April 22 - 26, 2024
Seattle, Washington
May 7 - 9, 2024 (Virtual)
Symposium Supporters
2024 MRS Spring Meeting & Exhibit
EL03/EL06/MT01.01

Exploring The Application of Hybrid DFTB-Molecular Mechanics Approach to Computing Optical Properties

When and Where

May 7, 2024
8:15am - 8:30am
EL06-virtual

Presenter(s)

Co-Author(s)

Ruicheng Li1,Gekko Budiutama1,Keisuke Kameda1,Sergei Manzhos1,Manabu Ihara1

Tokyo Institute of Technology1

Abstract

Ruicheng Li1,Gekko Budiutama1,Keisuke Kameda1,Sergei Manzhos1,Manabu Ihara1

Tokyo Institute of Technology1
The Density Functional based Tight Binding (DFTB) method has seen a rise in adoption for materials modeling, as it permits electronic structure-level modeling of large molecules, interfaces, nanostructures, etc. with much smaller computational cost and similar accuracy when compared to Density Functional Theory (DFT) methods. The cost reduction in DFTB compared to DFT is achieved by the pre-parameterization of the elements of the Hamiltonian matrix as well as the repulsion potential between all pairs of atoms. However, parameterization for new systems with accuracies competitive with DFT in specific applications requires specialized manpower and large computational resources. This prevents the application of the DFTB method to systems for which it was not parameterized.<br/>In our previous work (J. Chem. Theory Comput., 19, 5189-5198 (2023)), a DFTB-molecular mechanics (DFTB-MM) approach was introduced to expand the applicability of DFTB by modeling the interactions of the missing Slater-Koster parameters with an interatomic interaction potential fitted with machine learning. We have shown that when interactions between the atoms do not critically affect key mechanisms of interaction, we can obtain structures and partial densities of states with a similar accuracy to full DFTB. This in principle allows the use of DFTB in systems where not all pairs of atoms have been parameterized.<br/>Here, we further explore the capability of the DFTB-MM method to the calculation of optical properties with Time-Dependent (TD-) DFTB for the modeling of systems whose optical properties are of interest. TD-DFTB permits calculations that include a large number of excitations, which is required for systems with high densities of states, and would not be feasible with TD-DFT. In particular, we investigate interfacial charge transfer in systems such as TCNQ (tetracyanoquinodimethane) on TiO2, whose spectra are sensitive to the computational approach. We show that key spectral features, in particular transitions that involve the interfacial charge transfer band, are similar in full DFTB and DFTB-MM, indicating that DFTB-MM is a viable approach also for optical properties.

Keywords

absorption | oxide

Symposium Organizers

Aiping Chen, Los Alamos National Laboratory
Woo Seok Choi, Sungkyunkwan University
Marta Gibert, Technische Universität Wien
Megan Holtz, Colorado School of Mines

Symposium Support

Silver
Korea Vacuum Tech, Ltd.

Bronze
Center for Integrated Nanotechnologies, Los Alamos National Laboratory
Radiant Technologies, Inc.

Session Chairs

Aiping Chen
Rodrigo Freitas
Eiichi Kondoh

In this Session