Pairing Accuracy and Efficiency in a Multi-Reference Excited State Method Applicable in Non-Adiabatic Molecular Dynamics Simulations
Final Report Abstract
Excited-state processes underlie many fundamental natural phenomena such as photosynthesis and human vision as well as important technological applications, including photovoltaic systems and light-driven hydrogen production. A deep understanding of these mechanisms is hence essential, not only for comprehending nature but also for driving technological innovations needed to address the pressing environmental challenges of our time. This project focused on advancing the theoretical tools necessary for studying these processes at an atomic level – an endeavor beyond the reach of traditional experimental techniques. To this end, a novel quantum mechanical approach was developed, demonstrating the potential to overcome significant limitations of existing methods for describing exited states. The proposed approach can achieve high accuracy at relatively low computational cost and is additionally systematically improvable, providing a clear pathway to increase accuracy by trading computational resources. While further refinement and testing are required, this work has laid a solid foundation for future developments and holds promise to inspire continued progress in this direction. Given that excited-state methods often depend on ground-state calculations as their starting point, errors in these so-called reference calculations can propagate, potentially compromising the accuracy of subsequent excited-state procedures. To address this issue, several strategies were developed to effectively identify and correct errors in ground-state calculations. These strategies are valuable not only for ensuring reliable starting points for excited-state calculations, but also for investigating ground-state properties themselves, which underlie many important phenomena. For example, the developed methodologies were further applied to investigate the translational eigenstates of endofullerenes – an intriguing class of molecules with potential applications as qubits in spin-based quantum computing architectures.
Publications
-
Corrected density functional theory and the random phase approximation: Improved accuracy at little extra cost. The Journal of Chemical Physics, 159(17).
Graf, Daniel & Thom, Alex J. W.
-
Simple and Efficient Route toward Improved Energetics within the Framework of Density-Corrected Density Functional Theory. Journal of Chemical Theory and Computation, 19(16), 5427-5438.
Graf, Daniel & Thom, Alex J. W.
-
Research Data Supporting “Targeting spectroscopic accuracy for dispersion bound systems from ab initio techniques: translational eigenstates of Ne@C70”. Apollo - University of Cambridge Repository.
K. Panchagnula, D. Graf & E. Johnson
-
Research Data Supporting “Translational Eigenstates of He@C60 from four-dimensional ab initio Potential Energy Surfaces interpolated using Gaussian Process Regression”. Apollo - University of Cambridge Repository.
K. Panchagnula & D. Graf
-
Targeting spectroscopic accuracy for dispersion bound systems from ab initio techniques: Translational eigenstates of Ne@C70. The Journal of Chemical Physics, 161(5).
Panchagnula, K.; Graf, D.; Johnson, E. R. & Thom, A. J. W.
-
Translational eigenstates of He@C60 from four-dimensional ab initio potential energy surfaces interpolated using Gaussian process regression. The Journal of Chemical Physics, 160(10).
Panchagnula, K.; Graf, D.; Albertani, F. E. A. & Thom, A. J. W.
