Simulation BE-SONOS using TCAD Silvaco(Atlas Simulator)

[FarahAqila]

Good Morning. I have run BE-SONOS using silvaco. but the charge trap(nitride) is not doing the trapping job.

This is example of my coding.

# (c) Silvaco Inc., 2013
# Example, of dynamic sonos code,  charging by tunnelling
# followed by erasing by tunnelling.
go atlas

set vgate=16

mesh auto width=0.25

x.m  loc=0.0  spac=0.025
x.m  loc=0.1  spac=0.005
x.m  loc=0.2  spac=0.005
x.m  loc=0.3  spac=0.025

y.m  loc=-0.021  spac=0.0005
y.m  loc=-0.016  spac=0.00025
y.m  loc=-0.011  spac=0.00025
y.m  loc=-0.002  spac=0.00005
y.m  loc=0.0     spac=0.001
y.m  loc=0.1     spac=0.01
y.m  loc=1.0     spac=0.2

region num=1 silicon y.min=0.0
region num=2 oxide y.max=0.0
region num=3 nitride y.min=-0.016 y.max=-0.011 x.min=0.1 x.max=0.2
region num=4 material=Sapphire y.min=-0.011 y.max=-0.002 x.min=0.1 x.max=0.2

electrode name=gate y.min=-0.021 y.max=-0.021 x.min=0.1 x.max=0.2
electrode name=source y.min=0.0 y.max=0.0 x.max=0.05
electrode name=drain  y.min=0.0 y.max=0.0 x.min=0.25
electrode name=substrate bottom

doping uniform conc=1e18 p.type region=1
doping gauss   conc=1e20 n.type junc=0.05 x.right=0.1 ratio.lat=0.1 region=1
doping gauss   conc=1e20 n.type junc=0.05 x.left=0.2  ratio.lat=0.1 region=1

#struct outfile=sonosex02.str
#tonyplot sonosex02.str

interface n.i

# Reducing the effective mass in the barrier increases the tunnelling rate
material material=oxide semiconductor nc300=1e19 nv300=1e19 taun0=1.0 taup0=1.0 mc=0.4 mv=0.4

# Change the mobilities to get different transport inside the film
material material=nitride semiconductor nc300=1e19 nv300=1e19 mun=1.0e-1 mup=1.0e-1 mc=0.4 mv=0.4 taun0=1.0 taup0=1.0

#Material of sapphire 
material material=Sapphire semiconductor nc300=1e19 nv300=1e19 taun0=1.0 taup0=1.0 mc=0.5 mv=0.3 affinity=1.80 permitti=9.3 

contact name=gate n.poly

nitridecharge  nt.p=0.0 nt.n=1.0e10 tau.n=1.0e-4 tau.p=1.0e-4 sigmat.n=1.0e-16 sigmat.p=1.0e-14 sigman.p=1.0e-15 sigmap.n=1.0e-14 
#elec.depth=1.5 hole.depth=1.5 pf.barrier=2.5

# probe the total charged stored in the Nitride layer

probe name=nettcharge sonos.charge integrate y.min=-0.0165 y.max=-0.0105 x.min=0.1 x.max=0.2

###########################Charging by tunneling#############################################
#############################################################################################

models srh bgn cvt dt.cur dt.meth=1 dt.cbet F.BE=2.93e+08 F.AE=2.45e-06 print nearflg
#fldmob  boltzmann 

method carriers=2 

solve init sonos

solve prev 

output sonos.rates val.band con.band band.temp

#probe name=nettcharge sonos.charge integrate y.min=-0.015 y.max=-0.008 x.min=0.099 x.max=0.201

# Charge up the Nitride layer by tunnelling
log outf=sonosex03_charge.log j.tun sonos.curr 
solve vgate=$vgate ramptime=1e-9 tstep=1e-12 tfinal=1e-9  proj 

method dt.min=1.0e-9
solve tstop=1.0e-8 proj
save outf=sonosex03_charging_m8.str
solve tstop=1.0e-7 proj
method dt.min=1.0e-8
solve tstop=1e-6 proj
save outf=sonosex03_charging_m6.str
method dt.min=1.0e-7
solve tstop=1e-5 proj
method dt.min=1.0e-6
solve tstop=1e-4 proj
save outf=sonosex03_charging_m4.str
method dt.min=1.0e-5
solve tstop=1e-3 proj
method dt.min=1.0e-4
solve tstop=1e-2 proj
save outf=sonosex03_charging_m2.str
method dt.min=1.0e-3
solve tstop=1.0e-1 proj
solve tstop=1.0 proj
save outf=sonosex03_charging_m0.str master

# Solve for values of 0 V bias on each contact
# the sonos parameter on solve init preserves the
# large fixed charge in the nitride layer and prevents
# numerical problems caused by steady state compensation.


tonyplot -overlay sonosex03_charging_m8.str sonosex03_charging_m6.str sonosex03_charging_m4.str sonosex03_charging_m2.str sonosex03_charging_m0.str -set sonosex03_0.set

tonyplot -overlay sonosex03_charge.log -set sonosex03_2.set


quit

I have used VARIOT tunneling in order to enhance performance.

Can anyone help me fix this problem?