Frequency Dependent Polarizability for 1D Periodic Systems at Coupled Cluster with Single and Double Excitations Level

Marco Caricato, Taylor Parsons, Michael J. Frisch, Julia Abdoullaeva · 2025

Read Frequency Dependent Polarizability for 1D Periodic Systems at Coupled Cluster with Single and Double Excitations Level by Marco Caricato, Taylor Parsons, Michael J. Frisch, Julia Abdoullaeva (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-01266.txt.

Introduction

Marco Caricato and co-authors report a computing first in a 2025 chemistry paper. They computed how a repeating one-dimensional material responds to an electric field that changes with frequency. The property, called polarizability, was calculated with a high-accuracy method known as CCSD, never before applied this way to periodic systems.

The work was built into an open-source Python program, and the paper lays out the full equations. Test calculations show that moving from a single molecule to an endless chain flips which parts of the response dominate. The authors also test how the result settles as sampling grows, and discuss a known subtlety in defining electric response for endless systems. The advance opens accurate optical studies of solid materials.

Catalog record

CREATORMarco Caricato, Taylor Parsons, Michael J. Frisch, Julia Abdoullaeva
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE65 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-01266

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Frequency Dependent Polarizability for 1D Periodic Systems at Coupled Cluster with Single and Double Excitations Level Marco Caricato, Taylor Parsons, Michael J. Frisch, Julia Abdoullaeva Abstract We report the first implementation of the frequency-dependent electric dipole-electric dipole polarizability for 1D periodic systems computed with the coupled cluster with single and double excitations (CCSD) method with periodic boundary conditions (PBCs). The implementation is performed in the CCResPy open-source software, based on Python and the NumPy library. The complete equations and many details of the implementation are discussed. The test calculations show the impact on this linear response property of passing from a single molecule to a periodic chain, where the relative magnitude of the polarizability tensor elements is inverted. This work also explores the convergence towards the PBC thermodynamic limit with k k -space sampling, and some remaining issues in the definition of the electric dipole operator for periodic systems. This work represents a significant step forward for the simulation of optical response properties for solid-state materials with accurate and systematically improvable quantum mechanical methods. I Introduction The simulation of electronic and optical properties of solid state materials is paramount for the interpretation of the electronic response of these systems and their technological applications. The theoretical workhorse for these simulations is density functional theory (DFT) with periodic boundary conditions (PBC), which provides the best balance between reliability and computational cost. Evarestov 2012 However, DFT for the solid state suffers from the same issues as for molecules: the exact functional is unknown and any given approximate functional is not systematically improvable. Furthermore, a PBC simulation is inherently more computationally demanding than that on a molecular system of the same size as the simulation cell, which further limits the quality of the functional and basis set expansion that can be used compared to molecules. Benchmarking of approximate DFT methods is harder for solids than for molecules, because a direct comparison with experimental data is difficult, and high-level simulations (which are often used as reference in molecular benchmarking studies) are not available. For isolated molecules, when highly-accurate data are needed, the wave function (WF) formalism is invoked because systematically improvable electron correlation methods built on top of a reference WF can be formulated. Except for near-degeneracy situations where multi-reference methods are required, the reference WF is provided by the Hartree-Fock (HF) method. From there, post-HF methods lead towards the exact WF (within a set of constraints: the Born-Oppenheimer approximation, a finite and often incomplete basis set expansion, and a non-relativistic electronic Hamiltonian) with a sequence of well-defined improvements of increasing complexity and computational effort. The HF wave function is the best approximation, in a variational sense, of the exact WF when using a single Slater determinant built from optimized molecular orbitals (MOs). Post-HF methods improve upon this description using a linear or non-linear combination of Slater determinants built as a set of single, double, etc. electron excitations from occupied to unoccupied MOs. The exact WF is obtained when all possible excitations are accounted for within a finite basis set. Although the exact solution is practically unachievable for most real cases, a systematic path towards more accurate results obtained through the inclusion of more and more excited determinants is obviously a very desirable characteristic of these methods. Among these post-HF methods, those based on coupled cluster (CC) theory are the most successful in quantum chemistry. Bartlett and Musial 2007 ; Shavitt and Bartlett 2009 CC theory uses a non-linear expansion of the WF based on an exponential excitation operator e^{\hat{T}} , and approximate CC methods are defined in terms of the classes of excitations that are included in the \hat{T} cluster operator. The most common CC method includes single and double excitations (CCSD), because it provides the first important contribution to the correlation energy while maintaining a reasonable computational cost: O(N^{6}) where N is a measure of the system size. CC’s popularity stems from including higher-order excitations for a specific level of truncation (e.g., CCSD includes up to quadruple excitations thanks to products of single and double excitation operators) and the energy is size-extensive at each level of truncation, thus ensuring a balanced increase of the error with system size. Bartlett and Musial 2007 ; Shavitt and Bartlett 2009 The downsides of CC theory are that the complexity of the WF expansion grows quickly, both in terms of equations and of the corresponding computer code, and that the computational scaling is unfavorable compared to DFT: O(N^{6}) for CCSD vs O(N^{3}) for standard density functional implementations. Despite the steep computational cost, CC methods have been extended to the solid state, from the seminal work of Hirata et al. Hirata et al. 2004 ; Ohnishi and Hirata 2011 ; Gilliard, Sode, and Hirata 2014 ; Keçeli and Hirata 2010 ; Sode et al. 2009 ; Hirata and Ohnishi 2012 ; Hirata and Bartlett 2000 on CCSD energies to the more recent efforts from the Chan, Grüneis, Berkelback, and other groups. White et al. 2020 ; Pulkin and Chan 2020 ; Peng et al. 2021 ; White and Kin-Lic Chan 2020 ; Gao et al. 2020 ; White and Chan 2019 ; Zhu et al. 2019 ; White and Chan 2018 ; McClain et al. 2017 ; McClain et al. 2016 ; Gallo et al. 2021 ; Irmler and Grüneis 2019 ; Zhang and Grüneis 2019 ; Gruber et al. 2018 ; Grüneis 2018 ; Gruber and Grüneis 2018 ; Hummel, Tsatsoulis, and Grüneis 2017 ; Liao and Grüneis 2016 ; Grüneis 2015 ; Grüneis et al. 2011 ; Shepherd and Grüneis 2013 ; Schäfer et al. 2021 ; Xing and Lin 2024a ; Xing and Lin 2024b These developments include the calculation of ground and excited state energies, Pulkin and Chan 2020 ; McClain et al. 2017 ; Lange and Berkelbach 2020 ; Gallo et al. 2021 ; Wang and Berkelbach 2020 ; Katagiri 2005 ; Wang and Berkelbach 2021 ; Ye and Berkelbach 2024 ; Neufeld and Berkelbach 2023 ; Neufeld, Ye, and Berkelbach 2022 ; Vo, Wang, and Berkelbach 2024 and provide much needed high-level reference data for more approximate methods. These works also provide an opportunity to directly explore the importance of electron correlation in electronic properties of materials, going beyond the independent-particle approximation often invoked in condensed-phase physics. Excitation energies are important to predict the optical band gap in solid materials. However, the motion of the electron density in response to an external field is more generally described by the full linear response (LR) function. Christiansen, Jorgensen, and Hattig 1998 ; Helgaker et al. 2012 In this work, we present the first implementation for 1D periodic systems at CCSD level of the frequency-dependent LR function for the electric dipole operator in the length gauge (LG), i.e. the electric dipole-electric dipole polarizability tensor \bm{\alpha}(\omega) , where \omega is the frequency of the external field. The implementation was carried out in the CCResPy open source software, based on the Python language and the NumPy module library. git We report the equations and many details of the implementation, together with a set of proof-of-concept calculations on 1D chains of simple molecules. We explore how these electronic properties change going from an isolated molecule to a periodic

Read the complete text