Modeling of Polycrystalline Silicon Deposition by Siemens Process 

PolySim suite is a software designed specifically for the simulation of polysilicon deposition by Siemens process. It includes PolySim and PolySim 3D. The tools are largely oriented to engineers and researchers and do not require expertise in computational fluid dynamics from the users. The unique features of the tool are built-in thermal and chemical reactor models combined with simple and convenient Graphical User Interface (GUI), which works in “grower-friendly” terms similar to those used in real reactor operation, minimizing the possible errors and saving user time needed to set-up the problem.  Modeling capabilities include turbulent gas flow, resistive rod heating, thermal radiation, gas-phase and surface chemistry, formation of porous structure often referred to as “popcorn”, and particle (dust) formation. PolySim and PolySim 3D  are two complimentary parts of the software and can exchange data among themselves.

Right: Results of PlySim 3D visualized in built-in viewer for thorough analysis of the effect of the flow structure on the growth conditions

PolySim 3D is used for detailed study of the flow structure inside the reactor volume. Thus, it is the main tool for finding the optimal nozzle locations on base plate, nozzle diameters and flow rates that would provide uniform growth conditions. Popcorn simulation is available in PolySim3D as well, so that the user can find and analyze the popcorn locations on the rod surfaces and make optimizations in order to mitigate this issue in the most problematic spots.

PolySim is intended for parametric study and recipe development. It can be used to simulate the growth process by a specified recipe and obtain the process characteristics like productivity (kg/h), energy consumption (kWh/kg), silicon output (kg) and others. PolySim can also be used to study the effect of any process parameter (Rod temperature, TCS flow, TSC/H2 ratio, chamber size, number of rods, side wall emissivity, and many others) on the outcome, in other words, to study trends.

Software can be particularly beneficial in working on the following practical tasks:

  • Increase the reactor productivity
  • Reduce energy consumption
  • Increase triclorosilane-to-silicon conversion efficiently (yield)
  • Optimize the material quality
  • Develop new advanced technology
  • Reduce the polysilicon cost

Electric current recipe can be optimized to provide smooth temperature ramp down during the process, which is optimal for reducing popcorn formation

Overall growth process characteristics (Silicon output, Productivity, Energy consumption, Conversion etc.) as well as instantaneous
characteristics at specific time or rod diameter are available in the Results table