Computational intelligence method of determining the energy band gap of doped ZnO semiconductor

Abstract

Presence of a direct wide band gap and large exciton binding energy in ZnO contributes enormously to its diverse practical applications such as transparent electrodes in solar cells and guide electron devices to mention a few. Doping has been recognized as an effective way of improving the intrinsic defects associated with pure ZnO and further serves as a way of controlling the band gap of ZnO for specific application of interest. Since the direct signature of doping in a crystal structure is the distortion of lattice constants, this present work develops a support vector regression computational intelligence (SVRCI)-based model for estimating the energy band gap of doped ZnO using lattice parameters as the inputs to the model. The proposed model further investigates the effect of annealing and sample preparation conditions on the band gap of doped ZnO. The proposed model is characterized with mean absolute percentage error of 1%. The band gaps estimated by the developed model agree well with the experimental values. The precision exhibited by the developed model paves way for its application for quick estimation of energy band gap of doped ZnO and consequently relieves the experimental stress involved in band gap measurements.