No abstract

An alternative boron emitter diffusion process called rapid vapor-phase direct doping (RVD) is studied and applied to n-type silicon solar cells with a tunnel oxide passivated electron contact (TOPCon). The RVD emitter diffusion process occurs under an atmosphere containing only the dopant gas and hydrogen. Thus, compared with standard tribromide diffusion processes, no oxygen is present. Hence, no boron glasses form during the RVD process. Consequently, a faster diffusion process with fewer chemical treatments after the diffusion process compared with standard tribromide processes is possible. In this paper, three different RVD emitter surface dopant concentrations and dopant depths were achieved by process parameter variations. These RVD emitters were applied to TOPCon cells, and their cell characteristics were compared with profiles of TOPCon reference cells with standard boron-diffused emitters. Up to 24.0% cell efficiency, 697.6 mV open-circuit voltage, 41.8 mA/cm 2 short-circuit current density, and 82.1% fill factor were reached by the best TOPCon cell with an RVD emitter. Nevertheless, compared with the reference, all cells with RVD emitters exhibited efficiency losses. Hence, to further optimize cells with RVD emitters, in-depth characterizations were conducted. The cell efficiency of cells with an RVD emitter is mainly limited by two main reasons: First, effective carrier lifetime degradation was observed, resulting in voltage losses, and second, for RVD diffusion temperatures above 980 • C, a flattening of textured cell surfaces was detected leading to current losses. In order to overcome these issues, an adapted two-step RVD emitter diffusion process is suggested for future experiments.

No abstract

Cost reduction is still the driving force in photovoltaic industry. One promising way to tackle this issue in the crystalline silicon technology is to reduce the consumption of highly purified Si raw materials. To avoid kerf losses and to manufacture still high quality Si foils with thicknesses in the range of 50 m the porous Si approach [1, 2] is one potential solution. Several investigations could show that although Si is an indirect semiconductor such thin layers are sufficient for conversion efficiencies well above 20 % [3] given good material quality and optical confinement. At Fraunhofer ISE we have been working on high-throughput solutions for all necessary technologies such as porosification, reorganization plus Si epitaxy and lift-off. In this paper we will present the first reliable process to produce Si foils with reorganization and epitaxial thickening done in an industrially relevant inline tool. The paper will furthermore give a status report on the different technology steps

No abstract