Transcript Block Copolymer Micelle Nanolithography Roman Glass, Martin

Block Copolymer Micelle
Nanolithography
Roman Glass, Martin Moller and Joachim P Spatz
University of Heidelberg
IOP Nanotechnology (2003)
Erika Parra
EE235
4/18/2007
Motivation


Market Trends
Small features





Sub-10nm clusters deposited
Patterns 50nm to 250nm and greater
Lower cost of tedious fabrication processes for
conventional lithography
Increase throughput (from e-beam) – parallel process
Bottom line: bridge gap between traditional selfassembly and lithography
Process Overview
Dip wafer (Si) into
micelle solution

Retrieve at
12mm/min

Air-evaporate
solvent

Plasma (H2, Ar, or
O2) removes
polymer shell
Results:

Uniform

Hexagonal

2, 5, 6, or 8nm

Spherical

PS(190)-b-P[2VP(Au0.2)](190) PS(500)-b-P[2VP(Au0.5)](270)
PS(990)-b-P[2VP(Au0.5)](385) PS(1350)-b-P[2VP(Au0.5)](400)
Side view TEM – treated wafer
Au ~ HAuCl4
Diblock Copolymer Micelles





Dendrite shaped macromolecule
Corona is amphiphilic
Micelle MW and shape controlled by
initial monomer concentration
Polymer corona with “neutralized” core
(Au, Ag, AgOx, Pt, Pd, ZnOx, TiOx, Co, Ni,
and FeOx)
Nanodot “core” size is controlled by the
amount of metal precursor salt
PS
P2VP
Au
In this paper:
Water-in-oil micelle (toulene solvent)
Polystyrene(x)-b-poly(2-vinylpyridine)(y) (PS(x)-b-P2VP(y))
Au core from chloroauric precursor (HAuCl4)
Cluster Pattern Characterization
Low
PDI



MW tunes nanodot distance (max of 200 nm micelle)
Low polydispersity permits regularity
Higher MW decreased pattern quality and position precision
(softness in shell)
Guided Self-Assembly (>250nm)





Predefine topographies
using photo or e-beam
Spin-on concentrated
micelle solution
(capillary forces of
evaporating solvent
adheres them to sides)
Micelles are pinned to
the substrate by plasma
(100W, 0.4mbar, 3min)
Lift-off removes PR and
micelles
2nd plasma treatment
removes micelle
polymer (100W,
0.4mbar, 20min)
PS(1350)-b-P[2VP(Au0.5)](400)
D = 8nm, L = 85nm
Cluster Aggregation



Vary PR
thickness
Feature height
(volume) defines
cluster diameter
Figure: e-beam
200nm features
on 2um square
lattice
800nm
500nm
75nm
Line Patterning





Cylindrical micelle
Formed if corona
volume fraction < core
PS(80)-b-P2VP(330)
Length of several
microns
Substrate patterned
with grooves & dipped
in micelle solution
4nm line
Negative Patterning with E-beam




Spin-on micelles
Expose with e-beam (1KeV, 40050,000 μC/cm2), 200um width
Ultrasound bath + 30min plasma
Electrons stabilize micelle on Si due
to carbon species formed during
exposure
Micelles on Electrically Insulating Films



Glass substrate
desired in
biology
E-beam requires
conductive
substrate
Evaporate 5nm
carbon layer
Mechanical Stability of Nano-Clusters

Treated and unaffected by:



Pirahna, acids, many bases, alcohols, ultrasonic
water bath
Hypothesis: edge formed by the substratecluster borderline is partly wetted by surface
atoms during plasma treatment
Thermal

800 C evaporated clusters but no migration
occured
Conclusions



Simple process for
sub-10nm clusters
and lines
Block copolymer
micelle size
controls nanocluster
interspacing
Micelle size
controlled by
monometer
concentrations
Micelles as masks for diamond field emitters
F. Weigl et al. / Diamond & Related Materials 15 (2006)