Modality 2: Design of Materials with Targeted Functionality
Modality 2 applies to cases where we have numerous—perhaps thousands—of materials, each with a single (usually ground-state) configuration, and the desired target property is complex, so it currently cannot be computed on the fly.
In this case, we use "design principles"—derived quantities that can be calculated for each material and which suggest key materials parameters that need to be obtained to get the relevant functionality. This modality applies to known materials with yet unknown functionality.
Imagine that the target functionality of interest is a complex property such as transparent conductivity (combining defect physics, absorption, and doping) or photovoltaic absorbance (combining band-structure effects with non-radiative recombination) or water-splitting ability (requiring a needed gap and a non-trivial alignment of band edges with water redox).
Imagine further that the relevant space encompasses many materials with potentially different structures—say, hundreds or a few thousand A2BX4, or all ABX3—but that only the ground-state configuration or a couple of nearby metastable states are relevant.
The appropriate strategy to identify potential materials then consists of three main steps:
Develop design principles (DP) i.e., chemical and physical factors deemed to control the target property.
Compute via high-throughput methods the quantity reflected in the DP for hundreds of candidate materials, and sort out those that have a high score on the relevant DPs.
Narrow the list to ˜5–10 materials that best satisfy the DP conditions, and launch a detailed theoretical calculation of the functionality/property in parallel with high-throughput and directed experimental synthesis plus characterization investigation of these "best of class."
This modality has focused, to date, on the challenge of developing new materials for energy, especially for photovoltaic solar energy conversion, with the goals to
Identify materials for improved p-type transparent conducting oxides, where we have demonstrated >10x previous conductivities; and
develop Earth-abundant absorbers based on iron sulfide. Here, we have shown theoretically and experimentally that adding Si and Ge can enhance performance (eliminate defects) while maintaining the desirable properties of FeS.
Also check Modality 1 and Modality 3. The three modalities are a framework for the evolving application of inverse design.