Effect of Na-doped Mo layer as a controllable sodium reservoir and diffusion barrier for flexible Cu(In,Ga)Se2 solar cells

Abstract To overcome the absence of sodium (Na) and to prevent undesired impurity diffusion from stainless steel (STS) substrates, a bi-layer molybdenum (Mo) structure consisting of a 1- μ m-thick Mo and Na-doped Mo (Mo:Na) with variable thickness was utilized as the back contact layer in copper–indium–gallium–selenide (CIGS) solar cells. Na ions were supplied to the CIGS absorber layer from the Mo:Na layer, and the amount of Na ions was controlled by changing the thickness of Mo:Na layer. The Na doping to the CIGS layer is verified by the X-ray photoelectron spectroscopy (XPS) and the dynamic secondary ion mass spectroscopy (D-SIMS). The recombination process in the bulk CIGS and the CdS/CIGS junction has been analyzed by temperature dependent current density–voltage (J–V–T) measurement. The D-SIMS showed that the amount of Na ions in the CIGS layer grown on Mo/Mo:Na/STS is comparable with that in the conventional CIGS film deposited on Mo coated soda-lime glass. Moreover, the bi-layer Mo also acts as a diffusion barrier layer against the Fe atoms, which deteriorate the quality of CIGS. The higher content of Na ions and the lower content of Fe ions in the CIGS layer lowered the recombination activation energy, series resistance and ideality factor, leading to improved junction quality. As a result, the short circuit current density and power conversion efficiency of the CIGS on the Mo and 600-nm-thick Mo:Na were improved by 8.1 and 69.4%, respectively, compared with those of CIGS solar cells fabricated without a Mo:Na layer.