Researchers AI4Q, the world’s first nationwide AI and quantum computing initiative. Their goal is to create new battery materials that are safer, cheaper, and more environmentally sound. This collaborative effort, involving experts from various institutions, follows an innovative idea proposed by John P. Perdew in the summer of 2001. Surprisingly enough, over the course of 2023 and into early 2024, the project found strong headwind. Because of the massive scale, researchers turned to AI to screen more than 32 million possible battery materials, zeroing in on the best prospects.
Our close collaboration with the Pacific Northwest National Laboratory (PNNL) has been absolutely essential in getting AI to work for materials evaluation. In materials science, researchers are now better enabled to do rapid assessments in the field. This advanced technology transforms cumbersome data acquisition and analysis processes that were once just days, but often weeks or months. This process streamlines the speed of scientific discovery while improving the accuracy of material selection.
The Journey from Concept to Candidates
In the summer of 2001, John P. Perdew had one such thought-provoking idea. His approach would go on to revolutionize the field of materials science. Unbeknownst to them, this concept set the stage for a long, winding, but ultimately more fruitful road of evaluation that first started in earnest two decades later. First, we reviewed a huge body of work. In less than a week, we reduced that down to 500,000 stable potential candidates.
After this first round of filtration, researchers used complex algorithms and other techniques to further cut the list to 800 of the most promising candidates. It was this rigorous selection process that proved the power of combining human expertise with AI’s computational power.
“Set up on the earth, and the top of it reached to heaven. And behold the angels of God ascending and descending on it.” – Book of Genesis
This quote gets to the ambitious spirit of this project. Researchers are equally committed to making the most radical breakthroughs in battery technology.
The Role of AI in Material Evaluation
Artificial intelligence has shown to be an incredibly powerful tool in screening new promising battery materials. From 2023 through 2024, these AI models reviewed more than 32 million vetted resources. They provided insights that no age-old simulation could provide in the same time period. That’s a significant time savings, as classical computing methods can take days or weeks to provide such answers. In comparison, AI makes predictions, literally, at the speed of light.
By flattening the curve of scientific research, AI gives scientists the ability to hone in on “first-time right” candidates for synthesis and extensive testing. As resource efficiency and environmental concerns become increasingly important in material design, this capability will become even more crucial.
As our first joint seminar series between Smithsonian and PNNL demonstrated, it takes interdisciplinary collaboration to push this emerging field forward. Qiskit AI training as of now includes quantum-accurate data. This enabled researchers to gain an essential understanding to decide which materials should be studied in more detail, specifically studying their electronic properties and stability.
The Promise of Quantum Computing
The combined power of quantum computing and AI has the potential to revolutionize the field of materials science. Quantum computing will provide unique capabilities in modeling complex systems, especially those with exotic electronic behavior like high-temperature superconductors. An accurate description of electron correlation within these novel systems is critical for designing applications featuring them.
Compounds containing isolated specific metal atoms play an important and indispensable role for ionic catalytic reactions. Understanding these compounds is fundamental to developing the next generations of battery technology. With AI, quantum data can be analyzed faster and more precisely to identify optimal materials. This new capability increases researchers’ capacity to rapidly discover new materials with high-performance, safety, and environmental standards.
Making quantum computing truly reliable is a steep hill to climb. It would involve building fault-tolerant quantum computation systems, with quantum error correction making use of redundant encoding of quantum information in logical qubits. Further progress on this front will guarantee that quantum computers are producing complicated calculations in a reliable and reproducible manner.