Nanolithography with diblock copolymers:

1
Nanolithography with diblock copolymers What
Fingerprints taught us about getting long range
order in 2D.
Basic Idea Massively process 5-50nm scale
patterns using copolymer templates

• How striped patterns get ordered
• Hex Patterns
• Completely aligned nm patterns on
• cm scale
• Model for shear alignment

Toshibas disk drives
2
Whos to Blame
Mingshaw Wu, Vincent Pelletier, Dan Angelescu,
Matthew Trawick David Huse Princeton University,
Physics Department Judith Waller Oxford
University Daniel Vega ExxonMobil Christopher
Harrison Schlumberger-Doll Douglas
Adamson Princeton Materials Institute
PCCM-MRSEC NSF DMR-0213706 Petroleum Research
Fund 35207-AC5,7
A presentation of the Chaikin-Register Polymer
Lithography Group
3
(No Transcript)
4
ala Lew Fetters
5
L. Leibler
6
Diblock Copolymers
Cylinders
Spheres
Lamellae
7
Our Diblocks are usually a plastic and a rubber
Plastic
Rubbers
e.g.
8
collaboration with R.R. Li, P.D. Dapkus, and M.E.
Thompson (USC) ? use MOCVD to selectively grow
GaAs dots on substrate, through holes in
removable mask
polymer
SiNx (15 nm)
ozone
MOCVD
CF4 RIE
wet etch
9
GaAs Dots Have Narrow Size Distribution
Number of Dots
diameter 23 3 nm overall height 14 2 nm
tapping-mode atomic force microscopy (TMAFM)
Size (nm)
10
Kinetics of annealing from quench
11
Annealing of stripes Orientational correlation
length
Harrison
?
?
395 K
466 K
Orientational Correlation length
12
Fingerprint Identification
But we dont yet know how fingerprints develop,
e.g. they are not genetic.
13
(No Transcript)
14
Nloops Ntriradii2
Nloops Nfingers1
Ntriradii Nfingers-1
SH
15
The ridges are not produced by a field of
force,such a mechanism would invariably produce
a directed field and the triradius and loop
would be ruled out. Instead we must expect
something of a tensor character such as a stress
or strain
16
(No Transcript)
17
Annealing of stripes
?
?
395 K
466 K
Orientational Correlation length
Average Distance between Disclinations
?2 ?-1/2 ? t1/4
18
(No Transcript)
19
(No Transcript)
20
Annealing by quadrupole annihilation

-
dipole-dipole
Fp2/r3

-
Why not dipole annihilation? For stripes moving
disclination produces dislocations
21
Experimental setup
22
Wondershearfridge Alignment
23
3um
PS-PHMA 2 Hrs at 150 deg C Alignment ?
Angelescu Waller
24
(No Transcript)
25
Nano-rheometry
Polymer used monolayer of cylinder-forming
PS-PEP
T100C
Slow shearing (300 nm/sec)
26
Thin film viscosity measurement
S/EP 3/24 Bulk Data
J Sebastian et al Macromolecules 35 p.
2707-2713 (2002)
Cylinder and Sphere thin films
Applied stress
From the analysis of 2-layer cylinder and sphere
data we get an apparent viscosity on the order
of 1000 Pa s which is lower than expected for
bulk. We are in the stress regime where
microdomain order is destroyed by shear !
27
Assume misaligned cylinders rip off due to shear.
Similar to an angle dependent Order-Disorer
Temperature (TODT)
Possible model
Cylinders melt due to shear
PDMS movement
Cylinders will be melted unless the temperature
is below the effective TODT. Thus, the will only
be stable if they are within a narrow range of
angles centered around the shearing direction.
Otherwise they melt (or break) and reform.
Cylinders are stable
28
(S/EP 3/24)
Sphere Alignment 2 Layers
Spheres aligned over square cm !

• Only align if they are 2-layer thick.
• One lattice vector parallel
• to shearing direction (like colloids)
• Slip occurs at the interface
• between layers.

Top layer
Jamming
Bottom layer
500 nm
Easy slip
29
Shear Alignment of two layers of spheres Complete
alignment over cm, No Disclinations, Greatly
reduced dislocations
m
30
Transmitting Stress through a Fluid Layer
Rheometrics DSR-200 parallel plate
fixtures isometric view
cross-sectional view

• z gradient
• ? flow
• R neutral

z
z
?
Plate
R
R
PDMS oil 106 cSt h 100-400 µm
t
3 Si wafer with SEP 3/24 (78 nm for 3 layers)
31
Shearing PS-PHMA 21-64
1
Shear annealed at 135C for 30 minutes
0.8
0.6
0.4
Saturation stress
0.2
0
Sample is a quarter wafer
-0.2
3500Pa - 30min
Rheometer tools edge (scratched on sample)
-0.4
Threshold stress
5000Pa - 30min
-0.6
0
1000
2000
3000
4000
5000
6000
Shear stress (Pa)
at edge
Measured edge position
AFM images
(Those are the triangles in the plot above)
32
sthresh
sstep
33
(No Transcript)
34
Recrystallization T vs. TODT
Change in distribution
35
Free energy governing cell dynamics simulations
Q
Time to destroy by shear
Assembly time tas
Adapted from I. Hamley Macromol. Theory Simul. 9,
363 (2000)
36
Two Square Layers, Interacting
y1
yi dricos(2p(xi- xi,0))cos(2p(yi-yi,0))
is the order parameter for a square density wave
y2
TODT TODT - G /a
where a2a, u2u
37
Shear at an angle to lattice
Now comes the real fudge, time average over
relaxation time
38
Channel Flow Motivation
Since we can transmit the shear stress using PDMS
oil, we should be able to produce it with
pressurized flow.
39
Channel Flow Setup
PDMS oil
60mL syringe
Pressure gauge
Channel carved in Teflon
Sample
Hot plate
Syringe pump
40
Channel Flow VIN Channel
PS-PEP 5-13 120C 600Pa
41
Channel Flow Double Wedge
Preheating chamber
1
0.8
0.6
S
0.4
blue
0.2
0
red
-0.2
0
1000
2000
3000
4000
5000
Shear stress (Pa)
Get stress from flow rate and local channel width.
42
Model PS-PHMA
1
0.8
PS-PHMA 21-64
0.6
0.4
S
0.2
0
-0.2
3500Pa - 30min
-0.4
5000Pa - 30min
-0.6
0
1000
2000
3000
4000
5000
6000
Shear stress (Pa)
PS-PHMA 30-84
43

• easy alignment method over square cm
• for both cylinders and spheres
• system for studying the mechanisms
• responsible for shear alignment

PS-PHMA film
44
Characteristic correlation lengths (30nm
period) As cast 100nm 1 hr. anneal
1mm Anneal limit 5mm Zone
Refine 50mm Shear align gt2cm
45
Polarizers SEMs
1?m
SEM of a sample with aluminum wires on a Si wafer.
46
Quantitative Measurements of Polarization
? use laboratory UV-vis spectrometer (in air) ?
put two nanowire polarizers in train, measure
transmitted intensity with wires parallel
and crossed
LAMP/ GRATING
POL
DETECTOR
? POL ? 30 in the visible, goes to zero near
350 nm ? for wavelengths below the node
(closer to the metals plasma frequency),
light polarized transverse to the wires is
preferentially transmitted
data, pair 1 data, pair 2 polarization model
47
Polarizers UV Crystal Analyzer
Fit to dirty Al
measured at Toshiba
48
Polarizers The Movie
This is one of the samples, as viewed in an
optical microscope with a polarizer analyzer. The
field of view is 2.1mm by 1.6mm. This is the
edge of the shear-aligned region.
49
IBM Nanotechnology announcement at IEDM
IBM used self assembly to build a
nanocrystal FLASH memory device
Toshiba
SEM image of the patterned magnetic media with
a 40 nm diameter. The scale bar indicates 100nm.
Nano-Patterning for Patterned Media using
Block-Copolymer, Koji Asakawa, Toshito Hitaoka,
Hiroyuki Hieda, Masatoshi Sakurai, Yoshiyuki  
Kamata and Katsyuki Naito., J. Photopolymer
Science and Technology Vol.15, p. 465 (2002)
50
Conclusions

• Nanolithography with Diblock Copolymers WORKS
• 5-50 nm patterns, 1011 1012 /cm2
• Wide variety of substrates and transfers
• Standard Semiconducting Processing
• Ordering Kinetics
• T1/4 in both stripes and hexagons
• Quadrupole annihilation in stripes!
• Disclinations again in hexagonal
• Shear Alignment gives long range order
• Cylinders work in a monolayer, better in 2 layers
• Spheres work in two or more layers
• Melting/Recrystallization model with TODT(q)
• describes results well ???
• Alignment with channel flow, can pattern stripes

51
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                  
                                                 
52
(No Transcript)