Substrate geometry CdTe solar cells with catalytically-grown nano-rough surfaces
- PDF / 305,811 Bytes
- 6 Pages / 432 x 648 pts Page_size
- 65 Downloads / 188 Views
Substrate geometry CdTe solar cells with catalytically-grown nano-rough surfaces G. Papageorgiou, J.D. Major and K. Durose Stephenson Institute for Renewable Energy / Department of Physics, University of Liverpool, Chadwick Building, Peach St, Liverpool, L69 7ZF, United Kingdom ABSTRACT Substrate geometry CdTe solar cells have been modified with the addition of metalcatalysed nano-structures in order to influence their efficiency. Conditions for the growth of Auand Bi-catalysed nanostructures were explored. The substrate devices themselves comprised indium tin oxide/CdS/CdTe/Mo foil and were developed using the MgCl2 alternative to the usual CdCl2 processing – this yielded open circuit voltages of up to 740 mV. It was demonstrated that the addition of Au-catalysed nanowires to 200 nm thick CdTe films on glass substrates decreased their optical transmission by 10%, this being significantly higher than for thick films. However, reproducibility issues with forming Bi nanostructures limited the device modification tests to the use of Au-catalysed wires, and these always acted to depress photovoltaic performance. INTRODUCTION This work explores the influence of additional surface structures on the performance of substrate geometry CdTe thin film photovoltaic (PV) devices. The conventional geometry for CdTe PV devices is the “superstrate” configuration, in which the light passes through an upward facing glass plate and a transparent conductive oxide (TCO) electrode before reaching the CdS/CdTe heterojunction. This device is in commercial production and has achieved lab efficiencies of 21.5% [1]. Recently a process innovation using MgCl2 - as a cheaper and safer alternative to the well-known CdCl2 processing agent - has been reported [2]. The alternative “substrate” geometry inverts this structure with the CdTe being grown on an opaque metal substrate and the CdS/TCO being deposited on top. This architecture has the advantage of being compatible with manufacture on lightweight metal foils. Presently, laboratory efficiencies of up to 13.6% have been reported [3] for such devices fabricated on Mo layers on glass. An emerging approach is the use of nanowire (NW) or nano-textured CdTe as an absorber. This offers the potential advantages of low materials utilization, reduced reflectance loss and short carrier transit lengths [4], and in principle permits the use of low-grade starting material. Models predict up to ~20% efficiency with CdTe [4], while 15.3% has been achieved for nanowire GaAs [5]. However for CdTe core-shell nanowire PV devices a maximum of just 2.49% has been reported, this being for Au-catalysed vapour-liquid-solid (VLS) grown CdTe nanowire devices [6]. It was postulated that the low performance was due to Au forming deep level in the CdTe, and that this prevented the effective formation of a p-n junction in the coreshell wires themselves. Bi has been used as a catalyst for CdTe [7], but not for solar cells. This work compares the feasibility of using Bi and Au catalysed structures for use with CdTe PV devices
Data Loading...
