Scientists have successfully demonstrated that the independent method can discover complex new materials. This AI-based technique has led to the discovery of three new nanostructures, including one of its kind that has not been achieved before.
The newly discovered structures were formed through a process called self-assembly, in which molecules of matter organize themselves into specific patterns.
The team of Gregory Doerk and Kevin Yager, from the US National Laboratory in Brookhaven, already have experience directing self-assembly processes, creating templates that make materials assume the desired configurations for applications in microelectronics, catalysis, and more. Now, their obtaining said nanostructure of a type not achieved before, can be described as a nanoladder, in addition to the ease of obtaining other nanostructures expands the range of self-assembly applications.
In previous studies, these scientists showed that it is possible to create new kinds of patterns by mixing two self-assembling materials.
“The fact that we can now create a ladder structure, which no one thought possible, is amazing,” Yeager asserts. “Traditional self-assembly can only form relatively simple structures, such as cylinders, sheets, and spheres. But by mixing two materials and using the appropriate chemical template, we discovered that completely new structures are possible.”
Images, taken using an electron microscope, of the structures created by the new AI-based automated system. Scale bars measure 200 nm. (Images: Brookhaven National Laboratory)
Mixing self-assembling materials allowed Doerk and Yager’s team to achieve unprecedented nanostructures, but it also created new challenges. With so many parameters to control the self-assembly process, finding the right set of parameters to create new and useful structures takes an enormous amount of time. To speed up their research, the scientists have taken advantage of a new capability of AI: autonomous experimentation.
In collaboration with the Center for Advanced Mathematics for Energy Research Applications (CAMERA) at Lawrence Berkeley National Laboratory in the US, Dwerk and his colleagues developed an artificial intelligence system capable of independently identifying and executing all steps of an experiment. The camera’s gpCAM algorithm drives the system’s independent decision-making. The latest research is the first successful demonstration of the algorithm’s ability to discover new materials.
Writing in the academic journal Science Advances, Doerk’s team reveals technical details of the new nanostructures achieved as well as the AI procedure used to create them via self-assembly. Their report is titled “Independent discovery of morphologies emerging in the directed self-assembly of group 3 copolymer blends”. (Line: NCYT by Amazings)
