• Energy Research

Abstract The early stage growth mechanisms of sublimation-grown thin-film polycrystalline CdTe are evaluated by growth interrupts and ex-situ SEM/AFM analysis for growth under 100 Torr of inert gas. Development of island size, density and coverage demonstrates that growth proceeds by island nucleation, island growth and density increase, followed by coalescence, channel formation and secondary nucleation. Addition of material to the islands occurs partly by the ‘step-flow’ mechanism. Grains in the completed films are considered to arise from individual nuclei. Nucleation and coalescence models are used to explain the correlation between increased substrate temperature and increased CdTe grain size in sublimation deposited CdTe films.

ITO/ZnO/CdS/CdTe/Mo solar cells have been grown in the substrate configuration by a combination of close-space sublimation and RF sputtering. A peak efficiency of 8.01% was achieved. A two stage CdCl2 annealing process was developed, with the first stage contributing to CdTe doping and the second being linked to CdTe/CdS interdiffusion by secondary ion mass spectrometry analysis. The inclusion of a ZnO layer between CdS and ITO layers improved performance significantly (from η=6% to η=8%) by increasing the shunt resistance, RSH, from 563 Ω cm2 to 881 Ω cm2. Cross-sectional scanning electron microscopy highlighted the importance of the resistive ZnO layer as numerous pinholes and voids exist in the CdS film. Solar cell performance was also investigated as a function of CdTe thickness, with optimal thicknesses being in the range 3–6 μm. All devices were deemed to be limited principally by a non-Ohmic back contact, the Schottky barrier height being determined to be 0.51 eV by temperature dependent J–V measurements. Modelling of device performance using SCAPS predicted efficiencies as high as 11.3% may be obtainable upon formation of an Ohmic back-contact. SCAPS modelling also demonstrated that a quasi-Ohmic back-contact may be achievable via inclusion of a highly p-doped (~1018 cm−3) buffer layer, between CdTe and Mo, which also has an optimal electron affinity (4.2 eV). The evaluation of device processing and the in-depth characterisation presented here provides a number of insights towards the continued improvement of substrate cell performance.

There is renewed interest in cathodoluminescence (CL) in the transmission electron microscope, since it can be combined with low energy loss spectroscopy measurements and can also be used to probe defects, such as grain boundaries and dislocations, at high spatial resolution. Transition radiation (TR), which is emitted when the incident electron crosses the vacuum-specimen interface, is however an important artefact that has received very little attention. The importance of TR is demonstrated on a wedge shaped CdTe specimen of varying thickness. For small specimen thicknesses (<250nm) grain boundaries are not visible in the panchromatic CL image. Grain boundary contrast is produced by electron-hole recombination within the foil, and a large fraction of that light is lost to multiple-beam interference, so that thicker specimens are required before the grain boundary signal is above the TR background. This is undesirable for high spatial resolution. Furthermore, the CL spectrum contains additional features due to TR which are not part of the 'bulk' specimen. Strategies to minimise the effects of TR are also discussed.

A new approach to back contacting CdTe solar cells that uses an organic poly(3-hexythiophene-2,5-diyl) (P3HT) back contact layer is reported. The most striking benefit of P3HT was demonstrated to be through a “pinhole blocking” effect, significantly improving performance uniformity. This was demonstrated through comparison of open circuit voltage values for a large sample set (600 cells) and through measurement of a device with a graded absorber layer thickness (0.7–1.9 µm). The conversion efficiency achievable and the electrical barrier height of the contacts to the CdTe were also investigated for P3HT/Au and Au control contacts – both being tested with and without additional Cu. Temperature dependent JV measurement showed the use of P3HT reduced the barrier to (0.29–0.33 eV) from the value achievable with Au (0.39–0.42 eV), but inclusion of Cu into either of the structures gave the lowest barriers (0.21–0.22 eV). For the data sets recorded, P3HT/Au yielded higher peak efficiencies than the Au control contact. However, when Cu was included the peak performance of devices having P3HT/Cu/Au and Cu/Au contacts were comparable at 14.7% respectively but the P3HT/Cu/Au contact displayed a significantly higher average performance through increased uniformity of the device response.

The CdCl 2 treatment is a key step in CdTe solar cell fabrication. However, despite its near ubiquitous use, the process is nonideal as CdCl 2 is both expensive and potentially hazardous to utilize in processing. In this paper, we report on the development of a NH 4 Cl replacement to the CdCl 2 process, which is a low-cost noncarcinogenic alternative. Comparative cells were fabricated and compared via C -V, J-V, scanning electron microscopy, and external quantum efficiency analysis. Further process optimization led to device efficiencies of up to 11.5%, achieved using this new process, with V OC values of up to 832 mV, which is relatively high.

A simple approach, which can estimate the barrier height of non-Ohmic back contacts for CdS/CdTe solar cell by using its temperature dependent forward biased current-voltage data, is explained. The method involves modelling the forward J–V characteristics using a double exponential expression for the main junction and by a reverse biased Schottky barrier for the back contact. Cells processed with both CdCl2 and MgCl2 are compared, with the current transport phenomena in both kinds of cells also being analysed. Performance loss due to limitation of the forward bias hole current, and its dependence on the post-deposition chloride processing, is discussed. The forward current transport is mainly dominated by recombination at CdS/CdTe interfacial region with pronounced tunnelling effects. Classical Schottky-type conduction, as described by the Richardson-Schottky formula, is a good fit to the reverse biased current-voltage behaviour of an Au/CdTe junction above ∼240 K. Below this temperature, the current limiting effect due to the increasing contribution from interfacial defect states can be satisfactorily explained by Bardeen’s model for a modified Schottky type barrier at back contact interface.

Abstract The effect of (NH4)2S and CS2 chemical etches on surface chemistry and contacting in Sb2Se3 solar cells was investigated via a combination of x-ray photoemission spectroscopy (XPS) and photovoltaic device analysis. Thin film solar cells were produced in superstrate configuration with an absorber layer deposited by close space sublimation. Devices of up to 5.7% efficiency were compared via current–voltage measurements (J–V) and temperature-dependent current–voltage (J–V–T) analysis. XPS analysis demonstrated that both etching processes were successful in removing Sb2O3 contamination, while there was no decrease in free elemental selenium content by either etch, in contrast to prior work. Using J–V–T analysis the removal of Sb2O3 at the back surface in etched samples was found to improve contacting by reducing the potential barrier at the back contact from 0.43 eV to 0.26 eV and lowering the series resistance. However, J–V data showed that due to the decrease in shunt resistance and short-circuit current as a result of etching, the devices show a lower efficiency following both etches, despite a lowering of the series resistance. Further optimisation of the etching process yielded an improved efficiency of 6.6%. This work elucidates the role of surface treatments in Sb2Se3 devices and resolves inconsistencies in previously published works.