ECE6466ICEngineering

Dr.WandaWosik
Chapter6
Oxidation

:
SiliconVLSITechnology
Fundamentals,PracticeandModeling
byJ.D.Plummer,M.D.Deal,
andP.B.Griffin
UH;F2014

THERMAL OXIDATION - Chapter 6

Basic Concepts
SiO2 and the Si/SiO2 interface are the principal reasons for silicons dominance
in the IC industry.
O x id e
T h ic k n ess
1 m

0 .1 m

T h er m a lly G r o w n O x id es

D ep o sited O x id e s

F ield O x id es

B a c k en d I n su la to r s
B e tw e e n M e ta l L a y e r s

M a sk in g O x id es

M a sk in g O x id es

P a d O x id es
10 nm

G a te O x id es
T u n n e lin g O x id es
1 nm

C h em ic a l O x id e s fr o m C le a n in g
N a tiv e O x id e s

SiO2:
Easily selectively etched using
lithography.
Masks most common impurities
(B, P, As, Sb).
Excellent insulator
(
).
High breakdown field
7
-1
( 10 Vcm )
Excellent junction passivation.
Stable bulk electrical properties.
Stable and reproducible interface
with Si.

No other known semiconductor/insulator combination has properties that approach

the Si/SiO2 interface.
2

ThermalOxidationandSiO2Interface
Applications
SiO2growsonSi(also@RT);enablesveryeasyICformation;
ensuresstabilityandreliability.
Year of Production

1998

2000

2002

2004

2007

2010

2013

2016

2018

Technology Node (half pitch)

250 nm 180 nm 130 nm

90 nm

65 nm

45 nm

32 nm 22 nm 18 nm

MPU Printed Gate Length

100 nm

70 nm

53 nm

35 nm

25 nm

18 nm 13 nm 10 nm

512M

1G

4G

16G

32G

64G

128G

128G

MPU Transistors/Chip (x106)

550

1100

2200

4400

8800

14,000

Gate Oxide T ox Equivalent (nm)

1.2

0.9

0.7

0.6

0.5

0.5

1.5

1.2

0.9

0.8

0.7

0.7

170

230

330

1000

1670

1670

Newdielectrics
toavoidtunneling.
(highK)

<4

<4

<4

<4

<4

<4

LowKdielectrics

0.9-1.2

0.8-1.1

0.7-1-0

DRAM Bits/Chip (Sampling)

256M

Lowerthermalbudget
12nm

MPU
Gate Oxide T ox Equivalent (nm)
Low Operating Power
Gate Dielectric Leakage
(nA/m @100C) MPU
Thickness Control (% 3 )
Min Supply Voltage (volts)

1.8-2.5

1.5-1.8

1.2-1.5

06-0.9 0.5-0.8 0.5-0.7

HistoricalDevelopmentandBasicConcepts

OxidegrowthusingO16andO18
isotopesidentifiesthemechanism.
NeutralO2andH2Oand/orOHare
dominantspeciesinoxidation,notatoms
orionsO,O,O2,
VolumeofSiO2is30%largerthanSi.

1
1

1 .3

1
1

1
S i su b str a te

1 .2
1

1 .3
1 .3

1
S i su b str a te

Oxidation involves a volume expansion ( 2.2X).

Especially in 2D and 3D structures, stress
effects play a dominant role.

(1.3)3~2.2
volumeofthe
oxidecannotbe
accommodatedin
Silicon

SiliconConsumptionDuringOxidation(LOCOS)

D ep o sited P o ly silic o n

BirdsBeakformation
O r ig in a l S i S u r fa c e

V o lu m e
E x p a n sio n

StressattheSi/Si3N4
interface

S iO 2
S i S u b str a te

StructureofSilicaGlass
Shortrangeorder
maintained

Amorphous hydrogen
Nonbridgingoxygenin
material
fusedSilica(notpresentin
crystallineSiO2)

Networkmodifier>Qm

Sicanbereplacedby
deposits.B,P,AsorSb
=networkformers.

largecompressivestress(5*109dynes/cm2)existsinSiO2.Hightemperature

canreliefstressbyviscousflow.
LargedifferenceinthethermalexpansioncoefficientsofSiandSiO2.
Siliconintensionrefertocurvature
6

 SiO2 is amorphous even though it

grows on a crystalline substrate.

(Intel Web site)

Q m

K +
N a+

S iO 2

+
T r a n sitio n
R eg io n

S ilic o n

+
+

Q ot

x
Q it

Q f

Four charges are associated with

insulators and insulator/semiconductor
interfaces.
Qf - fixed oxide charge
Qit - interface trapped charge
Qm - mobile oxide charge
Qot - oxide trapped charge

SiO2/SiSystem:StructureandCharges
(Fig3.15)amorphous/crystallineinterfaceflat.(TEM)
RoughnesswithgrowthrateandT.
Trappedcharges
S iO 2
Qotimportantwhen
isolationradiation
ispresent(space
T r a n sitio n
applications,
R eg io n
plasmaprocessing)
EPROM

S ilic o n

Q m

K +
N a+

+
x

Detectdensityattheinterfaces
is~1091011cm2.

+
+

Q ot

x
Q it

Q f

Fixedcharge~1091011cm
2
is+anddoesnotchange
indeviceoperation.
Interfacecharge=traps
duetodanglingSibonds
changeinoperation
QpQitbothrelatedto
unoxidizedSiatoms.

Deal
triangle
Reducechargessincetheydegrade
deviceoperationT,H2anneal..

OxideChargesandTheirAnnealing
HfastthroughSiO2
Alannealing:Al+OHALO+H

Qit

ttoreach
equilibrium
withT

H22H

Si+HSiH
Increasingsurfaceroughnessincreasescharges

(100)SiusedinICs

ArbetterdoesnotreactwithSi.
9

10

Q u a r tz
T ube

W a fer s

Q u a r tz C a r r ier

Oxidation systems are conceptually

very simple.
In practice today, vertical furnaces,
RTO systems and fast ramp furnaces
all find use.

R esista n c e H ea tin g

O 2

H 2

Gate Oxides

LOCOS
or STI DRAM Dielectrics

Thermal oxidation can

potentially be used in many
places in chip fabrication.
In practice, deposited SiO2
layers are increasingly being
used (lower Dt).

11

ManufacturingMethodsandEquipment
Waferloadingshoulduse
cantileverorelavators
(perpendicular)toavoid
touchingthewalls.

3zones

Verticalfurnacesarealso
used.Betteruniformity,
easierautomation,
cleanernocontactwith
thetube

+0.5C

Dryorwet
oxidation

@1000C,inwater,

RampingofTfrom/to800C(10C/sec)
AddHClorTCAforgetteringpurpose(metals,Na+)

2Xforeach100C
Temperaturecontrol+0.5C
12

13

http://www.tf.unikiel.de/matwis/amat/elmat_en/kap_6/illustr/i6_2_4.html
14

MeasurementMethods
Physical(Destructive):Etchastepandmeasure(stylus,microscopewithmonochromaticlight,
AFM,SEM,TEM)
Optical(usuallynondestructive):thickoxides(colorchart,ellipsometry,reflectance)but
forthinoxides:ellipsometry
REFLECTANCE

whiteor
monochromatic
light

Refraction
indexes

Colorchart(xox>50nm)>notdestructive
interferencewillaffectthereflectedlight>
colorcorrelatedwiththicknessofadielectric
layer(1020nmaccuracy)use standards

Formonochromaticlightminimaandmaximainthereflectedbeamallowto
determinexox(fringes,spectrometerswithsweepingwavelengthforfixedwe can
2n x cos
min,max 1 ox
find extrema)
m
Goodforafewtensofnm

Ellipsometryusespolarizedlightanddetectthechangeinpolarization
ofthereflectedlightduetoafilm(thickness,indexofrefraction)

1
n sin
sin( o
)
n1

1,2,3max
1/2min

15

ElectricalMeasurements

C-V Measurements
+ V

a)

MOSCapacitor

Powerful technique for

characterizing semiconductor/
insulator structures.

C
C O
C O

Capacitancevoltage
method

e-

accumulation
A
COX OX
X OX

N S ilic o n
D o p in g = N D

V G

dQ
acsignal
dV

xoxobtained
ChargeDensity

- V

b)

QG QD ND X D

C
C O

C O

+ + + + + + +x D

C D

depletion
A
CD S
XD

e-

QG QD ND X D QT

c)

-- V

Q G
Q I
Q D

V G

Inversion
Equilibrium
conditions.

C O
C O

+ + + + + +x D M
+ ++ +
++ +
+

ax

C D M in

Ndobtained

eH o les

V TH

VG

16

CVMeasurements

Lowfrequency(~1Hz),highfrequency(100Khz11)AC
signalsusedforCVMeasurements.)
QIfollowsQGC=COX

XD=XDmaxQDfixed

*
XD>XDmax

HolesgeneratedintheD.Land
attractedbythegatesourcetheDL
when|VG|increases

Toavoiddeep
depletion*:

ni
qn W
,J gen i
G
G
dV J gen J gen qniW

0.1V /sec
dt
C
COX ACOX

HighfrequencyACsignal
changesfasterthanQIcan
respond(generationisslow)

17

Bias-Temperature (BT) Test

18

ChargesDerived
Qm,Qothavesimilareffect
asQf(shiftcharacteristics)

ntype

Duetotraps

EitEc

StatesnearEc
Qipresent

Midgapstates
Trapscannotchargeor
dischargedonot
respondtoHFsignal

EitEv
StatesnearEv
Alwayspresent

QirespondtoDCvoltagestretchout
changeinEF(VG),chargesatEit.
Stressoftheoxide(ex.chargeinjection,
radiation)CVdegradation(timeto
breakdown,chargetobreakdown)
19

Experiment:WhynoCVchangeisobserved?
Assumegoodoxide/Siinterface.
Maskingoxidewastoothinhighdopingunder
thegate(CD)
Supplyofcarrierswouldgive,CVHF=CVLF

20

ModelsandSimulation
FirstOrderPlanarGrowthKineticsLinearParabolicModel

Boundary
layer

InsteadystateF1=F2=F3

DealandGrooveModel

ReactionatSisurface

Henrylaw
C*CO(PGPS)
OxidantsolubilityinSiO2

hGmasstransfercoefficient

Gradient CoCi

h=hG/HkT

Transportofthe
oxidanttotheoxide
surface.

TransporttoSi=Diffusion
ofO2(H2O)throughthe
oxide

Sol.sol.inSiO2
ksx0/D<<1orksx0/D>>1

21

ModelsandSimulation
FirstOrderPlanarGrowthKineticsLinearParabolic
Model

transport
reaction

ksx0/D<<1

ksx0/D>>1

CIC*

CI0

Diffusionfastcompared
tochemicalreactionfor
thinoxides.
ks&D=f(t)
Transition region
@about50200nm

Fastreaction
diffusionlimits
oxidation
(thickoxides)

Fitting
parameters
xo(t)

Generaldependenceofoxide
thicknessontime.
Linearrateconstant parabolicRateconstant
Thinoxides
Thickoxides
22

LinearandParabolicRateConstants
Experimentalresults

DerivationofA&B

Arrheniusdependence
Parabolic
Linear
RepresentSiSibondbreaking

activationenergies

E2

Parabolic

Linear

diffusion
parabolic

Breaking
Sibonds

E1

RepresentdiffusionofO2orsteam

Bwet>BdrybecauseCwet*>>Cdry*
@1100C,3*1019cm3>>5*1016cm3!! 23

Deal&Grove model
24

Wetoxidationusedforthickeroxides

Dryoxidationusedupto100200nm
2

0 .7

C*|wet>>C*|dry
1 2 0 0 C

0 .5

O x id e T h ic k n ess - m ic r o n s

O x id e T h ic k n ess - m ic r o n s

0 .6

c)

0 .4

1 1 0 0 C

0 .3

a)

1 0 0 0 C

0 .2
0 .1

1 1 0 0 C

1 .5

1 0 0 0 C
1
9 0 0 C
0 .5

b)
8 0 0 C

9 0 0 C
8 0 0 C

4
T im e - h o u r s

Calculated (100) silicon dry O2

oxidation rates using Deal Grove.

10

7 0 0 C

0
0

4
5
6
T im e - h o u r s

10

Calculated (100) silicon H2O

oxidation rates using Deal Grove.

Example: Problem 6.13 in the text: a) 3 hrs in O2 @ 1100 C = 0.21 m + b) 2 hrs in

H2O @ 900 C = 0.4 m + c) 2 hrs in O2 @ 1200 C = 0.5 m total oxide
thickness.
25

Thin Oxide Growth Kinetics

A major problem with the Deal Grove model was recognized when it was first
proposed - it does not correctly model thin O2 growth kinetics.
Experimentally O2 oxides grow much faster for 20 nm than Deal Grove predicts.
MANY models have been suggested in the literature.

1. Reisman et. al. Model

(17)
Power law fits the data for all oxide thicknesses.
a and b are experimentally extracted parameters.
Physically - interface reaction controlled, volume expansion and viscous flow
of SiO2 control growth.

2. Han and Helms Model

(18)
Second parallel reaction added - fits the data for all oxide thicknesses.
Three parameters (one of the A values is 0).
Second process may be outdiffusion of OV and reaction at the gas/SiO2 interface.
26

3. Massoud et. al. Model

(19)
Second term added to Deal Grove model - higher dx/dt during initial growth.
L 7 nm, second term disappears for thicker oxides.
Easy to implement along with the DG model, used in process simulators.
Data agrees with the Reisman, Han and Massoud models. (800C dry O2 model
comparison below.)
.

0 .0 3

O x id e T h ic k n ess - m ic r o n s

0 .0 2 5
D ea l G r o v e M o d el ( = 8 h r s)

0 .0 2

R e ism a n e t. a l. M o d e l
0 .0 1 5
H a n & H elm s M o d el
0 .0 1
D e a l G r o v e M o d e l ( = 0 )

0 .0 0 5

0
0

4
6
T im e - h o u r s

10

27

OtherModelsforPlanarOxidationKinetics>IncorporateThinOxides
DGmodelfails
Forthinoxides<20nm

linear

Models:
O2coupleddiffusionwithholes

thermionicemissionofefromSi

micropores,channelsparallel
transport
paralleloxidationwithO2,O(diffusion

andreaction)

siliconsurfacewithadditionalsites

800C

Volumeexpansionattheinterface
providedbyviscousflowmodel
applicableto2Doxidationin
nonplanarstructures
a,b=f(T,p.press.)

1)
2)
1000C

3)

B/APn n=1H2O,n=.5DryO2

4)
Massoudetal.L=7nm
Goodforthinoxides

O2throughSiO2,O2+O
@Si=parallelreactions
Odominates@lowT
O2dominates@highT

Addedforthinoxides

28

EffectofVolumeMismatchinSi/SiO2System;RecessedLOCOS
Example:

ySi

H2O@1000C;Findtimetoget
planarsurface?
2.2Xvolumeexpansion>
45%yox=ySisoyox=ySi/.45

Total oxide thickness to be grown:

yox=ySi/0.45=ySi+0.5m
ySi=0.41m
yox=0.91m

For H2O

Time for dry oxidation would be unrealistically long

29

DependenceofGrowthKineticsonPressure

Dry O2
0.1 atm

if

C*PG

B, B/A PG

dxo/dt PG

DGmodelgivesgoodmatchforwetoxidationnotfordrylinear
parabolicmodelisincompleteinconsistentwithD&Gmodel:

BP B/APn
0.5<n<1.0
ManymodelsusemodifiedSi/SiO2interfacereactionnoneis
widelyaccepted

10 atm
AcceptedsolutionistheDGmodelwithcorrection:
B/A=(B/A)iPB=(B)iPforH2O)
B/A=(B/A)iPnB=(B)iPforO2
n0.70.8

30

MixedAmbientGrowthKinetics
RoleofClinOxidationProcesses
Mixed ambient

B/A
B

f(HCl)
Twotermsincludeindependentreactionsfor
*H2OandO2(notagoodmodel)
*H2O/HCl(firsttermonly)and
*O2/HClinteractionbetweenoxidants
dependence>pressurerole

31

OrientationEffectsinOxidation
wet

dry

(100),(111),andPolysilicon
Densityofatoms(bonds)in(111)>(100)
Noeffectoforientationinthe
parabolicregime
Verythinoxidegrowthathigh
pressureandlowT(ex.800C):
(100)>(111)

Inthelinearregime
Relatedtostress

Simulatedoxidegrowth
32

C. 2D SiO2 Growth Kinetics

a)

E tc h ed S i R in g

S i S u b str a te

S id e V iew s

These effects were investigated in

detail experimentally by Kao et. al.
about 15 years ago.
Typical experimental results below.

T o p V ie w s

b)

P o ly silic o n
c)
S iO 2
Si

d)

(Kao et.al)
33

2DSiO2GrowthKinetics
Differenceinvolume>problemswhen
expansionisrestricted(SiO2confined)

ExperimentsbyKaoetal.:
Retardationatsharpcorners(2Xfor500nmSiO2)
Retardationlarger@lowT(noeffect@1200C)
Interior(concave)cornersoxidizeslowerthan
Poly-Si
for
contrast

exterior(convex)butbothslowerthanflatSi

Reasons
Crystalorientation
DiffusionofoxidantthroughamorphousSiO2isthe
same>nodependenceondirection
Stress(volumedifference):SiO2underlarge
compressivestress>affectbothoxidanttransportand
reactionattheSisurface
34

In addition, the flow properties of the SiO2 need to be described by a stress

dependent viscosity

(23)

Where
is the shear stress in the oxide and VC is a fitting parameter.
- 0 .4
SUPREME uses many fitting parameters.
.

- 0 .2

( T ) - SiO2
( T ) - Si3N4

Value
0.0125 nm3
0.0065 nm3
0
3
0.3 nm @ 850C
0.72 nm3 @ 1050C
3.13 x 1010 exp(2.19 eV/kT) poise
4.77 x 1010 exp(1.12 eV/kT) poise

S iO 2

S i3 N 4

M ic r o n s

Parameter
VR
VD
VS, VT
VC

0 .2

S ilic o n

0 .4
0 .6
0 .8

- 0 .8

- 0 .4

0
0 .4
M ic r o n s

0 .8

- 0 .8

- 0 .4

0
0 .4
M ic r o n s

0 .8

These models have been implemented in modern process simulators and allow
them to predict shapes and stress levels for VLSI structures (above right).
ATHENA simulation: Left - no stress dependent parameters, Right - including
stress dependence.
35

OxidationofNonPlanarStructures
Example

(outer)
(inner)

nostress

Stressretardsoxidation;@highT
viscoelasticflowrelaxesstress
Oxideviscosity=f(stress,T)

Stress
included

Stress@T>Stress@T

Reaction
diffusion
solubility

stress
dependent
activation
values

Relievestress:
HighTnotgoodforICs
O2/NF3?
Coronadischarge?

Viscoelasticproperties(nonlinearinT)of
polySi,Si3N4alsohavetobeincludedin
simulationsofoxidation
HistoryofStressVERYIMPORTANT
StressinanoxidedependsongrowthT.
Insequentialprocessing,transientwill
appearinthenextstep@higherTfrom
theoriginalstress(=higheratlowerT)
whichsetstheoxidegrowthratebelow
thatathighT(lowerstress).
Also
TF(irst)>TS(econd)
StressF<StressS
GrowthF>GrowthEquilibrium

36

SegregationofDopantsattheSi/SiO2Interface

B in H2
Dry O2

P, As

Ga - not
useful as a
dopant

37

Point Defect Based Models

The oxidation models we have considered to this point are macroscopic models
(diffusion coefficients, chemical reactions etc.).

O 2

D if f u s io n

H 2O
I
O x id e

There is also an atomistic picture of

oxidation that has emerged in recent
years.
Most of these ideas are driven by the
volume expansion occurring during
oxidation and the need for
free volume.

S i li c o n

Oxidation Enhanced Diffusion (OED)

Oxidation Retarded Diffusion (ORD)
Both are related to the interface reaction (B/A or ks)
Si/SiO2 - generation of I ~ 1/103 Si atoms diffuse to
Si, others oxidized

In Chapter 3 we described internal

oxidation in the following way:

Surface oxidation can be thought of in the same

way.

38

 The connection between oxidation and other processes can then be modeled as
shown below.
IncreaseinconcentrationofSiI>
OEDalsolaterally

ATHENA simulation of OED.

Original B layer was uniform

S i3 N 4

S iO 2

0
I n e r t D i ff u s i o n

M ic r o n s

O E D
0 .5

1 .0

1 .5

-2

-1

0
M ic r o n s

Oxidation injects interstitials to create free volume for the oxidation process.
Oxidation can also consume vacancies for the same reason.
These processes increase I concentrations and decrease V concentrations in
nearby silicon regions.
Any process (diffusion etc) which occurs via I and V will be affected.
39

AdvancedPointDefectsBasedModelforOxidation
O2

S u r fa c e
R e c o m b in a tio n

Relationtodopantdiffusion

Interstitials (I) diffuse far

I n er t
D i ff u s i o n

B u lk R e c o m b in a tio n

B u r ied D o p a n t M a r k er L a y er

O ED

GRdetermineanetfluxofI>effect
ondiffusioneveninaburiedlayer.
ExperimentalGRforvariouslayers(both
dependonsurfacelayer)
Reasons:
kinksonthesurface;
generation/surface
regrowth/segregation/Si/SiO2

40

SubstrateDopingEffects

ConcentrationEnhancedOxidation(CEO)
5x faster
due to dopant

2x faster
NDopantP>oxidegrowthbyB/A
notbyB;especially@lowTabout3
4Xduetoconcentrations
PropertiesofoxidedonotchangeforP
butchangeforB

LowT
HighT

CEOstrongerforN+thanP+(B/Agrows,Bdoesnot)
CEOforBoronchangesBbutnotB/Adue to incorporation
in the oxide

OxidationneedsVforvolume
expansionsofordopant
concentrations,chargedV(V,V=
N+type;V+P+type)>B/A
Dopantsegregation
N+>toSi
P+>SiO2
Interfacechangesduringoxidation
>growthratechanges

41

RecessedLOCOSSimulation

42

Complete Process Simulation of Oxidation

Many of these models (and others in Chapter 6), have been implemented in
programs like SUPREM.
Simulation of an advanced
isolation structure (the SWAMI
process originally developed by
Hewlett-Packard), using
SSUPREM IV. The structure
prior to oxidation is on the top
left. A 450 min H2O oxidation at
1000 C is then performed which
results in the structure on the
top right. An experimental
structure fabricated with a
similar process flow is shown
on the bottom right. The stress
levels in the growing SiO2 are
shown at the end of the
oxidation on the bottom left.

M ic r o n s

0
.

0 .4

0 .8

M ic r o n s

0
0 .4
0 .8
-1

0
M ic r o n s

43

SilicideOxidationSimilartoSilicon

Silicides(TiSi2,CoSi2)usedinICsinsteadof
polysilicon(400cm)localinterconnects

OxidationresultsinformationofSiO2(formost
metals)ormostlyMOxforHfSi2,ZrSi2,TiSi2
Metalbondsbreak>MdiffusestotheSi(polysilicon)substrate,releasedSiformSiO 2
Bsilicide=Bsilicon

For TiSi2 Si atoms diffuse from the poly_Si

For CoSi2, NiSi2, PtSi - metals diffuse to poly_Si

B/Asilicide>>B/Asilicon
Parabolic growth

44

PolysiliconOxidation
Ratelargerthan(100)becauseofvarious
orientationsandsizesofgrains>B/A=f(grain
size)notB
Grainboundaryimportantindoping,regrowthathighT

Si3N4GrowthandOxidationKinetics
NirideisamaskagainstinLOCOS
Si3N4growthslowinNH32.57.5nm@900C

Kooieffect

orSi2N2O

SmalldiffusionofoxidantsthroughSi3N4
45

OxidationInducedStackingFaults(PreferentialEtching)
OISFwithtimeandtemperatureoftheoxidationprocess

PreferentialetchingrevealsSpitsandOISF

Inc

reas

Surfaceandbulk
OISF

ing

etc
hin
g

tim
e

46

FormationandAnnihilationofOISFinWetandDryOxidationProcesses
OISFdependonavailableIattheSisurface

DryOxidation

Wet

47

Summary of Key Ideas

Thermal oxidation has been a key element of silicon technology since its
inception.
Thermally, chemically, mechanically and electrically stable SiO 2 layers on silicon
distinguish silicon from other possible semiconductors.
The basic growth kinetics of SiO2 on silicon are controlled by oxidant diffusion
and Si/SiO2 interface chemical reaction.
This simple Deal-Grove model has been extended to include 2D effects, high
dopant concentrations, mixed ambients and thin oxides.
Oxidation can also have long range effects on dopant diffusion (OED or ORD)
which are modeled through point defect interactions.
Process simulators today include all these physical effects (and more) and are
quite powerful in predicting oxidation geometry and properties.

48