Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur Graph Search.
One of the key elements to improve mainstream crystalline silicon (c-Si) solar cell performance is surface passivation, which is at the center of the ongoing transition from cells with direct silicon-metal contacts to full area passivating contacts. In theory, these contacts combine an optimal passivation and current extraction. Among those, the fired passivating contact (FPC) studied in this thesis is a promising candidate. Based on a 1.3 nm thin tunneling oxide capped with a doped SiCx layer, it is compatible with the high temperature firing process typically used in industry for metallization. Moreover, its fabrication does not require a long annealing step, reducing its thermal budget.In this thesis, effective minority carrier lifetime measurements were combined with advanced characterization techniques, such as Secondary Ion Mass Spectrometry (SIMS), to investigate the surface passivation mechanism. Chemical passivation was found to be predominant, achieved through hydrogen diffusion from a SiNx:H reservoir layer to the c-Si/SiOx interface where it accumulates. The composition of the interfacial oxide was found to influence the surface passivation quality too, reaching implied open circuit voltage (iVoc) values around 710, 720 and 740 mV with chemical, UV-O3 and thick thermal oxides, respectively. After firing, the boron-doped SiCx(p) layer was observed to consist of 5-10 nm large c-Si grains embedded in a carbon rich amorphous matrix.Surprisingly, it was found that the firing step can induce shallow bulk defects within the used Float-Zone (FZ) wafers, which can be passivated by hydrogen. Kinetics experiments indicate a rapid passivation of the c-Si/SiOx interface, happening in less than 1 min, followed by a slower passivation of the bulk. The kinetics of the hydrogenation process seem to be limited by the available hydrogen supply, rather than by its diffusivity within the bulk. In addition, the possibility to hydrogenate both surfaces with a SiNx:H layer deposited on only one side of the wafer was demonstrated, providing further flexibility to the fabrication process of FPC based solar cells.Pursuing the same goal, diffusion of fluorine from SiCx(p):F layers upon firing was tested. The results indicate that this approach can also passivate the c-Si/SiOx interface even though the iVoc after firing seems to be limited by the creation of shallow bulk defects. Nevertheless, values over 730 mV could be reached with UV-O3 SiOx layers upon subsequent hydrogenation.Regarding the stability of this passivation, no degradation was observed for FPCs upon 80 days of light soaking at 80°C. Moreover, it remained stable upon thermal treatments of 30 min up to 170°C.Finally, metallization of these FPCs was studied. A layer stack with a total thickness of ~90 nm was developed to be contacted by a screen printed Al paste fired through the SiNx:H layer. Unfortunately, so far good contact resistivities could only be achieved at the expense of passivation. Closer microstructural investigations revealed diffusion of Al from the metallic paste into the SiNx:H layer, forming an Al-N compound, and in some cases accumulating in the SiOx layer. Nevertheless, using a low temperature metallization approach, solar cells featuring co-fired SiCx(n) and SiCx(p) FPCs at the front and rear were processed, reaching up to 20.5% in efficiency. This corroborates the potential of FPCs for high efficiency solar cells with a simple and low-cost fabrication process.
Christophe Ballif, Franz-Josef Haug, Andrea Ingenito, Audrey Marie Isabelle Morisset, Philippe Wyss, Mario Joe Lehmann, Père Roca i Cabarrocas