The 79th JSAP Autumn Meeting, 2018

Presentation information

Oral presentation

16 Amorphous and Microcrystalline Materials » 16.3 Bulk, thin-film and other silicon-based solar cells

[20a-133-1~8] 16.3 Bulk, thin-film and other silicon-based solar cells

Thu. Sep 20, 2018 9:30 AM - 11:45 AM 133 (133+134)

Katsuhiko Shirasawa(AIST)

11:15 AM - 11:30 AM

[20a-133-7] The impact of etching on the fracture strength of silicon wafers cut by diamond wires

〇(P)Halubai Sekhar1, Tetsuo Fukuda1, Katsuto Tanahashi1, Hidetaka Takato1 (1.Renewable Energy Research Center, AIST, Japan.)

Keywords:Diamond wire sawing, Si wafers, Fracture stress

Crystalline silicon is an essential material for photovoltaic (PV) applications. Cutting silicon bricks into wafers is the first processing step in solar cells fabrication. To reduce the cost of wafer preparation for PV use, sawing industry has challenged two goals to reduce the saw dust (silicon kerf) and the wafer thickness. To achieve these goals, sawing industry shifted from loose-abrasive slurry to fixed-diamond abrasives on thinner wires with smaller abrasives than before [1-3].
As-sawn wafers have asymmetry in fracture strength as we reported that the strength values were quite different in tests parallel and perpendicular bending directions to saw marks [2, 3]. And it is interesting and practically important to understand how much etching will make “the parallel strength” and “the perpendicular strength” match.
After cutting 200 µm wafers using a 100-µm-core wire with 6 - 12 µm diamond abrasives, we conducted bending tests of sample wafers whose surfaces were etched by 3, 5 and 10 µm in 25% KOH aqueous solution. Also as-sawn wafers were included in the test. Fig. 1 shows that the etching of 3 and 5 µm improved the fracture strength by factors of 1.96 and 2.9 times in parallel test. On the other hand, the strength improvement was quite small (1.03 and 1.31 times) in perpendicular test. The fracture strength was approximately the same in both bending tests of 5 µm etched wafers and exactly the same in 10 µm etched wafers. We will discuss the result in the upcoming meeting.