Google’s AI powerhouse finds millions of new crystals that could change the fate of humanity forever — and, for better or worse, it is just getting started
Google DeepMind researchers have used artificial intelligence to discover new crystals and inorganic materials that could power future technologies as part of a milestone study.
Using the deep learning tool Graph Networks for Materials Exploration (GNoME), researchers have found 2.2 million new crystals, including 380,000 stable materials.
The discovery could mark a milestone in the discovery of materials used to power modern technologies such as computer chips, batteries and solar panels, all of which rely on inorganic crystals.
Google AI development
Availability and stability of these materials is a common obstacle in the development of such technologies. However, researchers said that by using the GNoME AI tool they could “dramatically increase the speed and efficiency of discoveries by predicting the stability of new materials.”
“To enable new technologies, crystals must be stable, otherwise they can decompose, and behind each new, stable crystal there can be months of painstaking experimentation,” the study said.
“With GNoME, we have multiplied the number of technologically viable materials known to humanity. Of the 2.2 million predictions, 380,000 are the most stable, making them promising candidates for experimental synthesis.
“Among these candidates are materials that have the potential to develop future transformative technologies ranging from superconductors, powering supercomputers and next-generation batteries to increase the efficiency of electric vehicles.”
Transformative materials can drive the next wave of technological innovation
Google DeepMind's findings have been published in the Nature magazine, with the company noting that more than 28,000 new materials have been discovered over the past decade after extensive research.
However, traditional AI-based approaches to searching for new crystal structures have typically been an “expensive process of trial and error” that can take months before yielding minimal results.
“AI-guided approaches face a fundamental limit in their ability to accurately predict materials that could be experimentally viable,” the study said.
GNoME's recent discovery of 2.2 million materials would be “equivalent to approximately 800 years of knowledge,” researchers said, highlighting the transformative power and accuracy now afforded to scientists operating in the field.
As part of the project, approximately 52,000 new compounds similar to graphene have been discovered. According to the study, they could revolutionize the development of electronics and superconductors.
In previous years, only 1,000 of these types of materials had been identified through previous techniques.
“We also found 528 potential lithium-ion conductors, 25 times more than in a previous study, that could be used to improve the performance of rechargeable batteries.”
In the long term, Google DeepMind researchers said the GNoME project aims to “reduce the cost of discovering new materials.”
To date, external researchers have created 736 materials discovered through GNoME in a laboratory setting. The company also plans to release its database of newly discovered crystals and share its findings with the research community.
“By giving scientists the full catalog of the promising 'recipes' for new candidate materials, we hope this will help them test and enable the best ones.”