D. Kieven

Effect of internal surface area on the performance of ZnO∕In2S3∕CuSCN solar cells with extremely thin absorber

Solar cells with an extremely thin light absorber were realized by wet chemical preparation on arrays of ZnO nanorods. The absorber consisted of an In2S3 layer (∼20nm thickness) and its interface region with a transparent CuSCN hole conductor. By changing the length of the nanorods (0–3.3μm) and keeping the In2S3 layer thickness constant at ∼20nm, the short circuit current increased from about 2–10mA∕cm2. A marked increase of the external quantum efficiency at longer wavelengths is attributed to light scattering and a solar energy conversion efficiency of 2.5% has been demonstrated.

Current-voltage characteristics and transport mechanism of solar cells based on ZnO nanorods/In2S3∕CuSCN

Temperature dependent current-voltage characteristics in the dark and under illumination have been analyzed on up to 3.2% efficient solar cells with extremely thin absorber based on ZnO nanorods/In2S3∕CuSCN structures. The diode ideality factor and the open circuit voltage are strongly influenced on a thermal activation process. Significant enhancement of the devices efficiency by annealing at moderate temperatures has been demonstrated. After this annealing, the activation energy of the saturation current increased from 1.00to1.46eV (in the dark). Transport mechanisms at the In2S3∕CuSCN interface region are discussed.

Formation of the charge selective contact in solar cells with extremely thin absorber based on ZnO-nanorod/In2S3/CuSCN

Annealed and not-annealed solar cells with extremely thin absorber based on ZnO-nanorod /In2S3/CuSCN structures have been compared. Significantly higher external quantum efficiencies have been recorded on annealed devices. The temperature dependent current-voltage characteristics in the dark and under illumination were analyzed in detail. The short-circuit current density increased with the temperature and depended on the light intensity by a power law with a power coefficient of 0.85 that was independent of the annealing or measurement temperature. The temperature dependence of the ideality factor dominated the temperature dependencies of the diode saturation current density and of the open circuit voltage. The activation energies increased strongly after annealing. We propose that the limiting charge selective contact is driven away from the highly defective In2S3/CuSCN interface into the In2S3 layer due to stimulated by the annealing Cu diffusion.

Band alignment at Sb2S3/Cu(In,Ga)Se2 heterojunctions and electronic characteristics of solar cell devices based on them

Band offsets at Sb2S3/Cu(In,Ga)Se2 heterojunctions have been studied by x-ray and ultraviolet photoemission spectroscopy. The valence and conduction band offset have been estimated to −(0.6±0.3) eV and (0.2±0.3) eV, respectively. This result suggests Sb2S3 as a potential buffer layer material for chalcopyrite based solar cells. However, Cu(In,Ga)Se2/Sb2S3/ZnO solar cells have been investigated. While the open circuit voltage ranged up to ∼0.4–0.5 V, the short circuit current was limited to ∼1.8–4.9 mA/cm2. A photocurrent of about 30 mA/cm2 was found for negative bias. On the basis of bias dependent quantum efficiency measurements and calculations, limiting mechanisms are discussed.

Dry vacuum buffers for industrial chalcopyrite absorbers from a sequential absorber process route

First results are presented for chalcopyrite solar cells with Cd-free dry buffer layers prepared by two different vacuum deposition techniques on sequentially processed Cu(In, Ga)(S, Se)2 absorbers. The deposition techniques are namely thermal evaporation of In2S3 compound powder and the reactive sputtering of Zn(O, S) layers. Both approaches are suitable for an in-line assembly into an industrial production line. The influence of the variation of process parameters such as buffer layer thickness and annealing time on current-voltage characteristics and quantum efficiency is shown. Best cell results exceed 14% conversion efficiency for the In2S3 approach (>11% for the Zn(O, S) approach) which is comparable but slightly less than the standard CdS reference cells (>15%).