• Energy Research

Abstract Reducing the as-cut thickness of silicon wafers is one of the key issues to significantly lower the manufacturing costs of the photovoltaic industry. The pursuit of this objective is encouraged by the outstanding development of diamond wire sawing technology, which in addition to being twice more productive, also has great potential for further kerf reduction. However, in order to avoid higher breakage rates, it is crucial to understand how the sawing process affects the mechanical resistance of wafers as their thickness decreases. In this study, wafers of 180, 160 and 140 μm thickness were cut out of monocrystalline and multicrystalline silicon bricks. Their mechanical strength was evaluated by performing 4-line bending tests coupled with finite element simulations. The specimens were loaded in the parallel and perpendicular direction with respect to the saw marks. Because of the particular shape and orientation of the sawing-induced defects, all tested wafers are significantly weaker in parallel loading. While monocrystalline and multicrystalline wafers exhibit similar mechanical strength when bent perpendicular to the sawing marks, multicrystalline wafers are 30% less resistant in parallel loading. Finally, it is shown that the fracture stress of a wafer of a given silicon quality is independent of its thickness.

AbstractMonolike silicon wafers can achieve solar cells efficiencies close to those of CZ silicon. However, this performance is affected by the presence of 2D structural defects, especially sub-grain boundaries zones that expand in the upper part of the ingots. In the present work, the relations between the structure of different types of 2D defects, previously characterized by EBSD and synchrotron based X-ray topography, and their electrical activities are analyzed. The defects are generated independently by misorienting the seeds by various tilt angles Δθ relative to the growth direction . LBIC and PL imaging are used to quantify the surface recombination velocity (SRV) of isolated defects at different heights of the ingot. The relative effects of the differences in structure of these defects, position in ingot, and cell processing treatment are exemplified and discussed.

Passivating the contacts of crystalline silicon (c-Si) solar cells (SC) with a poly-crystalline silicon (poly-Si) layer on top of a thin silicon oxide (SiOx) is currently sparking interest for reducing carrier recombination at the interface between the metal electrode and the c-Si substrate. However, due to the interrelation between different mechanisms at play, a comprehensive understanding of the surface passivation provided by the poly-Si/SiOx contact in the final SC has not been achieved yet. In the present work, we report on an original ex-situ doping process of the poly-Si layer through the deposition of a B-rich dielectric layer followed by an annealing step to diffuse B dopants in the layer. We propose an in-depth investigation of the passivation scheme of the resulting B-doped poly-Si/SiOx contact by first comparing the surface passivation provided by ex-situ doped and intrinsic poly-Si/SiOx contacts at different steps of the fabrication process. The excellent surface passivation properties obtained with the ex-situ doped poly-Si(B) contact (iVoc = 733 mV and J0 = 6.1 fA cm−2) attests to the good quality of this contact. We then propose further STEM, ECV and ToF-SIMS characterizations to assess: i) the evolution of the microstructure and B-doping profile through ex-situ doping and ii) the diffusion profile of hydrogen in the poly-Si contact. Our results show a gradual filling of the poly-Si layer with active B dopants with increasing annealing temperature (Ta), which strengthens the field-effect passivation and enables an iVoc increase after annealing up to 800 °C. We also observe a diffusion of O from the SiON:B doping layer to the interfacial SiOx layer during annealing, that likely enhances the passivation stability of our ex-situ doped poly-Si contact with increasing Ta. Finally, we conclude that the mechanism dominating the surface passivation changes during the fabrication process of the poly-Si/SiOx contact from field-effect passivation after annealing (performed for B-diffusion in the contact) to chemical passivation after following hydrogenation of the samples (performed by depositing a H-rich silicon nitride layer)