A new type of excited-state intramolecular proton transfer: Proton transfer from phenol OH to a carbon atom of an aromatic ring observed for 2-phenylphenol

Article abstract of DOI:10.1021/ja0267831

The photochemical deuterium incorporation at the 2′-and 4′-positions of 2-phenylphenol (4) and equivalent positions of related compounds has been studied in D₂O (CH₃OD)-CH₃CN solutions with varying D₂O (CH₃OD) content. Predominant exchange was observed at the 2′-position with an efficiency that is independent of D₂O (MeOD) content. Exchange at the 2′-position (but not at the 4′-position) was also observed when crystalline samples of 4-OD were irradiated. Data are presented consistent with a mechanism of exchange that involves excited-state intramolecular proton transfer (ESIPT) from the phenol to the 2′-carbon position of the benzene ring not containing the phenol, to generate the corresponding keto tautomer (an o-quinone methide). This is the first explicit example of a new class of ESIPT in which an acidic phenolic proton is transferred to an sp²-hybridized carbon of an aromatic ring. The complete lack of exchange observed for related substrates 6-9 and for planar 4-hydroxyfluorene (10) is consistent with a mechanism of ESIPT that requires an initial hydrogen bonding interaction between the phenol proton and the benzene π-system. Similar exchange was observed for 2,2′-biphenol (5), suggesting that this new type of ESIPT is a general reaction for unconstrained 2′-aryl-substituted phenols and other related hydroxyarenes.

Full text of DOI:10.1021/ja0267831

Published on Web 07/23/2002
A New Type of Excited-State Intramolecular Proton Transfer:
Proton Transfer from Phenol OH to a Carbon Atom of an
Aromatic Ring Observed for 2-Phenylphenol1
Matthew Lukeman and Peter Wan*
Contribution from the Department of Chemistry, Box 3065, UniVersity of Victoria,
Victoria, British Columbia, Canada V8W 3V6
Received May 3, 2002
Abstract: The photochemical deuterium incorporation at the 2′- and 4′-positions of 2-phenylphenol (4)
and equivalent positions of related compounds has been studied in D2O (CH3OD)-CH3CN solutions with
varying D2O (CH3OD) content. Predominant exchange was observed at the 2′-position with an efficiency
that is independent of D2O (MeOD) content. Exchange at the 2′-position (but not at the 4′-position) was
also observed when crystalline samples of 4-OD were irradiated. Data are presented consistent with a
mechanism of exchange that involves excited-state intramolecular proton transfer (ESIPT) from the phenol
to the 2′-carbon position of the benzene ring not containing the phenol, to generate the corresponding keto
tautomer (an o-quinone methide). This is the first explicit example of a new class of ESIPT in which an
acidic phenolic proton is transferred to an sp2-hybridized carbon of an aromatic ring. The complete lack of
exchange observed for related substrates 6-9 and for planar 4-hydroxyfluorene (10) is consistent with a
mechanism of ESIPT that requires an initial hydrogen bonding interaction between the phenol proton and
the benzene π-system. Similar exchange was observed for 2,2′-biphenol (5), suggesting that this new type
of ESIPT is a general reaction for unconstrained 2′-aryl-substituted phenols and other related hydroxyarenes.
Introduction
Such “energy wasting” reactions have been exploited for use
as photostabilizers and other applications.4 The first example
Excited-state intramolecular proton transfer (ESIPT) is a
simple and fundamentally important photochemical process that
has received considerable attention.2,3 For aromatic organic
compounds, ESIPT generally arises when both acidic and basic
functionalities on the molecule experience simultaneous en-
hancement of the acidity and basicity of these groups upon
electronic excitation. This enhancement is often sufficient for
direct protonation of the basic site by the acidic group. In some
cases, the mediation of a protic solvent is required if the distance
between the two sites is too far for direct proton transfer. ESIPTs
generally involve heteroatoms (as part of a heteroaromatic ring)
serving as the basic site and OH or NH acids, either as a
substituent on or part of an aromatic ring system. ESIPT in these
systems is wholly reversible and in most cases extremely fast.
of an ESIPT between a phenol OH and a carbon atom (the
â-carbon of an alkene moiety) was first reported by Yates and
co-workers5 while studying the photohydration reactions of
styrenes and arylacetylenes (eq 1). Photolysis of o-hydroxy-
styrene (1) in aqueous CH3CN gave the hydration product 3
via o-quinone methide 2. The primary photochemical event from
S1 is believed to be ESIPT from the phenol to the â-carbon of
the alkene moiety. Reverse proton transfer is not observed, and
instead 2 is trapped quantitatively by water to give 3.
* To whom correspondence should be addressed. E-mail: pwan@uvic.ca.
Tel: (250) 721-8976. Fax: (250) 721-7147.
(1) Part of this work has appeared in preliminary form: Lukeman, M.; Wan,
P. J. Chem. Soc., Chem. Commun. 2001, 1004.
(2) (a) Ormson, S. M.; Brown, R. G. Prog. React. Kinet. 1994, 19, 45. (b) Le
Gourrierec, D.; Ormson, S. M.; Brown, R. G. Prog. React. Kinet. 1994,
19, 211. (c) Formosinho, S. J.; Arnaut, L. G. J. Photochem. Photobiol., A
1993, 75, 21.
(3) Some representative recent papers: (a) Chou, P.-T.; Liao, J.-H.; Wei, C.-
Y.; Yang, C.-Y.; Yu, W.-S.; Chou, Y.-H. J. Am. Chem. Soc. 2000, 112,
986. (b) Kyrychenko, A.; Herbich, J.; Izydorzak, M.; Wu, F.; Thummel,
R. P.; Waluk, J. J. Am. Chem. Soc. 1990, 112, 11179. (c) Fang, W.-H. J.
Am. Chem. Soc. 1998, 120, 7568. (d) Rios Rodriguez, M. C.; Penedo, J.
C.; Willemse, R. J.; Mosquera, M.; Rodriguez-Prieto, F. J. Phys. Chem. A
1999, 103, 7236. (e) Ameer-Beg, S.; Ormson, S. M.; Brown, R. G.;
Matousek, P.; Towrie, M.; Nibbering, E. T. J.; Foggi, P.; Neuwahl, F. V.
R. J. Phys. Chem. A 2001, 105, 3709. (f) Chou, P.-T.; Chen, Y.-C.; Yu,
W.-S.; Chou, Y.-H.; Wei, C.-Y.; Cheng, Y.-M. J. Phys. Chem. A 2001,
105, 1731. (g) Fischer, M.; Wan, P. J. Am. Chem. Soc. 1999, 121, 4555.
To our knowledge, there are no explicit examples of ESIPT
to a carbon atom that is part of an aromatic ring. In related
work on excited-state proton transfer (ESPT), Webb, Tolbert,
and co-workers6 have suggested that the geminate recombination
process in the ESPT reaction of 1-naphthol could result in
protonation at the C-5 and C-8 ring carbon positions, via solvent-
(4) (a) Kasha, M.; McMorrow, D.; Parthenopoulos, D. A. J. Phys. Chem. 1991,
95, 2668. (b) Kasha, M. J. Chem. Soc., Faraday Trans. 2 1986, 82, 2379.
(c) Kasha, M. Acta Phys. Pol. 1987, 71A, 717.
(5) (a) Isaks, M.; Yates, K.; Kalanderopoulos, P. J. Am. Chem. Soc. 1984,
106, 2728. (b) Kalanderopoulos, P.; Yates, K. J. Am. Chem. Soc. 1986,
108, 6290.
(6) Webb, S. P.; Philips, L. A.; Yeh, S. W.; Tolbert, L. M.; Clark, J. H. J. Am.
Chem. Soc. 1986, 108, 5145.
9
9458
J. AM. CHEM. SOC. 2002, 124, 9458-9464
10.1021/ja0267831 CCC: $22.00 © 2002 American Chemical Society

Excited-State Intramolecular Proton Transfer
A R T I C L E S
mediated ESIPT. This type of ESIPT would not be available in
the absence of a suitable solvent to assist the initial deproto-
nation. In addition, reports of the photoprotonation (by aqueous
acid) of the ring carbons of a variety of aromatic rings (usually
containing an electron-donating group) are well known.7 In
particular, we have shown that hydroxy- and methoxy-
substituted biphenyls are photoprotonated (in mild acid) exclu-
sively on the benzene ring not containing the substituent.7c These
results indicate that the aromatic ring carbons can be sufficiently
basic on photoexcitation to undergo ESPT from moderately
acidic aqueous acid. What is unknown is whether an explicit
ESIPT (from phenol) to an aromatic ring carbon could operate
in appropriately designed systems. We present results for the
photochemical deuterium exchange in 2-phenylphenol (4) and
related compounds that are entirely consistent with an explicit
ESIPT from the phenol moiety to the ring carbon (position 2′)
of the benzene ring not bearing the hydroxyl group, as first
examples of a new type of ESIPT process between phenol and
attached aromatic ring carbons.
Figure 1. 1H (360 MHz) NMR showing the spectrum (expanded aromatic
region) of 4 before (bottom) and after (top) photolysis (1 h, 254 nm) in 1:1
D2O-CH3CN. NMR integration indicates 55% and 15% deuteration at the
2′ (Ha)- and 4′ (Hc)-positions, respectively.
throughout the photolysis, we can reasonably conclude that the
residual decrease in signal of the Hg/Hc signal is exclusively
due to exchange of Hc (4′-proton). All other peaks remain
unchanged, consistent with the absence of deuterium exchange
at these positions on photolysis. Comparison of oxygen- and
argon-bubbled runs showed similar exchange yields at low
conversion (although significant photodecomposition was ob-
served in high conversion oxygen-bubbled runs). Analysis of
MS data confirms the extent of deuteration measured by NMR.
No observable exchange took place in the dark under otherwise
identical conditions.
The readily available 2,2′-biphenol (5) displayed similar
photoexchange behavior. When photolyzed under similar condi-
tions as 4, deuterium exchange was observed at the 6- and 6′-
positions to an extent of 35%, which is marginally lower than
the exchange observed for the equivalent position of 4. Exchange
at the para (4,4′)-position of 5 was not detectable by 1H NMR.
Irradiation of m and p-phenylphenols (6 and 7, respectively)
in 1:1 (v/v) D2O-CH3CN gave no measurable deuterium
incorporation (by NMR), even when extended photolysis times
were employed. Methyl ether analogues 8 and 9 were also
nonreactive. These analogues have no available acidic proton,
or one that is not at the ortho position. Their lack of
photoexchange is consistent with a mechanism of reaction of 4
and 5 that requires an o-phenyl-substituted phenol capable of
undergoing ESIPT to the adjacent ring.
Results and Discussion
Deuterium Exchange. Our studies of excited-state proton-
transfer-initiated reactions of a wide variety of phenol deriva-
tives8 led us to consider the possibility of ESIPT from a phenol
OH to an attached aromatic ring carbon, which led us to examine
the photochemistry of 2-phenylphenol (4), it being the simplest
compound that could undergo the reaction. This compound is
effectively the “aromatic” analogue (replacement of alkene by
phenyl) of o-hydroxystyrene (1), which was shown by Yates
and co-workers5 to undergo efficient ESIPT. Because 4 was
not expected to give an isolable photohydration product, we
sought to follow the anticipated proton transfer by deuterium
exchange (in D2O). Initial photolysis of 4 in 1:1 (v/v) D2O-
CH3CN (argon purged) for 1 h yielded as photoproducts (after
H2O wash) starting material deuterated at the ortho (2′) (4-2′D)-
and para (4′) (4-4′D)-positions (47 and 14% exchange, respec-
tively) of the aromatic ring not containing the hydroxyl group
(eq 2). The exchange was readily observable by 1H NMR (360
MHz), although this method of analysis cannot differentiate
between mono- or dideuteration of the two equivalent 2′-
positions. The reduction in the area of the NMR signal assigned
to the two protons of the 2′-position (δ 7.58, dd) (Ha) on
photolysis (1 h) is accompanied by formation of a broad doublet
(but no reduction in peak area) of the peak assigned to the
neighboring meta (3′) proton (δ 7.38, dd) (Hb) due to coupling
to deuterium that is incorporated at Ha (Figure 1). Protons at
the 2 (Hg)- and 4′ (Hc)-positions were not resolvable and appear
as a multiplet at δ 7.28. The area of this peak is also reduced
on photolysis in D2O although much less efficiently. Because
the signal of Hf (neighboring proton of Hg) remains unchanged
(7) (a) Shizuka, H. Acc. Chem. Res. 1985, 18, 141. (b) Wan, P.; Zhang, G.
Res. Chem. Intermed. 1993, 19, 119. (c) Shi, Y.; Wan, P. J. Chem. Soc.,
Chem. Commun. 1995, 1217.
(8) (a) Wan, P.; Barker, B.; Diao, L.; Fischer, M.; Shi, Y.; Yang, C. Can. J.
Chem. 1996, 74, 465. (b) Wan, P.; Brousmiche, D. W.; Chen, C. Z.; Cole,
J.; Lukeman, L.; Xu, M. Pure Appl. Chem. 2001, 73, 529.
To gain further insights into the exchange mechanism of 4
(and 5), the extent of exchange (by NMR and MS) was
9
J. AM. CHEM. SOC. VOL. 124, NO. 32, 2002 9459

A R T I C L E S
Lukeman and Wan
until very low D2O concentrations were reached. A nominal
concentration of D2O is necessary to exchange the phenolic
protons of the starting material (i.e., convert 4 to 4-OD), as well
as the fact that the CH3CN contains residual H2O. We interpret
the sharp rise (between 0 and 0.25 M D2O) in exchange yield
at the 2′-position observed at low D2O concentrations as arising
from the conversion of 4 to 4-OD, and when 4 is fully converted
to 4-OD, the exchange efficiency is independent of D2O
concentration. The lack of a “bulk” solvent dependence was
also observed when CH3OD was used in place of D2O. Again,
the exchange efficiency did not change with CH3OD concentra-
tion until very small concentrations were reached. The apparent
lack of D2O (CH3OD) concentration dependence is consistent
with exchange at the 2′-position arising from direct proton
transfer from the phenol to the adjacent phenyl ring in the
excited state. Similar “solvent-independent” behavior for ex-
change (at the 6,6′-positions) was observed for 5.
A much different solvent dependence was observed for
exchange occurring at the para (4′)-position of 4 (Figure 3).
Although the efficiency of exchange at the 4′-position is
considerably lower as compared to the exchange at the 2′-
position (vide supra), the amount of exchange rises steadily as
the concentration of D2O is increased (data obtained from
extended photolysis runs). Such behavior strongly indicates that
the mechanism for exchange at this position differs markedly
from that of the 2′-position. It is apparent that “bulk” D2O is
necessary for efficient exchange at the 4′-position. As the 4′-
position is too far from the acidic phenol for direct protonation
to take place, a proton transfer from the phenol to this position
would have to be solvent mediated. In addition, no exchange
was observed at the 4′-position when CH3OD was used as the
cosolvent (unlike the observed behavior at the 2′-position). These
observations suggest that exchange at the 4′-position is due to
reprotonation of the excited-state phenolate ion (at the 4′-
position) by D2O, because water is required for formation of
the excited-state phenolate (vide infra).
Figure 2. Plot of % deuterium exchange at the 2′ (9)- and 4′ (b)-positions
of 2-phenylphenol (4) with photolysis time in 1:3 D2O-CH3CN (as
determined by NMR). At lower conversions, exchange at the 2′-position is
due to monodeuterium incorporation (by MS). Because significant exchange
at the 4′-position is observed only at longer photolysis times, a combination
of 4′-monodeuterated and 2′,4′-dideuterated (and higher) products is
contributing.
Efficient deuterium exchange at the 2′-position of 4 was also
observed when the photolyses were carried out in neat EtOD
or cyclohexane (with 4-OD) with no observed exchange at the
4′-position. Exchange efficiencies in water and alcohols were
similar, whereas the exchange is considerably more efficient in
cyclohexane (about 2-fold more efficient as compared to that
in 1:1 (v/v) D2O-CH3CN). The latter observation suggests that
the presence of water or other protic solvents hinders ESPT to
the 2′-position. This is consistent with intrinsic ESIPT from the
phenol to the 2′-position as responsible for the observed
exchange. Additional evidence for this mechanism was the
finding that photolysis of powdered crystals of 4-OD resulted
in exclusive deuterium incorporation at the 2′-position (∼10%).
Because crystalline samples of 4 grown in H2O did not contain
included solvent, the exchange observed in the solid state result
is further evidence of an intrinsic ESIPT mechanism.
Figure 3. Plot of total % deuterium incorporation at the 2′ (9)- and 4′-
positions (b) of 4 versus D2O concentration (in CH3CN), as measured by
1H NMR. Because of the lower efficiency of exchange at the 4′-position,
photolysis times were tripled for these experiments to increase conversion.
Inset: expansion of data for exchange at the 2′-position at low D2O
concentration.
monitored as a function of photolysis time (Figure 2). The data
show that exchange at the 2′-position is significantly more
efficient than exchange at the 4′-position at all conversions.
Higher conversion runs led to significant proportions of 2′,2′-
and 2′,4′-dideuterated and small amounts of trideuterated
products. Quantitative exchange of the 2′-positions can be
achieved with extended photolysis (not shown in Figure 2).
It is well known9 that bulk water is generally required for
the efficient ESPT of phenols and other hydroxyarenes (to
solvent). Thus, decreasing the water (D2O) content of the solvent
system should retard ESPT and water-assisted ESIPT processes.
To explore the effect of water content on the deuterium
incorporation reaction of 4 and 5, exchange studies were carried
out in which the D2O concentration (in CH3CN) was varied.
The extent of exchange at the 2′- and 4′-positions of 4 as a
function of D2O content is shown in Figure 3. Decreasing the
concentration of D2O present in the solvent mixture led to no
measurable decrease in exchange efficiency at the 2′-position
Molecular modeling (Chem 3D, AM1 optimization) and
crystallographic data10 show that the ground-state structure of
4 is twisted with a dihedral angle between the aromatic rings
(9) (a) Moore, R. A.; Lee, J.; Robinson, G. W. J. Phys. Chem. 1985, 89, 3648.
(b) Lee, J.; Robinson, G. W.; Webb, S. P.; Philips, L. A.; Clark, J. H. J.
Am. Chem. Soc. 1986, 108, 6538. (c) Robinson, G. W. J. Phys. Chem.
1991, 95, 10386. (d) Tolbert, L. M.; Haubrich, J. E. J. Am. Chem. Soc.
1994, 116, 10593.
(10) Perrin, M.; Bekkouch, K.; Thozet, A. Acta Crystallogr., Sect. C 1987, 43,
980.
9
9460 J. AM. CHEM. SOC. VOL. 124, NO. 32, 2002

Excited-State Intramolecular Proton Transfer
A R T I C L E S
Figure 5. Fluorescence emission spectra of 4 and 10 in various solvents.
The emission at 420 nm observed only in water is assigned to phenolate
13.
Figure 4. Geometry optimized (Chem3D, AM1) structures of 4 and 10.
The dihedral angles between the aromatic rings (in the ground state) for 4
and 10 are 54° and 0°, respectively.
S-orbital of the acidic proton and the accepting π-system prior
to excitation.
Table 1. Photophysical and Photochemical Parameters
compound
Φex
Φfb
τc (ns)
pK*d
a
Quantum Yields of Deuterium Exchange. Quantum yields
for deuterium incorporation (Φex) for 4 and 5 were measured
using the deuterium exchange reaction of dibenzosuberene as
a secondary reference standard. Dibenzosuberene is known to
undergo deuterium exchange of the benzylic C-H hydrogens
4
5
6
7
0.041
0.034
0.32
0.28
0.35
0.20
0.65
1.9 (0.5)
1.8
4.5 (1.5)
7.7 (1.2)
3.5
1.15 ( 0.04
<0.001e
<0.001e
<0.001e
1.36 ( 0.02
1.83 ( 0.04
10
when irradiated in 1:1 D2O-CH3CN (Φ ) 0.030).13
Φex at the
a Quantum yield for deuterium incorporation in 1:3 D2O-CH3CN at the
2′-position of 4 and 6-position of 5. Estimated error (10% of quoted value.
Measured relative to the quantum yield for deuterium incorporation reported
for dibenzosuberene.13 b Fluorescence quantum yield in neat CH3CN.
Measured relative to the reported fluorescence quantum yield for fluorene
(Φf ) 0.68).14 Estimated error of (10% of quoted value. c Fluorescence
lifetime in neat CH3CN measured by single photon counting. Values in
brackets are the fluorescence lifetimes in water, from ref 15. Lifetimes
measured in this way have an estimated error of (0.2 ns. d pKa for the
excited singlet state, from ref 15. e No observable exchange by NMR under
conditions employed.
2′-position of 4 and the 6-position of 5 (measured in 1:3 D2O-
CH3CN) were 0.041 ( 0.004 and 0.034 ( 0.004, respectively.
For 4, the quantum yield of exchange at the para position was
much lower (less than 0.01) and hence not reliably measurable
using this technique. The quantum yield for the actual photo-
protonation step should be higher than the observed quantum
yield of exchange because the intermediate obtained from ESIPT
can also lose deuterium to regenerate nondeuterated starting
material.
Fluorescence Measurements. We have recently shown3g that
formal ESIPT (via S1) from phenol to side-chain carbon atoms
(of an alkene moiety) is mediated by water, where the use of
steady-state and time-resolved fluorescence studies was essential
for mechanistic characterization. In these systems,3g water is
essential for reaction, acting as a bridge for the formal transfer
of a phenol proton to the â-carbon of the alkene moiety;
otherwise there is no ESIPT. The results obtained so far for
deuterium exchange of 4 and 5 indicate that water is not essential
for exchange at the 2′ (6)-positions, but is required for exchange
at the 4′-position of 4. Fluorescence measurements should
provide additional insights into the requirement of water on
reactivity. Strong fluorescence emission (λmax ) 330 nm; Φf )
0.32 ( 0.03) was observed for 4 in neat CH3CN (Figure 5).
The addition of ∼10 M water caused minimal quenching of
this band (about 10%). Above 10 M water, the intensity of this
band was progressively reduced, along with concurrent growth
of a second emission band (λmax ) 420 nm) which was assigned
to the phenolate (by comparison with emission of authentic
phenolate ion). The fluorescence emission is essentially un-
changed in neat CH3OH (λmax ) 335 nm, Φf ) 0.32 ( 0.03)
as compared to neat CH3CN, with no observable phenolate
of ∼54° (Figure 4). In this conformation, the hydroxyl group
can engage in significant interaction with the π-system of the
other ring.11 Indeed, there is considerable evidence that such
an interaction exists for 4 and derivatives.10,12 As a preliminary
investigation into the geometric requirements of the ESIPT of
4 and 5, 4-hydroxyfluorene (10) was synthesized and studied.
Molecular modeling indicated that 10, unlike 4, is planar in the
ground state, with the OH group lying within the plane of the
ring with no possibility of interaction between the phenol OH
and the aromatic π-system (Figure 4). Photolysis of 10 in 1:1
(v/v) D2O-CH3CN gave no detectable deuterium incorporation,
even on prolonged irradiation, with quantitative recovery of
substrate. This result was somewhat unexpected because 10 has
all of the apparent structural requirements that 4 and 5 possess
with the added bonus of a longer singlet lifetime due to the
rigid fluorene backbone (Table 1). Its complete lack of reactivity
supports the notion that ESIPT between the phenol OH and the
2′-position requires some degree of orbital overlap between the
(11) This type of aromatic π-H interaction (in the ground state) has been
investigated in some detail, for example: Tarakeshwar, P.; Choi, H. S.;
Kim, K. S. J. Am. Chem. Soc. 2001, 123, 3323.
(12) (a) Oki, M.; Iwamura, H. J. Am. Chem. Soc. 1967, 89, 576. (b) Schaefer,
T.; Wildman, T. A.; Sebastian, R.; McKinnon, D. M. Can. J. Chem. 1984,
62, 2692.
(13) Budac, D.; Wan, P. J. Org. Chem. 1992, 57, 887.
9
J. AM. CHEM. SOC. VOL. 124, NO. 32, 2002 9461

A R T I C L E S
Lukeman and Wan
Scheme 1
emission. The absence of phenolate emission in methanol
indicates that this solvent is not sufficiently polar to mediate
proton transfer from the phenol of 4 (to solvent). Therefore,
the exchange observed exclusively at the 2′-position of 4 on
photolysis in neat CH3OD (vide supra) must arise from the
protonated form of 4, via ESIPT. A mechanism of exchange at
the 4′-position of 4 involving protonation of the phenolate is
supported by the fact that both the exchange at the 4′-position
and the formation of phenolate require significant amounts of
water.
Biphenol analogue 5 exhibited fluorescence behavior similar
to that observed for 4. The emission at 348 nm (in neat CH3-
CN) is reduced on addition of water with concomitant growth
of the phenolate emission band at 397 nm. The strong
fluorescence emission (λmax ) 320 nm, Φf ) 0.65 ( 0.07) of
10 (in neat CH3CN) was also reduced by the addition of water
with observation of phenolate emission at 390 nm. The
fluorescence quantum yields of 10 and for the parent fluorene
(Φf ) 0.68)14 are the same (within experimental error),
indicating that the presence of a hydroxyl group to the 4-position
apparently does not introduce any additional deactivational
pathway, consistent with the lack of observable exchange.
Fluorescence lifetimes for 4, 6, and 7 were measured in neat
CH3CN (Table 1) and were found to be 1.9, 4.5, and 7.7 ns,
respectively. Townsend and Schulman15 observed that the
fluorescence lifetime (in water) is shorter for 4 than for isomeric
phenylphenols 6 and 7 (Table 1), and they attributed the
difference to the higher efficiency of ESPT to water for 4 (more
acidic in S1). The much shorter lifetime observed for 4 as
compared to isomers 6 and 7 in neat CH3CN is not surprising
given that only 4 undergoes ESIPT.
) 520 nm, τ ≈ 50 ns) was observed for 4. However, transients
of similar intensity, λmax, and lifetime were also observed for
“unreactive” analogues 7 and 8. Quinone methides 14a and 14b
would be expected to absorb in the 500-600 nm region with
anticipated lifetimes that are significantly shorter than those
observed for related analogues 11 and 12. This is because of
the fact that both 14a and 14b can readily tautomerize, noting
that resonance forms such as 15a would expedite such a
mechanism, whereas 11 and 12 can only undergo nucleophilic
capture by water. The weakness of the transient as well as
observation of residual signals in the same region from both 7
and 8 leads us to conclude that it is premature to assign the
observed broad 520 nm transient as due to quinone methide
14a or 14b at this time.
Laser Flash Photolysis (LFP). A wide range of quinone
methide intermediates generated from phenol derivatives has
been characterized8 in our laboratory by nanosecond LFP,
including biphenyl quinone methides7c 11 (λmax ) 570 nm, τ
) 400 ns) and 12 (λmax ) 525 nm, τ ) 67 µs) in aqueous CH3-
CN. Exchange of 4 at the 2′-position via ESIPT (of 4-OD) would
give o-quinone methide 14a (or 14a-D). Similarly, exchange
at the 4′-position via initial ESPT to solvent H2O, followed by
reprotonation, would lead to the isomeric o-quinone methide
14b (or 14b-D) (Scheme 1). Both 14a and 14b are structurally
similar to 11 and 12, and hence we anticipated that they will
have visible bands in the 500-600 nm region, as was observed
for 11 and 12. The range in lifetimes (in water) for quinone
methides studied to date vary from stable (>minutes) to
subnanosecond (undetectable using nanosecond LFP). The point
to be addressed in the next section is whether 14a or 14b is
detectable by LFP.
LFP (266 nm) of 4, 6, and 7 (N2 purged) in neat CH3CN,
1:1 H2O-CH3CN, or cyclohexane gave a strongly absorbing
transient (λmax ≈ 350 nm, τ ≈ 1 µs) whose lifetime was
significantly reduced in the presence of oxygen. This transient
is most likely the triplet state of these phenols, although radical
intermediates (probably derived from two-photon processes)
cannot be excluded. The triplet state of 3-cyanophenol has been
reported at λmax ≈ 325 nm in EtOH.16 In neat cyclohexane or
neat CH3CN (oxygen purged), a broad short-lived transient (λmax
Mechanisms of Deuterium Incorporation. The proposed
mechanisms for the deuterium incorporation at the 2′ and 4′-
positions of 4 are outlined in Scheme 1. Singlet state reactivity
is inferred from the correlation of fluorescence data with
observed exchange, the lack of oxygen quenching of exchange,
and the known enhanced excited singlet state acidity of
phenylphenols, although residual triplet state reactivity cannot
be excluded. ESIPT cannot be readily distinguished from an
equivalent hydrogen transfer process, although this is a moot
point, especially in those systems where the reverse transfer is
very fast, and where no net deuterium incorporation is possible
(such as between heteroatoms). We argue that the ESIPT
observed in this work is best described as being true proton
transfer on the basis of the following points: (i) the phenolic
proton of phenylphenols is known to be quite acidic in S1 (pK*
< 2) (Table 1); (ii) hydroxy- and methoxy-substituted biphenyls
are photoprotonated (in mild acid) exclusively on the benzene
ring not containing the substituent;7c (iii) examination of HOMO
and LUMO coefficients of phenylphenols using simple Hu¨ckel-
type calculations predicts that the benzene ring not containing
the hydroxy substituent becomes more basic on excitation; (iv)
the exchange reaction is observed only for those o-phenyl-
phenols that are twisted in the ground state (i.e, 4 and 5 but not
for 10) which allows for significant OH-π hydrogen bonding,
further polarizing the O-H bond.
In the proposed mechanism for deuterium exchange (Scheme
1), singlet excited 4-OD (4-OD*) undergoes ESIPT to the 2′-
position of the other aromatic ring to form o-quinone methide
14a-D. This intermediate can then undergo reverse proton
(14) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photochemistry,
2nd ed.; Marcel Dekker: New York, 1993.
(15) Townsend, R.; Schulman, S. Chim. Oggi 1992, 10, 49.
(16) Bonnichon, F.; Richard, C. J. Photochem. Photobiol., A 1998, 119, 25.
9
9462 J. AM. CHEM. SOC. VOL. 124, NO. 32, 2002

Excited-State Intramolecular Proton Transfer
A R T I C L E S
from a bright orange to an olive green. Stirring was continued for a
further 90 min, at which time the olive color disappeared leaving a
deep orange solution. This solution was added dropwise (over 25 min)
to a boiling solution of 1:1 (v/v) concentrated H2SO4-H2O (80 mL).
After addition, the mixture was boiled for a further 15 min and then
allowed to cool. The solution was extracted with 3 × 50 mL of CH2-
Cl2, and the combined organic layers were extracted with 5 × 150 mL
of 0.1 M NaOH. The crude 4-hydroxyfluorenone was precipitated by
acidification (10% HCl) of the aqueous extract, and filtered and air-
dried (260 mg, 52%). The crude 4-hydroxyfluorenone (0.25 g, 1.3
mmol) was dissolved in 20 mL of EtOH to which was added 75 mg of
Pd/C. The solution was stirred for 24 h under 35 atm of H2. Filtration
followed by evaporation of the solvent gave a yellow-orange solid,
which was recrystallized from 95% EtOH, to give pure 10 (200 mg,
45% overall yield), mp 105-106 (lit. 108-109).17 1H NMR (CDCl3):
δ 3.91 (2H, s), 5.25 (broad s, exchangeable in D2O), 6.73 (1H, m),
7.13 (2H, m), 7.27 (1H, t, J ) 7.4 Hz), 7.37 (1H, dd, J ) 7.4, 7.4 Hz),
7.51 (1H, d, J ) 7.4 Hz), 8.10 (1H, d, J ) 7.4 Hz). 13C NMR
(CDCl3): δ 37.3, 113.7, 117.6, 123.5, 124.4, 125.9, 126.8, 127.5, 128.6,
140.6, 142.5, 145.7, 151.5.
2-Methoxybiphenyl (8). To a solution of 4 (5.1 g, 30 mmol) in
DMSO (50 mL) was added KOH (6.7 g, 120 mmol) and CH3I (6.5 g,
45 mmol). The mixture was stirred at room temperature for 5 h, at
which time the reaction was quenched by the addition of water (100
mL). This mixture was extracted with 2 × 50 mL of CH2Cl2 which
was subsequently washed with 4 × 100 mL of water. The organic layer
was dried (MgSO4), and the solvent was removed under vacuum. The
crude product (yellowish oil) was purified by bulb-to-bulb distillation,
to give a colorless oil (4.3 g, 78%). 1H NMR (CDCl3): δ 3.83 (3H, s),
7.04 (2H, m), 7.35 (3H, m), 7.44 (2H, dd, J ) 7.4, 7.4 Hz), 7.57 (2H,
d, J ) 7.4 Hz). 13C NMR (CDCl3): δ 55.6, 111.3, 120.9, 127.0, 128.0,
128.7, 129.6, 130.8, 138.6, 156.5.
transfer to generate 4-2′D. The reverse reaction could also be
viewed as a [1,5]-hydrogen shift, although if one considers that
the resonance form 15a should be a strong contributor to the
overall o-quinone methide structure, the reverse reaction may
be best viewed simply as a proton transfer. Studies involving
photolyses in solvents of varying D2O content in CH3CN and
in CH3OH, in neat cyclohexane, and in the solid state indicate
that the initial proton transfer to the 2′-position does not require
water.
ESPT to solvent from 4* (or 4-OD*) to generate 13 becomes
increasingly efficient as the water content of the solvent system
is increased, as evident from fluorescence studies (Figure 5;
observation of increasing phenolate ion emission). Protonation
of 13 by H2O (D2O) at the 4′-position will lead to formation of
quinone methide intermediate 14b (14b-D) which would result
in deuterium incorporation at the 4′-position on deprotonation
(to give 4-4′D). While protonation of 13 at the 2′-position might
also be occurring to some extent in solvent systems containing
significant amounts of H2O (D2O), it is not possible to determine
the contribution of this mechanism to the overall exchange
observed at the 2′-position at high water content.
Proper relative geometric orientation of the phenol group and
the accepting ring carbon is an important requirement for ESIPT
in 4. Ground-state hydrogen bond interactions between the donor
and acceptor groups are generally required for systems to
undergo efficient direct ESIPT reactions.2 For unreactive
analogue 10, the phenolic proton is constrained within the nodal
plane of the π-system resulting in no net overlap with the
π-system of the adjacent benzene ring. It appears that ESIPT
to the adjacent ring cannot occur unless the s-orbital of the
phenolic proton can sufficiently overlap with the basic π-orbitals
prior to proton transfer. The twisted ground-state geometries
of 4 and 5 allow for such an interaction and hence are both
able to undergo the observed ESIPT.
2,2′-Dimethoxybiphenyl (9). In a similar procedure as above starting
1
from 5, 9 was isolated in 85% yield (5.46 g). H NMR (CDCl3): δ
3.77 (6H, s), 6.98 (4H, m), 7.24 (2H, d, J ) 7.3 Hz), 7.32 (2H, dd, J
) 7.3, 7.3 Hz). 13C NMR (CDCl3): δ 55.7, 111.1, 120.3, 127.8, 128.6,
131.5, 157.0.
Product Studies. Solutions were prepared in a quartz vessel and
purged with argon (or O2) gas for 10 min prior to and continuously
during irradiation in a Rayonet RPR 100 photochemical reactor using
254 nm lamps. Cooling was achieved with a coldfinger (∼15 °C).
Following photolysis, the solution was extracted with CH2Cl2, washed
with water, and dried over MgSO4.
This work has revealed that a significant photochemical
pathway for 4 (and 5, and possibly many related compounds)
is ESIPT from the phenol proton to the 2′-carbon of the adjacent
benzene ring. This new type of ESIPT might be a general
photochemical pathway for many unconstrained o-aryl phenols
and o-aryl hydroxyarenes. As such functionalities are ubiquitous
in organic chemistry, the identification of this new pathway will
provide new insights into the photochemical behavior of many
other related systems providing topics for continued studies in
this area.
Photolysis of 4. A 15 mg sample of 4 was irradiated for 60 min in
1:3 D2O-CH3CN (40 mL), which cleanly gave 4 with 55% deuterium
incorporation at the 2′-position and 15% deuterium incorporation at
the 4′-position (42% overall exchange of all possible 2′- and 4′-positions
1
(3 in total)) (by 360 MHz H NMR). MS (after 13C correction) also
indicates an overall exchange of 42%, arising from a product distribution
of 4 (20%), 4-D (42%), 4-D2 (32%), and 4-D3 (6%). A second 15 mg
sample of 4 was irradiated for 5 min in 1:3 D2O-CH3CN (40 mL) and
cleanly gave 4 with 7% deuterium incorporation at the 2′-position and
4% deuterium incorporation at the 4′-position (5.8% overall exchange)
(by 360 MHz 1H NMR). MS (after 13C correction) indicates an overall
exchange of 5.0%, arising from a product distribution of 4 (86%), 4-D
(13%).
Experimental Section
1
General. H NMR spectra were recorded on Bruker AC 300 (300
MHz) and AM 360 (360 MHz) instruments. MS were recorded on a
Kratos Concept H spectrometer using the LSIMS ionization method
using m-nitrobenzyl alcohol. CH3CN was dried over CaH2 and distilled
prior to use. Other solvents were spectral grade from Aldrich and used
as received.
Photolysis of 5. Irradiation of a 15 mg sample of 5 dissolved in 1:1
D2O-CH3CN (40 mL) for 60 min cleanly gave 5 with deuterium
incorporated at the 6-position to an extent of 35% (by NMR).
Photolysis of 6-10. Solutions of each of 6-10 were prepared (1:1
D2O-CH3CN), and each was irradiated for 60 min. No deuterium
incorporation could be observed by 1H NMR (360 MHz) for any
derivative.
Materials. Phenylphenols 4-7 were purchased from Aldrich and
recrystallized before use. 4-Hydroxyfluorene (10) was synthesized from
commercially available 4-aminofluorenone (Aldrich). Methyl ethers 8
and 9 were prepared from 4 and 5, respectively, via Williamson ether
synthesis.
4-Hydroxyfluorene (10). 4-Aminofluorenone (500 mg, 2.5 mmol)
was suspended in 15 mL of water (ice bath) to which ∼20 mL of
concentrated H2SO4 was added, at which point all of the solid dissolved.
A solution of NaNO2 (0.22 g, 3.2 mmol) in 10 mL of water was added
dropwise over 2 min with stirring. The color of the solution turned
Laser Flash Photolysis (LFP). All transient spectra and lifetimes
were obtained using a Nd:YAG laser (Spectra Physics Quanta-Ray,
(17) Pan, H.; Fletcher, T. L. J. Org. Chem. 1960, 25, 1106.
9
J. AM. CHEM. SOC. VOL. 124, NO. 32, 2002 9463

A R T I C L E S
Lukeman and Wan
GCR, <30 mJ) with a pulse width of 10 ns and excitation wavelength
of 266 nm. Flow cells (7 mm path length) were used, and solutions
were purged with either nitrogen or oxygen for 10 min prior to
excitation. ODs at 266 nm were ∼0.3.
fluorene (Φf ) 0.68).14 Fluorescence lifetimes were measured on a PTI
LS-1 instrument using the time-correlated single-photon-counting
technique, and all decays were first order.
Acknowledgment. Support of this work was provided by the
Natural Sciences and Engineering Research Council (NSERC)
of Canada. M.L. thanks NSERC for a post-graduate scholar-
ship.
Steady-State and Time-Resolved Fluorescence Measurements.
All solutions (OD ≈ 0.1 at excitation wavelength) were purged with
argon for 5 min before analysis. Steady-state fluorescence spectra were
obtained using a Photon Technology International A-1010 (PTI) Quanta-
Master luminescence spectrometer. Fluorescence quantum yields were
determined by comparing the integrated emission bands with that of
JA0267831
9
9464 J. AM. CHEM. SOC. VOL. 124, NO. 32, 2002