Optimizing CZ Silicon Crystal Growth: Algorithmic Approach for Defect Minimization

Abstract

Silicon wafers with minimal defects are important for semiconductor manufacturing as they boost device performance and reliability. Reducing defects enhances electronic properties, decreasing device failures and increasing yields. A new algorithm is suggested in this study via computational modeling and simulation to ascertain the pulling speed necessary for silicon crystal growth with fewer defects. This algorithm can adjust the pulling speed dynamically to correctly position the boundary between vacancies and interstitials and allow enough recombination time during crystal growth before point defects reach the bulk defects nucleation temperature. Experimental validation is performed using polished silicon wafers taken from final silicon ingots, accompanied by laser particle count measurements. The results validate the effectiveness of the new algorithm in reducing defect formation during crystal growth. Furthermore, a comprehensive analysis of production speed, cost, and power consumption at different pulling speeds is provided. The suggested algorithm finds a middle ground between production efficiency and crystal quality, offering speed and cost optimization while maintaining minimal defect formation.