| Popis: |
The kesterite materials, Cu2ZnSn(S,Se)4, represent a promising class of absorber materials, for cheap, earth-abundant, non-toxic photovoltaic cells. However, the record efficiency of a device based on these materials is only 12.6 %, compared to 15 % required for the material to be commercially viable and 22.6 % for the related chalcopyrite material, CIGS. In this thesis, we consider the architecture of a typical kesterite solar cell, from the back contact to the window layer and identify possible causes of the open-circuit voltage deficit, and how this deficit can be reduced. We begin by investigating the necessity for photovoltaics as a result of climate change, caused by our use of fossil fuels, and how the technology of photovoltaics has developed. We also consider the physical principles of photovoltaics and then discuss kesterite materials and architecture and compare them to chalcopyrites. Thus, we begin by investigating the photoelectron spectroscopy of kesterite materials obtained from different synthetic routes, and determine, the band alignments of kesterite/buffer interfaces by the Anderson electron affinity rule and the more reliable Kraut method. Using the results, it is shown that the band offsets between CdS and the kesterite materials considered in this work are inappropriate for high-efficiency photovoltaics. In contrast, In2S3 shows advantageous band offsets in all cases. Simulations are then used to compare the CdS and In2S3 devices Another limit on the efficiency of kesterite photovoltaics is the formation of the n-type Mo(S,Se)2 at the back contact. The formation of this layer typically results in the formation of a reverse diode, opposed to the main photodiode, thus increasing the recombination rate of the photoholes. However, in CdTe devices, Mo is often used as a back contact, which can result in the formation of an analogous MoTe2 layer which does not seem to have this effect. By considering the effects of Ar+ ion induced defects upon single crystals or multilayers exfoliated from a single crystal, the possible reasons why MoTe2 does not have the same effect as the Mo(S,Se)2 layer are investigated. We will also consider the uppermost layer of the photovoltaic cell, the window layer, which usually consists of a transparent conducting oxide (TCO). However, most of the widely-used TCOs have considerable issues, such as scarcity, cost, and self-compensating defects. Hence the final experimental chapter will consider an alternative TCO: Ga2O3. This material has a considerably larger band gap than that of the other TCOs, making it of interest for a wide range of applications Despite this widespread interest, fundamental properties of the material are still poorly understood. Thus, in the final chapter we begin by investigating the fundamental surface properties of a β-Ga2O3 single crystal with a (2 ̅01) surface termination and show that contrary to previous reports, the material exhibits surface accumulation. We also investigate the properties of several of the other polymorphs of Ga2O3 later in the chapter. This thesis concludes by considering the impact of these findings upon the future of kesterite photovoltaics and describe the likely future development of the material and its prospects for commercial deployment. |
