Sub-50 nm feature sizes using positive tone molecular glass resists for EUV lithography

Article abstract of DOI:10.1039/b514065j

Extreme ultra violet (EUV) lithography is one of the most promising next generation lithographic techniques for the production of sub-50 nm feature sizes with applications in the semiconductor industry. Coupling this technique with molecular glass resists

Full text of DOI:10.1039/b514065j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Sub-50 nm feature sizes using positive tone molecular glass resists for EUV
lithography
Seung Wook Chang,*a Ramakrishnan Ayothi,a Daniel Bratton,a Da Yang,a Nelson Felix,a Heidi B. Cao,b
Hai Dengb and Christopher K. Ober*a
Received 4th October 2005, Accepted 16th January 2006
First published as an Advance Article on the web 1st February 2006
DOI: 10.1039/b514065j
Extreme ultra violet (EUV) lithography is one of the most promising next generation lithographic
techniques for the production of sub-50 nm feature sizes with applications in the semiconductor
industry. Coupling this technique with molecular glass resists is an effective strategy for high
resolution lithographic patterning. In this study, a series of tert-butyloxycarbonyl (t-Boc)
protected C-4-hydroxyphenyl-calix[4]resorcinarenes derivatives were synthesized and evaluated as
positive tone molecular glass resists for EUV lithography. The amorphous nature of these
molecules was confirmed using thermal analysis, FTIR and powder X-ray diffraction. Feature
sizes as small as 30 nm with low line edge roughness (4.5 nm, 3s) were obtained after patterning
and development.
Here, we report the synthesis and characterization of a
Introduction
novel positive-tone t-Boc protected C-4-hydroxyphenyl-calix-
The photolithographic fabrication of nanometre scale semi-
[4]resorcinarene (Scheme 1).
conductor devices requires increasingly high resolution tech-
These materials are designed with bulky aromatic side
niques in order to maintain pace with Moore’s Law.1,2 EUV
groups to ensure high Tg, to inhibit crystallization, and ensure
lithography has emerged as one of the most promising
good etch resistance in lithographic processing. We then
candidates to meet the growing demands for ever-smaller
describe the patterning of these novel molecular glasses using
feature sizes.3–5 Increasing attention has been given to
EUV lithography, resulting in the production of sub-50 nm
amorphous low molecular weight compounds, known as
features.
molecular glasses, as new chemically amplified resists.6–11
The use of these monomolecular resists leads to lower line edge
Experimental
roughness (LER) in the fabricated pattern, less swelling or
residual stress, and minimal chain entanglement compared to
Materials and equipment
common photoresist polymers. Coupling small molecular size
4-(Dimethylamino)pyridine (DMAP), di-tert-butyl dicarbo-
with short wavelength is expected to result in an effective
nate, 4-hydroxybenzaldehyde, resorcinol, propylene glycol
process for the production of nanoscale lithographic features.
methyl ether acetate (PGMEA) and 1-methyl-2-pyrrolidinone
Small organic glass forming molecules are well known in
(NMP) were purchased from Aldrich and used as received. All
the literature, and in particular the synthesis of non-planer
other solvents were from Fischer, and used as received unless
structures with bulky side groups are thought to lead to the
formation of stable glasses.12,13 Materials with higher glass
otherwise stated. Fourier transform infrared spectra (FTIR)
were measured using a Thermo Nicolet 870. Nuclear magnetic
transition temperatures (Tg > 100 uC) are generally more
desirable for positive tone resists to maintain the integrity
resonance (NMR) spectra were recorded on a Mercury Varian
Inova-300 spectrometer. Thermogravimetric analysis (TGA)
of the pattern during the post exposure bake (PEB) in the
lithographic process.
To date, it has been demonstrated that several calixarene
derivatives form molecular glasses with several desirable
characteristics for a resist, including high Tg.14–17 Calixarene
derivatives have been employed as both negative15,18,19 and
positive20 tone resists for E-beam lithography, and as positive
and negative tone resists using 365 nm UV radiation.21,22 To
date, no reports in the literature exist of a calixarene based
resist for EUV lithography.
aDepartment of Materials Science and Engineering, Cornell University,
Bard Hall, Ithaca, NY 14853, USA. E-mail: cober@ccmr.cornell.edu;
Fax: +1 607 255 2365; Tel: +1 607 255 8417
bIntel Corp., 5200 N.E. Elam Young Parkway, Hillsborough, OR 97124,
Scheme 1 The synthesis of t-Boc protected C-4-hydroxyphenyl
calix[4]resorcinarene.
USA
1470 | J. Mater. Chem., 2006, 16, 1470–1474
This journal is ß The Royal Society of Chemistry 2006

View Article Online
was performed on a Thermogravimetric/Differential Thermal
Analyzer TG/DTA 320 at a heating rate of 10 uC min21 under
nitrogen. The Tg of C-4-hydroxyphenyl-calix[4]resorcinarene
derivatives was measured on TA Instruments Q1000
Modulated differential scanning calorimeter (DSC) at a
heating rate of 10 uC min21 under nitrogen. Tg was determined
on the second heating cycle. Powder X-ray diffraction (XRD)
was carried out on a Scintag, Inc. Theta–Theta Diffractometer
at room temperature. EUV Lithography was conducted at the
Advanced Light Source, Lawerence Berkeley National Lab,
Berkeley, California. The patterned wafers were examined
using a high resolution Hitach SEM. Line edge roughness
(LER) was calculated using SuMMIT image analysis software,
EUV technology, Martinez, CA.
1481, 1471 (n CLC aromatic), 1168 (n C–O–C). 1H-NMR
(300 MHz, CDCl3, TMS) d (ppm) 1.2–1.6 (m, 86.4H, CH3),
5.7–6.5 (m, 4H, CH), 6.7–7.2 (m, 24H, aromatic H). Compound
D (70% protected). Yield: 67%. FTIR (film, cm21): 3502 (n
OH), 2983, 2939 (n CH), 1764 (n CLO), 1509, 1486, 1469 (n
CLC aromatic), 1155 (n C–O–C). 1H-NMR (300 MHz, CDCl3,
TMS) d (ppm) 1.2–1.6 (m, 75.6H, CH3), 5.5–6.3 (m, 4H, CH),
6.4–7.1 (m, 24H, aromatic H). Compound E (60% protected)
Yield: 63%. FTIR (film, cm21): 3490 (n OH), 2983, 2937 (n
CH), 1762 (n CLO), 1511, 1486, 1473 (n CLC aromatic), 1153 (n
1
C–O–C). H-NMR (300 MHz, CDCl3, TMS) d (ppm) 1.2–1.6
(m, 64.8H, CH3), 5.6–6.3 (m, 4H, CH), 6.4–7.1 (m, 24H,
aromatic H).
Synthesis of triphenyl sulfonium 2-(phenoxy)tetrafluoro-
ethane-1-sulfonate [C6H5OCF2CF2SO32 +S(C6H5)3] photoacid
generator. The photacid generator, triphenylsulfonium
2-(phenoxy)tetrafluoroethane-1-sulfonate (Fig. 1) was synthe-
sized by combining a protocol published independently by
Fiering and Wonchoba24 and Crivello and Lam.25 Detailed
synthesis and characterization of the novel PAG will be
published elsewhere.
Synthesis of C-4-hydroxyphenyl-calix[4]resorcinarene.
Resorcinol (16.5 g, 150 mmol) and conc. HCl (25 mL) were
dissolved in ethanol (150 mL) and was stirred at 0 uC for
10 min using ice water bath. P-Hydroxybenzaldehyde (18.3 g,
150 mmol) was added to the solution and the mixture was
refluxed for 6 h. After cooling to room temperature, the
precipitated solids were collected by vacuum filtration. The
crude solids were washed with ethanol (100 ml) two times
and acetone (150 ml) three times, followed by drying at 50 uC
in vacuo during overnight.
Briefly, potassium phenoxide was reacted with 1,2-dibro-
motetrafluoroethane, resulting in 1-bromo-2-(phenoxy)tetra-
fluoroethane. This was transformed to 2-(phenoxy)tetra-
fluoroethane-1-sulfinate on reaction with sodium dithionite
and sodium bicarbonate in aqueous acetonitrile. The sulfinate
was converted to 2-(phenoxy)tetrafluoroethane-1-sulfonate by
reaction with elemental chlorine in water to sulfonyl chloride,
then oxidation with lithium hydroxide in aqueous THF.
Subsequently an ion exchange reaction of sulfonate with
diphenyliodonium chloride in aqueous acetonitrile gave
diphenyliodonium 2-(phenoxy)tetrafluoroethane-1-sulfonate,
which by reaction with diphenyl sulfide and copper(II) acetate
was converted to triphenylsulfonium 2-(phenoxy)tetrafluoro-
ethane-1-sulfonate. This was purified by recrystallization
from dichloromethane–ether mixture. Yield: 70%. 1H-NMR
(300 MHz, acetone-d6) d (ppm) 7.37–7.30 (m, 3H, aromatic
protons, anion), 7.37–7.46 (m, 2H, aromatic protons, anion),
7.80–8.0 (m, 15H, aromatic protons, cation). 19F NMR
(282 MHz, acetone-d6) d (ppm) 282.40 (s, OCF2), 2118.01
(s, CF2SO32). Elemental analysis, C% calc = 58.2, C% found =
57.5, ESI MS (m/e) = 273.18 (C6H5OCF2CF2SO32), 263.38
((C6H5)3S+)
Yield: 75%. Characterization data was identical to that
reported by Tunstad et al.23
Synthesis of dodecyl-O-tert-butyl carbonated C-4-hydroxy-
phenyl-calix[4]resorcinarene (100% t-Boc protected C-4-
hydroxyphenyl-calix[4]resorcinarene). To
a
solution of
C-4-hydroxyphenyl-calix[4]resorcinarene (1 g, 1.17 mmol)
and DMAP (0.057 g, 0.47 mmol) in NMP (15 mL) di-tert-
butyl dicarbonate (3.44 g, 15.8 mmol) was added slowly
using a dropping funnel. The evolution of CO2 gas occurred
immediately. This mixture was stirred for 24 h at room
temperature, then precipitated into a four-fold excess of
distilled water, and the resulting solids were collected
by vacuum filtration. The crude product was dissolved in
ethyl acetate (30 mL) washed with water (3 6 50 mL) and
brine (1 6 50 mL). The collected organic layer was dried over
Mg2SO4, filtered and solvent was evaporated under reduced
pressure, resulting in the collection of a fine powder which was
dried at 50 uC under reduced pressure overnight. Yield: 1.7 g
(70%). The partially protected compounds were obtained using
a similar procedure, with varying amounts of di-tert-butyl
dicarbonate.
Compound A (100% t-Boc protected C-4-hydroxyphenyl-
calix[4]resorcinarene. Yield: 70%. FTIR (film, cm21): 2983,
2939 (n CH), 1765 (n CLO), 1508, 1486, 1471 (n CLC aromatic),
1
1157 (n C–O–C). H-NMR (300 MHz, CDCl3,TMS) d (ppm)
1.2–1.6 (m, 108H, CH3), 5.7–6.5 (m, 4H, CH), 6.7–7.1 (m,
24H, aromatic H). Compound B (90% protected) Yield: 62%.
FTIR (film, cm21): 3531 (n OH), 2983, 2966 (n CH), 1764
(n CLO), 1508, 1486, 1469 (n CLC aromatic), 1151 (n C–O–C).
1H-NMR (300 MHz, CDCl3, TMS) d (ppm) 1.2–1.6 (m,
97.2H, CH3), 5.7–6.5 (m, 4H, CH), 6.6–7.1 (m, 24H, aromatic
H). Compound C (80% protected) Yield: 65%. FTIR (film,
cm21): 3517 (n OH), 2984, 2939 (n CH), 1766 (n CLO), 1508,
Fig. 1 Triphenyl sulfonium 2-(phenoxy)tetrafluoroethane-1-sulfonate.
J. Mater. Chem., 2006, 16, 1470–1474 | 1471
This journal is ß The Royal Society of Chemistry 2006

View Article Online
Lithographic evaluation
Typically, the resist compounds were dissolved in propylene
glycol methyl ether acetate (PGMEA) making a 5% wt
solution. A novel PAG, triphenylsulfonium 2-(phenoxy)tetra-
fluoroethane-1-sulfonate, (5% wrt resist), trioctylamine
(0.12 wt% wrt resist) was added and the solution was filtered
through a 0.2 mm membrane filter three times. The solutions
where then spin coated onto a 4 inch HMDS primed silicon
wafer (2000 rpm, 30 s) leading to a film approximately 100 nm
thick. This was subjected to a post application bake (PAB) at
100 uC for 120 s, then exposed using EUV radiation. After
exposure, the wafer was baked at 80 uC for 30 s, before
development in an aqueous solution of 0.026 N tetramethyl-
ammoniumhydroxide (TMAH), rinsed with water then dried
with a stream of nitrogen.
Fig. 2 DSC analysis of C-4-hydroxyphenyl-calix[4]resorcinarene with
various t-Boc protecting compositions.
agreement with the theoretical t-Boc coverage in each case.
DSC analysis revealed that all of the compounds exhibited a
Tg around 120 uC, with compound E (60% protection) having
the highest Tg of 127 uC (Fig. 2).
Results and discussion
The synthesis of C-4-hydroxyphenyl-calix[4]resorcinarene
has been previously reported.26 We modified the synthetic
procedure to include ethanol as a solvent, and obtained the
unprotected products as fine pink powders. Protection of
the phenol groups using di-tert-butyl dicarbonate lead in
the desired structure, depicted in Scheme 1. The 1H NMR
spectrum of the protected species was recorded in CDCl3, and
showed multiplet t-Boc peaks around 1.2–1.6 ppm, revealing
the existence of a variety of different structural configurations
of the target molecule,23 a factor which helps to inhibit
crystallization. By controlling the amount of di-tert-butyl
dicarbonate added during the protection step, we were able to
obtain partially protected compounds in order to probe the
effects of resist structure on lithographic performance. In total,
five compounds with average t-Boc protection ranging
from 100% to 60% protected were synthesized. All protected
compounds showed good solubility in common lithographic
solvents, such as propylene glycol methyl ether acetate
(PGMEA).
FTIR was used to investigate hydrogen bonding in the
partially protected compound. The intensity of free hydroxyl
group peak at around 3500 cm21 increased gradually with
decreasing t-Boc protecting ratios, as expected. Compound E
showed the broadest peak around 3500 cm21 due to multiple
hydrogen bonding interactions. The frequency of the free
hydroxyl group peak shifts to lower wavenumbers with
decreasing t-Boc coverage, also indicating an increase in
hydrogen bonding (i.e., compound B, 90% protected, nOH =
3517 cm21 compared to compound E, 60% protected nOH =
3490 cm21). The spectra can be seen in Fig. 3.
However, the carbonyl band at 1765 cm21 was largely
unchanged regardless of the degree of protection. These results
indicate that hydroxyl–hydroxyl hydrogen bonding interac-
tions predominate over hydroxyl–carboxylic interactions
between adjacent molecules, possibly because intermolecular
hydrogen bonding between hydroxyl groups and carbonyl
groups are hindered by the presence of the bulky tertiary butyl
The thermal characteristics of the compounds were
examined by thermal gravimetric analysis (TGA) and differ-
ential scanning calorimetry (DSC). TGA showed that the
observed percentage mass loss of t-Boc upon thermal depro-
tection was in good agreement with those calculated for the
average degree of protection for each compound (Table 1).
TGA showed that thermal deprotection occurred at
temperatures higher than 150 uC in all cases, and the observed
weight loss upon deprotection was found to be in good
Table 1 Thermal characterization of fully and partially t-Boc
protected C-4-hydroxyphenyl-calix[4]resorcinarene
Compound
A
B
C
D
E
t-Boc (mol%)a
% Weight loss (calc)b
% Weight loss (found)c
Tg/uCd
100
90
56.4
56.9
80
53.4
52.8
70
50.0
50.0
60
46.2
47.6
58.9
58.7
117
118
116
118
127
a
b
Theoretical mol% protection by t-Boc. Theoretical % weight loss
upon thermal deprotection. Actual % weight loss upon thermal
Fig. 3 FTIR of fully and partially protected C-4-hydroxyphenyl-
calix[4]resorcinarene recorded as thin films. Note the increase in hydro-
gen bonding indicated by the broadening of the peak at 3500 cm21 as
the degree of protection decreases (A), 100%, to (E) 60%.
c
d
deprotection (measured by TGA). Measured by DSC, heating/
cooling rate 10 uC min21
.
1472 | J. Mater. Chem., 2006, 16, 1470–1474
This journal is ß The Royal Society of Chemistry 2006

View Article Online
Fig. 4 XRD analysis of fully or partially t-Boc protected C-4-
hydroxyphenyl-calix[4]resorcinarene, recorded at room temperature.
functionalities. We believe that the increase in hydrogen
bonding is responsible for the increase in Tg in compound E.
In order to confirm the amorphous properties of the
compounds, the powder XRD traces were recorded at room
temperature (Fig. 4).
All compounds show two halo peaks indicative of
amorphous materials (2h = approximately 6.3 and 18.3u).
The d-spacing of the two amorphous halo peaks at narrow
and wide angles are around 1.4 and 0.5 nm respectively.
Interestingly, compound (A) has a sharp peak at 2h = 5.3u and
a less intense peak at 2h = 10.5u. These two sharp peaks
suggest the existence of a lamellar structure. This result
may indicate the existence of a layered mesophase structure
in (A), and to a lesser extent in (B). Previous studies
on calix[4]resorcinarene derivatives have observed similar
phenomena.23 As the degree of t-Boc coverage decreases
(compounds C-E), the sharp peaks gradually disappear
indicating amorphous compounds, and suggesting that an
increase in the number of structural conformers in the material
results in a complete absence of short range ordering.13
Preliminary solubility tests during spin coating revealed that
compound C, 70% protected, was the optimal resist material
when spin coated onto a silicon wafer primed with HMDS.
The compounds with higher protection ratios (A and B)
showed poor adhesion to the silicon wafer, while compounds
with lower protection ratios (D and E) were slightly soluble in
aqueous base even prior to exposure, leading to a reduction in
image integrity upon development. Full details about the
preparation of the resists for lithographic evaluation can be
found in the experimental section. No crystallization of the
spun coated films was observed, even after storage for several
months, indicating stability of the amorphous resists.
Fig. 5 (A) and (B) SEM images of 30 nm patterns obtained using
compound C, 70% protected C-4-hydroxyphenyl-calix[4]resorcinarene
with EUV lithography, 5% PAG, dose 22 mJ cm22. (C) 100 nm lines
demonstrating low LER (4.6 nm, 3 s, from 50 nm dense lines).
Full details regarding lithographic evaluation can be found in the
experimental section.
ideal fitted line averaged across several features. The LER of
the features obtained via EUV lithography were calculated
from these images using SuMMIT image analysis software.
LER (3s) was calculated from 50 nm dense lines to be 4.6 nm.
With conventional polymeric resists, LER of 12–14 nm have
been encountered.27
SEM images demonstrate that 30 nm resolution was
achieved using EUV lithography, with a resist formulation
consisting of 5% by weight of the novel photoacid generator
triphenylsulfonium 2-(phenoxy)tetrafluoroethane-1-sulfonate
and an exposure dose of 22 mJ cm22 (Fig 5A and B).
Low line edge roughness (LER) is purported to be a major
advantage of using molecular glass resists.6,20 This quantity is
usually defined as a deviation in the edge of a feature from an
The exact reason for the low LER often encountered when
molecular glass resists are employed is still under debate. The
This journal is ß The Royal Society of Chemistry 2006
J. Mater. Chem., 2006, 16, 1470–1474 | 1473

View Article Online
conventional explanation is that the smaller size of these
resists compared to conventional polymeric resists results in
higher resolution features through smaller pixel sizes. Several
groups3,28 have examined the LER obtained using polymeric
resists, and showed that even a 12 fold increase in Mw
(from 2.9–33.5 kg mol21) does not affect LER, which
suggests molecular size alone is not the complete explanation.
One significant difference between molecular glass resists
and polymeric resists is polydispersity. Molecular glass
resists are monodisperse, compared to polymer resists which
always consist of a range of molecular weights. Lower
molecular weight polymers have faster dissolution rates
compared to higher molecular weight samples.3 This implies
that the dissolution behavior of a polymeric resist upon
development is a composite value depending on it molecular
weight distribution and degree of deprotection, which leads
to higher values of LER. We hypothesise that monodisperse
MG resists, in which dissolution rate should be comparatively
constant across the exposed regions of the sample, develop
in a more homogenous manner and hence return lower values
of LER.
References
1 E. Reichmanis, F. M. Houlihan, O. Nalamasu and T. X. Neenan,
Chem. Mater., 1991, 3, 394.
2 L. F. Thompson, C. G. Willson and M. J. Bowden, in Introduction
to Microlithography, ed. L. F. Thompson, C. G. Willson and
M. J. Bowden, ACS, Washington DC, 1994.
3 R. L. Brainard, J. Cobb and C. A. Cutler, J. Photopolym. Sci.
Technol., 2003, 16, 401.
4 H. B. Cao, J. M. Roberts, J. Dalin, M. Chandhok, R. P. Meagley,
E. M. Panning, M. K. Shell and B. J. Rice, Proc. SPIE-Int. Soc.
Opt. Eng., 2003, 5039, 484.
5 G. M. Blumenstock, C. Meinert, N. R. Farrar and A. Yen, Proc.
SPIE-Int. Soc. Opt. Eng., 2005, 5645, 188.
6 Y. Shirota, J. Mater. Chem., 2005, 15, 75.
7 J. W. Leon, M. Kawa and J. M. J. Frechet, J. Am. Chem. Soc.,
1996, 118, 8847.
8 D. C. Tully, K. Wilder, J. M. J. Frechet, A. R. Trimble and
C. F. Quate, Adv. Mater., 1999, 11, 314.
9 K. Tsuchiya, S. W. Chang, N. M. Felix, M. Ueda and C. K. Ober,
J. Photopolym. Sci. Technol., 2005, 18, 431.
10 T. Kadota, H. Kageyama, F. Wakaya, K. Gamo and Y. Shirota,
J. Photopolym. Sci. Technol., 1999, 12, 375.
11 T. Kadota, H. Kageyama, F. Wakaya, K. Gamo and Y. Shirota,
Chem. Lett., 2004, 33, 706.
12 I. Alig, D. Braun, R. Langendorf, H. O. Wirth, M. Voigt and
J. H. Wendorff, J. Mater. Chem., 1998, 8, 847.
13 Y. Shirota, J. Mater. Chem., 2000, 10, 1.
14 O. C. Myoung, H. H. Choi, Y. S. Ku, C. I. Ahn, G. W. Chung and
J. B. Lee, Kongop Hwahak, 2002, 13, 131.
Conclusions
15 J. Fujita, Y. Ohnishi, Y. Ochiai and S. Matsui, Appl. Phys. Lett.,
1996, 68, 1297.
16 O. Haba, K. Haga, M. Ueda, O. Morikawa and H. Konishi, Chem.
Mater., 1999, 11, 427.
17 T. Nishikubo, A. Kameyama, K. Tsutsui and M. Iyo, J. Polym.
Sci., Part A: Polym. Chem., 1999, 37, 1805.
18 M. Ishida, J.-i. Fujita, T. Ogura, Y. Ochiai, E. Ohshima and
J. Momoda, Jpn. J. Appl. Phys., Part 1, 2003, 42, 3913.
19 H. Sailer, A. Ruderisch, W. Henschel, V. Schurig and D. P. Kern,
J. Vac. Sci. Technol., B, 2004, 22, 3485.
20 S. W. Chang, D. Yang, J. Dai, N. Felix, D. Bratton, K. Tsuchiya,
Y.-J. Kwark, J.-P. Bravo-Vasquez, C. K. Ober, H. B. Cao and
H. Deng, Proc. SPIE-Int. Soc. Opt. Eng., 2005, 5753, 1.
21 M. Ueda, D. Takahashi, T. Nakayama and O. Haba, Chem.
Mater., 1998, 10, 2230.
In summary, t-Boc protected C-4-hydroxyphenyl-calix[4]re-
sorcinarenes were synthesized and studied as positive-tone
resists for advanced lithography. The compounds were found
to be amorphous materials with high Tg’s. Feature sizes as
small as 30 nm with excellent low LER were produced after
patterning with EUV wavelength radiation. Future research
goals will focus on the synthesis of new calixarene analogues as
target materials for EUV resists.
Acknowledgements
We thank Intel Corporation for making this work possible
through its financial support and International SEMATECH
for initial support in this research area. The authors thank Drs
Dong-kyu Lee and Young-Uk An for discussions regarding
XRD data interpretation and measurement. We also gratefully
acknowledge the Cornell Nanofabrication Facility (CNF) and
the Cornell Center for Materials Research (CCMR) for use of
their facilities and technical support. In particular, we wish to
thank Dr Patrick Naulleau and Lawrence Berkeley National
Laboratory for the EUV exposures and an academic license
for SuMMIT.
22 T. Nakayama, K. Haga, O. Haba and M. Ueda, Chem. Lett., 1997,
265.
23 L. M. Tunstad, J. A. Tucker, E. Dalcanale, J. Weiser, J. A. Bryant,
J. C. Sherman, R. C. Helgeson, C. B. Knobler and D. J. Cram,
J. Org. Chem., 1989, 54, 1305.
24 A. E. Feiring and E. R. Wonchoba, J. Fluorine Chem., 2000, 105,
129.
25 J. V. Crivello and J. H. W. Lam, J. Org. Chem., 1978, 43, 3055.
26 F. Weinelt and H. J. Schneider, J. Org. Chem., 1991, 56, 5527.
27 J. Shin, G. Han, Y. Ma, K. Moloni and F. Cerrina, J. Vac. Sci.
Technol., B, 2001, 19, 2890.
28 T. Yamaguchi, K. Yamazaki and H. Namatsu, J. Vac. Sci.
Technol., B, 2004, 22, 2604.
1474 | J. Mater. Chem., 2006, 16, 1470–1474
This journal is ß The Royal Society of Chemistry 2006