Lin-Wang Wang

Institute of Semiconductors

Beijing Lonxun KuanTeng KeJi Inc.

EXTENDED ABSTRACT: Facing extreme challenges from global warming to end of Moore's law in the current world, there is an urgent need to quickly develop new materials. However, traditionally material development based on the trial­and-error approach is extremely time consuming, in average, it takes more than a decade to discover and fully develop a new material in a particular industry. This calls for the change of paradigm in the new material development. One new approach is to use first principles material calculations. This new approach is made possible by the 10 million folds computer power increase in the past three decades during which the first principles calculation based on density functional theory has been developed and widely accepted in academic research. Unfortunately, the current adoption rate of the first principles calculations is still low in industry, despite the tremendous opportunity to fundamentally change the industry. Traditional industry software based on continuum medium, Newton's law, and finite elements method cannot help the material development, rather they describe the material properties using empirical or experimentally measured parameters. To help the new material development or improving the material properties, one needs the atomistic simulations based on quantum mechanics. There are a few approaches first principle material simulations can help the new material development: (1) using high throughput calculation material Genome method to build the material database, to prescreen new materials before experimental research; (2) to help to understand the underline mechanism of a material phenomenon, hence to come up with strategy to solve the problem or improve the property; (3) using multiscale approach to directly simulate, either an microelectronic device, or a material process. However, to realize the full potential of these approaches, especially the approach (3), we still need to overcome a tremendous gap between the capability we have today, and the needs from the industry. Specifically, we need to overcome the challenges in accuracy, the size and temporal scale, and the easy usability. In this talk, I will discuss the efforts in LonXun Kuan Teng Inc. to develop material simulation software for industry usages. Keywords: first principles simulations, material calculations, HPC, quantum mechanical calculations.