Superconductivity

“It is better to keep the optimism on a low level,” said J Georg Bednorz in 1988. He was speaking to Physics World a year after he and Alex Müller, his colleague at the IBM Zurich Research Laboratory in Switzerland, had been swiftly awarded the Nobel Prize for Physics for the discovery of the world’s first “high-temperature” superconductor. The breakthrough had set alight the condensed-matter community, with many physicists seeing it as the gateway to amazing applications such as lossless transmission lines, high-performance magnets and levitating trains. In the US, Time magazine was heralding “the superconductivity revolution”. Bednorz, on the other hand, was resisting the urge to “promise too much”.

Thirty years on, such restraint seems to have been justified. The field of high-temperature superconductivity has not exploded with applications. Nor, however, has it petered out into irrelevance. Patience has been rewarded by steady improvements in instrumentation and computation, to the point that experiments and theoretical calculations once thought impossible are now routine. And further breakthroughs have come. In the early years, these concerned the notching-up of transition temperatures and the establishment of theoretical approaches. More recently, they have related to the discoveries of entirely new families of superconductor. All in all, the field is “astonishingly different”, says J C Séamus Davis, a physicist at Cornell University in the US. “If we put ourselves back in 1988, what goes on now is inconceivably different to what was happening at that time.”

Disappearing resistance

Superconductivity has captured the imaginations of scientists and the public alike for a very long time. It was in 1911 that the Dutch physicist Heike Kamerlingh Onnes discovered that the electrical resistance of mercury suddenly disappears beneath a temperature of 4.2 K. The development in the 1920s of quantum mechanics, which explained atomic structure, did not, however, immediately provide an answer for this phenomenon in mercury and, later, other metals. Granted, the loosely bound outer electrons of metal atoms can carry a current, but it was believed that at any temperature there always ought to be random thermal fluctuations that keep the current in check. Only in 1957 did three theorists at the University of Illinois Urbana-Champaign in the US – John Bardeen, Leon Cooper and Robert Schrieffer – describe how an electron can deform the atomic lattice through which it moves, thereby pairing with a neighbouring electron. Being paired allows all the electrons in a superconductor to move as a single cohort, known as a condensate, prevailing over thermal fluctuations with ease.

Fast-forward to the observations of Bednorz, Müller and those who followed in their footsteps, however, and BCS theory (named after the surnames of its originators) was next to useless. For starters, BCS theory is widely interpreted to limit superconductivity to temperatures less than around 30 K, whereas the superconductors of Bednorz and others were, by 1987, transitioning at 93 K – warm enough to be cooled by liquid nitrogen rather than liquid helium. Moreover, these superconductors were not metals at all but insulators made of copper oxides, or “cuprates”.

The early days of high-temperature superconductivity were mired in inflated language and competitiveness, as physicists strove to rewrite what they thought they knew about condensed matter while often deriding the attempts of rivals who sought to do likewise. Because of the hype surrounding the new field, a 1987 meeting of the American Physical Society (APS) was dubbed the “Woodstock of physics”. Like the original pop-music festival, there were “false starts, confusion and eventually some squabbling” according to a recent account by Reinhardt Schuhmann, an editor at the APS journal Physical Review Letters.

Gradually things settled down. Although physicists did not (and still cannot) settle on an overarching theory of high-temperature superconductivity, they did assemble themselves in two camps: one believing that electrons paired through chemical valence bonds become mobile thanks to doping; and another believing that doping allows for a ripple in an antiferromagnetic alignments of electrons, drawing the electrons together.

A shared sense of mission has been helped by the ability to probe experimentally the electronic and magnetic behaviour of materials with much greater precision, using tools such as resonant inelastic X-ray scattering, angle-resolved photoemission spectroscopy and atomically resolved spectroscopic imaging, to name but three. “These techniques have helped to ‘clean up’ the field: everybody agrees on the landscape of experimental facts,” says Kees van der Beek from the École Polytechnique in Palaiseau, France.

The heat is on

The transition temperatures of the cuprates rose sharply in the late 1980s, then more slowly, to a peak of 135 K (–138 °C) at ambient pressure in the mid-1990s. There ended headline news for high-temperature superconductors until 2008, when Hideo Hosono and colleagues at the Tokyo Institute of Technology in Japan discovered another class of high-temperature superconductors based on iron and arsenic. Known as iron pnictides, these materials have exhibited only relatively low transition temperatures of up to 75 K. Nevertheless, they raised hopes that yet more classes of superconductor could be found, with transitions even closer to room temperature – the field’s primary goal.

As indeed they have been. In 2015 Mikhail Eremets, Alexander Drozdov and colleagues at the Max Planck Institute for Chemistry in Germany discovered that when they squeezed hydrogen sulphide – a substance commonly known for its stench of bad eggs – to pressures of 150 GPa, and cooled it below 203 K (–70 °C), it lost any trace of electrical resistance. The result smashed the previous record for the highest superconducting transition temperature (at high pressure) by 39 K. Even more surprisingly, it threw attempts to untangle the underlying physics of superconductivity up in the air once again, since hydrogen sulphide is conjectured to not be an unconventional superconductor like the cuprates or iron pnictides, but a conventional superconductor in the mould of BCS theory.

In a field that is full of surprises, it seems harder than anywhere else to predict what future developments are in store for high-temperature superconductivity. What is certain is that the recent discoveries of the iron pnictides and hydrogen sulphide have renewed hopes of future breakthroughs. “I am utterly and intimately convinced that room-temperature superconductivity will be discovered,” says van der Beek.

Quietly emerging

Even aside from that room-temperature goal, applications are emerging, if a bit later and with less fanfare than some had expected. High-temperature superconductors have reportedly been employed by the US military in the form of electric motors for propulsion, and as superconducting quantum-interference devices to detect very weak signals from submarines. Energy companies have begun to install them within fault limiters, which restrict electrical currents in the event of power faults; higher superconducting transmission temperatures are important here, because fault limiters heat up during their operation. And cities have begun to see high-temperature superconductors employed as power transmission cables. One of the first, in 2001, was Copenhagen in Denmark, in which a 30 m-long cable was given responsibility for the distribution of power to some 150,000 residents.

In April 2014 a section of power cable based on a cuprate was switched on in the German city of Essen. One kilometre long, it is thought to be the longest high-temperature superconducting cable ever laid in a power grid, designed to eliminate the energy loss associated with regular copper cables. Present at the ceremony was Bednorz, who could finally witness the patience of his approach bearing fruit. “What started to be a dream in the 1980s,” he said, “is now becoming a reality.”