We have investigated the surface and near-surface bulk chemical properties of differently stressed (Au/Cu/)CdTe/CdS thin-film solar cells. The employed Au/Cu contact design created unique samples in which both the back contact surface/interface and the (back) surface of the CdTe absorber were exposed. X-ray photoelectron spectroscopy and soft x-ray emission spectroscopy were employed to determine the chemical composition of the surface and near-surface bulk of both the exposed CdTe surface and the Au/Cu back contact surface of each sample. Sulfur is observed to have migrated (from the CdS) into the CdTe bulk, forming Au2S and Cu2-xS compounds under the back contact. Chlorine is found at the surface and in the near-surface bulk of both the back contact and the exposed CdTe absorber. Under the back contact, Cl is observed to form Cu-Cl bonds only in the samples that were heat stressed under AM 1.5 illumination.

Developing a high-quality transparent back contact, while maintaining efficient light transmission through the top absorber layer, are key components for achieving high-efficiency II–VI polycrystalline thin-film tandem solar cells. Combining these two elements, we fabricated ultra-thin bifacial CdTe solar cells (0.68μm) with ZnTe:N/ITO transparent back contact and achieved efficiencies of 5.7% and 5.0% with illumination from the glass and the contact side, respectively. Device analysis, using (J–V) and QE measurements, show that the loss in efficiency is due to higher RS and J0 as well as lower, side-dependent, photons absorption.

An AI-doped ZnO front contact was successfully used for the first time for the fabrication of high efficiency CdSiCdTe thin-film solar cells. The Zn0:Al films werc deposited on aluminosilicate glass by RF sputtering from a ZnO:Al,O, target. Thc ZnO:Al film has -95% average transmission in the visible spectrum with -3 ohmisquare sheet resistance. The CdS and CdTe thin films were also deposited by RF sputtering and devices were completed with a vapor CdCI, treatment and evaporated CuiAu back contacts. The highcst processing temperature was 387 “C, reached during the vapor CdC12 treatment, The devices were tested at NREL with efficiency of 14.0% which is a record for an all-sputtered CdSiCdTe solar cell. The ZnObased cell had Jsc of 23.6 mAicm’ compared to 20.7 mAlcm’ for our recent NREL-tested 12.6% cell on a commercial soda-liinc-glass/SnO~:F substrate. Othcr paramctcrs of the 14% ZnO based cell are: FF = 73.25% and V,, = 814 mV. The improved performance is almost entirely due to higher current because of better optical and electrical properties of Zn0:Al TCO. We report also on the relative stability between devices on Sn02:F and Zn0:AI TCO, under one-sun light soak at Vac.

Introduction Copper is an important dopant element in CdTe thinfilm solar cells. It is a deep acceptor in CdTe and is commonly used to obtain a heavily doped, low-resistance back-contact. However, Cu is also a fast diffuser in CdTe and can accumulate at the CdS/CdTe junction. It is suspected of leading to cell performance degradation in some cases. A substantial, thermally restorable “aging” behavior was also found in Cu-doped CdTe. The present study is designed to help identify the lattice location and amount of Cu in CdTe.

The effects of thermal annealing on CdTe/CdS heterojunctions have been investigated by means of the grazing incidence x-ray scattering technique. Changes in the interface morphology due to heat treatment are characterized quantitatively in terms of a set of parameters including the interfacial roughness and correlation lengths of the layer thickness fluctuations.

This report describes work done by the University of Toledo during the first year of this subcontract. During this time, the CdTe group constructed a second dual magnetron sputter deposition facility; optimized reactive sputtering for ZnTe:N films to achieve 10 ohm-cm resistivity and {approximately}9% efficiency cells with a copper-free ZnTe:N/Ni contact; identified Cu-related photoluminescence features and studied their correlation with cell performance including their dependence on temperature and E-fields; studied band-tail absorption in CdS{sub x}Te{sub 1{minus}x} films at 10 K and 300 K; collaborated with the National CdTe PV Team on (1) studies of high-resistivity tin oxide (HRT) layers from ITN Energy Systems, (2) fabrication of cells on the HRT layers with 0, 300, and 800-nm CdS, and (3) preparation of ZnTe:N-based contacts on First Solar materials for stress testing; and collaborated with Brooklyn College for ellipsometry studies of CdS{sub x}Te{sub 1{minus}x} alloy films, and with the University of Buffalo/Brookhaven NSLS for synchrotron X-ray fluorescence studies of interdiffusion in CdS/CdTe bilayers. The a-Si group established a baseline for fabricating a-Si-based solar cells with single, tandem, and triple-junction structures; fabricated a-Si/a-SiGe/a-SiGe triple-junction solar cells with an initial efficiency of 9.7% during the second quarter, and 10.6% during the fourth quarter (after 1166 hours of light-soaking under 1-sun light intensity at 50 C, the 10.6% solar cells stabilized at about 9%); fabricated wide-bandgap a-Si top cells, the highest Voc achieved for the single-junction top cell was 1.02 V, and top cells with high FF (up to 74%) were fabricated routinely; fabricated high-quality narrow-bandgap a-SiGe solar cells with 8.3% efficiency; found that bandgap-graded buffer layers improve the performance (Voc and FF) of the narrow-bandgap a-SiGe bottom cells; and found that a small amount of oxygen partial pressure ({approximately}2 {times} 10{sup {minus}5} torr) was beneficial for growing high-quality films from ITO targets.

The high reflectivity, fluid “plasma annealing” phase of semiconductors, particularly Si, subjected to short, intense pulses of laser or electron or ion beam irradiation is known to exhibit a combination of properties for which no adequate explanation has previously been given. These include an almost crystalline Raman spectrum, a lattice temperature that is an extremely non-linear function of absorbed energy density, an optical band gap with no detectable free carrier absorption, a flat absorption spectrum above the gap. We propose a bose condensation of the carriers excited by the irradiation into a state having properties similar to those of a superconductor to explain these anomalies.