One of the biggest promises of Artificial Intelligence (AI) is faster progress towards new scientific discoveries. That is no longer merely a promise - it is already happening.
In June 2023, Microsoft launched Azure Quantum Elements with the aim of “accelerating scientific discovery with the power of Artificial Intelligence (AI)” combined with high-performance computing (HPC). Early outcomes are now starting to appear.
“Showing what can be built with AI is not the same as proving that this technology can identify something new.”
Dr. Nathan Baker, Product Lead, Azure Quantum Elements, Microsoft
Microsoft chose to apply this technology to the investigation and development of new battery chemistries, particularly solid-state batteries, and the initial signs are encouraging.
Why Microsoft built Azure Quantum Elements
The approach brings AI together with HPC so that vast numbers of candidate materials can be assessed far more quickly than traditional routes that rely on lengthy computation and extensive lab work.
First steps
The Azure Quantum team “joined forces” with the US Department of Energy’s Pacific Northwest National Laboratory (PNNL), and in August 2023 they presented their first results.
After a few months of work - including screening 32 million potential materials that could be used in battery manufacturing - the AI system concluded that more than 500,000 were “stable”. That established a starting point.
From there, those 500,000 materials were narrowed down to a single candidate by weighing multiple factors and functional properties relevant to producing a solid-state battery. Find out more about this type of battery:
Once they had characterised the structure of the chosen material, measured its connectivity, and put it through tests at various temperatures, they evaluated its technical viability using an experimental battery.
New solid-state battery material
After nine months of research, PNNL identified a material - not found in nature - that uses 70% less lithium than today’s batteries, partially addressing the scarcity of this raw material. Even though there is still a long road ahead, AI made it possible to “reduce this research process from years to weeks and from weeks to days”.
What previously demanded extensive high-performance computing (HPC) calculations and costly, time-consuming laboratory research was, with the help of AI, brought down to just nine months.
“PNNL has demonstrated that the potential of new HPC and AI approaches significantly accelerates the innovation cycle.”
Dr. Nathan Baker, Product Lead, Azure Quantum Elements, Microsoft
Beyond helping to synthesise insights from millions of materials tests, the AI also enabled the discovery of a material that uses less lithium by replacing it with sodium.
What the new chemistry could change for lithium-ion batteries
This newly identified combination of materials addressed some of the key drawbacks of lithium-ion batteries - safety and scarcity - while also promising higher energy density.
In other words, what is at stake is a new battery chemistry that is cheaper, more sustainable, and offers greater energy density.
“By using less lithium this battery will have less impact on the planet, be safer and at the same time bring greater economic benefits”.
Dr. Nathan Baker, Product Lead, Azure Quantum Elements, Microsoft
However, this is only the beginning of a development process that may become shorter with AI’s support.
Even so, there is still a long way to go before we see this new battery chemistry applied to our cars, computers, smartphones, and other devices that need an internal source of energy storage.
Source: Microsoft
Comments
No comments yet. Be the first to comment!
Leave a Comment