edu11
Newbie level 4
I'm trying to simulate a SONOS gate capacitor with n substrate. The idea is to charge it with FN or DT tunneling and then observe the CV. So far the code i've written is this:
go atlas
mesh width=1.0
x.m loc=0.0 spac=0.005
x.m loc=0.5 spac=0.005
y.m loc=-0.1 spac=0.02
y.m loc=-0.03 spac=0.005
y.m loc=-0.02 spac=0.002
y.m loc=-0.005 spac=0.002
y.m loc=0.0 spac=0.002
y.m loc=0.1 spac=0.01
region num=1 silicon y.min=0.0
region num=2 oxide y.min=-0.005 y.max=0.0 x.min=0 x.max=0.5
region num=3 nitride y.min=-0.02 y.max=-0.005 x.min=0 x.max=0.5
region num=4 oxide y.min=-0.04 y.max=-0.02 x.min=0 x.max=0.5
electrode name=gate x.min=0 x.max=0.5 y.min=-0.06 y.max=-0.04
electrode name=substrate bottom
doping uniform conc=1e17 n.type region=1
material material=oxide semiconductor taun0=1.0 taup0=1.0 mc=0.35 mv=1.0 nc300=1e19 nv300=1e19
contact name=gate n.poly
# This enables the sophisticated SONOS model, DYNASONOS (this flag is true by default)
interface n.i dynasonos
# The trap densities in the nitride are set by nt.n (electrons)
# and nt.p (holes). If the trap density for a carrier type is
# set to zero then the trap equation is not solved for that carrier
# type.
# The capture cross-sections (sigmat.n and sigmat.p) can be used to
# control the position of the charge centroid in the Nitride. The
# opposite carrier cross-sections (sigman.p and sigmap.n) are less
# important. The emission rates obtained from tau.n and tau.p will
# affect the overall trapped charge density, but are fixed for this example.
nitridecharge nt.p=0.0e20 nt.n=5.0e19 tau.n=1.0e-4 tau.p=1.0e-4 elec.depth=1.5 sigmat.n=${xs} sigman.p=0.0 sigmap.n=0.0
# the sonos.charge field on the probe statement retuns the net trapped charge
probe name=nettcharge sonos.charge integrate y.min=-0.0075 y.max=-0.002 x.min=0.1 x.max=0.2
method dt.max=1.0e-5
models srh
solve init
solve prev
#struct outfile=history01.str
#tonyplot history01.str
models cvt srh print
output con.band val.band t.quantum
method carr=2 maxtrap=10
solve outfile=sonos1_temp.str master
log outf=sonos_1.log
solve vgate=0 vstep=0.1 vfinal=5 name=gate ac freq=1e6 direct
save outf=sonos_1.str
log off
tonyplot sonos_1.str
extract init inf="sonos_1.log"
extract name="tox_cl" (3.9*8.85e-10*1.0e-4)/max(c."gate""substrate")
Can somebody give me certain hints? How will i introduce the FNORD current? Also, i'm having a hard time with the dynasonos model. Am i declaring the parameters correctly?
go atlas
mesh width=1.0
x.m loc=0.0 spac=0.005
x.m loc=0.5 spac=0.005
y.m loc=-0.1 spac=0.02
y.m loc=-0.03 spac=0.005
y.m loc=-0.02 spac=0.002
y.m loc=-0.005 spac=0.002
y.m loc=0.0 spac=0.002
y.m loc=0.1 spac=0.01
region num=1 silicon y.min=0.0
region num=2 oxide y.min=-0.005 y.max=0.0 x.min=0 x.max=0.5
region num=3 nitride y.min=-0.02 y.max=-0.005 x.min=0 x.max=0.5
region num=4 oxide y.min=-0.04 y.max=-0.02 x.min=0 x.max=0.5
electrode name=gate x.min=0 x.max=0.5 y.min=-0.06 y.max=-0.04
electrode name=substrate bottom
doping uniform conc=1e17 n.type region=1
material material=oxide semiconductor taun0=1.0 taup0=1.0 mc=0.35 mv=1.0 nc300=1e19 nv300=1e19
contact name=gate n.poly
# This enables the sophisticated SONOS model, DYNASONOS (this flag is true by default)
interface n.i dynasonos
# The trap densities in the nitride are set by nt.n (electrons)
# and nt.p (holes). If the trap density for a carrier type is
# set to zero then the trap equation is not solved for that carrier
# type.
# The capture cross-sections (sigmat.n and sigmat.p) can be used to
# control the position of the charge centroid in the Nitride. The
# opposite carrier cross-sections (sigman.p and sigmap.n) are less
# important. The emission rates obtained from tau.n and tau.p will
# affect the overall trapped charge density, but are fixed for this example.
nitridecharge nt.p=0.0e20 nt.n=5.0e19 tau.n=1.0e-4 tau.p=1.0e-4 elec.depth=1.5 sigmat.n=${xs} sigman.p=0.0 sigmap.n=0.0
# the sonos.charge field on the probe statement retuns the net trapped charge
probe name=nettcharge sonos.charge integrate y.min=-0.0075 y.max=-0.002 x.min=0.1 x.max=0.2
method dt.max=1.0e-5
models srh
solve init
solve prev
#struct outfile=history01.str
#tonyplot history01.str
models cvt srh print
output con.band val.band t.quantum
method carr=2 maxtrap=10
solve outfile=sonos1_temp.str master
log outf=sonos_1.log
solve vgate=0 vstep=0.1 vfinal=5 name=gate ac freq=1e6 direct
save outf=sonos_1.str
log off
tonyplot sonos_1.str
extract init inf="sonos_1.log"
extract name="tox_cl" (3.9*8.85e-10*1.0e-4)/max(c."gate""substrate")
Can somebody give me certain hints? How will i introduce the FNORD current? Also, i'm having a hard time with the dynasonos model. Am i declaring the parameters correctly?