By Carolyn Mathas

There’s a dramatic increase in global demand for lithium-ion batteries, however it’s combined with rising costs and dwindling materials. Predicting lithium-ion battery performance and degradation has been an ongoing challenge given the complexity of the electrochemical system inside the cell, which depends on cell chemistry, temperature, cycle rate, operational voltage windows, as well as physical design. To date, the only way to determine the cycle life of a battery is by performing extensive real-time tests. Drawbacks of this path include the rapid pace of innovation eclipsing battery chemistries and designs now under test—making them obsolete.

A collaboration was just announced by NOVONIX and Alphabet spin-off SandboxAQ leveraging SandboxAQ’s AI-driven chemical simulation software and NOVONIX’s Ultra-High Precision Coulometry (UHPC) technology. The enhanced data and analytics capabilities of the collaboration complements NOVONIX’s UHPC testing equipment, R&D prototyping, and testing services to provide actionable information faster. Accurately predicting cell life using short-term, high-throughput UHPC testing with advanced AI models will accelerate innovation for new battery chemistries.

The collaboration partnership is SandboxAQ’s first foray into quantum-powered materials science. According to According to Ang Xiao, Technical Lead for AQ Application Discovery at SandboxAQ, “The ability to create digital twins of chemical compounds and rapidly simulate and observe their chemical interactions and properties at the molecular level creates tremendous opportunities for scientists and researchers to create next generation materials for use in a wide range of industries and use-cases.”

Xiao explained that, “SandboxAQ has a track record of developing highly effective strategies to scale up quantum chemical simulations using such tools as large scale linear algebra with TPUs, large scale DFT results and a Post-Hartree-Fock method called DMRG, scaled up to unprecedented size. The tools provide a foundation for a suite of solutions combining the accuracy of quantum mechanical calculations with the speed and power of AI, running on classical computing hardware.”

While global demand for EVs is booming, the charging infrastructure has not caught up. EV manufacturers need batteries that are longer lasting, powerful, efficient, durable, and lighter, pushing more driver-friendly vehicles for consumer and commercial use.

“The demand among the electronics industry for smaller, more powerful, longer-lasting, durable and eco-friendly batteries for use in mobile phones, portable electronics and IoT devices is staggering,” said Xiao. “Advances in materials science will help battery manufacturers improve their chemical formulas for greater power, longevity, and performance. It will help them design smaller batteries and identify or create new materials to reduce their reliance on cobalt or lithium, which are of limited supply and difficult to extract or procure.”

Expectations are that SandboxAQ’s predictive modeling technologies, and NOVONIX’s industry-leading expertise, will transform the battery industry’s ability to make informed decisions around chemistries, processes, cells, and technologies at every stage of research and manufacturing. The companies expect that the models will be used for data products and services as soon as the first half of 2024. A press release with additional information can be found here.

September 14, 2023