An AI-kernel discovering redox-stable organic electrode materials for alkali-ion batteries

• Data-driven approaches have been revolutionizing materials science and materials discovery in the past years. Especially when coupled with other computational physics methods, it can be applied in complex high-throughput schemes to discover novel materials, for example for batteries. In this direction, this work presents a robust AI-driven framework, the AI-kernel, working as a platform to accelerate the discovery of novel organic-based materials for Li-, Na- and K-ion batteries. This platform was able to predict the open-circuit voltage of the respective battery and provide an initial assessment of the material’s redox stability. The kernel was employed to screen 45 million small molecules in the search for novel high-potential cathodes, resulting in a proposed shortlist of 3202, 689 and 702 novel compounds for Li-, Na- and K-ion batteries, respectively, when only considering the redox-stable candidates.

Subject headings

NATURVETENSKAP  -- Kemi -- Fysikalisk kemi (hsv//swe)

NATURAL SCIENCES  -- Chemical Sciences -- Physical Chemistry (hsv//eng)

Publication and Content Type

vet (subject category)

ovr (subject category)