Two-Step Machine Learning Approach for Charge-Transfer Coupling with Structurally Diverse Data

被引:0
作者
Lin, Hung-Hsuan [1 ,5 ]
Wang, Chun-, I [1 ,6 ]
Yang, Chou-Hsun [1 ]
Secario, Muhammad Khari [1 ,2 ,3 ]
Hsu, Chao-Ping [1 ,4 ]
机构
[1] Acad Sinica, Inst Chem, Taipei 115, Taiwan
[2] Acad Sinica, Taiwan Int Grad Program Sustainable Chem Sci & Tec, Inst Chem, Taipei 115, Taiwan
[3] Natl Yang Ming Chiao Tung Univ, Dept Appl Chem, Hsinchu 300, Taiwan
[4] Natl Ctr Theoret Sci, Div Phys, Taipei 106, Taiwan
[5] Yuan Ze Univ, Dept Comp Sci & Engn, Innovat Ctr Big Data & Digital Convergence, 135 Yuan Tung Rd, Taoyuan, Taiwan
[6] Univ Illinois, Dept Chem, Urbana, IL 61801 USA
关键词
POTENTIAL-ENERGY SURFACES; KERNEL RIDGE-REGRESSION; ELECTRON-TRANSFER; MOLECULAR-DYNAMICS; TRANSPORT; SCATTERING; DNA;
D O I
10.1021/acs.jpca.3c04524
中图分类号
O64 [物理化学(理论化学)、化学物理学];
学科分类号
070304 ; 081704 ;
摘要
Electronic coupling is important in determining charge-transfer rates and dynamics. Coupling strength is sensitive to both intermolecular, e.g., orientation or distance, and intramolecular degrees of freedom. Hence, it is challenging to build an accurate machine learning model to predict electronic coupling of molecular pairs, especially for those derived from the amorphous phase, for which intermolecular configurations are much more diverse than those derived from crystals. In this work, we devise a new prediction algorithm that employs two consecutive KRR models. The first model predicts molecular orbitals (MOs) from structural variation for each fragment, and coupling is further predicted by using the overlap integral included in a second model. With our two-step procedure, we achieved mean absolute errors of 0.27 meV for an ethylene dimer and 1.99 meV for a naphthalene pair, much improved accuracy amounting to 14-fold and 3-fold error reductions, respectively. In addition, MOs from the first model can also be the starting point to obtain other quantum chemical properties from atomistic structures. This approach is also compatible with a MO predictor with sufficient accuracy.
引用
收藏
页码:271 / 280
页数:10
相关论文
共 74 条
  • [1] Predicting the DNA Conductance Using a Deep Feedforward Neural Network Model
    Aggarwal, Abhishek
    Vinayak, Vinayak
    Bag, Saientan
    Bhattacharyya, Chiranjib
    Waghmare, Umesh, V
    Maiti, Prabal K.
    [J]. JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2021, 61 (01) : 106 - 114
  • [2] Machine Learning Prediction of Electronic Coupling between the Guanine Bases of DNA
    Bag, Saientan
    Aggarwal, Abhishek
    Maiti, Prabal K.
    [J]. JOURNAL OF PHYSICAL CHEMISTRY A, 2020, 124 (38) : 7658 - 7664
  • [3] Machine Learning Approach to Calculate Electronic Couplings between Quasi-diabatic Molecular Orbitals: The Case of DNA
    Bai, Xin
    Guo, Xin
    Wang, Linjun
    [J]. JOURNAL OF PHYSICAL CHEMISTRY LETTERS, 2021, 12 (42) : 10457 - 10464
  • [4] First Principles Neural Network Potentials for Reactive Simulations of Large Molecular and Condensed Systems
    Behler, Joerg
    [J]. ANGEWANDTE CHEMIE-INTERNATIONAL EDITION, 2017, 56 (42) : 12828 - 12840
  • [5] Perspective: Machine learning potentials for atomistic simulations
    Behler, Joerg
    [J]. JOURNAL OF CHEMICAL PHYSICS, 2016, 145 (17)
  • [6] Charge-Transfer Landscape Manifesting the Structure-Rate Relationship in the Condensed Phase Via Machine Learning
    Brian, Dominikus
    Sun, Xiang
    [J]. JOURNAL OF PHYSICAL CHEMISTRY B, 2021, 125 (48) : 13267 - 13278
  • [7] Bypassing the Kohn-Sham equations with machine learning
    Brockherde, Felix
    Vogt, Leslie
    Li, Li
    Tuckerman, Mark E.
    Burke, Kieron
    Mueller, Klaus-Robert
    [J]. NATURE COMMUNICATIONS, 2017, 8
  • [8] Reproducing global potential energy surfaces with continuous-filter convolutional neural networks
    Brorsen, Kurt R.
    [J]. JOURNAL OF CHEMICAL PHYSICS, 2019, 150 (20)
  • [9] Detailed balance, internal consistency, and energy conservation in fragment orbital-based surface hopping
    Carof, Antoine
    Giannini, Samuele
    Blumberger, Jochen
    [J]. JOURNAL OF CHEMICAL PHYSICS, 2017, 147 (21)
  • [10] Evolutionary Approach to Constructing a Deep Feedforward Neural Network for Prediction of Electronic Coupling Elements in Molecular Materials
    Caylak, Onur
    Yaman, Anil
    Baumeier, Bjorn
    [J]. JOURNAL OF CHEMICAL THEORY AND COMPUTATION, 2019, 15 (03) : 1777 - 1784