A Breakthrough in Plating for Solar Cell Metallization

Abstract

Today’s world PV market is dominated by standard crystalline solar cells (so-called Al-BSF cells) and part of the market is shifting to PERC solar cells. The shift is obtained by introducing three additional process steps to the standard process (rear side cleaning, passivation and laser opening), and allows a gain of typically 1% absolute in efficiency. Next generation c-Si technologies should feature higher voltage solar cells with higher efficiency and less processing steps in the manufacturing, allowing for further cost reduction, both at the PV panel level and for the final cost of solar electricity. AMPERE, H2020 funded project, (http://www.ampere-h2020.eu/) focuses on technologies with such a potential and capitalizes on the high tech investments made in Europe over the last decade for establishing advanced manufacturing processes for crystalline silicon heterojunction (SHJ) solar cells and modules. A main cost factor in such cell processing, as expected, was found to be the silver based metallization. Estimating a silver paste consumption of 200mg, for a bifacial SHJ cell, and a silver paste cost conservatively estimated at 800 USD/kg for low-temperature paste, in a 23% efficiency cell, the silver cost will be around 3.0c$/Wp accounting for almost the 20% of the total industrial cost of the SHJ bifacial cell. Silver should be replaced on a large-scale basis by a more cost-effective material. Copper (Cu) is the intended substitute. The expected introduction of Cu into mass production is delayed mainly due to the big progress in screen printing and also due to some technical issues in reliability, adhesion that have to be resolved as also appropriate equipment and processes need to be available with the appropriate footprint and productivity. Comparing the state art of PV industrial plating a breakthrough plating process for SHJ copper metallization will be presented. Selective processing technique based on dynamic liquid drop/meniscus (DLD/DLM) allows metallization of solar cell fingers and bus bars without using any kind of lithography [1]. Figure 1.A presents the 2D view of how the DLD works. The system is composed, in a principle implementation, by an internal jetting outlet where a liquid flow is forced, and an external recalling inlet where by a depression the liquid is recalled back into the system. The input channel, confined by rigid wall (i.e. solid material), pumps a constant liquid flux that, depending on the input nozzle dimension, fixes the velocity of the liquid exiting the input channel. Due to a lower pressure in the surrounding of output channel, the airflow (gas) sustains the liquid (figure 1.A: black arrows in the red liquid) forming the DLD. The dynamic characteristics is due to the constantly refreshment of the liquid, during time, inside the drop. As the substrate gets in contact with the DLD, a DLM is formed, as shown in figure 1.B. Such technique allows to touch the surface of a solar cell only in specific defined positions and to perform electrochemical plating treatments in a localized manner. In DLM speed of liquid is in the meter per second range, like in a jet-plating technique, allowing to have fast deposition rate of metals. Due to the fast speed of liquid flow, the plating rate of copper is in the range of micron per seconds. In figure 1.C is shown an 100 micron width meniscus formed on glass substrate and the direction of movement of the DLM is indicated by the arrow. The ongoing result of EU Horizon2020 AMPERE project relative to copper metallization on ITO with fingers down to less than 60 micron width will be presented showing how by special reduction technique of ITO plus a thin barrier layer of nickel (figure 1.E and 1.D shows SEM cross-section of ITO reduced with thin layer of plated Ni), is possible without damaging the solar cell life time, obtain very good specific contact resistance (i.e. < 0.3mΩcm2) with 10 micron of copper and good adhesion (i.e. >2N/mm). Moreover, due to the fast plating condition, it will be shown that the equipment footprint can be interestingly reduced down to less than 18sqm for a 100MW line managing very low amounts of chemical. Due to its unique characteristics, DLD/DLM technique thanks to speed of processing, localized processing, small footprint and reduced chemical consumption, will represents a breakthrough in the market of plating for the solar cell applications. [1] M. Balucani, et al.; “New Selective Processing Technique for Solar Cells”, Energy Procedia (2013) 43, 54-65 Figure 1