PhD Position: Molecular Thermodynamics and Transport Modeling for the Energy Transition — ETH Zürich

Role overview

PhD Position: Molecular Thermodynamics and Transport Modeling for the Energy Transition 100%, Zurich, fixed-term print Drucken The Molecular Engineering Thermodynamics (MET) group at ETH Zurich is looking for a doctoral student to develop and improve computational tools for the molecular scale description of mass transport in membranes with an application to separation processes.

The MET group at ETH Zurich, led by Philipp Rehner, is dedicated to linking rigorous physical molecular models to the design of sustainable processes in chemical engineering.

To bridge the scale from molecules to processes, we apply state-of-the art mathematical concepts and tools combined with highly efficient computational methods.

A particular focus is on the modeling of interfacial phenomena in process design applications.

Our technological focus is on emerging technologies for the energy transition.

Project background A sustainable supply of our energy and materials demands must be built on novel processes that feature renewable feedstocks, green energy supply, and improved energy efficiency.

An efficient design of novel processes needs to account for the interactions of molecules and materials with the process performance that occur at interfaces: e.g., adsorbent materials, heat exchanger surfaces, or membranes.

The ProMote project establishes an integrated material and process design workflow that – for the first time – incorporates rigorous molecular models for interfacial phenomena directly into the evaluation and design of processes.

To bridge the gap between the continuum world of process design and the stochastic nature of molecules, the ProMote project proposes the application of classical density functional theory – a molecular-scale continuum description of inhomogeneous systems – in process design and, therefore, to fuse the scales from molecules to processes.

To overcome the computational challenge of applying molecular models at process scales, the project combines efficient mathematical concepts like automatic differentiation with backpropagation – the same concept that powers machine learning and artificial intelligence everywhere – with rigorous physical models that are robust and interpretable due to their physical constraints.