---

The quality of silicon wafers has a strong impact on the efficiency of silicon solar cells. Two critical material characteristics are crystallization related defects and micro-cracks. The detection, quantification and rating of both types of defects in photoluminescence images by means of pattern recognition techniques are in the foci of this work. Let´s see how precise we can rate the quality of an as-cut wafer!

---

(Topic I) Micro-cracks in silicon wafers reduce the strength of the wafers and can lead to critical failure within the solar-cell production. Especially micro-cracks which are induced before emitter diffusion strongly influence the current-voltage characteristics of the solar cell. To improve accuracy of crack detection in photoluminescence and infrared transmission images of as-cut wafers machine learning techniques are applied. Moreover, the comprehensive set of wafers allows the impact of crack morphology on wafer strength and electrical quality to be investigated and to derive sorting criteria. (Topic II) The efficiency of mc-Si silicon solar cells is sensitive to variations in electrical material quality. For these reasons, a rating procedure based on photoluminescence imaging has been developed within this work. The material quality is characterized by the distribution of crystallization-related defects, which are successfully correlated with the solar cell quality. This is demonstrated by an evaluation of a broad spectrum of currently available materials in a true blind test.

---