Design changes can mitigate skyrocketing metals prices.
The Holy Grail of photovoltaic production (and, therefore, adoption) is reducing cost so that the price tag for solar-generated electricity is comparable to that of traditional grid power. The industry has been on good pace to deliver on this requirement, and still is, except for a small speed bump. The bump’s name is silver, and its meteoric price escalation has thrown a little rod into the PV price model wheel. After silicon, printed metallization inks are the second-most expensive component in silicon solar cell manufacture. And, over the past 12 months, the price of silver has more than doubled (see "Market Watch," pg. 16). (As of this writing, the price was over $35 per troy ounce.) In fact, the cost of the silicon cell front-side silver paste is now in the region of $0.20 per wafer, whereas a year ago it was less than $0.10. But take heart. By employing some novel technologies and in the future perhaps some different materials, silver consumption can be significantly reduced.
The accuracy and repeatability of modern screen technology and metallization print platforms have already helped reduce silver consumption significantly by condensing the front-side conductor width to 70 µm and lower. Indeed, three years ago when the average width was in the range of 120 to 140 µm, many thought this unachievable. And, while sophisticated printing technology has vastly improved on required paste volume and cell efficiency, even more can be done to lower requisite silver use.
The first approach for silver volume reduction is achieved by disassociating the main use of front-side silver – the busbars – from the printing of the collector fingers by doing the printing in two stages. This technique, known as Print-on-Print (PoP), enables manufacturers to control the amount of silver used in the busbars independently from the volume used in the rest of the grid. The PoP process utilizes a two-stage print approach that yields higher, narrower collector fingers for reduced shadowing and greater conversion efficiency. By design, this technique then also permits independent control of the busbar height, as the busbars are generally printed with the second material print, where a thinner layer of material can be applied, thereby reducing the silver volume used for the busbars. By my calculations, for a standard two-busbar design at 2.0mm wide each on a 156mm wafer, simply reducing the busbar height from 30 µm to 20 µm will yield a reduction in silver paste consumption of approximately 28mg. This represents a significant cost savings of more than 10%.
Another strategy for silver reduction is a process originally developed by the Energy Research Centre of the Netherlands (ECN) and commonly referred to as dual printing. This method, which has seen a growing level of interest in the solar industry, uses a standard mesh screen to print the busbars in the first operation and a single layer electroformed nickel stencil to print high-aspect-ratio collector fingers in the second print pass. Like PoP, this enables the cell manufacturer to control the volume of the silver in the busbars, reducing it to the bare minimum required to achieve ample conductivity and provide a connection point for the tabbing ribbons. Another advantage to the dual print technique is that the fingers only need to be printed once to achieve a high aspect ratio, which eliminates the requirement for precise matching of the stencil and screen. There is a disadvantage to dual print at the moment, which is that there are currently very few pastes on the market optimized for stencil printing, which requires a minimum of post-print paste slump to achieve the best possible aspect ratios. The paste should shear thin when the squeegee moves across the stencil, pour into the apertures and then recover instantly to rebuild its structure. Development of materials that can accomplish this is certainly possible. It just hasn’t been perfected yet and, therefore, makes current adoption of dual print a bit challenging.
Last, many solar scientists are evaluating a complete change of material – either to nickel or to copper – eliminating silver entirely. The challenge with copper is that it leaches into the silicon, so a passivation layer or a diffusion barrier is required, and nickel has to be plated prior to copper plating. And, because copper is subject to various corrosive elements, it then has to be passivated and made solderable, which generally involves a top layer plating of tin. So, now, instead of one screen printing process for silver, there are three electroplating processes with various chemistries and cleaning steps inbetween. At least for the near term, the additional process steps of this approach may negate any savings potentially achieved by moving to copper. According to the International Technology Roadmap for Photovoltaics (ITRPV), however, there are implications that copper plating may start hitting the mainstream solar market by 2015.
Though many predict that silver consumption will be dramatically reduced (perhaps halved) come 2014, implementing some of the aforementioned conservation measures now will go a long way toward total solar power cost reductions today.
Tom Falcon is a senior process development specialist at DEK Solar (dek.com); This email address is being protected from spambots. You need JavaScript enabled to view it.. His column runs bimonthly.
