Modeling of Doping Profile for FT-IR Based Control of Epi Silicon Layer Growth

The objective of this project is to develop a model of the growth (thickness and doping profile) of epi-silicon layers. This model will be used in conjunction with on-line Fourier Transform Infrared (FT-IR) reflectance measurements to control epi-silicon growth in a cluster tool. AFR has successfully performed a field test at Texas Instruments for single-wafer epi-silicon film deposition in the cool-down chamber of a Centura 5200 cluster tool. This field test demonstrated the ability of FT-IR based on-line wafer-by-wafer measurements of thickness and doping density. With a development of appropriate models, the FT-IR reflectance measurements could be employed to control the cluster tool deposition process on a wafer-by-wafer base. The models include: a) A model for the complex dielectric function correlated with the doping density; b) A model for the reflectance from a layer with a duffuse interface; c) A model for the diffuse interface doping profile based on the thermal diffusion and epi-layer growth rate during the film growth process; d) A model for the control of the diffusion coefficient and growth rate. The proposed technology can also be applied to improve both wafer quality and yields for the production of poly-silicon, silicon oxide, silicon nitride, and silicon shallow junctions. This project will be led by the joint venture of AFR and On-Line Technologies, Inc. (On-Line). Texas Instruments, Inc. (TI) will be a subcontractor. Phase I will develop the models for FT-IR based monitoring and control for the production of epi-silicon films and prove the feasibility of data collection during production. To validate the models, measurements will be made using an On-Line FT-IR with calibration samples fabricated and characterized by TI. AFR will measure the ex-situ film characteristics and develop the analysis method and software. On-Line will design and construct the physical and optical interface for TI's cluster tool. Field tests will be done at TI. Phase II will modify the model based on actual manufacture processes, and plan for the implantation of the model in production.