Photochemical generation of thiophene analogs of 9-fluorenyl cations

Article abstract of DOI:10.1002/poc.1754

The 9-fluorenyl cation is a member of the 4N Hueckel antiaromatic series of intermediates, first observed by time-resolved spectroscopy on UV photo-excitation of 9-fluorenol. Thiophene analogs of 9-fluorenol in which one of the annelated benzene rings is

Full text of DOI:10.1002/poc.1754

Research Article
Received: 9 April 2010,
Revised: 14 May 2010,
Accepted: 19 May 2010,
Published online in Wiley Online Library: 29 July 2010
(wileyonlinelibrary.com) DOI 10.1002/poc.1754
Photochemical generation of thiophene
analogs of 9-fluorenyl cations
Mathew Bancerza, Lawrence A. Huckb, William J. Leighb,
Gabriela Mladenovaa, Katayoun Najafiana, Xiaofeng Zenga
and Edward Lee-Ruffa
*
The 9-fluorenyl cation is a member of the 4N Hu¨ckel antiaromatic series of intermediates, first observed by
time-resolved spectroscopy on UV photo-excitation of 9-fluorenol. Thiophene analogs of 9-fluorenol in which one
of the annelated benzene rings is replaced by a thiophene were prepared in three regioisomeric forms, and subjected
to preparative and laser flash photolysis. Photoproduct studies in methanol indicated products derived from the
corresponding fluorenyl cations and radical intermediates. Time-resolved spectroscopy showed transients which were
assigned to the corresponding cations as evident from methanol quenching. The lifetimes and methanol quenching
rates of these transients were compared with those of parent fluorenyl cation, and were found to be about two orders
of magnitude more stable than the latter. On the other hand thermodynamic stabilities obtained from theoretical
calculations based on isodesmic reactions of the open form cations, and NICS indicate that two of these ions (2 and 4)
are less stable than 9-fluorenyl cation, whereas the ‘‘side’’ isomer 3 is slightly more stable. Copyright ß 2010 John
Wiley & Sons, Ltd.
Keywords: antiaromaticity; fluorenyl cations; laser flash photolysis; fluorenol photochemistry
INTRODUCTION
The concept of antiaromaticity to describe the destabilization
of cyclic p-systems has been in existence for more than four
decades,[1,2] and has been extended to account for the
properties of isoelectronic ionic and radical intermediates.[3]
The 9-fluorenyl cation 1 is a member of the general class
of cyclic 4N p-systems exhibiting destabilizing properties.
Whereas these intermediates have been postulated in solvolysis
reactions, their direct observation by classical low temperature
NMR detection from alcohol precursors in superacids has never
been realized. The direct observation of the 9-fluorenyl cation as
a short-lived transient (t < 20 ꢀ 10ꢁ12 s in methanol) by laser
flash excitation of 9-fluorenol in neutral solutions was first
reported in 1989.[4] The lifetime of this species can be extended
These regioisomeric cations are denoted as the up-2, side-3,
in polar non-nucleophilic solvents such as 1,1,1-trifluoroethanol
and the down-4 species. Thiophene exhibits enhanced nucleo-
or 1,1,1,3,3,3-hexafluoroisopropanol.[5] The pKRþ value for
philicity compared to benzene. This is reflected in the aromatic
9-fluorenyl cation is 4.2 less than the benzhydryl cation
stabilization energies (22.4 kcal/mol for thiophene vs. 26.7 kcal/
indicating the destabilization of the former relative to the
mol for benzene)[9] and the proton affinities (196.5 kcal/mol for
open form.[6] This factor is also borne out by calculated
thiophene vs. 181.3 kcal/mol for benzene).[10] Thus replacement
destabilization energies of 8–10 kcal/mol relative to
of one or both of the phenyl rings in the 9-fluorenyl cation with a
benzhydryl cation for isodesmic reactions involving hydride
or hydroxide ion transfer.[6,7,8] This is equivalent to a rate
retardation of a factor of 106 to 107 at 25 8C for 9-fluorenyl cation
formation in solvolysis reactions compared to the benzhydryl
system.
*
Correspondence to: E. Lee-Ruff, Department of Chemistry, York University,
Toronto, Ontario, Canada, M3J 1P3.
E-mail: leeruff@yorku.ca
The p-isoelectronic structure of thiophene with benzene
suggests similar cations could be generated from their
alcohol precursors upon replacing one or both benzene rings
of 9-fluorenyl cation. However the lower symmetry of
this heterocycle leads to formation of three possible
regioisomers upon replacement of a single benzene with a
thiophene-ring.
a
M. Bancerz, G. Mladenova, K. Najafian, X. Zeng, E. Lee-Ruff, K. Najafian,
X. Zeng, E. Lee-Ruff
Department of Chemistry, York University, Toronto, Ontario, Canada, M3J
1P3, Canada
b
L. A. Huck, W. J. Leigh
Department of Chemistry, McMaster University, Hamilton, Ontario, Canada
L8S 4K1, Canada
J. Phys. Org. Chem. 2010, 23 1202–1213
Copyright ß 2010 John Wiley & Sons, Ltd.

PHOTOCHEMICAL GENERATION OF THIOPHENE ANALOGS OF 9-FLUORENYL CATIONS
thiophenyl ring might be expected to lead to enhanced
thermodynamic stability of the 12p annulene system. Conversely,
the presence of sulfur in the thiophene derivatives would be
expected to confer a decreased reactivity at the nucleofugal
carbon of ions 2–4 relative to fluorenyl cation 1 as a result of the
enhanced nucleophilicity of thiophene relative to benzene. Thus,
it was of interest to compare relative thermodynamic and kinetic
stabilities between these cations. Given that the fluorenyl cation
can be generated from electronic excitation of 9-fluorenol, the
likely precursors necessary for the generation of cations 2–4
would be the corresponding indenothiophenols 5–7.
singlet at d 1.8 ppm for the alcohol proton. Polyphosphoric acid
(PPA) was used to cyclize the central 5-membered ring, forming
8H-indeno[2,1-b]thiophene (10) in 69% yield. The product was
identified by 1H-NMR spectroscopy. The 13C-NMR spectrum
shows the expected ten aromatic carbon signals with 4 weak
signals associated with the quaternary carbons.
The preparation of up-5 involved a two-step reaction using
NBS bromination of 10 in carbon tetrachloride followed by
hydrolysis to the alcohol with aqueous silver nitrate. The 1H-NMR
spectrum of up-5 was consistent with the structure showing a 6:1
integration of the aromatic to aliphatic protons signals. The C-8
proton appears at d 5.4 ppm with the alcohol peak at 2.3 ppm,
both appearing as doublets. The 13C-NMR spectrum shows the
expected ten aromatic signals at d 149.5, 146.8, 146.7, 137.5,
131.5, 131.0, 128.9, 125.8, 124.8, 119.3, 118.7 ppm and one
aliphatic carbon signal at d 72.6 ppm. The mass spectrum shows
the molecular ion at m/z ¼ 188 with a fragment ion correspond-
ing to loss of hydroxide at m/z ¼ 171.
Preparation of down-7 followed the same protocol as used for
the synthesis of 5, with the use of2-thiophene boronic acid inplace
of 3-thiophene boronic acid in the Suzuki coupling (Scheme 2).
Reaction conditions: a. Pd(PPh3)4/C6H6 (66%); b. LiAlH4/THF
(75%); c.PPA, 120 8C (17%); d.NBS/CCl4, hy; e. AgNO3/acetone/H2O
(39%).
1
RESULTS
Ester 11 was obtained in 66% yield. The H-NMR spectrum is
consistent with the structure showing an integration of 7:3 of the
aromatic to the aliphatic methyl proton signals (d 3.7 ppm) of the
ester. Reduction of 11 with lithium aluminum hydride in THF
forms the corresponding alcohol 12 in 75% yield. Dehydrative
cyclization of 12 was carried out by the addition of a CH2Cl2
solution of 12 to neat pre-heated PPA (120 8C). The reaction
proceeded in significantly lower yield (17%) as compared to that
of 9, due presumably to the lower nucleophilicity at C-3
compared to C-2 of the thiophene ring. The remainder of the
isolated organic material consisted of some unreacted material
Synthesis of the fluorenol precursors
The preparation of the ‘‘up-‘‘ isomer 5 followed the synthetic
sequence outlined in Scheme 1.
Reaction conditions: a. Pd(PPh3)4/C6H6 (98%); b. LiAlH4/THF
(42%); c.PPA, 120 8C (69%); d.NBS/CCl4, hy; e. AgNO3/acetone/H2O
(44%).
The synthesis of methyl 2-(3-thienyl)benzoate (8) was
accomplished using a Suzuki coupling reaction between methyl
2-iodobenzoate (1.5 eq.) and thiophene-2-boronic acid.[11] The
yield for this reaction was quantitative as evident from the NMR
spectrum of the crude reaction mixture. Alcohol 9 was obtained
from LAH reduction of ester 8 in 42% yield. The proton NMR
spectrum of alcohol 9 showed peaks in the aromatic region that
was shifted upfield relative to the starting material, along with a
2H singlet at d 4.7 ppm for the methylene hydrogens and 1H
1
along with unidentified side products. The H-NMR spectrum is
consistent with the structure. The same cyclization was
attempted on the methyl ester of this compound without
success.
The last step of the synthesis of the down alcohol 7 was
performed in the same manner as with 5. However in this case
O
O
S
a
b
OCH3
OH
OCH3
S
+
I
HO B
S
OH
9
8
c
d,e
S
5
10
Scheme 1. Synthetic pathway for the preparation of 8H-indeno[2,1-b]thiophen-8-ol (5).
J. Phys. Org. Chem. 2010, 23 1202–1213
Copyright ß 2010 John Wiley & Sons, Ltd.
View this article online at wileyonlinelibrary.com

M. BANCERZ ET AL.
O
I
O
OH
HO B
S
a
b
OCH3
OH
OCH3
S
+
S
12
11
c
d,e
7
S
13
Scheme 2. Synthetic scheme for the preparation of 4H-indeno[1,2-b]thiophen-4-ol (down-7).
roughly half of the product obtained after treatment of the
bromide with silver nitrate was the target alcohol and the other
half was the corresponding ketone. The oxidation likely occurred
due to the use of excess silver nitrate. Nevertheless, the ketone
could be easily reduced with sodium borohydride in methanol to
give additional amounts of alcohol 7. The overall yield was 39%.
The 1H-NMR spectrum of 7 is consistent with its proposed
structure, exhibiting aromatic signals integrating to six protons,
with the C-4 proton appearing at d 5.4 ppm and the alcohol
proton at d 3.85 ppm. The 13C-NMR spectrum shows the expected
ten aromatic carbon signals at d 150.6, 149.0, 143.6, 136.8, 128.8,
127.8, 126.0, 124.8, 122.7, 118.8 ppm as well as the C-4 aliphatic
carbon signal at d 72.1 ppm. The mass spectrum shows the
molecular ion at m/z ¼ 188 with loss of hydroxide at m/z ¼ 171.
The initial Suzuki coupling was done using a 1:1 stoichiometric
molar ratio in order to maximize the yield of the mono-coupled
product. The product yield was 68%. The 1H-NMR spectrum
was consistent with the structure. Carboxylation of
3-bromo-4-phenylthiophene was carried out by lithiation with
n-butyl lithium followed by addition of gaseous CO2. The yield
was reasonably high (87%), and 14 was obtained in sufficient
purity that column chromatography was not necessary. The
integration ratios in the 1H-NMR spectrum were consistent with
its molecular structure. The carboxylic acid proton appeared
unusually upfield at d 5.2. Direct cyclization of acid 14 could be
effected in hot (200 8C) PPA. The 1H-NMR spectrum is consistent
with ketone 15, and coupling constants were used to distinguish
between protons on the thiophene ring and those on the phenyl
ring. Long range coupling of about 2 Hz was observed for the
protons on the thiophene ring. The reduction step to alcohol 6
was effected with sodium borohydride. The yield was near
quantitative (96%) without the need for purification by
chromatography. The 1H-NMR spectrum of 6 was consistent
with its structure. The same thiophene long range coupling
constants were observed as with 8H-indeno[1,2-c]thiophene.
The C-8 proton appears at d 5.5 ppm with the alcohol proton
A
slightly different approach was undertaken for the
preparation of the middle alcohol precursor 6. Using
3,4-dibromothiophene, an attempt was made to introduce the
electrophilic moiety for the cyclization reaction on the thiophene
ring and then carry out the Suzuki coupling to a phenyl ring. This
proved inefficient. The preferred pathway involved initial Suzuki
coupling to phenyl boronic acid followed by carboxylation
(Scheme 3).
S
Pd(PPh3)4
C6H6
1. n-BuLi
Br
+
CO2H
OH
B
2. CO2
3. H+
Br
Br
OH
S
S
14
PPA/200 oC
O
NaBH4
6
S
MeOH
15
Scheme 3. Synthesis of 8H-indeno[1,2-c]thiophen-8-ol (6).
View this article online at wileyonlinelibrary.com
Copyright ß 2010 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2010, 23 1202–1213

PHOTOCHEMICAL GENERATION OF THIOPHENE ANALOGS OF 9-FLUORENYL CATIONS
OH
OCH3
O
hν/MeOH
hν/MeOH
hν/MeOH
S
S
S
S
5
16
10
19
13
17
OH
OCH3
O
S
S
S
S
6
18
15
OH
OCH3
S
O
S
S
S
7
20
21
Scheme 4. Photoproducts from irradiation of alcohols 5, 6, and 7.
appearing at d 2.2 ppm both as doublets (J ¼ 9 Hz). The 13C-NMR
spectrum showed the expected ten aromatic carbon signals with
the benzylic C-8 carbon at d 70.3 ppm. The mass spectrum is
similar to the previous two isomers showing a molecular ion at
m/z ¼ 188 and a peak at m/z ¼ 171 corresponding to loss of
hydroxide.
lists the isolated yields of the major identified products from these
experiments.
Laser flash photolysis
Laser flash photolysis experiments were carried out using a KrF
excimer laser for excitation (248 nm, ꢂ25 ns, ꢂ100 mJ) and
deoxygenated 0.15–0.20 mM solutions of alcohols 5, 6, and 7 in
HFIP. Transient spectra were recorded using a flow system, while
rate constants for methanol (MeOH) quenching were measured
using a series of static samples containing different concen-
trations of the added substrate. Experiments with 9-fluorenol in
HFIP were carried out using a series of static samples.
Photoproduct studies
Solutions of alcohols 5, 6 and 7 in methanol (8 ꢀ 10ꢁ3 M) were
irradiated for a common time period and the products were then
isolated. The intermediacy of the carbocations was established by
the formation of the methyl ether photosolvolysis products. Other
products were formed from corresponding radical intermediates.
These included the corresponding fluorene analogs 10, 19, and 13,
respectively, from hydrogen atom abstraction, and the fluorenone
analogs17, 15, and21fromradicaldisproportionationoroxidation
pathways (Scheme 4).[12] The structural assignments of all photo-
products were based on comparison with samples prepared
independently. The methanolysis products 16, 18, and 20 were
prepared from methanol solvolysis of the corresponding benzylic
bromides prepared from fluorenes 10, 19, and 13, respectively.
Small amounts (<5%) of the bifluorenyl analogs were observed
from alcohols 5 and 6 which were not unequivocally characterized.
Alcohol 7 gave the smallest amount of photomethanolysis product,
with radical-derived products being the most predominant. Table 1
8H-Indeno[2,1-b]thiophen-8-ol (5)
Laser photolysis of 5 gave rise to two detectable transient species
with overlapping absorption spectra (see Fig. 1a). The dominant
absorption exhibited lmax ꢂ 500 nm and lifetime t ꢂ 20 ms, while
the weaker one was much longer lived (t ¼ 385–740 ms) and
exhibited lmax ꢂ 470 nm. Addition of MeOH to the solution
caused a shortening of the lifetime of the 500 nm species and a
reduction in the intensity of the long-lived (470 nm) component
of the decay; the latter was eliminated completely in the pre-
sence of 1.94 M MeOH. This behavior suggests that the 470 nm
species is a product of reaction of the 500 nm species, its
formation being suppressed in the presence of a substrate (i.e.
Table 1. Photoproduct distribution for thiophene fluorenol analogs 5–7a (1 h irradiation)
Alcohol
Methanolysis products (%)
Fluorene analog (%)
Ketone (%)
Unreacted alcohol (%)
5
6
7
16(13)
18(30)
20(2)
10(14)
19(35)
13(27)
17(28)
15(12)
21(69)
(35)
(17)
(1)
1
a Yields are based on H-NMR integrations with pure reference standards.
J. Phys. Org. Chem. 2010, 23 1202–1213
Copyright ß 2010 John Wiley & Sons, Ltd.
View this article online at wileyonlinelibrary.com

M. BANCERZ ET AL.
Figure 1. (a) Time-resolved UV–Vis absorption spectra recorded by laser flash photolysis of a deoxygenated 0.2 mM solution of 5 in HFIP, 0.32–0.96 ms
(*), 27.5–28.8 ms (&), and 155.5–156.8 ms (D) after excitation. (b) Plot of the pseudo-first order rate constant for decay of the 500 nm species vs. [MeOH];
the solid line is the fit of the data to Eqn. 1.
MeOH) that reacts competitively with its direct precursor. The
short-lived species is assigned to carbocation 2; the identity of
the longer-lived transient is more difficult to assign, but may be a
dimeric cationic species derived from electrophilic attack of the
primary formed cation with 5. Such bimolecular reactions have
been reported for the 9-fluorenyl cation 1.[13,14]
A plot of the pseudo-first order rate constant for decay of the
500 nm species (kdecay) vs. [MeOH] (Fig. 1b) exhibited positive
curvature, indicating that the species reacts with the alcohol via a
mechanism in which the order with respect to MeOH is higher
than one. The data were thus fit to the quadratic expression of
Eqn. 1, where k0 is the pseudo-first order decay rate constant in
the absence of added substrate, and k1 and k2 are overall second-
and third-order rate coefficients, respectively, for reaction of the
species with MeOH; a reasonable fit of the data required that k0
and k1 be constrained to positive values. The analysis afforded k0
and k1 values equal to zero within the estimated error limits, and a
k2 value of k2 ¼ (4.8 ꢃ 1.9) ꢀ 105 Mꢁ2 sꢁ1, where the error is
expressed as twice the standard deviation from the non-linear
least squares fit.
2
kdecay ¼ k0 þ k1½MeOHꢄ þ k2½MeOHꢄ
8H-Indeno[1,2-c]thiophen-8-ol (6)
(1)
Two transients with overlapping absorption spectra were also
observed for this isomer (Fig. 2a): a short-lived species exhibiting
lmax ꢂ 500 nm and lifetime t ꢂ 15 ms, which is formed with the
laser pulse and is assigned to carbocation 3, and a longer-lived
transient with lmax ꢂ 460 nm and t ꢂ 140 ms. The longer-lived
transient absorption (or at least part of it) shows a discrete
growth that occurs over a similar timescale as the decay of the
Figure 2. (a) Time-resolved UV/Vis absorption spectra recorded by laser flash photolysis of a deoxygenated 0.15 mM solution of 6 in HFIP, 0.32–0.64 ms
(^), 30.7–31.7 ms (*), and 107.5–108.5 ms (&) after excitation. (b) Plot of the pseudo-first order rate constant for decay of the 500 nm species (monitored
at 520 nm) vs. [MeOH]; the solid line is the fit of the data to Eqn. 1.
View this article online at wileyonlinelibrary.com
Copyright ß 2010 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2010, 23 1202–1213

PHOTOCHEMICAL GENERATION OF THIOPHENE ANALOGS OF 9-FLUORENYL CATIONS
first-formed transient, suggesting that it is a product of the latter
species. However, the absorption was not completely eliminated
upon addition of up to 2 M MeOH; a trace recorded at 460 nm in
the presence of 2 M MeOH was roughly 30% as intense as in the
absence of the alcohol, and exhibited t ¼ 20 ms. This suggests
that the longer-lived transient absorption consists of contri-
butions from at least two species, one whose formation is
quenched by added MeOH, and one that is most likely formed
along with 3 during the laser pulse and whose lifetime is relatively
insensitive to the presence of added MeOH.
A plot of kdecay vs. [MeOH] for the short-lived 500 nm species
again exhibited positive curvature (see Fig. 2b), but the fit to Eqn. 1
was essentially quantitative and afforded a k1 value equal to zero,
consistent with a pure second-order dependence on MeOH
concentration. The analysis afforded k2 ¼ (4.6 ꢃ 0.5) ꢀ 105 Mꢁ2 sꢁ1
where the error is again expressed as ꢃ2s.
,
4H-Indeno[1,2-b]thiophen-4-ol (7)
Laser photolysis of alcohol 7 resulted in the formation of two
transients with well-separated absorption spectra and markedly
different temporal behaviors (Fig. 3a). The shorter-lived absorp-
tion, tentatively assigned to carbocation 4, was formed with the
laser pulse and exhibits lmax ¼ 480 nm (t ¼ 8 ms). The second
species (lmax ¼ 360 nm) grew in after the laser pulse over a similar
timescale as the decay of the 480 nm absorption, again
suggesting it is a product of the first-formed transient. Addition
of MeOH led to discrete reductions in the lifetime of the 480 nm
species, and a plot of kdecay vs. [MeOH] again showed positive
curvature. The fit to Eqn. 1 was relatively poor, but afforded
k2 ¼ (4.5 ꢃ 3.5) ꢀ 105 Mꢁ2 sꢁ1 and k0 and k1 values indistinguish-
able from zero. The effect of added MeOH on the 360 nm
transient absorption was not determined.
Figure 4. Plot of the pseudo-first order rate constant for decay of the
9-fluorenyl cation in deoxygenated HFIP at 25 8C vs. [MeOH]. The solid line
is the non-linear least squares fit of the data to Eqn. 1.
with good pseudo-first order kinetics and exhibited t ꢂ 10 ms. This
is in good agreement with the results of McClelland et al.,[5] who
assigned the species obtained under these conditions to the
9-fluorenyl cation (lmax ¼ 515 nm). Addition of MeOH resulted in a
shortening of the transient lifetime, and a plot of kdecay vs. [MeOH]
exhibited positive curvature over the 0–0.3 M concentration range
in added MeOH (see Fig. 4). Analysis of the data according to Eqn. 1
afforded k0 and k1 values indistinguishable from zero within
experimental error, and k2 ¼ (4.8ꢃ 3.6) ꢀ 107 Mꢁ2 sꢁ1
.
Figure 5 shows plots of kdecay vs. [MeOH]2 for all four of the
carbocations studied in this work, along with the results of linear
least squares analysis of the data according to Eqn. 2. As can be
seen in Fig. 5a, the plots for 2–4 all show very good linearity
9-Fluorenol
Laser flash photolysis of static solutions of 9-fluorenol in HFIP
afforded strong transient absorptions at 500 nm, which decayed
Figure 3. (a) Time-resolved UV/Vis absorption spectra recorded by laser flash photolysis of a deoxygenated 0.17 mM solution of 7 in HFIP, 0.16–0.64 ms
^ and 37.8–38.2 ms (&) after excitation. (b) Plot of the pseudo-first order rate constant for decay of the 500 nm species (monitored at 520 nm) vs. [MeOH];
the solid line is the non-linear least squares fit of the data to Eqn. 1.
J. Phys. Org. Chem. 2010, 23 1202–1213
Copyright ß 2010 John Wiley & Sons, Ltd.
View this article online at wileyonlinelibrary.com

M. BANCERZ ET AL.
Figure 5. Plots of the pseudo-first order rate constants for decay of (a) cations 2–4 and (b) the 9-fluorenyl cation (1) in deoxygenated HFIP at 25 8C vs.
[MeOH]2. The solid lines are the linear least squares fits of the data for [MeOH] ꢅ 2.0 M to Eqn. 2.
over the 0–2.0 M concentration range in added MeOH. The
corresponding k2 values obtained from least squares analysis of
the data over this concentration range are listed in Table 2
the rings and clusters. NICS are now widely employed to
characterize the aromaticity and antiaromaticity of two- as well as
three-dimensional species and transition states.[16–20] Negative
NICS values (given in ppm) imply aromaticity (diatropic ring
currents), and positive NICS values correspond to antiaromaticity
(paratropic ring currents). In contrast to other aromaticity criteria,
NICS does not require reference molecules, increment schemes,
or equations for evaluation.[15] Thus the degree of antiaromaticity
of the four cations can also be assessed from the NICS data
(Table 3) which suggest that cations 2 and 4 are more
antiaromatic than fluorenyl (1) from the perspective of the
middle ring, whereas ion 3 exhibits less antiaromatic character
compared to 1.
2
kdecay ¼ k0 þ k2½MeOHꢄ
(2)
Theoretical calculations of cation stabilities
In order to assess the relative stabilities of cations 1–4, energies for
the isodesmic reactions for hydride transfer between the fluorene
analogs and the corresponding benzhydryl analogs (Scheme 5)
**
were computed at the B3LYP/6-311G level of theory.
The computed energy for the hydride transfer between
fluorene and benzhydryl cation agrees well with previous
computations at lower levels of theory.[8] It can be seen that
thiophene cations 4 and 2 are thermodynamically less stable
than 1 relative to the open form with respect to the hydride
transfer isodesmic reactions depicted in Scheme 3. On the other
hand, cation 3 is calculated to be more stable relative to the
corresponding acyclic cation than 9-fluorenyl. Between the
isomeric cations 2–4, side- 3 is more stable than up-2 and down-3
by 1.6 and 5.2 kcal/mol, respectively.
DISCUSSION
As expected based on the extensive studies that have been
reported on the photochemistry of 9-fluorenol under various
conditions,[4,5,12,21] direct irradiation of the thiophene fluorenol
analogs 5, 6, and 7 in pure MeOH solution involves both
heterolysis and homolysis of the C—OH bond, to produce
photoproducts derived from the corresponding carbocation (i.e.
the methyl ethers 16, 18, and 20) and radical intermediates
(i.e. hydrocarbons 10, 13, and 19 and ketones 15, 17, and 21),
respectively. The ketones are produced from aerobic oxidation of
the corresponding fluorenyl radicals by way of their hydroper-
oxide intermediates. The parent 9-fluorenyl hydroperoxide is
known to undergo decomposition to fluorenone and fluorenol
with elimination of molecular oxygen.[22] It was found that
purging the solutions with argon prior to irradiation suppressed
the production of ketones 15, 17, and 21 with concomitant yield
increase of the reduced products 10, 13, and 19. The yields of the
methyl ethers derived from 5 and 6 are higher than that from 7,
which appears to produce mainly radical-derived photoproducts.
However, care must be exercised in the interpretation of the
product yields reported in Table 1, as the methyl ethers can
undergo secondary photolysis to yield the same radical-derived
products as the alcohols,[12] and no attempt was made to
determine the true primary photoproduct yields in either MeOH
or MeOH/HFIP mixtures. The distribution of homolysis and
Nucleus-independent chemical shifts (NICS), introduced in
1996, provide a simple and efficient magnetic criterion of
aromaticity.[15] NICS is based on the negative absolute magnetic
shielding computed at the cluster centers or above the centers of
Table 2. Lifetimes and second order methanol quenching
rate constants for 9-fluorenyl cation (1) and the thiophene
fluorenyl cation analogs 2, 3, and 4 in deoxygenated HFIP
solution at 25 8C.
Cation
Lifetime (ms)
kMeOH M
ꢁ2 sꢁ1
9-Fluorenyl cation (1)
10
19
15
8
(7 ꢃ 1) ꢀ 107
(4.0 ꢃ 0.2) ꢀ 105
(4.6 ꢃ 0.2) ꢀ 105
(3.4 ꢃ 0.4) ꢀ 105
2
3
4
View this article online at wileyonlinelibrary.com
Copyright ß 2010 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2010, 23 1202–1213

PHOTOCHEMICAL GENERATION OF THIOPHENE ANALOGS OF 9-FLUORENYL CATIONS
Scheme 5. Computed relative energies for thiophene cations 2, 3, 4, and fluorenyl cation 1.
heterolysis products from photolysis of 9-fluorenol is known to
depend critically on solvent polarity and acidity, with the relative
yield of the cation increasing with increasing ionizing ability and
acidity;[21] cation 1 is formed essentially exclusively on photolysis
of 9-fluorenol in the strongly acidic, non-nucleophilic solvent
HFIP.[12,21] Thus, the fact that the methyl ethers 16 18, and 20
are observed in the photolyses of 4–6 in MeOH, even after
extended irradiation, provides reasonably strong support for the
expectation that the corresponding cations (2–4) should be
the major primary products upon photolysis of the compounds
in HFIP. Indeed, like 9-fluorenol itself,[12] the yield of the
methanolysis product 18 increased from 30 to 48%, at the
expense of the radical-derived products 19 and 15, upon
irradiation of 6 in 3:1 MeOH-H2O under similar conditions to
those employed for the irradiations in pure MeOH.
times (irradiation for 30 min gave 11% of 20 (vida supra Table 2))
or by the addition of HFIP (20% of 20 in a 25% v/v of HFIP/
methanol for 15 min UV exposure time). However, the limited
solubility of alcohol 7 in HFIP, as well as the lack of an authentic
HFIP solvolysis product precluded preparative photoproduct
studies in this solvent. Such solvent dependent behavior on the
relative extent of heterolytic and homolytic C—O bond scission in
9-fluorenol has been reported from differences in lifetimes and
decay features measured for the corresponding cation and
radical transients.[21]
Our assignment of the promptly formed transients observed in
the laser flash photolysis experiments with 5–7 to the correspond-
ing carbocations rests mainly on the similarity of their absorption
spectra to that of 1 and the fact that they are quenched by MeOH,
albeit 20–30 times less rapidly than 1 under similar conditions. We
were initially concerned that these reactions appear to proceed via
overall third-order kinetics (first order in cation, second order in
alcohol) in all four cases, because of the report by McClelland et al
of a second-order rate constant for reaction of 1 with water in HFIP
solution (kH2O ꢂ 1.5 ꢀ 106 Mꢁ1 sꢁ1).[5] However, this value was
derived from data recorded over the 0–0.5 M concentration
range in added water, where the variation in the cation decay
rate constant with [H2O] was ‘‘approximately linear’’. In fact a plot of
the reported kdecay values vs. [H2O] for this system over the
complete concentration range studied reveals an almost
perfect squared dependence of the rate on concentration, similar
to what we have found in the present work for quenching of 1–4
by MeOH; a plot of kdecay versus [H2O]2 yields a slope of
kH2O ¼ (9.41 ꢃ 0.07)ꢀ 105 Mꢁ2 sꢁ1. This is roughly 20 times lower
than the value measured here for MeOH quenching under similar
conditions, and is line with the known relative reactivities of water
and methanol solvents toward carbocations.[5,23,24]
This was also found to be the case for the ethers 16, 18, and 20
which underwent secondary photodecomposition to radical
derived products. The relatively low yield of 20 from irradiation of
alcohol 7 in methanol was enhanced using shorter irradiation
Table 3. Calculated NICS 1 A8 above the five- and six-
membered ring centers (denoted as NICS (1) ppm) for fol-
**
lowing compounds at B3LYP/6-311 þ G levels
NICS(0)total/NICS(1)total
Cation
6MR
5MR
5MR(S)
1
2
3
4
þ10.1/4.90
þ10.10/5.37
þ4.07/0.10
þ27.0/18.4
þ30.22/21.44
þ17.79/10.99
þ36.47/ þ 26.33
þ3.27/2.85
þ2.80/4.39
þ3.63/ þ 4.3
While McClelland et al suggested that the non-linear
quenching behavior might be due to solvation effects,[5] we
þ13.45/ þ 8.27
J. Phys. Org. Chem. 2010, 23 1202–1213
Copyright ß 2010 John Wiley & Sons, Ltd.
View this article online at wileyonlinelibrary.com

M. BANCERZ ET AL.
Scheme 6. Mechanism for methanol quenching of cations 2–4.
suggest that a more reasonable explanation is provided by the
mechanism of Scheme 6, in which initial nucleophilic attack is
followed by general base catalyzed deprotonation of the
first-formed intermediate,[25–27] with a second molecule of
alcohol acting as the general base.
of hydride transfer to the open form cation equivalents. Based on
this scheme, cations 2 and 4 are less stable than 9-fluorenyl cation
(1), whereas 3 is more stable than 1. These data are corroborated
by the calculations of the NICS values for the middle ring which
show that 2 and 4 are more antiaromatic than 9-fluorenyl cation
whereas 3 is less antiaromatic than 1. The kinetic reactivity of the
three regioisomeric cations are roughly two orders of magnitude
less than 9-fluorenyl cation, and is likely associated with the lower
positive charge density of the nucleofugal carbon as a result of
charge delocalization to the sulfur.
Computed charge densities for 9-fluorenyl cation 1 shows a
þ0.145 positive charge at C-9 with the remaining charge
delocalized about the other atoms. However, for ions 2–4
significant positive charge resides on sulfur (þ0.59 for 2, þ0.54 for
3, and þ0.63 for 4) with much less charge on the nucleofugal
carbon. This is broadly consistent with the observation that
cations 2–4 are all 20–30 times less reactive toward MeOH than
the parent cation, 1; the decrease in positive charge density at the
nucleofugal carbon results in an increase in the kinetic barrier for
nucleophilic attack. It should be noted that while the differences
in positive charge density at sulfur in 2–4 are relatively small, they
appear to correlate (inversely) with the rate constants for reaction
of the three cations with MeOH, with the smallest rate constant
being obtained for cation 4, in which positive charge density at
sulfur is highest (and that on the nucleofugal carbon is lowest).
The antiaromatic character of the fluorenyl cation has been
convincingly established by theoretical and experimental
studies.[28,29] The recent theoretical studies by Herndon and
Mills[28] indicate an average destabilization energy for cation 1 of
16.3 ꢃ 1.6 kcal/mol with specific values dependent on the
reference system which corrects for strain, non-aromatic
stabilization and other structural effects.[28] Our value of
9.7 kcal/mol for the destabilization energy of cation 1 falls
somewhat lower due to the different reference system chosen in
our study for comparison with the corresponding benzhydryl
analogs which is the common standard for computing aromatic
stabilization energies of fluorenyl ions. While it may be argued
that the differences of computed relative stabilization energies
for ions 2, 3, and 4 fall within the deviation of 1.6 kcal/mol[28]
depending on the choice of the reference system and thus not
significant, the independent NICS values determined for these
ions corroborate the trend for the destabilization energies.
EXPERIMENTAL
All reactions were done in dried glassware. All solvents used for
the reactions were dried and distilled. THF and diethyl ether were
dried using sodium metal. 1,1,1,3,3,3-Hexafluoro-2-propanol
(Sigma–Aldrich) was purified by distillation over sodium
borohydride. Proton NMR spectra were obtained on a Bruker
ARX-400 spectrometer and 13C NMR was carried out on a Bruker
300 spectrometer. Solutions in 99.8% CDCl3 with 1% TMS as an
internal standard were used. Methanol-d4 and DMSO-d6 were
used for certain samples for solubility reasons, and are noted
where appropriate. Photolyses were carried out using a Hanovia
450W medium pressure mercury arc lamp. Melting points were
determined on a Reichert melting point apparatus and are
uncorrected. Mass Spectra were collected (EI mode) using a
Varian Series 4000 GC/MS/MS spectrometer.
Methyl 2-(3-thienyl)benzoate (8)11
In a 250 mL round-bottom flask with a magnetic stirrer, was
added methyl 2-iodobenzoate (5.3 mL) benzene (120 mL). 2 M
Sodium carbonate (36 mL) was added to the benzene solution.
3-Thiophene boronic acid (3.07 g) dissolved in ethanol (22 mL)
was added to this solution. The mixture was purged with argon.
After 30 min, 270 mg of tetrakis(triphenylphosphine)palladium(0)
was introduced. Under argon atmosphere, the mixture was
stirred and refluxed overnight. The flask and condenser were
wrapped in aluminum foil to protect the reaction from light. After
the reaction was complete, the mixture was poured into a
separatory funnel. The organic layer was extracted with
dichloromethane (100 mL) and washed with saturated sodium
chloride solution (50 mL) then washed with water (3 ꢀ 100 mL).
Methyl 2-(3-thienyl)benzoate was purified by column chroma-
tography (ethyl acetate:hexane 5:95 by volume). An oily material
was obtained (5.54 g, quantitative). 1H NMR d: 7.82 (d, J ¼ 7.0 Hz,
1H), 7.52 (m, 1H), 7.4 (m, 4H), 7.28 (m, 1H), 3.75 (s, 3H). These data
are inconsistent with those reported.[30] The literature describes
this compound as a solid (m.p. 151–152 8C) and showing a
molecular ion mass of m/z ¼ 192 which is inconsistent with the
molecular mass for this compound. The correct assignment for
the structure of this ester was confirmed by hydride reduction to
the known alcohol 9 (see below).
SUMMARY AND CONCLUSIONS
Precursor alcohols for the photogeneration of the three
regioisomeric thiophene cation equivalents of 9-fluorenylcation
were prepared. These alcohols on photoexcitaion generated the
corresponding cations which were quenched in methanol as
evident from the formation of the methanolysis products. In
addition, products derived from the corresponding radicals were
evident from the presence of reduced (indenothiophenes) as well
as oxidized (indenothiophenones) materials in the photomixture
which distribution was dependent on irradiation times and
solvent composition. Time resolved photoexcitation permitted
the direct observation of the three regioisomeric cations which
were quenched by methanol following second order kinetics. The
relative thermodynamic stabilities of the fluorenyl cation analogs
were assessed using MO calculations for the isodesmic reactions
View this article online at wileyonlinelibrary.com
Copyright ß 2010 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2010, 23 1202–1213

PHOTOCHEMICAL GENERATION OF THIOPHENE ANALOGS OF 9-FLUORENYL CATIONS
[2-(3-Thienyl)phenyl]methanol (9)
mixture was poured into a separatory funnel. The organic layer
was extracted with diethyl ether (150 mL) and washed with
saturated sodium chloride solution (50 mL) then washed with
water (3 ꢀ 100 mL). Methyl 2-(2-thienyl)benzoate was purified by
column chromatography (ethyl acetate:hexane 5:95 by volume).
An oily product (2.24 g, 66%) was obtained. 1H NMR d: 7.7 (d,
J ¼ 7.5 Hz, 1H), 7.45 (m, 2H), 7.3 (m, 2H), 7.0 (m, 2H), 3.7 (s, 3H). The
NMR data are consistent with those reported in the literature.[33]
Methyl 2-(3-thienyl)benzoate (5.54 g) was dissolved in dry THF
(150 mL). Lithium aluminum hydride (1.0 g) was added and the
solution was left to stir for 1 h. Water (10 mL) was then added to
quench the reaction. The THF was partially evaporated and the
product was acidified with 1 M HCl (2 mL), extracted with
dichloromethane (100 mL) and washed with water (100 mL). Yield
2.03 g of oily material (42%) which was used for the next step
1
without further purification. H NMR d: 7 ꢆ 57 (d, 1 H, J ¼ 7.1 Hz),
[2-(2-Thienyl)phenyl]methanol (12)
7.48 (d, 1 H, J ¼ 7.0 Hz), 7.43–7.34 (m, 3 H), 7.20 (d, 1 H, J ¼ 3.0 Hz),
7.13 (t, 1 H), 4.78 (s, 2 H), 1.80 (s, br, 1 H); 13C NMR d 141.5, 138.6,
133.7, 130.9, 128.9, 128.3, 127.9, 127.5, 127.0, 125.8, 63.5, 4.7 (s,
2H), 1.8 (s, 1H, OH). These spectra were identical with those for an
authentic sample obtained commercially.
Methyl 2-(2-thienyl)benzoate (2.24 g) was dissolved in THF (50 mL).
A molar equivalent of lithium aluminum hydride (390 mg) was
added and the solution was left to stir for 40min at room
temperature. Another equivalent of lithium aluminum hydride
(390 mg) was added and the solution was left to stir for another
30 min. Water (10 mL) was then added to quench the reaction. The
solvent was partially evaporated and the product was extracted
with diethyl ether (100 mL) and washed with water (100 mL). 1.47 g
of [2-(2-thienyl)phenyl]methanol was obtained in 75% yield. M.p.
60–628C. 1H NMR d: 7 ꢆ 54 (m, 1 H), 7.3–7.5 (m, 5 H), 7.20 (d, 1 H,
8H-Indeno[2,1-b]thiophene (10)
PPA (50 mL) was heated to 120 8C with stirring. [2-(3-Thienyl)
phenyl]methanol (2.03 g) was added along with some dichlor-
omethane for quantitative transfer and heated for 2 min. This
mixture was then poured into ice water (300 mL) and extracted
with dichloromethane (3ꢀ 100 mL). The product was purified by
column chromatography twice (ethyl acetate:hexane 5:95 by
volume) yielding 1.27g (69%) of a white solid m.p. 65–67 8C
J ¼ 3.0 Hz), 7.24 (m, 1 H), 4.69 (d, 2 H, J ¼ 3.7 Hz), 1.81 (s, br, 1 H).; 13
C
NMR: d: 140.8, 138.1, 136.2, 130.1, 128.9, 127.9, 127.7, 125.5, 123.2,
63.5. This sample was identical in all respects with an authentic
commercial sample.
1
(lit.[31] m.p. 66–678C). H NMR d: 7.57 (d, J ¼ 7.5 Hz, 1H), 7.51 (d,
J ¼ 7.4 Hz, 1H), 7.35 (m, 3H), 7.21 (t, J ¼ 7.4 Hz, 1H), 3.85 (s, 2H).
4H-Indeno[1,2-b]thiophene (13)
8H-Indeno[2,1-b]thiophen-8-ol (5)
PPA (20 mL) was heated to 120 8C with stirring.
[2-(2-Thienyl)phenyl]methanol (1.47 g) was added with some
dichloromethane for quantitative transfer and heated for 4 min.
This mixture was then poured into ice water (50 mL) and extracted
with dichloromethane (3ꢀ 100 mL). The product was purified by
column chromatography (ethyl acetate:hexane 5:95 by volume).
4H-Indeno[1,2-b]thiophene (13) was obtained (229.4 mg, 17%) as a
8H-Indeno[2,1-b]thiophene (10) (1.27 g) was dissolved is carbon
tetrachloride (200 mL). One molar equivalent of NBS (1.31 g)
was added and the solution left to stir and reflux with a 100 W
lamp directed on the solution for 3 h. The solution was filtered
and washed with water (100 mL) in a separatory funnel, and then
the solvent was evaporated under vacuum. The crude
8-bromo-8H-indeno[2,1-b]thiophene was dissolved in a 50:50
solution of acetone and water (200 mL) containing a 1 mol eq. of
silver nitrate (1.25 g) and allowed to stir for 15 min. After the
reaction was complete the solution was filtered and extracted
with diethyl ether (100 mL) and washed with water (2 ꢀ 100 mL).
The organic extract was dried over anhydrous magnesium
sulfate. The crude product was purified by column chromatog-
raphy (ethyl acetate:hexane 5:95 by volume) and yielded
604.5 mg of 8H-indeno[2,1-b]thiophene-8-ol in 44% yield.
m.p.: 115–117 8C (lit.[32] m.p. 116 8C). IR: n ¼ 3292, 3194,
1
white solid, m.p. 67–688C (lit.[34] m.p. 68–698C). H NMR d: 7.52
(dxd, J ¼ 7.4 Hz, 7.6 Hz, 2H), 7.35 (m, 2H), 7.22 (t, J ¼ 7.4 Hz, 1H), 7.15
(d, J ¼ 4.7 Hz, 1H), 3.75 (s, 1H).
4H-Indeno[1,2-b]thiophen-4-ol (7) via
4-bromo-4H-indeno[1,2-b]thiophene
4H-Indeno[1,2-b]thiophene (229.4 mg) was dissolved is carbon
tetrachloride (125 mL). One molar equivalent of NBS (237 mg)
was added and the solution was left to stir and reflux with a
100 W lamp directed on the solution for 2.5 h. Following this the
solution was filtered and washed with water (100mL) in a
separatory funnel, and then the solvent was evaporated under
vacuum. The crude 4-bromo-4H-indeno[1,2-b]thiophene was
dissolved in a 50:50 solution of acetone and water (160 mL)
containing a molar equivalent of silver nitrate (225 mg) and
allowed to stir for 10 min. After the reaction was complete the
solution was filtered and extracted with diethyl ether (100mL),
washing with water (2 ꢀ 100 mL). The extract was dried with
anhydrous magnesium sulfate. The crude product was purified
by column chromatography (ethyl acetate:hexane 20:80 by
volume) and yielded 50 mg of 4H-indeno[1,2-b]thiophen-4-ol as
well as 52 mg of 4H-indeno[1,2-b]thiophen-4-one (21). For
4H-indeno[1,2-b]thiophen-4-ol: m.p.: 148–150 8C. IR: n ¼ 3369,
1026 cmꢁ1
;
1H NMR d: 7.56 (d, J ¼ 7.4 Hz, 1H,), 7.44 (d,
J ¼ 4.8 Hz,1H), 7.41 (d, J ¼ 7.35 Hz, 1H), 7.33 (t, J ¼ 7.35 Hz, 1H),
7.24 (t, J ¼ 7.4 Hz, 1H), 7.23 (d, J ¼ 4.9 Hz, 1H), 5.4 (br s, 1H), 2.3
(br s, 1H, OH). 13C NMR d: 149.5, 146.82, 146.75, 137.5, 131.0,
128.9, 125.8, 124.8, 119.3, 118.7, 72.7; MS: m/z (% intensity): 188
(M þ,100), 171 (42), 115 (41).
Methyl 2-(2-thienyl)benzoate (11)11
In a 250 mL round-bottom flask with a magnetic stirrer, methyl
2-iodobenzoate (3.44 mL) was added to benzene (80 mL). To this
solution was added 2 M sodium carbonate (24 mL), and
2-thiophene boronic acid (2.0 g) dissolved in ethanol (15 mL).
The mixture was purged with argon. After 20 min, 180 mg of
tetrakis(triphenylphosphine)palladium(0) was introduced into
the mixture. Under argon atmosphere, the mixture was stirred
and refluxed overnight. After the reaction was complete, the
1
3304, 1053 cmꢁ1; H NMR d: 7.53 (d, J ¼ 8.4 Hz, 1H), 7.3 (m, 3H),
7.2 m 1H), 7.18 (d, J ¼ 4.8 Hz, 1H), 5.4 (s, 1H), 3.85 (s, 1H, OH). 13C
NMR d: 150.3, 148.7, 143.9, 136.9, 129.0, 128.2, 126.2, 124.9, 122.7,
J. Phys. Org. Chem. 2010, 23 1202–1213
Copyright ß 2010 John Wiley & Sons, Ltd.
View this article online at wileyonlinelibrary.com

M. BANCERZ ET AL.
119.0, 72.3. MS: m/z: 188 (100), 171 (26), 115 (60); Anal. Calc. for
C11H8OS: C, 70.21; H, 4.26. Found: C, 69.89; H, 3.88.
8H-Indeno[1,2-c]thiophene-8-ol (6)
8-H-Indeno[1,2-c]thiophene-8-one (237 mg) was dissolved in
methanol (50 mL). To this solution was added sodium borohy-
dride (50 mg) and the solution allowed to stir for 10 min. The
reaction mixture was then quenched with water (10 mL) and the
methanol partially evaporated. The product was extracted with
diethyl ether (100 mL) and washed with water (100 mL). Yield
230 mg. (96%) m.p.: 135–137 8C. IR: n ¼ 3302, 3202, 1026 cmꢁ1; 1H
NMR d: 7.58 (d, 1H, J ¼ 7.5 Hz), 7.51 (d, 1H, J ¼ 7.2 Hz), 7.3 (m, 3H),
7.2 (d, 1H, J ¼ 2.2 Hz), 5.5 (d, 1H, J ¼ 8.5 Hz), 2.2 (d, 1H, J ¼ 9.1 Hz).
13C NMR d: 150.5, 149.1, 145.2, 136.1, 129.0, 127.5, 126.0, 120.7,
119.6, 112.9, 70.4.MS: m/z: 188 (Mþ, 100), 171 (17), 115 (40). Anal.
Calc. for C11H8OS: C, 70.21; H, 4.26. Found: C, 70.02; H, 3.98.
4H-Indeno[1,2-b]thiophen-4-ol (7) via
4H-indeno[1,2-b]thiophen-4-one (21)
The remaining 4H-indeno[1,2-b]thiophen-4-one (52 mg) from
the previous step was dissolved in methanol (100 mL) and
three equivalents of sodium borohydride (30 mg) were added
and allowed to stir for 1.5 h. The solution was then quenched
with water and the methanol partially evaporated. This
mixture was extracted with diethyl ether (100 mL) and washed
with water (100 mL) in a separatory funnel. The purified
product was obtained by column chromatography (ethyl
acetate:hexane 20:80 by volume) yielding 47.5 mg of
4H-indeno[1,2-b]thiophene-4-ol and when combined with
the previous 4H-indeno[1,2-b]thiophene-4-ol resulted in a
total yield of 97.5 mg (39% yield).
Photolysis in methanol
Approximately 40mg of each isomer: 8H-indeno[2,1-b]thiophene-
8-ol, 8H-indeno[1,2-c]thiophen-8-ol and 4H-indeno[1,2-b]
thiophen-4-ol was dissolved in methanol (25 mL) and placed
in separate quartz tubes. These solutions were purged with
argon for approximately 10 min. The tubes were then attached
to a cooling immersion well containing a Hanovia 450 W
medium-pressure mercury arc lamp and irradiated for 1 h. The
irradiated solutions were evaporated in vacuo and the residue
was separated by preparative TLC (10:90 ethyl acetate:hexane).
1H NMR spectra were acquired for each fraction and compared
with reference samples prepared above.
3-Bromo-4-phenylthiophene11
To a benzene solution (120 mL) of 3,4-dibromothiophene (4.44 g)
was added 2 M aqueous sodium carbonate solution (37.5 mL).
Phenyl boronic acid (2.25 g), dissolved in ethanol (22.5 mL) was
added. The mixture was purged with argon. After 20 min, 225 mg
of tetrakis(triphenylphosphine)palladium(0) was introduced
into the mixture. The mixture was stirred and refluxed under
argon overnight. The flask and condenser were wrapped in
aluminum foil to protect the reaction from light. After the
reaction was complete, it was extracted with dichloromethane
(100 mL) and washed with saturated sodium chloride solution
(50 mL) then with water (3 ꢀ 100 mL). The product was purified
by column chromatography (hexane), (2.98 g, 68% yield). m.p.:
63–67 8C (lit.[34] m.p. 69–70 8C) 1H NMR d: 7.47 (m, 2H), 7.35 (m,
4H), 7.2 (m, 1H).
Preparation of methyl ethers 16, 18 and 20
8-Methoxy-8H-Indeno[2,1-b]thiophene (16)
8H-Indeno[2,1-b]thiophene (10) (45 mg) was dissolved in carbon
tetrachloride (5 ml). NBS (52 mg) was added and the mixture left
to stir and reflux with a 100 W lamp directed on the mixture for
90 min. After cooling, the reaction mixture was filtered and
washed with hexane, and then the solvents were removed by
evaporation under vacuum. The residue was dissolved in a small
amount of acetone, followed by addition of methanol (20 ml) and
AgNO3 (49 mg). The resulting suspension was stirred for 30 min at
room temperature. After filtering, the filtrate was evaporated
under vacuum, and the crude product was purified on silica
column using gradient elution of 0–10% ether in hexanes,
4-Phenylthiophene-3-carboxylic acid (14)
Under an argon atmosphere, n-butyl lithium (5 mL, 2.5 M in
hexanes) was added to a solution of 3-bromo-4-phenylthiophene
(2.98 g) dissolved in dried diethyl ether (60 mL) and was slowly
added over a period of 10 min at ꢁ78 8C. This reaction mixture
was left to stir for 20 min. Dry carbon dioxide was bubbled into
the reaction mixture for 1 h. Water (10 mL) was then added to
quench the reaction and the solution was washed with diethyl
ether (100 mL) in a separatory funnel. The aqueous phase was
then acidified with 1 M HCl (2 mL) and extracted with diethyl
ether (100 mL). Yield 2.21g (87% yield). m.p.: 195–197 8C
(lit.[34] m.p. 206–208 8C). 1H NMR d: 8.31 (m, 1H), 7.4(m, 5H),
7.21(m, 1H), 5.2(s, 1H, COOH).
1
resulting in 30 mg of 16 as a gummy residue (56%). H NMR d
7.54 (d, 1H, J ¼ 7.3 Hz), 7.42–7.47 (m, 2H), 7.36 (t, 1H, J ¼ 7.3 Hz),
7.20–7.27 (m, 2H), 5.54 (s, 1H), 3.28 (s, 3H), 13C-NMR d 148.1, 146.5,
143.3, 138.4, 131.2, 128.9, 125.6, 125.2, 119.3, 118.7, 79.4, 53.5; MS
(GC/MS): m/z 202.1 (Mþ, 74), 187 (100), 171 (80), 115 (40). Anal.
Calc. for C12H10OS: C, 71.29; H, 4.95. Found: C, 71.10; H, 5.12.
8-Methoxy-8H-indeno[1,2-c]thiophene (18)
8H-Indeno[1,2-c]thiophen-8-one (15)
8-H-Indeno[1,2-c]thiophen-8-ol (6) (24 mg) was treated with two
drops of SOCl2 in CDCl3 (1 ml) and left to stir for 0.5 h at room
temperature. After removing solvents by rotary evaporation
under reduced pressure, methanol (10 ml) was added to dissolve
the residue, and then a solution of AgNO3 (27 mg) in methanol
was added. The resulting suspension was stirred for 15 min at
room temperature. After filtering, the filtrate was evaporated
under reduced pressure, and the crude was purified on silica
column using gradient elution of 0–10% ether in hexanes,
PPA (15 mL) was heated to 200 8C in a large test tube using a
Bunsen burner. 4-Phenylthiophene-3-carboxylic acid (1 g) was
added to the PPA and heated for 2–3 min. The hot mixture was
poured into ice water (50 mL) and extracted with dichloro-
methane (3 ꢀ 75 mL). The same procedure was repeated on the
remaining 1.11 g of starting material. The product was separated
and purified by column chromatography (ethyl acetate:hexane
20:80). A white solid was obtained (237 mg, 12%), m.p. 90–93 8C
(lit.[34] m.p. 92–93 8C). 1H NMR d: 7.76 (d, 1H, J ¼ 1.71 Hz), 7.67 (d,
1H, J ¼ 7.48 Hz), 7.5 (m, 2H), 7.28 (m, 1H), 7.14(m, 1H).
1
resulting in 22 mg of 18 as a gummy residue (85%). H-NMR d
7.56 (t, 1 H), 7.38 (t, 1 H), 7.34-7.28 (m, 2 H), 7.24 (d, 1 H, J ¼ 1.8 Hz),
View this article online at wileyonlinelibrary.com
Copyright ß 2010 John Wiley & Sons, Ltd. J. Phys. Org. Chem. 2010, 23 1202–1213

PHOTOCHEMICAL GENERATION OF THIOPHENE ANALOGS OF 9-FLUORENYL CATIONS
5.51 (s, 1 H), 3.28 (s, 3 H).; 13C-NMR d 147.7, 146.2, 145.9, 137.1,
129.1, 127.3, 126.5, 120.7, 119.8, 112.7, 77.1, 53.6; MS (GC/MS) m/z
202.1 (Mþ, 79), 187 (60) 171 (80), 115 (35). Anal. Calc. for
C12H10OS: C, 71.29; H, 4.95. Found: C, 71.51; H, 4.72.
[3] A. Allen, T. T. Tidwell, Chem. Rev. 2001, 101, 1333–1348.
[4] S. L. Mecklenburg, E. F. Hilinski, J. Am. Chem. Soc. 1989, 111,
5471–5472.
[5] R. A. McClelland, N. Mathivanan, S. Steenken, J. Am. Chem. Soc. 1990,
112, 4857–4861.
[6] T. Amyes, M. Richard, M. Novak, J. Am. Chem. Soc. 1992, 114,
8032–8041.
[7] H. Jia, Pv. R. Schleyer, Y. Mo, M. A. McAllister, T. T. Tidwell, J. Am. Chem.
Soc. 1997, 119, 7075–7083.
[8] C. F. Rodriquez, D. L. Vukovic, A. C. Hopkinson, J. Mol. Struct.
(THEOCHEM) 1996, 363, 131–138.
[9] A. R. Katritzky, M. Karelson, S. Sild, T. M. Krygowski, K. Jug, J. Org.
Chem. 1998, 63, 5228–5231.
[10] S. G. Lias, J. F. Liebman, R. D. Levin, J. Phys. Chem. Ref. Data 1984, 13,
695–808.
[11] Hark. Richard, R Hauze. Diane, B Petrovskaia. Olga, Joullie. Madeleine,
M Tetrahedron Lett. 1994, 35(42), 7719–7722.
4-Methoxy-4H-Indeno[1,2-b]thiophene (20)
This isomer was prepared by using the same procedure as that for
16 except that 4H-indeno[1,2-b]thiophene (13) was used as the
starting material. 1H NMR d 7.54 (d, 1H, J ¼ 7.4 Hz), 7.26–7.19 (m,
2H), 5.42 (s, 1H), 3.22 (s, 3H); 13C-NMR d 147.4, 145.7, 144.8, 137.7,
129.0, 127.8, 125.9, 125.2, 123.1, 119.0, 78.7, 53.3; MS (GC/MS) m/z
202.1 (Mþ, 91), 187 (95), 171 (100), 115 (38). HRMS calc. for
C12H10OS (M) 202.0452, found 202.0454.
[12] P. Wan, E. Krogh, J. Am. Chem. Soc. 1989, 111, 4887–4895.
[13] R. A. McClelland, F. L. Cozens, J. Li, S. Steenken, J. Chem. Soc., Perkin
Trans. 2: Phys. Org. Chem. 1996, 1531–1543.
[14] F. Cozens, J. Li, R. A. McClelland, S. Steenken, Angew. Chem. 1992,
104, 753–755 (See also Angew. Chem., Int. Ed. Engl., 1992, 31,
743-5).
[15] Pv. R. Schleyer, C. Maerker, A. Dransfeld, H. Jiao, N. J. Rv. E. Hommes, J.
Am. Chem. Soc. 1996, 118, 6317.
[16] S. Patchkovskii, W. Thiel, J. Mol. Model. 2000, 6, 67 and references
cited
[17] D. Sawicka, S. Wilsey, K. N. Houk, J. Am. Chem. Soc. 1999, 121, 864
[18] M. Buhl, Chem. Eur. J. 1998, 4, 734.
[19] A. R. Lera, R. Alvarez, B. Lecea, A. Torrado, F. P. Cossio, Angew. Chem.,
Int. Ed. 2001, 40, 557
[20] S. P. Verevkin, H.-D. Beckhaus, C. Ruckhardt, R. Haag, S. I. Kozhushkov,
T. Zywietz, A. deMeijere, H. Jiao, P. v. R. Schleyer, J. Am. Chem. Soc.
1998, 120, 11130.
[21] G. G. Gurzadyan, S. Steenken, Chem. Eur. J. 2001, 7, 1808–1815.
[22] L. Zhang, G. D. Mendenhall, J. Photochem. Photobiol. A: Chem. 1997,
107, 169.
[23] J. P. Richard, M. E. Rothenburg, W. Jencks, J. Am. Chem. Soc. 1984, 106,
1361.
[24] R. A. McClelland, V. M. Kanagasabapathy, N. Banait, S. Steenken, J. Am.
Chem. Soc. 1988, 110, 6913.
[25] F. L. Cozens, N. Mathivanan, R. A. McClelland, S. Steenken, J. Chem.
Soc. Perkin Trans. 2, 1992, 2083.
[26] J. R. Gandler, J. Am. Chem. Soc. 1985, 107, 8218.
[27] R. Ta-Shma, W. P. Jencks, J. Am. Chem. Soc. 1986, 108, 8040.
[28] W. C. Herndon, N. S. Mills, J. Org. Chem. 2005, 70, 8492–88496.
[29] S. Pogodin, I. Agranat, J. Org. Chem. 2007, 72, 10096–10107.
[30] N. M. Chaignon, I. J. S. Fairlamb, A. R. Kapdi, R. J. K. Taylor, A. C.
Whitwood, J. Mol. Catal. A 2004, 219, 191–1199.
[31] D. W. H. MacDowell, T. B. Patrick, J. Org. Chem. 1967, 32, 2441–2445.
[32] C. L. Arcus, G. C. Barrett, J. Chem. Soc. 1960, 2098–2102.
[33] F. Sieber, P. Wentworth, K. Janda, J. Comb. Chem. 1999, 1,
540–546.
Laser flash photolysis
Nanosecond laser flash photolysis experiments employed the
pulses (248 nm; ca. 20 ns; ca. 100 mJ) from a Lambda-Physik
Compex 120 excimer laser filled with F2/Kr/He mixtures, and a
Luzchem Research mLFP-111 laser flash photolysis system,
modified as described previously.[35] Solutions were prepared
at concentrations such that the absorbance at the excitation
wavelength was between ca. 0.7 and 0.9. For transient spectra,
solutions were flowed continuously through a thermostatted
7 ꢀ 7 mm Suprasil flow cell connected to a calibrated 100mL
reservoir, fitted with a glass frit to allow bubbling of argon gas
through the solution. A MasterflexTM 77390 peristaltic pump fitted
with Teflon tubing (Cole-Parmer Instrument Co.) was used to pump
the solution from the reservoir through the sample cell at cont-
rolled rates. Solution temperatures were measured with a Teflon-
coated copper/constantan thermocouple inserted into the thermo-
stated sample compartment in close proximity to the sample cell.
Quenching experiments were carried out in 7 ꢀ 7 mm Suprasil
cells, which were replaced regularly throughout each experiment
with fresh solution in order to maintain adequate transient
signals. Appropriate amounts of MeOH were added by microlitre
syringe, as the neat liquid in the cases of 2–4 or as a standard
solution in HFIP in the case of 2-fluorenol. Transient decay
constants were calculated from traces recorded at or near the
absorption maxima using the manufacturer’s software, or by
non-linear least squares analysis of the absorbance-time profiles
using the Prism 5.0 software package (GraphPad Software, Inc.)
and the appropriate user-defined fitting equations, after
importing the raw data from the Luzchem mLFP software. Rate
constants were calculated by linear or non-linear least-squares
analysis of decay rate-concentration data (at least 7 points) that
spanned as large a range in transient decay rate as practically
possible. Errors are quoted as twice the standard deviation
obtained from the least-squares analyses.
[34] D. W. H. MacDowell, A. T. Jeffries, J. Org. Chem. 1970, 35,
2441–2445.
[35] W. J. Leigh, C. R. Harrington, I. Vargas-Baca, J. Am. Chem. Soc. 2004,
126, 16105.
[36] Gaussian 2003: M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T.
Vreven, K. N. Kudin, J. C. Burant, J. M. M. Millam, S. S. Lyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M.
Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J.
Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A.
Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A.
D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al_Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian 03, Revision
B.05; Gaussian,Inc.: Pittsburgh, PA, (2003).
Theoretical Calculations
**
Calculations were carried out at the B3LYP/6-311 þ G //B3LYP/
6-311 þ G level using GAUSSIAN 2003.[36] The energies were
**
corrected using calculated zero-point vibrational energies.
REFERENCES
[1] R. Breslow, Chem. Eng. News 1965, June 28, 90–99.
[2] R. Breslow, Acc. Chem. Res. 1973, 6, 393–398.
J. Phys. Org. Chem. 2010, 23 1202–1213
Copyright ß 2010 John Wiley & Sons, Ltd.
View this article online at wileyonlinelibrary.com