A deep dive into material sciences from MIT Technology Review, August 21:
The story of neodymium reveals many of the challenges we’ll likely face across the supply chain in the coming century and beyond.
Leaving aside meteorites that strike Earth’s surface and spacecraft that get flung out of its orbit, the quantity of materials available on this planet isn’t really changing all that much.
That simple fact of our finite resources becomes clearer and more daunting as the pace of technological change advances and our society requires an ever wider array of material inputs to sustain it. So for nearly as long as we’ve systematically extracted these substances, we’ve been trying to predict how long they will be able to meet our demand. How much can we pump from a well, or wrest from a mine, before we need to reconsider what we’re building and how?
Those predictions have grown increasingly complicated. And now it’s also a matter of how much we can pull from manufactured and discarded objects. Can we recycle parts of that iPhone, or the guts of that massive wind turbine? How much of any given object can we recirculate into our churning technological economy?
Estimates of how much material we’ll have access to in the future tend to have a tricky, often implicit assumption at their center: that we’ll be making roughly the same products with the same materials as today. But technology moves quickly, and by the time we understand what we might need next, or develop a specialized system to mine or recycle it, the next generation of tech might render all our assumptions obsolete.
We’re in the middle of a potentially transformative moment. The materials we need to power our world are beginning to shift from fossil fuels to energy sources that don’t produce the greenhouse-gas emissions changing our climate. Metals discovered barely more than a century ago now underpin the technologies we’re relying on for cleaner energy, and not having enough of them could slow progress.
Take neodymium, one of the rare earth metals. While far from a household name, it’s a metal that humans have relied on for generations. Since the early 20th century, neodymium has been used to give decorative glass a purplish hue. Today, it’s used in cryogenic coolers to reach ultra-low temperatures needed for devices like superconductors and in high-powered magnets that power everything from smartphones to wind turbines.
Demand for neodymium-based magnets could outstrip supply in the coming decade. The longer-term prospects for the metal’s supply aren’t as dire, but a careful look at neodymium’s potential future reveals many of the challenges we’ll likely face across the supply chain for materials in the coming century and beyond.
Peak panic
Before we get into our material future, it’s important to point out just how hard it’s always been to make accurate predictions of this kind. Just look at our continuous theorizing about the supply of fossil fuels.
One version of the story, told frequently in economics classes, goes something like this: Given that there’s a limited supply of oil, at some point the world will run out of it. Before then, we should reach some maximum amount of oil extraction, and then production will start an irreversible decline. That high point is known as “peak oil.”
This idea has been traced back as far as the early 1900s, but one of the most famous analyses came from M. King Hubbert, who was a geologist at Shell. In a 1956 paper, Hubbert considered the total amount of oil (and other fossil fuels, like coal and natural gas) that geologists had identified on the planet. From the estimated supply and the amount the world had burned through, he predicted that oil production in the US would peak and begin declining between 1965 and 1970. The peak of world oil production, he predicted, would come a bit later, in 2000.
For a while, it looked as if Hubbert was right. US oil production increased until 1970, when it reached a dramatic peak. It then declined for decades afterward, until about 2010. But then advances in drilling and fracking techniques unlocked hard-to-reach reserves. Oil production skyrocketed in the US through the 2010s, and as of 2023, the country was producing more oil than ever before.
Peak-oil panic has long outlived Hubbert, but every time economists and geologists have predicted that we’ve reached, or are about to reach, the peak of oil production, they’ve missed the mark (so far).
Now there’s a new reason we might see fossil-fuel production actually peak and eventually fall off: the energy transition. That’s shorthand for the grand effort to shift away from energy sources that produce greenhouse gases and toward renewables and other low-carbon options.
Hubbert’s theory suggested that a fixed supply would force production to decline from a peak. But as the world wakes up to the dangers of climate change, and as low-carbon energy sources like wind, solar, and nuclear take off, we may wind up leaving some coal, oil, and natural gas in the ground. Simply put, production might head back down because of a lack of demand, not a lack of supply.
Those newly ascendant energy sources, though, are ironically a new source of “peak” panic. Solar panels, wind turbines, and batteries may not require fuel, but they do require a host of metals, including lithium, copper, steel, and rare earths like neodymium.
If we extract, process, use, and discard these metals, conceptually there must be some point in the future when we run out of them. And as the energy transition has gotten underway, plenty of forecasts have attempted to understand which metals we should worry about and when they might start to be depleted. But experts say that understanding the availability of resources in this sector is much more complicated than picking out a single future peak.
“The peak modeling thing is something that doesn’t really apply to metals,” says Simon Jowitt, director of the Center for Research in Economic Geology at the University of Nevada, Reno. It’s nearly impossible to understand whether we’ve reached a peak in production for any given material, or even whether those peaks can be predicted, as Jowitt said in a 2020 paper....
....MUCH MORE, he's just getting started.