Inverse band structure optimization of (InAs)/(GaAs) (001) nanostructures for thermophotovoltaics

Thermophotovoltaic materials converting black-body thermal radiation to electricity often require conversion efficiency for materials with direct band gaps of 0.6 eV. Random In$_{0.53}$Ga$_{0.47}$As alloy lattice $matched$ to InP have a gap around 0.76 eV, too big for this application. Therefore, difficult to grow lattice-$mismatched$ In-rich InGaAs alloys have been attempted in the past. Here we suggest to use (InAs)$_n$/(GaAs)$_m$ {\it ordered superlattices} (rather than random), lattice $matched$ to InP substrates. Using empirical pseudopotential calculations and genetic algorithm methods we look for the sequence of InAs and GaAs {\it pure} layers that simultaneously lead to a target band gap of 0.6 eV and has a minimum in-plane stress (strain balance condition). Further, since for (InAs)$_n$ layers with n$>$5 the two-dimensional growth is unstable and SK quantum dots are seen to form, we restrict the value for the period of the InAs layers to be always lower than 5.