Don't let China hold rare-earths to ransom again
There's plenty to go around if we're clever
Opinion We're all screwed over this China rare earths thing, aren't we? After five weeks of withholding shipments, the Chinese seem to have let some of them go again, but if they can keep their precious cargo to ransom once, they can do it again, can't they?
The problem is, as those who stayed awake in chemistry class will know, that much of these whizzy modern electronic thingies that we all make our living with depend upon the lanthanides (those 15 elements in the funny little box at the bottom of the periodic table). Hard drives and windmills need neodymium for the magnets, CFL bulbs need terbium, MRIs need lutetium, CRTs need europium and so on – all the way down to metal halide bulbs needing the scandium that I deal with.
The further problem is that China currently produces 95 to 97 per cent of the world's supply of these metals – and if they're playing silly buggers with supply then WTF are we going to do?
While they say that they're restricting exports for environmental reasons (and to some extent they are, rare earth mining is a horrible, messy business), there's more than a suspicion that it's all part of industrial planning. If you cannot export the metals or the salts, but you can export things made from them, then all the work, the industry, of turning the salts and metals into products will migrate to China. This is along with all sorts of fat, highly paid manufacturing jobs – and it is not as if we really have a surplus of those at the moment anyway.
All of which would be a serious concern if it were not for two things. Firstly, rare earths (REs) aren't actually rare (nor are they "earths"). Secondly, technology isn't static. And there is a very plausible way in which we could obtain the REs we need for manufacturing.
Step one: As I've mentioned here before, it is possible to extract REs from that toxic red sludge that went splat over Hungary a few weeks back. We almost always find RE deposits associated with the titanium and zircon sands we already dig up for those respective metals. There's even a bit in the 150 million tonnes of coal flue dust we dump each year globally. This means that everyone and their mother who can float a junior mining company (which, given the current hysteria, really is just about everyone who can talk to a stockbroker without gagging) is hacking away at some part of the earth or other. Greenland, under frozen lakes in Northern Canada, parts of India. Someone even had a peek into an old mine in Derbyshire just to see what was there (erm, me actually). Then there's also various serious and sensible people, like Molycorp in California, Lynas at Mount Weld in Australia, and Arafura in Oz again: any one of those three alone can and will supply 20 per cent of current global demand once up-and-running.
Lutetium is used in the manufacture of magnetic
resonance imaging systems.
So while we may have a little bit of stuttering over the next couple of years, there's no long-term shortage of REs. Well, OK, there's no long-term shortage of the basic ores and concentrates. Here is where we get to the technology part of it all. There are three stages to the industry: the mining, as above, then there's the separation of the lanthanides, each from the other. Chemistry is all about the number of electrons in the outer ring of the atom and as the lanthanides all have that same number, we can't really use chemistry to separate them. We have to use some physical property. What we currently do, and China is really the only place left that does it (there's an outpost in Malaysia) is boil the ores up in lovely strong acids and then use multiple iterations of solvent extraction. To fully separate the ore can take thousands of iterations. This is a labour-intensive, dangerous and polluting activity: just the sort of thing that China is going to be good at.
A battle to keep using rare earths on our terms
Crystallised orange-pink "butterfly" twinned crystals of Monazite from Bolivia.
Even if we can get our ores and concentrates then, we're still going to be at the mercy of China: either we've got to use their plants to do this or we've got to build our own ($400m each!) and yet we'll still be competing with China on the cost of labour and environmental damages: not a battle we're going to win.
Well, yes, that's true, but only if we assume that technology is static: and that we would have to rebuild the industry using that 1960s technology of solvent extraction. What would make more sense is for us to use what we're good at (capital and, you guessed it, high tech!) to design a new system to find some other physical property which we could use as a separation mechanism. We've got a good candidate, the Kroll Process.
Step two: Turn all the metals into a halide (Kroll uses chlorides, but bromides, iodides and fluorides all work best/better for different mixtures of metals), and then a bit of fractional distillation and we're done. While no one has built a plant to do this sort of thing yet it has been tested: the RE industry has fallen behind on the implementation of new technologies in recent decades for, well, why care? China would feed us all we want at prices we were happy to pay. Now that that has changed, all we've got to do is pick up the research that has been done and do the engineering to make the actual machines to do it. And as for distillation: well, if that toothless guy in all the Burt Reynolds movies can do it, makin' the white lightnin' before Burt drives it through the police cordon, well, yes, we probably do have a firm grasp on the basics of the technique.
So it's entirely possible (and I know people other than myself who are working on it). This is how we can sort out the second part of the RE industry: and do it better and cheaper that the competition.
Step three: The third part of the industry is turning our now separated salts into the metals and actual products that we desire. Here, again, there's good news: we've got better methods available, ones that beat the traditional. The usual way to turn an RE oxide into an RE metal is to reduce it with calcium metal: this requires the consumption of Ca metal of course. A rough rule of thumb is that an RE metal will cost twice the price of the same purity of RE oxide. Part of this of course is the loss of the oxygen (La2O3 obviously loses the weight of the O3 when made into a metal) but a goodly part of it is simply the cost of the process.
Just as one example, the company Metalysis, up in Rotherham could help to expedite this. Sadly, given that they use a PR agency for journalistic-type questioning, I couldn't get any answers from them but then I've known about the process they're using for longer than they have. It's the FFC process from Cambridge University.
Instead of reducing with Ca metal, why not use a cathode/anode arrangement in molten salt? Should be cheaper and crucially, you'll have a lot less problem with residual gases (yes, a potentially major problem). Better and cheaper: just what we want as technology marches on. BTW, the reason they're in Rotherham is simply because, while the big steel mills have gone, that's still where the UK's weird metals industry is. It's what the economists call “clustering”: where are all the likely suppliers, where's the trained workforce, where are the lawyers, accountants, engineers, who understand the basics of your business? You stick a weird metals business in Rotherham/Sheffield just as you would a new pottery in Stoke on Trent (as happened a few weeks back), something with chlorine in Hartlepool or a new silicon design set-up in Cambridge or Silicon Valley.
Note that I'm not saying that all of this is easy, that we can just fall off a log and it'll be all right. I'm only saying that China really doesn't have a long-term lock on RE supplies. The seeds of how we get out of this half nelson are already there, ready to be nurtured. We can and will deal with competition and industrial dominance from a low-wage/high-pollution country or company by doing what all of you people around here do. We can use brains rather than brawn to design better, more mechanised, ways of ensuring our supply of what we need to keep industry on its feet.
The method eventually used to make this happen may not be the specific techniques I've outlined (although I'd suggest that it is the way to bet) but the general principle will hold. By being clever about how we mine and process rare earths we'll be able to free ourselves from reliance upon those who use simple brute force to do so. ®