Passivation Layers for Solar Cells
Future generations of high-efficiency solar cells rely on excellent passivation of the silicon back surface to ensure minimal surface recombination losses. Considerable R&D effort is currently being pursued to develop an alternative to the back surface field, which is today’s industrial standard for silicon solar cells. Recently, it was shown that Al2O3 films synthesized by ALD reached an excellent level of surface passivation. On high-quality float-zone wafers, excellent minority carrier lifetimes were demonstrated, corresponding to very low surface recombination rates of less than 10 cm/s.
NaMLab investigates different strategies for the Al2O3 deposition including the Physical Vapor Deposition (PVD) and the Atomic Layer Deposition (ALD) using H2O and O3 as oxidizer. Furthermore, different doping strategies were evaluated to further optimize the passivation layers in terms of recombination rate and high-temperature stability. NaMLab focus on photo-carrier lifetime (MDP) and electrical (CV) measurements to characterize the passivation layers. The applied electrical methodology is very similar to that applied in microelectronics e.g. for MOSFET structures. This allows the use of a broad expertise and specialized equipment, which is set up for microelectronic research at NaMLab. The excellent passivation of Al2O3 could be linked to a combination of a low density of interface traps (Dit) and high density of negative fixed charges (Qfix). The latter generate an electrical field to repel electrons from the surface.
Systematical Ti and Si doping of Al2O3 passivation layers is a relatively new research field. Doping in the percent-range influences the chemical coordination within the crystal and creates new defects types, which potentially carry the targeted negative charge. It could be shown that the Al2O3 passivation improves with 3 at% Si-doping, resulting in a slightly enhanced effective carrier lifetime (Fig. 2). Al2O3 passivation layers were also doped with Ti. However, no improvement in the effective carrier lifetime was observed for this material. The process window for the doping concentration was limited by 10 at% and 0.5 at% for Si and Ti, respectively. The electrical evaluation of these critical concentrations reveals the formation of interface traps at the silicon surface. These traps act as recombination centre for photo-generated carriers and significantly reduce the effective carrier lifetime. The understanding of the doping limitations is a key input for the ongoing investigations on doped Al2O3 passivation layers.
Above measurements were performed on double side ALD coated Czochralski type Si-wafers. Currently, these processes are repeated on Flow Zone substrates where charge carrier lifetimes of more than 5ms are achieved for wafers passivated with 10nm undoped Al2O3 (Fig. 3).
Cooperation: Technische Universität Dresden (IHM), FILK, FSU Jena
Contact: Dr. Ingo Dirnstorfer