Requirements for Minerals and Metals for 100% Renewable Scenarios

11.2.1 Solar PV

A typical crystalline silicon (c-Si) PV panel, which is currently the dominant technology, with over 95% of the global market, contains about 76% glass (panel surface), 10% polymer (encapsulant and back-sheet foil), 8% aluminium (frame), 5% silicon (solar cells), 1% copper (interconnectors), and less than 0.1% silver (contact lines) and other metals (e.g., tin and lead). Thin film technologies, copper–indium–gallium–(di)selenide (CIGS) and cadmium telluride (CdTe), make up the remainder of the market. These technologies require less material overall than crystalline silicon. For CdTe panels, the composition is 96–97% glass, 3%–4% polymer, and less than 1% semi-conductor materials (CdTe) and other metals (e.g., nickel, zinc, tin). CIGS contain about 88%–89% glass, 7% aluminium, 4% polymer, and less than 1% semi-conductor material (indium, gallium, selenium) and other metals (e.g., copper) (Weckend et al. 2016). Figure 11.1 provides a simplified diagram of the PV supply chain, including key materials and sub-components.

Open image in new window

11.2.2 Wind

The major raw materials required for the manufacture of wind turbine components are bulk commodities: iron ore, copper, aluminium, limestone, and carbon. Wind turbines use steel for the towers, nacelle structural components, and the drivetrain, accounting for about 80% of the total weight. Some turbine generator designs use direct-drive magnetics, which contain the rare earth metals neodymium and dysprosium (Fig. 11.2). The development of direct-drive permanent magnet generators (PMG) by major producers (e.g., Siemens and General Electric) simplifies the design by eliminating the gearbox, and this is attractive for offshore applications because it reduces maintenance (Zimmermann et al. 2013). It is estimated that about 20% of all installed wind turbines (both onshore and offshore) use rare earth magnets (CEMAC 2017).

Open image in new window

11.2.3 Batteries and Electric Vehicles

This study focuses on lithium ion batteries (LIBs), which power almost all electric vehicles (EVs) on the market today and are also an important technology for stationary energy-storage applications. LIBs are made of two electrodes (anode and cathode), current collectors, a separator, electrolyte, a container, and sealing parts. The anode is typically made of graphite, with a copper foil current collector. The cathode is typically a layered transition metal oxide, with an aluminium foil current collector. In between the electrodes is a porous separator and electrolyte. All of these components are typically housed in an aluminium container. LIBs are generally referred to by the material content of the cathode, which accounts for 90% of the material value and about 25% of the total weight (Gratz et al. 2014).

The size and type of the LIB has the greatest impact on the material requirements. Since commercialization in the 1990s, a range of different types (‘chemistries’) have been developed for different applications, named according to the metals in the cathode. The most common LIB types for EV applications are nickel–manganese–cobalt (NMC), lithium–iron phosphate (LFP), nickel–cobalt–aluminium (NCA), and lithium–manganese oxide (LMO) (Vaalma et al. 2018). NMC is the most common battery type for passenger vehicles, and NCA is also common, with a small share for LMO. However, in China, LFP is the dominant chemistry. Electric buses have traditionally used LFP batteries (BNEF 2018) and lead–acid batteries are most commonly used for two-wheel vehicles. However, the application of LIBs in this market sector is growing (Yan et al. 2018).

For energy storage, NMC and NCA are the most commonly used chemistries. A simplified overview of the lithium-ion battery supply chain, including its key metals (for the NMC chemistry) and sub-components, is shown in Fig. 11.3. Rare earth permanent magnets using neodymium and dysprosium are common in most electric vehicle motors. Other motor technologies, or those that replace rare earths with lower-cost materials, are under development and are already used in some vehicles. However, rare earth magnets are expected to remain the standard for the foreseeable future because of their higher performance characteristics (Widmer et al. 2015).

Open image in new window

11.3.1 Electricity and Transport Scenarios

The energy scenarios presented here were developed to achieve the climate target in the 2015 Paris Climate Agreement of limiting anthropogenic climate change to a maximum of 1.5 °C above pre-industrial levels. This projection is more ambitious than the IEA’s annual World Energy Outlook (WEO) scenarios, which project that current policies, and therefore the global development of renewable power and electric mobility, will not change. In contrast, many scenarios have been proposed in the academic literature that extend as far as zero emissions, whereas the IEA’s own “Energy Technology Perspectives” (ETP) scenarios include both a 2.0 °C scenario (2DS) and a more ambitious “Beyond 2.0 °C” scenario (B2DS), which aim to achieve the Paris targets. The scenario proposed here includes both a shift to 100% renewable electricity and a shift to renewable electricity and fuels in the transport sector.

In this scenario, solar PV will account for more than one-third of the installed capacity by 2050, with the remainder from wind and other renewables. Lithium-ion batteries will account for approximately 6% of energy storage (which will be dominated by pumped hydro and hydrogen).

In the transport system, we focus on the material requirements for the batteries used in road transport, because other types of transport do not require batteries to power drivetrains or are assumed to rely on other forms of energy (e.g., biofuels for aviation). In 2050, most of the energy for road transport will come from electricity (55%) and hydrogen (22%), and the remainder will be from biofuels and synfuels. In the 1.5 °C Scenario, the batteries required to electrify road transport are specified for electric buses and passenger cars, including battery electric vehicles (BEV), plug-in hybrid electric vehicles (PHEV), and commercial vehicles. Passenger cars will account for 92% of vehicles and 51% of battery capacity, whereas commercial vehicles are projected to account for 48% of battery capacity, although they will make up only 8% of the total fleet of vehicles. This is because the battery sizes for commercial vehicles (assumed to be 250 kWh in 2015 and rising to 600 kWh in 2050) are larger than those for passenger vehicles (5–15 kWh for PHEV and 38–62 kWh for BEV). Buses will account for a small percentage (1%) of both vehicles and batteries. Electric bikes and scooters have been excluded, because although they are currently a growing market in Asia, by 2050, their share of electricity consumption will be negligible compared with the predicted uptake of electric passenger and commercial vehicles. Lead–acid batteries are also the main type of battery for electric bikes and scooters, although lithium-ion may replace these in future.

11.3.2 Resource Scenario Development

11.3.3 Technology Assumptions

The total metal demand for renewable energy and battery technology each year is estimated based on the metal intensity of a specific technology and the capacity of each technology introduced in a specific year. This introduced stock will accounts for the new capacity and the replacement of technologies at the end of their lives, based on a lifetime distribution curve for the average lifetime.

The values for metal intensity are given in tonnes/GW (for solar PV) or tonnes/GWh (for batteries) of capacity, and two values are given for each metal to evaluate the impact of improving the material efficiency (for solar PV) and technology shifts (for batteries). To evaluate the impact of recycling, the primary demand is estimated by multiplying the discarded products by the recycling rate. The recycling rate is obtained by multiplying the collection rate by the recovery efficiency of a metal from a specific technology. This recycling rate is also varied to obtain a current rate and a potential rate. The potential recycling rate is technologically possible, but is not currently applicable because it is not economic. The detailed assumptions for batteries and solar PV are explained in detail in the following section.

11.3.4 Batteries

We have assumed a collection efficiency of 100% for all batteries and a recovery rate of 90% for Co and Ni (Georgi-Maschler et al. 2012). A 10% recovery rate is assumed for Li, acknowledging that pyrometallurgical processing routes account for most of the current global capacity, and Li recovery may not be possible by this route (King et al. 2018). For scenarios that assume a ‘potential future recycling’ rate, we have assumed 95% recovery for all metals. This is reasonable given that 100% recovery has been reported in the laboratory (Gratz et al. 2014).¹ However, some losses during processing seem unavoidable.²

11.3.5 Solar PV

For solar PV, we assume that the technology types do not change until 2050, and that they retain their current market shares, so that crystalline silicon will remain the dominant technology. We have modelled the potential to offset demand through increases in material efficiency and increases in recycling.

A high and a low value are given for silver to show the impact of material efficiency on silver demand. The current data on silver intensity are from a survey of the PV industry (ITRPV 2018), and the future material efficiency is based on an assumed minimum amount of silver (Kavlak et al. 2015). The ‘current recycling rate’ scenarios assume a current collection rate of 85% for all panels, consistent with the target of the EU WEEE Directive.³ This should be considered an average rate, noting that remote and/or distributed roof-top systems will be more costly to collect and transport than large utility-scale PV. We assume that no recycling is currently occurring for silver from PV, but that 95% recovery may be possible. Therefore, for the ‘potential recycling rate’ scenarios, we assume a 95% recovery efficiency and an 85% collection rate, which will result in an 81% metal recycling rate.

11.3.6 Metal Assumptions

11.4.1 Cobalt

The annual demand for cobalt from EVs and storage could exceed the current production rates in around 2023 (in all scenarios). In the ‘future technology’ scenario, shifting to Li–S instead of LIB will reduce the demand for cobalt. However, recycling, rather than shifting technologies, will have the greatest impact on reducing the primary demand in both the current technology and future technology scenarios.

The cumulative total demand to 2050 (with current technology and no recycling) could exceed current reserves by 400%, and exceed current resources by 20%. Even with recycling and a shift to technologies that use less cobalt, the cumulative demand will still exceed reserves. However, in these scenarios, the demand will remain below the resource levels (Figs. 11.6 and 11.7).

Open image in new window

Open image in new window

11.4.2 Lithium

The annual demand for lithium for EVs and storage could exceed the current production rates by around 2022 (in all scenarios). In the ‘future technology and no recycling’ scenario, a shift to Li–S will increase the demand for lithium, because these batteries have a higher amount of lithium. Increasing recycling from its current low levels (which are assumed to be 10%) will offer the greatest potential to offset the primary demand for lithium.

The cumulative demand for lithium by 2050 will be below the resource levels for all scenarios, but will exceed the reserves unless there is a shift to a high recycling rate. The cumulative demand could be as high as 170% of the current reserves with the current technology, and could be 280% of current reserves with a switch to Li–S batteries (Figs. 11.8 and 11.9).

Open image in new window

Open image in new window

11.4.3 Silver

The total annual demand for silver could reach more than 40% of the current production rates by 2050, assuming no recycling and that the materials efficiency does not change (Fig. 11.10). The cumulative demand to 2050 could reach around half the current reserves with the current technology, and around one-quarter if the technology improves (Fig. 11.11). The reduction in material intensity in the ‘future technology’ scenario, in which silver use decreases from 20 to 4 tonnes/GW, has the greatest potential to reduce demand.

Open image in new window

Open image in new window

11.5.1 Limitations

This study only focuses on the metal demand for renewable energy and transport. It does not take into account other demands for these metals. It is expected that with the increase in renewable energy, renewable energy technologies will consume a greater share of these metals. In our modelling, the recycled content used in technologies comes only from metals from the same technologies at the end of their lives. However, the demand could potentially be offset by accessing other secondary sources of the metal. At the same time, these technologies are only one end-use within the overall economic demand for these metals. The expansion of renewable energy and storage technologies could have a significant effect on the overall market dynamics, including influencing prices, which may feed back to the efforts to reduce material intensity. These results only focus on cobalt, lithium, and silver because of their importance. However, further analysis is required of other metals that will be important in the renewable energy transition.

Another important limitation is that this analysis does not consider the impact of the demand for a mineral on the mining of that mineral and therefore on the amount of energy required for this mining and processing activity. In particular, as the potential ore grade declines, and polymetallic ore processing and the mining of deeper ore bodies increase, it is possible that this feedback loop could have a more-than-marginal influence on the overall sectoral energy consumption. To examine this is an area accurately will require much more-complex modelling.

11.5.2 Comparison with Other Studies

A large number of studies have examined scenarios for renewable energy and storage technologies that will mitigate climate change. In recent years, there has also been an upsurge in studies of mineral ‘criticality’, which have paralleled the present study in terms of the high penetration of renewable and storage technologies and the potential constraints that certain minerals may impose. This increased interest has been prompted to some extent by China’s rare earth export restrictions of 2009–2011, which reflect the sense that mineral supply chains are still quite insecure. Most of these studies have addressed specific technologies or specific countries or regions, rather than global climate targets. A number of studies have specifically and directly addressed the Paris Agreement targets (ensuring that the temperature rise does not exceed 2.0 °C), although the modelling frameworks have been slightly different. Some of the authors of the present chapter have been involved in these studies (Tokimatsu et al. 2017; Watari et al. 2018).

The results for silver are in the middle of the range of results given by other studies. Our scenario includes a higher installed capacity of PV in 2050 than the other studies. However, this study also assumes a lower metal intensity than previous studies, because new data have been published based on the current material use by the PV industry (ITRPV 2018).

11.6.1 Geopolitical Landscape

The geopolitical shift underway in the supply of the resources required for the globe’s future energy mix is clearly illustrated in Fig. 11.12. Whereas the value of the lithium industry is much less than the value of the oil industry, this comparison highlights a distinct shift in the energy commodities that society values. Oil’s rate of use is projected to decline somewhat in the decade ahead (Mohr et al. 2015), whereas lithium’s production is expected to grow rapidly (Mohr et al. 2012).

Open image in new window

Although it is not a key focus of this particular chapter, the dominance of China in the supply of rare earths has encouraged manufacturing countries to look at diversifying their supply sources, including through the recovery of rare earths from recycled material. A similar situation of heavily concentrated supply occurs for cobalt, where the Democratic Republic of the Congo is the largest supplier, at around 66,000 tonnes per year. The next four largest cobalt-producing countries (China, Canada, Russia, and Australia) only produce 5–7000 tonnes per year each. The supply chain is also concentrated downstream, with around 50% of cobalt smelted and refined in China.

11.6.2 Social and Environmental Impacts

The mining and supply chain for these metals can have adverse social and environmental consequences for workers, local communities, and the environment. These impacts are most significant for the cobalt mined in the Democratic Republic of Congo, where there are human rights violations, child labour, and severe environmental pollution affecting health (Florin and Dominish 2017).

These types of impacts at mine sites and along the supply chain also influence the availability of primary resources. For example, whereas Australia and Chile are large producers of lithium, large deposits remain undeveloped in Bolivia, due in part to local concerns over the social and environmental impacts. The global silver market receives less media attention than the market for lithium, but the world’s second-largest silver mine (Escobal) in Guatemala is currently closed by a constitutional court ruling that the Xinca Indigenous peoples were not adequately consulted before a mine licence was granted (Jamasmie 2018).

The increased use of materials such as lithium, cobalt, and silver has economic implications for the future of battery manufacture. For example, the cost of cobalt has risen dramatically from US$20000/t in 2016 to US$80000 in 2018 (Tchetvertakov 2018). This is prompting manufacturers to look for alternatives, such as nickel, vanadium, and zinc (Tchetvertakov 2018). At the same time, a significant proportion of these price fluctuations are attributable to non-industrial factors, such as investment and speculation in metal markets, which may also require greater regulation to avoid unnecessary restrictions on renewable energy and storage technologies.