Metal oxide/semiconductor heterojunctions as carrier-selective contacts for photovoltaic applications

Abstract

Solar radiation is a vast, distributed, and renewable energy source which Humanity can utilize via the photovoltaic effect. The goal of photovoltaic technology is to minimize the true costs, while maximizing the power conversion efficiency and lifetime of the cell/module. Interface-related approaches to achieving this goal are explored here, for two technologically-important classes of light absorbers: crystalline-silicon (c-Si) and metal halide perovskite (MHP). The simplest solar cell consists of a light absorber, sandwiched between two metals with dissimilar work functions. Carrier-selective contacts (CSC’s), which are ubiquitous in modern solar cells, are added to improve the electrical performance. Solar cells require asymmetric carrier transport within the cell, which can be effected via electrostatic and/or effective fields, and CSC’s augment the asymmetry by selectively transporting holes to one contact, and electrons to the other contact.

The proper design and implementation of a CSC is crucial, as the performance, lifetime, and/or cost reduction of a solar cell can be hampered by a single interface or layer. A framework, consisting of eight core requirements, was developed from first-principles to evaluate the effectiveness of a given CSC. The framework includes some requirements which are well-recognized, such as the need for appropriate band offsets, and some requirements which are not well-recognized at the moment, such as the need for effective valence/conduction band density of states matching between the absorber and CSC.

The application of the framework to multiple silicon-based and MHP-based CSC’s revealed the difficulties of effectively designing and implementing a CSC. A poly(3-hexylthiophene)/c-Si heterojunction was found to be a near ideal hole-selective contact (HSC). Three metal oxide/c-Si heterojunctions initially expected to yield comparable electron-selective contacts (ESC’s), titanium dioxide/c-Si (TiO2/c-Si), zinc oxide/c-Si (ZnO/c-Si), and tin dioxide/c-Si (SnO2/c-Si), were instead discovered to be widely different. The TiO2/MHP heterojunction was found to be a moderately ideal ESC, and the nickel oxide/MHP (NiOX/MHP) heterojunction is expected to be a good HSC. If interfacial lead di-iodide (PbI2) is intentionally or unintentionally deposited at the interfaces of a MHP solar cell, it is expected to be detrimental to the operation of the NiOX/MHP HSC, but not to the TiO2/MHP ESC.