The Encyclopedia will display computed materials properties, based on millions of calculations that have been performed worldwide. Whenever new data will be available - concerning additional calculations, new materials or novel methodology - the NOMAD Encyclopedia will make them visible to the outside world. The information about individual materials will span from structural and mechanical characteristics to their thermal behavior and from basic electronic properties to various kinds of spectroscopy. Likewise, complex queries will enable to find all possible materials in the database that exhibit desired features. To this extent, the NOMAD team develops a search engine together with a graphical user interface.
Our main source of raw data is the NoMaD Repository that already hosts results from millions of calculations. These have been produced by a variety of methods, like density-functional theory (DFT), (ab initio) molecular dynamics, time-dependent DFT, and many-body perturbation theory, utilizing many different computer packages of the materials-science community. As such heterogenous data do not allow for direct comparison, the NOMAD CoE is processesing them into a code-independent form. That way, results from different computer packages become comparable and build the backbone of the NOMAD database.
Materials properties and functions
In-depth characterization of materials concerns knowledge of their features from different perspectives. Thus we collect information about their structure and behavior upon mechanical distortion or change of temperature, their electronic structure and magnetic characteristics, their response to light and other excitations, and more. Putting this knowledge together creates insight and tells whether a material is useful in view of a certain application. It may also give us hints what kind of modification would improve its function.
Hand-shake with data analytics
New knowledge generated by the big-data analytics toolkit will be incorporated directly into the Encyclopedia. This not only concerns classification of materials, error analysis, trust level of underlying methods, but also parameters that turn out most relevant for characterizing a certain feature or material class (descriptors). The Encyclopedia will offer an interface for data processing, be it for external users or the big-data analytics team.
contact concerning general aspects of the CoE: Kylie O'Brien
contact concerning the NOMAD Encyclopedia: Georg Huhs