Discrete and Continuous Methods for Modelling and Simulation of Polymer Materials

Pfaller S (2021)

Publication Language: English

Publication Type: Thesis

Publication year: 2021

Publisher: Friedrich-Alexander-Universität Erlangen-Nürnberg

Edited Volumes: Schriftenreihe Technische Mechanik

City/Town: Erlangen, Germany

ISBN: 2190-023X

DOI: 10.25593/opus4-fau-18036


The present thesis subsumes recent research activities of the "Capriccio" group, which has been established by the author at the Institute of Applied Mechanics of the Friedrich-Alexander-Universität Erlangen-Nürnberg in 2018. It focuses on multiscale modelling and simulation of polymers and is interested, among others, in particle-based material descriptions and molecular dynamics pseudo experiments, constitutive modelling of polymers, material characterization and model calibration, polymer nanocomposites and fracture of polymers. To this end, the group further advances the Capriccio method, which is a domain-decomposition multiscale technique specifically designed for amorphous materials and which has proven to be a suitable simulation tool to couple a continuum with a particle domain treated by molecular dynamics (MD). This approach subdivides the entire domain of interest into sub-domains of different resolution and thus confines the computationally expensive particle-based treatment to regions of specific interest: Typically, these are located in the vicinity of interfaces, crack tips, or other types of discontinuities. In contrast, the remaining regions are modelled as a continuum requiring significantly less computational effort. Such an approach, however, requires a careful consideration of the transition region between the different domains involved, but is a powerful tool for material modelling and simulation. With such a technique at hand, the system sizes which molecular dynamics is usually capable of can be significantly enlarged. Hence, representative volume elements of, e.g., heterogeneous materials like polymer nanocomposites, come within reach and form the basis for further upscaling techniques. The first part of this theses concentrates on an enhanced version of the Capriccio method in a purely elastic setting. The second part provides a profound data basis by MD pseudo experiments of a polymer model system at coarse-grained resolution. It turns out that the material behaviour is by far too complex to be captured by simple elastic descriptions and even available inelastic models are not able to adequately reproduce the MD simulation data. Hence, another part of this thesis focuses on the development of a viscoelastic-viscoplastic constitutive law which is able to appropriately model the material behaviour. With this novel constitutive description at hand, the Capriccio method is extended to inelasticity and yields promising results. Finally, the present thesis applies the multiscale framework introduced here to polymer nancomposites and detects material parameter profiles in the interphase between the filler particles and the bulk polymer. The approach presented here is rather unconventional and uses goal data obtained from coupled simulations using the Capriccio method. There, the interphase regions are described at particle resolution and base exclusively on the interactions between the particles. By means of an appropriate, purely continuum surrogate model, the material property profiles in the interphase are eventually detected. This procedure enables insights into regions which are not accessible by experiments and thus is a promising, novel methodology to identify the material behaviour informed by molecular dynamics simulations. Based on these findings, this thesis closes with an outlook to future research activities of the Capriccio group and embeds them into a larger scientific context. There, envisioned studies in the fields of multiscale treatment of nanocomposites and fracture, integration of modelling, simulation, and experiments as well as of advanced design of polymer materials are addressed.

Authors with CRIS profile

Related research project(s)

How to cite


Pfaller, S. (2021). Discrete and Continuous Methods for Modelling and Simulation of Polymer Materials (Habilitation).


Pfaller, Sebastian. Discrete and Continuous Methods for Modelling and Simulation of Polymer Materials. Habilitation, Erlangen, Germany: Friedrich-Alexander-Universität Erlangen-Nürnberg, 2021.

BibTeX: Download