Computational Material Design
Development of a Computational "Toolbox"
The predictive simulation-based optimisation of advanced materials represents a central challenge within simulation technology. Besides applications in the classical fields of engineering, the trend for miniaturisation of everyday-objects leads to the demand for highly sophisticated new materials also within the disciplines of computer science or chemical and bio-engineering. As a result, advanced simulation technology is facing the ever-growing need to provide a “toolbox” that allows the development of “tailor-made molecules and materials” through the optimisation of material behaviour predicted in large-scale simulations of processes in engineering and natural science.
Linking of traditionally separated disciplines
The toolbox links quantum mechanical, discrete and continuum modelling for the description of hard and soft matter and serves as the foundation for holistic concepts of computational material design. Our ultimate vision is to construct knowledge-based virtual test laboratories, which will allow us to optimise the microstructure of materials at the atomic, microscopic or mesoscopic scale. As a result, the overall response of newly designed hybrid material systems will be controllable and optimisable by numerical simulations.
Key ingredient 1: The Different Scales
- Molecular simulation of pulling
C.antarctica lipase B away
from a tributyrin substrate interface.
Physical phenomena on the individual scales
The physical phenomena governing the overall behaviour of materials appear on a wide range of length and time scales. The understanding of the multifield phenomena on the individual scales is hence the necessary starting point for multi-scale material modelling. Examples are atomic void clustering, dislocation nucleation, phase transitions, polymer network redistributions or crack propagation.
Computational tools on the individual scales
The computational tools developed in SimTech on the different length and time scales range from density functional theory based methods for electronic structure calculations over molecular dynamics based methods for atomistic simulations to finite element based methods for macroscopic continuum computations.
Key ingredient 2: The Scale Bridging Concept
- Multiscale simulation of necking and
texture evolution of a polycrystalline
metallic rod under tension.
Linking of methods through scale bridging
To understand the overall material behaviour, one needs to understand the interplay of the phenomena at the different scales. This challenging task is accomplished through a variety of scale bridging methods. Spatial and temporal information of various physical phenomena are coupled through the integrative toolbox based on fundamental concepts of multiscale material modelling.
Top-Down and Bottom-Up Approaches
Whereas engineers tend to incorporate information from smaller scales into their continuum scale simulations (top-down approach), chemists and physicists operate oppositely (bottom-up approach). The collaborations of the SimTech experts from these traditionally separated areas contribute to the final goal of the combination of these approaches and the bridging of all time and length scales.
Key ingredient 3: The Virtual Testing Concept
- Virtual test of the sound wave emission
by a propagating crack through a
molecular dynamic simulation.
Change of material properties
With the integrated toolbox at hand, one can now not only predict the overall behaviour of existing materials from their microstructure, but one can also test the behaviour of new materials with virtual microstructure at the atomic, microscopic or mesoscopic scale.
Virtual applications
The construction of such “virtual test laboratories” allows virtual experiments with entirely new materials and molecules. Advanced functional materials with unprecedented properties can then be identified by a systematic exploration of microstructure modifications e.g. for multi-field problems involving thermo-electro-chemo-mechanical coupling effects spanning a multi-scale cascade of length and time scales.
Key ingredient 4: Optimise the Material
- Precipitation of Cu clusters in Fe-Cu
systems is simulated with a kinetic
Monte-Carlo approach.
Design the material
The last step towards the vision of computational material design is the optimisation of the microstructure of a material with respect to desired properties and functionalities. The solution of this inverse problem by use of the virtual test laboratories and a successively growing knowledge base finally allows the design of new “tailor-made molecules and materials”.
Material systems design
The overall response of newly designed hybrid material systems can now be both controlled and optimised. Our Excellence Cluster thus significantly contributes to the future quality of integrative material systems design.
Demonstrator: Holistic Concepts for Computational Material Design
We demonstrate our vision through our proposed holistic concepts for computational material design. As described in the key ingredient,s the experts in SimTech develop a toolbox of methods for different scale modelling, scale bridging, virtual testing and optimisation. This toolbox can be applied to hard matter and soft matter.
The integrative toolbox is linked together by researchers of SimTech to develop and analyse new materials with optimised properties that are validated in predictive large-scale analyses of processes in engineering and natural science.
Directly the demonstrator of the vision "Computational Material Design"
Contributions to the vision "Computational Material Design"