Sheet resistance characterization of laser-doped lines on crystalline silicon wafers for photovoltaic applications

• Published in: Solar energy materials and solar cells 2011-03, Vol.95 (3), p.974-980
• Main Authors: Wang, Kee Soon, Tjahjono, Budi S., Wong, Johnson, Uddin, Ashraf, Wenham, Stuart R.
• Format: Article
• Language: English
• Subjects: Applied sciences; Crystal structure; Direct energy conversion and energy accumulation; Electrical engineering. Electrical power engineering; Electrical power engineering; Electrical resistivity; Energy; Exact sciences and technology; Forming; Four-point probe; Laser doping; Lasers; Natural energy; Photoelectric conversion; Photovoltaic cells; Photovoltaic conversion; Sheet resistance characterization; Silicon; Solar cell; Solar cells; Solar cells. Photoelectrochemical cells; Solar energy; TLM; Wafers
• ISSN: 0927-0248; 1879-3398
• DOI: 10.1016/j.solmat.2010.12.002

An improved technique of measuring the upper sheet resistance ( R sheet) limit of a single laser-doped (LD) line has been presented. Probing pads of a structure based on the transfer length method (TLM) principle are formed by laser doping without any further metallization step, forming a fast, simple, and inexpensive measurement technique that is applicable to both textured and planar silicon wafers with or without any silicon nitride (SiN x ) dielectric. It has been demonstrated that within the scope of using a laser to form lines with R sheet less than 15 Ω/□, this TLM method seems to be an accurate and reliable method to characterize the sheet resistance of a single laser-doped line. Investigations here also suggest that the sheet resistance of a laser-doped rectangular area (LD box) formed by overlapping multiple laser-doped lines on crystalline silicon wafers and measured using the four-point probe cannot accurately or reliably represent the sheet resistance of a single LD line that is laser doped under the same laser-scribing condition. This is mainly because of (1) the tendency of the high power laser to introduce electrically detrimental cracks onto the laser-doped box, (2) the unknown appropriate spacing between adjacent LD lines that should be used when forming the LD box, and (3) the ambiguity of overdoping that makes an LD box unsuitable to represent a single LD line at all. The TLM method proposed here not only avoids these problems but is also able to self-justify its own predicted upper limit values. ▪