The Hard Math of Minerals
Today’s plans to decarbonize global energy systems, which center on a massive expansion in the use of solar, wind, and battery technologies, need to better account for the high environmental and economic costs of materials and minerals.
The great twentieth-century physicist Richard Feynman once said that “it is important to realize that in physics today, we have no knowledge what energy is.” But we do know one unequivocal fact: delivering useful energy services to society has always been about materials.
Today’s plans to decarbonize global energy systems center on a massive expansion in the use of solar, wind, and battery technologies, with the goal of these becoming the dominant means to power society. But scaling up these energy sources entails a radically heavier materials footprint than is associated with fossil fuels, paradoxical though it may seem. The unavoidable scale of materials demand will have significant impacts on commodities markets and prices, as well as on the environment. Most policy formulations fail to account for these implications. The country is long overdue for thoughtful and realistic planning that honestly acknowledges the tradeoffs and consequences arising from the materials needed to accelerate what is being called the energy transition.
It has long been known that building solar and wind systems requires roughly a tenfold increase in the total tonnage of common materials—concrete, steel, glass, etc.—to deliver the same quantity of energy compared to building a natural gas or other hydrocarbon-fueled power plant. Beyond that, supplying the same quantity of energy as conventional sources with solar and wind equipment, along with other aspects of the energy transition such as using electric vehicles (EVs), entails an enormous increase in the use of specialty minerals and metals like copper, nickel, chromium, zinc, cobalt: in many instances, it’s far more than a tenfold increase. As one World Bank study noted, the “technologies assumed to populate the clean energy shift … are in fact significantly MORE material intensive in their composition than current traditional fossil-fuel-based energy supply systems.”
Today, the material intensity of solar and wind systems and EVs is still of minimal consequence because those technologies account for only a few percentage points of the global energy system. But the material demands will become hard to ignore if the world’s economies all simultaneously pursue similarly ambitious policies to displace the fossil fuels that currently supply over 80% of all energy. The vision plan from the International Energy Agency (IEA), which has been adopted and even exceeded by some policymakers, has solar and wind providing some 60% of net new global energy supply over the coming two decades.
Installing so much wind and solar generation capacity worldwide has profound materials implications, not to mention land requirements, which will soon become problematic. Replacing the energy output from a single 100 megawatt (MW) natural gas-fired turbine (producing enough electricity for 75,000 homes) requires at least 20 wind turbines, each about 500 feet tall and collectively requiring some 30,000 tons of iron ore and 50,000 tons of concrete, as well as 900 tons of nonrecyclable plastics for the turbine blades. The gas turbine, by contrast, requires only about 300 tons of iron ore and some 2,000 tons of concrete. The 20 wind turbines also require 1,000 tons of specialty metals and minerals such as copper, chromium, zinc, etc., versus about 100 tons embodied in the gas turbine. Moreover, the gas turbine is about the size of a residential house, while those 20 wind turbines require 10 square miles of land. And although a solar installation would require one-third as much land as wind, the aggregate tonnage of cement, steel, and glass used is about 150% greater than wind.
And if solar and wind are to become the primary sources of power, then utility-scale electricity storage and additional generating capacity will be required to meet demand and to produce excess energy to be stored. Thus, replacing a 100 MW gas turbine would necessitate at least 200 MW of solar or wind capacity, more than doubling the hardware and materials requirements—along with yet more materials associated with building about 10,000 tons of batteries for energy storage.
Scaling up solar, wind, and batteries also means scaling up the mining of the refined minerals they require. There is a significant environmental impact associated with the sheer tonnage of earth that must be moved and processed to produce these refined minerals. To produce one ton of a purified element, a far greater quantity of ore must be extracted and processed. Copper ores, for example, typically contain only about 0.5% by weight of the element itself: roughly 200 tons of ore are dug up, moved, crushed, and refined to produce 1 ton of copper. The rare earth element neodymium, which is used in wind turbines, requires mining from 20 to 160 tons of ore to obtain 1 ton. Cobalt (used in most batteries) occurs at a grade typically lower than 1 ton of the element per 1,500 tons of ore. The calculus of the upstream environmental footprint should also include the overburden—the necessary removal of even more tons of rocks and dirt to access a single ton of the buried mineral-bearing ore.
The energy transition, as it’s being conceived today, will create a need for tens of gigatons of materials for solar and wind generation, grid storage, and car batteries. The IEA terms this a “shift from a fuel-intensive to a material-intensive energy system.” The agency estimates that an energy plan more ambitious than implied by the 2015 Paris Agreement, but one that remains far short of eliminating the use of fossil fuels, would increase demand for minerals such as lithium, graphite, nickel, and cobalt rare earths by 4,200%, 2,500%, 1,900% and 700%, respectively, by 2040.
Can the world meet the minerals and mining demands of these collective goals? The IEA report is not alone in pointing out that the required mining and processing infrastructure capacities don’t yet exist to meet the demand for essentially every category of mineral necessary for the transition path.
In a recent report from the Geological Survey of Finland, researchers considered the minerals implications for achieving a so-called full transition; that is, using solar and wind to electrify all ground transport as well as to produce hydrogen for both aviation and chemical processes. They found the resulting demand for nearly every necessary mineral, including common ones such as copper, nickel, graphite, and lithium, would exceed not just existing and planned global production capabilities, but also known global reserves of those minerals.
A recent analysis by the Wood Mackenzie consultancy found that if EVs are to account for two-thirds of all new car purchases by 2030, dozens of new mines must be opened just to meet automotive demands—each mine the size of the world’s biggest in each category today. But 2030 is only eight years away and, as the IEA has reported, opening a new mine takes 16 years on average.
Despite these and similar analyses, many countries, and many US states, are now proposing to accelerate deployment of solar, wind, and battery technologies without clear plans for overcoming the material shortfalls. One study sponsored by the Dutch government offered a blunt statement of reality: “Exponential growth in [global] renewable energy production capacity is not possible with present-day technologies and annual metal production.”
Another area of concern for these new technologies is their future cost, which will be inseparable from the velocity and scale of their entry into the market. Today, future plans for solar, wind, and battery technologies assume costs will continue to fall significantly, as they have over the last decade. But the implications of record-breaking demands for mineral commodities suggest the reverse is more likely.
Consider batteries, which underpin hopes to displace fossil fuels both in transportation and in enabling solar- and wind-dominated grids. Numerous estimates (exact data are proprietary) suggest that commodity materials comprise 60 to 70% of the cost to produce a battery. Thus, modest increases in commodity prices can wipe out gains in the smaller share of costs associated with assembly, electronics, and labor, leading to overall higher costs. The IEA’s analysis in early 2021 of “energy transition minerals” noted as much, concluding that future mineral price escalations could “eat up the anticipated” reductions in manufacturing costs expected from the “learning effects” in further scaling up battery production. In fact, 2021 saw high material costs lead to overall lithium battery prices declining by only 6%. That was a dramatic slowdown from the decadal trend, and less than half the decline rate in each of the prior two years. Although EVs comprise only 5% of the market for automobiles, the price index of EV battery metals has already increased by more than 200% over the past two years.
Commodity inflation has begun to escalate the cost to build wind and solar systems as well, slowing or reversing long-run cost declines. As with batteries, progress in manufacturing efficacy has reduced solar module production costs so much that commodity inputs now make up about 70% of the overall price of modules. These inputs include not only copper, silver, and aluminum but also, in no small irony, coal. The energy-intensive fabrication of polysilicon, a key raw material in solar modules, takes place mainly in China (with its two-thirds share of all polysilicon supply) on its low-cost, coal-dominated grid. The combination of mineral commodity inflation and the jump in coal prices pushed solar module prices up nearly 50% over 2020. Wind turbine manufacturers were similarly stung by higher material costs (which make up 20% of their cost) with many now planning to sell turbines with clauses that will “pass through” commodity price hikes onto buyers.
Many analysts claim that materials demand can be greatly alleviated with recycling. The ideal is described as a circular economy achieving nearly complete reuse of materials from discarded hardware. Although a worthy aspiration, myriad practical and economic factors impede getting close to that goal in general, not just with solar, wind, and batteries. And, as one United Nations study observed: “Less than one-third of some 60 metals studied have an end-of-life recycling rate above 50% and 34 elements are below 1% recycling, yet many of them are crucial to clean technologies.” Even if far greater levels of recycling were mandated, the vast quantity of solar and wind equipment required for the energy transition will for decades overwhelm any marginal additions to materials supply that could come from recycling the far smaller quantity from worn-out hardware.
Some proponents of the transition pin their hopes on innovation to reduce materials intensity through improvements to the underlying operating efficiency of the systems: higher photovoltaic conversion efficacy and battery chemistries with higher energy density, for example. But in these realms, gains of 10% or so are hard won. To have a meaningful impact on materials demands would require, rather than 10% efficiency gains, leaps of tenfold over existing solar, wind, and battery technologies—gains that aren’t even theoretically feasible.
There is, in short, no escaping the fact that the astonishing scale of global materials production needed for proposed energy transition plans will almost certainly place severe limits on aspirations for expanding the use of wind, solar, and battery systems. But even before those limits are reached, the pursuit of a materials-heavy energy infrastructure will cause economic impacts that ripple beyond energy markets, inflating the cost of nonenergy uses for the same minerals in computers, conventional manufacturing equipment, everyday consumer appliances, and more.
Beyond economics, there are also the practical and geopolitical challenges arising from realignments of energy material supply chains. For example, the United States today is dependent on imports for 100% of some 17 critical minerals and, for 28 others, net imports account for more than half of existing domestic demand. Assembling batteries or solar hardware in the United States won’t change the underlying dependencies any more than assembling automobiles domestically would if the key components and all the fuel were imported.
Finally, there are the social and moral implications associated with a radical shift in the types and locations of environmental impacts that comes from replacing drilling (for fossil fuels) with a massive expansion in mining, much of which will occur in emerging markets and fragile ecosystems. For example, Australia’s Institute for Sustainable Futures noted in its analysis that the global gold rush for minerals to meet ambitious transition plans could take miners into “some remote wilderness areas [that] have maintained high biodiversity because they haven’t yet been disturbed.”
Meanwhile, little attention has been afforded the social and humanitarian implications of this shift. Jennifer Dunn, a pioneer in social life cycle assessment (S-LCA) and associate director of the Center for Engineering Sustainability and Resilience at Northwestern University’s McCormick School of Engineering, has noted that “technologies that are designed to solve grand challenges such as climate change must consider both their environmental and social impacts to understand their true consequences.” As Dunn and her collaborators observe in a recent analysis focused on cobalt as a case study, while environmental life cycle assessment is a “mature widely-used tool,” social and humanitarian considerations remain nascent and “the lack of regionally or locally specific data and guidance for collecting them are significant barriers to robust and effective S-LCA.”
Policymakers are limited in what they can do to alleviate the materials challenges arising from an overreliance on solar, wind, and battery technologies. While the long history of maintaining military stockpiles for critical minerals may seem like a precedent to emulate, stockpiles don’t solve a systemic supply problem. In any case, the quantities of materials required in the energy sector are many orders of magnitude greater than for defense purposes, rendering that option economically, if not functionally, impossible—even for the security feature that stockpiles are intended to address.
The European Union has acknowledged the need for additional mining, specifically on its own continent, and has even proposed development incentives. But the few attempts thus far to open new mines in EU countries have quickly met with fierce environmental opposition. In the United States, neither Congress nor the administration has proposed anything meaningful to help expand domestic mining industries. Instead, proposals for new mines continue to be blocked.
The obvious approach for avoiding the creation of unsustainable demands for minerals is to adopt more moderate and longer-term deployment targets for solar, wind, and battery hardware. This would necessitate a far less aggressive imposition of mandates and subsidies directed at accelerating market adoption. More realistic policies could not only avoid triggering hyper-inflation in commodity markets, but they would also have the salutary benefit of a more cost-effective, natural evolution of new energy technologies.
The downside to this approach is that it leaves a gap in aspirations for reducing the use of fossil fuels. It bears noting that over the past decade of already accelerated transition policies, hydrocarbon consumption has risen and is forecast by the IEA to continue rising for the usefully foreseeable future. To address this, policies could more productively focus on support for the expanded use of different kinds of technologies, especially those that radically improve fossil fuel efficiencies.
For the longer term, policymakers might take heed of the reality that a goal of “net-zero” will require new technologies that don’t exist today. That reality points to the need for a greater focus on basic scientific research. Unfortunately that path doesn’t have a “predictor function” (to use Bill Gates’s locution) and one cannot, in effect, order up elusive breakthrough technologies. One can imagine but not predict when someone will discover, for example, a low-cost, room-temperature superconductor that would make storing electricity as easy and cheap as storing petroleum, or a metamaterial that synthesizes hydrogen at a scale and cost rivaling natural gas.
Based on today’s physics and technology, the only path to an energy system with a material intensity lower than hydrocarbons would be one focused on nuclear fission. In the pantheon of energy-producing machines, none is more remarkable than the nuclear reactor. Nuclear fission offers a potential hundredfold reduction in material intensity over combustion, and a thousandfold reduction over solar and wind. Here too, though, even if policies are implemented that are conducive to a nuclear renaissance, meaningful expansion will take decades longer than the rapid transition timelines popular today.
The material realities associated with solar, wind, and storage technologies do not obviate an expanded, or even a substantial, role for these energy systems. However, believing that such technologies make possible a rapid and wholesale replacement of fossil fuels ignores the underlying physics, engineering, and economics. Even more troublesome, putting so much effort and money into those technologies will lead the world down a path that won’t meet targets to reduce carbon dioxide emissions, but would cause massive collateral damage to economies and the environment. If Feynman were alive today, one suspects he would repeat another of his favored aphorisms: “For a successful technology, reality must take precedence over public relations, for nature cannot be fooled.”