Simulation of chemistry-driven growth phenomena for metastable materials
Schloss Rauischholzhausen, Germany
More details and registration is found here:
The continuing advances of computational resources and algorithms provide increasing modelling access to realistic systems. In the field of electronic structure theory, DFT has become the main workhorse for material scientists. Algorithmic challenges reside in the formulation of ever improved approximate exchange-correlation functionals to describe both short-ranged and longer-ranged/dispersive electronic interactions. Efficiency challenges reside in the modelling of non-periodic defects (e.g. kinks, steps that are generally expected to be important nucleation sites) and charged systems (e.g. charged defects) with the usual supercell geometry approach to surfaces. Thermodynamic effects are crucial at the usually elevated growth temperatures and are currently mainly considered in terms of ab initio thermodynamics (AITD) or molecular dynamics simulations. Kinetic modelling can be either based on mean-field rate equation approaches or more accurate kMC simulations that fully resolve the spatial distributions at the surface. Challenges at this level generally reside in the overwhelming number of in principle conceivable elementary processes. If kinetic parameters for these processes are derived from ab initio calculations, present kMC modelling is typically limited to lattice approaches, preventing access to incommensurate film growth or the assessment of potential surface morphological changes during growth. Alternative phase-field approaches that effectively describe such situations are riddled by unexplored or questionable averaging procedures. CFD for the description of the macroscale fluid patterns in the growth chamber are generally well established. Reactive gas-phase chemistry introduces challenging stiffness to the underlying differential equations and needs specific attention, though. Information about the growing surface enters as a boundary condition. Production work is presently restricted to empirical mean-field kinetic models for this boundary conditions, with first couplings to kMC only recently achieved.[7a,16] Corresponding methodological work is presently geared towards application in heterogeneous catalysis, and rarely found in growth modelling.[7b]
The need for low temperatures to enable growth of metastable materials is one of the main challenges in modern MOVPE procedures. Here, the incorporation of carbon into the material grown must be controlled. This leads to the necessity to understand decomposition channels for precursors in detail. For ALD, the corresponding challenge resides in the occurrence of sub-monolayer growth which can often be traced back to steric or electronic effects and can thereby be approached by first principles methods.[4c]
Objectives and Format
The main idea of this workshop is to bring together experts which work on the simulation of growth phenomena at the different scales and which due to the different employed methodology hitherto meet usually in separate conference and workshop formats. The central theme will be to introduce the current state-of-the-art and current challenges in the individual fields, and to discuss ideas of bridging between them.
The workshop will be restricted to 40-50 participants to foster intense and informal discussions. Invited experts will cover the various key modelling areas, while a few eminent experimental colleagues will be asked to present keynotes overviewing contemporary materials and processes, and defining the targets for the modelling. The intended format is three full days with 30-minute talks and ample time for discussion. To stir these discussions we plan on specifically inviting distinguished colleagues to act as discussion leaders. Young researchers will have the opportunity to present their own research projects during a poster session at the first evening.
The main aim here is thus to develop a joint strategy for an improved modelling of this complex phenomenon which is in line with the aims of the Psi-k Workgroup 2 Multiscale Methods – Multiscale Modelling. This will lead to an improvement in the networking between the different communities.
Joerg Neugebauer (Max-Planck-Intitut fuer Eisenforschung) - Organiser
Karsten Reuter (Fritz-Haber-Institut der MPG) - Organiser
Ralf Tonner (Philipps-Universität Marburg) - Organiser
Simon Elliott (Tyndall National Institute) - Organiser