Researchers from NUST MISIS and Kazan Federal University have presented optimized quantum algorithms that allow simulating the behavior of electrons in molecules dozens of times faster. This opens up new opportunities for the development of drugs, catalysts, batteries, and other materials. The results of the work are published in the journal Quantum Reports.
The researchers proposed a modification of the variational quantum algorithm, which allows reducing the number of operations while maintaining the accuracy of calculations. The scientists excluded electrons that do not affect chemical properties from the simulation, reduced the number of qubits, and simplified quantum circuits. This made it possible to reduce the number of operations from 600 thousand to 12 thousand without losing accuracy.
In particular, the fact that with the help of physically justified approximations, it is possible to significantly (by one or two orders of magnitude) reduce the number of operations that carry a high probability of error, without losing accuracy.
The technology has already been tested on simple molecules, such as methylamine and formic acid, which are used in the pharmaceutical, food and textile industries. In the future, the algorithm can be adapted for:
- Searching for new drugs – accelerating the screening of candidate molecules.
- Development of catalysts – modeling reactions for the chemical industry.
- Creation of materials for batteries – calculation of the properties of new compounds.
The development is promising for the use of quantum simulations in molecular design tasks, especially with organic compounds important for pharmaceuticals and materials science.
Modern chemistry and materials science face problems that require accurate calculation of the electronic structure of molecules. However, classical supercomputers do not always cope with such calculations due to their high complexity. Quantum computers are theoretically capable of solving such problems more efficiently, but their practical application is still limited due to errors and high resource intensity.
The research results were published in the scientific journal Quantum Reports. The study was conducted in the context of the strategic technological project of NUST MISIS "Quantum Internet", which is being implemented as part of the "Priority-2030" program.
Read more materials on the topic:
Generative AI from Russian scientists can speed up DNA analysis