On September 21, 2020, Artem Oganov and Dmitry Semenok from the Skolkovo Institute of Science and Technology presented the results of their research at a joint laboratory seminar led by Valery Shvetsov. The reports of the scientists were devoted to a very exciting topic: the possibilities of accurate prediction of the structure and properties of new substances and the latest advances in this field, in particular, the synthesis and properties of superconducting hydrides discovered in recent years. A separate agenda of the seminar was the discussion of joint investigations in this area using the FLNP instrumentation base and neutron scattering methods.

“What we are working on can be called the prediction of new materials using artificial intelligence,” said the world-renowned scientist Artem Oganov, winner of numerous prizes, whose research, according to many scientists, has long deserved the Nobel Prize. (Immediately after the seminar, the scientist received another high appraisal of his achievements in world science – a letter informing that he has been elected a Fellow of the American Physical Society.) The USPEX method developed by Artem Oganov and his team deals with self-learning evolutionary algorithms. This "evolution" of crystal structures is modeled on a computer in search of compounds that are optimal in terms of the required physical properties (for example, hardness) or thermodynamic stability.

“The method makes it possible to predict with great accuracy the structure, composition and properties of substances that have not yet been produced, including those that do not fit into the framework of classical chemistry. Generalization of the method to the case of low-dimensional materials allows us not only to simulate the surface structures of crystals, polymers, and nanoparticles, but also predict their stable compositions, which are often nontrivial ones. “Among the users of the patented method are many thousands of researchers, as well as large companies,” Oganov said.

A large number of materials simulated by the scientists were subsequently obtained experimentally. The problem of predicting crystal structures (for example, room-temperature superconducting compounds or other materials with unique properties) is considered one of the most important problems in theoretical crystallography, physics, chemistry, geology and mineralogy.

For example, recently, scientists have found a connection between the position of a chemical element in the Periodic Table and its ability to produce high-temperature superconducting hydrides. Dmitry Semenok spoke at the seminar about the properties and synthesis of such new superconductors discovered in recent years. Current investigations are performed for ternary metal hydride systems based on hydrides of lithium, lanthanum, yttrium, uranium, thorium, magnesium, and barium. Magnetic hydrides of europium, gadolinium, and samarium are also actively studied.

“There is a clear correlation between stability and superconductivity, and also between stability and the number of the period and group of an element. Superhydrides of heavy elements, lanthanides and actinides are the most stable. These elements form unique chemical compounds. Therefore, hydrides, deuterides and tritides of such elements as radium, actinium, protactinium, promethium, neptunium, plutonium, americium and curium are of great interest for high-pressure chemistry and superconductivity physics,” noted Dmitry Semenok.

The scientists are confident that in the coming years, hydride superconductivity will expand beyond the ultrahigh pressure range, which will make it possible to create electronic devices on their basis and, in the future, produce materials that conduct current without losses at room temperature and normal pressure.

Denis Kozlenko, Head of the Scientific and Experimental Department of Neutron Investigations of Condensed Matter (FLNP JINR) spoke about the possibilities that neutron scattering methods can provide for studies of new materials: "The specific features of neutron research techniques offer a number of advantages over synchrotron and x-ray methods in studying the atomic structure of materials containing light atoms, especially against the background of heavy elements. However, there are also a number of difficulties. For example, superconducting hydrides are synthesized under high pressure of several hundred thousand atmospheres. This explains the microscopic amount of the resulting material produced in a high-pressure cell (on the order of one hundredth of a cubic millimeter or even less). At the same time, the radiation fluxes of neutron sources are many orders of magnitude lower than those of synchrotron radiation sources, which makes research at high pressures a very difficult task.

Nevertheless, in recent years, high-pressure neutron diffraction techniques have been actively developed in the world's leading neutron centers, including FLNP JINR at the IBR-2 pulsed high-flux research reactor. We have recently designed and constructed a specialized DN-6 instrument for neutron diffraction studies of the atomic and magnetic structure of materials. The first successful experiments were carried out at a pressure of up to 350 000 atm. In the near future, we plan to further increase the working pressure range up to 500 000 atm, which is sufficient for studying the atomic structure of a number of superconducting hydrides. After the seminar, we discussed with our colleagues from Skoltech the prospects for joint research and possible options for starting experimental studies."

For his part, Dmitry Semenok added: “At present, at DN-6, in cooperation with the Frank Laboratory of Neutron Physics, we are already conducting calibration measurements of the components for the synthesis of metal polydeuterides, which should show the possibility of determining the structure of the deuterium sublattice in our compounds. During this year, we plan to choose the most convenient source of deuterium for synthesis in diamond cells and make a series of neutron diffraction measurements with lanthanide polydeuterides (La, Ce, Pr, Nd). Determination of the true structure of the deuterium/hydrogen sublattice in this new class of compounds is extremely important for understanding their superconducting properties. At present, this structure is known only from theoretical calculations."