Ten billion. That is how many commercially obtainable molecules are currently available. Start looking at them in groups of five (the typical combination used to make electrolyte materials in batteries) and it will increase to 10 to the 47th power.
For those who count, that’s a lot.
All of those combinations are important in the world of batteries. Find the right mix of electrolyte materials and you could end up with faster charging and a more energy-dense battery for an electric vehicle, the grid, or even an electric airplane. Does he lower it? Just like the drug discovery process, it can take more than a decade and thousands of failures to find the right drug.
That’s where the founders of the startup Aionica They say their ai tools can speed things up.
“The problem is that there are too many candidates and not enough time,” Aionics co-founder and CEO Austin Sendek told TechCrunch during the recent Up Summit event in Dallas.
Electrolytes, meet the ai
Lithium-ion batteries contain three fundamental components. There are two electrodes, an anode (negative) on one side and a cathode (positive) on the other. Typically, an electrolyte sits in the middle and acts as a messenger to move ions between the electrodes during charging and discharging.
Aionics focuses on the electrolyte and uses a suite of artificial intelligence tools to accelerate discovery and ultimately deliver better batteries. The aionic approach to catalyst discovery has also attracted investors. The Palo Alto-based startup, founded in 2020, has raised $3.5 million to date, including a $3.2 million seed round from investors that included UP.Partners.
The startup is already working with several companies, including Porsche’s battery manufacturing subsidiary Cellforce. The company has also worked with energy storage company Form Energy, Japanese materials and chemicals maker Showa Denko (now Resonac) and battery technology company Cuberg.
This entire process starts with a company’s wish list (or performance profile) for a battery. Aionics scientists, using ai-accelerated quantum mechanics, can perform experiments on an existing database of billions of known molecules. This allows them to consider 10,000 candidates every second, Sendek said. That ai model learns to predict the outcome of the next simulation and helps select the next candidate molecule. Each time it runs, more data is generated and problem resolution improves.
Enter generative ai
Aionics has gone a step further, in some cases bringing generative ai into the mix. Instead of relying on billions of known molecules, Aionics this year began using generative ai models trained on data from existing battery materials to create or design new molecules targeted at a given application.
The company is boosting its efforts by using software developed in Carnegie Mellon University’s Computational Accelerated Electrochemical Systems Discovery program. Venkat Viswanathan, an associate professor at CMU and director of that program, is co-founder and chief scientist of Aionics.
Aionics has also started using large language models built in OpenAI’s GPT 4 to help its scientists identify millions of possible formulations before they even start running them through the database. This chatbot tool, which has been trained on chemistry textbooks and scientific papers curated by Aionics, is not used for actual discovery, but scientists can use it to eliminate certain molecules that would not be useful in a particular application, Sendek explained. .
Once trained with those textbooks, LLMs allow the scientist to consult the model. “If you could talk to your textbook, what would you ask it?” Sendek said. But he was quick to point out that this is doing nothing different than a person who curates scientific articles. “This just provides next-level interaction,” he said, adding that everything is verifiable by pointing to the sources used to train the chatbot.
“I think the good thing for our field is that we’re not looking for specific facts, but rather for design principles,” he said while explaining the chatbot’s function.
Pick a winner
Once the billions of candidates have been selected and narrowed down to just a couple, or designed using the generative ai model, Aionics sends samples to its customers for validation.
“If we don’t make it to the first round, we repeat it and we can do some clinical trials to test it until we get to the winner,” Sendek said. “And once we find the winner, we work with our manufacturing partners to scale that manufacturing and bring it to market.”
Curiously, this process is even being used in some new sectors such as cement. chemoa startup co-founded by Viswanathan and which is also partnered with Aionics, is working on ways to use electricity and renewable raw materials to drive chemical reactions to make emissions-free products like cement.