Quantum embedding for molecular systems : a projection-operator approach

Density functional theory (DFT) is widely used to describe the electronic structure of molecular systems and, thanks to the simplicity of its theoretical framework, it is particularly suitable for the development quantum embedding schemes. In this dissertation a novel embedding scheme, based on the...

Full description

Bibliographic Details
Main Author: Stella, Martina
Published: University of Bristol 2015
Subjects:
540
Online Access:http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.691179
id ndltd-bl.uk-oai-ethos.bl.uk-691179
record_format oai_dc
spelling ndltd-bl.uk-oai-ethos.bl.uk-6911792017-03-16T16:24:22ZQuantum embedding for molecular systems : a projection-operator approachStella, Martina2015Density functional theory (DFT) is widely used to describe the electronic structure of molecular systems and, thanks to the simplicity of its theoretical framework, it is particularly suitable for the development quantum embedding schemes. In this dissertation a novel embedding scheme, based on the employment of a projection operator, is presented. This method allows one to embed one sub-region of a given molecular system in its environment and treat these regions at different level of theory (e.g. CCSD(T)in- DFT). Thanks to the use of a projection technique that enforces the Pauli principle between subsystems, the complications associated with the appearance of non-additive kinetic energy contributions are overcome. First, I show a general software implementation of the method and the features that allow the analysis of a variety of chemical problems (e.g. organic reactions, transition metal complexes). Next, I apply the method to a wide range of benchmarking examples chosen to assess accuracy and performance. Namely, the SN2 reaction of I-propylchloride with CI- , phenol molecule deprotonation reaction, association of' iron(II) to ethylamine, Diels-Alder cycloaddition, and Stone' Vales rotation reaction are investigated. I show that , for such examples, this framework is able to reproduce the accuracy of highly correlated wave-function (WF) methods with reduced computational cost, by performing WF-in-DFT calculations. In addition, by exploring several simulation conditions, e.g. different functionals, localisation schemes, basis sets, I demonstrate the performance of the method displays a fairly independent behaviour with respect to simulation conditions. Finally, once the robustness of the code has been tested, I extend applications to more realistic chemical systems of technological and experimental interest, i.e. adsorption of cobalt on coronene. A further improvement of the method is also described. I assess a new version of the code that enables further reduction of the computational cost and the possibility of enlarging the size of the systems studied by performing an intelligent truncation of the atomic basis set used in the WF-based calculation.540University of Bristolhttp://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.691179Electronic Thesis or Dissertation
collection NDLTD
sources NDLTD
topic 540
spellingShingle 540
Stella, Martina
Quantum embedding for molecular systems : a projection-operator approach
description Density functional theory (DFT) is widely used to describe the electronic structure of molecular systems and, thanks to the simplicity of its theoretical framework, it is particularly suitable for the development quantum embedding schemes. In this dissertation a novel embedding scheme, based on the employment of a projection operator, is presented. This method allows one to embed one sub-region of a given molecular system in its environment and treat these regions at different level of theory (e.g. CCSD(T)in- DFT). Thanks to the use of a projection technique that enforces the Pauli principle between subsystems, the complications associated with the appearance of non-additive kinetic energy contributions are overcome. First, I show a general software implementation of the method and the features that allow the analysis of a variety of chemical problems (e.g. organic reactions, transition metal complexes). Next, I apply the method to a wide range of benchmarking examples chosen to assess accuracy and performance. Namely, the SN2 reaction of I-propylchloride with CI- , phenol molecule deprotonation reaction, association of' iron(II) to ethylamine, Diels-Alder cycloaddition, and Stone' Vales rotation reaction are investigated. I show that , for such examples, this framework is able to reproduce the accuracy of highly correlated wave-function (WF) methods with reduced computational cost, by performing WF-in-DFT calculations. In addition, by exploring several simulation conditions, e.g. different functionals, localisation schemes, basis sets, I demonstrate the performance of the method displays a fairly independent behaviour with respect to simulation conditions. Finally, once the robustness of the code has been tested, I extend applications to more realistic chemical systems of technological and experimental interest, i.e. adsorption of cobalt on coronene. A further improvement of the method is also described. I assess a new version of the code that enables further reduction of the computational cost and the possibility of enlarging the size of the systems studied by performing an intelligent truncation of the atomic basis set used in the WF-based calculation.
author Stella, Martina
author_facet Stella, Martina
author_sort Stella, Martina
title Quantum embedding for molecular systems : a projection-operator approach
title_short Quantum embedding for molecular systems : a projection-operator approach
title_full Quantum embedding for molecular systems : a projection-operator approach
title_fullStr Quantum embedding for molecular systems : a projection-operator approach
title_full_unstemmed Quantum embedding for molecular systems : a projection-operator approach
title_sort quantum embedding for molecular systems : a projection-operator approach
publisher University of Bristol
publishDate 2015
url http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.691179
work_keys_str_mv AT stellamartina quantumembeddingformolecularsystemsaprojectionoperatorapproach
_version_ 1718423551488294912