Electron-Phonon Interactions and Charge Transport in Organic Crystals and Transition Metal Oxides from First-Principles Calculations

Citation

Chang, Benjamin K (2024) Electron-Phonon Interactions and Charge Transport in Organic Crystals and Transition Metal Oxides from First-Principles Calculations. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/6zkq-2p13. https://resolver.caltech.edu/CaltechTHESIS:03232024-174719340

Abstract

Electron-phonon (e-ph) interactions play a critical role in determining material properties, such as charge and heat transport, optical response, and superconductivity. Recent advances in first-principles calculations based on density functional theory (DFT) enable quantitatively predictive studies of e-ph interactions and charge transport in a wide range of simple semiconductors and metals. However, certain technologically important materials, such as organic crystals and transition metal oxides (TMOs), remain less explored. Organic molecular crystals, known for their versatile electronic and mechanical properties, typically require high charge carrier mobility for practical applications. Yet accurately predicting the mobility and engineering approaches to improve it are challenging in organic crystals, because of their complex crystal structures with large unit cells and various charge transport regimes induced by e-ph interactions. Similarly, TMOs, both conventional and strongly correlated, are materials with broad applications and unique physics. A notable example are copper oxides (cuprate) superconductors, which are central to the study of high-temperature superconductivity and other exotic physical phenomena. Extensive experimental studies, particularly using photoemission techniques, have been employed to indirectly probe the e-ph interactions in TMOs. Nevertheless, many results are not fully understood, and calculations of e-ph coupling in TMOs are still scarce. This is mainly due to the strong correlation induced by d- and f-electrons posing a significant challenge to modeling.

This thesis aims to develop state-of-the-art first-principles calculations to accurately describe e-ph interactions and the associated physical properties in organic crystals and TMOs. We focus on three research topics. First, we investigate the high-mobility bandlike transport regime in organic crystals. Using the formalism of the Boltzmann transport equation with electronic collisions computed from first principles, we study the mobility and its temperature dependence in benzene, anthracene, tetracene, pentacene, and biphenyl. Our results are in excellent agreement with experiments in all cases, and our pentacene calculation (72 atoms per unit cell) sets the record for the largest first-principles e-ph calculation to date. We find that the mobility is mainly regulated by e-ph scattering from low-frequency intermolecular phonons. Our analysis evidences the effectiveness of strain-based engineering to improve the mobility of organic crystals. Second, we propose a computational approach to study the intermediate polaronic transport regime in organic crystals. This method combines a finite-temperature cumulant-expansion approach for calculating electron spectral functions with the Kubo formula to compute the electronic conductivity and mobility. We show calculations of electron mobility in a naphthalene crystal in excellent agreement with experiments, and find that polaron effects, encoded in the satellites of the spectral functions, are induced by strong e-ph coupling of intramolecular hydrogen-atom vibrations. In the third and final topic, we study quantitatively the e-ph interactions in cuprate superconducting materials. Using the framework of Hubbard-corrected DFT, we focus on the prototypical parent (undoped) cuprate compound La₂CuO₄, which becomes superconducting upon doping. We show the first quantitative evidence of strong Fröhlich-type e-ph interactions between holes and oxygen atomic vibrations, as well as polaron effects in hole spectral functions. Our findings explain a range of observations in photoemission experiments on both undoped and doped cuprates, suggesting the strong e-ph coupling is an intrinsic feature of the parent compounds rather than being induced by doping. The computational workflow presented in this work can be easily extended to a broad class of strongly-correlated oxides and insulators more generally. In summary, this thesis pushes the boundaries of first-principles calculations of e-ph interactions and transport, paving the way for a microscopic understanding of materials with large and complex unit cells, strong electronic correlations, and strong e-ph interactions.

Thesis Files