Contribution to force field modeling & energy minimization of nanostructures

Chatzieleftheriou, Stavros

dc.contributor.author	Chatzieleftheriou, Stavros	en
dc.date.accessioned	2018-05-02T09:43:18Z
dc.date.issued	2018-05-02
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/46915
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.2876
dc.rights	Default License
dc.subject	Energy minimization	en
dc.subject	analytic hessian matrix	en
dc.subject	potential energy finite elements	en
dc.subject	molecular mechanics	en
dc.subject	Ελαχιστοποίηση δυναμικής ενέργειας	el
dc.subject	Αναλυτικός υπολογισμός εσσιανού μητρώου	el
dc.subject	Πεπερασμένα στοιχεία δυναμικής ενέργειας	el
dc.subject	Μοριακή δυναμική	el
dc.subject	Μοριακή μηχανική	el
dc.subject	molecular dynamics	en
dc.title	Contribution to force field modeling & energy minimization of nanostructures	en
dc.contributor.department	Institute of Structural Analysis and Antiseismic Research	el
heal.type	doctoralThesis
heal.classification	Nanotechnology	en
heal.classification	Modeling	en
heal.classification	Chemical engineering--Mathematics	en
heal.classificationURI	http://skos.um.es/unescothes/C02659
heal.classificationURI	http://id.loc.gov/authorities/childrensSubjects/sj96005954
heal.classificationURI	http://id.loc.gov/authorities/subjects/sh2009118797
heal.dateAvailable	2019-05-01T21:00:00Z
heal.language	en
heal.access	embargo
heal.recordProvider	ntua	el
heal.publicationDate	2018-03
heal.abstract	The potential energy of molecules and nanostructures is commonly calculated in the molecular mechanics formalism by superimposing bonded and nonbonded atomic energy terms. In this work a new, generalized numerical simulation is presented for studying the mechanical behaviour of three-dimensional nanostructures at the atomic scale. The energy gradient and Hessian matrix of such assemblies are usually computed numerically; a potential energy finite element model is proposed herein where these two components are expressed analytically. The global tangent stiffness matrix for any nanostructure is formed as an assembly of the generalized finite elements and is directly equivalent to the Hessian matrix of the potential energy. The advantages of the proposed model are identified both in terms of accuracy and computational efficiency. In the case of popular force fields (e.g. CHARMM) the computation of the Hessian matrix by implementing the proposed method is of the same order to that of the energy computation. Furthermore a new energy minimization strategy is presented, that achieves excellent convergence and is suitable for dealing with large-scale molecular systems. The basis of the proposed minimization strategy (SimNano) is a trust region algorithm based on exact second order derivative information. In order to present the efficiency of the proposed energy minimization strategy several test examples are examined and the results achieved are compared with those obtained by one of the most popular molecular simulations software, i.e. the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS). The results indicate that the proposed minimization strategy depict superior convergence properties compared to those of the algorithms that are generally employed in the field (i.e. the non-linear conjugate gradient, Broyden– Fletcher–Goldfarb–Shanno (BFGS) algorithms, etc.). Finally an application in computational material science (stress strain curve and computation of elastic moduli of atactic polystyrene based on the quasi-harmonic approximation) that benefits from fast and rigorous energy minimization is examined.	en
heal.advisorName	LAGAROS, NIKOS	en
heal.committeeMemberName	Lagaros, Nikos	en
heal.committeeMemberName	Theodorou, Doros	en
heal.committeeMemberName	Koumousis, Vlasis	el
heal.committeeMemberName	Provatidis, Christopher	en
heal.committeeMemberName	Bathe, Mark	en
heal.committeeMemberName	Papadopoulos, Vissarion	en
heal.committeeMemberName	Papathanasiou, Athanasios	en
heal.academicPublisher	Σχολή Πολιτικών Μηχανικών	el
heal.academicPublisherID	ntua
heal.numberOfPages	211
heal.fullTextAvailability	true