The 80th JSAP Autumn Meeting 2019

Presentation information

Oral presentation

15 Crystal Engineering » 15.7 Crystal characterization, impurities and crystal defects

[18a-C212-1~11] 15.7 Crystal characterization, impurities and crystal defects

Wed. Sep 18, 2019 9:00 AM - 12:00 PM C212 (C212)

Naoki Fukata(NIMS), Kazuhisa Torigoe(SUMCO), Wataru Sugimura(SUMCO Corporation)

11:15 AM - 11:30 AM

[18a-C212-9] Transient global modeling of oxygen and carbon segregations during the pulling process of Czochralski silicon crystal growth

Xin Liu1, Xue-Feng Han1, Satoshi Nakano1, Hirofumi Harada1, Yoshiji Miyamura1, Koichi Kakimoto1 (1.RIAM, Kyushu Univ.)

Keywords:Czochralski, Impurity, Simulation

Transient global modeling of the Czochralski silicon (CZ-Si) crystal growth process is essential for understanding the dynamic behaviors of the heat and mass transport in the crystallization set-up. Furthermore, segregation of impurities and dopants could also be predicted dynamically by the transient global simulation. Excessive oxygen (O) and carbon (C) can both degrade the crystal and wafer by forming the precipitates and micro-defects. Therefore, transport and segregation of O and C must be investigated for the pulling process.
The transient global model for the crystal pulling process was developed for CZ-Si growth with the cusp-shaped magnetic field (CMF). Generation, transport, and segregation of O were considered for the crystal growing process. Incorporation, accumulation, and segregation of C were also predicted at the basis of our previous studies for CZ-Si growth. The distributions of O at the growth interface were dynamically predicted for different pulling stages. The segregation curves of O and C were plotted as the function of crystal length. The radial and axial homogeneities of O and C were investigated for the growing crystal. The O concentration along the axis decreased with the increase of the length of the crystal. Due to the continuous contamination and the lower segregation coefficient, the C concentration increased with the increase of the crystal length. The developed transient global model is also applicable for the segregation prediction of other dopants and impurities in CZ-Si growing process.