Disorder and cavity evolution in single-crystalline Ge during implantation of Sb ions monitored in-situ by spectroscopic ellipsometry

Ion implantation has been a key technology for the controlled surface modification of materials in microelectronics and generally, for tribology, biocompatibility, corrosion resistance and many more. To form shallow junctions in Ge is a challenging task. In this work the formation and accumulation of shallow damage profiles was studied by in-situ spectroscopic ellipsometry (SE) for the accurate tracking and evaluation of void and damage fractions in crystalline Ge during implantation of 200-keV Sb + ions with a total fluence up to 1016 cm−2 and an ion flux of 2.1 × 1012 cm−2s−1. The consecutive stages of damage accumulation were identified using optical multi-layer models with quantitative parameters of the thickness of modified layers as well as the volume fractions of amorphized material and voids. The effective size of damaged zones formed from ion tracks initiated by individual bombarding ions can be estimated by numerical simulation compared with the dynamics of damage profiles measured by ion beam analysis and ellipsometry. According to our observations, the formation of initial partial disorder was followed by complete amorphization and void formation occurring at the fluence of about 1 × 1015 cm−2, leading to a high volume fraction of voids and a modified layer thickness of ≈200 nm by the end of the irradiation process. This agrees with the results of numerical simulations and complementary scanning electron microscopy (SEM) measurements. In addition, we found a quasi-periodic time dependent behavior of amorphization and void formation represented by alternating accelerations and decelerations of different reorganization processes, respectively. For the understanding and prevention of adverse void formation and for controlled evolution of subsurface nanocavities or cellular surface texture the in-situ monitoring of the dynamics of structural damage accumulation by the developed SE method is essential.