Mineral Resource Usage for the Renewable Energy Transition
Our civilisation is still mostly powered by fossil fuels; this energy source is limited, we are about half way through the known and obviously non-renewable resources; most importantly, however, their use and combustion has detrimental effects on life on Earth in many ways, including but not limited to climate change. A transition to an alternative source in the form of renewable energy is therefore needed, i.e. away from mining vast amounts of fossil fuels to be burnt in power plants, vehicles and homes, to mining less but specific minerals for an energy infrastructure without continuous fuel input. The question of how this transition would be feasible in terms of the amount of minerals needed has puzzled me for a long time (Renewable Energy Transition Modelling).
This story tries to communicate an idea of the order of magnitude for minerals needed during the transition and beyond, derived from various sources such as the International Energy Agency’s “The Role of Critical Minerals in Clean Energy Transitions” and Updated Mining Footprints and Raw Material Needs for Clean Energy. The good news is that such a transition should be feasible; but there are some important reservations.
Status Quo: World Powered by Fossil Fuels
In 2023, according to Our World In Data, the global annual primary energy consumption of 164,000 TWh consisted of 40,100 TWh in gas, 54,560 TWh in oil and 45,560 TWh in coal. Let’s look at the order of magnitude in terms of mass of this fossil fuel consumption given their specific energy density. The global total consumption in 2022 according to the World Energy & Climate Statistics Yearbook was:
- oil: 4.4·10¹²kg = 4,400Mt = 4.4Gt, for transport (64%), i.e. petrol and diesel, non-energy use (16%), industry (8%); percentages from Breakdown of oil consumption by sector.
- gas: 2.9·10¹²kg = 2,911Mt = 2.9Gt, for industrial use (30%), electricity generation (28%), domestic heat (20%); percentages from Natural Gas Global Consumption by End Use.
- coal: 8.5·10¹²kg = 8,500Mt = 8.5Gt, for electricity generation (58%), metallurgy (16%), residential heat (14%), industrial use (12%); percentages from The world coal consumption divided into subgroups.
Taking the numbers and energy related percentages from above, the portions related directly or indirectly to energy such as transport, electricity generation or heat (domestic and industrial) consumed about 2.8Gt oil, 1.4Gt gas and 6.1Gt coal (in 2022); the non-energy related consumption of fossil fuels is not in scope here but looked into in From Fossil to Solar Powered Chemistry in a Circular Economy. By 2025, the consumption is slightly higher, but the order of magnitude would probably be still the same.
It is worth mentioning that fossil fuels for energy purposes are mostly combusted in order to extract the useful final energy, e.g. electricity in fossil-fuel based power generation, or motion in the internal combustion engine of cars and trucks. Given the thermodynamic efficiency of processes in heat machines, on average only about 1/3 of the primary energy is actually extracted to do some useful work; the other 2/3 are being wasted as shown in the diagram below. Of course, some processes such as a combined cycle gas turbine may be about 50–60% efficient, but on the other end automotive ICE may operate only at around 25% or less.
Objective: World Powered by Renewable Energy
Let’s assume that the world transitions from a negligible amount of renewable energy today (in form of electricity from Wind, Water & Solar, WWS¹) to 100% with the next 25 years while completely ramping down the equivalent fossil fuel consumption over the same period (from 164,000 TWh primary energy per year to zero). That would also mean ramping down CO₂ emissions through burning fossil fuels from about 40Gt/annum to zero, or an accumulated 25*40/2 = 500Gt, roughly the allowance to contain CO₂ induced warming within 1.5°C with a probability of about 50% used to be; this is the reason to choose 25 years for the transition from 2025 to 2050.
The consumption of 164,000 TWh primary energy per year equates to approximately 18.7TW as average continuous power. In contrast, once the world is completely electrified in 2050, i.e. all the services above, e.g. transport, domestic and industrial heat, only about 1/3 will be needed, at least in theory. This translates to 55,000 TWh/annum or about 6.24TW continuous power in 2023 or 2025. For simplicity, let’s assume that the same amount of energy has to be produced by renewable resources in 2050 (no growth²). The questions now are, a) what is the material resource cost of that transition, and b) what additional material resources are still needed for their maintenance of that infrastructure beyond 2050.
Transitioning to the New Energy World
First, an important issue needs to be addressed: how can an annual energy production of 55,000 TWh/annum (6.4TW average continuous power) be realised using WWS given the inherent intermittency of sun and wind, and in extreme cases, the dreaded Dunkelflaute. It is in the end a question of quality of service that can be achieved at a certain cost in statistical terms with a combination of the following provisions:
- Overbuild: installed WWS capacity needs to be a multiple of the required power, perhaps a factor of 3, i.e. about 18TW or 160,000 TWh/annum, i.e. incidentally compensating the gain in energy in-efficiency of burning fossil fuels. Since water and wind have limited capacity, perhaps up to 2TW combined (source), the majority would have to be solar, i.e. about 16TW of installed solar capacity alone.
- Storage: in order to smooth over peaks and troughs of the difference in consumption and production, electricity storage is needed. In my studies to produce power for UK homes with wind alone I found that a capacity of 10% of the total annual energy would be needed. Water & overbuild solar from above will reduce that amount perhaps to a few percent, to the order of 4,000 TWh, all depending on other measures.
- Large grids: while storage spreads electrical load in time, large grids can spread electrical load in space. Significant grid extension is needed, ideally to span north-south for seasonal variation, and east-west for daily variations: a global grid. As the sun will always shine somewhere, production can be shifted around, e.g. Xlinks. From the top of my head, I reckon that about several circumferences of the Earth would be needed in new HVDC lines, perhaps of the order of 100,000km.
- Demand response: as mentioned in a dedicated article, the price and availability of any commodity cannot be constant in a dynamic world of production and consumption. A global electricity market is needed to have the energy price reflect availability with suitable predictions of a cost profile into the future businesses can plan for.
Now, to the essence of this article: what is the material requirement to build 160,000 TWh/annum WWS capacity (18TW), about 4,000 TWh storage and at least 100,000km HVDC lines for a global grid? Given the target year of 2050 as a first approximation, this capacity would have to be built over a period of 25 years from 2025 onwards; every year from 2025 up to 2050, 1/25 of the target capacity has to be built and added: 6,400 TWh/annum WWS capacity (730GW, China alone has added 216GW capacity in 2023), 160TWh storage and 4,000km HVDC lines. If the infrastructure is assumed to have conveniently a life time of 25 years, the same capacity has to be replaced every year beyond 2050.
Combining the power generation material requirement per as kg/MW from the IEA publication with the target capacity installation of 730GW per year one would get the following estimate if the capacity would be realised with solar only (which is likely to be the dominant energy source):
- copper: at 3t/MW the material required for solar PV is 2.2Mt/annum which is about 1/10 of the 22Mt world’s copper production in 2023; hence, the copper production needs to increase by 10% for solar panels alone (not talking into account other infrastructure).
- silicon: at 4t/MW the material required is 2.9Mt/annum which is about 33% of the 8.8Mt world’s silicon production in 2023; consequently, the silicon production needs to increase by 1/3.
These two examples are an indication of magnitude only; more minerals are needed for other forms of renewable energy generation, grid extension and energy storage. In due course I would like to model all that in more detail, but it can be expected that all requirements are of the same order of magnitude; they constitute an addition to the current world production for every mineral, usually in the low single digit Mts. These numbers do sound feasible, even though it must not be forgotten that a lot of earth has to be moved in mining depending on the ore grade (which might degrade).
On the upside, during this very transition, every year 1/25th of the current annual fossil fuel consumption (2.8Gt oil, 1.4Gt gas and 6.1Gt coal, in 2022) can be reduced by 112Mt oil, 56Mt gas and 244Mt coal every year assuming a pro-rate distribution, or around 400Mt/annum less fossil fuel every year (and the reduction in CO₂ emissions). Therefore, the amount of fossil fuel avoided is an order of magnitude more than the additional minerals that have to be mined. There are some secondary effects:
- Not having to mine and ship 112Mt oil, 56Mt gas and 244Mt coal less year after year until 2050 frees up quite some material, e.g. steel, copper and other metals not needed for ships, trains and pipelines; they can be decommissioned and recycled.
- Not having to process 112Mt oil, 56Mt gas and 244Mt coal less year after year until 2050 frees up energy and land: e.g. oil and gas fields, land for pipelines, refineries power plants and all the energy they use for their operation, and most importantly less CO₂ emitted.
- Of the 120Mt hydrogen produced every year more than half is used for hydro-cracking; 1/25 of these 60Mt = 2.4 Mt/annum every year will not be needed either, including the natural gas used to produce that grey hydrogen in the first place, even less gas.
- There may be even more positive side effects of not having to mine and process less and less fossil fuels every year, but there will be losers. It is therefore understandable that the fossil fuel industry is fighting tooth and nail to hold on its grip on all national economies.
In terms of energy, some fossil fuels are actually needed to enable the renewable energy transition, i.e. energy needed to mine the necessary materials and to produce the infrastructure. It is interesting to see, that this energy investment is a small portion of the current fossil fuel energy consumption as shown in the following plot. Once the build-out is completed, it can be expected that even further build-out and maintenance can be powered by renewable energy, too.
Perhaps an important note is due on personal transport which consumes a significant portion of energy (mostly in form of oil). Replacing the 1.5 billion ICE powered cars with battery electric vehicles over 25 years would required significant resources per year. More here:
Maintenance & Substitution
Suppose the energy transition started in earnest in 2025 and 18TW installed capacity has been achieved by 2050 with energy storage and a global grid in place as well as an energy demand response implemented to provide a decent final energy service across the world. Earlier it was assumed that the life time of that infrastructure was about 25 years; this means that by 2051, the first instalment built in 2025 will have to be replaced plus an allowance for growth in energy consumption. The good news is that not all resources have to be mined but the old infrastructure can be reused in the best case, or refurbished, or recycled. At a recycling rate of say 85%, only 15% of the original material needs to be mined for the next generation. And as recycling improves, less and less material needs to be mined to maintain the same amount of infrastructure. Only new capacity will have to be mined, probably even an order of magnitude less than the initial expenditure in the transition.
The situation might even get better: during the 25 year’s time of the transition, materials will have improved significantly, e.g. energy and power density for storage. Some materials that are expensive to mine may have been substituted, e.g. through different battery chemistry. Perhaps it will be common practice to use copper-clad aluminium for cables for transmissions and motor windings alike, perhaps sprinkled with some nano-carbon particles; who knows.
Conclusion
Critics of the renewable energy transition often question its feasibility or the impact on nature and the collateral damage for the planet. My counter question is: what else can we do? Fossil fuels will eventually run out while destroying the planet even more; they would postpone the inevitable phase-out; any delay will leave a sicker planet to deal with in the end. In my opinion the renewable energy transition is the best shot we have; a industrial civilisation powered by renewable energy from WWS is far from being 100% clean, but it is less dirty than a fossil fuel powered one and it needs fewer resources, by one or two orders of magnitude.
Caveat
The renewable energy transition is a necessary but not sufficient step; it may address CO₂ emissions and some aspect of climate change; there are other, perhaps even bigger issues to address: deforestation, aquifer depletion, soil erosion and degradation, ocean acidification, to name a few. Ignoring those may lead to a temporary green economy overshoot followed by a Seneca collapse. We must not suffer from carbon tunnel vision!
If we loose sight of all issues above, the renewable energy transition may turn into a Pyrrhic victory: a lot of effort resulting in eventual defeat by pulling the rug under out feet when the biophysical basis for our existence is destroyed. We need to stay within planetary boundaries, the planet’s limited waste assimilation and resource regeneration capabilities. Social changes will be necessary to allow truly sustainable life on a finite planet, perhaps by inducing a managed economic contraction during the renewable energy transition.
Footnotes
- Note that only WWS are considered as serious contenders for renewable energy generation, with solar on top. Mark Z. Jacobson has made a very good case in “No Miracles Needed: How Today’s Technology Can Save Our Climate and Clean Our Air” as has Michael Barnard in The Short List of Climate Actions That Will Work.
- Let’s allow for a bit of growth at the rate of the usual 2.5% increase of final energy consumption per year; over 25 years that would be 85% more in 2050, or about 100,000 TWh/annum which is equivalent to about 11.4TW continuous power, i.e. about half of the current world’s primary energy consumption.
