Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Wafer identification using laser marked data matrix codes

: Krieg, A.; Schmitt, C.; Rein, C.; Nübling, S.; Schneider, P.

Volltext urn:nbn:de:0011-n-1564451 (492 KByte PDF)
MD5 Fingerprint: 61c67d5547cb2c8cbd236bc4001e2074
Erstellt am: 20.12.2014

European Commission:
25th European Photovoltaic Solar Energy Conference and Exhibition, EU PVSEC 2010. Proceedings : 5th World Conference on Photovoltaic Energy Conversion, 6-10 , September 2010, Valencia, Spain
München: WIP-Renewable Energies, 2010
ISBN: 3-936338-26-4
European Photovoltaic Solar Energy Conference and Exhibition (EU PVSEC) <25, 2010, Valencia>
World Conference on Photovoltaic Energy Conversion <5, 2010, Valencia>
Konferenzbeitrag, Elektronische Publikation
Fraunhofer ISE ()
PV Produktionstechnologie und Qualitätssicherung; Silicium-Photovoltaik; Pilotherstellung von industrienahen Solarzellen; Messtechnik und Produktionskontrolle; Industrielle und neuartige Solarzellenstrukturen; Produktionsanlagen und Prozessentwicklung

In this study, the robustness of wafer identification using very small Data Matrix codes with a size of only 0.8 x 3.2 mm² on mono- and multi-crystalline silicon wafers is investigated. The quality of different codes concerning marking depth, laser-induced damage and robustness of code detection by automatic vision inspection is investigated for the whole variety of different surfaces relevant for solar cell manufacturing. The laser marked Data Matrix codes are read-out with the SICK Image Code Reader. To allow identification of Data Matrix codes in every process step, codes with a groove depth of 18 µm minimum after texturisation are necessary. With respect to laserinduced damage the codes should not be deeper than 22 µm. We developed two laser processes which generate robust code structures while introducing only little laser damage which can be removed within an etch removal of dE = 11 µm as it is typical for a standard texturisation process. For cell processes with surface texture on mono- and multi-crystalline silicon average detection rates of 98 % are already achieved over the whole process sequence with Data Matrix codes with an element size of 150 µm.