The Long Read: The search for superconductivity

This article was first published in the June issue of the ESRFnews magazine.

There is no subject in materials science that has generated quite so much recent excitement – and controversy – as high-temperature superconductivity. In the last few years, the field has been set alight with results for a series of high-pressure materials that superconduct closer and closer to room temperature, the longstanding goal. At the same time, there have been false starts, confusion and – in the case of one high-profile scientist – alleged fraud.

Superconductivity is a hot topic for a good reason. Get it to work successfully in ambient conditions and it could well be a technological panacea – a route to lossless electricity distribution, and a major weapon against climate change. Yet the direction of much of the latest research has been almost guaranteed to end in turmoil. High-pressure synthesis often produces complex, mixed-phase samples, so that it is not always clear what is actually being studied. It also requires specialist experimental knowledge, meaning there is less scope for results to be reproduced independently. Worst of all, its very nature interferes with the key tests upon which claims of superconductivity can be made.

This is where the ESRF can help. Fed by the EBS source, beamlines such as ID15B and ID27 are able to extract crystallographic data at micron resolution, so that the makeup of samples is clear, no matter how heterogeneous they are. They can also do this repeatedly at an array of temperatures and pressures, to expose how complex changes in structure relate to the emergence of superconductivity. Best of all, the ESRF has its own specialist tools and expertise, so that even users without them can get involved.

“We want to have more people working in this field,” says Gaston Garbarino, the ESRF scientist in charge of ID15B. “And now we have tools that are extremely powerful, to perform full crystallographic analysis in all samples at extreme conditions. We can solve the crystal structures and, importantly, obtain results that are reproducible.”

Long journey

The history of superconductivity goes back to 1911, when the Dutch physicist Heike Kamerlingh Onnes discovered that at just four degrees above absolute zero the electrical resistance of mercury vanishes. Over the next few decades, the same phenomenon was found in several other metals, but it was not until nearly half a century after Kamerlingh Onnes’s original discovery that a trio of US physicists, John Bardeen, Leon Cooper, and John Robert Schrieffer, figured out why it occurred.

According to BCS theory, which was named after their initials, an electron travelling through a conductor attracts nearby positive atomic nuclei. This higher density of positive charge attracts another electron, and the two electrons become connected as a so-called Cooper pair. In this way, the electrons become immune from lattice vibrations, as each electron can balance any random kicks suffered by its partner. The result is that they travel without resistance, and the ordinary conductor becomes a “super” conductor.

Conventional superconductivity usually occurs at very low temperatures. That is because at higher temperatures, atomic lattices have enough energy to overpower the distortions of positive charge required for Cooper pairing. The maximum operating temperature of conventional superconductors is not set in stone, though, for if the atoms are less massive, it is easier for the conducting electrons to attract them, and thereby preserve the Cooper pairing when the lattice has more energy. On that basis, a lattice made of solid metallic hydrogen, the lightest element, theoretically ought to superconduct easily at room temperature, but scientists have been trying and failing to make that happen for decades.

A turning point came in the late 2000s, when theorists Neil Ashcroft and Roald Hoffmann predicted that hydrogen combined with other elements could be a realistic alternative to metallic hydrogen. Their prediction was proved correct in 2015, when a group led by Mikhail Eremets at the Max Planck Institute for Chemistry in Mainz, Germany, synthesised hydrogen sulfide (H₃S) at a pressure of 90 GPa, and claimed that it turned superconducting at a record 203 K (–83°C). Then came the synthesis of superhydrides, starting with FeH₅ at the ESRF (see Figure 1). Another superhydride based on lanthanum (LaH₁₀) turned out to be very promising. In 2019, a group led by Russell Hemley at the University of Illinois Chicago in the USA claimed that it turned into a superconductor at a pressure of around 200 GPa and a temperature of 260 K (–13°C).

Fig. 1: In 2017, a group led by Paul Loubeyre at the CEA and the Université Paris-Saclay in France synthesised and studied FeH₅, the first ever superhydride, at the ESRF. Their X-ray diffraction results allowed them to construct models of how increasing pressure allows more and more hydrogen to be packed into an iron hydride’s structure, until above 130 GPa and the formation of FeH₅, in which the hydrogen forms metallic “slabs” between quasi-cubic FeH₃ units [1]. Though it is debated, and awaiting experimental investigation, some theoretical studies have suggested that this buried metallic hydrogen within FeH₅ could make the material a superconductor.

The attainment of room-temperature superconductivity looked to be just a matter of time – and so it was, apparently, in two extraordinary claims by a group led by Ranga Dias at the University of Rochester in New York, USA. The first claim was for a compound made of carbon, sulfur and hydrogen (CSH), which was said to superconduct at a pressure of 267 GPa and a temperature of about 287 K (14°C), while the second was for lutetium hydride doped with nitrogen, which was said to work at just 1 GPa and 294 K (21°C). However, both papers were soon retracted due to concerns over data integrity, and this year an internal investigation at Rochester reportedly found Dias guilty of misconduct.

While the Dias case has been exceptional, his claims are not the only ones to have come under close scrutiny. The trouble is that the two major signatures of superconductivity – plummets in electrical resistivity and magnetic susceptibility – are both very difficult to observe in high-pressure experiments. If it follows a strong theoretical prediction, an experimental claim is much more persuasive, but that relies on a positive match between theoretical and experimental crystal structures. In the past, due to limitations in beam focusing, synchrotron X-ray diffraction in extreme conditions has not had the resolution to discern different structures within highly heterogeneous samples.

A new brilliance

That has changed with the ESRF–EBS. Thanks to the extremely low emittance and high brilliance of the upgraded light source, beamlines such as ID15B and the newly refurbished ID27 are able to focus X-ray beams down to less than a micrometre, making it possible to record high-quality diffraction patterns from even the smallest individual crystals at extreme pressures and temperatures.

The power of the single-crystal X-ray diffraction (SCXRD) technique was demonstrated earlier this year, when researchers at the University of Bayreuth came to the ESRF to study a promising system, yttrium and hydrogen, at pressures from 87 to 171 GPa, and discovered five previously unknown phases [2].

A recent study by the long-term user Leonid Dubrovinsky (above) and colleagues has shown just how complex high-pressure samples can be. Photo: ESRF/S.Candé

According to one of the team members, Dominique Laniel at the University of Edinburgh in the UK, the new phases were unanticipated by theory, and could very well not have ever been discovered at all – had the researchers not come to the ESRF. “Using ID27, ID15B and ID11, the very high flux and small beam sizes allowed us to identify and solve the crystal structure of even the tiniest crystallites,” he says.

Such beam properties have other benefits too, Laniel adds. When measuring electrical resistivity, a fine map of the sample cavity can expose whether there is actually a path from electrode to electrode comprised of the same phase, or whether there are heterogeneities. Moreover, the high flux opens up the unprecedented possibility – so long as the electron count of the other elements is not too high – of directly determining the position of hydrogen atoms in the structural model, which is usually very difficult with synchrotron data. 

ESRF SCXRD will help to clarify the structures of other complex hydride systems. Whatever the structure of a sample, however, superconductivity must still be established. Difficulties with the usual direct measurements of electrical resistivity, usually noise, can to some extent be assuaged with the use of indirect, alternating-current techniques. But measurements of magnetic susceptibility are inherently tough, as the diamond anvil cells (DACs) used to generate extreme pressures will always themselves have a residual magnetic response.

Back in 2016, Eremets and colleagues designed a novel way to circumvent this problem, in which they immersed a foil enriched with the tin isotope ¹¹⁹Sn in a sample of H₃S, before placing the lot in a DAC. Once at the correct pressure and temperature, the researchers subject the device to synchrotron Mössbauer spectroscopy at the ESRF’s former ID18 beamline.

Specific probe

Mössbauer spectroscopy is unique in that it can measure the nuclear resonance from a specific isotope – in this case the ¹¹⁹Sn – while ignoring any other source of magnetism. In this way, Eremets and colleagues could unambiguously demonstrate the expulsion of the magnetic field by the surrounding H₃S when it reached a relatively high temperature, and thereby prove its superconductivity beyond doubt [3].

The experiment is difficult to perform, but with the completed three-year transfer and upgrade of the Mössbauer beamline to the ESRF’s ID14 port, scientists have the benefit of an order-of-magnitude resolution boost, from 10 μm to 0.8 μm. “Now, instead of studying the entire sample in integral, one can map a sample, giving a two-dimensional profile of a superconducting state,” says Alexander Chumakov, the scientist in charge of ID14.

Mössbauer spectroscopy is not the only way to study magnetism within DACs. In the past few years, groups led by Jean-François Roch and Paul Loubeyre at the Université Paris-Saclay in France and others have been developing a method that exploits a defect in diamonds known as the nitrogen vacancy (NV) centre. The ability of an NV to fluoresce in a microwave field is dependent on its spin, the energy of which, in turn, depends on an external magnetic field. Therefore, performing microwave spectroscopy at a diamond anvil tip within a DAC allows the mapping of the contained sample’s magnetism.

In 2020, Loubeyre and colleagues demonstrated that a compact NV magnetic microscope could be installed on an XRD platform at the SOLEIL synchrotron in France. Now that ID27 is up and running, he plans to bring the system to the ESRF, to have the best in structure and magnetism detection in superconducting hydride research. “Both techniques have a similar micrometre space resolution,” he says.

Anna Pakhomova, beamline scientist at ID27, helps users to prepare high-pressure superconductivity experiments. Photo: ESRF/S.Candé

So far, all the superconducting hydrides investigated experimentally stabilize at high pressures, up to around 300 GPa, which is more or less the squeeze present at the centre of the Earth. Naturally, such pressures rule out practical applications, whatever the transition temperature. For that reason, much of the new focus is on compounds of hydrogen with more than one other element – so-called ternary or quaternary hydrides. Indeed, some of the very first studies of ternary hydrides have been performed at the ESRF. “The hope is that we can obtain a superhydride superconductor that is metastable at ambient pressure,” says Loubeyre. “A few have already been predicted by ab initio calculations.”

“Of course, all synchrotrons can perform measurements at high pressures and temperatures,” says Garbarino. “But here at the ESRF with the machine, the new beamlines, the new detectors and the on-hand sample environments – we’re in a privileged situation.” 

Text
Jon Cartwright

References
[1] C.M. Pépin et al., Science 357, 382 (2017); https://doi.org/10.1126/science.aan0961
[2] A. Aslandukova et al., Sci. Adv. 10, eadl5416 (2024); https://doi.org/10.1126/sciadv.adl5416
[3] I. Troyan et al., Science 351(6279), 1303 (2016); https://doi.org/10.1126/science.aac8176