Optimizing the design of CIGS-based solar cells: a computational approach

We have used a program called analysis of microelectronic and photonic structures-one dimensional (AMPS-I D) to design two distinct types of CuIn(1-x)GaxSe2 (CIGS)-based solar cells and study their device performance. The first type of design was a single layer CuIn(1-x)GaxSe2 absorber layer. The thickness (d) varied from 1.0 to 5 mum and device parameters like fill factor (FF), open-circuit voltage (V-oc), short-circuit current density (J(sc)) and efficiency (eta) were studied as a function of the thickness of the absorber layer and the gallium content, x. In the second type of design, the absorber layer was a bi-layered heterojunction with CuGaSe2 (CGS) as the top layer and CuInSe2 (CIS) as the bottom layer. Once again the device parameters were studied as a function of the total absorber layer thickness and the relative thicknesses of the absorber layers. In both cases, 30 nm of cadmium sulfide (CdS) and 150 nm of zinc oxide (ZnO) were used as window layer and top contact layer. respectively. The thicknesses of these layers were held constant. The front and the back contacts were assumed to be ohmic. It was found that the efficiency eta depends both on the gallium content, x, and the thickness of the absorber layers. For bi-layer cells, it is also a function of the relative thicknesses of the layers. eta, for single layered cells, was found to depend both on the absorber layer thickness and the gallium content, x. As the thickness was increased (x = constant), eta increased from similar to16% (1 mum thick absorber layer) to similar to20% (for 5 mum thick absorber layer). Any change of eta was minimal for absorber layers containing between 20 and 60% gallium. For bi-layers eta dropped sharply when a thin layer of CGS was added as a top layer. eta then increased steadily until the CGS thickness was about 90% of the total thickness. (C) 2004 Elsevier B.V . All rights reserved.