Ion Implantation for SiGe Semiconductor Technology

The low cost of processing silicon complementary metal oxide semiconductor (CMOS) field effect transistors (FETs) have allowed silicon to completely dominate the present semiconductor market. The increasing need for always higher speed-performance has brought silicon-germanium (SiGe) and silicon-germanium-carbide SiGeC alloys in the focus of semiconductor research. These are promising materials for the fabrication of a wide range of devices, in particular high-speed CMOS devices can be fabricated by using standard manufacturing processes and equipment. The SiGe heterojunction bipolar transistor (HBT) has already entered the market place and the heterojunction FET (HFET) is expected to be the next SiGe device to reach commercial markets. The formation of ultra shallow junctions by ion implantation is essential for the successful processing of devices based on strained silicon on relaxed SiGe CMOS technology. The goal of this project is to analyze the behavior of ion implantation into SiGe and SiGeC alloys with different germanium fraction by experiment and simulation. It is intended to study arsenic as a n-type and boron as a p-type dopant in SiGe and SiGeC targets. The experimental data (SIMS measurements) will be used to calibrate the Monte Carlo ion implantation simulator MCIMPL-II, in order to extend the applicability of the simulator to SiGe and SiGeC target materials. The simulator is based on the binary collision approximation (BCA) method and can handle arbitrary multi- dimensional device structures consisting of various amorphous and some crystalline materials. An empirical model is used within the simulator to describe the electronic stopping of the ions. The parameters for this stopping model have to be calibrated for each dopant and target composition since these parameters strongly depend on the electronic density of the target material. Moreover, special attention will be put on the implantation- induced crystal damage which mainly depends on the dopant atom species and the implantation dose. The accurate modeling of the generated vacancies and interstitials in SiGe and SiGeC is also important for the subsequent diffusion simulation.

The low cost of processing silicon complementary metal oxide semiconductor (CMOS) field effect transistors (FETs) have allowed silicon to completely dominate the present semiconductor market. The increasing need for always higher speed-performance has brought silicon-germanium (SiGe) and silicon-germanium-carbide SiGeC alloys in the focus of semiconductor research. These are promising materials for the fabrication of a wide range of devices, in particular high-speed CMOS devices can be fabricated by using standard manufacturing processes and equipment. The SiGe heterojunction bipolar transistor (HBT) has already entered the market place and the heterojunction FET (HFET) is expected to be the next SiGe device to reach commercial markets. The formation of ultra shallow junctions by ion implantation is essential for the successful processing of devices based on strained silicon on relaxed SiGe CMOS technology. The goal of this project is to analyze the behavior of ion implantation into SiGe and SiGeC alloys with different germanium fraction by experiment and simulation. It is intended to study arsenic as a n-type and boron as a p-type dopant in SiGe and SiGeC targets. The experimental data (SIMS measurements) will be used to calibrate the Monte Carlo ion implantation simulator MCIMPL-II, in order to extend the applicability of the simulator to SiGe and SiGeC target materials. The simulator is based on the binary collision approximation (BCA) method and can handle arbitrary multi- dimensional device structures consisting of various amorphous and some crystalline materials. An empirical model is used within the simulator to describe the electronic stopping of the ions. The parameters for this stopping model have to be calibrated for each dopant and target composition since these parameters strongly depend on the electronic density of the target material. Moreover, special attention will be put on the implantation-induced crystal damage which mainly depends on the dopant atom species and the implantation dose. The accurate modeling of the generated vacancies and interstitials in SiGe and SiGeC is also important for the subsequent diffusion simulation.