A new AI tool speeds up the search for really novel materials

The University of Liverpool researchers have developed a collaborative artificial intelligence (AI) platform that minimizes the time and effort necessary to identify really novel materials. Artificial intelligence is a broad field of computer science concerned with creating intelligent computers that can accomplish activities that would normally need human intellect.

The new method, which was published in the journal Nature Communications, has already led to the identification of four novel materials, including a new family of solid-state lithium conductors. Solid electrolytes like this will be crucial in the development of solid-state batteries that will provide greater range and safety for electric cars. More intriguing materials are in the works.

The technology combines artificial intelligence (AI) and human expertise to prioritize uncharted chemical areas where novel useful materials are most likely to be discovered. Machine learning and deep learning are two sub-fields of AI that are commonly cited in combination with artificial intelligence.

Because there is an unlimited range of potential materials accessible by mixing all of the elements in the periodic table, and it is unknown where new materials exist, discovering new functional materials is a high-risk, difficult, and frequently protracted process.

To date, a common and powerful approach has been to design new materials by close analogy with existing ones, but this often leads to materials that are similar to ones we already have.

Professor Matt Rosseinsky

A team of academics from the University of Liverpool’s Department of Chemistry and Materials Innovation Factory, lead by Professor Matt Rosseinsky, created the new AI tool to solve this problem.

The tool investigates the links between known materials on a scale that humans can’t match. These connections are used to find and rank combinations of elements that are likely to generate new materials numerically.

Scientists utilize the rankings to direct targeted research of the vast unknown chemical space, making experimental investigation significantly more efficient. The ultimate judgments are made by those scientists, who are aided by the AI’s unique insight.

Artificial Super Intelligence (ASI), sometimes known as super-intellect, would outperform the human brain’s intelligence and abilities. While strong AI is currently purely theoretical, with no practical applications to yet, that doesn’t mean AI researchers aren’t working on it.

The lead author of the paper Professor Matt Rosseinsky said: “To date, a common and powerful approach has been to design new materials by close analogy with existing ones, but this often leads to materials that are similar to ones we already have.”

“We, therefore, need new tools that reduce the time and effort required to discover truly new materials, such as the one developed here that combines artificial intelligence and human intelligence to get the best of both.”

This collaborative technique combines computers’ capacity to look at the links between hundreds of thousands of known materials, a scale unreachable by humans, with human researchers’ specialist knowledge and critical thinking, resulting in creative breakthroughs. This technique is one of several collaborative artificial intelligence technologies that scientists will likely benefit from in the future.

Our ability to design and make materials with targeted functions, such as better solar absorbers for better solar panels or superior battery materials for longer-range electric cars, or replacing existing materials with less toxic or scarce elements, limits society’s ability to solve global challenges such as energy and sustainability.

These novel materials assist society by enabling new technologies to address global concerns, as well as revealing new scientific phenomena and knowledge.

The materials in lithium-ion batteries, which were developed in the 1980s, enable all modern portable electronics, emphasizing how a single materials class can transform how we live: defining accelerated routes to new materials will open up previously unimaginable technological possibilities for our future.

Topic : Article