Computational Materials Physics

We investigate fundamental quantum-physical properties of materials and their potential impact on applications, using leading-edge computational techniques in the research group of Prof. Cesare Franchini and Prof. Georg Kresse.

Main Research branches:

Nature Reviews Materials -- cover July 2021
Polarons

Polarons are quasiparticles that easily form in polarizable materials due to the coupling of excess electrons or holes with ionic vibrations [Nature Reviews Materials 2021]. By adopting advanced computational methods [Springer Handbook 2020], we study the intrinsic nature of polarons [PRX 2017] and their impact on applications [PRL 2019].

KTaO3(001) -- Science 2018
Surface Science

Surface of solids show properties different than the corresponding bulk materials, which might have a dominant impact on applications. We study the morphology and electronic properties of oxide surfaces [Science 2018], with particular interest for their chemical properties [PNAS 2020], and for the influence of surface charge states on polar terminations and their coupling with ferroelectricity [Nat. Comm. 2022, Sc. Adv. 2022].

Unfolding
DFT and Machine Learning

Leading-edge scientific research requires constant development of the investigation techniques. Our well established experience with Density Functional Theory (DFT) and the implementation of useful tools, such us the unfolding scheme [JPC-C 2021], support us in in the study of a wide range of quantum materials properties [Sc. Adv. 2019, Nat. Comm. 2022]. Additionally, we contribute to develop Machine Learning models tailored for polaronic materials [npj Comp. Mat. 2022] and surface science problems [TACO SFB project].

Teaching Activities

[Winter Semesters, Lab-Course (LP), Bachelor Program]

Simulations of quantum-mechanical many-electron systems, with special focus on solid state physics and materials science.

The course is structured in two parts.
In the first part, lectures consist of brief theoretical introduction to physical properties and computational techniques, followed by hands-on sessions; during these practical sessions, students try to solve weekly assignments in groups (typically pairs).
In the second part of the course, students work on a final project on a topic of their preference (chosen in agreement with dedicated supervisors).

Simulations are performed in the density-functional theory framework, by using the Vienna ab initio software package (VASP).
Fundamental physical properties, such as the electronic band structure and density of states, effective forces acting on ions, vibrational frequencies, mechanical properties, thermodynamic properties, as well as magnetic properties are discussed and calculated.
Upon request, during the final project, students may also write programs in selected areas of computational science (Monte Carlo, molecular dynamics, Schrödinger equation solvers, Machine Learning).

Link (W-2020)

[Summer Semesters, Frontal Lecture plus Practice (VU), Master and PhD Programs]

The course aims to introduce students to fundamental aspects of Surface Science.
The topics range from the theoretical study of surface crystallography and facets of mono-atomic and oxide compounds, up to polarity instabilities, structural reconstructions and chemical activity. The theoretical description of these topics is accompanied by the review of established and modern experimental techniques (such as scanning probe microscopy, spectroscopy, electron diffraction) and computational methods (mostly density functional theory and machine learning).

Link (S-2021)

[Summer Semesters, Frontal Lecture plus Practice (VU), Bachelor Program]

The course focuses on the application of Data Science methods in Physics, that is the combination of interdisciplinary activities (such as scientific, statistical and computational tools) required to elaborate data-centered analysis on relevant physical quantities. Data Science is a topic of increasing interest in the scientific community, due to the growing power of modern computational machines and the associated creation of large databases: The valuable information stored in such large databases can be extracted by Data Science methods, i.e., by combining statistics with advanced computational methods, including machine learning.

This course aims to guide students through the basic theoretical concepts regarding Data Science in Physics, and to provide them with the ability to successfully face practical applications in this field. Specifically, the lectures cover the following topics: (i) collection and manipulation of data via computational tools (mostly in python environments), (ii) effective visualization of relevant information extracted from data, (iii) scientific analysis and physical interpretation of data, (iv) advanced computational techniques.
The course is structured in theoretical lectures, followed by practical lectures.

Link (S-2021)

[Winter Semesters, Frontal Lectures (VO), Master Program]

Leading Lecturer: Prof. Cesare Franchini

Introduction to computational problems in physics with emphasis on fundamental concepts in classical and quantum mechanics (Newton equation, Maxwell equations, Schrödinger equation). The course provides a comprehensive introduction to the basics methods and algorithms of computational physics from a multidisciplinary perspective, merging together fundamental physics, mathematics and computer sciences concepts (including machine learning) conveyed in the language of a natural scientist. Using several examples, the course will illustrate step-by-step how to construct computer programs to carry out simulations and solve realistic problems.

Link (W-2021)

[Summer Semesters, Laboratory (LP), Master Program]

Leading Lecturer: Prof. Cesare Franchini

In this practicum you will use numerical methods to solve classical, statistical or quantum mechanical many body problems. The students can chose between implementing a code or applying an existing code such as VASP to a materials sciences problem.

Link (2023-S)

Willing to prepare your Thesis in the field of Computational Materials Physics?
Aiming to work on a research project at our institution?
I would gladly discuss the opportunity to work together with motivated students.
Thesis Topics are usually related to our Research field: have a look at the list below.

Record of supervised Thesis and collaborations:

(present)
"Artificial-Intelligence driven interpretation of LEED experiments" by T. Sebastian (Bachelor Thesis).
(present) "Automatized AFM experiments" by H. Forman (external project).
(present) "Acceleration of Polaron Dynamics via Machine Learning Force Field" by S. Trivisonne (Master Thesis).
(present) "Characterization of polarons on Nb-doped TiO2(110) surfaces" by A. Veternik (Bachelor Thesis).
(present) "Lattice Defects in Density Functional Theory" by V. Rosenzweig (Bachelor Thesis).
(present) V. Birschitzky's Doctorate on "Machine Learning for Polarons" [co-supervising].
(present) M. Corrias' Doctorate on "Computer Vision for Surface Science" [co-supervising]
(present) F. Ellinger's Doctorate on "Polarons on Perovskite Surfaces" [co-supervising].
(2023) "Machine Learning the Potential Energy Surface of Molecules" by A. Zier (Bachelor Thesis) [co-supervisors: M. Sahre, D.Lemm].
(2023) "Machine Learning meets LEED-IV Curves" by I. Grabner (Bachelor Thesis).
(2023) "Electronic State Unfolding for Plane Waves" by D. Dirnberger (Master Thesis).
(2022) J. Huber's Master Thesis on "Ferroelectric Stability of KTaO3" [co-supervised].
(2022) "Polaron Dynamics via Machine Learning with Atom-Centered Symmetry Functions" by H. Forman (Bachelor Thesis).
(2022) "Polaron Dynamics via Machine Learning with Gaussian Regression" by M. Janach (Bachelor Thesis).
(2021) D. Freinberger's training project on "Polaron Dynamics via Machine Learning".
(2021) "Polaron Dynamics in SrTiO3" by M. Corrias (Master Thesis, Erasmus internship) [co-supervised].
(2021) "Polaron Stability via Machine Learning" by M. Prezzi (Bachelor Thesis).
(2021) "Polaron Hopping in Hematite" by F. Six (Bachelor Thesis).
(2021) "Electronic Properties of the WO2(001) Polar Surface of Tungsten Trioxide" by A. Angeletti (Master Thesis, Erasmus internship) [co-supervised].
(2020) "Polaron Configurational Energies using Machine Learning" by V. Birschitzky (Master Thesis) [co-supervised].
(2020) "Solving the Schrödinger Equation by Machine Learning" by D. Freinberger (Bachelor Thesis).
(2020) "Follow the Polarons: Molecular Dynamics and Machine Learning" by S. Trivisonne (Bachelor Thesis).
(2019) "Effects of Doping in LaMnO3" by A. Lümbacher (Bachelor Thesis) [co-supervised].
(2018) "Doping-induced Insulator-to-Metal Transition in the Spin-orbit Oxide NaOsO3" by S. Dobrovits (Master Thesis) [co-supervised].
(2017) "Computational Tools for Surface Modeling" by P. Flauger (Training Project).
(2016) "Polaron Dynamics" by P. Flauger (Bachelor Thesis) [co-supervised].
(2016) "Effective Band Structure" by D. Dirnberger (Bachelor Thesis) [co-supervised].

My Tweets (login to your X account to visualize the content)

Created with Mobirise web template