Photoreaction between 2-benzoylthiophene and phenol or indole

Article abstract of DOI:10.1021/jo034225e

Laser flash photolysis, density functional theory (DFT) calculations, and product studies have been performed to understand the mechanism of photoreduction of 2-benzoylthiophene (BT) in the presence of phenol or indole. Time-resolved experiments showed th

Full text of DOI:10.1021/jo034225e

P h otor ea ction betw een 2-Ben zoylth iop h en e a n d P h en ol or In d ole
J ulia Pe´rez-Prieto,*,† Francisco Bosca´,‡ Raquel E. Galian,† Agust´ın Lahoz,‡
Luis R. Domingo,† and Miguel A. Miranda*,‡
Departamento de Quı´mica Orga´nica/ ICMOL, Universidad de Valencia, Av. Vicent Andre´s Estelle´s s/ n,
46100 Burjassot, Valencia and Departamento de Qu´ımica/ Instituto de Technologı´a Quı´mica UPV-CSIC,
Universidad Polite´cnica de Valencia, Camino de Vera s/ n, 46071 Valencia
julia.perez@uv.es
Received February 20, 2003
Laser flash photolysis, density functional theory (DFT) calculations, and product studies have been
performed to understand the mechanism of photoreduction of 2-benzoylthiophene (BT) in the
presence of phenol or indole. Time-resolved experiments showed that BT ketyl (BTH) and phenoxy
(PhO) or indolyl (In) radicals are generated with high rate constants and quantum yields close to
1. However, low conversions (specially in the case of indole) of the starting reagents are obtained
upon prolonged lamp irradiation, indicating that recombination within the radical pairs must occur
to a large extent, regenerating the starting materials. The solvent-dependence of the quenching
rate constants, together with DFT theoretical studies, indicate fundamental differences between
the mechanisms of the reaction of BT triplet with phenol and indole. Thus, data for phenol agree
with the involvement of a hydrogen-bonded exciplex BT...HOPh, where concerted electron and proton
transfer leads to the BTH...OPh radical pair. However, in the case of indole, electron transfer at
the BT...HIn stage precedes proton transfer. Finally, C-C cross-coupling products have been isolated
and characterized in the preparative irradiation of BT in the presence of phenol and indole. The
structures of the products have been confirmed by alternative synthesis.
In tr od u ction
the photochemical behavior of the benzoylthiophene
chromophore.
The 2-benzoylthiophene (BT) chromophore is interest-
ing for a number of reasons. Compounds containing BT
are used as components of polymerization catalysts,1 as
crystals for generating higher harmonics in Nd:YAG
lasers,2 as UV absorbers for photographic materials,3 and
as photostabilizers of polyolefins.4 This chromophore is
also present in the nonsteroidal antiinflammatories
tiaprofenic acid (TPA) and suprofen (SUP). Both TPA and
SUP are known to produce detrimental side-effects on
humans upon exposure to sunlight, either after topical
or systemic administration.5 Such effects involve lipid
peroxidation,6 photohemolysis,7 drug-protein photobin-
ding,8 photooxidation of proteins,9 and DNA lesions.10
The lowest lying triplet of BT has a π,π* configuration
involving mainly the thenoyl group and is generated with
a quantum yield close to unity.11 BT undergoes [2 + 2]
photocycloaddition to olefins through either the CdO or
the thiophene CdC bonds. Thus, Arnold et al.12 have
reported formation of oxetanes by photocycloaddition of
isobutylene to the carbonyl group of BT using a quartz
filter; the resulting products are thermally unstable and
eliminate formaldehyde. By contrast, Cantrell has found
that Pyrex-filtered irradiation leads to cycloaddition to
a thiophene CdC bond.13 Furthermore, reaction at the
(6) Castell, J . V.; Go´mez-Lecho´n, M. J .; Grassa, C., Mart´ınez, L.,
A.; Miranda, M. A.; Tarrega, P. Photochem. Photobiol. 1994, 59, 35.
(7) De Guidi, G.; Chillemi, R.; Costanzo, L. L.; Giufrida, S.; Sortino,
S.; Condorelli, G. J . Photochem. Photobiol., B: Biol. 1994, 23, 125.
(8) Castell, J . V.; Herna´ndez, D.; Go´mez-Lecho´n, M. J .; Lahoz, A.;
Miranda, M. A.; Morera, I. M.; Pe´rez-Prieto, J .; Sarabia, Z. Photochem.
Photobiol. 1998, 68, 660.
Despite its relevance to the above-mentioned photo-
biological and phototechnological applications, very lim-
ited information is available in the literature regarding
* To whom correspondence should be addressed. Tel: (34)-96-
3543050.
(9) (a) Miranda, M. A.; Castell, J . V.; Herna´ndez, D.; Go´mez-Lecho´n,
M. J .; Bosca, F.; Morera, I. M.; Sarabia, Z. Chem. Res. Toxicol. 1998,
11, 172. (b) Figueiredo, A.; Ribeiro, C. A. F.; Goncalo, M.; Baptista, A.
P.; Teixeira, F. J . Photochem. Photobiol. B: Biol. 1993, 18, 161.
(10) (a) Agapakis-Causse, C.; Bosca´, F.; Castell, J . V.; Herna´ndez,
D.; Mar´ın, M. L.; Marrot, L.; Miranda, M. L. Photochem. Photobiol.
2000, 71, 499. (b) Starrs, S. M.; Davies, R. J . H. Photochem. Photobiol.
2000, 72, 291. (c) Artuso, T.; Bernadou, J .; Meunier, B.; Piette, J .;
Paillous, N. Photochem. Photobiol. 1991, 54, 205.
(11) (a) Becker, R. S.; Favaro, G.; Poggi, G.; Romani, A. J . Phys.
Chem. 1995, 99, 1410. (b) Arnold, D. R.; Birtwell, R. J . J . Am. Chem.
Soc. 1973, 95, 4599.
(12) Arnold, D. R.; Birtwell, R. J .; Clarke, B. M., J r. Can. J . Chem.
1974, 52, 1681.
† Universidad de Valencia.
‡ Universidad Polite´cnica de Valencia.
(1) Wrzyszczynski, A.; J anota, H. Polymer 1996, 41, 560.
(2) Vizgert, R. V.; Davydov, B. L.; Kotovshchikov, S. G.; Starodubt-
seva, M. P. Kuantovaya Electron (Moscow) 1982, 9, 380. Chem. Abstr.
1982, 96, 226199r.
(3) Leppard, D. G.; Burdeska, K. Eur. Pat. Appl., 1993. Chem. Abstr.
1993, 119, 82785x.
(4) Zimin, Y. B.; Levin, P. I.; Matveeva, E. A.; Kozodoi, A. A.;
Sotnikova, L. M. Plast. Massy 1970, 20. Chem. Abstr. 1970, 72,
101351g.
(5) Bosca´, F.; Mar´ın, M. L.; Miranda, M. A. Photochem. Photobiol.
2001, 74, 637.
(13) Cantrell, T. S. J . Chem. Soc. 1972, 155.
10.1021/jo034225e CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/22/2003
5104
J . Org. Chem. 2003, 68, 5104-5113

Photoreduction of 2-Benzoylthiophene
carbonyl group also occurs in the [2 + 2] photocyclo-
addition of BT to 2,5-dimethylthiophene, while methyl-
maleic anhydrides react with a thiophene CdC bond.14
Quenching of the BT triplet state by methyl methacrylate
(MMA) has been postulated to occur via formation of 1,4-
biradicals; however, only formation of uncharacterized
polymers (and no oxetanes) has been reported.15
On the other hand, excited triplet BT is able to abstract
hydrogen from suitable donors, though with low ef-
ficiency. Traynard and Blanchi have reported quantum
yields of 0.17 for the disappearance of BT upon irradia-
tion in 2-propanol, much lower than the value of 1.23 for
benzophenone.16 The reaction products, assumed to be
pinacols but not isolated, were only assigned on the basis
of UV spectra. By contrast, the pinacols obtained after
irradiation of the analogue 2-benzoyl-5-ethylthiophene
(Et-BT) in 2-propanol have been isolated and fully
characterized.17,18 The involved reductive process is medi-
ated by the second excited triplet, of n,π* character.
Temperature dependence experiments agree with the
reaction occurring after thermal population of the reac-
tive upper n,π* ketone triplet.18
F IGURE 1. A: Transient absorption spectra recorded 50 ns
after laser excitation (355 nm) of BT in deaerated methanol
(0), acetonitrile (O), or dichloromethane (4) solutions. B:
Transient absorption spectra recorded 8 µs after laser excita-
tion (355 nm) of BT in deaerated methanol (9), acetonitrile
(b), or dichloromethane (2) solutions.In part B, absorbances
have been multiplied by two.
In this paper, our aim was to perform a detailed study
on the photoreactivity of BT toward phenol and indole.
Using the laser flash photolysis technique, we have been
able to observe clearly the transient absorption spectrum
of BT ketyl radical, together with formation of phenoxy
and indolyl radicals. Product studies have shown that
not only the ketone hydrodimers, but also cross-coupling
products can be isolated and characterized in both cases.
Furthermore, theoretical studies on the interaction of BT
triplet state with phenol and indole have been performed;
some interesting differences between ketone/phenol and
ketone/indole exciplexes are found.
In accordance with the low efficiency of hydrogen
abstraction by the benzoylthiophene triplet,19 its lifetime
i
in neat ButOH (τ ) 4.5 µs) is only slightly shorter than
in 2,2,2-trifluoroethanol (5.2 µs).19b The long lifetime and
intensity of BT triplet, as well as the overlap of triplet
and ketyl absorptions, have made the clear observation
and study of the latter intermediate difficult.
Formal hydrogen abstraction by the BT chromophore
can also occur by sequential proton-electron-transfer
processes. As a consequence, hydrodimers have been
obtained when compounds containing the BT chro-
mophore are irradiated in the presence of electron donors
such as p-cresol.9a,17 In similar cases, the high reactivity
of π,π* triplet ketones toward phenolic hydrogen abstrac-
tion has been explained as due to rate-determining
electron transfer within the phenol-ketone exciplex,
followed by proton transfer.20,21 Although phenoxy/ketyl
radical pairs are the expected primary intermediates, it
is remarkable that up to now C-C cross-coupling prod-
ucts have only been detected and isolated when using a
phenol unable to form the hemiketals resulting from C-O
coupling, such as the sterically hindered 2,6-di-tert-
butylphenol.17 No product studies of this type have been
performed with indolic derivatives, although indoles are
also able to quench BT-derived triplets.9a
Resu lts a n d Discu ssion
Tim e-Resolved Stu d ies. Dynamic studies on the
reaction of benzoylthiophene with phenol and indole were
performed in methanol, acetonitrile, and dichloromethane
using 355 nm laser excitation (Nd:YAG).
Laser flash photolysis of deaerated BT solutions led
to a transient absorbing in the 300-700 nm range, with
maxima at 350 (ꢀmax ) 4800 M-1 cm-1) and 600 nm
(ꢀmax ) 2900 M-1 cm-1). In the three solvents, the spectra
obtained at short times after the laser pulse matched that
previously reported for BT triplet state.11a After longer
delay times, a residual absorption peaking at 350 and
580 nm followed triplet disappearance in methanol. This
absorption was ascribed to BT ketyl radical formation,
due to hydrogen abstraction from the donor solvent (see
Figure 1 for a comparison between the absorption spectra
obtained in the three solvents at two different times after
the laser pulse). Triplet lifetimes were measured in
oxygen-free solutions and found to be 1.9 µs in dichlo-
romethane, 1.6 µs in acetonitrile, and 1.4 µs in methanol
(a lifetime of 1.4 µs in acetonitrile has been reported in
the literature).11a
Quenching of the triplet by phenol and indole was
observed in the three solvents at room temperature. The
time-evolution of the transient spectra in dichloromethane
is shown in Figure 2. Similar spectra were obtained in
the other two solvents (not shown). In the presence of
phenol the disappearance of the BT triplet was followed
by formation of transient(s) with absorption maxima at
(14) (a) Rivas, C.; Bolivar, R. A.; Gonzalez, R.; Machado, R.; Vargas,
F. Monats. Chem. 1996, 127, 529. (b) Rivas, C.; Bolivar, R. A. J .
Heterocycl. Chem. 1973, 10, 967.
(15) Timpe, H. J .; Kronfeld, K. P.; Schiller, M. J . Photochem.
Photobiol. A: Chem. 1991, 62, 245.
(16) Traynard, P.; Blanchi, J . P. J . Chim. Phys. Physicochim. Biol.
1972, 69, 284.
(17) Miranda, M. A.; Pe´rez-Prieto, J .; Lahoz, A.; Morera, I. M.;
Sarabia, Z.; Mart´ınez-Man˜ez, R.; Castell, J . V. Eur. J . Org. Chem. 1998,
11, 172.
(18) Encinas, S.; Miranda, M. A.; Marconi, G.; Monti, S. Photochem.
Photobiol. 1998, 67, 420.
(19) (a) Bosca, F.; Miranda, M. A.; Morera, I. M.; Samadi, A. J .
Photochem. Photobiol. B: Biology 2000, 58, 1. (b) Ortica, F.; Romani,
A.; Favaro, G. J . Phys. Chem. 1999, 103, 1335.
(20) (a) Miranda, M. A.; Lahoz, A.; Bosca´, F.; Metni, M. R.;
Abdelouahab, F. B.; Castell, J . V.; Pe´rez-Prieto, J . Chem. Commun.
2000, 2257. (b) Miranda, M. A.; Lahoz, A.; Mat´ınez-Man˜ez, R.; Bosca´,
F.; Castell, J . V.; Pe´rez-Prieto, J . J . Am. Chem. Soc. 1999, 121, 11569.
(21) (a) Biczo´k, L.; Be´rces, T.; Linschitz, H. J . Am. Chem. Soc. 1997,
119, 11071. (b) Leigh, W. J .; Lathioor, E. C.; Pierre, M. J . S. T. J . Am.
Chem. Soc. 1996, 118, 12339.
J . Org. Chem, Vol. 68, No. 13, 2003 5105

Pe´rez-Prieto et al.
SCHEME 1. Ra d ica ls F or m ed a fter Qu en ch in g of
BT Tr ip let by P h en ol or In d ole
(Figures 3B and 3C). On the other hand, the BT ketyl
radical absorption spectrum was obtained by laser flash
photolysis of deaerated methanolic solutions of BT, once
the BT triplet had completely disappeared (Figure 3A).
It presents maxima at 350 (ꢀmax ) 9100), 540 (shoulder),
and 580 nm (ꢀmax ) 2100 M-1 cm-1). It is remarkable that,
while formation of the ketyl radical from benzophenone
triplets is associated with a bathochromic shift of ca. 20
mm (from 525 to 545 nm),25 in the case of BT the
analogous process leads to a hypsochromic shift of the
same magnitude (from λmax ) 600 to 580 nm).
Molar absorption coefficients of BT triplet and ketyl
radical (see above) were determined by a comparative
method, using solutions of benzophenone and 2-benzoyl-
thiophene with the same absorbances at the monitoring
wavelength. This procedure also allowed us to establish
that the efficiencies for the hydrogen transfer from phenol
or indole to BT triplet are close to the unity (see
Experimental Section).
F IGUR E 2. A: Transient absorption spectra of a dichoro-
methane solution of BT (0.9 mM) and phenol (4mM) at 0.1 µs
(O), 0.5 µs (b), and 2.0 µs (4) after the laser pulse. B: Transient
absorption spectra of a dichoromethane solution of BT (0.9
mM) and indole (4 mM) at 0.1 µs (O), 0.5 µs (b) and 2.0 µs (4)
after the laser pulse.
350 (strong), 390 (shoulder), 540 (shoulder), and 580 nm
(medium intensity), and no absorption beyond 625 nm
(Figure 2A). Similar spectra were observed when using
indole as quencher, but without the shoulder at 390 nm
and with an enhanced broad band in the 450-550 nm
region (Figure 2B).
It has been reported that the transient BT ketyl radical
absorption spectrum possesses two maxima located at
355 and 580 nm.19b The transient absorption spectra for
phenoxy and indolyl radicals have also been described.
Quenching of BT triplet by phenol and indole was
measured in the three solvents (methanol, acetonitrile,
and dichloromethane) by determining the triplet lifetime
in the absence and in the presence of several quencher
concentrations. The rate constants kq were obtained from
Stern-Volmer plots (Figures 4A and 4B), according to
eq 1
The former displays two maxima at 385-405 nm (ꢀ405
)
22
3200 M-1 cm-1
)
and the latter exhibits peaks at 320
(ꢀ320 ) 4000 M-1 cm-1) and 520 nm (ꢀ520 ) 2000 M-1
cm-1).23 Hence, the spectra obtained at long times after
laser irradiation of BT in the presence of phenol and
indole might well be the result of an overlap of ketyl
(BTH) and phenoxy (PhO) or indolyl (In) radicals absorp-
tions (Scheme 1).
Laser flash photolysis of aerated solutions could in
principle be used to demonstrate the formation of oxygen
(phenoxy)- and nitrogen (indolyl)-centered radicals, tak-
ing into account their low reactivity toward oxygen
compared to that of ketone triplets and ketyl radicals,
which are quenched at close to diffusion controlled
rates.24 In agreement with expectations, both phenoxy
and indolyl absorption spectra were clearly detected at
2 µs after the laser pulse, in the presence of oxygen
kd ) 1/τ0 + kq[quencher]
(1)
The quenching rate constants for phenol in the differ-
ent solvents decreased in the order dichloromethane (2.30
× 109 M-1 s-1) > acetonitrile (0.14 × 109 M-1 s-1) >
methanol (0.06 × 109 M-1 s-1) (Figure 4A). The values
obtained when phenol and indole were used as BT triplet
quenchers in dichloromethane were similar (2.01 × 109
M-1 s-1 for indole). However, kq for indole was higher in
methanol (6.00 × 109 M-1 s-1) and lower in acetonitrile
(24) (a) Bejan, E. V.; Font-Sanchis, E.; Scaiano, J . C. Org. Lett. 2001,
3, 4059. (b) Chattopadhyay, S. K.; Kumar, C. V.; Das, P. K. J .
Photochem. 1995, 30, 81. (c) Smith, R. C.; Reeves, J . C.; Dage, R. C.;
Schnettler, R. A. Biochem. Pharmacol. 1987, 36, 1457. (d) Burton, G.
W.; Ingold, K. U. Acc. Chem. Res. 1986, 19, 194. (e) Wilkinson, F.;
Garner, A. Photochem. Photobiol. 1978, 27, 659.
(25) (a) Bosca, F.; Miranda, M. A. J . Photochem. Photobiol. B: Biol.
1998, 43, 1. (b) Baral-Tosh, S.; Chattopadhyay, Das, P. K. J . Phys.
Chem. 1994, 88, 1404. (c) Porter, G.; Windsor, M. W. Proc. R. Soc.
London Ser. A 1958, 245, 238.
(22) (a) Mere´nyi, G.; Lind, J .; Shen, X. J . Phys. Chem. 1988, 92,
134. (b) Bansal, K. M.; Fessenden, R. W. Radiat. Res. 1976, 67, 1. (c)
Schuler, R. H.; Neta, P.; Zemel, H.; Fessenden, R. W. J . Am. Chem.
Soc. 1976, 98, 3825.
(23) J ovanovic, S. V.; Steenken, S. J . Phys. Chem. 1992, 96, 6674.
5106 J . Org. Chem., Vol. 68, No. 13, 2003

Photoreduction of 2-Benzoylthiophene
F IGURE 3. A: Transient absorption spectra of a deaerated
methanol solution of BT (0.9 mM) obtained 8 µs after the laser
pulse. B: Transient absorption spectra of an aerated dichlo-
romethane solution of BT (0.9 mM) and phenol (4 mM) 2 µs
after the laser pulse. C: Transient absorption spectra of an
aerated dichloromethane solution of BT (0.9 mM) and indole
(4 mM) 2 µs after the laser pulse.
F IGURE 4. A and B: Stern-Volmer plots of the quenching
of the BT triplet generated from 355 nm laser flash photolysis
of BT in deaerated solutions of acetonitrile (9), methanol (b),
or dichloromethane (1) by phenol (A) or indole (B), monitored
at 610 nm. C and D: correlation between the calculated and
experimental quenching rate constants by phenol (C) and
indole (D).
(0.60 × 109 M-1 s-1) (Figure 4B). The different trends
observed for BT triplet quenching in the three solvents
could indicate that there are some differences in the
mechanisms operating in the reaction of BT triplet with
phenol as compared to indole.
In this context, the influence of the solvent on the
kinetics of electron-transfer processes has received con-
siderable attention. The Taft model for the solvent-
dependent term in the Marcus theory provides a useful
approach for prediction of the electron-transfer kinetics.26
This model recognizes the importance of solvent-solute
interactions in the development and evolution of the
transition state. Thus, there is partial reorientation of
the solvent molecules surrounding the reacting species,
to form the transition state and the products; these steps
depend on the acidities and basicities of the solvent
molecules, reactants, and solutes. However, when the
actual electron-transfer process occurs, reorientation
accompanying the change in charge distribution is pre-
cluded by the Franck-Condon principle; nonetheless,
electronic rearrangements can still take place. The latter
will be dependent upon the polarizability of the solvent
molecules. Hence, the experimental rate constants can
be correlated with the solvent properties by means of the
Taft model by using a set of simultaneous equations of
the type in eq 2.
ln kexp,j ) aRj + bâj + cπ*j
(2)
where R is an index of solvent HBD (hydrogen bond
donor) acidity, â is an index of solvent HBA (hydrogen
bond acceptor) basicity, and π* is a measure of solvent
polarity/polarizability. The coefficients a, b, and c for
phenol and indole can be obtained by solving the corre-
sponding simultaneous equations (eq 2), using the known
(26) (a) Abbott, A. P.; Rusling, J . F. J . Phys. Chem. 1990, 94, 8910.
(b) Kamlet, M. J .; Dickinson, C. Taft, R. W. Chem. Phys. Lett. 1981,
77, 69.
J . Org. Chem, Vol. 68, No. 13, 2003 5107

Pe´rez-Prieto et al.
SCHEME 2. P r ocesses In volved in th e
P h otoa ctiva tion -Dea ctiva tion of BT in th e
Absen ce a n d in th e P r esen ce of a Hyd r ogen Don or
(DH)
TABLE 1. Solven t P a r a m eter s (Ta ft m od el) a n d
Exp er im en ta l Ra te Con sta n ts for Electr on Tr a n sfer fr om
P h en ol or In d ole to BT Tr ip let
solvent (j)
methanol
dichloromethane 0.13 0.10 0.82
acetonitrile 0.19 0.40 0.75
R
â
π* phenol ln kexp indole ln kexp
0.98 0.66 0.60
17.91
21.56
18.76
22.52
21.42
20.21
R, â, and π* values for the three solvents (j), namely
methanol, dichloromethane, and acetonitrile27 (Table 1).
The results are shown in eqs 3 and 4.
phenol: ln kj ) 5.28Rj - 4.32âj + 25.98π*j (3)
indole: ln kj ) 7.54Rj + 0.28âj + 24.88π*j (4)
important in the case of indole. The difference in reactiv-
ity between indole and phenol toward BT triplet might
arise from the lower ionization potential and acidity of
indole.22a,23,33
Figures 4C and 4D show the correlation between the
experimental kexp,j and the calculated kj from eqs 3 and
4. It can be clearly seen that for both phenol and indole
the correlation using the Taft model is quite satisfactory.
A clear consequence of these equations is that the
contribution of solvent polarity/polarizability (π*j) is by
far the most important factor determining the value of
the quenching rate constants for both phenol and indole.
The equations also indicate some contribution of hydro-
gen bonding by HBD solvents (Rj). Interestingly, contri-
bution of hydrogen bonding of the quencher to HBA
solvents (âj) is very small for indole (b ) 0.28) and
becomes even negative in the case of phenol (b ) -4.32).
The last finding agrees well with retardation of bimo-
lecular phenolic hydrogen abstraction by the BT triplet
due to the basicity of the solvent as expected from a
decreased concentration of free phenol. Similar effects are
observed in the rate constants for phenolic hydrogen
abstraction by benzophenones.28,29
Th eor etica l Ca lcu la tion s. The mechanism of the
reaction between photoactivated BT and phenol or indole
was theoretically studied using density functional theory
(DFT) methods. Excitation of BT results in formation of
the excited singlet state (S1), which undergoes fast
intersystem crossing (ISC) to the lower lying excited
triplet state (T1). Subsequent radiationless decay would
lead to the ground state (S0) (Scheme 2). In the presence
of a donor (DH), hydrogen abstraction can in principle
take place from the excited triplet state. Therefore, the
mechanism of photoinduced hydrogen abstraction be-
tween triplet BT and a phenol molecule was studied. The
UB3LYP/6-31G* total and relative energies of the reac-
tants and intermediates are given in Table 2, while the
corresponding energy profile is represented in Figure 5.
The geometries of the optimized triplets are given in
Figure 6 (see Computational Methods in Experimental
Section).
The UB3LYP/6-31G* energy for vertical excitation of
the BT chromophore to the singlet state S1 is 82.0 kcal/
mol. The excited triplet state T1 (61.2 kcal/mol relative
to the ground-state S0) lies ca. 20 kcal/mol below. These
results are in reasonable agreement with the literature
data.11a In the presence of a phenol molecule hydrogen-
bonded to the carbonyl oxygen atom of BT, hydrogen
abstraction by BT (T1) to give the triplet radical pair
BTH‚‚‚OPh can be envisaged. This intermediate can be
converted into BT‚‚‚HOPh (S0) by hydrogen back-dona-
tion to the phenoxy radical. The BT‚‚‚HOPh (T) is 0.5
kcal/mol lower in energy than BT (T1). It is remarkable
that the triplet radical pair BTH‚‚‚OPh (T) lies ca. 16
kcal/mol below BT‚‚‚HOPh (T); this supports the feasibil-
ity of the proposed hydrogen abstraction from the BT
triplet state. All attempts to locate the corresponding
transition state for hydrogen abstraction were unsuc-
cessful. These results suggest that this is a barrierless
process from the triplet state. The analogous reaction was
also studied for indole, obtaining similar results. The
energies of the hydrogen-bonded triplets BT‚‚‚HIn (T) and
Leigh et al.29 have proposed that the key mechanistic
feature of hydrogen abstraction from phenol by triplet
aromatic ketones is hydrogen bonding between the
excited triplet ketone and the phenol, which has the effect
of adjusting the reduction and oxidation potentials of the
reactants to the extent where electron transfer becomes
thermodynamically favorable. A similar rationalization
would be applicable to the photoreaction between BT and
phenol or indole, as the actual electron-transfer step
would be endergonic according to the Weller equation.30,31
This would be more than counterbalanced by the acidity
of phenol or indole radical cation and the basicity of the
ketone anion radical.32 Such effect appears to be less
(27) Marcus, Y. Chem. Soc. Rev. 1993, 409.
(28) Das, P. K.; Encinas, M. V.; Scaiano, J . C. J . Am. Chem. Soc.
1981, 103, 4154.
(29) Lathioor, E. C.; Leigh, W. J .; Pierre, M. J . S. J . Am. Chem. Soc.
1999, 121, 11984.
(30) Rehm, D.; Weller, A. Isr. J . 1970, 8, 259.
(31) The E0 values for phenol and indole, measured by cyclic
voltammetry, are 1.4 and 1.1 V vs SCE, respectively. The E0 of the
2-benzoylthiophene is -1.8 V vs SCE. The triplet energy of BT is 62.7
kcal/mol.11a By fitting these data into the Weller equation,30 the free
energy changes (∆Get) associated with the electron-transfer step would
be ca. +11 (phenol) and +4 (indole) kcal/mol, neglecting the Coulombic
contribution to ∆Get.
(32) The pKa values for the radical cations of phenol and indole have
been found to be -8.1 (J ennings, P.; J ones, A. C.; Mount, A. R.;
Thomson, A. D. J . Chem. Soc., Faraday Trans. 1997, 93, 3791) and
+4.9 (Bordwell, F. G.; Cheng, J . J . Am. Chem. Soc. 1991, 113, 1736),
respectively. The pKa of the ketyl radical has been found to be 9.3 for
the benzophenone analogue (Hayon, E.; Ibata, T.; Lichtin, N. N.; Simic,
M. J . Phys. Chem. 1972, 76, 2072).
(33) (a) Canonica, S.; Hellrung, B.; Wirz, J . J . Phys. Chem. A 2000,
104, 1226. (b) Lagier, C. M.; Olivieri, A. C.; Harris, R. K. J . Chem.
Soc., Perkin Trans. 2 1998, 104, 1791. (c) Barbou, N.; Feitelson, J . J .
Phys. Chem. 1984, 88, 1065. (d) Dixon, W. T.; Murphy, D. J . Chem.
Soc. Faraday Trans. 2 1976, 72, 1221. (e) Yagil, G. Tetrahedron 1967,
23, 2855.
5108 J . Org. Chem., Vol. 68, No. 13, 2003

Photoreduction of 2-Benzoylthiophene
TABLE 2. UB3LYP /6-31G* Tota l a n d Rela tive En er gies, Geom etr ica l P a r a m eter s of th e Rea cta n ts a n d In ter m ed ia tes,
a n d Ch a r ge Tr a n sfer a t th e BT F r a gm en t in th e Tr ip let Sta tes a n d th e Tr ip let Ra d ica l P a ir s, a fter Excita tion /
Dea ctiva tion of BT in th e Absen ce a n d in th e P r esen ce of Hyd r ogen Don or s
chemical entity
total energy (au)
∆E (kcal/mol)
BT-H (Å)
D-H (Å)
D1a (degree)
D2a (degree)
CTb (e)
BT (S0)
BT (S1)
BT (T1)
BT-HOPh (S0)
BT-HOPh (T)
BTH-OPh (T)
BT-HIn (S0)
BT-Hin (T)
BTH-In (T)
-897.389704
-897.259024
-897.292160
-1204.869927
-1204.773139
-1204.798473
-1261.217229
-1261.125197
-1261.141127
82.0
61.2
1.851
1.748
0.986
1.961
1.689
0.997
60.7
44.8
0.986
1.804
6.9
13.6
29.5
23.7
-0.02
-0.05
57.8
47.8
1.041
1.797
5.7
12.6
30.8
24.9
-0.28
-0.06
a
b
D1 and D2 are the (O-C-C-S) and (O-C-C-C) dihedral angles, respectively. Charge transfer located at the BT(H) fragment.
between this hydrogen and the oxygen or nitrogen atoms
of the donors (D-H). For the excited triplet states
BT‚‚‚HOPh (T) and BT‚‚‚HIn (T) these distances are
1.748 and 1.689 Å (BT-H) and 0.986 and 1.041 Å (D-H),
respectively. These BT-H distances are shorter than
those found at the ground states BT‚‚‚HOPh (S0) and BT‚
‚‚HIn (S0) (1.851 and 1.961 Å, respectively). The shorter
BT-H distance found for BT‚‚‚HIn (T) as compared with
BT‚‚‚HOPh (T) agrees with the larger stabilization for
the former. In the case of the triplet radical pairs
BTH‚‚‚OPh (T) and BTH‚‚‚In (T) the corresponding
F IGURE 5. Energy profile (in kcal/mol) for the photoactiva-
distances are 0.986 and 0.997 Å (BT-H), and 1.804 and
tion/deactivation of BT in the absence and in the presence of
1.797 Å (D-H), respectively. The shorter BT-H distances
found in these species confirm that the hydrogen atom
has been transferred from the donor to the BT triplet,
while the PhO and In radicals are now hydrogen bonded
to the BTH radical. The D1 (O-C-C-S) and D2 (O-C-
C-C) dihedral angles (see Table 2) show that the
thiophene and benzene rings are twisted with respect to
the plane of the carbonyl group of BT. This deviation from
the planar arrangement of the two aromatic rings is a
consequence of the steric hindrance between the ortho
hydrogen atoms. In the BT‚‚‚HD (T) states D1 is smaller
than D2, indicating that conjugation of the carbonyl
group with the thiophene ring is larger than with the
benzene ring. In the BTH‚‚‚D (T) state D1 increases while
D2 decreases, indicating a higher participation of the
phenyl group in the delocalization of the system. In the
isolated BT molecule, in the absence of any interaction
with DH donors, the carbonyl group is π-conjugated (and
almost coplanar) with the thiophene ring, while the
degree of twisting of the benzene ring is ca. 71° in the
ground state.34
a donor (DH) molecule.
To understand the electronic features of step 5 in
Scheme 2, the degree of charge transfer at the BT‚‚‚HD
(T) and BTH‚‚‚D (T) species was analyzed. The natural
atomic charges have been shared between the BT/HD and
the BTH/D fragments. The results given in Table 2
suggest that the mechanism for the hydrogen abstraction
step is different for phenol and indole. Thus, while for
phenol both species present a similar (and almost neg-
ligible) charge, suggesting that the hydrogen atom, one
electron, and one proton are transferred along the
process, in the case of indole a larger charge transfer is
found in BT‚‚‚HIn (T) (0.28 e). This indicates that for
indole electron-transfer precedes proton transfer and is
F IGURE 6. Structures of the triplet states. The lengths of
the hydrogen-bonds are given in angstroms.
BTH‚‚‚In (T) are 57.8 and 47.8 kcal/mol, respectively,
above the ground-state BT‚‚‚HIn (S0). When compared
with the case of phenol, the hydrogen abstraction process
for indole is 5.9 kcal/mol less exothermic (10.0 vs 15.9
kcal/mol).
Figure 6 shows the geometries of the species relevant
to hydrogen abstraction by BT from phenol and indole.
The most significant data are the distances between the
acidic hydrogen of phenol and indole and the carbonyl
oxygen atom of BT (BT-H), as well as the distances
(34) (a) Abu-Eittah, R.; Hilal, R. Bull. Chem. Soc. J pn. 1978, 51,
2718. (b) Cheng, C. L.; J ohn, I. G.; Ritchie, G. L. D.; Gore, P. H.; Farnell,
L. J . Chem. Soc., Perkin Trans. 2 1975, 744.
J . Org. Chem, Vol. 68, No. 13, 2003 5109

Pe´rez-Prieto et al.
SCHEME 3. P r od u ct F or m a tion in th e P h otor ea ction of BT a n d P h OH
SCHEME 4. P r od u ct F or m a tion in th e
P h otor ea ction of BT a n d In H
in agreement with the fact that indole is better electron-
donor than phenol, explaining the lower energy found for
the species BT‚‚‚HIn (T).
Stea d y-Sta te Stu d ies. Irradiation of a deaerated
acetonitrile solution of benzoylthiophene and phenol,
using the Pyrex-filtered light from a medium-pressure
mercury lamp, led to a complex mixture containing the
known pinacols 135 along with minor amounts of pina-
colone 2 (from rearrangement of the pinacols)35b and
cross-coupling products with M+• 264, 266, and 282.
1
Analysis of the isolated pinacols by H NMR allowed a
reliable structure assignment by comparison with the
data reported for analogous compounds (Et-BT pinacol
diastereoisomers).17 On the other hand, the spectral data
agreed with structures 3, 4, and 536 for the cross-coupling
photoproducts with M+. 282, 264, and 266, respectively
(Scheme 3). Such assessment was confirmed by compari-
son of their spectra with those of similar products
obtained upon irradiation of Et-BT in the presence of the
2,6-di-tert-butylphenol. An easy interconversion (dehy-
dration/hydration) was observed between products 3 and
4.
Likewise, irradiation of BT in the presence of indole,
led to a complex mixture consisting mainly of the BT
pinacols. However, in this case compounds 6 (M+• 303),
and 7 (M+• 289) were also present in the photomixtures
as minor products (Scheme 4).
The structural assignment of 5, 6, and 7 was confirmed
by unambiguous synthesis, using well-established meth-
ods.37 Thus, compound 5 was synthesized by sodium
borohydride reduction of fuchsone 4 (Scheme 5a). The
synthesis of compound 6 was accomplished through
benzoylation of the thienyl group of 3-thien-2-yl-1H-
indole, which was obtained in one step by adding thien-
2-yllithium to a refluxing mixture of isatin in ether,
followed by treatment with LiAlH4, (Scheme 5b). Com-
pound 7 was obtained by acid-catalyzed condensation of
phenyl(thien-2-yl)methanol with indole (Scheme 5c).
Related photochemical reactions of phenols with aro-
matic ketones have been previously studied.17,37b Thus,
photolysis of aromatic ketones in the presence of 2,6-
disubstituted phenols, followed by coupling of the gener-
ated ketyl and phenoxy radicals and tautomerization
gives rise to triaryl carbinols. These carbinols are con-
verted into 3,5-disubstituted fuchsones in a nonphoto-
chemical process. Furthermore, Becker has demonstrated
that benzophenone (BP) sensitizes secondary photochemi-
cal reactions of the generated fuchsones such as reduc-
tion. Thus, the fuchsone triplet state can abstract hy-
drogen atoms from the solvent or oxidize phenols,
ultimately leading to the triarylmethyl radical. Dimer-
ization of this intermediate (a reaction analogous to
formation of pinacols) does not occur due to steric
hindrance; instead, it disproportionates to fuchsone and
triarylmethane. Analogous reaction steps would explain
formation of all the products obtained in the photoreac-
(35) (a) Foulatier, P.; Salaun, J . P.; Caullet, C. C. R. Acad. Sc. Paris
1974, 279, 779. (b) Kegelman, M. R.; Brown, E. V. J . Am. Chem. Soc.
1953, 75, 5961.
(36) Lapkin, I. I.; Belonovich, M. I. Metody Poluch. Khim. Reaktivov
Prepr. 1969, 142. Chem. Abstr. 1971, 74, 1414126h.
(37) (a) Bergman, J . Acta Chem. Scand. 1971, 25, 1277. (b) Becker,
H. D. J . Org. Chem. 1967, 32, 2115.
5110 J . Org. Chem., Vol. 68, No. 13, 2003

Photoreduction of 2-Benzoylthiophene
SCHEME 5. Alter n a tive Syn th esis of
P h otop r od u cts 5, 6, a n d 7
radicals with high rate constants and quantum yields
close to 1. However, long irradiation times are needed
(mainly in the case of indole) to achieve important
conversions in the steady-state irradiation experiments.
These data indicate that ketyl and phenoxy or indolyl
radicals largely react to regenerate starting materials,
as it has been proposed for the photoreactions of ben-
zophenones with phenols and amines. Moreover, the
efficiency of radical formation reflects that when electron-
transfer precedes proton transfer in the exciplex (as with
indole), subsequent back electron transfer is slow as
compared with proton transfer.
The different trends observed for the solvent-depen-
dence of the quenching rate constants, together with DFT
theoretical studies, indicate fundamental differences
between the mechanisms of the reaction of BT triplet
with phenol and indole. Thus, for the BT/phenol pair,
data agree with the involvement of a hydrogen-bonded
exciplex, where concerted electron and proton transfer
leads to the radical pair. Conversely, in the case of indole
electron-transfer precedes proton transfer. This marked
reactivity difference must arise from the lower ionization
potential and acidity of indole, together with the lower
acidity of the indole radical cation, when compared to the
corresponding values for phenol and its radical cation.
Finally, it is remarkable that C-C cross-coupling
products have been isolated and characterized in the
irradiation of BT in the presence of phenol and indole.
The structures of the products have been confirmed by
alternative synthesis where needed.
SCHEME 6. Mech a n istic P r op osa l for th e P r od u ct
F or m a tion in th e P h otor ea ction of
2-Ben zoylth iop h en e w ith P h en ol a n d In d ole.
Exp er im en ta l Section
1
Gen er a l. H and 13C NMR spectra were recorded in a 300
MHz spectrometer; chemical shifts (δ) are reported in ppm
relative to TMS. The coupling constants (J ) are in hertz (Hz).
Column chromatography was performed on silica gel (230-
400 mesh). HPLC was carried out using a C-18 Bondapack
column or a Silica column. High-resolution mass spectra were
conducted at the SCSIE in Valencia, Spain.
La ser F la sh P h otolysis. Laser flash photolysis experi-
ments were carried out by using the third harmonics (355 nm)
of a pulsed Nd:YAG laser. The pulse duration was 10 ns, and
the energy of the laser beam was 10 mJ /pulse, respectively. A
Lo255 Oriel xenon lamp was employed as detecting light
source. The laser flash photolysis apparatus consisted of the
pulsed laser, the Xe lamp, a 77200 Oriel monochromator, and
an Oriel photomultiplier (PMT) system made up of 77348 side-
on PMT, 70680 PMT housing, and a 70705 PMT power supply.
The oscilloscope was a TDS-640A Tektronix. The output signal
was transferred to a personal computer for data analysis.
The molar absorption coefficients of BT triplet excited state
in MeCN were determined by a comparative method (eq 5),
using the triplet absorption of benzophenone (BP) in MeCN
as an actinometer:
tion of BT with phenol and indole, except ketone 6. The
latter would originate from coupling of the delocalized
ketyl radical, through the thienyl 5-position, with the
indolyl radicals generated in the process (Scheme 6).
Remarkably, analysis of the product distribution revealed
some differences between the reactions with phenol and
indole. Thus, the compound analogous to the ketone 6,
isolated in the reaction with indole, was hardly detected
as traces by GC/MS in the case of phenol.
BT
BP
BP
φisc ) φisc × ∆ABT × ꢀ525BP/∆A525 × ꢀBT
(5)
where ∆ABT and ∆A525 values refer to the net absorbances
of the BT triplet excited state at 350 or 600 nm and BP triplet
state at 525 nm, respectively. The BP molar absorption
coefficient (ꢀ525BP) and triplet yield (φiscBP) in acetonitrile were
BP
taken to be 6500 M-1 cm-1 and 1, respectively. The BT triplet
BT
yield in MeCN (φ ) 1)25 has also been described. Hence, BT
Con clu sion
isc
molar absorption coefficients of ca. 2900 M-1 cm-1 at 600 nm
and ca. 4800 M-1 cm-1 at 350 nm were obtained. Similar values
were calculated for dichloromethane and methanol as solvents.
Our results demonstrate that photoreduction of BT in
the presence of phenol or indole generates BT ketyl
J . Org. Chem, Vol. 68, No. 13, 2003 5111

Pe´rez-Prieto et al.
On the other hand, molar absorption coefficients of BT ketyl
radical (BTH) were also determined by a comparative method,
using two solutions with the same absorbance (0.3) at 355
nm: BP in MeCN and BT containing indole (4 × 10-4 M) in
dichloromethane (ca. 95% of BT triplet quenching). The
quantum yield for the formation of indolyl radical (φIn) was
calculated by using eq 6:
4-[hydroxy(phenyl)thien-2-ylmethyl]phenol (3), and [phenyl-
(thien-2-yl)]methylidencyclohexa-2,5-dien-1-one (4) (4 mg, 1.4%
in weight).
(b) A degassed acetonitrile solution of BT (188 mg, 1 mmol)
and phenol (94 mg, 1 mmol) in a Pyrex tube was irradiated
for 8 h with a 125-W medium-pressure mercury lamp inside a
quartz immersion well, under continuous magnetic stirring.
Addition of 10 drops of concentrated HCl led to the precipita-
tion of a solid (180 mg) which was filtered and chromato-
graphed (hexane/ethyl acetate, 10/1) to give pinacolone 2 (50
mg, 17.8%) and BT. On the other hand, the filtrate was
evaporated and column-chromatographed (dichloromethane/
hexane, 10/1) to give pinacolone 2 (6 mg, 2.2%), BT and (4-
hydroxyphenyl)-phenyl-(2-thienyl)methane (5, 2 mg, 0.9%).
BP
In
BP
In
φIn ) φisc × ∆A460 × ꢀ525BP/∆A525 × ꢀ460
(6)
where ∆A460 and ꢀ460 (850 M-1 cm-1) are the net absorbance
and the molar absorption coefficient of In radical at 460 nm.
It was assumed that BTH radical does not absorb at 460 nm.
Thus, for φIn a value of ca. 0.95 in dichloromethane was
calculated. Then, since φIn ) φ BTH, eq 7 allowed for the
calculation of the molar absorption coefficients of BTH radical:
In
In
1
1a : H NMR (CDCl3): δ 3.20 (s, 2 H), 6.82-6.86 (m, 2H),
6.94 (bb, 2H), 7.04-7.20 (m, 8H), 7.25-7.28 (m, 4H). 13C NMR
(CDCl3): δ 82.1 (s), 124.9 (d), 125.4 (d), 126.1 (d), 126.4 (d),
126.6 (d), 126.7 (d), 140.8 (s), 147.3 (s).
× ꢀ525BP/∆A525 × ꢀ580
1b: H NMR (CDCl3): δ 3.36 (s, 2 H), 6.82-6.84 (m, 4H),
BP
(BTH+In)
BP
(BTH+In)
1
φBTH ) φisc × ∆A580
7.13-7.17 (m, 8H), 7.27-7.30 (m, 4H). 13C NMR (CDCl3): δ
82.3 (s), 124.6 (d), 125.5 (d), 126.1 (d), 126.2 (d), 126.8 (d), 127.8
(d), 140.1 (s), 147.3 (s).
(7)
where ∆A580 value refers to the net absorbance of the ketyl
and indolyl radicals at 580 nm. Thus, the BTH molar absorp-
tion coefficient was found to be ca. 2100 M-1 cm-1 at 580 nm.
By the same methodology following eq 8, a value of ca. 0.95
was calculated for φBTH in a dichloromethane solution of BT
containing 4 × 10-3 M phenol (ca. 95% of BT triplet quench-
ing):
2: 1H NMR (CDCl3): δ 6.65 (dd, J 1 ) 3.6, J 2 ) 1.1 Hz, 2H),
6.83 (dd, J 1 ) 5.1, J 2 ) 3.6 Hz, 2H), 7.14 7.21 (m, 9H), 7.32 (t,
J ) 7.4, 1H), 7.55 (d, 2H, J ) 7.4 Hz, 2H). 13C NMR (CDCl3):
δ 64.3 (s), 125.0 (d), 125.3 (d), 126.7 (d), 126.8 (d), 127.2 (d),
128.0 (d), 128.1 (d), 129.9 (d), 131.1 (d), 135.9 (s), 142.8 (s),
146.5 (s), 196.5 (s). MS m/z 360 (M+ 89), 283 (21), 255 (100),
221 (36).
3: 1H NMR (CDCl3): δ 2.85 (s, 1H), 5.05 (s, 1H), 6.68 (d,
J ) 3.3 Hz, 1H), 6.76 (d, J ) 8.4 Hz, 2H), 6.91 (dd, J 1 ) 5.1,
J 2 ) 3.6 Hz, 1H), 7.19-7.32 (m, 8H).
BP
(BTH)
BP
(BTH)
φBTH ) φisc × ∆A580
× ꢀ525BP/∆A525 × ꢀ580
(8)
In this case, ∆A580 refers only to the BT ketyl radicals since
phenoxy radicals have no absorption at this wavelength.
Com p u ta tion a l Meth od s. Density functional theory38
calculations have been carried out using the B3LYP or
UB3LYP39 exchange-correlation functional, together with the
standard 6-31G* basis set.40 The geometry optimizations were
carried out using the Berny analytical gradient optimization
method.41 For minimized triplet states, the UB3LYP wave
functions showed no spin contamination ( s2 ca. 2.0). A recent
investigation of Z/E isomerization of polyenes has shown that
DFT methods (UB3LYP/6-31G**) give relaxed triplet energies
that are close to experimental values and compare well with
results from high level ab initio methods.42 Atomic charges
were obtained using the natural bond orbital (NBO) method.43
All calculations were carried out with the Gaussian 98 suite
of programs.44
La m p Ir r a d ia tion of 2-Ben zoylth iop h en e in th e P r es-
en ce of P h en ol. (a) A degassed acetonitrile solution of BT
(188 mg, 1 mmol) and phenol (94 mg, 1 mmol) in a Pyrex tube
was irradiated for 27 h with a 125-W medium-pressure
mercury lamp inside a quartz immersion well, under continu-
ous magnetic stirring. After evaporation of the solvent, the
residue was chromatographed in a silica gel column (hexane/
ethyl acetate, 10/1) and then submitted to semipreparative
HPLC, to give 2-benzoylthiophene (18 mg, 6.4%), 1,2-diphenyl-
1,2-dithien-2-ylethane-1,2-diol (1a and 1b ) (33 mg, 11.8%),
4: 1H NMR (CDCl3): δ 6.35 (dd, J 1 ) 9.3, J 2 ) 2.1 Hz, 1H),
6.48 (dd, J 1 ) 10.0, J 2 ) 2.1 Hz, 1H), 7.09-7.15 (m, 2H), 7.20-
7.22 (m, 1H), 7.27 (d, J ) 6.8 Hz, 2H), 7.39-7.46 (m, 3H), 7.62
(dd, J 1 ) 5.1, J 2 ) 1.1 Hz, 1H), 7.77 (dd, J 1 ) 10.0, J 2 ) 2.7
Hz, 1H). MS m/z 264 (M+ 100), 235 (59), 202 (29). HRMS Calcd
for C17H12OS: 264.0609. Found: 264.0716
5: 1H NMR (CDCl3): δ 4.57 (s, 1H), 5.55 (s, 1H), 6.60-6.61
(m, 1H), 6.70 (d, J ) 8.7 Hz, 2H), 6.86 (dd, J 1 ) 5.1, J 2 ) 3.6
Hz, 1H), 7.01 (d, J ) 8.7 Hz, 2H), 7.12-7.15 (m, 3H), 7.20-
7.23 (m, 2H). MS m/z 266 (M+ 100), 249 (6), 189 (56), 115 (6).
HRMS Calcd for C17H14OS: 266.0765. Found: 266.0753.
La m p Ir r a d ia tion of 2-Ben zoylth iop h en e in th e P r es-
en ce of In d ole. A degassed acetonitrile solution of BT (188
mg, 1 mmol) and indole (117 mg, 1 mmol) in a Pyrex tube was
irradiated for 100 h with a 125-W medium-pressure mercury
lamp inside a quartz immersion well, under continuous
magnetic stirring. After evaporation of the solvent, the residue
was chromatographed in a silica gel column (hexane/ethyl
acetate, 10/1) and then submitted to semipreparative HPLC,
to give indole (22 mg, 7.2%), 1,2-diphenyl-1,2-dithien-2-yl-
ethane-1,2-diol (1a and 1b) (17.5 mg, 5.8%), 2-benzoylth-
iophene (84 mg, 27.7%), and [5-(1H-indol-3-yl)thien-2-yl)]-
(phenyl)methanone (6) (8.5 mg, 2.8%).
6: 1H NMR (CDCl3): 7.22-7.24 (m, 2H), 7.28 (d, J ) 3.9
Hz, 1H), 7.36-7.53 (m, 4H), 7.55 (d, J ) 2.8 Hz, 1H), 7.58 (d,
J ) 3.9 Hz, 1H), 7.82 (d, J ) 8.5 Hz, 2H), 7.97-8.00 (m, 1H),
8.45 (bb, 1H). 13C NMR (CDCl3): δ 111.9 (s), 112.1 (d), 120.4
(38) (a) Parr, R. G.; Yang, W. Density Functional Theory of Atoms
and Molecules; Oxford University Press: New York, 1989. (b) Ziegler,
T. Chem. Rev. 1991, 91, 651.
(39) (a) Becke, A. D. J . Chem. Phys. 1993, 98, 5648. (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(40) Hehre, W. J .; Radom, L.; Schleyer, P. v. R.; Pople, J . A. Ab initio
Molecular Orbital Theory; Wiley: New York, 1986.
(41) (a) Schlegel, H. B. J . Comput. Chem. 1982, 3, 214. (b) Schlegel,
H. B. Geometry Optimization on Potential Energy Surface. In Modern
Electronic Structure Theory; World Scientific Publishing: Singapore,
1994.
(42) (a) Brink, M.; J onson, H.; Ottonson, C.-H. J . Phys. Chem. A
1998, 102, 6513. (b) Abu-Hasanayn, F.; Herstroeter, W. G. J . Phys.
Chem. Phys. A 2001, 105, 1214.
(44) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98, revision A.6; Gaussian, Inc.: Pittsburgh, PA,
1998.
(43) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J . Chem. Phys. 1985,
83, 735.
5112 J . Org. Chem., Vol. 68, No. 13, 2003

Photoreduction of 2-Benzoylthiophene
(d), 121.8 (d), 123.3 (d), 123.7 (d), 123.9 (d), 125. 3 (s), 128.8
(d), 129.4 (d), 132.3 (d), 136.8 (d), 136.9 (s), 139.0 (s), 140.0
(s), 148.6 (s), 188.4 (s). MS m/z 303 (M+ 100), 226 (50), 154
(33), 105 (11). HRMS Calcd for C19H13NOS: 303.071786.
Found: 303.0717.
7: 1H NMR (CDCl3): 1H NMR (CDCl3): δ 5.82 (s, 1H), 6.71-
6.73 (m, 2H), 6.86 (dd, J 1 ) 5.1, J 2 ) 3.2 Hz, 1H), 6.95 (t, J )
6.4 Hz, 1H), 7.10-7.12 (m, 2H), 7.18-7.30 (m, 7H), 7.90 (bb,
1H). 13C NMR (CDCl3): δ 44.4 (d), 111.5 (d), 119.9 (d), 120.1
(d), 120.3 (s), 122.6 (d), 123.9 (d), 124.4 (d), 126.0 (d), 126.9
(d), 127.0 (d), 127.1 (s), 128.8 (d), 128.9 (d), 137.0 (s), 144.2
(s), 148.7 (s). MS m/z 289 (M+ 100), 256 (7), 212 (50), 171 (8).
HRMS Calcd for C19H15NS: 289.0925. Found: 289.0925.
Syn th esis of 5. To a solution of fuchsone 4 (8 mg, 0.03
mmol) in methanol (7 mL) was added sodium borohydride (38
mg, 0.01 mmol). After 25 min, water was added, and the
reaction mixture was extracted with ether and dried over
anhydrous sodium sulfate. Solvent was removed, and the
Ack n ow led gm en t. We thank Generalitat Valenci-
ana and the Spanish Government (Projects GV00-020-
2, BQU2001-2725, BQU2002-00377, BQU2002-01032)
for generous support of this work. We also thank
Fundacio´n J ose´ y Ana Royo and the Spanish MECD for
postdoctoral support to R.E.G. Thanks are also due to
Rafael Ballesteros for his comments on the synthetic
aspects.
1
reaction crude was analyzed by H NMR showing the forma-
tion of 5 in a 30% yield.
Syn th esis of 6. A solution of 3-thien-2-yl-1H-indole37b (597
mg, 3 mmol) and benzoyl chloride (0.3 mL, 3 mmol) in
dichloromethane (10 mL) at - 10 °C was treated during 45
min with AlCl3 (420 mg, 3 mmol), stirred for 12 h at room
temperature and decomposed with ice and HCl, to give an oil
which was column-chromatographed (hexane/ethyl acetate,
5/1) leading to 6 (89 mg, 10% yield).
Syn th esis of 7. A mixture of phenyl(thien-2-yl)methanol35b
(190 mg, 1 mmol) and indole (234 mg, 2 mmol) in acetic acid
(15 mL) was stirred at refluxing temperature for 4 h and then
cooled by means of an ice bath. Subsequently, water was
added, and the reaction mixture was extracted with ether and
dried over anhydrous sodium sulfate. Solvent was removed,
and the crude was chromatographed in a silica gel column
(hexane/ethyl acetate, 5/1), to give 3-[phenyl(thien-2-yl)m-
ethyl]-1H-indole (7) (49 mg, 17%).
Su p p or tin g In for m a tion Ava ila ble: NMR spectra for
compounds 1a , 1b, 2-7 and the B3LYP/6-31G* computed total
energies and Cartesian coordinates for the stationary points
BT(S0), BT(T1), BT‚‚‚HOPh (S0), BT‚‚‚HOPH (T), BTH‚‚‚OPh
(T), BT‚‚‚HIn (S0), BT‚‚‚HIn (T), and BTH‚‚‚In (T) are included.
This material is available free of charge via the Internet at
http://pubs.acs.org.
J O034225E
J . Org. Chem, Vol. 68, No. 13, 2003 5113