Simulations of superconductors accelerated enormously - research shows promising insights for high-tech materials
A team of mathematicians and physicists from Saarland University, Eindhoven University and the DLR Institute for Software Technology has developed a method that can significantly improve the simulation of the electrical properties of superconductors. The method is based on a formula published in 2022 by two members of the research team, Dr Andreas A. Buchheit (Saarland University) and Dr Torsten Keßler (Eindhoven University).
The formula describes so-called long-range interactions as they often occur in nature. For physicist Dr Benedikt Fauseweh from the DLR Institute for Software Technology, it was obvious to apply the formula to the behaviour of superconductors at the atomic level. In superconductors, pairs of electrons, known as Cooper pairs, form larger complexes known as Bose-Einstein condensates. The oscillations of these complexes are easier to calculate than the states of each individual electron.
Specifically, the interdisciplinary team developed a method to calculate and simulate the physical properties of a superconductor based on the mathematical formula developed by Andreas Buchheit. Torsten Keßler worked on the development of a code that translates the mathematical formula into simulation calculations. Dr Peter K. Schuhmacher, also a graduate of Saarland University and a research assistant at the DLR Institute for Software Technology, contributed solid-state physics expertise together with group leader Benedikt Fauseweh.
The method is not only interesting because it can be applied to superconductors - what is particularly exciting, even if initially surprising, is the exponentially decreasing computation time of the simulations as more and more simulated particles are included in a system. It is not the forces on individual particles that are calculated, but their behaviour in larger structural units. In practice, this means that calculations on the behaviour of electrons in superconductors no longer require the computing power of a supercomputer, but can run on ordinary PCs.
This means that simulations can be carried out faster and in greater numbers without the need for a supercomputer, which in the next step could give a huge boost to the development of functional materials for high-tech applications. Superconductors are already used in quantum computer hardware, magnetic resonance imaging and other high-power energy applications. The newly developed method allows the properties of superconductors to be verified in advance through simulations, which in turn means a significant reduction in the physical production and measurement in the development process.
Further information: