Ryoichi Tanaka
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ryoichi Tanaka has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Mikito Toda
Research Institute for Electronic Science, Hokkaido University 2 , Sapporo 001-0020
1 shared publication - Takuro Tsutsumi
Department of Chemistry, Faculty of Science Hokkaido University Sapporo Japan
1 shared publication - Tamiki Komatsuzaki
Hokkaido University
1 shared publication - Tetsuya Taketsugu
Hokkaido University
1 shared publication - Yuta Mizuno
National Institute of Advanced Industrial Science and Technology
1 shared publication
Latest Publications
- 15 October 2024
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 161 (15)
Low-dimensional projection of reactivity classes in chemical reaction dynamics using supervised dimensionality reduction
Transition state theory (TST) provides a framework to estimate the rate of chemical reactions. Despite its great success with many reaction systems, the underlying assumptions such as local equilibrium and nonrecrossing do not necessarily hold in all cases. Although dynamical systems theory can provide the mathematical foundation of reaction tubes existing in phase space that enables us to predict the fate of reactions free from the assumptions of TST, numerical demonstrations for large systems have been yet one of the challenges. Here, we propose a dimensionality reduction algorithm to demonstrate structures in phase space (called reactive islands) that predict reactivity in systems with many degrees of freedom. The core of this method is the application of supervised principal component analysis, where a coordinate transformation is performed to preserve the dynamical information on reactivity (i.e., to which potential basin the system moves from a region of interest) as much as possible. The reactive island structures are expected to be reflected in the transformed, low-dimensional phase space. As an illustrative example, the algorithm is scrutinized using a modified Hénon–Heiles Hamiltonian system extended to many degrees of freedom, which has three channels leading to three different products from one stable potential basin. It is shown that our algorithm can predict the reactivity in the transformed, low-dimensional coordinate system better than a naïve coordinate system and that the reactivity distribution in the transformed low-dimensional space is considered to reflect the underlying reactive islands.
- 15 October 2024
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 161 (15)
Low-dimensional projection of reactivity classes in chemical reaction dynamics using supervised dimensionality reduction
Transition state theory (TST) provides a framework to estimate the rate of chemical reactions. Despite its great success with many reaction systems, the underlying assumptions such as local equilibrium and nonrecrossing do not necessarily hold in all cases. Although dynamical systems theory can provide the mathematical foundation of reaction tubes existing in phase space that enables us to predict the fate of reactions free from the assumptions of TST, numerical demonstrations for large systems have been yet one of the challenges. Here, we propose a dimensionality reduction algorithm to demonstrate structures in phase space (called reactive islands) that predict reactivity in systems with many degrees of freedom. The core of this method is the application of supervised principal component analysis, where a coordinate transformation is performed to preserve the dynamical information on reactivity (i.e., to which potential basin the system moves from a region of interest) as much as possible. The reactive island structures are expected to be reflected in the transformed, low-dimensional phase space. As an illustrative example, the algorithm is scrutinized using a modified Hénon–Heiles Hamiltonian system extended to many degrees of freedom, which has three channels leading to three different products from one stable potential basin. It is shown that our algorithm can predict the reactivity in the transformed, low-dimensional coordinate system better than a naïve coordinate system and that the reactivity distribution in the transformed low-dimensional space is considered to reflect the underlying reactive islands.