In mos2d example, we could simply adjust doping condition, but what if we want to
simulate different implant condition.
My questions are:
(1) Could DEVSIM simulate implant dose doping concentration varied by depth with difference implant condition (like different energy, angle or concentration, etc.)? or even different oxide thickness on a device?
(2) Is there an example or python package for reference?
(3) what’s formula for general use in this problem (concentration vs depth) ? any paper mentioned this?
Hello @king , Thanks for joining the forum. Unfortunately I do not have too much experience with implant conditions. Do you have access to a process simulator? There was a university software named UT Marlowe that could do implant simulation. I am not sure if FLOOPS or SUPREM-IV has this capability .
If you are able to generate numerical or analytic doping profiles, they can be input into DEVSIM via interpolation or expression. I also have a mesh refinement flow for the 2D mos example here:
Thanks for your reply. DEVSIM, this forum and your professional reply really help users a lot!
(1) I don’t have access to UT Marlowe. Is that open source? I have google it but couldn’t find specific resource for free.
(2) It seems like DEVSIM didn’t build up implant simulation but only general diffusion.
(3) I have tried 2D mos example but what i really want to know is Implant concentration profile with different screen oxide thickness (model’s formula, etc.). I found couple resource here for reference. (1.Diffused Ion Implantation Profile Calculator and Graph 2.https://alan.ece.gatech.edu/ECE6450/Lectures/ECE6450L5-Ion%20Implantation.pdf)
You are welcome. If you are interested in a process simulator capable of implant simulation, there is one here:
FLOOPS is the basis for Synopsys Sentaurus Process. The principal author of this software, Professor Mark Law, is also the author of SUPREME-IV.
UT Marlowe was a monte carlo based implant simulator from the University of Texas. I don’t know if it is still under active developement.
In principal you could model diffusion in the simulator, but it has not been done yet.
Thanks for sharing the references. There is no monte carlo simulation in DEVSIM, as shown in the figure 1 of the second reference. The
erfc functions are available and additional math functions can be easily added through the Python interface.