Supply insecurity, high prices, technical optimization, regulatory bans: there are many reasons why companies must replace certain raw or base materials. The search for suitable alternatives is time-consuming and often ends without results. A research team from the Fraunhofer IPA has now developed an AI-assisted tool for material substitution.
A particularly current example of the need for material substitution is the group of per- and polyfluorinated alkyl substances (PFAS). These chemicals are increasingly under scrutiny by regulatory authorities due to their extreme persistence, potential health risks, and resistance to degradation. PFAS also pose a significant challenge in the industrial valve sector – for instance, as components in seals, coatings, or membranes. A possible ban or restriction on their use necessitates the development and evaluation of suitable replacement substances. This is exactly where the AI-assisted tool comes in, helping to identify and assess functional and less problematic regulatory alternatives.
Cobalt is used in lithium-ion batteries for electric vehicles, making it a key material for the energy transition. However, for several reasons, cobalt is classified as a critical raw material.
Cobalt is rare. Its proportion in the Earth's crust is just 0.004 percent.
Considering the specific use case in the company
The world's known cobalt reserves are estimated at 7.2 million tons.
Over half of them—around four million tons—are located in the Democratic Republic of the Congo.
Working conditions in the mines of this politically unstable Central African country are often poor, and the environmental damage from ore extraction is significant.
Whether due to supply insecurity, high prices on the global market, ethical concerns, regulatory bans, or product innovations with improved material properties: companies have many reasons to look for alternative materials.
“There are databases that product developers can use for research. But these often fail to deliver usable results because they don’t take the company’s specific application into account,” says Charlotte Schmidt from the research team Sustainability and Material Compliance Management at the Fraunhofer Institute for Manufacturing Engineering and Automation IPA.
To make the search easier and deliver more relevant results, Schmidt and two colleagues developed an AI-assisted tool for material substitution.
Via an input mask, users must first specify the material or raw substance they wish to replace, along with the required properties of the alternative material and contextual information about the intended use.
Material Substitution: AI Research Using a Database
This is followed by an AI-driven search that scours the “Semantic Scholar” database based on the specific user data and requirements.
By comparing the user input with the information available in the database, the AI identifies suitable alternative materials.
The AI component for material substitution is just one of several building blocks the researchers use to support companies in the search for alternative raw materials, base substances, or chemicals.
Once the AI has completed its task, the suggested substitutes and the original materials are subjected to a comprehensive evaluation that takes into account legal, ecological, and social aspects, as well as supply security.
In close cooperation with the respective company, the researchers then examine how well the proposed materials meet the specific requirements.
At the end of the process, a report is compiled. It presents the most suitable substitutes along with the evaluation of the various criteria.
This provides companies with a solid decision-making foundation.
Initial tests show: AI integration is promising
As an alternative to cobalt, the AI-assisted material substitution tool proposes, among others, iron.
“It’s not exactly a new insight that lithium iron phosphate can be used instead of lithium nickel manganese cobalt oxide for battery cathodes,” says Schmidt. “But this and other results have shown us that AI integration is a promising approach when searching for alternative materials.”
The AI-supported material substitution tool was developed as part of the research project “Ultraefficiency Factory – Deep Dive.”
It started in April 2024 and will run until the end of August 2025. The research project is supported by the Ministry of the Environment, Climate and Energy of Baden-Württemberg with a total of 1.4 million euros.
