The stuff problem
How much mined material will we need to build a 100-per-cent renewable world? Danny Chivers works it out.
The problem with wind turbines, solar panels, ground-source heat pumps and electric cars is that they’re all made of stuff. When people like me make grand announcements (and interactive infographics) explaining how we don’t need to burn fossil fuels because fairly shared renewable energy could give everyone on the planet a good quality of life, this is the bit of the story that often gets missed out. We can’t just pull all this sustainable technology out of the air – it’s made from annoyingly solid materials that need to come from somewhere.
So how much material would we need to transition to a 100-per-cent renewable world? For my new NoNonsense book, Renewable Energy: cleaner, fairer ways to power the planet, I realized I needed to find an answer to this question. It’s irresponsible to advocate a renewably powered planet without being open and honest about what the real-world impacts of such a transition might be.
In this online article, I make a stab at coming up with an answer – but first I need to lay down a quick proviso. All the numbers in this piece are rough, ball-park figures, that simply aim to give us a sense of the scale of materials we’re talking about. Nothing in this piece is meant to be a vision of the ‘correct’ way to build a 100-per-cent renewably powered world. There is no single path to a clean-energy future; we need a democratic energy transition led by a mass global movement creating solutions to suit people’s specific communities and situations, not some kind of top-down model imposed from above. This article just presents one scenario, with the sole aim of helping us to understand the challenge.
How much aluminium, copper, iron and cement would we need?
In October 2014, a joint academic study between researchers from Norway, the US, the Netherlands, Chile and China made an assessment of the main materials needed to build renewable generators: steel, concrete, copper and aluminium.1 They looked at the materials required for renewables to provide 40 per cent of global energy use by 2050, and concluded that this would be feasible within current rates of global resource use.
I’ve taken their figures and attempted to go a step further. How much material would be needed for a transition to a 100-per-cent renewable world, where everyone had access to 13,000 KWh of energy per year? (This is one estimate of the amount of energy needed for an eco-efficient version of a “modern” lifestyle – it’s less than half of the energy currently used per person in the EU). For this calculation, I assumed that 3,000 KWh per person would be provided by non-electric generation (rooftop solar heat collectors, heat pumps, geothermal heat, waste gas, maybe energy crops). I then assumed we would build the following generation sources to provide 10,000 KWh of electricity for nine billion people (these totals all fit comfortably within realistic estimates of the amount that could be sustainably generated from these sources using current technology)2:
If we transitioned to 100-per-cent renewable energy by 2040 – thus giving ourselves a decent chance at avoiding runaway climate change – we would need the materials laid out in the table below to build and maintain this amount of generation.
The table shows that this is a serious undertaking and that we’re cutting things rather fine – particularly with regard to aluminium and copper – but also that the amounts of material required fall within current production totals and so are certainly possible to obtain.3 Once these materials have been extracted once, the metals can theoretically be recycled indefinitely, meaning that we’re talking about a short-term burst of new material use to get everything installed, from which point onwards we’ll be able to get most of what we need from recycling the old turbines, panels and so on.
What if we had to do more mining to achieve this?
Ideally, we would get these materials by diverting production away from less socially useful consumer junk into the sustainable technology that we actually need, so there’d be no net increase in mining. However, what if that isn’t possible? What if our shift to a renewable future requires us to pull an extra four billion tonnes of material out of the ground over the next 25 years? There is no such thing as zero-impact mining; it is one of the most notoriously destructive, poisonous and corrupt industries in the world.
Let’s look at this worst-case scenario. The final amount of raw material produced is just the tip of the extraction iceberg; every tonne of metal or cement requires many more tonnes of rock and ore to be hauled out of the ground in the mining and production process. Making four billion tonnes of copper, aluminium, iron and cement will require 50 billion tonnes of real-life extraction.
However, we need to look at the other side of the equation too. Phasing out fossil fuels over the next 25 years will mean a huge reduction in the amount of oil, coal and gas extracted over that period. Based on IEA projections, shifting to 100-per-cent renewables would avoid the need for around 230 billion tonnes of fossil fuels between now and 2040. Coal, tar sands and heavy oil, like metals, require the extraction of large amounts of extra rock and earth; when all this is added in, our transition would prevent 1,850 billion tonnes of fossil-related extraction up to 2040.
So even if we needed the full 50 billion tonnes of new extraction to build our new electricity generators, we’d still be creating a large reduction in the amount of destructive extractive industry taking place worldwide. We might be able to reduce the damage further by recycling the materials from all the oil and gas rigs, pipelines, and fossil-fuel power stations that we’ll no longer need, providing raw materials for our sustainable alternatives.4
Rare Earth Elements
As well as the high-volume materials, there are also a number of rarer minerals (known as ‘Rare Earth Elements’ or REEs) that we need to watch out for. These include indium, gallium and tellurium, which are used as semi-conductors in some types of solar panel. These metals have important uses in other technologies too (for example, indium is used in solder and flat-screen technologies, and gallium is used in computing components and LEDs), and are relatively rare; this means that there is likely to be a limit to how many solar panels can be made with these particular semiconductors. Luckily, this only affects certain specific designs of panel (not including our familiar black silicon panels),5 and so shouldn’t prevent us from rolling out the amount of solar power we need.
There’s a similar issue with dysprosium, which is used for making magnets in many modern wind turbines. The rarity of this element is likely to constrain the number of turbines that can be made this way. There are, however, alternative ways of making magnets without dysprosium, and so this shouldn’t act as a serious constraint either.
What about the materials needed for the rest of our sustainable transition? A typical ground-source heat pump weighs around 200 kg; air-source units tend to be a little lighter.6 If 200-kg heat pumps were installed – slightly excessively – in three billion buildings around the world, that would require 0.6 billion tonnes of materials. If we also installed three billion solar water heaters, weighing 100 kg each, that would give us another 0.3 billion tonnes. So the rest of our power generation would come in at less than a billion tonnes of material. Even if this required 10 times as much extracted material, bringing our total (when added to electricity generation, above) up to 60 billion tonnes, it would still leave us with a huge material saving thanks to the 1,850 billion tonnes of fossil-fuel extraction that we’re preventing.
A worst-case scenario would involve having enough storage facilities and back-up generators to support our wind and photovoltaic solar generation, making sure that the lights stay on even when the sun sets and the wind drops. Assuming that these facilities required similar quantities of material per KWh as a gas-fired power station, this would add another 0.4 billion tonnes of material, and three billion tonnes of mining.
What about electric vehicles? Well, there are currently more than a billion road vehicles in the world. Currently we are on a path of pure expansion, with the number of cars on the road expected to double in the next 20 years. In 2014, for example, the world manufactured over 80 million new cars, buses and trucks.7
A billion vehicles are probably enough. If distributed more fairly around the world, with the priority on buses and car-sharing schemes, they are likely to give us all the mobility we need. Consider, for example, that cities considered to be well served with buses such as London, Rio and Hong Kong contain between 650 and 1,700 buses per million inhabitants.8 If we decided to err on the side of caution and provide 2,000 buses per million people globally, that would require around 20 million buses. Add in a few billion bicycles (most of which probably already exist) and we’ll have sorted out most people’s daily transport needs. The remaining 980 million vehicles should then be enough to plug the global transport gaps as shared cars, taxis, and trucks for freight.
So what if, instead of doubling the number of vehicles globally in the next 20 years, we instead gradually replaced the existing fleet with renewably powered vehicles? This would require no increase in manufacturing overall, just a change in what we manufactured and where. We could even provide a large amount of the necessary raw materials by recycling old fossil-powered vehicles at the same rate as clean-energy vehicles emerge from the factories.
The point is that a genuine transition to a sustainable transport system wouldn’t require an increase in manufacturing, but a redirection of existing manufacturing. This would need a significant shift from our current position though; out of the 80-90 million vehicles currently manufactured per year, only 200,000-300,000 are fully electric.9
Of course, we should check in with the worst-case scenario too: what if we ended up manufacturing a billion renewably powered vehicles in a way that added to global material use? Well, a typical car weighs around 1.5 tonnes; trucks and buses, though smaller in number, are larger, so let’s be cautious and say an average vehicle weighs two tonnes. This would add two billion tonnes onto our material demand, and thus around 20 billion tonnes onto our grand extraction total, bringing it up 80 billion tonnes. This is still far less than the 1,850 billion tonnes of fossil-fuel extraction that we would prevent.
In addition, there are certain elements used in electric cars that we need to be particularly aware of. One of them is copper – a typical electric car contains around 60 kg of copper, compared with 20 kg in a fossil-fuelled car. If we build a billion of these vehicles over 20 years, we’ll need 0.003 billion tonnes of copper per year. This compares with 0.002 billion tonnes per year that’s already being used for manufacturing conventional cars; if we succeed in phasing out fossil-fuel car production and only building clean-energy vehicles, then we’ll only be increasing overall copper demand by 0.001 billion tonnes per year – much of which should be obtainable from recycling old vehicles. In the worst-case scenario, with no recycling, mining the extra copper needed for a billion electric cars would add another nine billion tonnes of mining onto our extraction total,10 still leaving us way below the fossil-fuelled business-as-usual amount.
Rare elements in electric cars
A recent study by Delucchi et al into the material components of electric cars identified a number of rare elements that could potentially limit their growth.11 The first is neodymium, an element used in electric motors and also in the generators of many wind turbines. Maintaining a billion electric vehicles and obtaining a quarter of our energy from wind turbines could exhaust global neodymium supplies in less than 100 years; however, there are alternative ways of building motors and generators without neodymium, which means that this needn’t be a constraining factor.
The second group of potentially problematic elements are rare metals and minerals such as lithium, cobalt, nickel, manganese, phosphorous and titanium. These are used in the rechargeable batteries in electric cars, and potentially in other energy storage systems too. All of these batteries use lithium, combined with other elements. The Delucchi et al study found that cobalt and nickel reserves, in particular, could be rapidly depleted by a mass rollout of electric cars using batteries containing these elements. Using titanium-based batteries would be unlikely to exhaust global titanium reserves but would involve multiplying the rate of extraction of this metal by more than 100 times, which might create practical difficulties. Fortunately, manganese, iron and phosphorous are much more abundant, and so we should be able to make the batteries we need without relying on cobalt, nickel or titanium.
Lithium itself is more likely to be a problem. The Delucchi et al study suggests that a mass rollout of electric cars could exhaust proven lithium reserves within 100 years – not counting the extra lithium that might be needed for improved electricity storage systems in homes and communities. This means that humanity should be able to obtain enough lithium to make the initial transition to an electrified transport system, but to maintain it beyond the second half of the century we’ll need to either get very good at recycling it, find more supplies, or find safe and affordable ways to extract lithium from the oceans (where it is abundant, but dispersed).
Avoiding a colonialist mindset
There’s another serious issue here. This is one of those moments where it’s easy to slip accidentally into a colonialist mindset, when referring casually to ‘reserves’ of minerals ‘available’ to the world. Whether or not those materials are dug out of the ground should not be a decision for someone like me, a white guy typing on a computer in Europe; it should be up to the communities that live in the area concerned and would be affected by the extraction. Although the quantities of lithium required for everyone in the world to have decent access to electrified transport are relatively small when compared to high-volume mined materials like iron or coal, the necessary mines would no doubt loom large in their local landscape. Most of the world’s known lithium reserves are located in Bolivia and Chile. These are real places, inhabited by real people – including Indigenous peoples whose lives, livelihoods and culture are intimately bound up with the land they live on. Will it be possible to obtain enough lithium for an electrified world without trampling over the rights of local communities? If not, then we’ll need to find a different path to our renewably powered future.
Renewable Energy: cleaner, fairer ways to power the planet by Danny Chivers is published by New Internationalist and available at nin.tl/nononsenserenewables
- Hertwich et al, ‘Integrated life-cycle assessment of electricity-supply scenarios confirms global environmental benefit of low-carbon technologies’, PNAS Sep 2014. ↩
- The 2014 international study only provides material usage for solar, wind and hydro power. For wave and tidal, I have assumed the same material use per TWh as for offshore wind; for geothermal, I have assumed the same material use per TWh as for a typical gas power station. ↩
- At current extraction rates, there are more than enough of all these materials in proven reserves to last for decades to come; once extracted, the metals can in theory be recycled indefinitely. ↩
- This recycling process would be unlikely to provide more than a few percent of the raw materials required, however, because wind and solar power require far more building material per MWh than oil, gas or coal power. See greet.es.anl.gov/publication-oil-gas-prod-infra ↩
- Hertwich et al (2014) - Annex ↩
- nin.tl/groundsourcepumps ↩
- Wall Street Journal, nin.tl/carsalesrising ↩
- nin.tl/brazilbuses ↩
- Forbes, nin.tl/electriccars2014 ↩
- Copper mining is particularly wasteful, with 310 tonnes of rock extracted for every tonne of metal produced. ↩
- MA Delucchi, C Yang, AF Burke, JM Ogden, K Kurani, J Kessler and D Sperling, ‘An assessment of electric vehicles’, Phi Trans R Soc A 2014 372, 2013. ↩