Research & Publications

Research overview

Computational Chemistry & Molecular Simulation

My research combines machine learning-enhanced molecular simulation and GPU-accelerated methods to study nuclear quantum effects and excited-state dynamics, with a focus on accuracy and computational efficiency.

COMPUTATIONAL CHEMISTRY

Research Topics

Machine learning force fields
Molecular dynamics
Path integral

ResearchGate profile

PUBLICATIONS

Journal Articles

2026Journal article

Analytical Excited-State Gradients and Derivative Couplings in TDDFT with Minimal Auxiliary Basis Set Approximation and GPU Acceleration

Zhichen Pu, Xiaojie Wu, Yuanheng Wang, Cheng Fan, Wen Yan, Zehao Zhou^*, Yi Qin Gao^*, Qiming Sun^*

Journal of Chemical Theory and Computation, 22(4), 1793–1810 (2026)

Analytical TDDFT excited-state gradients and derivative couplings with minimal auxiliary basis approximations and GPU acceleration.

DOI

2026Journal article

GPU Accelerated Minimal Auxiliary Basis Approach TDDFT for Large Organic Molecules

Zehao Zhou^*, Xiaojie Wu, Yanheng Li, Xinran Wei, Cheng Fan, Fusong Ju, Qiming Sun^*, Yi Qin Gao^*

Journal of Chemical Theory and Computation (2026)

GPU-accelerated TDDFT with a minimal auxiliary basis approach for large organic molecules.

Google Scholar

2026Journal article

Path-Integral Molecular Dynamics of Muonium in Diamond Using Machine-Learning Force Field

Yue Yuan^†, Cheng Fan^†, Li Deng, Ziwen Pan^*, Yi Qin Gao^*, Bangjiao Ye^*

Physica B: Condensed Matter, 418749 (2026)

Path integral molecular dynamics with a machine learning force field to study muonium in diamond.

DOI

2025Journal article

Performing Path Integral Molecular Dynamics Using an Artificial Intelligence-Enhanced Molecular Simulation Framework

Cheng Fan, Maodong Li, Sihao Yuan, Zhaoxin Xie, Dechin Chen, Yi Isaac Yang^*, Yi Qin Gao^*

Journal of Chemical Theory and Computation, 21(15), 7279–7289 (2025)

Combining machine learning force fields with a modular simulation framework to study nuclear quantum effects in reactions and phase transitions at reduced computational cost.

DOI Preprint

2025Journal article

Multiscale Force Field Model Based on a Graph Neural Network for Complex Chemical Systems

Zhaoxin Xie, Yanheng Li, Yijie Xia, Jun Zhang, Sihao Yuan, Cheng Fan, Yi Isaac Yang^*, Yi Qin Gao^*

Journal of Chemical Theory and Computation, 21(5), 2501–2514 (2025)

A graph neural network-based multiscale force field model for complex chemical systems.

Google Scholar

2025Journal article

Effective Nucleation Size for Ice Crystallization

Maodong Li, Yupeng Huang, Yijie Xia, Dechin Chen, Cheng Fan, Lijiang Yang, Yi Qin Gao^*, Yi Isaac Yang^*

Journal of Chemical Theory and Computation, 21(4), 1990–1996 (2025)

Investigating the effective nucleation size in ice crystallization.

Google Scholar

2025Journal article

Catalytic Hydrogenation Dominated by Concerted Hydrogen Tunneling at Room Temperature

Qingyuan Wu^†, Pengxin Liu^†, Xia-Guang Zhang^†, Cheng Fan^†, Ziwen Chen, Ruixuan Qin, Yi Qin Gao^*, Yi Zhao^*, Nanfeng Zheng^*

ACS Central Science, 11(11), 2180–2187 (2025)

Investigating concerted hydrogen tunneling in catalytic hydrogenation at room temperature.

DOI

2024Journal article

Generating High-Precision Force Fields for Molecular Dynamics Simulations to Study Chemical Reaction Mechanisms Using Molecular Configuration Transformer

Sihao Yuan, Xu Han, Jun Zhang, Zhaoxin Xie, Cheng Fan, Yunlong Xiao, Yi Qin Gao^*, Yi Isaac Yang^*

The Journal of Physical Chemistry A, 128(21), 4378–4390 (2024)

High-precision force fields generated with a molecular configuration transformer for molecular dynamics studies of chemical reaction mechanisms.

Google Scholar

† Equal contribution; * Corresponding author

PREPRINTS

Preprints

2026Preprint

TDDFT Gradients and Nonadiabatic Couplings with Minimal Auxiliary Basis Set Approximation for Fewest-Switches Surface Hopping Dynamics

Cheng Fan^†, Zhichen Pu^†, Zehao Zhou, Yuanheng Wang, Yi Qin Gao^*, Qiming Sun^*

arXiv:2605.06489 (2026)

Efficient TDDFT-based surface hopping dynamics in GPU4PySCF using minimal auxiliary basis approximations and GPU acceleration.

arXiv

† Equal contribution; * Corresponding author

Journal articles are listed from the supplied Google Scholar profile; preprints appear separately. More onGoogle Scholar · ResearchGate