Phosphorus diffusion and activation in silicon: Process simulation based on ab initio calculations
- PDF / 157,450 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 10 Downloads / 240 Views
1070-E03-03
Phosphorus diffusion and activation in silicon: Process simulation based on ab initio calculations Beat Sahli1, Kilian Vollenweider1, Nikolas Zographos2, Christoph Zechner2, and Kunihiro Suzuki3 1 Integrated Systems Laboratory, ETH Zurich, Zurich, 8092, Switzerland 2 Synopsys Switzerland LLC, Zurich, 8050, Switzerland 3 Fujitsu Laboratories Ltd, Atsugi, 243-0197, Japan ABSTRACT We present the results of extensive ab initio calculations for phosphorus clustering and diffusion in silicon and the application of these results in a state-of-the-art process simulator. The specific defects and the parameters that are investigated are selected according to the needs of diffusion and activation models, taking into account the availability of experimental data, the capabilities of current ab initio methods and the requirements for advanced technology development. The calculated formation energies, binding energies and migration barriers are used to determine a good starting point for the calibration of a diffusion and clustering model implemented in an atomistic process simulator. The defect species V, I, P, PV, PI, PI2, P2, P2V, P2I, P3, P3V, P3I and P4V are considered in all relevant charge states. The ab initio results are discussed as well as the transfer of this information into the process simulation model and the impact on model quality. INTRODUCTION Phosphorus is used for n-type doping of source/drain regions of nMOSFETS due to its high activation. However, phosphorus suffers strong transient enhanced diffusion (TED), and therefore a good knowledge of its diffusion mechanism is needed, but yet not fully established. Generally, the need for increasingly accurate and predictive modeling of dopant diffusion and activation leads to a move away from phenomenological models to increasingly detailed physics based models, e.g. atomistic simulations based on the kinetic Monte Carlo (KMC) approach or continuum based simulations with a large number of defect species. However, such models also require a large number of physical model parameters, for many of which there is no specific experimental data available. The process of calibrating an initial model using a database of dopant profiles from varying processing conditions is well developed. However, the development of the initial model including the large number of physical model parameters is still a very challenging task. The aim of this work is to explore the use of systematic ab initio calculations to support model development. The ab initio calculations are performed according to the needs of model development, taking into account the capabilities and limitations of current ab initio methods and the availability of experimental data. The ab initio results are used together with experimental data to determine the initial parameter values for the model, which is then calibrated with the traditional approach. This is done for the example of phosphorus diffusion and clustering, but the methodology and the required software tools are developed in such a way that t
Data Loading...
