powerex05.in : Guard Ring Breakdown Analysis
Requires: SSuprem 4/S-Pisces
Minimum Versions: Athena 5.22.3.R, Atlas 5.34.0.R
This example demonstrates the fabrication and electrical analysis of a protection structure using guard rings. It shows:
- Protection device with guard rings process simulation using Athena
- Proper guard ring contact definition for breakdown analysis using Atlas
- Transport model definition including impact ionization
- Breakdown test definition
The breakdown structure considered here consists of a series of four guard rings sandwiched between cathode and anode electrodes. The Athena simulation begins by defining a phosphorous doped silicon substrate and its associated mesh. Silicon dioxide is then deposited and etched to expose silicon for guard ring and anode implantation. Boron is implanted and diffused to define the guard ring and anode junction depth. Boron doped polysilicon is then deposited and etched to form the guard ring plate electrodes. Aluminum is deposited and etched to create the anode and cathode electrodes. These electrodes are specified using the electrode statement. This includes the name and position. Notice that a substrate electrode is added by specifying the backside option. The completed structure is saved and plotted using TonyPlot.
The Atlas breakdown simulation is now described. In this simulation, the structure created by Athena will be automatically loaded into Atlas when the command go atlas is reached. For this breakdown simulation including guard rings, the guard ring contact must be allowed to float. The contact statement specifies that a 1E20 ohm series resistor be added to each contact. This allows each guard ring contact to float and ensures that there is no appreciable current flow. The bipolar option to the models statement selects a default set of transport models for bipolar devices. The default bipolar models are conmob: concentration dependent mobility, fldmob: lateral electric field dependent mobility, bgn: band gap narrowing effects, consrh: concentration dependent Shockley-Read-Hall recombination lifetimes, and auger: recombination accounting for high level injection effects. Since breakdown in semiconductor devices can occur in the presence of impact ionization, this model should be enabled. In this case, the impact selb statement enables the Selberherr impact ionization model. Please refer to the Atlas user's manual for a detailed description of these and other models.
When simulating devices with external components like the guard ring plate resistors, the newton solution method is required. When using large bias steps, or when operating a device near breakdown, it is advised to specify the trap option to the method statement. With this option enabled, Atlas automatically reduces the applied bias when convergence is not achieved. The simulation then continues at this new bias point. If convergence is achieved, Atlas tries the original bias point again. If not, the applied bias is reduced again. This will continue until convergence is achieved or until a maximum number of reductions takes place. Once convergence occurs, Atlas will automatically work its way back to the original bias point where the initial reduction occurred.
The Atlas solve statement is now used to construct the breakdown test. As in most cases, an initial zero bias solution is performed by selecting the init option of the solve statement. This gives Atlas an initial guess for subsequent simulations. An output log file is opened and will contain terminal characteristics for each bias point selected for test or until another output log file is opened. For this breakdown test, the anode voltage is stepped from -1V to -900V in three stages: -1V to -5V, -5V to -25V, and -25V to -900V. This arrangement is designed to provide good convergence while minimizing total simulation time.
The solution at the final bias point is saved as a structure file and TonyPlot is invoked to plot the results. Anode, cathode and substrate currents can be plotted to identify the voltage where breakdown occurs. Various contour quantities such as electric field, potential, and impact generation rate can be displayed from the structure plot.
To load and run this example, select the Load button in DeckBuild > Examples. This will copy the input file and any support files to your current working directory. Select the Run button in DeckBuild to execute the example.
# (c) Silvaco Inc., 2022 go athena # line x loc=0.00 spac=1.0 line x loc=95.0 spac=.5 line x loc=96.0 spac=3.0 line x loc=99.0 spac=.5 line x loc=110.0 spac=.5 line x loc=119.0 spac=.5 line x loc=127.0 spac=.5 line x loc=130.0 spac=3.0 line x loc=133.0 spac=.5 line x loc=155.0 spac=.5 line x loc=163.0 spac=.5 line x loc=166.0 spac=3.0 line x loc=169.0 spac=.5 line x loc=193.0 spac=.5 line x loc=201.0 spac=.5 line x loc=204.0 spac=3.0 line x loc=207.0 spac=.5 line x loc=233.0 spac=.5 line x loc=241.0 spac=.5 line x loc=244.0 spac=3.0 line x loc=247.0 spac=.5 line x loc=280.0 spac=.5 line x loc=295.0 spac=3.0 line x loc=310.0 spac=3.0 line x loc=315.0 spac=3.0 line x loc=335.0 spac=3.0 line x loc=345.0 spac=3.0 line x loc=350.0 spac=5.0 # line y loc=0.00 spac=0.1 line y loc=2.00 spac=0.5 line y loc=12.0 spac=2.0 line y loc=50.0 spac=5.0 line y loc=120.0 spac=10. # init c.phosphor=6e13 orientation=100 space.mult=4 # deposit oxide thick=2.50 dy=0.40 # etch oxide left p1.x=99 # etch oxide start x=127 y=-3 etch oxide cont x=127 y=0 etch oxide cont x=133 y=0 etch oxide done x=133 y=-3 # etch oxide start x=163 y=-3 etch oxide cont x=163 y=0 etch oxide cont x=169 y=0 etch oxide done x=169 y=-3 # etch oxide start x=201 y=-3 etch oxide cont x=201 y=0 etch oxide cont x=207 y=0 etch oxide done x=207 y=-3 # etch oxide start x=241 y=-3 etch oxide cont x=241 y=0 etch oxide cont x=247 y=0 etch oxide done x=247 y=-3 # implant boron dose=1.0e15 energy=100 # method fermi compress diffus time=420 temp=1100 nitro # etch oxide right p1.x=335.00 # deposit poly thick=.8 dy=0.40 c.boron=1e19 # etch poly left p1.x=95.00 # etch poly start x=119 y=-4 etch poly cont x=119 y=0 etch poly cont x=130 y=0 etch poly done x=130 y=-4 # etch poly start x=155 y=-4 etch poly cont x=155 y=0 etch poly cont x=166 y=0 etch poly done x=166 y=-4 # etch poly start x=193 y=-4 etch poly cont x=193 y=0 etch poly cont x=204 y=0 etch poly done x=204 y=-4 # etch poly start x=233 y=-4 etch poly cont x=233 y=0 etch poly cont x=244 y=0 etch poly done x=244 y=-4 # etch poly start x=280 y=-4 etch poly cont x=280 y=0 etch poly cont x=310 y=0 etch poly done x=310 y=-4 # etch poly right p1.x=345.00 # deposit alumin thick=.8 dy=0.40 # etch aluminum start x=110.0 y=-8.0 etch aluminum cont x=110.0 y=0.0 etch aluminum cont x=315.0 y=0.0 etch aluminum done x=315.0 y=-8.00 # electrode name=anode x=0 electrode name=cathode x=350 electrode name=plate1 x=150 electrode name=plate2 x=180 electrode name=plate3 x=220 electrode name=plate4 x=260 electrode name=substrate backside structure outfile=powerex05_0.str tonyplot powerex05_0.str -set powerex05_0.set # go atlas contact name=plate1 resist=1e20 contact name=plate2 resist=1e20 contact name=plate3 resist=1e20 contact name=plate4 resist=1e20 models bipolar print impact selb output e.field method newton trap maxtraps=10 solve init log outf=powerex05_1.log solve vanode=-1 vstep=-1 vfinal=-5 name=anode solve vanode=-5 vstep=-5 vfinal=-25 name=anode solve vanode=-25 vstep=-25 vfinal=-900 name=anode save outf=powerex05_2.str tonyplot powerex05_1.log -set powerex05_1.set tonyplot powerex05_2.str -set powerex05_2.set quit