Influence of the Back Contact Thickness in Laser Patterning for Monolithic Interconnection of Devices Based on a-Si:H

25th European Photovoltaic Solar Energy Conference and Exhibition / 5th World Conference on Photovoltaic Energy Conversion, 6-10 September 2010, Valencia, Spain; 3026-3031
A method for the evaluation of electrical losses in the manufacturing of optoelectronic devices, with diode pumped solid-state lasers with temporal width of nanoseconds and wavelength in the visible spectrum (532 nm) in backscribing geometry is presented. By synchronizing laser micromachining systems and ultrafast electronics systems, the interconnection process can be controlled very precisely. This helps avoiding failures caused by mechanical and kinetics effects of the positioning systems, which play a decisive role in the interconnection and isolation of thin film photovoltaic cells of hydrogenated amorphous silicon a-Si:H in superstrate configuration. This work focuses on the back contact patterning process, commonly known as P3. The success in the P3 scribes depends in several process parameters. Film removal thresholds, number of scribes, groove-width, pulse repetition or scanning speed clearly influence the quality of the scribe. An incorrect set of parameters for the P3 scribe can lead to important electrical losses in the scribe and even result in a completely unworkable module. In this work, in addition to the previously mentioned parameters, back metallic contacts of different thickness are investigated. As will be shown, gradually increasing the contact layer thickness can lead to the worsening in the quality of an otherwise perfect P3 scribe. The investigation is primarily carried out in single solar cells. Several optical imaging techniques are used to assess the quality of the P3 scribes. In addition, by monitoring the electrical characteristics of the patterned solar cells the effect of the contact thickness can be evaluated. Finally, a-Si:H modules are fabricated using the optimum combination of back contact thickness and P3 process parameters with efficiencies 7.2 %.