Search Results

Project B7: Automated model building and representation learning for multiscale simulations Project B7 addresses applications of machine learning techniques to multi-scale simulation of soft-matter systems. Multi-scale methods address the problem that the complexity of high-resolution base-line models grows too quickly for problems at relevant scales. Thus, they assumed that there is a coarser-resolution structure emerging from the details that can be efficiently computed with many fewer operations but that can still inform us about relevant behavioural aspects of the system. Machine learning can help in discovering such simplified surrogate computations by fitting a restricted computational model (such as a parametrized model, a kernel regressor, or a deep feed-forward network) to example results obtained from a full-resolution simulation. Conceptually, this involves two aspects: The first is to build a coarser-grained (CG) model. Learning of a CG model can take the form of just parametrizing a force-field or a mapping procedure motivated […]

Projects – C: Bridging the particle-continuum gap • C1: Using molecular fields to bridge between particle and continuum representations of macromolecular systems • C3: Spinodal decomposition of polymer-solvent systems • C4 (E): Nonlocal electrostatics of biomolecular systems • C5: Adaptive hybrid multiscale simulations of soft matter fluids • C6 (E): Linking hydrodynamics and microscopic models of wet active matter with anisotropic particles • C7: Dense active suspensions in the chaotic regime • C8: Numerical approximation of high-dimensional Fokker-Planck equations

Admission and Qualification Admission Students funded directly by the TRR146 are automatically admitted to the IRTG. External students can apply for admission by presenting a CV and a one-page project plan to the TRR146 Office where they explain how their project fits TRR146 topics. The application will be evaluated by the PIs of the IRTG. We anticipate that workshop and conference travel funding for admitted external students will be limited and accessible only upon application. Qualification plan The most important training element of the IRTG is the research on the project, assisted by efficient supervision . In addition, the integrated training group serves as a mean to provide students and young postdoctoral researchers with the training required for working within the CRC-TR. The training is made necessary by the interdisciplinary nature of the CRC-TR, where chemistry, physics, mathematics and computer science are intertwined in a non-standard combination, which is usually […]

IRTG – Activities The IRTG fosters its objectives through a series of activities (see items below), which the members can attend/exploit in relation to their needs. Participation to the activities of the IRTG should not require more than 15% of a student’s working time. The activities are coordinated by the coordination office of the CRC-TR together with the elected student/postdoc speakers.

IRTG Organization Currently, the student/postdoc speakers are • Rebecca Steiner (further information) • Fabio Frommer (further information) • Moritz Mathes (further information) • Maarten Brems (further information)

Project A2: Dynamically consistent coarse-grained models The aim of this project is to develop methods that endow chemically-specific coarse-grained (CG) simulation models with consistent dynamical properties. To this end, CG models with conservative and dissipative interactions are derived from a higher-resolution model using bottom-up coarse-graining methods that retain a highmlevel of chemical specificity. In the first two funding phases, we have developed methods for deriving Markovian and non-Markovian CG models that successfully represent the dynamics of molecular liquids, polymer solutions, and star-polymer melts on diffusive time scales. The Markovian method uses a dissipative particle dynamics (DPD) thermostat that is parameterised by means of a bottom-up approach using the microscopic dynamics. While successful in CG simulations of molecular liquids where only the friction due to the relaxation of atomic vibrations needs to be accounted for, it fails to describe the dynamics of polymer melts and the dynamics of small molecules in […]

Project A3: Coarse-graining frequency-dependent phenomena and memory in colloidal systems The purpose of this project is to develop numerical strategies for dynamic coarse-graining in situations where the separation of time scales is incomplete and memory effects are important. This entails the reconstruction of coarse-grained dynamical equations that include memory (generalized Langevin equations, GLE), the efficient simulation of coarse-grained models with memory and the application to colloidal dispersions at equilibrium and non-equilibrium. This project is complementary to project A2, where related problems are addressed in the context of dynamic coarse-graining of molecular liquids. In the second funding period, we have extended our previous work on iterative memory reconstruction for single colloids (first funding period) to systems containing multiple colloids, where pair memory effects must be taken into account. A benchmark simulation of 125 colloids in solution showed that a speedup of at least three orders of magnitude can be obtained by […]

Project A4 (Completed): Understanding Water Relaxation Dynamics at Interfaces The aim of the project is to develop multiscale approaches to understand the mechanisms of vibrational energy relaxation in water at interfaces and in confined environment. In the first funding period, we have developed an efficient method to describe molecular vibrational relaxation based on single molecule excitations and the use of new descriptors. In the second funding period, we plan to include nuclear quantum effects (NQEs), which may be important in water. We aim to develop a multi resolution scheme where the electronic structure is included with an effective force field, which accurately reproduces high-level ab initio calculations, while the NQEs are explicitly addressed with the path integral formalism. This project has ended in June 2022.

Project A5 (Completed): Heat transfer in polymer nanocomposites A multiscale approach to heat transfer in soft matter will be developed. In particular, coarse-grained models of polymer nanocomposites including graphite flakes will be built and employed to obtain and characterize relaxed structures of such materials. Atomistic details will be reinserted in these structures and heat transfer will be characterized at this level of description to obtain reference data. Then, the question will be addressed how the coarse-grained models have to be modified in order to characterize heat transport in the nanocomposites directly at the coarsened level of description. This project has ended in June 2018.

Project A6: Coarse-grained models for dynamically asymmetric liquid mixtures under non-equilibrium conditions he main goal of this project is to gain better insight into the mapping of time-dependent properties of complex molecular systems, when studied using multiscale simulations. While the mapping of length scales is inherently defined by the coarse-graining procedure, the mapping of dynamic processes involves a complex combination of factors due to both the removal of degrees of freedom as well as approximations made in determining the coarse-grained (CG) interactions based on a reference all-atom (AA) model. As a consequence, the development of dynamically-consistent CG models is particularly challenging when various dynamic processes on different time scales coexist. To investigate these issues, we have focused on two important classes of systems, liquid crystals (LCs) and ionic liquids (ILs), which pair technological relevance with appropriate dynamics, and still show well defined modes of motion despite their significant complexity. In […]