To meet future energy demand and limit global warming, researchers and industrials aim to make solar technologies more efficient and cost effective. Si-wafer based photovoltaic technology accounted for about 95 % of the total production in 2017 [1] and one of the key processing step allowing to reduce cost and increase efficiency of these solar cells is the metallization step [2]. Nowadays, screen printing is the leading metallization technology as it is simple, fast and suitable for mass production. However, this method relies on silver containing pastes, an expensive metal with a fluctuating cost [3]. Moreover, screen-printing has almost reached its limitations in term of contact quality and different ways to apply metallic contacts are required for high efficiency cell architectures and advanced cells concepts. In this context, nickel copper (Ni/Cu) electrochemical metallization, also called “plating”, is under investigation as it not only gives a solution to metal cost reduction by using cheaper metals but also has the potential to overcome screen-printing limitations [4]. Indeed, Ni/Cu plating is suitable for advanced silicon solar cell architectures and, among others interests, form very thin and non-porous metallic contacts with low contact resistivity and better aspect ratio. Despite this, Ni/Cu plating requires more improvements, especially in terms of reliability and adhesion before being implemented in mass production. Plating process optimization is only achievable through a complete understanding of electrochemical reactions mechanisms involved which is the purpose of this work. We have investigated the Ni/Cu electrochemical metallization for n-PERT (Passivated Emitter Rear Totally diffused) bifacial silicon solar cells having a front textured side and rear polished side. The different steps of the Ni/Cu plating process are the following: 1) localized laser contact opening of dielectric layers using a UV picosecond laser, 2) silicon surface deoxydation to remove silicon oxide, 3) palladium activation of the silicon surface through galvanic displacement, 4) electroless nickel deposition, 5) electrolytic copper deposition, 6) final annealing to form NiSi metallic contact. Firstly, we have demonstrated that the silicon surface morphology and composition were non-homogenous after localized laser ablation over all the laser ablation conditions. Indeed, after laser ablation, silicon oxide is still present on the silicon surface with different thicknesses over the laser opened area geometry. Consequently, palladium activation appears to be non-homogeneous with a higher number of palladium particles on the border of laser opened area. Since nickel deposition rate depends on the number of palladium particles deposited, border effects of the nickel layer have been observed. Adherence and quality of the metallic contacts are thus adversely affected. Accordingly, we have investigated in details the roles of the different parameters involved in surface deoxydation and palladium galvanic displacement reactions in order to obtain a more homogeneous activated silicon surface before nickel deposition. This allowed us to control palladium deposition process by promoting palladium nucleation over growth, enabling to have more homogeneous silicon activated surface. A changer: “”is currently being done for nickel electroless deposition and copper electrolytic deposition to increase the quality of metallic contact and silicon metal interface through a better control of deposition conditions. Moreover, a direct plating route without palladium activation, which is a challenge for metallization of bifacial solar cells, will be presented.
Read full abstract