In a recent study, a team of researchers successfully evaluated the frictional properties of multiple types of lubricants by using their new proposed method of combining machine learning and molecular dynamics. The research team included first-year doctoral student Ikki Yasuda and second-year doctoral student (at the time of this study) Katsuhiro Endo from the Keio University Graduate School of Science and Technology, as well as Assistant Professor (at the time of this study) Yusei Kobayashi, Associate Professor Noriyoshi Arai, and Professor Kenji Yasuoka from the same university. The remaining members of the research group, Associate Professor Kazuhiko Fujiwara, Professor Kuniaki Yajima, and the late Professor Yoshihiro Hayakawa were from the National Institute of Technology, Sendai College.
Lubricants are used in mechanical interfaces to reduce any friction that may occur. It is a well-known fact that the speed of mechanical processes has an impact on lubricants’ frictional properties. However, it has been difficult for scientists to use molecular dynamics to perform exhaustive analyses on these frictional properties for lubricants with diverse molecular structures. In this study, the team of researchers was able to decipher the molecular movements associated with common frictional properties in various molecules by using a method that employs machine learning to analyze the molecular dynamics of the molecules in lubricants. The research method proposed in this study provides a new way of performing a simplified and comprehensive analysis of various lubricants' characteristics. It is expected to make considerable contributions to a wide range of industries that rely on lubricants.
The outcomes of this study were published in the international academic journal, ACS Applied Materials & Interfaces, on January 30, 2023.