Computational Electrochemistry

According to the two-volume "Modern Electrochemistry" written by Bockris and Reddy, there are two kinds of electrochemistry. The first one is "The physical chemistry of ionically conducting solutions" and the second one is "The physical chemistry of electrically charged interfaces". We are working on the fundamental sides of these problems in energy storage/conversion with a focus on method developments.

Modeling electrochemical interfaces with finite field molecular dynamics
A realistic representation of an electrochemical interface requires treating electronic, structural and dynamic properties on an equal footing. The density functional theory-based molecular dynamics (DFTMD) method is perhaps the only approach that can provide a consistent atomistic description. However, the challenge for DFTMD modeling of material’s interfacial dielectrics is the slow convergence of the polarization P, where P is a central quantity to connect all dielectric properties of an interface.

Our contribution is to develop finite-field molecular dynamics simulation techniques for modeling electrochemical systems [1, 2]. Constant D Hamiltonian, originally designed for treating spontaneous polarization in groundstate ferroelectric systems, is a new statistical mechanics ensemble. We showed that the advantage of constant D simulations in computational electrochemistry is two-fold:   a) It significantly speeds up simulations and makes the dielectric constant calculations of polar liquids possible using DFTMD simulations; b) It also eliminates the finite size effect for modelling the electric double layer due to the periodic boundary condition. This methodology was further extended to treat the charge compensation between polar surfaces and the electrolyte solution.

Charge transportation in battery electrolytes
Lithium batteries are electrochemical devices which involve multiple time-scale and length-scale to achieve its optimal performance and safety requirement. In terms of the electrolyte which serves as the ionic conductor, a molecular-level understanding of the corresponding transport phenomenon is crucial for the rational design. Currently, we are working on molecular dynamics simulations of ionic conductivities in different types of electrolytes from aqueous electrolytes to polymer electrolytes (with Daniel Brandell) which are relevant to battery applications [3].

Developing atomic neural network for modeling molecules and materials
Machine learning is becoming increasingly important in computational chemistry and materials discovery. Atomic neural networks, which constitute a class of ML methods, have been very successful in predicting physico-chemical properties and approximating potential energy surfaces. 

Recently, we have taken the initiative and developed an open-source Python library named PiNN (, allowing researchers to easily develop and train state-of-the-art ANN architectures specifically for making chemical predictions. In particular, we have designed and implemented an interpretable and high-performing graph convolutional neural network architecture PiNet, and demonstrate how the chemical insight “learned” by such a network can be extracted [4].

[1] Zhang, C., Hutter, J. and Sprik, M.  J. Phys. Chem. Lett., 2016, 7: 2696 DOI:0.1021/acs.jpclett.6b01127

[2] Zhang, C., Hutter, J. and Sprik, M. J. Phys. Chem. Lett., 2019, 10: 3871 DOI:10.1021/acs.jpclett.9b01355

[3] Shao, Y., Hellström, M., Yllö A., Mindemark, J., Hermansson K., Behler, J. Zhang, C. Phys. Chem. Chem. Phys., 2020, DOI: 10.1039/C9CP06479F (Communication)

[4] Shao, Y., Hellström, M., Mitev, P. D., Knijff, L., Zhang, C. J. Chem. Inf. Model., 2020, DOI: 10.1021/acs.jcim.9b00994

Keywords: Molecular Dynamics, Density Functional Theory, Machine Learning, Electrolytes, Solid Electrolyte Interfaces