Featured Discovery

Home > Featured Discovery > SJTU Doctoral Student Published Research Paper in Nano Energy

SJTU Doctoral Student Published Research Paper in Nano Energy

March 04, 2020      Author: Wei Han

Recently, Professor Bao Hua of University of Michigan-Shanghai Jiao Tong University Joint Institute (JI) and Professor Xiulin Ruan's research team of Purdue University jointly published their latest research paper titled "Genetic algorithm-driven discovery of unexpected thermal conductivity enhancement by disorder" in Nano Energy, an international authoritative academic journal. The paper displays their findings in using machine learning to reveal new thermal conductivity mechanism in disordered nanoporous graphene. Wei Han, a doctoral student from JI and selected into Zhiyuan Honor Program, is the first author. Professor Bao Hua and Professor Xiulin Ruan are co-correspondence authors of the paper.

Machine learning algorithms have been widely applied in data mining, computer vision, natural language processing, biometric identification, search engines, medical diagnosis, etc. This study provides a new research paradigm with machine learning to discover exceptions, contributing to further theoretical development. This work is sponsored by Guangdong Province Key Area R&D Program (2019B010940001) and the National Natural Science Foundation of China (No. 51676121). Is has also received support from the Materials Genome Initiative Center of Shanghai Jiao Tong University and Center for Higher Performance Computing, SJTU.


Discovering exceptions has been a major route for advancing sciences but a challenging and risky process. Machine learning has shown effectiveness in high throughput search of materials and nanostructures, but using it to discover exceptions has been out of the norm. Here we demonstrate the use of genetic algorithm to discover unexpected thermal conductivity enhancement in disordered nanoporous graphene as compared to periodic nanoporous graphene. Recent studies have concluded that random pores in nanoporous graphene lead to reduced thermal conductivity than periodic pores, due to phonon Anderson localization. This work, however, aims to challenge this accepted knowledge by searching for exceptions. A manual search was first shown to be expensive and unsuccessful. An efficient "two-step" search process coupled with genetic algorithm was then designed, and unexpected thermal conductivity enhancement was successfully discovered in certain structures with random pores, at a fraction of the computational cost of the manual search. Through structural analysis, we proposed that such unusual enhancement is due to the effect of shape factor and channel factor dominating over that of the phonon localization. Our work not only provides insights in thermal transport in disordered materials but also demonstrates the effectiveness of machine learning to discover small probability events and the intriguing physics behind.

Link: https://doi.org/10.1016/j.nanoen.2020.104619


Translated by Zhou Rong      Reviewed by Wang Bingyu