S
1
Supplementary Materials For:
Photolithographic Olefin Metathesis Polymerization
Raymond A. Weitekamp, Harry A. Atwater, Robert H. Grubbs
JACS 2013
Materials and Methods
Supplementary Experiments
Figures S1
–
S3
S
2
Materials and Methods:
(H
2
IMes)(PPh
3
)(Cl)
2
RuCHPh was recieved as a research gift from Materia Inc.
and converted to
1
via literature procedure.
1
All other chemicals were purchased from
Sigma Aldrich. Printed photomasks were purchased from CAD/Art Services, Inc.
(http://outputcity.c
om).
Silicon coupons were ordered as a pre
-
diced 4” wafer from Ted
Pella (Part # 16006). Dichloromethane, ethyl vinyl ether and
5
-
ethylidene
-
2
-
norbornene
were first degassed by bubbling argon through for 15 minutes.
The lamp used was an 8
-
watt “MRL
-
58 Multiple Ray Lamp” from Ultra Violet
Products (#UVP
95
-
0313
-
01). The 254 nm bulb used was a General Electric germicidal
bulb (#
GEG8T5
, from
http://bulbtronics.com
)
, with a FWHM of
only a few nm
. The 352
nm bulb was an Eiko blacklight bulb (#
WKF8T5BL
, from
http://bulbtronics.com
), with a
FWHM of approximately 50 nm.
Samples were placed approximately 1.5” away from the
bulb during exposure.
A sta
ndard PLOMP resist recipe:
Compound
1
(1.3 mg) was placed under argon and dissolved in 2 mL
dichloromethane. To this catalyst solution was quickly added 1.5 mL 1,5
-
cyclooctadiene,
the solution became a semi
-
solid in 10 seconds and was allowed to react for
1 minute
before quenching with 3 mL ethyl vinyl ether. The viscous solution was slowly stirred for
5 minutes, sealed under argon, and sonicated for 1 hour.
The
volatiles were removed on a
rotary evaporator, to yield
semisolid
poly(COD), colored light yello
w by the quenched
catalyst (the photoactive vinyl ether complex). Ethylidene norbornene (10 mL) was added
to this mixture, which was cooled to 0 °C and sonicated for 1 hour. The partially
dissolved mixture was placed on an ice bath and stirred until fully
dissolved, while
allowing the bath to warm to room temperature. The result is a light yellow, viscous
solution weighing approximately 10 grams.
General Film Casting Procedures
1x1 cm silicon coupons were cleaned in a piranha solution (3:1 concentrated
H
2
SO
4
: 30 % H
2
O
2
), rinsed with deionized water (“Nanopure”), isopropanol and
acetone.
(Caution! Piranha solution reacts violently with organic matter.)
Before spin
casting, the coupons were heated to 140
-
150 °C for 1
-
2 minutes to drive off adventitious
mois
ture
, cooled to room temperature under a stream of argon gas, and quickly loaded
onto the spinner. While this pre
-
heating step was not always necessary, it led to the most
reproducible results. Samples were spun between 1500
–
7000 RPM for 60 seconds to
ac
hieve films of varying thickness. These cast films should be exposed and developed
quickly, prolonged delay after spinning lead to inconsistent results.
Specific conditi
ons
for the samples in Figures 2 &
3
:
The
samples in Figures 2 &
3
were prepared by using various dilutions of the
standard resist preparat
ion outlined above. For Figure 2
,
0.7
5 mL of the standard resist
de
scribed above was diluted with 1.2
5 mL ENBE
. The
2” diameter
wafer was cleaned
using the procedure outlined for the
coupons, heated to 150 °C for 2 minutes and cooled
1
Love, J. A.; Morgan, J. P.; Trnka, T. M.; Grubbs, R. H.
Angew
.
Chem
.
Int
.
E
d.
2002
,
41
, 4035
–
4037.
S
3
under a st
ream of argon. Approximately 0.9
mL of the solution was used to cover the
en
tire wafer, which was spun at 35
00 RPM for 60 seconds. The film was irradiated
through the mask for
15
minutes, and dev
eloped in hexanes for
90 seconds
.
For Figure 3
,
1.00 mL of the standard resist described above was diluted with 0.1 mL ENBE.
Approximately 0.1 mL of this solution was used to cover the 1 cm
2
coupons, which were
spun at 7000 RPM for 60 seconds. The films we
re irradiated through the grid test mask
for 10 minutes, and developed in 10% dichloromethane/hexanes for 2 minutes
.
Analytical equipment:
Profilometry was performed on a Bruker
D
ektak
XT
stylus profiler. Optical
micrographs were obtained on a
Zeiss
Axio
O
bserver
inverted
microscope equipped with
a 10
×
objective.
NMR spectra were recorded at room temperature on a Varian Inova 500
(at 500 MHz). The NMR spectra were analyzed on MestReNova software and are
reported relative to CD
2
Cl
2
(
δ
= 5.320
).
Supplementary
Experiments:
To show that the catalyst is necessary for the resist to function, the standard resist
prep was used except the polymer was precipitated into methanol to extract the quenched
catalyst. Compound
1
(1.3 mg) was placed under argon and dissolved
in 2 mL
dichloromethane. To this catalyst solution was quickly added 1.5 mL 1,5
-
cyclooctadiene,
the solution became a semi
-
solid in 10 seconds and was allowed to react for 1 minute
before quenching with 3 mL ethyl vinyl ether. The viscous solution was very
slowly
stirred for 5 minutes, after which 5 mL methanol was added. The suspension was
sonicated for 20 minutes, the brown solution was decanted and the off
-
white
solid
polymer was washed three times with 10 mL of methanol. The polymer was dried
i
n
vacuo
,
and dissolved in 10 mL
5
-
ethylidene
-
2
-
norbornene
to afford a very pale yellow,
viscous solution. This solution was cast as before and exposed for 6 minutes at 254 nm (4
times the standard exposure for the analogous resist) with no evidence of pattern
forma
tion. After developing with hexanes, a clean Si surface was recovered.
As well,
p
rolonged irradiation of pure ENBE at both 254 nm and 352 nm did not render any
change in viscosity or other evidence of crosslinking.
The addition of BHT (
2,6
-
Di
-
tert
-
butyl
-
4
-
methylphenol
) to
the
PLOMP re
sist
did not
appear to
change
its
behavior, which
suggests that the mechanism does not involve radicals.
To support the hypothesis that the ruthenium vinyl ether complex is intact inside
the PLOMP resist, the
1
H NMR spectra of
a PLOMP resist and complex
2
were
compared. The resist was prepared by the standard recipe above. Complex
2
was
prepared as reported by Louie.
2
The spectra strongly support the proposed composition of
the PLOMP photoresist; the alkylidene protons in each
spec
trum are less than 1 ppm
apart (Figure S2).
2
Louie, J.; Grubbs, R. H.
Organometallics
2002
,
21
, 2153
–
2164.
S
4
Figure S1
-
The chemical structure of c
omplex
2
.
Figure S2
–
1
H spectra in CD
2
Cl
2
of complex
2
and a PLOMP photoresist. The region of
the alkylidene proton is shown to highlight the similarity between the two. No other peaks
were observed in the downfield region (
δ
=11
–
22 ppm), suggesting that no other
ruthenium alkylidene species are present in
any significant quantity.
To support the hypothesis that dative bonding from the highly olefinic resist
stabilizes the photoactive complex, two experiments were performed. First, the formally
14
-
electron ruthenium vinyl ether species can be prepared by qu
enching
the 2
nd
generation Grubbs
-
Hoveyda catalyst
with ethyl vinyl ether.
3
This complex immediately
crosslinks ENBE, suggesting that there is a ligand present in the resist that stabilizes
the
photoactive catalyst. O
bvious candidates for this ligand
include
the original pyridine
ligands from complex
1
or the olefins in the viscous resist material.
Wenzel
and
coworkers have
demonstrated the ability to remove pyridine ligands using acids such as
3
Li, J.
Caltech Ph.D. Thesis
2012
N
N
Ru
Cl
PCy
3
OEt
Cl
13.35
13.40
13.45
13.50
13.55
13.60
13.65
13.70
13.75
13.80
ppm
N
N
Ru
Cl
PCy
3
OEt
Cl
N
N
Ru
Cl
OEt
Cl
S
5
trifluoroacetic acid (TFA).
4
If only the removal of a pyri
dine ligand was required to
reactivate the complex, we would expect that crosslinking could also be triggered by
acid. However, the addition of TFA lead to no change in viscosity after 24 hours. In fact,
the PLOMP resist was still able to function in the p
resence of TFA; no change in
behavior was observed for 254 nm
photopatterning
30 minutes after adding TFA. While
these experiments do not explicitly rule out the presence of a pyridine
-
coordinated
complex, they strongly suggest that dative bonding from the
surrounding olefins is the
more likely mechanism of catalyst stabilization. This olefin could belong to the
poly(COD), ENBE or excess ethyl vinyl ether.
Figure S3
–
A cartoon depicting
the
proposed
crosslinking process of a PLOMP resist.
4
Wenzel, A. G.; Blake, G.; VanderVelde, D. G.; Grubbs, R. H.
J. Am. Chem. Soc.
2011
,
133
, 6429
–
6439.
N
N
Ru
Cl
Cl