34
ChE 384T / 323 Lecture 16 Chemical Engineering for Micro/Nano Fabrication Next: 193nm Lithography
ChE 384T / 323
Lecture 16Chemical Engineering for Micro/Nano Fabrication
Next: 193nm Lithography
ChE 384T / 323
Meta-cresol novolak
Poly-(4-hydroxystyrene)
Polyacrylates
(aliphatic)
200 225 250 275 300 325 375350 400175
0.4
0.8
1.2
1.6
2.0
2.4
2.8
0.0
Wavelength [nm]
Ab
so
rptio
n c
oe
ffic
ien
t [1
/µm
]
ArF
193 nmKrF
248 nm
i-line
365 nm
Source: R.D. Allen et al., IBM J. Res. Develop. 41 (1/2), 95-104 (1997)
Absorption of Photoresist Polymers
ChE 384T / 323
Exposure
Step
Final
Profile
Impact of Photoresist Absorbance
on Developed Image Profile
Moderate
absorbance
High
absorbance
ChE 384T / 323
Relative Etch Rate of Polymers
Relative Etch R
ate Aliphatic
Aromatic
Carbon
ChE 384T / 323
The Ohnishi Number
Watanabe, F. and Ohnishi, Y., J. Vac. Soc. Technol. B,422 (1986)
ChE 384T / 323
• 248 and 365 nm resists are unsuitable for 193 nm
imaging because they are opaque at this wavelenth
• Etch resistance requires high carbon/hydrogen
ratio but aromatics are precluded because of their
absorption
• How do you achieve both 193nm optical
transparency and etch resistance?
193 nm Resist Materials
Challenge:
ChE 384T / 323
High C:H Ratio of Alicyclic Hydrocarbons
Structure:
Formula:
Unsaturation
Number:
C6H6 C12H16 C7H10 CnH2n+2
CH3(CH2)nCH3
4 5 3 0
The key!
ChE 384T / 323
APEX 248nm Resist Design
Tethering Function Acid Lability Base Solubility
O tBOC
CH CH2 CH
OH
CH2
Etch Resistance
ChE 384T / 323
UTexas193nm Design
Tethering Function
Etch Resistance
Acid Lability Base Solubility
C O
O
C O
C(CH3)3OH
[[ ] ]
ChE 384T / 323
Early Lithography
•Excellent image quality
•Adhesion failure
C O
O
C(CH3)3
C O
OH
0.80 0.20
Resist
+ Ø3SSbF
6
poly(NBCA-co-CBN)
•Synthesis requires metal catalyst!
ChE 384T / 323
x
O
O
OOOTHF
+
O
O
OO O
V601
70C
Trading Etch Resistance for Adhesion:
Alternating Systems: COMA
Yield :Mn :Mw :Pd :
60%
4,660
6,860
1.472
N N
COOCH3
H3C
COOCH3
CH3
CH3CH3
Waco Chemical
No Metal!
V601
Shelf Life issues?
ChE 384T / 323
Images in UT 193nm COMA Resist
Alternating co-polymer
Uzo Okoroanyanwu, Jeff Byers
ChE 384T / 323
Improving Etch Resistance
Dioxane
O
O
x
OOO
O
O
+
OO O
V601
70C
UV Absorbance0.44 mm-1
@ 193 nm
Yield :Mn :Mw :Pd :
55 %
3,400
4,340
1.276
DNBC
With JSR
ChE 384T / 323
Resist and Process Development
I-line
248nm
193nm Issues!
Basic
Chemistry
Formulation
and Process
Development
Optimization
Per
form
ance
Time
ChE 384T / 323
Fujitsu’s Acrylic Platform
CH2 C
H
C O
O
nCH2 C
H
C O
O
OO
m oCH2 C
H
C O
O
O O
pCH2 C
H
C O
OH
Acrylate Copolymers …
Free radical polymerizations
No metal
ChE 384T / 323
Acrylic Polymer Platform
HOO
OO
OO
HOCF3
CF3
OO
OO
OO
O
O
OO
F3C
CF3
OH
OO
F3C
CF3
O
O
O
Fujitsu IBM,JSR, etc.
ChE 384T / 323
Types of PAGs Used For 193 nm Lithography
AsF6-
PF6-
Rf SO3-
F3C SO3-
SO3-
S+
O
CH3
N
O
O
O S
O
O
Ar
N
O
O
O S
O
O
CF3
F3C
CF2
CF2
CF2
SO3-
NEALS
I+
S+ DPI
TPS
R
R
R
R
IONICVOLATILE
NON-VOLATILE
: too weak
for 193 nm
NON-IONIC
O OH O S
O
O
N
O
O
O S
O
O
CF3
MBT
NIT
MDT
SIT
“nonaflate”
O
SO3-
ALS
NAT.OTf
S+
O CH3
O
S+
ChE 384T / 323
193nm Resist Challenges
Pattern Collapse
Line Edge Roughness (LER)
Etch Resistance
Heisenberg Principle issue
New Defect Types
Pattern Collapse
LERm Bridging
ChE 384T / 323
Line edge Roughness
193 248
ChE 384T / 323
Simulation of a PE Bake
Blocked sites
Unreacted polymer
Latent Image Edge
Gerard Schmid
ChE 384T / 323
Influence of Base on LER
x
I(x)
Base
Acid
No base With base
J. E. Meiring, T. B. Michaelson, G. M. Schmid and C. G. Willson, Proc. SPIE, 5753(2005), to be published
Base quencher can decrease the acid sphere of influence in low contrast
regions, thereby reducing LER.
ChE 384T / 323
Exploring Base Effects
I(x)
x x x
J. E. Meiring, T. B. Michaelson, G. M. Schmid and C. G. Willson, Proc. SPIE, 5753(2005), to be published
To add base quencher seems to make the contrast higher, thereby LER
reducing.
0% base 15% base 30% base
6.61 nm RMS 5.47 nm RMS 3.89 nm RMS
ChE 384T / 323
Typical KrF system always shows moderate variation of DR……
0.01
0.1
1
10
100
1000
10000
100000
1.0 10.0 100.0
DOSE(mJ/cm2)
dis
so
lutio
n r
ate
(A/s
ec)
total
surface
middle
bottom
0< DR @ Unexposed area
Gradually increasing DR
Low contrast value
Moderate Rmax
Uniform DR inside film
0.01
0.1
1
10
100
1000
10000
100000
1.0 10.0 100.0
DOSE(mJ/cm2)
dis
solu
tion r
ate
(A/s
ec)
total
surface
middle
bottom
Typical ArF system always shows dramatic variation of DR……
No DR @ Unexposed area
No DR changes until this point
Suddenly the dissolution begins
The contrast is very high
Quite high Rmax
Different DR inside film
Comparison of ArF and KrF
ChE 384T / 323
Dissolution behavior
OO
OO
O
OD
R
KrF System
ArF System
OH
OO
This difference in the
contrast amplifies small
variations in the blend
region. “Digital” On/Off
switching phenomenon
makes the line edge rough.
ChE 384T / 323
The ArF system changes from insoluble to
soluble over a very narrow dose range
Small fluctuations are amplified and cause
huge changes in dissolution rate.
Stochastic process noise becomes line edge
roughness
The KrF system has lower contrast
Small fluctiuations cause small changes in
dissolution rate..
Is there such a thing astoo much contrast?
DR
ChE 384T / 323
Can “it” be done again at 157nm???
Perhaps….but
This time it would have been really hard!!
Could have been
ChE 384T / 323
Wavelength (nm)
157.6 193 248
248 resist 6.84 0.37
193 resist 6.86 0.47
Polystyrene* 6.20
Polynorbornene* 6.10
PMMA* 5.69
Fluorocarbon* 0.70
Absorption (mm-1) of Common Polymers
* R.R.Kunz,et al., Proc. SPIE 3678, 13 (1999).
Transmission
0.00001%!!!
• Vacuum UV
• O2, H2O absorbs at this wavelength
• Even hydrocarbons like butane and
polyethylene absorb strongly
ChE 384T / 323
n nF
FF
F
Opaque Transparent
Fluorination of Norbornane Skeleton
F
248 nm 193 nm 157 nm
?
?
How many fluorines and where to fluorinate?
ChE 384T / 323
150 155 160 165 170 175 180
Wavelength (nm)
Ab
sorb
ance
(m
To
rr-1
cm-1
)
Selective Fluorination of Norbornane
Geminal substitution at the two carbon bridge is the most
effective fluorination pattern -CF3 acrylates
F F
F
F
F
F
ChE 384T / 323
Surprising Serendipitous Discovery
A157 = 2.57 mm-1
NBHFABOC and NBHFA are surprisingly transparent
Ni (II)
n
CF3
F3C O
O
O
CF3
CF3
O
O
O CF3
F3C OH
n
H+
A157 = 1.15 mm-1
T. Chiba, et. al., J. Photopolym. Sci. Technol , 13 (2000) 657-664
ChE 384T / 323
0
1
2
3
4
5
6
7
150 160 170 180 190 200
Wavelength (nm)
Abso
rbance (m
m-1
)
(2.68 mm-1)
(2.44 mm-1)
(1.15 mm-1)
Absorbance of Fluorinated Polymers
Hexafluoroisopropyl and -trifluoromethylcarboxylic acid
are groups surprisingly transparent!
O
O
CH2 C
CF3
COOMe
CF3
CF3
O
O
O
CF3
CF3
OH
CH2 C
CH3
COOMe
(6.02 mm-1)
(5.42 mm-1)
ChE 384T / 323
Some Imaging Results
40nm
80nm90nm100nm
60nm
ChE 384T / 323
Resist and Process Development
I-line
248nm
193nm
Basic
Chemistry
Formulation
and Process
Development
Optimization
Per
form
ance
Time
We are well off the base line
60nm
157 nm
ChE 384T / 32334
Images in UT 157nm Resists
70nm
85nm
60nm
Issues include: Pace, Resist, Pellicles,
CaF2, Birefringence, Cost, etc.
Intel Announcement
