Efficient photoacids based upon triarylamine dialkylsulfonium salts

Article abstract of DOI:10.1021/ja011186k

New triarylamine dialkylsulfonium salts that are photosensitive in the near-ultraviolet have been prepared. The quantum yields of photoacid generation were found to be ～0.5 and are independent of the counterion. On the other hand, the efficiencies of the sulfonium salts toward the photopolymerization of cyclohexene oxide depend on the counterion and the sulfonium substituents. Photopolymerization kinetic studies demonstrate that these triphenylamine sulfonium salts are highly efficient cationic photoinitiators.

Full text of DOI:10.1021/ja011186k

Published on Web 02/09/2002
Efficient Photoacids Based upon Triarylamine
Dialkylsulfonium Salts
Wenhui Zhou, Stephen M. Kuebler, Dave Carrig, Joseph W. Perry,* and
Seth R. Marder*
Contribution from the Department of Chemistry, UniVersity of Arizona, Tucson, Arizona 85721
Received May 14, 2001. Revised Manuscript Received October 29, 2001
Abstract: New triarylamine dialkylsulfonium salts that are photosensitive in the near-ultraviolet have been
prepared. The quantum yields of photoacid generation were found to be ∼0.5 and are independent of the
counterion. On the other hand, the efficiencies of the sulfonium salts toward the photopolymerization of
cyclohexene oxide depend on the counterion and the sulfonium substituents. Photopolymerization kinetic
studies demonstrate that these triphenylamine sulfonium salts are highly efficient cationic photoinitiators.
Scheme 1
Introduction
Photoacids are used extensively in both positive- and nega-
tive-tone photoresist formulations in the microelectronics in-
dustry.1 In particular, iodonium and sulfonium salts have found
widespread use as initiators for cationic photopolymerization
of epoxides and vinyl ethers,2 and for the cleavage of tert-
butoxycarbonyl, t-BOC, esters in chemically amplified resists.3
Saeva4 has shown that for arylsulfonium salts, the photofrag-
mentation can be initiated either by direct excitation of a π-σ*
(sulfur-carbon bond) transition or by excitation of a π-π*
transition followed by intramolecular electron transfer to the
σ* orbital of the sulfur-carbon bond. In both cases, subsequent
homolytic cleavage of the sulfur-carbon bond leads to forma-
tion of a cation-radical on the sulfur-containing fragment, and
a neutral radical (Scheme 1, top). These species undergo
secondary reactions, such as radical coupling or H-transfer, to
produce Brønsted acid, which are important in the decomposi-
tion of some sulfonium salts. In the past, this reaction has been
facilitated by stabilizing the radical fragment.4-7 In contrast,
phenyldimethyl sulfonium salts having a σ* LUMO are not
photoactive toward acid generation. This has been attributed to
the strength of the S-CH3 bond (or, in other words, the
instability of a methyl radical), which allows radiative and
nonradiative decay pathways to compete effectively with C-S
bond cleavage.6 Provided that the energy of the excited state is
sufficient enough to populate σ*, it then follows from the
Hammond postulate that both the driving force and rate for bond
homolysis should be strongly affected by the stability of both
the radical cation and the neutral radical. Herein, we explore
the hypothesis that the quantum efficiency of photoacid genera-
tion can also be enhanced by incorporating triarylamine func-
tionalities that stabilize the radical cation intermediate formed
upon homolysis of the carbon-sulfur bond of aryl dialkyl
sulfonium salts (Scheme 1, middle and bottom). Triarylamine
sulfonium salts were selected as the initial targets to test our
hypothesis because (i) triarylamine radical cations are known
to be very stable; (ii) protonated triarylamino groups are
extremely strong acids (Ph3N+H, pKa ) -5),8 thus the presence
of this amine functionality should not inhibit reactions such as
the ring-opening polymerization of epoxides; and (iii) the
conditions used in the preparation of aryl dialkyl sulfonium salts9
are compatible with those of the triarylamine functional group.
(1) (a) Roffey, C. Photogeneration of ReactiVe Species for UV Curing;
Wiley: New York, 1997. Fouassier, J.-P. Photoinitiation, Photopolymer-
ization, and Photocuring: Fundamentals and Applications; Hanser/
Gardner: Cincinnati, OH, 1995. (b) Shirai, M.; Tsunooka, M. Prog. Polym.
Sci. 1996, 21, 1. (c) Photopolymerisation and Photoimaging Science and
Technology; Allen, N. S., Ed.; Elsevier: London, 1989.
(2) Crivello, J. V.; Lam, J. H. W. Macromolecules, 1977, 10, 1307. Crivello,
J. V.; Lee, J. J. Polym. Sci. Polym. Chem. Ed. 1989, 27 3951. Crivello, J.
V.; Lam, J. H. W.; Volante, C. N. J. Rad. Curing 1977, 4, 2. Pappas, S. P.;
Pappas, B. C.; Gatechair, L. R.; Jilek, J. H. Polym. Photochem. 1984, 5, 1.
Welsh, K. M.; Dektar, J. L.; Garcia-Garibay, M. A.; Hacker, N. P.; Turro,
N. J. J. Org. Chem. 1992, 57, 4179. Crivello, J. V.; Kong, S. Macromol-
ecules 2000, 33, 825.
(3) Ito, H.; Willson, C. G.; Frechet, J. M. J.; Farrell, M. J.; Eichler, E.
Macromolecules 1983, 16, 510. Ito, H.; Willson, C. G.; Frechet, J. M. J.,
US Patent 4,491,628, 1985.
(4) Saeva, F. D. AdV. Electron-Transfer Chem. 1994, 4, 1.
(5) Saeva, F. D.; Brelin, D. T. J. Org. Chem. 1989, 54, 712.
(6) Saeva, F. D.; Brelin, D. T.; Martic, P. A. J. Am. Chem. Soc. 1989, 111,
1328.
(7) Andrieux, C. P.; Robert, M.; Saeva, F. D.; Save´ant, J.-M. J. Am. Chem.
Soc. 1994, 116, 7864.
Experimental Section
Absorption spectra were recorded on a Hewlett-Packard model 8453
spectrophotometer. Fluorescence spectra were collected on a Jobin Yvon
Spex Fluorolog-III fluorimeter. Gas chromatography-mass spectrom-
etry spectra were obtained on a Hewlett-Packard model 6890
(8) March, J. AdVanced Organic Chemistry, 3rd ed.; John Wiley & Sons Ltd:
New York, 1985; p 225.
(9) Saeva, F. D.; Morgan, B. P. J. Am. Chem. Soc. 1984, 106, 4121.
9
10.1021/ja011186k CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. VOL. 124, NO. 9, 2002 1897

A R T I C L E S
Zhou et al.
GC system with a model 5973 mass selective detector. NMR spectra
were recorded on a Varian Unity Plus-500 (500 MHz) spectrometer.
Combustion elemental analyses were performed by Desert Analytics.
Standard reduction potentials were measured on a BAS Model 100B/W
cyclic voltammetry unit. A glassy-carbon electrode was used as the
working electrode along with a platinum auxiliary electrode and a
standard Ag/AgCl reference electrode. Spectrophotometric grade ac-
etonitrile (Aldrich) was used as received in the measurement of
absorption and fluorescence spectra. Triphenylsulfonium hexafluoro-
antimonate (10) was synthesized following a literature procedure10 and
recrystallized from 2-propanol. Phenyl dimethylsulfonium triflate was
synthesized following a literature procedure.9 All other commercially
available reagent grade chemicals were obtained from Aldrich and used
without further purification.
13C NMR (125 MHz, CDCl3) δ 148.3, 147.5, 139.3, 129.4. 129.2, 124.3,
122.9, 121.5, 120.5, 120.2, 15.6 ppm. EIMS (rel intensity %) 291 (100,
M+).
Preparation of 3-Benzylthiotriphenylamine (4). 3-Bromophenyl
benzyl sulfide (2) (1.86 g, 6.67 mmol) was added to a solution of tris-
(dibenzylideneacetone)dipalladium (Pd2(dba)3) (0.15 g, 0.16 mmol) and
bis(diphenylphosphino)ferrocene (DPPF) (0.11 g, 0.20 mmol) in dry
toluene (20 mL) under nitrogen atmosphere at room temperature. The
resultant mixture was stirred for 10 min. Sodium tert-butoxide (1.42
g) and diphenylamine (1.13 g, 6.69 mmol) were then added to this
solution which was stirred at 90 °C for 24 h. The reaction mixture was
poured into 20 mL of water, extracted three times with ether (3 × 60
mL), and dried over anhydrous magnesium sulfate. The product was
purified by flash column chromatography with use of 2% ethyl acetate
in hexanes as eluent to give 1.53 g of a pale yellow oil (62.5%). The
purity of the product was estimated greater than 95% by 1H NMR and
Preparation of 3-Bromothioanisole (1). 3-Bromobenzenethiol (5.0
g, 26.44 mmol) was added to a solution of sodium methoxide (1.43 g,
26.48 mmol) in 20 mL of anhydrous methanol. The mixture was stirred
for 30 min under nitrogen at room temperature and then a solution of
methyl iodide (4.51 g, 31.77 mmol) in 20 mL of anhydrous methanol
was added dropwise. The reaction mixture was stirred overnight under
nitrogen at room temperature, poured into 2 M aqueous sodium
hydroxide, and extracted with ether (3 × 50 mL). The combined organic
layer was washed with saturated brine and dried over anhydrous
magnesium sulfate. After removal of solvent, the product was purified
by distillation at 80 °C (0.3 mmHg) to give 4.2 g of a colorless oil
1
was used in the next step without further purification. H NMR (500
MHz, CDCl3) δ 7.2-7.3 (m, overlap, 9H), 7.10 (t, J ) 8.0 Hz, 1H),
7.0-7.06 (m, overlap, 7H), 6.93 (d, br, J ) 8.0 Hz, 1H), 6.86 (d, br,
J ) 8.0 Hz, 1H), 4.04 ppm (s, 2H, CH2). 13C NMR (125 MHz, CDCl3)
δ 148.2, 147.4, 137.4, 137.2, 129.4, 129.2, 128.7, 128.4, 127.1, 124.4,
123.4, 123.0, 121.6. 38.7 ppm (one carbon was not observed). EIMS
(rel intensity %) 367 (100, M+).
Preparation of [3-(N,N-Diphenyl)amino]phenyl Dimethyl Sulfo-
nium Trifluoromethanesulfonate (5). 3-Methylthiotriphenylamine (3)
(1.77 g, 6.08 mmol) was dissolved in 30 mL of dry methylene chloride
and cooled to -78 °C in the dark. To this solution was added via syringe
0.76 mL (1.10 g, 6.70 mmol) of methyl trifluoromethanesulfonate under
nitrogen with stirring for 30 min while the temperature was maintained
at -78 °C. The resultant mixture was then stirred overnight at room
temperature. Ether (60 mL) was added to the mixture resulting in the
slow formation of white crystals. The crystals were collected by
filtration and washed three times with ether. The product was purified
by recrystallization from methylene chloride and ether at room
1
(86.1%). H NMR (500 MHz, CDCl3) δ 7.37 (s, br, 1H), 7.26 (d, br,
1H), 7.12-7.20 (m, 2H), 2.50 ppm (s, 3H, CH3). 13C NMR (125 MHz,
CDCl3) δ 140.9, 130.0, 128.6, 127.9, 124.9, 122.9, 15.6 ppm. Electron
ionization mass spectrometry (EIMS, rel intensity %) 202, 204 (1:1,
100, M+).
Preparation of 3-Bromophenyl Benzyl Sulfide (2). 3-Bromoben-
zenethiol (5.0 g, 26.44 mmol) was added to a solution of sodium
methoxide (1.43 g, 26.48 mmol) in 20 mL of anhydrous methanol.
The mixture was stirred for 30 min at room temperature under nitrogen
and then a solution of benzyl bromide (4.53 g, 26.48 mmol) in 20 mL
of anhydrous methanol was added dropwise. The reaction mixture was
stirred overnight under nitrogen at room temperature, poured into a 2
M aqueous solution of sodium hydroxide, and extracted with ether (3
× 60 mL). The combined organic layer was washed with saturated
brine and dried over anhydrous magnesium sulfate. After removal of
solvent, the product was purified by distillation at 161 °C (0.3 mmHg)
to give 6.31 g of colorless oil (85.5%). 1H NMR (500 MHz, CDCl3) δ
7.44 (t, J ) 1.7 Hz, 1H), 7.27-7.35 (m, 5H), 7.26 (m, 1H), 7.02 (dt,
J ) 8.0, 1.5 Hz, 1H), 7.11 (t, J ) 8.0 Hz, 1H), 4.17 ppm (s, 2H, CH2).
13C NMR (125 MHz, CDCl3) δ 138.8, 136.7, 131.7, 130.1, 129.2, 128.8,
128.6, 127.8, 127.4, 122.6, 38.7 ppm. EIMS (rel intensity %) 278, 280
(1:1, 30.8, M+).
1
temperature to give a yield of 2.45 g (88.6%). H NMR (500 MHz,
d6-DMSO) δ 7.61 (d, br, J ) 8.5 Hz, 1H), 7.58 (d, J ) 8.5 Hz, 1H),
7.55 (t, J ) 1.7 Hz, 1H), 7.38 (t, br, J ) 8.0 Hz, 4H), 7.13-7.20 (m,
3H), 7.09 (d, br, J ) 7.5, 4H), 3.19 ppm (s, 6H, CH3). Anal. Calcd for
C21H20NO3S2F3: C, 55.37; H, 4.42; N, 3.09. Found: C, 55.26; H, 4.59;
N, 3.19.
Preparation of [3-(N,N-Diphenyl)amino]phenyl Dimethyl Sulfo-
nium Hexafluorophosphate (7). A fresh 20 mL solution of KPF6 (1.14
g, 5.98 mmol) in water was added to a solution of [3-(N,N-diphenyl)-
amino]phenyl dimethyl sulfonium trifluoromethanesulfonate (5) (1.3
g, 2.86 mmol) in 15 mL of acetone. The mixture was stirred for 2 h at
room temperature in the dark. The solid was collected by filtration
and redissolved in 10 mL of acetone. The above anion-exchange was
repeated three times. The resultant white solid was washed three times
with water and ether. The solid was purified by two precipitations from
a 10 mL acetone solution through the addition of 50 mL of diethyl
ether. The final product yield was 1.08 g (83.6%). 1H NMR (500 MHz,
d6-DMSO) δ 7.62 (d, J ) 7.5 Hz, 1H), 7.57 (d, J ) 8.0 Hz, 1H), 7.55
(s, 1H), 7.38 (t, br, J ) 8.2 Hz, 4H), 7.13-7.20 (m, 3H), 7.09 (d, J )
8.5, 4H), 3.19 ppm (s, 6H, CH3). Anal. Calcd for C20H20NSF6P: C,
53.22; H, 4.47; N, 3.10. Found: C, 53.20; H, 4.29; N, 3.13.
Preparation of 3-Methylthiotriphenylamine (3). 3-Bromothioani-
sole (1) (2.0 g, 9.85 mmol) was added to a solution of tris-
(dibenzylideneacetone)dipalladium (Pd2(dba)3) (0.28 g, 0.306 mmol)
and bis(diphenylphosphino)ferrocene (DPPF) (0.245 g, 0.442 mmol)
in dry toluene (20 mL) under nitrogen atmosphere at room temperature.
The resultant mixture was stirred for 10 min. Sodium tert-butoxide
(2.17 g) and diphenylamine (1.67 g, 9.85 mmol) were then added and
stirred at 90 °C for 24 h. The reaction mixture was poured into 20 mL
of water, extracted three times with ether (3 × 60 mL), and dried over
anhydrous magnesium sulfate. The product was purified by flash column
chromatography with use of 2% ethyl acetate in hexanes as eluent to
give 1.87 g of a pale yellow oil (65.2%). The purity of the product
was estimated greater than 95% by 1H NMR and was used in the next
Preparation of [3-(N,N-Diphenyl)amino]phenyl Dimethyl Sulfo-
nium Hexafluoroantimonate (8). A fresh 20 mL portion of NaSbF6
(1.14 g, 4.40 mmol) in water was added to a solution of [3-(N,N-
diphenyl)amino]phenyl dimethyl sulfonium trifluoromethanesulfonate
(5) (1.0 g, 2.19 mmol) in 10 mL of acetone. The mixture was stirred
for 2 h at room temperature in the dark. The solid was collected by
filtration and redissolved in 10 mL of acetone. The above anion-
exchange was repeated three times. The resulting white solid was
washed three times with water and ether. The product was purified by
two precipitations from a 10 mL acetone solution through the addition
of 50 mL of diethyl ether. The final product yield was 1.02 g (85.9%).
1
step without further purification. H NMR (500 MHz, CDCl3) δ 7.25
(t, br, J ) 7.5 Hz, 4H), 7.14 (t, J ) 8.2 Hz, 1H), 7.08 (d, br, J ) 8.2
Hz, 4H), 7.02 (t, br, J ) 7.5 Hz, 2H), 6.97 (t, J ) 2.0 Hz, 1H), 6.87
(d, br, J ) 8.0, 1H), 6.82 (d, br, J ) 8.0, 1H), 2.20 ppm (s, 3H, CH3).
(10) Crivello, J. V.; Lam, J. H. W. J. Org. Chem. 1978, 43, 3055.
9
1898 J. AM. CHEM. SOC. VOL. 124, NO. 9, 2002

Triarylamine Dialkylsulfonium Salts
A R T I C L E S
1H NMR (500 MHz, d6-DMSO) δ 7.62 (d, J ) 7.5 Hz, 1H), 7.57 (d,
J ) 8.0 Hz, 1H), 7.55 (s, 1H), 7.38 (t, br, J ) 8.2 Hz, 4H), 7.13-7.20
(m, overlap, 3H), 7.09 (d, J ) 8.5, 4H), 3.19 ppm (s, 6H, CH3). Anal.
Calcd for C20H20NSF6Sb: C, 44.30; H, 3.72; N, 2.58. Found: C, 44.29;
H, 3.76; N, 2.54.
Photopolymerizations. Aliquots (1.0 mL) of a solution containing
3.16 × 10-3 mol/L of either 5, 7, 8, 9, or 10 and 7.91 mol/L of
cyclohexene oxide in methylene chloride were sealed in 4-mL capped
vials and irradiated at 300 nm in a merry-go-round holder. After various
intervals the sample tubes were withdrawn from the irradiation chamber
and any ionic reactions were quenched by the immediate addition of 1
mL of 2 M ammonia in methanol. Methanol was added and the solutions
were stirred until the polymers precipitated as powders. The powders
were collected by filtration and dried in vacuo for 40 h. The conversion
of monomer was determined gravimetrically.
Preparation of [3-(N,N-Diphenyl)amino]phenyl Benzyl Methyl
Sulfonium Hexafluoroantimonate (9). 3-Benzylthiotriphenylamine
(1.48 g, 4.03 mmol) was dissolved in 30 mL of dry methylene chloride
and cooled to -78 °C in the dark. To this solution was added via syringe
0.46 mL (0.67 g, 4.07 mmol) of methyl trifluoromethanesulfonate under
nitrogen. The mixture was then stirred for 30 min while the temperature
was maintained at -78 °C. The mixture was then allowed to warm to
room temperature and was stirred overnight. Ether (60 mL) was added
resulting in the formation of a light green oil. The solvent was decanted
and the oil was dried in vacuo at room temperature. The product was
used in the ion-metathesis described below without further purification.
A fresh 20 mL portion of aqueous NaSbF6 (2.10 g, 8.11 mmol) was
added to a solution of [3-(N,N-diphenyl)amino]phenyl benzyl methyl
sulfonium trifluoromethane-sulfonate (6) (1.3 g, 2.86 mmol) in 10 mL
of acetone. The mixture was stirred for 2 h at room temperature in the
dark. The solid was collected by filtration and redissolved in 10 mL of
acetone. The above anion-exchange was repeated three times. The
resultant white solid was washed three times with water and ether. The
product was purified by reprecipitation twice from a 10 mL acetone
solution by the addition of 50 mL of diethyl ether giving a final yield
of 1.08 g (71.8%). 1H NMR (500 MHz, CD3COCD3) δ 7.58-7.62 (m,
overlap, 2H), 7.55 (t, J ) 7.7 Hz, 1H), 7.46 (t, J ) 7.5 Hz, 2H), 7.36
(t, J ) 7.7 Hz, 4H), 7.25-7.32 (m, 4H), 7.17 (t, J ) 7.7 Hz, 2H), 7.02
(d, J ) 8.0, 4H), 5.25 (d, J ) 12.5 Hz, 1H, CH2), 5.02 (d, J ) 13.0
Hz, 1H, CH2), 3.52 ppm (s, 3H, CH3). Anal. Calcd for C26H24NSF6Sb:
C, 50.51; H, 3.91; N, 2.26. Found: C, 50.42; H, 3.98; N, 2.19.
Photolysis of Triphenylamine Sulfonium Salt and Identification
of Photoproducts. A 2 mL sample of a 0.05 M solution of 5 in
acetonitrile in a glass vessel was purged with nitrogen for 30 min,
sealed, and irradiated for 1 h with a 300 nm photochemical lamp. The
reaction mixture was poured into 5 mL of water and extracted twice
with 10 mL of ether. The combined extract was concentrated to 10
mL under reduced pressure and analyzed by GC-MS. The presence of
the H-transfer product 3-methylthio triphenylamine was confirmed by
comparison with an authentic sample, and the other coupling products
were identified by their mass spectra. The relative product distribution
was estimated from the peak areas, neglecting differences in their
response factors.
Results and Discussion
Synthesis. Triphenylamines bearing sulfonium groups were
prepared from 3-bromothiophenol in four steps (Scheme 2).
Alkylation of sodium 3-bromothiophenoxide with alkyl halides
in methanol produced 3-bromophenyl alkyl sulfides 1 and 2 in
yields of 86.1% and 85.5%, respectively. Pd-catalyzed coupling
of bromophenyl alkyl sulfides 1 and 2 with diphenylamine
following the method of Buchwald and Hartwig13 gave triph-
enylamine sulfide precursors 3 and 4 in yields of 65.2% and
62.5%, respectively. Finally, alkylation of 3 and 4, with methyl
trifluoromethanesulfonate employing a modified version of the
method of Saeva,9 gives triphenylamine sulfonium triflate salts
5 and 6 in yields generally above 80%. For epoxide polymer-
ization, it is known that the most active photoacid generators
-2
have nonnucleophilic anions, such as PF6-, AsF6-, or SbF6
.
Therefore, we exchanged the triflate anion for PF6-, AsF6-, or
SbF6-, by anion-exchange reactions that were driven to comple-
tion by repeating the metathesis three times using an excess of
either the potassium or sodium salt of the anion.
Scheme 2
Measurement of Photoacid Generation Quantum Yields. The
photoacid generation quantum yields were measured by using Rhodamine
B base as an acid indicator,11 which is protonated by acid produced in
the photolysis to give a characteristic absorbance at 555 nm in
acetonitrile. The acid concentration in acetonitrile was determined by
a calibration curve of rhodamine B titrated with p-tolunesulfonic acid
in acetonitrile. The monochromated lamp output of the Fluorolog-III
fluorimeter was used as the irradiation source. The photon doses at
245 and 294 nm were determined by ferrioxalate actinometry and the
uncertainty of the measured dose is (3%.12 Acetonitrile solutions of
5, 7, 8, and 9 were irradiated at 294 nm, and solutions of 10 were
irradiated at 245 nm. The optical density of the samples was greater
than 2.5 at the irradiation wavelength, so all photons incident on the
solution could be assumed to be absorbed. The dose rates were kept
sufficiently small that the maximum conversion was lower than 15%.
The uncertainties in the photoacid generation quantum yields are
estimated to be (10%.
Photophysical and Polymerization Studies. To estimate the
driving force for photoinduced electron transfer from the
triphenylamine group to the sulfonium moiety, we calculated
the free energy change (∆Get) using the Rehm-Weller equation
(eq 1).14 Here, Eox(D/D+•) and Ered(A-•/A) are the oxidation
potential of the electron donor and reduction potential of the
electron acceptor, respectively. E*(0,0) is the energy of the first
excited state and Ecoul. is the Coulombic energy.
∆Get ) Eox(D/D+•) - Ered(A-•/A) - E*(0,0) - Ecoul. (1)
The energy of the excited singlet state E*(0,0) was obtained
from the wavelength of intersection of the normalized absorption
and emission spectra, and was estimated to be 349 kJ/mol for
each sulfonium salt. Each sulfonium salt has one irreversible
(11) Ortica, F.; Scaiano, J. C.; Pohler, G.; Cameron, J. F.; Zampini, A. Chem.
Mater. 2000, 12, 414. Pohler, G.; Scaiano, J. C. Chem. Mater. 1997, 9,
3222.
(12) Parker, C. A. Photoluminescence of Solutions: with Applications to
Photochemistry and Analytical Chemistry; Elsevier: Amsterdam, The
Netherlands, 1968. Murov, S. L.; Carmichael, I.; Hug, G. L Handbook of
Photochemistry, 2nd ed.; Marcel-Dekker: New York, 1993.
(13) Wolfe, J. P.; Rennels, R. A.; Buchwald, S. L. Tetrahedron 1996, 52, 7525.
Wolfe, J. P.; Wagaw, S.; Buchwald, S. L. J. Am. Chem. Soc. 1996, 118,
7217. Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1996, 118, 72117.
(14) Rehm, D.; Weller, A. Isr. J. Chem. 1970, 8, 259.
9
J. AM. CHEM. SOC. VOL. 124, NO. 9, 2002 1899

A R T I C L E S
Zhou et al.
Table 1. Parameters for Free Energy Change of Photoinduced
Intramolecular Electron Transfer and the Relative Fluorescence
Quantum Yield of Triphenylamine Sulfonium Salts
a
a
E
,
E
,
E*(0,0),
kJ/mol
∆G ,
kJ/mol
λmax,fl,
nm
ox1
ox2
et
V
V
E
red,a V
Ιflb
+
ΦH
5
7
8
9
1.03
1.02
1.01
1.03
1.23
1.24
1.22
1.30
-1.79
-1.72
-1.74
349
349
349
349
-77.3
-85.3
-83.6
379
376
375
376
0.02
0.02
0.02
0.02
0.48
0.48
0.47
0.41
a Peak values vs Ag/AgCl (300 mV/s scan rate) at the constant
concentration of 10-3 mol/L of sulfonium salts with 0.1 N tetrabutylam-
monium hexafluorophosphonate as electrolyte. b The peak intensity of
fluorescence (excited at λmaxabs 294 nm) relative to triphenylamine (excited
abs
at λmax 297 nm) under the same conditions.
Figure 2. Photoacid concentration of 8 as a function of the absorbed photon
dose at 294 nm in acetonitrile. Concentration, 1.5 × 10-4 mol/L; OD245
∼2.5; maximum conversion is less than 15%.
of that of triphenylamine), and the emission maximum shifts
to shorter wavelengths. These data are consistent with decom-
position of the sulfonium fragment and formation of fluorescent
neutral triarylamines upon irradiation. Analysis of the photolysis
products by GC-MS showed that the products consisted of 52%
3-methylthiotriphenylamine (3) (identified by comparison with
an authentic sample), ∼10% of isomers of methyl substituted
3-methylthiotriphenylamine, consistent with previous findings
for other photoacid systems, and two products (∼18% each)
whose molecular mass is consistent with intramolecular coupling
of aryl rings to form carbazoles.
The photoacid generation quantum yields were measured by
the method reported by Scaiano.11 For triphenylamine sulfonium
salt 8 a plot of the acid concentration versus absorbed photon
dose (Einsteins) was linear, and the slope of the line was used
to determine the photoacid generation quantum yield (Figure
2). Similarly the quantum yield for triphenylsulfonium salt 10
was measured and the value of 0.50 was obtained. This is in
good agreement with the literature value of 0.53.16 The values
of acid generation quantum yields determined in this manner
for salts 5, 7, 8, and 9 are 0.48, 0.48, 0.47, and 0.41, respectively.
Thus, each one of these salts is an efficient photoacid.
These results are very different from those observed for a
simple phenyl dimethyl sulfonium salt, which, as noted above,
is not active in photoacid generation, but are consistent with
Saeva’s observation that the anthracenyl-substituted dimethyl
sulfonium salt 11 is an active photoacid.17 In the case of a simple
phenyl dimethyl sulfonium salt, the σ*orbital (S-C bond) can
be populated either by direct π-σ*excitation or by π* to σ*
intramolecular electron transfer. The dissociation of the π-σ*
excited state is energetically unfavorable due to the high total
energy of the methyl radical and the phenyl methyl sulfide
radical cation species. In contrast, the high acid generation
quantum yields of the triphenylamine sulfonium salts reported
Figure 1. UV-visible absorption spectra of triphenylamine and sulfonium
8 and 9 in acetonitrile. Concentration, 5.0 × 10-5 mol/L.
electrochemical reduction wave and two irreversible oxidation
waves. The ∆Get values were calculated by using the Rehm-
Weller equation, wherein the peak values of the reduction wave
and the first oxidation wave were used to estimate Ered(A-•/A)
and Eox(D/D+•), respectively.15 The Coulombic energy Ecoul. was
neglected in the calculation. Despite the uncertainty in determin-
ing the exact oxidation and reduction potential for the com-
pounds, due to the oxidation and reduction waves being
irreversible, the results, shown in Table 1, nonetheless indicate
that for these sulfonium salts electron transfer from the aromatic
amine to the C-S σ* orbital is unfavored in the ground state
(∆Get > 0) and that there is a large driving force for electron
transfer (∆Get < 0) from the first excited singlet state.
The positions of the absorption maxima are virtually identical
for all the sulfonium salts (Figure 1). It should be noted that
the absorption spectra of these sulfonium salts are similar to
that of triphenylamine, but have an additional feature on the
low energy side. In previous studies on aryl sulfonium ions, a
low energy band relative to the aryl π-π* band has been
assigned to a π-σ* transition5 (Figure 1). Irradiation of the
triphenylamine sulfonium salts at 294 nm results in a rapid
increase in the fluorescence intensity (which is initially ∼2%
(16) Hacker, N. P.; Dektar, J. L. J. Am. Chem. Soc. 1990, 112, 6004.
(17) Saeva, F. D.; Garcia, E.; Martic, P. A. J. Photochem. Photobiol. A: Chem.
1995, 86, 149.
(15) As Eox1/2 < Eoxp and Ered1/2 > Eredp, the real ∆Get value may be more negative
than the value calculated from peak potential values.
9
1900 J. AM. CHEM. SOC. VOL. 124, NO. 9, 2002

Triarylamine Dialkylsulfonium Salts
A R T I C L E S
Scheme 3
here show that dissociation is energetically favorable. We
suggest that the electron donating ability of the diphenylamine
moiety strongly activates the dissociation of the S-C bond by
stabilizing the triphenylamine methyl sulfide cation product
species, such that the total product energy now favors dissocia-
tion from the π-σ* excited state.
In the approach used by Saeva, increasing the size of the
π-conjugated electron donor apparently provides the energetic
stabilization of the radical cation product needed for bond
dissociation to be favorable. For example, for compound 11
(Scheme 3), which has a large delocalized chromophore (an
anthracenyl group), there is still adequate driving force for
excited-state electron transfer from π* of anthracene to σ* of
the C-S bond (∆G ) -19.6 kJ/mol), as evidenced by a low
fluorescence quantum yield of <0.0117 and a photoacid genera-
tion quantum yield of 0.27.17 However, a further increase in
the size of the chromophore results in a lowering of the energy
of the π* state, thus decreasing the driving force for the initial
electron-transfer reaction (∆G ) 7.5 kJ/mol). As a result, an
analogous naphthacene photoacid, 12, shows negligible photo-
acid generation (the photoacid generation quantum yield is
0.003), and this was attributed to a low forward rate of electron
transfer that cannot compete with the rate of radiative decay
and other nonradiative processes, as evidenced by the relatively
high fluorescence quantum yield of 0.47.17 In the approach that
we have employed, substitution of the phenyl group attached
to the sulfonium with an electron rich diphenylamine provides
both stabilization of the radical cation and a large driving force
for electron transfer, as shown in Table 1.
Figure 3 shows a comparison of the photopolymerization of
cyclohexene oxide with 5, 7, 8, and 9. In accord with previous
observations2 large anion-dependent variations in the photopo-
lymerization rate were observed. Sulfonium salt 5, which bears
a triflate anion, does not initiate polymerization, even after 1 h
of irradiation. In contrast, 8 and 9, having a SbF6- anion, rapidly
initiate polymerization, reaching 90% conversion to polymer
in 130 s. Although 8 and 9 both have SbF6- as the counterion,
have similar UV absorption spectra, and have comparable acid
generation quantum yields, the polymerization rate of 9 is higher
than that of 8. Park and co-workers suggested that in addition
to electron-transfer mediated acid-generation, benzyl-substituted
sulfonium salts undergo heterolytic bond cleavage in the excited
state to yield a benzylic cation that can itself initiate polymer-
ization efficiently.18 This could account for why 9 exhibits a
higher rate of photopolymerization relative to 8, as formation
of the methyl cation is 316 kJ/mol unfavorable relative to benzyl
cation (in the gas phase).19 Interestingly, 8 and 9 were found to
initiate polymerization with extremely short induction times
Figure 3. Photopolymerization kinetics of cyclohexene oxide initiated by
different triphenylamine sulfonium salts in dichloromethane irradiated at
300 nm. Concentration of monomer, 7.91 mol/L; concentration of initiator,
3.16 × 10-3 mol/L. Lines are a guide to the eye.
relative to triphenylsulfonium hexafluoroantimonate, when
irradiated at 300 nm, under otherwise identical conditions. This
demonstrates that triphenylamine sulfonium salts are highly
efficient cationic initiators that are sensitive to 300 nm radiation.
Conclusions
These sulfonium salts, bearing nonnucleophilic counterions,
form acids very efficiently upon exposure to 300 nm radiation.
Stabilization of the radical cation on the sulfur-containing
fragment greatly enhances the rate and efficiency of homolytic
cleavage of the carbon-sulfur bond. The acids produced in this
manner are sufficiently strong to initiate polymerization of
epoxides, and thus these sulfonium salts should be useful for
photocuring highly viscous media or coatings, as well as for
acid-activated material processing and fabrication in the solid
phase. Finally, this approach suggests a new design strategy
for the synthesis of very efficient photoacids that could be
excited over a broad spectral range by either single-photon or
multiphoton excitation.20
Acknowledgment. Support of this research by the NSF
(Chemistry Division) and the Office of Naval Research (through
CAMP) is gratefully acknowledged.
Supporting Information Available: 1H NMR spectra of
triphenylamine sulfide 3 and 4 and 1H NMR spectra of
triphenylamine sulfonium 5, 7, 8 and 9 (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
JA011186K
(19) Lowry, T. H.; Richardson, K. S. Mechanisms and Theory in Organic
Chemistry, 3rd ed; Harper and Row: New York, 1987.
(20) Cumpston, B. H.; Ananthavel, S. P.; Barlow, S.; Dyer, D.; Erhlich, J. E.;
Erskine, L. L.; Heikal, A. A.; Kuebler, S. M.; Lee, I.-Y. S.; McCord-
Maughon, D.; Qin, J.; Ro¨ckel, H.; Rumi, M.; Wu, X.-L.; Marder, S. R.;
Perry, J. W. Nature 1999, 398, 51.
(18) Park, J.; Kihara, N.; Ikeda, T.; Endo, T. Macromolecules 1997, 30, 3414.
9
J. AM. CHEM. SOC. VOL. 124, NO. 9, 2002 1901