A transformative partnership is taking place between artificial intelligence (AI) and quantum data. Together, this partnership is poised to revolutionize materials science, particularly for battery technology. With help from AI, PNNL researchers are significantly reducing the time it takes to sift through large databases of new potential materials. This strategy enhances the efficiency and effectiveness of the discovery process. This new and creative approach greatly amplifies research in a scalable way. It hopes to make future applications safer, cheaper, and more environmentally friendly.
AI is so much more than simple data analysis. Its goals include recognizing “first-time right” candidates and ruling out unworthy materials. AI has helped whittle down an original list of more than 32 million potential battery materials to just 800 materials with high promise. This innovation is a prime example of how AI can transform the process through which scientists synthesize and test new compounds. This new level of efficiency frees researchers to focus their attention on the most promising solutions. In so doing, they can shorten the timeline for innovation.
The Role of AI in Material Discovery
AI’s increasing role in scientific research is changing the speed and scale of discovery across all scientific disciplines. Traditional approaches to material evaluation can be a cumbersome process, with simulations taking days to weeks to produce answers. AI helps to speed run those simulations at lightning speeds. This enables it to make fast predictions across similar systems, and it does this at a small percentage of the price of classical compute.
The next frontier for how AI models will be used to predict the performance of new battery materials is a perfect example. By utilizing large datasets, AI is able to take into account variables and performance measures that haven’t traditionally been thought of. This capability is fundamental to determining which materials best meet the sustainable design goals of today while still satisfying increasingly rigorous performance requirements.
This success represents the latest success of AI to help researchers sift through billions of potential battery materials to find the best candidates, showcasing real-world applications. Results show that AI is a powerful tool to identify materials that provide performance improvements while reducing environmental impacts. This joint collaboration is a major step forward in aligning transformative technology to scientific research.
Quantum-Enhanced AI: A New Frontier
Beyond just these aspects, the future of AI in chemistry will be in its ability to work in tandem with quantum computing. Quantum-enhanced AI will be key to addressing the world’s biggest challenges. This includes addressing climate change and a number of diseases, and it would do so several years sooner than previously expected. This next leap demands fault tolerance via the redundant encoding of quantum information in logical qubits.
Researchers project that to do anything interesting outside the classical computation boundaries, we’ll need hundreds, if not thousands of quality high-fidelity qubits. Ideally, these qubits have error rates on the order of 10^-15 to sustain reliable computation. As a point of comparison, to get to that level of reliability in quantum computing could take on the order of a million physical qubits.
The quantum computing implications for AI are huge, potentially groundbreaking, actually. This incredible technology now allows researchers to investigate the structures of very complex molecules containing many hundreds of atoms. It empowers them to pursue efforts that need computational capacity well outside the reach of today’s classical approaches.
Accelerating Research Through Collaboration
This continued partnership between AI and researchers at PNNL reinforces the importance of interdisciplinary partnerships for pushing revenue science forward. Using their background in AI and material science, the researchers are developing ways to quickly process large amounts of data. This helps them easily whittle down to the candidates that fit the bill.
This collaboration is a testament to how AI can reimagine traditional approaches to research. Today, researchers can do much more comprehensive evaluations in a fraction of the time it used to take. This helps scientists to rapidly narrow down the most promising candidates. This increased capacity allows them to minimize extraneous resource expenditures and focus efforts on synthesizing and testing material.
These technologies accelerate each other’s development, powering revolutions in materials science. As advances in AI grow exponentially, they have the power to disrupt all industries, including energy storage and healthcare. Their applications are huge, not only enhancing current enabling technologies but creating new ones as well.