Designing a non-volatile imaging switch for mass-persistent, chemically amplified photolithography: A model study

Article abstract of DOI:10.1039/b209695a

An acid catalysed rearrangement that transforms a bicyclic lactone into a phenolic carboxylic acid has been tested for potential use in chemically amplified microlithographic imaging.

Full text of DOI:10.1039/b209695a

Designing a non-volatile imaging switch for mass-persistent, chemically
amplified photolithography: a model study
John M. Klopp, Nicolas Bensel, Zachary M. Fresco and Jean M. J. Fréchet*
Department of Chemistry, University of California, Berkeley, CA 94720-1460.
E-mail: http://www.frechet.com/ E-mail: frechet@cchem.berkeley.edu; Fax: +510 643 3079; Tel: +510 643
3078
Received (in Cambridge, UK) 4th October 2002, Accepted 29th October 2002
First published as an Advance Article on the web 13th November 2002
An acid catalysed rearrangement that transforms a bicyclic
lactone into a phenolic carboxylic acid has been tested for
potential use in chemically amplified microlithographic
imaging.
Standard silicon-based microchips are manufactured through an
iterative photolithographic process where individual compo-
nents are transferred into silicon via projection through a series
of patterned masks. Critical to the success of this technique is a
carefully designed polymeric photoresist. Coated on the silicon
substrate, the resist undergoes a solubility change only where
selectively irradiated and allows pattern transfer from the
Scheme 2
mask.1
In the early 1980’s we developed the highly sensitive
chemically amplified photolithographic resist system2 that
relies on the entropically favourable release of volatile by-
products to provide the driving force for pattern transfer via a
catalytically activated solubility switch. For example, phenolic
polymers containing tert-butyloxycarbonyloxy pendant groups
or polymers with tert-butyl ester groups are selectively
unmasked in imagewise fashion by a catalytic amount of a
photogenerated acid (PAG) releasing carbon dioxide and/or
2-methylpropene. While chemically amplified resists3–7 de-
rived from this early invention2 are used almost universally in
the fabrication of microelectronic devices today, the demand for
more powerful computers continues to push the limits of this
process for the production of ever smaller computer chips.8 One
way to accomplish this reduction is to alter traditional
photolithographic techniques to incorporate shorter irradiation
wavelengths.9 However, as exposure tools become more
expensive in order to accommodate this shift, damage to the
optics by the release of organic volatiles from the resist itself
becomes a major issue.10 To reduce the release of gaseous
components, we designed a new solubility switch model
employing an enthalpic rather than entropic driving force as
shown in Scheme 1. We proposed that, in the presence of a
catalytic amount of acid, lactone 18 would be converted to
antithetically soluble phenolic acid 2 because of the enthalp-
ically favourable formation of an aromatic ring. Being acid
catalysed, the rearrangement would also be compatible with the
existing type of PAG’s used today with chemically amplified
(CA) resists in photolithography.
bromine were attempted. Although this route only provides a
fair yield of the desired ketoester 1, it provided the best results
and enabled the production of a substantial amount of
product.11
Preliminary evaluation of the acid catalysed rearrangement
was accomplished by dissolving ketoester 1 in deuterated
dimethyl sulfoxide. This choice of solvent allows heating of the
solution of 1 to just under its boiling point of 189 °C. Several
different acids used in catalytic amount were tested for this
reaction and the various solutions were heated with monitoring
of reaction progress by NMR spectroscopy.12 In the cases where
the rearrangement occurred, both the desired phenolic acid 2
and some of the corresponding cyclised aromatic lactone by-
product 6 were formed (Scheme 3).
Scheme 3
Product distributions for each set of conditions are shown in
Table 1. As expected, no measurable conversion of 1 into 2 and/
or 6 was observed in the absence of acid catalyst upon heating
to a temperature of 185 °C. In the presence of triflic acid the
rearrangement takes place with initial formation of phenolic
acid 2 as the major product. Prolonged heating increases the
To test the feasibility of this proposed acid catalysed
rearrangement, we synthesized model lactone 1 as shown in
Scheme 2. Regioselective Birch reduction of 2-methoxypheny-
lacetic acid 3 provided exclusively enol ether 4, which was
acidified on workup to give ketoacid10 5 in approximately 80%
yield. Subsequent bromolactonization followed by in situ
elimination generated ketoester 1. It should be noted that
various lactonization conditions with sources of both iodine and
Table 1 Product distributions obtained upon heating ketoester 1 with
various acidsa
150 °C for
Acid
1 min
185 °C for 2 min
185 °C for 15 min
None
1
1
1
1
1
1
1
N/A
N/A
HNO3
HClO4
CF3SO3H
PAGb
7% 2 + 93% 1
35% 2 + 10% 6 + 55% 1
26% 2 + 8% 6 + 66% 1 48% 2 + 48% 6 + 4% 1
10% 2 + 90% 1 25% 2 + 8% 6 + 67% 1
a Product distributions determined by integration of 1H NMR yielded the
relative ratios of compounds 1, 2, and 6. 1H NMR spectra of compounds 2
and 6 were obtained from the literature.13 b Photoactive acid generator
(diphenyl [4-(phenylthio)phenyl] sulfonium hexafluoroantimonate salt).
Scheme 1
2956
CHEM. COMMUN., 2002, 2956–2957
This journal is © The Royal Society of Chemistry 2002

Technol., 1993, 6, 575; (d) R. D. Allen, G. M. Wallraff, R. Di Pietro, D.
Hofer and R. R. Kunz, J. Photopolym. Sci. Technol., 1994, 7, 507; (e) R.
D. Allen, I. Wan, G. M. Wallraff, R. Di Pietro and D. Hofer, J.
Photopolym. Sci. Technol., 1995, 8, 623; (f) E. Hasegawa, K. Maeda and
S. Iwasa, Polymers for Advanced Technologies, 2000, 11, 560.
5 (a) W. Hinsberg, G. Wallraff, F. Houle, M. Morrison, J. Frommer, R.
Beyers and J. Hutchinson, ACS Symposium Series, 1998, 695, 344; (b)
U. Okoroanyanwu, T. Shimokawa, J. Byers and C. G. Willson, Chem.
Mater., 1998, 10, 3319; (c) U. Okoroanyanwu, J. Byers, T. Shimokawa
and C. G. Willson, Chem. Mater., 1998, 10, 3328; (d) J. M. Klopp, D.
Pasini and J. M. J. Fréchet, Chem. Mater., 2001, 13, 4136; (e) J. M.
Klopp, D. Pasini, J. D. Byers, C. G. Willson and J. M. J. Fréchet, Chem.
Mater., 2001, 13, 4147; (f) H. Wu and K. E. Gonsalves, Adv. Mat., 2001,
13, 670.
6 (a) R. D. Allen, R. Sooriyakumaran, J. Opits, G. M. Wallraff, R. Di
Pietro, G. Breyta, D. Hofer, R. R. Kunz, S. Jayaraman, R. Shick, B.
Goodall, U. Okoroanyanwu and C. G. Willson, Proc. SPIE, 1996, 2724,
334; (b) R. D. Allen, R. Sooriyakumaran, J. Opits, G. M. Wallraff, R. Di
Pietro, G. Breyta, D. Hofer, R. R. Kunz, U. Okoroanyanwu and C. G.
Willson, J. Photopolym. Sci. Technol., 1996, 9, 465; (c) J. H. Lee, K. D.
Ahn and I. Cho, Polymer., 2001, 42, 1757.
7 (a) E. Reichmanis, O. Nalamasu and F. M. Houlihan, Acc. Chem. Res.,
1999, 32, 659; (b) F. M. Houlihan, T. I. Wallow, O. Nalamasu and E.
Reichmanis, Macromolecules, 1997, 30, 6517; (c) S.-J. Choi, Y. Kang,
D.-W. Jung, C.-G. Park and J.-T. Moon, Proc. SPIE, 1997, 3049, 104;
(d) A. H. Gabor, O. Dimov, A. N. Medina, M. O. Neisser, S. G. Slater,
R. H. Wang, F. M. Houlihan, R. A. Cirelli, G. Dabbagh, R. S. Hutton,
I. L. Rushkin, J. R. Sweeney, A. G. Timko, O. Nalamasu and E.
Reichmanis, Proc. SPIE, 1999, 3678, 221; (e) J. B. Kim, J. J. Lee and
J. S. Kang, Polymer., 2000, 41, 6939.
relative amount of aromatic lactone 6 that is formed, and, after
15 min, the reaction is almost complete affording an equal
mixture of 2 and 6. While the temperature required for this
rearrangement to take place clearly exceeds that used with
today’s CA resists, the fact that it occurs with triflic acid is
encouraging since the counterions of many photogenerated
acids are fluorinated sulfonates. In addition, past experience
with various generations of commercial CA resists has shown
that a reaction affecting only a small number of pendant groups
in a CA resist is sufficient to effect the desired solubility
switch.3b Differential scanning calorimetric measurements
suggest that the uncatalysed rearrangement of 1 into 2 takes
place near 225 °C with subsequent transformation of 2 into
lactone 6 taking place near 250 °C.
Although not yet optimised, ketoester 1 clearly demonstrates
that a rearrangement that unmasks both a phenol and a
carboxylic acid, affording the solubility switch necessary for
chemically amplified photolithography, can, in principle, take
place in ‘mass persistent’ fashion without the concomitant
evolution of large amounts of volatile organic material as
observed with today’s CA resists. In order to implement this
chemistry into an actual photoresist, structure 1 would have to
be incorporated as a pendant group in a polymer or copolymer
used in a chemically amplified formulation3b in conjunction
with a photoacid generator such as a triarylsulfonium salt.
Exposure of a coating of this formulation to radiation would
then generate acid within the resist, and a subsequent baking
step would accomplish the desired solubility change by
transforming the pendant group derived from 1 to one derived
from 2 (hence soluble in typical aqueous base developers) in the
exposed areas of the coating.
In summary, an organic soluble bicyclic lactone that can
undergo acid catalysed rearrangement to unmask moieties
soluble in aqueous base without extensive release of volatile
organic by-products has been designed and synthesized. This
design incorporates the enthalpically favourable formation of an
aromatic ring as the driving force rather than the typical
entropically favoured release of small molecules arising from
removal of a protecting group. Preliminary experiments per-
formed in solution have demonstrated that conversion of the
model compound can be achieved with aromatization and
change in solubility properties upon heating in the presence of
a catalytic amount of acid.
Funding from SEMATECH International and AFOSR is
gratefully acknowledged. Fellowship support (JK) from SRC is
also acknowledged with thanks. The authors also thank
Professor C. Grant Willson and members of his research group
at the University of Texas-Austin for stimulating discussions.
8 Semiconductor Industry Association (SIA), The National Technology
Roadmap for Semiconductors, San Jose, CA, 1997.
9 R. Hung, H. V. Tran, B. Trinque, T. Chiba, S. Yamada, D. Sanders, E.
Connor, R. Grubbs, J. Klopp, M. J. Fréchet, B. Thomas, G. Shafer, Des
Marteau, W. Conley and C. G. Willson, Proc. SPIE, 2001, 4345, 385; C.
G. Willson, B. C. Trinque, B. P. Osborn, C. R. Chambers, Y. Hsieh, T.
Chiba, P. Zimmerman, D. Miller and W. Conley, J. Photopolym. Sci.
Technol., 2002, 15, 583.
10 (a) A. C. Cefalas, E. Sarantopoulou, E. Gogolides and P. Argitis,
Microelec. Eng., 2000, 53, 123; (b) M. M. Chauhan and P. F. Nealey, J.
Vac. Sci. Technol. B, 2000, 18, 3402; (c) R. R. Kunz, V. Liberman and
D. K. Downs, Microlithogr. World, 2000, 9, 2.
11 (6-Oxocyclohex-3-en-1-yl)acetic acid (5). To 100 mL of condensed
ammonia stirring at 255 °C under argon was added a solution of 5.0 g
(30 mmol) of 2-methoxyphenylacetic acid and 7.2 mL (75 mmol) of
tert-butyl alcohol in 20 mL of tetrahydrofuran. Small pieces of 15 cm of
Li wire (97 mmol, 45 mg/cm) were added in three portions. The first
addition was immediate; the second followed it by one half hour; and the
third followed the second by one hour. The blue color persisted only
after the third addition. After the mixture had stirred overnight,
methanol was slowly added to discharge the remaining blue color. The
mixture was warmed to room temperature and 200 mL of 1N aqueous
HCl was added. This addition was followed by dropwise addition of
concentrated HCl until a pH of 0 was measured. The solution was stirred
at pH 0 for one minute and then raised to pH 4 with slow addition of a
1M aqueous NaOH solution. The aqueous layer was immediately
extracted with Et2O (2 3 200 mL). The combined organic layers were
washed with brine (1 3 100 mL), dried over MgSO4, and filtered. The
solvent was removed in vacuo to yield 3.67 g (79%) of acid 5.
12 The Aldrich Library of 13C and 1H FT NMR Spectra, ed. C. J. Pouchert
and J. Behnke, Aldrich Chemical Company, Inc., 1993, vol. 2, 1005B,
1306B.
13 2-Oxabicyclo[3.3.1]non-7-ene-3,6-dione (1). To a solution of 10.0 g (52
mmol) of acid 5 and 1.37 g (5.2 mmol) of 18-crown-6 ether in 260 mL
of dry CH2Cl2 stirring at 0 °C under N2 was added 14.3 g (104 mmol)
of potassium carbonate followed by 3.5 mL (67 mmol) of Br2 dropwise.
After the mixture had warmed to room temperature overnight, 200 mL
of water was added. Extraction of the crude material with CH2Cl2 (2 3
200 mL), evaporation of the solvent and chromatography of the
resulting oil in 3:2 ethyl acetate/hexanes yielded 2.68 g (34%) of ester
1 as a white solid.
Notes and references
1 For an extensive overview of microlithography see: Introduction to
Microlithography, ed. L. F. Thompson, C. G. Willson and M. J.
Bowden, American Chemical Society, Washington, DC, 1994.
2 (a) J. M. J. Fréchet, H. Ito and C. G. Willson, Proceedings Microcircuit
Engineering, 1982, 260; (b) J. M. J. Fréchet, E. Eichler, H. Ito and C. G.
Willson, Polymer, 1983, 24, 995; (c) H. Ito, C. G. Willson and J. M. J.
Fréchet, U.S. Patent, 4, 491, 628, 1985.
3 (a) C. G. Willson, H. Ito, J. M. J. Fréchet, T. G. Tessier and F. M.
Houlihan, J. Electrochem. Soc., 1986, 133, 181; (b) S. A. MacDonald,
C. G. Willson and J. M. J. Fréchet, Acc. Chem. Res., 1994, 27, 151.
4 (a) Y. Kaimoto, K. Nozaki, S. Takechi and N. Abe, Proc. SPIE, 1992,
1672, 66; (b) R. R. Kunz, R. D. Allen, W. D. Hinsberg and G. M.
Wallraff, Proc. SPIE, 1993, 1925, 167; (c) R. D. Allen, G. M. Wallraff,
W. D. Hinsberg, W. E. Conley and R. R. Kunz, J. Photopolym. Sci.
CHEM. COMMUN., 2002, 2956–2957
2957