Frank J. Bottari

Rapid metallization paste firing of crystalline silicon solar cells

Co-firing of crystalline silicon solar cell metal contacts in infrared conveyor furnaces is the standard of the industry today. Typical ramp rates of 60–80°C./s. and total firing times of approximately 16 to 20 seconds are used due to limitations inherent in currently available production equipment. We report on a novel industrial-scale process utilizing ramp rates as high as 400°C./s. and high cooling rates which result in total firing times of 1.09 to 1.72 seconds. Cells have been produced with this process with measured fill factors in excess of 80% and high shunt resistance. At the lower firing times in this experimental series, high fill factors were maintained but open circuit voltage (Voc) reduced indicating non-optimal back surface field (BSF) formation. This study addresses the requirements for aluminum BSF formation in very rapid co-firing.

High-ramp industrial firing processes for the metallization of crystalline silicon solar cells

Rapid firing of screen-printed metallization is a critical step in the manufacture of crystalline silicon solar cells. Improved cell results are obtained with rapid heat-up and cool-down but optimum conditions have not yet been determined due to the limitations of conventional firing furnaces in achieving high ramp rates. We report results of solar cells processed by several firing profiles. Changes in solar cell parameters were observed as small changes were made to the firing profiles. Changes in cell parameters are correlated to the physical structure of the fired contacts, for both front silver and backside aluminum, as determined by optical microscopy and SEM analysis of wafer cross sections.

PSG trapping of metal contaminants during belt furnace inline phosphorus diffusion in crystalline Si wafers

Inline wet chemical phosphorus coating and thermal diffusion in a conveyor furnace is a high-throughput alternative to the conventional batch POCl3 process. There is now significant interest in the adoption of inline diffusion on an industrial scale in order to achieve a continuous flow, low-cost process which interfaces well with other inline processing steps. A concern about inline diffusion in conveyor furnaces has been the potential for conveyor belts to contribute to wafer contamination. The objective of this study is to use Secondary Ion Mass Spectrometry (SIMS) to (i) determine if commonly used nickel alloy conveyor belts are a potential source of wafer metal contamination during the diffusion firing and (ii) to investigate the effectiveness of phosphorus coating and the resulting post-diffusion phosphosilicate glass (PSG) to act as a contaminant isolating barrier, or trapping layer. We report SIMS depth and concentration profiles of Ni, Cr, and Fe, which indicate that the phosphosilicate glass produced during the diffusion process acts as an effective trap of metal contaminants.

P‐8: Conductive and Adhesive Properties of Z‐axis Adhesives for Tail Bonding

The change in resistance of an anisotropic conducting adhesive (Z-axis adhesive) solderless joint with increasing strain in a shear configuration is investigated using a modified mechanical testing machine. No change in resistance is evident up to a load of 25 Newtons above this load the resistance increases rapidly and electrical continuity is lost. We also report on the change in resistance with strain after ageing at 85°C and 85 % Relative humidity for six days. The anisotropic conducting adhesive performs well in shear after ageing.