powerex24.in : Silicon Carbide (SiC) CoolMOS using Realistic Processing.
Requires: Victory Process - Victory Mesh - Victory Device Minimum Versions: Victory Process 7.76.1.R, Victory Mesh 1.9.0.R, Victory Device 1.20.0.R
In this example, we demonstrate realistic Monte-Carlo implantation, diffusion, activation and oxidation of a silicon carbide (SiC) CoolMOS device. A CoolMOS device is a modification of a standard trench gate design, but with added super-junction features to reduce the compromise between maximum breakdown voltage and low on resistance.
As with all super-junction based devices, the device breakdown and on current (I_on) characteristics are very sensitive to achieving the optimum charge balance in the super-junction region of the device. Charge balance depends on the ACTIVE dopant, not just the implanted doping levels, so care has to be taken to ensure activation levels are correctly simulated. If the super-junction charge balance pinches off the channel at too low a drain voltage, a high breakdown voltage can be achieved, but the on resistance will be too high. Conversely, if the channel pinches off at too high a drain voltage, a high on current can be achieved, but the breakdown voltage will suffer severely.
A secondary effect in these devices that must be taken into consideration, is channel punch through. This effect can be mistaken for early "soft" breakdown, since the effects are similar, but is not the result of impact ionization, merely the reduction of internal barrier heights, to the point where significant current flow occurs. An additional critical part of optimizing this device, was therefore engineering the exact location of the super-junction regions near the active drain metallurgic junction. If the p-doped pillars are located too close to the channel at the drain end, then on current suffers significantly. Place the p-doped pillars too far from the channel at the drain end and early onset punch-through occurs, which looks similar to a soft breakdown.
Also, to ensure near constant gate control of the channel, and to reduce the compromise arising from the angle of the trench, the p-doped pillar adjacent to the active channel was tapered to match the angle of the trench etch, resulting in a near constant p- channel depth.
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.
Input Deck
# (c) Silvaco Inc., 2022
# 3D CoolSiC Trench MOSFET
go victoryprocess
# Initialize Structure
INIT material=4h-SiC dopants=nitrogen dopingvalues=1e19 layout=powerex24.lay \
from="0" to="5" at=0.5 gasheight=15 depth=2 resolution=0.1 meshdepth=2
set n_factor=0.9
set p_factor=0.9
set p_pillar=-0.15
# Create basic grid
line x location=0 spacing=0.2
line x location=0.9 spacing=0.1
line x location=1.5 spacing=0.05
line x location=1.8 spacing=0.05
line x location=2.5 spacing=0.1
line x location=3.2 spacing=0.05
line x location=3.5 spacing=0.05
line x location=4 spacing=0.1
line x location=5 spacing=0.2
Line y location=0 spacing=0.05
Line y location=5 spacing=0.05
Line z location=-12.4
line z location=-12.2 spacing=0.05
line z location=-6 spacing=0.1
line z location=-3 spacing=0.5
Line z location=0 spacing=0.1
Line z location=2 spacing=0.5
cartesian mask=pColumn spacing=0.1
cartesian mask=nContac spacing=0.1
Interface material1=4H-SiC material2=SiO2 dif.parameter="segregation/N/Trn" \
dif.parametervalue="0/0"
Interface material1=4H-SiC material2=SiO2 dif.parameter="segregation/N/Seg" \
dif.parametervalue="0/0"
Interface material1=4H-SiC material2=SiO2 dif.parameter="segregation/Al/Trn" \
dif.parametervalue="0/0"
Interface material1=4H-SiC material2=SiO2 dif.parameter="segregation/Al/Seg" \
dif.parametervalue="0/0"
# n-drift Epitaxy
Epitaxy material=SiC-4H thickness=3 time=18 minutes \
dopants=nitrogen dopingvalues=3e16*$n_factor temperature=950
# n-drift multiepitaxy with p-columns implants
loop steps=14
# Epi is low n-type
Epitaxy material=SiC-4H thickness=0.5 time=3 minutes \
dopants=nitrogen dopingvalues=3e16*$n_factor temperature=950
# These implants form the p+ columns
mask mask=pColumn deltacd=$p_pillar
Implant aluminum n.ion=200000 dose=1e12*$p_factor energy=320 bca \
rotation=27 tilt=7 temperature=500
Implant aluminum n.ion=200000 dose=1e12*$p_factor energy=320 bca \
rotation=207 tilt=7 temperature=500
Implant aluminum n.ion=100000 dose=5e11*$p_factor energy=140 bca \
rotation=27 tilt=7 temperature=500
Implant aluminum n.ion=100000 dose=5e11*$p_factor energy=140 bca \
rotation=207 tilt=7 temperature=500
Implant aluminum n.ion=50000 dose=2.5e11*$p_factor energy=15 bca \
rotation=27 tilt=7 temperature=500
Implant aluminum n.ion=50000 dose=2.5e11*$p_factor energy=15 bca \
rotation=207 tilt=7 temperature=500
strip material=resist
l.end
diffuse temp=700 time=10
set dcd=0.03
loop steps=5
# Epi is low p type
Epitaxy material=SiC-4H thickness=0.5 time=3 minutes \
dopants=aluminium dopingvalues=1e15 temperature=700
# implants are the same p-dose and energies
mask mask=pColumn deltacd=($dcd)+($p_pillar)
Implant aluminum n.ion=400000 dose=5e11*$p_factor energy=320 bca \
rotation=27 tilt=7 temperature=500
Implant aluminum n.ion=400000 dose=5e11*$p_factor energy=320 bca \
rotation=207 tilt=7 temperature=500
Implant aluminum n.ion=200000 dose=2.5e11*$p_factor energy=140 bca \
rotation=27 tilt=7 temperature=500
Implant aluminum n.ion=200000 dose=2.5e11*$p_factor energy=140 bca \
rotation=207 tilt=7 temperature=500
Implant aluminum n.ion=100000 dose=1.25e11*$p_factor energy=15 bca \
rotation=27 tilt=7 temperature=500
Implant aluminum n.ion=100000 dose=1.25e11*$p_factor energy=15 bca \
rotation=207 tilt=7 temperature=500
strip material=resist
set dcd=(0.03+$dcd)
l.end
# Etch Trench
mask mask=Trench
Etch material=4h-SiC dry thickness=3 angle=86
Etch material=4h-SiC wet thickness=0.5
strip material=resist
deposit material=oxide thickness=1 min
# Trench Oxidation
Diffuse time=30 temperature=1250 dryo2
# p contact layer
mask mask=p_plus
Implant aluminum n.ion=2000000 dose=5e13 energy=60 bca rotation=27 tilt=7 \
temperature=500
Implant aluminum n.ion=2000000 dose=5e13 energy=60 bca rotation=207 tilt=7 \
temperature=500
Implant aluminum n.ion=2000000 dose=5e13 energy=120 bca rotation=27 tilt=7 \
temperature=500
Implant aluminum n.ion=2000000 dose=5e13 energy=120 bca rotation=207 tilt=7 \
temperature=500
strip material=resist
# P plus Implant Activation RTA
Diffuse time=20 temperature=1800
# n contact layer
mask mask=nContac
Implant nitrogen n.ion=4000000 dose=1e14 energy=40 bca rotation=207 tilt=7 \
temperature=500
Implant nitrogen n.ion=4000000 dose=1e14 energy=40 bca rotation=27 tilt=7 \
temperature=500
strip material=resist
# N plus Implant Activation RTA
Diffuse time=10 temperature=1800
# Deposit and Etch-back Polysilicon gate
Deposit material=polysilicon thick=1 conformal \
dopants=phosphorus dopingvalues=1e19
# Etch Excess Poly
Etch thickness=1.05 dry
# Isolation oxide Oxidation
Deposit material=oxide thick=0.75 conformal
Etch material=oxide mask="T_OUTER" dry thickness=0.8 CORNERRADIUS=0.5
# Depo Alum
Deposit material=Aluminum min thickness=1
# Electrodes
ELECTRODE NAME=source x=2.5 material=Aluminum
ELECTRODE NAME=gate x=2.5 material=Poly
ELECTRODE NAME=drain substrate
SIMULATIONMODE FLOW.DIM=3D
# Save for Victory Mesh
SAVE name=powerex24_vp
######### =================== Mesh Creation =================== #########
go victorymesh
load in=powerex24_vp
remesh delaunay
refine max.size=0.3 regions="*"
refine max.interface.size=0.02 grading="quadratic"
refine max.junction.size=0.05 grading="quadratic"
save out=powerex24_0.str
tonyplot3d powerex24_0.str -set powerex24_0.set
go victorydevice simflags="-80"
mesh infile=powerex24_0.str
contact name=gate n.poly
models material=4H-SiC consrh
models material=4H-SiC consrh auger fermi ni.fermi print
output band.temp band.param
method pas itlimit=50 norm.scaling.local \
px.tol=1e-10 cx.tol=1e-10 pr.tol=1e-20 cr.tol=1e-20
# Vt Unsaturated
solve init
solve previous
solve vdrain=0.1
solve vstep=0.25 vfinal=0.75 name=gate
log outfile=powerex24_0.log
solve vstep=0.25 vfinal=10 name=gate
log off
# Vt Saturated
solve init
solve previous
solve name=drain vstep=0.5 vfinal=10
log outfile=powerex24_1.log
solve vgate=0
solve name=gate vstep=0.25 vfinal=10
tonyplot powerex24_0.log powerex24_1.log
log off
go victorydevice simflags="-160"
mesh infile=powerex24_0.str
models material=polysilicon consrh
models material=4H-SiC consrh auger fermi ni.fermi print
impact material=4H-SiC selb e.side
contact name=gate n.poly
output band.temp band.param
method pam.gmres itlimit=50 norm.scaling.local \
ir.tol=1.e-45 ix.tol=1.e-45 \
px.tol=1.e-25 pr.tol=1.e-40 \
cx.tol=1.e-25 cr.tol=1.e-33
solve init
solve previous
log outfile=powerex24_2.log
solve name=drain vstep=0.5 vfinal=10
solve name=drain vstep=2 vfinal=150
solve name=drain vstep=25 vfinal=1600
method pam.gmres itlimit=50 norm.scaling.local \
cx.tol=1e-10 px.tol=1e-10 \
ix.tol=1e-33 ir.tol=1e-33 \
cr.tol=1e-21 pr.tol=1e-21
contact name=drain current
solve previous
solve istep=1.25 ifinal=1e-6 imult name=drain
save outfile=powerex24_1.str
tonyplot powerex24_2.log -set powerex24_1.set
tonyplot3d powerex24_1.str -set powerex24_2.set
