Modality 1: Inverse Band Structure
Modality 1 applies to cases where we have a single material system, but an astronomical number of configurations, and where the target properties can be calculated on the fly.
The approach is also called Inverse Band Structure (IBS). The IBS approach began a dozen years ago within the Solid-State Theory group at the National Renewable Energy Laboratory (NREL), under support from the U.S. Department of Energy's Office of Basic Energy Sciences.
Imagine that you have a constrained material system such as an alloy, impurity, or a nanostructure. In each of these cases, we might have a huge number of spatial configurations, (e.g., different realizations of A and B atoms on a lattice with N points).
In principle, each configuration (alloy, nanostructure, or impurity system) would have a different value of the physical property of interest (e.g., bandgap, Curie temperature, oscillator strength for absorption, defect level depth). Clearly, one cannot hope to calculate all configurations, but must calculate a few and infer the rest. The computational strategy most appropriate is the use of sampling methods (such as Simulated Annealing or Genetic Algorithms) in conjunction with the relevant quantum mechanics. Thus, we calculate a few thousand configurations and seek to find the one with the target property out of ˜1014 configurations.
In Modality 1, we assume that a configuration can be realized synthetically even if it is not a ground state, e.g., superlattice sequences or different core-shell nanostructures. Historically, NREL's Solid-State Theory group developed the IBS methodology for alloys and impurities prior to the Energy Frontier Research Center.
Also check Modality 2 and Modality 3. The three modalities are a framework for the evolving application of inverse design.