Harvard University has jointly screened more than 1,000 blue light molecules for organic light-emitting diodes (OLEDs) with Samsung and MIT. The new material, if used in OLEDs, can significantly reduce its production costs and improve the performance of OLED screens.
The researchers developed a computer-aided molecular space shuttle screening process that combines experimental chemistry, machine learning, and chemical informatics to quickly identify new OLED molecules that are higher than industry standards.
Aspuru-Guzik, the team leader, said that researchers have long believed that organic light molecules should come from a very small molecular area that people find difficult to find. But with sophisticated molecular builders and the most advanced machine learning technology, they discovered high-performance blue OLED materials.
According to project researchers, in order to reduce the cost of OLED production, the biggest challenge is to emit blue light.
OLEDs, like LCDs, also require the combination of green, red, and blue sub-pixels to achieve all the colors in the screen. However, the current problem is that organic molecules do not emit blue light efficiently. In order to increase the intensity of blue light, OLED manufacturers use some transition metals (such as germanium) to enhance molecules with phosphorescence to produce an organic metal molecule. However, this solution is too costly and does not guarantee the stability of blue light.
Therefore Aspuru-Guzik and his team wanted to completely replace organic metal compounds with organic molecules. First, the researchers established a molecular library of more than 1.6 million candidate molecules. In order to narrow the scope, Ryan Adams, assistant professor of computer science at Harvard University, led the research team to develop a new machine learning algorithm to predict which molecules can produce better results. And give priority to the virtual testing of these molecules. In this way, the cost of search calculations is reduced by more than ten times. They conclude that the combination of chemistry and machine learning makes it easier for researchers to predict the color and brightness of molecules.
Adams mentioned that machine learning has already begun to be used in a large number of scientific fields. The collaboration of machine learning and homes of various disciplines will not only help promote the development of computer science, but also develop more new materials.
To find out more precisely these super molecules, researchers combined a theoretical model and experimental practice to create a web application that allows collaborators to study more than 500,000 quantum chemical simulations. After this phase was completed, the team got hundreds of molecules that are equivalent to the current best metal-free OLEDs.
From the current point of view, these molecules can not be directly commercialized, but with the development of this technology, the future is very likely to play a very big impetus for the replacement of OLED display screens.
Cable Gland,Power Cable,Cable Clips,Cable Organizer
Changshu Herun Import & Export Co.,Ltd , https://www.herunchina.com