Competing excited state intramolecular proton transfer pathways from phenol to anthracene moieties

Article abstract of DOI:10.1021/jo0524728

Four new 9-(2′-hydroxyphenyl)anthracene derivatives 7-10 were synthesized and their potential excited state intramolecular proton transfer (ESIPT) reaction investigated. Whereas 7 reacted via the anticipated (formal) ESIPT reaction (proton transfer to the 10-position of the anthracene), derivatives 8-10 reacted via ESIPT to both 9- and 10-positions, giving rise to two types of intermediates, quinone methides (e.g., 29) and zwitterions (e.g., 30). These intermediates are trapped by solvent (water or methanol) giving addition products that can readily revert back to starting material. However, on extended photolysis, the products that are isolated can best be rationalized as being due to competing elimination and intramolecular cyclization of zwitterions 30 and 37. These results show that it is possible to structurally tune ESIPT in (hydroxyphenyl)anthracenes to either result in a completely reversible reaction or give isolable anthracene addition or rearrangement products.

Full text of DOI:10.1021/jo0524728

Competing Excited State Intramolecular Proton Transfer Pathways
from Phenol to Anthracene Moieties
Nikola Basaric´† and Peter Wan*
Department of Chemistry, Box 3065, UniVersity of Victoria, British Columbia, Canada, V8W 3V6
pwan@uVic.ca
ReceiVed NoVember 30, 2005
Four new 9-(2′-hydroxyphenyl)anthracene derivatives 7-10 were synthesized and their potential excited
state intramolecular proton transfer (ESIPT) reaction investigated. Whereas 7 reacted via the anticipated
(formal) ESIPT reaction (proton transfer to the 10-position of the anthracene), derivatives 8-10 reacted
via ESIPT to both 9- and 10-positions, giving rise to two types of intermediates, quinone methides (e.g.,
29) and zwitterions (e.g., 30). These intermediates are trapped by solvent (water or methanol) giving
addition products that can readily revert back to starting material. However, on extended photolysis, the
products that are isolated can best be rationalized as being due to competing elimination and intramolecular
cyclization of zwitterions 30 and 37. These results show that it is possible to structurally tune ESIPT in
(hydroxyphenyl)anthracenes to either result in a completely reversible reaction or give isolable anthracene
addition or rearrangement products.
Introduction
basic site.7 These ESIPT reactions require initial hydrogen
bonding interaction between the phenol OH and the π-system
of the adjacent ring that is possible only in twisted biaryl
structures.7a In the excited singlet state, the phenolic proton
becomes sufficiently acidic to facilitate proton transfer to a
carbon atom of an adjacent aromatic ring, giving rise to quinone
methide (QM) intermediates (as shown in eq 1 for the parent
system, 2-phenylphenol (1)7a). In the (hydroxyphenyl)anthracene
(HPA) system 4, water-mediated proton transfer to position 10
of the anthracene ring gives quinone methide 5. The later is
Excited state intramolecular proton transfer (ESIPT) is a
fundamental process that has received considerable attention1
since its first discovery by Weller.2 ESIPT arises when a
molecule contains both an acidic and a basic site which upon
electronic excitation experience an enhancement in acidity or
basicity, respectively. In most examples of ESIPT processes an
OH or NH is the acidic site, while a basic site usually contains
a heteroatom such as heterocyclic nitrogen or a carbonyl group.
Since most of the ESIPT processes are reversible, these systems
have found applications as photostabilizers,3 laser dyes,4 scin-
tillators,5 and solar collectors.6
(3) (a) Stu¨ber, G. J.; Kieninger, M.; Schettler, H.; Busch, W.; Go¨ller,
B.; Frnke, J.; Kramer, H. E. A.; Hoier, H.; Henkel, S.; Fischer, P.; Port,
H.; Hirsch, T.; Rytz, G.; Birbaum, J.-L. J. Phys. Chem. 1995, 99, 10097.
(b) Keck, J.; Kramer, H. E. A.; Port, H.; Hirsch, T.; Fischer, P.; Rytz, G.
J. Phys. Chem. 1996, 100, 14468.
We recently reported the first examples of ESIPT between
phenol as an acidic site and an aromatic carbocyclic ring as a
(4) (a) Chou, P.-T.; McMorrow, D.; Aartsma, T. J.; Kasha, M. J.
Phys. Chem. 1984, 88, 4596. (b) Parthenopoulos, D. A.; McMorrow, D.
P.; Kasha, M. J. Phys. Chem. 1991, 95, 2668. (c) Kasha, M. Acta Phys.
Pol. 1987, 71A, 717. (d) Park, S.; Kwon, O.-H.; Kim, S.; Park, S.; Choi,
M.-G.; Cha, M.; Park, S. Y.; Jang, D.-J. J. Am. Chem. Soc. 2005, 127,
10070.
(5) Pla-Dalmau, A.; Bross, A. D. Mater. Res. Symp. Proc. 1994, 148,
163.
(6) Vollmer, F.; Rettig, W. J. Photochem. Photobiol., A 1996, 95, 143.
(7) (a) Lukeman, M.; Wan, P. J. Am. Chem. Soc. 2002, 124, 9458. (b)
Lukeman, M.; Wan, P. J. Am. Chem. Soc. 2003, 125, 1164. (c) Flegel, M.;
Lukeman, M.; Huck, L.; Wan, P. J. Am. Chem. Soc. 2004, 126, 7890.
* Address correspondence to this author.
† On leave from the Ruder Bosˇkovic Institute, Zagreb, Croatia.
(1) (a) Ireland, J. F.; Wyatt, P. A. H. AdV. Phys. Org. Chem. 1976, 12,
131. (b) Klo¨pffer, W. AdV. Photochem. 1977, 10, 311. (c) Arnaut, L. G.;
Formosinho, S. J. J. Photochem. Photobiol., A 1993, 75, 1. (d) Formosinho,
S. J.; Arnaut, L. G. J. Photochem. Photobiol., A 1993, 75, 21. (e) Ormson,
S. M.; Brown, R. G. Prog. React. Kinet. 1994, 19, 45. (f) Le Gourrierec,
D.; Ormson, S. M.; Brown, R. G. Prog. React. Kinet. 1994, 19, 211. (g)
Tolbert, L. M.; Solntsev, K. M. Acc. Chem. Res. 2002, 35, 19. (h) Agmon,
N. J. Phys. Chem. A 2005, 109, 13.
(2) Weller, A. Z. Elektrochem. 1956, 60, 1144.
10.1021/jo0524728 CCC: $33.50 © 2006 American Chemical Society
Published on Web 03/09/2006
J. Org. Chem. 2006, 71, 2677-2686
2677

Basaric´ and Wan
trapped by water giving anthracene addition product 6 in
reactions were carried out in toluene/methanol mixture in the
presence of K2CO3 as base and Pd(PPh3)4 as catalyst. In some
reactions palladium-catalyzed debromination took place, result-
ing in slightly lower yields of arylated products (formation of
15, 18, and 22). However, all the couplings were achieved in
good to high yields (>60%).
relatively high yield (eq 2).7c
Product Studies. All the HPA derivatives 7-10 show typical
anthracene absorption with a strong UV-vis absorption band
at 250 nm and a weaker but characteristic structured absorption
band between 300 and 400 nm. We first examined their
photoreactivity by monitoring for changes in the UV-vis
absorption spectra upon photolysis. Compounds were irradiated
in argon-purged 1:1 H2O-CH3OH or H2O-CH3CN in UV
cuvettes at 350 nm. Photolyses of 7-9 were characterized by a
decrease in anthracene absorption at all wavelengths (Figure
1a), which was somewhat faster in H2O-CH3OH than in H2O-
CH3CN. If the photolyses were carried out at sufficiently low
conversion, the decrease in absorbance was reversible. For
example, the recovery of absorbance at 375 nm after photolysis
is shown for 9 in 1:1 H2O-CH3CN (Figure 1b). The rate of
recovery was faster in H2O-CH3CN (several minutes) than in
H2O-CH3OH (>10 min). These observations suggest that
photolysis of those HPA’s results in formation of anthracene
addition products (attack of H2O or CH3OH) that are thermally
unstable (unlike 6 in eq 2). Under similar photolysis conditions
as for 7-9, 10 showed no observable bleaching of UV-vis
absorption. This could imply that 10 is photochemicaly inert
or that any addition product formed reverts back to starting
material very quickly.
In this study we further explore the potential of ESIPT in
other HPA systems. Formation of quinone methides from HPA’s
is of special interest since it can be facilitated by light of longer
wavelengths (350-400 nm), where the anthracene moiety
absorbs. Hence, HPA’s may find potential use in biological
systems (e.g., as DNA alkylating agents8), since anthracenes
can be excited at sufficiently long wavelengths without interfer-
ence of absorption of light by most biological molecules. Four
new HPA derivatives 7-10 were synthesized and their ESIPT
To characterize potential irreversibly formed products, pre-
parative irradiations were performed. Thus, on irradiation of
HPA derivative 8 in 1:4 H2O-CH3CN, products 23-25
(Scheme 2) were detected by NMR, formed in ∼20%, ∼10%,
and ∼30% yield, respectively, after 4 h of photolysis with ∼50-
60% conversion of the starting material. Products 23 and 24
were isolated by TLC on silica gel, while 25 proved to be
unstable and readily reverted back to 8. Although we were not
able to isolate hydration products cis- and trans-25 in pure form,
their structures were assigned according to the characteristic
chemistry investigated. Compound 7 is different from the parent
anthracene derivative 4, by bearing an additional electron
donating and potentially acidic substituent (OH group). All of
HPA derivatives 8-10 have a substituent at position 10 of the
anthracene ring. This was designed to explore the effect of
increased steric hindrance (and extended conjugation in 9 and
10) on the anticipated ESIPT to position 10. One additional
question we wanted to address was whether ESIPT can be
directed to two or more different anthracene ring positions.
1
signals present in the H NMR spectrum of the crude photo-
chemical mixture (quartets at δ ∼4.25, doublets at 1.65, and
singlets at 9.10 and 10.00 ppm corresponding to the hydrogen
bonded phenolic OH protons). Hydration products cis- and
trans-26 were not detected; they may have been formed but
just too reactive for isolation under our conditions. Since 26 is
not stabilized by intramolecular hydrogen bonding (as in 25),
it would readily undergo elimination of water giving back
starting material.
Results and Discussion
Synthesis. The HPA derivatives 7-10 were prepared from
the corresponding methoxy compounds (15, 16, 20, and 22,
respectively) by treatment with BBr3. The methoxy derivatives
were synthesized by Suzuki biaryl coupling reactions and their
preparations are summarized in eq 3 and Scheme 1. Suzuki
The structures of isolated compounds 23 and 24 were
characterized by NMR. Compound 23 shows in the aliphatic
region of three 1H NMR singlets at δ 5.78, 5.66, and 4.68 ppm
1
in the ratio 2:1:1. There is no signal in H NMR that would
correspond to the protons of the methyl group. The aromatic
region shows the presence of eight protons, two of which are
in the higher field. In the 13C NMR spectrum, there is only one
doublet in the aliphatic region at δ 44.6 ppm, while the aromatic
region shows eight doublets and a triplet at 110.3 ppm. All the
data obtained from NMR point to structure 23 characterized by
an exocyclic double bond. Compound 24 is characterized by
1
the presence of two singlets in the aliphatic region of the H
(8) (a) Freccero, M. Mini-ReV. Org. Chem. 2004, 1, 403. (b) Rietjens, I.
M. C. M.; Boersma, M. G.; van der Woude, H.; Jeurissen, S. M. F.; Schutte,
M. E.; Alink, G. M. Mutat. Res. 2005, 574, 124.
NMR spectrum at δ 4.59 and 2.32 ppm (in the ratio 1:3) and
the presence of a doublet and a quartet in the 13C NMR at δ
2678 J. Org. Chem., Vol. 71, No. 7, 2006

Competing Excited State Proton Transfer Pathways
SCHEME 1
SCHEME 2
53.0 and 21.7 ppm. The aromatic region of the proton spectrum
shows eight protons, four of which are shielded, which is in
agreement with the 13C spectrum characterized by eight aromatic
doublets. The structure was also characterized by 2D NMR tech-
niques. Thus, in the HMBC spectrum the following important
interactions can be seen: C7-H5 and C7-H9 (see Experimental
Section and the Supporting Information). The NOESY spectrum
showed interactions between the following protons: CH3-H12,
H7-H5, and H7-H9. All of these interactions are in agreement
with the proposed bridged structure 24.
8) was recovered; no other product was detectable by NMR or
TLC. However, irradiation of 8 in neat CH3CN in the presence
of Bu4NClO4 (3 mM) gave 23 as the only product in 10% yield.
Compound 24 was not detected; if formed, it may have reverted
to 23 in the presence Bu4NClO4. In any case, this result shows
that a protic solvent is not required for the formation of 23, but
the media has to be sufficiently polar, which can be achieved
by either the addition of water or salts. The quantum yield of
formation of 23 in 1:4 H2O-CH3CN was determined by using
a secondary reference (photolysis of 4 in CH3CN-D2O giving
deuterated 4 and 6; Φ ) 0.09).7c In this way, the quantum yield
of formation of 23 was estimated to be 0.006, which is about
10 times smaller than the reference reaction. Hydration products
25, as well as bicyclic compound 24, are not thermally stabile
and readily revert to 8. Therefore, the amounts of 24 and 25
detected by NMR after photolyses were not reproducible, so
their quantum yield of formation is not reported. In addition,
for the same reasons as described above, we have not attempted
to measure quantum yields for loss of substrate as this is also
problematic due to reversion of products back to starting
material. Hence, the quantum yield of about 0.006 for formation
of 23 may be regarded as a lower limit of overall reactivity of
this system.
Whereas the hydration product 25 requires the presence of
water, it is not clear whether water is needed for the formation
of 23 and 24. Therefore, photolysis of 8 was carried out in neat
CH3CN. After 4 h of irradiation, only starting material (HPA
Preparative irradiations of 8 in CH3OH were also carried out.
Methanol is a better nucleophile than H2O, so the addition
products should be formed with higher efficiency. Additionally,
CH3OH adducts should exhibit higher chemical stability than
the corresponding hydration products. After 1 h of photolysis
of 8 in neat CH3OH, there was ∼65% conversion of the starting
material and formation of four adducts: cis- and trans-27
(∼40%) and cis- and trans-28 (∼15%). Besides 27 and 28,
FIGURE 1. (a) Absorption spectra showing loss of the anthracene
absorption band on photolysis (350 nm) of HPA 9 in 1:1 CH3OH-
H2O taken after the following irradiation times: 1, 2, 4, 8, 16, and 32
min. (b) Time dependence of the recovery of absorbance monitored at
375 nm after initial photolysis for HPA 9 (1:1 CH3CN-H2O) carried
out under low conversion. At higher conversions, the recovery is not
as clean due to formation of products that do not revert back to substrate
(vide infra) and also some residual photodecomposition.
J. Org. Chem, Vol. 71, No. 7, 2006 2679

Basaric´ and Wan
TABLE 1. Relative Yieldsa (%) of the Photochemical Products
Formed by Photolysis of 8 in CH3OH (CH3OD)
which were the major products, some smaller amounts of 23
(<10%) and 24 (traces) were also formed (Scheme 2, Table 1).
We were not able to isolate 27 and 28 because they readily
undergo elimination of methanol on silica, giving starting
material. However, evidence of their formation was obtained
photolysis conditions
8
23
24
27
28
350 nm, 1 h
35
75
92
96
55
80
9
5
1
1
4
3
1
trace
40
15
5
3
27
12
15
5
2
350 nm, 1 hb
1
from the H NMR spectra of crude photochemical mixtures.
300 nm, 2 h
254 nm, 2 h
350 nm, 1 h; 300 nm, 15 minc
350 nm, 1 h; 254 nm, 15 mind
Thus, cis- and trans-27 are characterized by the presence of
hydrogen-bonded OH signals at δ 10.33 and 10.09 ppm (ratio
1:10), methoxy signals at ∼3 ppm, quartets at ∼4.3 ppm, and
doublets at ∼1.65 ppm. On the other hand, cis- and trans-28
show in the 1H NMR spectrum triarylmethyl singlets at δ ∼5.7
ppm, methoxy singlets at 2.80 and 2.78 ppm (ratio 1:2), and
methyl singlets at ∼1.94 ppm.
14
5
a Relative yields obtained from the ratio of integrals of signals in 1H
NMR (CDCl3). b Photolysis carried out in CH3OD. c After 1 h of irradiation
at 350 nm, the same solution was then irradiated 15 min at 300 nm. d After
1 h of irradiation at 350 nm, the same solution was then irradiated 15 min
at 254 nm.
To investigate whether products 23, 24, 27, and 28 could be
ascribed to initial ESIPT (formal or direct), irradiation of the
(methoxyphenyl)anthracene derivative 16 in CH3OH was carried
out. No photoreaction was detected after 1 h of photolysis. This
indicates that the phenolic OH group in 8 is required (partici-
pates) in the photochemical reaction giving rise to the above
products, presumably via a solvent-mediated ESIPT mechanism.
Additionally, 8 was also irradiated in CH3OD. It gave a mixture
containing starting material 8 (75%), without any deuterium
incorporated in the molecule, and deuterated products 23 (5%),
24 (traces), 27 (15%), and 28 (5%) (Scheme 2, Table 1).
Deuterium incorporation in the products was observed from the
The photochemistry of 8 in CH3OH was investigated further
by changing the irradiation wavelength. The relative yields of
1
the photoproducts obtained from the H NMR spectra of the
crude mixtures are summarized in Table 1. From the data in
Table 1 it can be seen that the highest yields of products were
obtained upon photolysis at 350 nm, while irradiation at 300
and 254 nm resulted mostly in recovery of the starting material.
This observation can be rationalized by absorption of light (at
300 and 254 nm) by photoproducts which revert to the starting
material by a photochemical reaction. After prolonged irradia-
tion, a photostationary state is reached, determined by the ratio
of absorption coefficients, and rates of interconversions, of
products and the starting material. Thus, after irradiation of 8
at 350 nm where the products are formed with maximal
efficiency, subsequent irradiation (of the same solution) at 300
or 254 nm gives back the starting material and hence a lower
observed yield of photoproducts (Table 1).
1
disappearance of the signals in H NMR, comparing to the
spectra after photolysis in CH3OH at δ 5.66, 4.59, 4.30, and
5.70 ppm, respectively. In addition to disappearance of those
signals, the doublet at ∼1.6 ppm (from the compound 27)
became a broad singlet. Formation of deuterated products
additionally indicates that ESIPT (or solvent-mediated ESIPT)
to position 9 or 10 of the anthracene ring is one of the steps in
their formation. Moreover, the conversion of the starting material
after 1 h of photolysis in CH3OD was ∼25%; that is about 2.6
times lower than in CH3OH (Table 1). The lower conversion is
in agreement with the slower transfer of deuterium in ESIPT
due to an apparent primary isotope effect.
Formation of two types of CH3OH adducts (regioisomers
27 and 28), and incorporation of deuterium in the products
at position 9 or 10 of the anthracene ring upon photolysis in
CH3OD, indicates the presence of two types of intermediates
that are formed in the ESIPT of HPA 8, the QM 29 and the
zwitterion (ZI) 30. Hydration products 25 and the CH3OH
adducts 27 probably arise by solvent attack of the QM 29. On
the other hand, 23, 24, and 28 are probably formed from the ZI
30, by elimination reaction, intramolecular cyclization, and
nucleophillic attack of CH3OH, respectively. The CH3OH
adducts 27 and 28 are formed in a 3:1 ratio, as revealed by the
1
Photolysis of the HPA derivative 9 in 1:4 H2O-CH3CN gave
bridged product 31 (eq 4) as the only detectable product, formed
in ∼10-20%, after 4 h of irradiation, with ∼20% conversion
of the starting material. Compound 31 was isolated by TLC on
silica gel and characterized by 1D and 2D NMR. All the NMR
data are similar to those for compound 24 and in accordance
with the structural characteristics of the bridged structure 31.
Hydration products 32 and 33 were not detected by NMR upon
photolysis of 9 in H2O-CH3CN. They were probably formed
but reverted to the starting material fast enough to avoid
detection. To investigate the intriguing possibility that ESIPT
might take place to the phenyl ring at the 10 position, 9 was
irradiated in 4:1 CH3CN-D2O. After 1 h of photolysis, the
recovered starting material 9 show no readily detectable
exchange by NMR.
integration of methoxy singlets in H NMR, giving the ap-
proximate ratio of the rates of these two ESIPT pathways. Thus,
under the employed conditions (irradiation in CH3OH), long-
range ESIPT to position 10 of the anthracene ring giving 29 is
about 3 time faster than the short-range ESIPT to position 9
giving 30.
2680 J. Org. Chem., Vol. 71, No. 7, 2006

Competing Excited State Proton Transfer Pathways
TABLE 2. Photophysical Characteristics of the HPA Derivatives
For compound 9, deuterium exchange experiments cannot
provide information regarding ESIPT to positions 9 and 10 of
the anthracene ring. However, formation of CH3OH adducts will.
After 1 h of irradiation, two CH3OH adducts were detected by
NMR, probably cis- and trans-34 (eq 4) which were formed in
a 1:1 ratio, in ∼40% yield with ∼40-50% conversion of the
starting material. Any attempt to isolate cis- and trans-34 by
chromatography resulted in recovery of 9 and isolation of a trace
of 31. The isomers 34 readily undergo elimination of CH3OH
on silica gel giving 9. However, the presence of cis- and trans-
34 was indicated from the characteristic signals in the 1H NMR
spectrum of the crude photochemical mixture, corresponding
to phenolic OH protons at δ 10.24 and 10.18 ppm, triarylmethyl
protons at 5.38 and 5.32 ppm, and methoxy groups at 3.18
and 2.84 ppm. The adducts 35 were not detected by NMR, but
they could have been formed but too reactive for isolation.
Detection of the CH3OH adducts 34 and isolation of the bicyclic
product 31 confirms that 9 also undergoes ESIPT, giving two
types of intermediates (as for 8): solvent-mediated long-range
proton transfer to position 10 of the anthracene ring gives
QM 36, while short-range proton transfer to position 9 gives
ZI 37.
7-10a
compd
λ
abs (nm)b
λfl (nm)c
Φfd
τe (ns)
7
8
254, 346, 365, 384
258, 355, 374, 394
420
0.27
0.62
3.7
402, 423, 445 (sh),
475 (sh)
11.3
9
10
258, 354, 372, 392
259, 355, 373, 393
405, 425, 450 (sh)
416, 430 (sh)
0.93
0.64
8.7
4.2
a Measurements were performed in CH3CN at 20 °C. b Maxima in the
absorption spectra. c Maxima in the fluorescence spectra. d Fluorescence
quantum yield measured relative to quinine sulfate in 1.0 N H2SO4 (Φfl
)
0.55).10 Estimated error is (0.05. e Singlet excited-state lifetime, measured
by single photon counting, estimated error (0.1 ns.
already indicated by the above results for 7; we can assume
with some confidence that it is operating to full effectiveness
for 10.
Fluorescence Measurements. All the HPA derivatives 7-10
were characterized by relatively strong fluorescence emission
in CH3CN with bands between 400 and 500 nm, typical for
anthracene fluorescence (Table 2). Stokes shifts for all the
compounds were relatively small indicating that there is only a
small perturbation of the geometry for the fluorescent state. The
fluorescence of 7 was the most perturbed from classical
anthracene emission. It was structureless and characterized by
a lower quantum yield of fluorescence compared to 4 (Φf )
0.50, τ ) 5.9 ns).7c The introduction of a methyl substituent at
the 10 position in 8 resulted in an increase of the fluorescence
quantum yield and lifetime, contrary to expectation. However,
the singlet lifetime and emission spectrum of 9 are similar to
those reported for other diphenylanthracenes (∼8 ns).9 From
fluorescence and UV-vis spectra of 7-10 we may conclude
that these compounds are characterized by weak conjugation
(interaction) between the anthracene and phenol moieties in both
ground and excited singlet states.
The fluorescence emissions of all the HPA’s are quenched
upon addition of water. One example of quenching by water is
shown in Figure 2 for the derivative 10. The quenching (for all
the HPA’s 7-10) is characterized by a decrease of the
fluorescence intensity without any shift of the maxima. New
bands that might arise from emission of phenolates were not
detected, indicating that water quenching is due to solvent-
mediated ESIPT (giving rise to intermediate QM’s and ZI’s),
rather than simply proton transfer to the solvent (giving excited
state phenolates). That fact that fluorescence from phenolates
was not observed does not rule out their formation. They may
be formed but are very short lived (i.e., very reactive). For
example, the excited state phenolate may undergo protonation
of the anthracene at the 9 or 10 positions giving the observed
Unlike HPA derivatives 8 and 9, photolysis of 7 and 10 in
CH3OH or H2O-CH3CN (1:4), resulted in only recovery of
starting material. We were unable to detect any hydration or
CH3OH addition product by NMR. The explanation probably
lies with the fast and efficient elimination of H2O or CH3OH
from the corresponding adducts. This is in accord with the UV-
vis experiments (Figure 1) which showed significantly faster
recovery to starting material upon bleaching of the UV-vis
absorption of these substrates on photolysis. However, photolysis
of 7 in D2O-CH3CN (1:4) resulted in deuterium incorporation
in the molecule. After 15 min, 23% of the hydrogen at position
10 of the anthracene ring was replaced by deuterium, whereas
1 h of irradiation furnished 60% of deuterated 7. Formation of
deuterated HPA 7 is explained by formaton of QM 38 via
ESIPT. The quantum yield of deuterium incorporation in
position 10 was estimated by using a secondary reference
(photolysis of 4 in CH3CN-D2O, giving deuterated 4 and 6; Φ
) 0.09)7c that gave a quantum yield of 0.06, which is similar
to the efficiency of deuterium incorporation in the parent 4.
Although we were not able to detect any product upon
photolysis of 10 in CH3CN-H2O or CH3OH, in analogy with
the ESIPT chemistry observed for 7-9, one might assume that
10 also undergoes this type of photochemistry, resulting in
protonation of positions 10 and 9 of the anthracene ring, giving
rise to intermediates QM 39 and ZI 40, respectively. Both 7
and 10 are substituted by a p-hydroxyl group on a phenyl ring.
Such a group is a well-know electron-donating group in the
ground state and may destabilize any addition products that
may be formed in these systems, thus making them much
prone to decomposition (reversion back to substrate). This is
(9) (a) Schoof, S.; Guesten, H.; Von Sonntag, C. Ber. Bunsen-Ges. 1978,
82, 1068. (b) Meech, S. R.; Phillips, D. J. Photochem. 1983, 23, 193.
J. Org. Chem, Vol. 71, No. 7, 2006 2681

Basaric´ and Wan
affected by oxygen, and negative signals (bleaching) in the 350-
400 nm region due to the absorption of the laser flash by the
substrate. The bleaching signals for all the derivatives recovered
almost completely to baseline indicating good reversibility of
the processes initiated by the laser flash. In addition to these
bands, the anthracene triplet-triplet absorption was detected.
It was significantly influenced (shorter lifetimes) by the presence
of oxygen and situated with the maximum at ∼430 nm, in
agreement with the literature.12
In both solvents, we also detected transient absorptions at
longer wavelengths (500-700 nm), where the absorption of
QM’s was anticipated. Those bands were not affected by the
presence of oxygen and were especially pronounced for the
HPA’s 8 and 9. However, we observed similar transient
absorptions for the (nonreactive) methoxy compound 16.
Decreasing the laser power resulted in higher relative signal
intensity at 430 nm compared to the signals in the 500-700
nm region. This suggests that the 500-700 nm transient arises
via a two-photon process (possibly forming radical cations).
Addition of ethanolamine, a known QM quencher,13 resulted
in no observable changes in any of the observed transients
obtained for 7 (in CH3CN-H2O). We conclude that none of
the observed transient absorptions correspond to QM’s. Al-
though QM’s are probably formed as transients, they are
probably too weak to be observed in these systems that also
have strong bleaching signals and triplet-triplet absorptions as
well as transients that arise via two-photon processes. Attempts
were also made to characterize ZI intermediates 30, 37, and
40. LFP studies were carried out in trifluoroethanol, a polar
but less nucleophilic solvent, with the expectation that the
transient if formed would be longer lived. However, no
detectable transients were observed in the 450-500 nm region
where di- or triarylmethyl carbocations are known to absorb.14
Thus, the proposed ZI’s are characterized by short lifetimes,
which in addition to their low quantum yields of formation
would make their detection very difficult.
FIGURE 2. Quenching of the fluorescence of HPA 10 in CH3CN
(λex ) 375 nm) by addition of water ([H2O] ranging from 0 to 10 M).
FIGURE 3. Stern-Volmer plots of fluorescence quenching vs water
concentration for HPA’s 7-10 (in CH3CN).
photochemistry reported above. Nevertheless, this pathway may
still be regarded as a formal ESIPT mechanism. In general,
however, excited state proton transfer to solvent requires much
higher water content for most phenols. The observation of
fluorescence quenching at relatively low water content as
observed in these phenol systems would suggest that a different
water-mediated quenching pathway is operative, viz., a water-
mediated ESIPT to the anthracene 9 and 10 positions.
Mechanisms of Reaction. Upon excitation with near-visible
light, all of the HPA’s 7-9 (and probably 10 also) undergo
formal ESIPT in CH3CN-H2O or CH3OH. Singlet state
reactivity has been inferred and is in accord with previous results
of related systems.7 For HPA 7, only one ESIPT pathway was
observed, protonation of the 10 position of the anthracene ring,
giving QM 38 (Scheme 3). Short-range ESIPT to position 9
was not observed as this would be intrinsically less favored in
the formation of a ZI rather than a neutral and fully conjugated
QM. In addition to reverting back to 7 via tautomerization (with
deuterium incorporation at the 10 position if carried out in
deuterated solvents), QM 38 can react with solvent, giving
addition products 41 and 42.
Stern-Volmer plots (Figure 3) show nonlinear dependence
of the quenching efficiency on water concentration. With a small
amount of water (up to 5 M) quenching could be described as
a polynomial function, which is in agreement with a transfer of
proton to clusters of water molecules, seen in similar water-
mediated ESIPT systems.11 Water quenching is the most
efficient for the HPA derivative 7 (∼10 times more efficient
than for HPA’s 8-10). Relatively lower quenching efficiencies
for 8-10 indicate lower quantum yields of their water-mediated
ESIPT reactions. However, considering the high quantum yields
of fluorescence, especially for HPA’s 8 and 9, ESIPT is a quite
important nonradiative deactivation pathway from the singlet
excited state.
One of the results that we concluded from our fluorescence
data (vide supra) was that there is very little interaction between
the phenyl and the anthracene chromophores in ground and
excited singlet states of these compounds. However, it is clear
that formation of a QM intermediate requires the phenyl and
Nanosecond Laser Flash Photolysis. Nanosecond LFP was
used to probe for transient intermediates, viz., QM’s (29, 36,
38, and 39) and ZI’s, (30, 37, and 40) formed by ESIPT in
HPA’s 7-10 (Scheme 3). Transient absorption spectra were
recorded in 1:1 CH3CN-H2O, where the formation of the
intermediates was anticipated, and in neat CH3CN, which was
the control run (where the intermediates are not formed, based
on product studies). The transient absorption spectra for all
compounds (in both solvents) were characterized by strong
transient absorptions in the 250-350 nm region that were not
(12) Carmichael, I.; Helman, W. P.; Hug, G. L. J. Phys. Chem. Ref. Data
1987, 16, 239.
(13) (a) Brousmiche, D. W.; Wan, P. J. Photochem. Photobiol., A 2002,
149, 71. (b) Brousmiche, D. W.; Xu, M.; Lukeman, M.; Wan, P. J. Am.
Chem. Soc. 2003, 125, 12961. (c) Lukeman, M.; Veale, D.; Wan, P.;
Munasinghe, V. R. N.; Corrie, J. E. T. Can. J. Chem. 2004, 82, 240.
(14) (a) McClelland, R. A.; Kanagasabapathy, V. M.; Banait, N. S.;
Steenken, S. J. Am. Chem. Soc. 1989, 111, 3966. (b) Cozens, F. L.;
Kanagasabaphaty, V. M.; McClelland, R. A.; Steenken, S. Can. J. Chem.
1999, 77, 2069.
(10) Olmsted, J. J. Phys. Chem. 1979, 83, 2581.
(11) (a) Fischer, M.; Wan, P. J. Am. Chem. Soc. 1998, 120, 2680. (b)
Fischer, M.; Wan, P. J. Am. Chem. Soc. 1999, 121, 4555.
2682 J. Org. Chem., Vol. 71, No. 7, 2006

Competing Excited State Proton Transfer Pathways
SCHEME 3
anthracene rings to interact better, to give a more conjugated
system as implied in the QM structure. The formation of ZI’s
may also require enhanced electronic communication between
these rings to facilitate the transfer of the proton although it
may be somewhat less in this case. That is, how and when is
the proton transferred and how is it coupled with the required
torsional motion between the phenyl and anthracenyl moieties?
We have no way of directly probing these events, but in a
previous paper,7c several possible pathways were delineated that
will not be repeated here.
Substitution of the anthracene ring at position 10 with a
simple methyl group or a phenyl group (derivatives 8 and 9,
respectively) results in an overall decrease in ESIPT efficiency
that may be attributable to a simple steric effect as well as
observation of a new ESIPT pathway: short-range ESIPT to
the 9 position. The overall decrease in ESIPT efficiency was
manifested in lower efficiencies of fluorescence quenching by
water. ESIPT to the 10 position gives rise to QMs 29 and 36
which may be trapped by solvent. However, in these systems,
a short-range ESIPT operates to give ZI’s 30 and 37. This ESIPT
may be the direct proton-transfer pathway since it operates even
in neat CH3CN with added Bu4NClO4 (vide supra). The
operation of this pathway may be simply rationalized as the
result of steric congestion imposed at the 10 position on adding
a methyl or a phenyl group.
variety of addition products (25-28, 32-35). The ZI intermedi-
ates also react by intramolecular ring closure giving bridged
products 24 and 31, and also undergo elimination (feasible with
the methyl derivative 30) giving exocyclic alkene 23. The
relative yield of CH3OH adducts 27 and 28 formed by ESIPT
of HPA 8 (under conditions where formation of 23 and 24 is
negligible) gives the approximate ratio of the rates of the two
competing ESIPT pathways. In neat CH3OH, long-range ESIPT
to position 10 of the anthracene ring is about 3 times faster
than the short-range ESIPT to position 9. Thus, we anticipate
then that in the case where the methyl group is removed, the
long-range ESIPT pathway should dominate, as it does in the
case of 7 and 4.
Finally, these results amply demonstrate that the overall
chemistry in ESIPT from phenol to the anthracene moiety in
these compounds is highly sensitive to structural changes in
the chromophore. It is possible to have a totally reversible
process in which no products are isolable or it can result in the
formation of numerous interesting anthracene addition or
rearrangement products. The formation of ZI’s was shown to
occur via a pathway that was not initially anticipated but may
ultimately be useful for preparing bridged dibenzooxepines. To
the best of our knowledge there were only two reports for the
preparation of such a ring system.15
(15) (a) Mori, A.; Kawakami, H.; Kato, N.; Wu, S.-P.; Takeshita, H.
Org. Biomol. Chem. 2003, 1, 1730. (b) Mori, A.; Yokoo, H.; Hatsui, T.
Tetrahedron 2004, 60, 8783.
Both ZI and QM intermediates may revert back to starting
material or be trapped by solvents (H2O or CH3OH) giving a
J. Org. Chem, Vol. 71, No. 7, 2006 2683

Basaric´ and Wan
128.6, 127.8, 127.6, 127.3, 127.1, 125.1, 121.0, 111.5, one singlet
was not seen; MS (EI) m/z 360 (M+, 10), 284 (100), 278 (50), 277
(100), 268 (30); HRMS calculated for C27H20O 360.1514, observed
360.1513.
9-(2′-Methoxyphenyl)-10-(4′-methoxyphenyl)anthracene (22).
Starting from 500 mg (1.10 mmol) of 9-bromo-10-(2′-methoxyphe-
nyl)anthracene (18) and 230 mg (1.51 mmol) of 4-methoxyphenyl
boronic acid (21), the reaction furnished 370 mg (68.7%) of product
22.
Colorless crystals, mp 208-209 °C; 1H NMR (CDCl3, 300 MHz)
δ 7.68-7.78 (m, 2H), 7.58-7.66 (m, 2H), 7.54 (dd, 1H, J ) 2.2
Hz, J ) 8.1 Hz), 7.45 (dd, 2H, J ) 6.6 Hz, J ) 9.5 Hz), 7.26-
7.33 (m, 5H), 7.10-7.20 (m, 4H), 3.95 (s, 3H), 3.64 (s, 3H); 13C
NMR (CDCl3, 75 MHz) δ 159.2, 158.3, 137.0, 133.9, 133.2, 132.70,
132.66, 131.5, 130.6, 130.3, 129.5, 127.9, 127.3, 127.1, 125.1,
125.0, 121.0, 114.0, 111.5, 56.0, 55.6, one doublet more due to
not equivalent carbons; MS (EI) m/z 391 (M+, 30), 390 (M+, 100),
321 (20), 97 (20), 83 (20); HRMS calculated for C28H22O2
390.1620, observed 390.1621.
Cleavage of the Methyl Group by Use of BBr3: General
Procedure. The methoxy compound was dissolved in dichlo-
romethane and the solution was cooled to 0 °C with an ice-water
bath. By use of a syringe a dichloromethane solution (1M) of 5
equiv of BBr3 per equiv of methoxy substituent was added dropwise
under nitrogen. The ice bath was removed and the reaction mixture
was stirred at room temperature for an additional 2 h under nitrogen.
Reaction was quenched by addition of water and the layers were
separated. The water layer was extracted two more times with use
of CH2Cl2. Combined extracts were dried over anhydrous MgSO4,
solvent was removed, and the residue was chromatographed on a
column of silica gel with hexane/CH2Cl2 as eluent.
Experimental Section
Suzuki Reaction: General Procedure. A solution of 1 equiv
of bromo compound in toluene was added to a methanol solution
of 1 equiv of boronic acid. To the reaction mixture was added 2
equiv of potassium carbonate and the solution was purged with
N2. To the reaction mixture was added 0.01-0.02 equiv of Pd-
(PPh3)4 and the solution was purged again with N2. The reaction
mixture was heated at reflux temperature for 24 h under N2. To
the cooled reaction mixture was added water and extractions with
CH2Cl2 were carried out. Collected extracts were dried over
anhydrous MgSO4, solvent was evaporated, and the residue was
chromatographed on a column of silica gel with hexane/CH2Cl2
mixture as eluent.
9-Bromo-10-(2′-methoxyphenyl)anthracene (18). Starting from
1.00 g (6.6 mmol) of 2-methoxyphenyl boronic acid (13) and 2.21
g (6.6 mmol) 9,10-dibromoanthracene (17) the reaction furnished
1.46 g (60.9%) of product 18.
Yellow crystals, mp 135-136 °C; 1H NMR (CDCl3, 300 MHz)
δ 8.48 (d, 2H, J ) 8.8 Hz), 7.50 (d, 2H, J ) 8.8 Hz), 7.38-7.48
(m, 3H), 7.23 (ddd, 2H, J ) 1.5 Hz, J ) 6.6 Hz, J ) 8.8 Hz), 7.12
(dd, 1H, J ) 2.2 Hz, J ) 7.4 Hz), 7.05 (ddd, 1H, J ) 2.2 Hz, J )
7.4 Hz, J ) 7.4 Hz), 7.00 (d, 1H, J ) 8.1 Hz), 3.46 (s, 3H); 13C
NMR (CDCl3, 75 MHz) δ 158.2, 135.0, 133.0, 131.5, 130.7, 130.0,
128.2, 127.6, 127.2, 125.8, 123.1, 121.1, 111.6, 56.0, one singlet
was not seen; MS (EI) m/z 365 (M+, 30), 364 (M+, 100), 363 (M+,
30), 362 (M+, 100), 268 (50), 239 (40); HRMS, calculated for
C21H15BrO 362.0306, observed 362.0305.
9-(2′,4′-Dimethoxyphenyl)anthracene (15). Starting from 706
mg (2.75 mmol) of 9-bromoanthracene (11) and 500 mg (2.75
mmol) of 2,4-dimethoxyphenyl boronic acid (14) the reaction
furnished 470 mg (54.3%) of product 15.
9-(2′,4′-Dihydroxyphenyl)anthracene (7). Starting from 540 mg
(1.72 mmol) of 9-(2′,4′-dimethoxyphenyl)anthracene (15), the
reaction furnished 480 mg (97.5%) of product 7.
Colorless crystals, mp 228-229 °C; 1H NMR (CDCl3, 300 MHz)
δ 8.46 (s, 1H), 8.01 (d, 2H, J ) 8.8 Hz), 7.64 (d, 2H, J ) 8.8 Hz),
7.43 (dd, 2H, J ) 6.6 Hz, J ) 8.8 Hz), 7.32 (dd, 2H, J ) 6.6 Hz,
J ) 8.8 Hz), 7.15 (d, 1H, J ) 8.1 Hz), 6.71 (d, 1H, J ) 2.2 Hz),
6.69 (dd, 1H, J ) 2.2 Hz, J ) 8.1 Hz), 3.94 (s, 3H), 3.58 (s, 3H);
13C NMR (CDCl3, 75 MHz) δ 161.0, 159.2, 133.9, 133.4, 131.7,
131.0, 128.6, 127.1, 126.6, 125.3, 125.2, 120.0, 109.8, 104.8, 99.2,
55.9, 55.7; MS (EI) m/z 315 (M+, 30), 314 (M+, 100), 255 (10),
226 (10); HRMS, calculated for C22H18O2 314.1307, observed
314.1308.
Yellow crystals, mp 161-162 °C; 1H NMR (CDCl3, 300 MHz)
δ 8.53 (s, 1H), 8.04 (d, 2H, J ) 8.0 Hz), 7.70 (d, 2H, J ) 8.8 Hz),
7.48 (dd, 2H, J ) 6.6 Hz, J ) 8.0 Hz), 7.40 (dd, 2H, J ) 6.6 Hz,
J ) 8.8 Hz), 7.09 (d, 1H, J ) 8.1 Hz), 6.65 (d, 1H, J ) 2.2 Hz),
6.62 (dd, 1H, J ) 2.2 Hz, J ) 8.1 Hz), 5.03 (s, 1H, OH), 4.53 (s,
1H, OH); 13C NMR (CDCl3, 75 MHz) δ 157.2, 155.0, 133.1, 131.8,
131.5, 129.6, 128.8, 128.2, 126.6, 126.3, 125.8, 116.8, 108.4, 103.0;
MS (EI) m/z 287 (M+, 21), 286 (M+, 100), 202 (10), 119 (10);
HRMS calculated for C20H14O2 286.0994, observed 286.0997.
9-(2′-Hydroxyphenyl)-10-methylanthracene (8). Starting from
1.40 g (4.70 mmol) of 9-(2′-methoxyphenyl)-10-methylanthracene
(16), the reaction furnished 1.33 g (99%) of product 8.
Yellow crystals, mp 134-135 °C; 1H NMR (CDCl3, 300 MHz)
δ 8.38 (d, 2H, J ) 8.8 Hz), 7.67 (d, 2H, J ) 8.8 Hz), 7.53 (dd, 2H,
J ) 6.6 Hz, J ) 8.8 Hz), 7.46 (ddd, 1H J ) 1.5 Hz, J ) 7.5 Hz,
J ) 8.1 Hz), 7.40 (dd, 2H, J ) 6.6 Hz, J ) 8.8 Hz), 7.24 (dd, 1H,
J ) 1.5 Hz, J ) 7.4 Hz), 7.15 (d, 1H, J ) 8.1 Hz), 7.13 (dd, 1H,
J ) 7.4 Hz, J ) 7.5 Hz), 4.50 (s, 1H, OH), 3.19 (s, 3H); 13C NMR
(CDCl3, 75 MHz) δ 154.1, 132.6, 132.2, 130.8, 130.3, 130.0, 128.3,
127.0, 126.0, 125.7, 125.2, 124.7, 120.9, 115.8, 14.5; MS (EI) m/z
285 (M+, 25), 284 (M+, 100), 268 (20), 239 (20); HRMS calculated
for C21H16O 284.1201, observed 284.1203.
9-(2′-Hydroxyphenyl)-10-phenylanthracene (9). Starting from
550 mg (1.53 mmol) of 9-(2-methoxyphenyl)-10-phenylanthracene
(20), the reaction furnished 510 mg (96.4%) of product 9.
Yellow crystals, mp 212-213 °C; 1H NMR (CDCl3, 500 MHz)
δ 7.68-7.72 (m, 4H), 7.58-7.62 (m, 2H), 7.55 (tt, 1H, J ) 1.5
Hz, J ) 7.3 Hz), 7.44-7.51 (m, 3H), 7.33-7.40 (m, 4H), 7.30
(dd, 1H, J ) 1.8 Hz, J ) 7.5 Hz), 7.18 (dd, 1H, J ) 1.1 Hz, J )
8.1 Hz), 7.16 (ddd, 1H, J ) 1.1 Hz, J ) 7.3 Hz, J ) 7.5 Hz), 4.59
(s, 1H, OH); 13C NMR (CDCl3, 75 MHz) δ 154.1, 138.89, 138.86,
132.5, 131.4, 131.3, 130.8, 130.4, 130.1, 129.9, 128.7, 127.9, 127.6,
126.34, 126.27, 125.6, 124.5, 121.0, 116.0, one doublet more due
to not equivalent carbons; MS (EI) m/z 347 (M+, 28), 346 (M+,
9-(2′-Methoxyphenyl)-10-methylanthracene (16). Starting from
1.75 g (6.5 mmol) of 9-bromo-10-methylanthracene (12)16 and 0.98
g (6.5 mmol) of 2-methoxyphenyl boronic acid (13), the reaction
furnished 1.73 g (89.2%) of product 16.
Colorless crystals, mp 150-151 °C; 1H NMR (CDCl3, 300 MHz)
δ 8.33 (d, 2H, J ) 8.8 Hz), 7.60 (d, 2H, J ) 8.8 Hz), 7.52 (ddd,
1H J ) 1.5 Hz, J ) 7.5 Hz, J ) 8.1 Hz), 7.48 (dd, 2H, J ) 6.6 Hz,
J ) 8.8 Hz), 7.31 (dd, 2H, J ) 6.6 Hz, J ) 8.8 Hz), 7.23 (dd, 1H,
J ) 1.5 Hz, J ) 7.4 Hz), 7.15 (dd, 1H, J ) 7.4 Hz, J ) 7.5 Hz),
7.13 (d, 1H, J ) 8.1 Hz), 3.60 (s, 3H), 3.16 (s, 3H); 13C NMR
(CDCl3, 75 MHz) δ 158.3, 133.3, 130.3, 130.2, 129.4, 128.0, 127.7,
125.1, 125.0, 124.9, 120.9, 111.4, 55.9, 14.4, one singlet was not
seen; MS (EI) m/z 299 (M+, 30), 298 (M+, 100), 268 (30); HRMS
calculated for C22H18O 298.1358, observed 298.1358.
9-(2′-Methoxyphenyl)-10-phenylanthracene (20). Starting from
400 mg (1.10 mmol) of 9-bromo-10-(2′-methoxyphenyl)anthracene
(18) and 150 mg (1.21 mmol) of phenyl boronic acid (19), the
reaction furnished 390 mg (98.5%) of product 20.
Colorless crystals, mp 222-223 °C; 1H NMR (CDCl3, 500 MHz)
δ 7.65-7.79 (m, 2H), 7.60-7.64 (m, 2H), 7.56-7.60 (m, 2H),
7.49-7.56 (m, 3H), 7.45 (dd, 1H, J ) 2.2 Hz, J ) 7.4 Hz), 7.25-
7.31 (m, 5H), 7.17 (dd, 1H, J ) 1.1 Hz, J ) 7.3 Hz), 7.16 (dd, 1H,
J ) 7.3 Hz, J ) 7.8 Hz) 3.65 (s, 3H); 13C NMR (CDCl3, 75 MHz)
δ 158.3, 139.4, 137.2, 134.0, 133.2, 131.64, 131.60, 130.2, 129.5,
(16) Duan, S.; Turk, J.; Speigle, J.; Corbin, J.; Masnovi, J.; Baker, R. J.
J. Org. Chem. 2000, 65, 3005.
2684 J. Org. Chem., Vol. 71, No. 7, 2006

Competing Excited State Proton Transfer Pathways
100), 252 (10), 206 (10), 163 (10); HRMS calculated for C26H18O
346.1358, observed 346.1359.
9-(2′-Hydroxyphenyl)-10-(4′-hydroxyphenyl)anthracene (10).
Starting from 350 mg (0.90 mmol) of 9-(2′-methoxyphenyl)-10-
(4′-methoxyphenyl)anthracene (22), the reaction furnished 310 mg
(95.4%) of product 10.
H-12), 7.36 (dd, 2H, J ) 1.5 Hz, J ) 7.4 Hz, H-9), 7.26 (ddd, 2H,
J ) 1.5 Hz, J ) 7.3 Hz, J ) 7.4 Hz, H-11), 7.22 (ddd, 2H, J ) 1.5
Hz, J ) 7.3 Hz, J ) 7.4 Hz, H-10), 7.16 (dd, 1H, J ) 1.5 Hz, J )
7.5 Hz, H-5), 6.98 (ddd, 1H, J ) 1.5 Hz, J ) 7.3 Hz, J ) 8.1 Hz,
H-3), 6.71 (ddd, 1H, J ) 1.1 Hz, J ) 7.3 Hz, J ) 7.5 Hz, H-4),
6.54 (dd, 1H, J ) 1.1 Hz, J ) 8.1 Hz, H-2), 4.59 (s, 1H, H-7),
2.32 (s, 3H, CH3); 13C NMR (CDCl3, 125 MHz) δ 154.0 (s, C-1),
144.4 (s, C-8), 137.4 (s, C-13), 128.9 (d, C-3), 128.6 (d, C-10),
127.21 (d, C-5), 127.17 (d, C-11), 126.8 (s, C-6), 124.8 (d, C-9),
124.2 (d, C-12), 120.4 (d, C-4), 119.7 (d, C-2), 53.0 (d, C-7), 21.7
(q), signal of C-14, covered by CDCl3;
Photolysis of 9-(2′-Hydroxyphenyl)-10-phenylanthracene (9).
A 100 mL sample of the 4:1 CH3CN-H2O solution, containing
100 mg of the HPA derivative 9, was irradiated 4 h at 350 nm.
From the TLC with a hexane/CH2Cl2 (1:1) mixture the following
compounds were isolated: 70 mg (70%) of the starting material
and 2.6 mg (2.6%) of the bicyclic product 6-phenyl-6,11-(1,2-
phenyleno)-6,11-dihydro-[b,e]dibenzooxepine (31).
Yellow crystals, mp 260-261 °C decomposition; 1H NMR
(acetone-d6, 300 MHz) δ 8.66 (s, 1H, OH), 7.97 (s, 1H, OH), 7.68-
7.80 (m, 4H), 7.46 (ddd, 1H, J ) 1.5 Hz, J ) 7.3 Hz, J ) 8.1 Hz),
7.33-7.40 (m, 4H), 7.22-7.33 (m, 3H), 7.20 (d, 1H, J ) 8.1 Hz),
7.15 (d, 2H, J ) 8.8 Hz), 7.12 (dd, 1H, J ) 7.3 Hz, J ) 7.4 Hz);
13C NMR (acetone-d6, 75 MHz) δ 158.0, 156.6, 138.2, 134.3, 133.6,
133.3, 133.2, 131.42, 131.38, 130.6, 130.4, 127.9, 127.8, 126.2,
126.0, 125.9, 120.8, 117.0, 116.41, 116.39; MS (EI) m/z 363 (M+,
30), 362 (M+, 100), 268 (10); HRMS calculated for C26H18O2
362.1307, observed 362.1307.
Preparative Irradiation Experiments: General Procedure.
A 100 mL sample of the solution (4:1 CH3CN-H2O) containing
typically 100 mg of the HPA (c ∼ 10-3 M) was purged with argon
for 15 min and irradiated 1 to 4 h at 350 nm in a Rayonet reactor.
During irradiation, solution was continuously purged with argon
and cooled by water. After irradiation, extractions with CH2Cl2 were
carried out, the extracts were dried over anhydrous MgSO4, solvent
was evaporated, and the residue was chromatographed on a thin
layer of silica (TLC) with hexane/CH2Cl2 mixture as eluent.
Photolysis of 9-(2′-Hydroxyphenyl)-10-methylanthracene (8).
A 100 mL sample of the 4:1 CH3CN-H2O solution, containing
150 mg of the HPA 8, was irradiated 4 h at 350 nm. From TLC
with a hexane/CH2Cl2 (1:1) mixture the following compounds were
isolated: 30 mg (20%) of the starting material, 8 mg (5.3%) of the
bicyclic product 6-methyl-6,11-(1,2-phenyleno)-6,11-dihydro[b,e]-
dibenzooxepine (24), and 30 mg (20%) of the vinyl product 2-(10-
methylene-9,10-dihydroanthracen-9-yl)phenol (23).
6-Phenyl-6,11-(1,2-phenyleno)-6,11-dihydro-[b,e]dibenzoox-
1
epine (31): yellow crystals, mp 244-245 °C decomposition; H
NMR (CDCl3, 500 MHz) δ 8.38 (d, 1H, J ) 7.8 Hz, H-16), 7.64
(dd, 1H, J ) 7.4 Hz, J ) 7.8 Hz, H-17), 7.48 (tt, 1H, J ) 1.3 Hz,
J ) 7.4 Hz, H-18), 7.42 (dd, 1H, J ) 7.4 Hz, J ) 7.8 Hz, H-17),
7.39 (dd, 2H, J ) 1.5 Hz, J ) 7.3 Hz, H-9), 7.21 (ddd, J ) 1.5
Hz, J ) 7.3 Hz, J ) 7.4 Hz, H-11), 7.18 (dd, 1H, J ) 1.8 Hz, J )
7.5 Hz, H-5), 7.10 (d, 1H, J ) 7.8 Hz, H-16), 7.09 (ddd, 2H, J )
1.5 Hz, J ) 7.3 Hz, J ) 7.8 Hz, H-10), 7.04 (ddd, 1H, J ) 1.8 Hz,
J ) 7.3 Hz, J ) 8.1 Hz, H-3), 6.84 (dd, 1H, J ) 1.3 Hz, J ) 7.8
Hz, H-12), 6.78 (dd, 1H, J ) 1.1 Hz, J ) 8.1 Hz, H-2), 6.74 (ddd,
1H, J ) 1.1 Hz, J ) 7.3 Hz, J ) 7.5 Hz, H-4), 4.70 (s, 1H, H-7);
13C NMR (CDCl3, 125 MHz) δ 153.0 (s, C-1), 144.1 (s, C-8), 139.6
(s, C-15), 138.0 (s, C-13), 130.0 (d), 129.0 (d), 128.64 (d), 128.57
(d), 128.03 (d), 128.00 (s), 127.9 (d), 127.6 (d), 127.11 (s, C-6),
127.06 (d, C-5), 127.0 (d, C-10), 124.3 (d, C-9), 120.8 (d, C-4),
120.2 (d, C-2), 82.5 (s, C-14), 52.7 (d, C-7); MS (EI) m/z 347 (M+,
30), 346 (M+, 100), 252 (15); HRMS calculated for C26H18O
346.1358, observed 346.1360.
2-(10-Methylene-9,10-dihydroanthracen-9-yl)phenol (23): yellow
1
crystals, mp 92-94 °C; H NMR (CDCl3, 300 MHz) δ 7.76 (d,
2H, J ) 8.8 Hz, H-12), 7.18-7.31 (m, 6H), 7.06 (ddd, 1H, J )
1.5 Hz, J ) 7.4 Hz, J ) 8.1 Hz, H-3), 6.97 (dd, 1H, J ) 1.5 Hz,
J ) 8.1 Hz, H-5), 6.81 (d, 1H, J ) 8.1 Hz, H-2), 6.77 (dd, 1H, J
) 7.4 Hz, J ) 8.1 Hz, H-4), 5.78 (s, 2H, H-15), 5.66 (s, 1H, H-7),
4.68 (s, 1H, OH); 13C NMR (CDCl3, 75 MHz) δ 152.8 (s, C-1),
141.7 (s, C-14), 137.6 (s), 135.2 (s), 131.1 (s), 131.0 (d, C-5), 128.4
(d), 128.3 (d), 128.1 (d, C-3), 127.2 (d), 124.5 (d), 121.5 (d, C-4),
116.5 (d, C-2), 110.3 (t, C-15), 44.6 (d, C-7); MS (EI) m/z 285
(M+, 25), 284 (M+, 100), 268 (10), 191 (10); HRMS calculated
for C21H16O 284.1201, observed 284.1201.
Laser Flash Photolysis (LFP). All LFP studies were conducted
at the University of Victoria LFP facility employing a YAG laser,
with a pulse width of 10 ns and excitation wavelength 355 nm.
Flow cells (0.7 cm) were used and solutions were purged with
nitrogen or oxygen for 30 min prior to measurements. Optical
densities at 355 nm were ∼0.4.
Steady-State and Time-Resolved Fluorescence Measurements.
The samples were dissolved in CH3CN and the concentrations were
adjusted to have optical densities at the excitation wavelength (355
or 375 nm) <0.1. Solutions were purged with nitrogen for 15 min
prior to analysis. The measurements were performed at 20 °C and
were not corrected. Fluorescence quantum yields were determined
by comparison of the integral of emission bands with that of quinine
bisulfate (Φ ) 0.55 in 1.0 N H2SO4).10 Typically three absorption
traces were recorded (and averaged) and two fluorescence emission
traces, exciting at two different wavelengths (i.e., 355 and 375 nm).
Two quantum yields were calculated and the mean value reported.
Fluorescence decay histograms were obtained on an instrument
equipped with a hydrogen flash lamp, using the time-correlated
single photon counting technique in 1023 channels. Histograms of
6-Methyl-6,11-(1,2-phenyleno)-6,11-dihydro-[b,e]dibenzoox-
epine (24): colorless crystals, thermally unstable, rearranges to 9-(2′-
1
hydroxyphenyl)-10-methylanthracene (8) (mp 134-135 °C); H
NMR (CDCl3, 500 MHz) δ 7.60 (dd, 2H, J ) 1.5 Hz, J ) 7.4 Hz,
J. Org. Chem, Vol. 71, No. 7, 2006 2685

Basaric´ and Wan
the instrument response functions (using a LUDOX scatterer) and
sample decays were recorded until they typically reached 3 × 103
counts in the peak channel. The half width of the instrument
response function was typically ∼1.5 ns. The time increment per
channel was 0.049 or 0.098 ns. Obtained histograms were fitted as
sums of the exponential, using Gaussian-weighted nonlinear least-
squares fitting based on Marquardt-Levenberg minimization
implemented in the software package of the instrument. The fitting
parameters (decay times and preexponential factors) were deter-
mined by minimizing the reduced ø2g. An additional graphical
method was used to judge the quality of the fit that included plots
of surfaces (“carpets”) of the weighted residuals vs channel number.
Acknowledgment. Support of this work was provided by
the National Sciences and Engineering Research Council
(NSERC) of Canada.
Supporting Information Available: Detailed experimental
1
procedures, H and 13C NMR spectra of the isolated compounds,
1H NMR spectra of some representative photochemical reaction
mixtures, UV and fluorescence spectra of 7-10, and representative
LFP runs. This material is available free of charge via the Internet
at http://pubs.acs.org.
JO0524728
2686 J. Org. Chem., Vol. 71, No. 7, 2006