Categories: Chemistry in History |  Comments
Guest post by Andy Extance
In 1935, two scientists working at Princeton University in the US made a prediction that chemists and physicists are still striving to make a reality. Writing in the Journal of Chemical Physics, Eugene Wigner and Hillard Bell Huntington foresaw a strange-sounding form of hydrogen. Rather than its familiar molecular form, H2, with two atoms joined by a covalent bond, at high pressures hydrogen might switch to a metallic lattice of individual atoms.
In researching my feature on high-pressure chemistry for the August issue of Chemistry World, this prediction came up regularly. Natalia Dubrovinskaia from the University of Bayreuth, Germany called it a point of honour, ‘like Fermat’s Last Theorem was for mathematicians’. That’s in part because, on top of the original predictions, later calculations suggested that metallic hydrogen should become a superconductor with relatively little cooling.
Today, research finally seems to be nearing this long-elusive goal. Yet while Wigner and Huntington are regularly cited in this work, their prediction doesn’t get the respect it deserves. Or at least that’s what Artem R. Oganov, from Stony Brook University in New York, US, and the Skolkovo Institute of Science and Technology in Moscow, Russia, felt. Oganov discussed the magnitude of the 1935 metallic hydrogen achievement with me in detail during our interview for my feature. He told the story so well, it deserves to be shared – and so we’re sharing it here. Oganov starts by explaining how great an achievement their paper was.
‘Wigner and Huntington used Wigner’s method for computing the electronic structure of solids, the classical cell method. It’s very approximate, one has to admit. It’s a very crude quantum mechanical method, but for that time, 1935, this was an absolute breakthrough. This was just a few years after the formulation of Schrodinger’s equation, before people even had computers. It’s a real tour-de-force. Now we look at that and smile, respectfully, at least I do.’
‘They couldn’t do accurate quantum mechanical calculations, they took very rough approximations. They could not predict crystal structures. If you cannot predict crystal structures, what kind of information can you provide about high pressure phases of hydrogen? Now we wouldn’t even try to do it. We’d start by trying to predict crystal structures. But at that time the apparatus for doing so was absent. They were not even the least dismayed by that.’
‘They said, OK, let’s try with a body-centered cubic structure. Why? Because maybe this is the structure that occurs under pressure. Alkali metals take this structure, sodium and so on. Like them, hydrogen also has one valence electron. So, for this simple structure they solved the quantum mechanical problem, they compared the estimated energy of molecular hydrogen, and found that there should be a phase transition from molecular hydrogen to the metallic monoatomic phase of hydrogen.’
‘They, to their credit, have recognised that the calculation that they are doing is extremely crude. They said that they estimate that metallisation of hydrogen will occur at pressures not lower than 25 GPa. It could be anything else, it could be 100GPa, it could be 400GPa. When they wrote that work, people thought that pressures of 25GPa sounded astronomically high. Now probably hundreds of laboratories around the world can do that. But at the time people said “We probably will never be able to reach that pressure”.’
‘Over time experimental techniques improved, and so did computational. They reached 25GPa and found that very little change happens in hydrogen. People were very bitterly disappointed, and some started blaming Wigner and Huntington. Usually in papers their result is quoted incorrectly. People say “Wigner and Huntington predicted that at 25GPa hydrogen metallises” but this is not right. Wigner and Huntington predicted metallisation above 25GPa and indeed it is above 25GPa. It’s maybe 20 times higher.’
‘Then there were many other predictions which placed metallisation at higher pressures, but experiments eventually reached those pressures and did not find metallisation. And again people were very disappointed. It looks like now we’re coming to a more-or-less convergent picture that metallisation will occur at pressures somewhere above 275GPa. I personally feel that at that pressure it’s pure semiconducting, with genuine metallisation occurring probably around 400GPa or maybe 500GPa.’
While it’s difficult, those pressures are now potentially achievable, and so the debate may soon be finally put to rest. For context on why Oganov feels the way he does about the status of metallic hydrogen research, see Matt Gunther’s article from the May issue of Chemistry World. And, if you have access, my feature delves into the many other bizarre chemical phenomena that occur at high pressure.