The scientists also accelerated the development process by using a predictive approach that lopped many months – and could lop years – off the typical timeline. For the most part, developing a new organic electronic material has been a time-intensive, somewhat hit-or-miss process, requiring researchers to synthesize large numbers of candidate materials and then test them.
The Stanford and Harvard-led group decided to try a computational predictive approach to substantially narrow the field of candidates before expending the time and energy to make any of them.
"Synthesizing some of these compounds can take years," said Anatoliy Sokolov, a postdoctoral researcher in chemical engineering at Stanford, who worked on synthesizing the material the team eventually settled on. "It is not a simple thing to do."
Sokolov works in the laboratory of Zhenan Bao, an associate professor of chemical engineering at Stanford. Alán Aspuru-Guzik, an associate professor of chemistry and chemical biology at Harvard, led the research group there and directed the theory and computation efforts.
The researchers used a material known as DNTT, which had already been shown to be a good organic semiconductor, as their starting point, then considered various compounds possessing chemical and electrical properties that seemed likely to enhance the parent material's performance if they were attached. They came up with seven promising candidates. Semiconductors are all about moving an electrical charge from one place to another as fast as possible. How well a material performs that task is determined by how easy it is for a charge to hop onto the material and how easily that charge can move from one molecule to another within the material.
Using the expected chemical and structural properties of the modified materials, the Harvard team predicted that two of the seven candidates would most readily accept a charge. They calculated that one of those two was markedly faster in passing that charge from molecule to molecule, so that became their choice.