy 2013;55:556–64. [20] Stamp A, Wäger PA, Hellweg S. Linking energy scenarios with metal demand modeling-the case of indium in CIGS solar cells. Resour, Conservation Recycling 2014;93:156–67. [21] World Wide Fund for Nature (WWF) and Ecofys. Critical materials for the transition to a 100% sustainable energy future; 2014. http://awsassets.panda. org/downloads/critical_materials_report___jan_2014_lr.pdf [accessed December 15, 2016]. [22] Kim J, Guillaume B, Chung J, Hwang Y. Critical and precious materials consumption and requirement in wind energy system in the EU 27. Appl Energy 2015;139:327–34. [23] Wigley TML, Richels R, Edmonds J. Economics and environmental choices in the stabilization of atmospheric CO2 concentrations. Nature 1996;379:240–3. [24] McLellan B, Corder GD, Ali SH. Sustainability of rare earths-an overview of the state of knowledge. Minerals 2013;3(3):304–17. [25] McLellan B, Corder G, Ali S, Golev A. Rare metals, unconventional resources, and sustainability. Geological Society of America Special Papers 520; 2016. [26] Mohr SH, Mudd G, Giurco D. Lithium resources and production: critical assessment and global projections. Minerals 2012;2(1):65–84. [27] European commission critical raw materials for the EU; European Commission: 30th July, 2010 http://www.catalysiscluster.eu/wp/wp-content/ uploads/2015/05/2010_Critical-raw-materials-for-the-EU-2010.pdf [accessed January 15, 2017]. [28] National Research Council. 2008, Minerals, critical minerals, and the US economy. Washington: National Academies Press; 2008. [29] Rabe W, Kostka G, Smith Stegen K. China’s supply of critical raw materials: Risks for Europe’s solar and wind industries? Energy Policy 2017;101:692–9. [30] Viebahn P, Soukup O, Samadi S, Teubler J, Wiesen K, Ritthoff M. Assessing the need for critical minerals to shift the German energy system towards a high proportion of renewables. Renew Sustainable Energy Rev 2015;49:655–71. [31] Grandell L, Lehtilä A, Kivinen M, Koljonen T, Kihlman S, Lauri LS. Role of critical metals in the future markets of clean energy technologies. Renew Energy 2016;95:53–62. [32] Harper EM, Diao Z, Panousi S, Nuss P, Eckelman MJ, Graedel TE. The criticality of four nuclear energy metals. Resour, Conserv Recycling 2015;95:193–201.

15

[33] Hurd AJ, Kelley RL, Eggert RG, Lee M-H. Energy-critical elements for sustainable development. MRS Bull 2012;37(04):405–10. [34] May D, Prior T, Cordell D, Giurco D. Peak minerals: theoretical foundations and practical application. Nat Resour Res 2012;21(1):43–60. [35] Northey S, Mohr S, Mudd GM, Weng Z, Giurco D. Modelling future copper ore grade decline based on a detailed assessment of copper resources and mining. Resour, Conserv Recycling 2014;83:190–201. [36] Mohr S, Giurco D, Yellishetty M, Ward J, Mudd G. Projection of iron ore production. Nat Resour Res 2015;24(3):317–27. [37] Mohr SH, Wang J, Ellem G, Ward J, Giurco D. Projection of world fossil fuels by country. Fuel 2015;141:120–35. [38] Mudd GM, Jowitt SM, Werner TT. The world’s by-product and critical metal resources part I: Uncertainties, current reporting practices, implications and grounds for optimism. Ore Geol Rev; 2016 [online 5 May 2016]. [39] Werner TT, Mudd GM, Jowitt SM. The world’s by-product and critical metal resources part II: A method for quantifying the resources of rarely reported metals. Ore Geol Rev 2017;80:658–75. [40] Speirs J, McGlade C, Slade R. Uncertainty in the availability of natural resources: Fossil fuels, critical metals and biomass. Energy Policy 2015;87:654–64. [41] Hein JR, Mizell K, Koschinsky A, Conrad TA. Deep-ocean mineral deposits as a source of critical metals for high- and green-technology applications: Comparison with land-based resources. Ore Geol Rev 2013;51:1–14. [42] Binnemans K, Jones PT, Blanpain B, Van Gerven T, Yang Y, Walton A, Buchert M. Recycling of rare earths: a critical review. J Cleaner Prod 2013;51:1–22. [43] Habib K, Wenzel H. Exploring rare earths supply constraints for the emerging clean energy technologies and the role of recycling. J Clean Prod 2014;84:348–59. [44] Schüler D, Buchert, M, Liu D.-I.R, Dittrich D.-G.S, Merz D.-I.C. Study on rare earths and their recycling. Öko-Institut eV Darmstadt; 2011. [45] Elshkaki Ayman, Graedel TE. Solar cell metals and their hosts: A tale of oversupply and undersupply. Appl Energy 2015;158:167–77. 15 November. [46] Speirs J, Houari Y, Gross R. Materials availability: Comparison of material criticality studies - methodologies and results Working Paper III. UK Energy Research Centre; 2013. [47] United States Geological Survey (USGS). Mineral commodity summaries 2016, USGS and United States Department of Interior; 2016 https://minerals. usgs.gov/minerals/pubs/mcs/2016/mcs2016.pdf [accessed December 15, 2016]. [48] Gordon RB, Bertram M, Graedel TE. Metal stocks and sustainability. Proc Natl Acad Sci 2006;103(5):1209–14. [49] Tilton JE, Lagos G. Assessing the long-run availability of copper. Resour Policy 2007;32(1–2):19–23. [50] Gordon RB, Bertram M, Graedel TE. On the sustainability of metal supplies: A response to Tilton and Lagos. Resour Policy 2007;32(1–2):24–8. [51] Habib K, Wenzel H. Reviewing resource criticality assessment from a dynamic and technology specific perspective – using the case of direct-drive wind turbines. J Clean Prod 2016;112(Part 5):3852–63. [52] Kushnir D, Sandén BA. The time dimension and lithium resource constraints for electric vehicles. Resour Policy 2012;37(1):93–103. [53] Alonso E, Field FR, Kirchain RE. Platinum availability for future automotive technologies. Environ Sci Technol 2012;46(23):12986–93. [54] Glaister BJ, Mudd GM. The environmental costs of platinum-PGM mining and sustainability: Is the glass half-full or half-empty? Miner Eng 2010;23 (5):438–50. [55] Goe M, Gaustad G. Identifying critical materials for photovoltaics in the US: A multi-metric approach. Appl Energy 2014;123:387–96. [56] Bustamante ML, Gaustad G. Challenges in assessment of clean energy supplychains based on byproduct minerals: A case study of tellurium use in thin film photovoltaics. Appl Energy 2014;123:397–414. [57] Simoes S, Fortes P, Dias L, Seixas J. Assessing critical metal needs for a low carbon energy system in 2050. In: 2015 12th international conference on the European energy market (EEM), 19–22 May, Lisbon, Portugal; 2015. p. 1–5. [58] McLellan B, Yamasue E, Tezuka T, Corder G, Golev A, Giurco D. Critical minerals and energy-impacts and limitations of moving to unconventional resources. Resources 2016;5(2):19. [59] Tao CS, Jing J, Tao M. Natural resource limitations to terawatt-scale solar cells. Sol Energy Mater Sol Cells 2011;95:3176–80. [60] Grandell L, Thorenz A. Silver supply risk analysis for the solar sector. Renew Energy 2014;69:157–65. [61] Fraunhofer ISE. Photovoltaics report. Fraunhofer Institute for Solar Energy Systems ISE; 2016 https://www.ise.fraunhofer.de/en/news/news-archive/ news-2012/fraunhofer-ise-publishes-photovoltaics-report [accessed January 15, 2017]. [62] Kavlak G, McNerney J, Jaffe RL, Trancik JE. Metal production requirements for rapid photovoltaics deployment. Energy Environ Sci 2015;8:1651–9. [63] Jean J, Brown PR, Jaffe RL, Buonassisi T, Bulovi V. Pathways for solar photovoltaics. Energy Environ Sci 2015;8:1200–19. [64] Tokimatsu K, Murakami S, Adachi T, Ii R, Yasuoka R., Nishio M. Long-term demand and supply of non-ferrous mineral resources by a mineral balance model, Mineral Economics, in press, DOI: 10.1007/s13563-017-0109-8 [65] Tokimatsu K, Konishi S, Ishihara K, Tezuka T, Yasuoka R, Nishio M. Role of innovative technologies under the global zero emissions scenarios. Appl Energy 2016;162:1483–93.

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151

16

K. Tokimatsu et al. / Applied Energy xxx (2017) xxx–xxx

[66] Davidsson S, Höök M. Material requirements and availability for multiterawatt deployment of photovoltaics. Energy Policy 2016, in press. [67] Lea A. (President, International Copper Association), 2016, The longterm fundamentals of copper demand, Copper 2016, 2016.11.14., Kobe, Japan. [68] Hernández D. (Chairman, SONAMI (Chilean Chamber of Mines), Chile), 2016, The Future of Copper and Sustainable Development, Copper 2016, 2016.11.14., Kobe, Japan. [69] ITRPV. International technology roadmap for photovoltaic (ITRPV) 2014 results. No. Sixth Edition, Revision 1, July 2015; 2015. http://www.itrpv.net/ Reports/Downloads/2015/ [accessed Jan 15, 2017]. [70] Hashimoto and Murakami. Mitigation technologies for climate change and critical metals. Energy Resour 2013;34(5):291–5. [71] Fulton L, Cazzola P, Cuenot F. IEA mobility model (MoMo) and its use in the ETP 2008. Energy Policy 2009;37(10):3758–68. [72] Organisation for Economic Co-operation and Development/International Energy Agency 1(OECD/IEA). World energy outlook 2016. Paris: OECD; 2016. p. 2016. [73] Organisation for Economic Co-operation and Development/International Energy Agency (OECD/IEA). Energy Technology Perspectives 2016. Paris: OECD; 2016. p. 2016. [74] GEA (Global Energy Assessment). Global energy assessment – toward a sustainable future, Cambridge University Press, Cambridge UK and New York, NY, USA and the International Institute for Applied Systems Analysis, Laxenburg, Austria; 2012. http://www.iiasa.ac.at/web/home/research/ Flagship-Projects/Global-Energy-Assessment/Chapters_Home.en.html [accessed December 15, 2016]. [75] World Wide Fund for Nature (WWF), Ecofys, office for metropolitan architecture (OMA). The energy report 100% renewable energy by 2050; 2011. http://wwf.panda.org/about_our_earth/all_publications/energy_report/ [accessed December 15, 2016].

[76] Jacobson and Delucchi. Providing all global energy with wind, water, and solar power, PartI: Technologies, energy resources, quantities and areas of infrastructure, and materials. Energy Policy 2011;39(3):1154–69. [77] Delucchi and Jacobson. Providing all global energy with wind, water, and solar power, PartII: Reliability, system and transmission costs, and policies. Energy Policy 2011;39(3):1170–90. [78] Blanco G, Gerlagh R, Suh S, Barrett J, de Coninck HC, Diaz Morejon CF, Mathur R, Nakicenovic N, Ofosu Ahenkora A, Pan J, Pathak H, Rice J, Richels R, Smith SJ, Stern DI, Toth FL, Zhou P. Drivers, Trends and Mitigation. In: Edenhofer O, Pichs-Madruga R, Sokona Y, Farahani E, Kadner S, Seyboth K, Adler A, Baum I, Brunner S, Eickemeier P, Kriemann B, Savolainen J, Schlömer S, von Stechow C, Zwickel T, Minx JC, editors. Climate change 2014: mitigation of climate change. Contribution of working group III to the fifth assessment report of the intergovernmental panel on climate change. Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press; 2014. [79] Bruckner T, Bashmakov IA, Mulugetta Y, Chum H, de la Vega Navarro A, Edmonds J, Faaij A, Fungtammasan B, Garg A, Hertwich E, Honnery D, Infield D, Kainuma M, Khennas S, Kim S, Nimir HB, Riahi K, Strachan N, Wiser R, Zhang X. Energy Systems. In: Edenhofer O, Pichs-Madruga R, Sokona Y, Farahani E, Kadner S, Seyboth K, Adler A, Baum I, Brunner S, Eickemeier P, Kriemann B, Savolainen J, Schlömer S, von Stechow C, Zwickel T, Minx JC. editors. Climate change 2014: mitigation of climate change. Contribution of working group III to the fifth assessment report of the intergovernmental panel on climate change. Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press; 2014. [80] Joint Research Centre (JRC). Critical metals in strategic energy technologies – assessing rare metals as supply-chain bottlenecks in low-carbon energy technologies, European Commission Joint Research Centre, Report EUR 24884; 2011. http://publications.jrc.ec.europa.eu/repository/bitstream/JRC65592/reqno_ jrc65592_critical%20metals%20in%20strategic%20energy%20technologies%20 (online).pdf [accessed December 15, 2016].

Please cite this article in press as: Tokimatsu K et al. Energy modeling approach to the global energy-mineral nexus: A first look at metal requirements and the 2 °C target. Appl Energy (2017), http://dx.doi.org/10.1016/j.apenergy.2017.05.151