Photogeneration of 1,5-naphthoquinone methides via excited-state (formal) intramolecular proton transfer (ESIPT) and photodehydration of 1-naphthol derivatives in aqueous solution

Article abstract of DOI:10.1139/v03-184

The photochemistry of naphthols 1, 2, 4, 5 and 9, and phenol 10 has been studied in aqueous solution with the primary aim of exploring the viability of such compounds for naphthoquinone and quinone methide photogeneration, along the lines already demonstr

Full text of DOI:10.1139/v03-184

240
Photogeneration of 1,5-naphthoquinone methides
via excited-state (formal) intramolecular proton
transfer (ESIPT) and photodehydration of 1-
naphthol derivatives in aqueous solution1
Matthew Lukeman, Duane Veale, Peter Wan, V. Ranjit N. Munasinghe, and
John E.T. Corrie
Abstract: The photochemistry of naphthols 1, 2, 4, 5 and 9, and phenol 10 has been studied in aqueous solution with
the primary aim of exploring the viability of such compounds for naphthoquinone and quinone methide
photogeneration, along the lines already demonstrated by our group for phenol derivatives. 1-Naphthol (1) is known to
be substantially more acidic than 2-naphthol (2) in the singlet excited state (pKa* = 0.4 and 2.8, respectively) and it
was expected that this difference in excited-state acidity might be manifested in higher reactivity of 1-naphthol deriva-
tives for photochemical reactions requiring excited-state naphtholate ions, such as quinone methide formation. Our re-
sults show that three types of naphthoquinone methides (26a, 26b, and 27) are readily photogenerated in aqueous
solution by irradiation of 1-naphthols. Photolysis of the parent 1-naphthol (1) in neutral aqueous solution gave 1,5-
naphthoquinone methide 26a as well as the non-Kekulé 1,8-naphthoquinone methide 26b, both via the process of ex-
cited-state (formal) intramolecular proton transfer (ESIPT), based on the observation of deuterium exchange at the 5-
and 8-positions, respectively, on photolysis in D2O–CH3CN. A transient assignable to the 1,5-naphthoquinone methide
26a was observed in laser flash photolysis experiments. The isomeric 2-naphthol (2) was unreactive under similar con-
ditions. The more conjugated 1,5-naphthoquinone methide 27 was formed efficiently via photodehydroxylation of 4;
isomeric 5 was unreactive. The efficient photosolvolytic reaction observed for 4 opens the way to design related
naphthol systems for application as photoreleasable protecting groups by virtue of the long-wavelength absorption of
the naphthalene chromophore.
Key words: photosolvolysis, excited-state intramolecular proton transfer, quinone methide, photorelease,
photoprotonation. 253
Résumé : Opérant en solution aqueuse, on a étudié la photochimie des naphtols 1, 2, 4, 5 et 9 et du phénol 10 afin de
déterminer la viabilité de tels composés pour la photogénération de méthides de naphtoquinone et de quinone suivant
les voies déjà mises de l’avant par notre groupe pour les dérivés du phénol. Il est bien connu que, dans l’état excité
singulet, le 1-naphtol (1) est beaucoup plus acide que le 2-naphtol (2) (pKa* de 0,4 et 2,8, respectivement) et on
s’attendait à ce que cette différence d’acidité dans l’état excité pourrait se manifester par une plus grande réactivité des
dérivés du 1-naphtol pour les réactions photochimiques impliquant des ions naphtanolates dans l’état excité, tel que la
formation de méthide. Nos résultats démontrent que l’irradiation des 1-naphtols en solutions aqueuses conduit facile-
ment à la formation de trois types de méthides de naphtoquinones (26a, 26b et 27). L’irradiation du 1-naphtol fonda-
mental (1), en solutions aqueuses neutres, conduit au méthide de 1,5-naphtoquinone (26a) ainsi qu’au méthide de 1,8-
naphtoquinone (26b), une structure qui ne répond pas aux normes de Kekulé; sur la base d’expériences de photolyse
dans un mélange de D2O–CH3CN qui conduisent à un échange de deutérium dans les positions respectivement 5 et 8,
ces deux méthides se produisent par un processus de transfert intramoléculaire de proton dans l’état excité (formel).
Lors d’expériences de photolyse éclair au laser, on a observé une entité transitoire attribuable au méthide de la 1,5-
naphtoquinone. Dans des conditions semblables, le 2-naphtol (2) isomère n’est pas réactif. Le méthide de la 1,5-
naphtoquinone (27) qui est plus conjuguée se forme facilement par photodéhydroxylation du composé 4; son isomère 5
n’est pas réactif. La réaction efficace de photosolvolyse du composé 4 ouvre la voie au développement de systèmes de
naphtols qui pourraient être utilisés comme groupes protecteurs qui pourraient être éliminés photochimiquement en
Received 23 June 2003. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 30 January 2004.
M. Lukeman, D. Veale, and P. Wan.2 Department of Chemistry, Box 3065, University of Victoria, Victoria, BC V8W 3V6,
Canada.
V.R.N. Munasinghe and J.E.T. Corrie.3 National Institute for Medical Research, The Ridgeway, Mill Hill, London, NW7 1AA, U.K.
1This article is part of a Special Issue dedicated to Professor Ed Piers.
2Corresponding author (e-mail: pwan@uvic.ca).
3Corresponding author (e-mail: jcorrie@nimr.mrc.ac.uk).
Can. J. Chem. 82: 240–253 (2004)
doi: 10.1139/V03-184
© 2004 NRC Canada

Lukeman et al.
241
vertu de la grande longueur d’onde de l’absorption du chromophore du naphtalène.
Mots clés : photosolvolyse, transfert intramoléculaire de proton dans l’état excité, méthide de quinone, photoélimina-
tion, photoprotonation.
[Traduit par la Rédaction] Lukeman et al.
Introduction
whether such a process could operate in the naphthols (1 and
2), since naphthols are formally made up of two benzene
rings, one of them containing a phenol moiety. Also ad-
dressed is whether suitably designed 1-naphthol derivatives
4 and 5 will react in an analogous manner as reported by us
(1) for the parent hydroxy-substituted benzyl alcohols with
respect to naphthoquinone methide formation. Seiler and
Wirz’s earlier results (7) suggest that this would indeed be
possible.
Quinone methides are reactive intermediates of general in-
terest in organic chemistry (1–3). Recent studies from our
laboratory (1, 2, 4, 5) have demonstrated that suitably de-
signed simple phenol derivatives give rise to quinone
methide-type intermediates efficiently on photolysis in aque-
ous solution. The substrates that we have investigated so far
all possess a phenol and a benzylic-type alcohol or styryl
moiety. The enhanced acidity of the phenol moiety in S1 in
these systems is essential for reaction, as previously de-
scribed (1, 2, 4, 5). We have not investigated simple naphthol
derivatives extensively except in some biaryl systems con-
taining a 2-naphthol chromophore, where the quinone
methides photogenerated are of the biaryl type (6).
Some time ago, Seiler and Wirz (7) reported the photo-
hydrolysis (to the corresponding carboxylic acids) of a vari-
ety of trifluoromethyl-substituted 1-and 2-naphthols (as well
as several phenols) and invoked the initial formation of
naphthoquinone and quinone methides in their photo-
hydrolysis mechanism (e.g., eq. [1]). Webb et al. (8) have re-
ported a detailed study of the excited-state proton transfer
dynamics of 1-naphthol (1) and have shown that it is a much
stronger acid (in S1) than 2-naphthol (2) (pKa* values of 0.4
and 2.8, respectively). Moreover, it was shown that the 5-
and 8-positions of 1 are much more basic in S1, as evidenced
by deuterium incorporation at these positions when 1 was ir-
radiated in acidic D2O. The formation of two isomeric
naphthoquinone methide intermediates was proposed to ac-
count for the photoexchange but whether such intermediates
were formed via a water-mediated excited-state (formal)
intramolecular proton transfer (ESIPT) (from the naphthol
OH to the carbon 5- or 8-position) was not explicitly ad-
dressed. We have recently reported (9) the first explicit ex-
ample of an ESIPT from a phenolic OH to an aromatic ring
carbon (on the ring not containing the phenolic OH) in 2-
phenylphenol (3) (eq. [2]). This result led us to investigate
The photochemistry of phenols is also involved in the p-
hydroxyphenacyl photodeprotection system reported by
Givens and co-workers (10). An example designed to
photorelease acetic acid from 6 is shown in eq. [3] (11). The
essentials of the mechanism involve formation of a spiro-
ketone intermediate 7 from photolysis of p-hydroxyphenacyl
derivatives bearing good leaving groups, such as p-
hydroxyphenacyl acetate 6. Ring opening of 7 by water
gives the observed product 8. Release of the protected moi-
ety (e.g., acetic acid from 6) is believed to occur in the pri-
[1]
[2]
[3]
© 2004 NRC Canada

242
Can. J. Chem. Vol. 82, 2004
Scheme 1. Reagents: (a) MeOH, H2SO4, 74%; (b) NaOMe,
CuBr, MeOH, 70%; (c) NaOH, MeOH–H2O, 85%; (d) MeLi,
THF–Et2O, 78%; (e) PhSH, K2CO3, N-methylpyrrolidinone, 72%;
(f) NaBH4, EtOH, 55%.
mary photochemical step. Although the question of triplet
vs. singlet state reactivity (which might be dependent on the
nature of the leaving group at the α-position) has been a
topic of debate (10c, 11), the involvement of excited-state
(singlet or triplet) deprotonation of the phenol moiety in the
reaction sequence for formation of intermediate 7 seems
likely. We felt that related substrates 9 and 10 might react
similarly. The results of investigations of a variety of
naphthols 1,2,4, 5 and 9 (and phenol 10) aimed at providing
further insights to all of the above related photochemical re-
actions are reported herein.
Results and discussion
Materials
1- and 2-Naphthols (1 and 2) were purchased from
Aldrich and were sublimed and recrystallized from hexanes
prior to use. The hydroxyethylnaphthols 4 and 5 were syn-
thesized as shown in Schemes 1 and 2. Although the corre-
sponding hydroxyketones are both known compounds, the
synthetic routes previously described were not particularly
convenient or led to mixtures of regioisomers. The routes
shown in the schemes were straightforward but, particularly
for the 5-isomer, were only reached after some exploration
of other published methods. For synthesis of the 5-isomer 4
(Scheme 1), the readily prepared 5-bromo-1-naphthoic acid
11 (12) was esterified and the 5-bromo group was displaced
with methoxide in a copper(I)-catalyzed procedure (13). Fur-
ther steps to the phenolic ketone 16 and its subsequent re-
duction readily gave the alcohol 4. Note however that the
work up procedure for the thiophenoxide-mediated
demethylation of the methyl ether 15 (and of 17 in the route
to the 4-hydroxy compound 5), which used a literature pro-
cedure (14), was modified by addition of hydrogen peroxide
to oxidize excess thiophenol, thereby minimizing the odour
associated with the procedure. The route to the 4-isomer 5
(Scheme 2) requires no further comment, except to note that
direct borohydride reduction of the phenolic ketone 18 gave
only dark, uncharacterized material and it was found neces-
sary to protect the phenol as its TBDMS ether 19 during this
step. No attempts were made in this study at synthesizing
derivatives of 1 with substituents at the 8-position, although
these compounds would indeed be worthwhile to investigate
in the future. The naphthacyl acetate 9 was prepared as
shown in Scheme 3 and the 3-hydroxyphenacyl acetate 10
was made by a route similar to those for the corresponding
4-hydroxy esters previously described (11).
Scheme 2. Reagents: (a) PhSH, K2CO3, N-methylpyrrolidinone,
59%; (b) TBDMS-Cl, imidazole, DMF, 80%; (c) NaBH4, MeOH,
89%; (d) TBAF, THF, 88%.
compound having a more diffuse excited state electronic dis-
tribution. Moreover, charge transfer from the oxygen to the
ring is localized at the 5- and 8-positions of 1. To demon-
strate the enhancement of basicity at these positions of 1,
they irradiated 1 in acidic D2O–CH3CN and showed that po-
sitions 5 and 8 were efficiently exchanged (Φ = 0.11 in 99%
D2O, pD 3; Table 1) while similar irradiation of 2 resulted in
no exchange. Photolysis in acidic solution raises the possi-
bility that the exchange may be due to photoprotonation of 1
by hydronium ion, a process that is well known for 1-
Deuterium exchange studies of 1 and 2
As noted in the Introduction, Webb et al. (8) have shown
that 1-naphthol (1) is considerably more acidic than 2-
naphthol (2) in S1, due to the enhanced (localized) charge
transfer character available for 1 but not for 2, the latter
© 2004 NRC Canada

Lukeman et al.
243
1
Scheme 3. Reagents: (a) oxalyl chloride, DMF, CH2Cl2, 100%;
(b) TMSCHN2, CH2Cl2–THF–Et3N, 74%; (c) aq. NH3, THF–
MeOH, 47%; (d) HCl, HOAc, 82%; (e) HOAc, DBU, toluene,
67%.
Fig. 1. 500 MHz H NMR spectra of 1 before (bottom), and af-
ter (top) 20 min irradiation at 300 nm in D2O–CH3CN (1:1),
showing decreases in the integrated area of the peaks at 7.80 and
8.22 ppm assigned to ring protons at the 5- and 8-positions, re-
spectively.
Table 1. Photophysical and photochemical parameters.
Compound
ΦR
Φfa
0.20 (0.174)
(0.27)
0.17
τb (ns)
1
2
4
5
0.11 0.05c
(10.6)
(13.3)
5.4
[4]
<0.001c
0.84d
<0.001d
0.23
5.9
aFluorescence emission quantum yield in neat CH3CN. Measured rela-
tive to the reported fluorescence quantum yield for 1-naphthol in cyclohex-
ane (Φf = 0.174) (15) with correction to account for differences in the
indices of refraction of the different solvents. Estimated error of 10% of
quoted value. Values in brackets are the fluorescence quantum yields in
cyclohexane, from ref 15.
1
500 MHz H NMR spectra (Fig. 1); the areas of the peaks
assigned to protons at the 5- and 8-positions (7.81 and
8.23 ppm, respectively) decreased smoothly with irradiation.
The two overlapping triplets at 7.47 and 7.45 ppm assigned
to the protons at the 6- and 7- ring positions exhibited
changes consistent with diminished coupling to the adjacent
protons at the 5- and 8-positions as they are replaced with
deuterium. MS analysis of the photolyzed sample indicated
an isotope distribution of 37% nondeuterated, 52% mono-
deuterated, and 11% dideuterated 1-naphthol, consistent
with the overall deuterium incorporation measured by NMR.
The observation of deuterium incorporation in neutral solu-
tion suggests that protonation by acid (D3O+) is not respon-
sible for the observed reaction, and that the exchange might
be due to an ESIPT process. Formation of 1-5D and 1-8D
from an ESIPT to the 5- and 8-positions would necessarily
proceed via quinone methides 26a-D and 26b-D, which have
been previously proposed by Webb et al. (8). 2-Naphthol (2)
did not undergo any observable exchange, even on pro-
longed irradiation, and the compound was completely recov-
ered.
bFluorescence lifetime in neat CH3CN measured by single photon count-
ing. Lifetimes measured in this way have an estimated error of 0.2 ns.
Values in brackets are the fluorescence lifetimes in cyclohexane from ref.
15.
cQuantum yield of total deuterium exchange (of all ring positions) in
99% D2O (pD 3) from ref. 8.
dQuantum yield of photomethanolysis in H2O–CH3OH (1:1). Estimated
error 10% of quoted value. Measured relative to the quantum yield for
photomethanolysis of o-hydroxybenzhydrol in the same solvent (5).
methoxynaphthalene (16) and other aromatic compounds
(17–19), but does not occur for 2-methoxynaphthalene (16).
Therefore, we decided to study photochemical deuterium in-
corporation of 1 and 2 in aqueous solution, initially under
neutral conditions and then over a wide pH range and as a
function of water content. To our knowledge, this kind of
study has not been previously reported for the naphthols. A
similar study for 3 (eq. [2]) (9) resulted in the discovery of a
new type of ESIPT (vide infra).
Irradiations of 1 and 2 were carried out in 1:1 D2O–
CH3CN (~10–3 mol L–1, pD 7, Rayonet photochemical reac-
tor, 300 nm lamps, <30 min). Irradiation of 1 for 20 min re-
sulted in efficient incorporation of deuterium at positions 5
(55%) and 8 (20%) to give 1-5D and 1-8D, respectively
(eq. [4]). This was readily evident by examination of
To better our understanding of the effect of acid on the
deuterium incorporation reaction of 1, photolyses were car-
ried out in which the pD of the aqueous portion was varied
over the pD range 1.5–11.4 (Fig. 2). While acid catalysis is
observed at pD < 2.5, exchange efficiency was essentially
constant from pD 2.5 through to pD 9. Since the hydronium
ion concentration varies by many orders of magnitude over
© 2004 NRC Canada

244
Can. J. Chem. Vol. 82, 2004
Fig. 2. Plot of % deuterium exchange at the 5- (triangles) and 8-
(squares) positions vs. pD (in D2O–CH3CN (1:1)) for 1. Solu-
tions (20 mg substrate in 40 mL solution) were irradiated at
254 nm (16 lamps) for 15 min.
Fig. 3. Plot of % deuterium exchange at 5- (triangles) and 8-
(squares) positions vs. D2O content (pD 7).
to apply this strategy to naphthoquinone methide generation,
derivatives 4 and 5 were studied. The naphthoquinone
methides produced by net photodehydration of 4 or 5 (27
and 28, respectively) were not expected to be sufficiently
stable to isolate, so we instead sought to obtain indirect evi-
dence of their presence by isolating their photosubstitution
products arising from nucleophilic attack. For example, in
methanol–water solutions, the expected photoproduct from 4
is methyl ether 29 (eq. [5]). Although formation of 29 is
this pH range, it is very unlikely that the deuterium incorpo-
ration observed within this range results from protonation by
D3O+ present in solution. The exchange efficiency decreases
above pD 9, which corresponds with the ground state pKa of
1 (pKa = 9.2 (8)), indicating that the deprotonated form 1–
does not undergo exchange. Consistent with this, no ex-
change was observed at high pD. This result rules out a step-
wise mechanism in which 1 first undergoes excited-state
proton transfer (ESPT) to give excited phenolate (1–*) that is
protonated by D2O in a subsequent step, since every excita-
tion of 1– would necessarily lead to an excited phenolate. It
is apparent that the naphthol OH is required for reaction,
suggesting that an ESIPT might be taking place.
[5]
To investigate the role of solvent water in the ESIPT reac-
tion of 1, photolyses were performed in which the concen-
tration of D2O (in CH3CN cosolvent) was varied (Fig. 3).
The deuterium exchange efficiency for 1 was negligible at
low D2O concentrations, but increased dramatically as more
D2O was added. The enhancement of reaction efficiency ap-
peared to reach saturation at about 1:1 D2O–CH3CN. The
need for water to mediate the proton transfer indicates that
the ESIPT is not intrinsic, in contrast to the ESIPT observed
for 3. This is probably because the distance between the
acidic proton and both basic carbons is too large to allow
ground state hydrogen bonding interactions, which are
thought to be a prerequisite for intrinsic ESIPTs (9b). The
efficiency for exchange at the 8-position plateaus at lower
water concentrations than does exchange at the 5-position.
This could be because the OH functionality is much closer
to the 8-position, so less water is required to mediate the
proton transfer.
consistent with the presence of 27 as an intermediate, reac-
tion through a cationic intermediate would give the same
product and cannot be excluded on the basis of product stud-
ies alone.
Irradiation of 4 in 1:1 H2O–CH3OH (~10–3 mol L–1,
Rayonet photochemical reactor, 300 nm lamps, 1 min) did
indeed give 29 in 48% yield (eq. [5]), while solutions kept in
the dark did not react. The photosubstitution is clean and has
a remarkably high quantum yield for production of 29 from
4, measured to be 0.84 0.04 (Table 1). This value repre-
sents the highest reported quantum yield to date for
photomethanolysis of benzyl alcohol derivatives. The high
quantum yield for formation of 29 implies that attack by wa-
ter on 27 to regenerate starting material must be very ineffi-
cient. This is in line with the known high nucleophilicity of
methanol relative to that of water (20). Pure samples of 29
were obtained by chromatographic separation of the photo-
lysate, and photolysis of these samples in 1:1 aqueous aceto-
nitrile regenerated 4. The methyl ether product can therefore
also undergo photosubstitution, presumably through formal
Photodehydroxylation and photodehydration of 4 and 5
We have previously shown that the photolysis of phenols
containing benzylic alcohol groups gives rise to ortho-, meta-,
and para-quinone methides in high yield (1). In an attempt
© 2004 NRC Canada

Lukeman et al.
245
loss of methanol to regenerate 27. This result shows that
ether and ester derivatives of 4 might prove useful as pro-
tecting groups for photorelease of alcohols and acids.
Results from the deuterium incorporation studies of 1
show that electron density in singlet excited 1-naphthols is
primarily localized at the 5-position, and to a lesser extent at
the 8-position. Naphthols 4 and 29 both react through the
substituent located at the 5-position. In contrast to these ex-
amples, the substituent in 5 is located at the 4-position. No
deuterium was incorporated at this position on irradiation of
1, indicating that minimal charge is transferred to this posi-
tion on excitation. When 5 was irradiated under similar con-
ditions as 4, no reaction was observed, showing that
photodehydroxylation requires excited state charge transfer
from the naphthol OH.
tion was studied to provide mechanistic insights. If the
ESIPT reaction occurs from S1, as do most ESPTs from phe-
nols, addition of water to the solvent mixture should lead to
quenching of the fluorescence intensity by enhancing the
proton transfer pathways. Indeed, the fluorescence emission
of 1 (in CH3CN), after an initial red shift from λmax 340 nm
to λmax 355 nm on addition of small amounts of water, was
progressively quenched by added water, concomitant with a
growth in fluorescence intensity at 460 nm. In pure water,
the 355 nm band was completely quenched and only emis-
sion from the 460 nm band was observed. The band at
460 nm results from emission from the excited 1-
naphtholate ion (1–), which arises from ESPT from the OH
proton of 1 to solvent (water), and is consistent with the pre-
vious assignment (8). Relatively large amounts of water
were required for efficient formation of 1–, just as the ESIPT
of 1 to give 1-5D and 1-8D required large amounts of D2O.
This correlation is not surprising as the two processes are
closely related. Thus, quenching of the fluorescence emis-
sion of 1 in the presence of water arises from a combination
of two competing pathways from S1, namely ESPT to sol-
vent water and water-mediated ESIPT to the 5- and 8-
positions.
Photochemical studies of 9 and 10
The proposed mechanism for the photosolvolytic rear-
rangement of 6 (11) that leads to the release of acetic acid
appears to require either ESPT to solvent water or an ESIPT
from the phenol to the carbonyl oxygen mediated by water,
to generate a reactive excited phenolate or a related enol
species, respectively, prior to formation of spiroketone 7. It
was of interest to determine whether derivatives such as
naphthol 9 and phenol 10 would show similar photosolvo-
lysis. Since the parent 1-naphthol (1) has already been
shown to undergo a water-mediated ESIPT to the 5- and 8-
positions, it seemed reasonable to expect that a similar pro-
cess could operate in 9, with the difference that the proton
transfer would end up at the carbonyl oxygen. This could ac-
tivate the compound for release of acetic acid in a manner
similar to the reactivity observed for 6. In the case of phenol
10, it is well known that there is substantial charge transfer
through meta-positions on the benzene ring in S1 that allows
efficient formation of m-quinone methides (1, 2, 5). Such
charge transfer would be expected to enhance ESIPT from
the phenol to the carbonyl moiety of 10, resulting in a reac-
tion similar to that observed for 6. To our surprise, both 9
and 10 were found to be completely unreactive in 1:1 H2O–
CH3CN even on extended irradiation. Complete recovery of
the substrates was achieved in all cases. What was surprising
was the lack of photodecomposition even on extended
photolysis, which leads us to suggest that ESPT or ESIPT
processes are occurring for these compounds, but lead to no
net chemistry. In the case of 10, subsequent formation of the
non-Kekulé spiroketone intermediate seems to be prohibitive
as it would be a ground-state species. For 9, formation of the
requisite spiroketone would seem possible but on closer ex-
amination, such a reaction would result in loss of two aro-
matic rings and formation of a strained spiroketone.
Moreover, the excited singlet or triplet energy of 9 (a 5-
hydroxy-1-acetylnaphthalene) is expected to be substantially
lower than that of 6 (a 4-hydroxyacetophenone). One of
these factors probably explains the lack of reaction of 9. No
further studies of 9 and 10 were pursued due to their
photoinertness in aqueous solution.
The fluorescence behaviour of 5 closely resembled that of
1. Small amounts of water led to an initial red shift of the
fluorescence emission band from λmax 358 nm in neat
CH3CN to λmax 375 nm (Fig. 4). Larger amounts of water
led to quenching of this fluorescence band and to a growth
in emission at 465 nm (from naphtholate ion 5–). This infor-
mation suggests that there are no major pathways that deac-
tivate the singlet excited state of 5 other than those available
for 1.
The fluorescence behaviour of 4 is quite different from
that observed for 1 and 5. Small amounts of water led to a
red shift of the fluorescence emission band from λmax
345 nm in neat CH3CN to λmax 360 nm, though significant
quenching of this band was achieved with less water than re-
quired for 1 or 5 (Fig. 5). Naphtholate (4–) emission (λmax
460 nm) was observed only at high water concentrations and
was much weaker than the observed emission of 1– or 5–.
Modified Stern–Volmer quenching plots of the emission in-
tensity of 4 and 5 with added water are shown in Fig. 6. The
enhanced quenching by water for 4 relative to 5 can be
clearly seen. Since the pKa values for 1, 4, and 5 are ex-
pected to be approximately equivalent, an additional water-
dependent deactivation pathway must be operating for 4.
The net dehydration reaction to give 27 that is available for
4 but not for 1 or 5 can explain this additional quenching if
deprotonation of the naphthol and dehydroxylation of the
benzylic alcohol are concerted from S1. This pathway would
also compete with ESPT to solvent, explaining the weaker
phenolate emission observed. Both Stern–Volmer plots ex-
hibit a squared dependence on water concentration. Our
group has previously reported nonlinear fluorescence
quenching by water for related phenols, an effect attributable
to involvement of water clusters in the ESPT (4).
Fluorescence quantum yields and lifetimes were measured
for 4 and 5 in CH3CN and are listed in Table 1, along with
literature values for 1 and 2. Both 4 and 5 exhibit similar
photophysical behaviour to 1 in CH3CN, indicating that no
Fluorescence measurements
As the ESIPT reaction of 1 is strongly mediated by water,
the fluorescence behaviour with respect to water concentra-
© 2004 NRC Canada

246
Can. J. Chem. Vol. 82, 2004
Fig. 4. Representative fluorescence quenching traces of 5 by
added water in CH3CN (λex 300 nm). Water concentrations are
given in the legend. Naphtholate (5–) emission grows in at
λem 460 nm with increasing water content.
Fig. 6. Stern–Volmer plots for steady-state fluorescence quench-
ing of 4 (triangles) and 5 (diamonds) by H2O (in CH3CN). The
lines drawn are fits to I0/I = 1 + KSV[H2O]2.
Fig. 5. Representative fluorescence quenching traces of 4 by
added water in CH3CN (λex 300 nm). Water concentrations are
given in the legend. Naphtholate (4–) emission grows in at λem
465 nm with increasing water content.
Table 2. Summary of laser flash photolysis transient data.a
τc (µs)
kq (106 M–1 s–1)
b
d
Transient
λmax (nm)
26a
27
32
350, 530
410, 700
410, 700
350, 530
17
33
1.7
45
0.90 0.10
1.0 0.1
>>50e
33
Not determined
aLFP was performed on flowing solutions that were ~10–5 mol L–1 in
the appropriate substrate. Excitation was achieved with an excimer laser
(308 nm, ~ 10 ns, <30 mJ/pulse).
bAbsorption maxima in H2O–CH3CN (1:1) solution.
cTransient lifetime in H2O–CH3CN (1:1) solution, oxygen purged. The
estimated error is 10% of the reported value.
dBimolecular quenching rate constant with added ethanolamine in H2O–
CH3CN (1:1) solution, according to the following equation: kobs = k0 +
kq[Q].
eComplete quenching was observed with 5.5 mol L–1 ethanolamine.
Transients with lifetimes as short as 20 ns are readily detectable on the
system used, so for complete quenching to be realized with 5.5 mol L–1
quencher, kq must be at least 5 × 107 M–1 s–1, and is probably at least an
order of magnitude higher.
λmax 360 nm, both of which were assigned to cation 30
obtained via photoprotonation (from HFIP) of 1-
methoxynaphthalene at the 5-position (eq. [6]). This tran-
major deactivation pathways open up on addition of the
hydroxyethyl substituent to the 1-naphthol chromophore ex-
cept when water is present.
[6]
Laser flash photolysis (LFP)
LFP experiments were performed for 1, 4, and 5 to gain
direct evidence of the suspected naphthoquinone methide in-
termediates. The resulting data are summarized in Table 2.
Mathivanan et al. (19b) previously studied 1-
methoxynaphthalene by LFP in hexafluoroisopropyl alcohol
(HFIP) and observed a transient with a broad absorption
band at λmax 550 nm and a second, more narrow band at
sient could not be detected in aqueous acid, which the au-
thors proposed was because rapid reaction with water
reduced the lifetime of the cation to a value shorter than
their system could detect. This intermediate bears structural
© 2004 NRC Canada

Lukeman et al.
247
Fig. 7. Transient absorption observed on LFP of 1 in H2O–
CH3CN (1:1) (triangles) and in neat CH3CN (diamonds)
(λex 308 nm, O2 purged for both). For both spectra, the decay
trace was taken immediately after the laser pulse (~10 ns).
resemblance to suspected naphthoquinone methide 26a re-
sulting from ESIPT to the 5-position in 1. Chiang et al. (21)
observed naphthoquinone methide 31 by LFP in dilute aque-
ous acid solution (generated by γ-hydrogen abstraction in 5-
methyl-1,4-naphthoquinone (eq. [7])). This transient had a
broad visible absorption band at λmax 600 nm with a lifetime
[7]
of ≈1 µs (lifetime shortened as pH was decreased), and is
structurally related to 27 produced on photodehydration of 4.
Nanosecond LFP of 1 in 1:1 H2O–CH3CN (O2 saturated)
resulted in the formation of a transient with a broad absorp-
tion band at 530 nm and a second, more narrow band at
340 nm (Fig. 7). Both peaks appeared immediately after the
laser pulse (≈10 ns) and decayed to baseline. The decay ki-
netics of both peaks fit a monoexponential curve, giving a
lifetime of 16 2 µs. This transient spectrum is very similar
to that assigned to 30 (19b), though blue-shifted by about
20 nm. We assign this absorption to be 26a formed on
ESIPT of 1. The observed lifetime of 16 2 µs in aqueous
solution indicates that this transient is much less reactive
with water than 30, which is expected because of the greater
electrophilic nature of 30. Naphthoquinone methide 26b is
non-Kekulé and is expected to decay within the time of the
laser pulse (10 ns). LFP of 1 in neat CH3CN (O2 purged) led
to formation of a weaker transient (Fig. 7) with a broad visi-
ble band (λmax 510 nm) and narrower bands with λmax 390
and 350 nm. We assign this transient to the 1-naphthoxy rad-
ical based on the similarity of the observed absorption spec-
trum to that known for this species (22).
Fig. 8. Transient absorption observed on LFP of 5 in H2O–
CH3CN (1:1) (triangles) and in neat CH3CN (diamonds) (λex
308 nm, O2 purged for both). The decay traces were taken im-
mediately after the laser pulse (~10 ns).
=
Nanosecond LFP of 4 in 1:1 H2O-CH3CN (O2 saturated)
resulted in the formation of a strongly absorbing transient
immediately after the laser pulse, with peaks at 410 and
700 nm (Fig. 8). The decay kinetics gave a biexponential fit
with a short-lived minor component (τ = 1.5 0.5 µs) and a
longer lived major component (τ = 34 2 µs) that both de-
cayed cleanly to baseline. This biphasic decay was observed
at all wavelengths, demonstrating that the two decaying spe-
cies have similar absorbance profiles. The position of the
bands is consistent with what is expected for 27: the red
shift relative to 26a results from the extra unit of conjuga-
tion and the extra methyl substituent. The long-wavelength
band is similar to that reported for related naphthoquinone
methide 31 by Chiang et al. (21). The major (long-lived)
component of the observed absorption is assigned to 27 on
this basis and on the results of the product studies. The
shorter lived species can be assigned to cation 32 that would
result from dehydroxylation without prior deprotonation of
the phenol, a process well known for methoxy-substituted
benzyl alcohols. This cation possesses a similar degree of
conjugation as 27, which would explain its similar
absorbance characteristics. Its much shorter lifetime is con-
sistent with the expected enhanced reactivity of a charged
species relative to a neutral and closed-shell naphthoquinone
methide. No strong transients were observed on LFP of 4 in
neat CH3CN (O2 purged) in the absence of water. Instead, a
weak transient was observed with a broad band at 580 nm,
and with narrower bands at 400 and 350 nm (Fig. 8). This
species closely resembles the transient observed for 1 in neat
CH3CN in the position, structure, and intensity of the ab-
sorption bands. This transient is therefore assigned to the 1-
naphthoxy radical of 4.
A strong transient was also observed on LFP of 5 (in
H2O–CH3CN (1:1)), which exhibited a broad absorption
© 2004 NRC Canada

248
Can. J. Chem. Vol. 82, 2004
Fig. 9. Quenching plot showing the increase in the decay rate of
27 at 680 nm with added ethanolamine in H2O–CH3CN (1:1).
Line is the least-squares fit to the data and gives kq = 1.0 ×
106 M–1 s–1. Estimated error is 10%.
band at 520 nm, a small band at 410 nm, and a strong nar-
row band at 350 nm. Again, all peaks displayed first-order
decay kinetics, each with the same lifetime of 45 2 µs.
Product studies showed that
5
does not undergo
photosubstitution; therefore, this transient is unlikely to be
28. Nevertheless, the spectrum closely resembles that as-
signed to 26a. Indeed, 5 should be able to form a similar
naphthoquinone methide resulting from ESIPT to the 5-
position of the naphthol ring. On this basis, we assign this
absorption to naphthoquinone methide 33. The enhanced
lifetime of 33 relative to 26a is probably due to steric shield-
ing of the 5-position by the hydroxyethyl substituent that re-
tards deprotonation at this site. LFP of 5 in neat CH3CN
shows formation of small amounts of the corresponding 1-
naphthoxy radical.
Additional support for our assignment of the major tran-
sients observed for 1 and 4 as naphthoquinone methides was
sought through quenching studies. We have recently shown
(23) that ethanolamine can efficiently quench quinone
methide intermediates. For 27 (or cation 32), the expected
quenching mechanism is nucleophilic attack at the benzylic
position to furnish the ethanolamine adduct, while for
quinone methide 26a, the expected quenching mechanism is
base-catalyzed tautomerization. Quenching analyses were
first performed for the transients arising from LFP of 4
(Fig. 9). At the smallest ethanolamine concentration used
(5.5 mM in H2O–CH3CN (1:1)), the short-lived component
of the decay (assigned to 32) was completely quenched,
leaving only the longer lived component (assigned to 27). As
carbocations are known to react very rapidly with added
nucleophiles, this observation supports our assignment of the
short-lived component to 32. The dynamic quenching of
quinone methide 27 with added ethanolamine is plotted in
Fig. 9 and a linear correlation was observed, giving a bimo-
lecular quenching rate constant of 1.0 0.1 × 106 M–1 s–1. A
similar quenching analysis was performed for the transient
formed on LFP of 1 (assigned to 26a), and results are pre-
sented in Fig. 10. The decay of this transient was also accel-
erated with added ethanolamine, giving a good linear fit with
kq at 9.0 1.0 × 105 M–1 s–1.
Fig. 10. Quenching plot showing the increase in the decay rate
of 26a at 530 nm with added ethanolamine in H2O–CH3CN
(1:1). Line is the least-squares fit to the data and gives kq =
9.0 × 105 M–1 s–1. Estimated error is 15%. Outlined point was
not included in the least-squares analysis.
Mechanisms of reaction
The deuterium incorporation reaction of 1 on photolysis
in D2O–CH3CN (1:1) is proposed to occur via a water-
mediated ESIPT from the naphthol OH to the aromatic ring
carbons at the 5- and 8-positions, to give naphthoquinone
methide intermediates 26a and 26b, respectively,
(Scheme 4). The enhanced acidity of the naphthol OH in the
excited state (pKa* = 0.4) and the enhanced excited-state ba-
sicity of the 5- and 8-positions of 1-alkoxy- and 1-hydroxy-
substituted naphthalenes suggest that an intramolecular
acid–base reaction might be possible. The propensity of 1 to
undergo excited-state proton transfer reactions in aqueous
conditions is further demonstrated in fluorescence studies,
with emission from (1–)* observed on photolysis of 1 in
aqueous conditions. The protonation of the 5- and 8-
positions in the pH range 2.5–9 does not result from
electrophilic attack by H3O+ (or D3O+) since the exchange
efficiency is independent of pH (pD) throughout this large
pH range. Solvent water alone cannot be acting as the acid,
since no exchange was observed for irradiated (basic) solu-
tions of 1–. Under basic conditions, the naphthalene still has
a powerful electron-donating group at the 1-position, and the
concentration of water is still very high. It is thus evident
that the reaction requires the OH functionality, suggesting
that the reaction proceeds through ESIPT from the OH to the
ring carbons. The need for water to mediate the proton trans-
fer was clearly borne out by the product studies, with effi-
cient reaction only being observed at fairly high water
concentrations. This is because the OH group is too far from
the basic sites to allow any direct spatial interaction, so wa-
ter is needed to “solvate” the proton as it is transferred. We
have previously implicated water as mediating ESIPTs be-
tween remote sites in other phenols (4). Direct evidence for
© 2004 NRC Canada

Lukeman et al.
249
Scheme 4.
Scheme 5.
naphthoquinone methide intermediate 26a that would arise
from ESIPT to the 5-position was provided by LFP studies,
lending further support to our mechanistic proposal.
An additional consideration of interest in the ESIPT of 1
is the possible presence of other intermediates prior to
naphthoquinone methide formation. We recently reported an
example of a simple hydroxybiaryl for which ESIPT to two
aromatic carbon positions was observed (24). ESIPT to both
positions was strongly water dependent, as is the ESIPT of 1
in the present work. Furthermore, the efficiency of exchange
of the two positions had almost identical dependencies on
water content, which is unlikely if the two routes are wholly
independent. To account for this, we suggested that ESIPT
first leads to a “proton – π complex”, which can rearrange to
either of the two possible quinone methides (24). The ESIPT
in 1 appears not to occur through this route. Deuterium ex-
change efficiencies at the 5- and 8-positions appear to have
different dependencies on water concentration, with ex-
change at the 8-position reaching saturation with less water
than exchange at the 5-position. This observation favours a
mechanistic picture where two separate ESIPTs are occuring
that do not share a common intermediate prior to QM forma-
tion.
The photosubstitution reaction of 4 is mechanistically re-
lated to the ESIPT reaction of 1 in that it relies on excited-
state charge transfer from the naphthol OH to the 5-position.
The photochemical reactivity of 4 in aqueous solution can be
explained as occuring through three different pathways
(Scheme 5): (i) ESPT of the naphthol proton to solvent wa-
ter to give 4–; (ii) dehydroxylation of the benzylic alcohol
moiety to give 32; and (iii) photodehydration to form 27.
Evidence for the first pathway comes from the fluorescence
studies: emission from 4–* was observed on excitation of 4
at high water concentrations, indicating that an adiabatic
ESPT is occuring. As for the second pathway, a short-lived
transient observed in LFP studies was assignable to the cat-
ion 32 that would result from simple dehydroxylation. Cat-
ion 32 might also undergo deprotonation to give 27. The
third and major pathway involves deprotonation of the
naphthol OH concerted with dehydroxylation of the benzylic
site to give 27 directly. The formation of the naphtho-
quinone methide intermediate 27 was inferred from the re-
sults of the product studies and the observation of a transient
on LFP of 4 that was assignable to 27. That the reaction to
give 27 is concerted was inferred primarily from the fluores-
cence studies. The enhanced quenching of the fluoresence
emission of 4 in aqueous solution by water over 1-naphthol
derivatives 1 and 5 shows that an additional deactivation
pathway is available for 4 that is not available for 1 or 5,
with this pathway being photodehydration to form 27. If the
ESPT to solvent preceded dehydroxylation of the benzylic
alcohol, no enhanced quenching would be observed, since 1,
4, and 5 would be expected to have similar ESPT rates. The
LFP quenching studies for 4 demonstrated that very small
amounts of ethanolamine rapidly quenched the cation. If the
cation was a precursor to 27, the presence of ethanolamine
would substantially reduce the amount of 27 that is ob-
served, and that was not the case. Therefore, the only re-
maining option to explain the reactivity is via a concerted
mechanism.
The reactivity of 5 (Scheme 6) is different from that of 4
in that no photosubstitution was observed in the product
studies. Photodehydration to form naphthoquinone methide
28 therefore does not occur. The ESIPT in 1 and the photo-
substitution reaction of 4 are both interpreted as arising from
enhanced localized charge transfer from the naphtholic OH
to the 5- and 8- ring positions. The failure of 5 to react
through the 4-position confirms that this charge transfer is
crucial for photodehydration. A strong transient was ob-
served on LFP of 5 and was assigned to 33, indicating that 5
reacts though an ESIPT pathway similar to that observed for
© 2004 NRC Canada

250
Can. J. Chem. Vol. 82, 2004
Scheme 6.
about 5 mL, diluted with water and extracted with dichloro-
methane. The extracts were washed with brine, dried and
evaporated, and crystallized from hexanes to give methyl 5-
methoxy-1-naphthoate 13. Yield 3.43 g (70%); mp 76–
76.5 °C (lit. (25) mp 74.5–75.5 °C). This material (7.7 g)
was saponified in aqueous methanolic sodium hydroxide to
give 5-methoxy-1-naphthoic acid 14. Yield: 6.1 g (85%); mp
225–227 °C (from ethanol) (lit. (26) mp 227–229 °C).
A solution of 14 (2.02 g, 10 mmol) stirred under nitrogen
in anhyd THF (40 mL) was diluted with anhydrous diethyl
ether (110 mL) and a 1.5 mol L–1 solution of methyl lithium
in diethyl ether (16.7 mL, 25 mmol) was added over 20 min
at room temperature. The mixture was stirred for 1.5 h and
acetone (17 mL) was added. After 5 min, the solution was
diluted with 2 mol L–1 aqueous hydrochloric acid and ex-
tracted with diethyl ether. The organic phase was washed
with water, aqueous sodium hydrogen carbonate, water and
brine, dried and evaporated under reduced pressure. The res-
idue was crystallized from petroleum ether (bp 60–80 °C) to
give 1-acetyl-5-methoxynaphthalene 15. Yield: 1.56 g
(78%), mp 85–87 °C (lit. (27) mp 88 to 89 °C).
1. Despite containing a potentially reactive substituent at the
4-position, the chemistry is still dominated by the localized
charge transfer to the 5- and 8-positions.
Dry N-methyl-2-pyrrolidinone (10.5 mL) was added under
nitrogen to a mixture of 15 (4.2 g, 21.0 mmol) and anhy-
drous potassium carbonate (0.145 g, 1.05 mmol) and the
mixture was stirred for 5 min at room temperature.
Thiophenol (3.23 mL, 31.5 mmol) was added and the mix-
ture was heated at 190 °C for 45 min and cooled. Methanol
(105 mL) and triethylamine (4.2 mL) were added, followed
by 30% hydrogen peroxide (4.2 mL) in four equal portions
over ~5 min. The solution was stirred for 10 min, then con-
centrated under vacuum to ~10 mL, diluted with 5% aque-
ous sodium hydroxide and extracted with diethyl ether. The
aqueous layer was acidified to pH 2 with 6 mol L–1 hydro-
chloric acid and extracted with diethyl ether and the organic
extract was washed with brine, dried and evaporated under
reduced pressure. The residue was crystallized from EtOAc–
hexanes to give 1-acetyl-5-hydroxynaphthalene 16 as pale
yellow crystals. Yield: 2.82 g (72%); mp 173–174.5 °C (lit.
Experimental
General
1H NMR spectra were obtained on Bruker AC300
(300 MHz), AM360 (360 MHz), AVANCE 500 (500 MHz),
Varian AM400 (400 MHz), or JEOL FX90Q (90 MHz) in-
struments in CDCl3 unless otherwise specified. Low-
resolution mass spectra were obtained on a Finnigan 3300
(CI) and HR-MS were obtained on a Kratos Concept H (EI)
instrument. UV–vis spectra were recorded on a Varian Cary
5 instrument. Acetonitrile (HPLC or spectral grade) used for
fluorescence and LFP experiments was dried by distillation
from calcium hydride. Organic extracts were dried over an-
hydrous magnesium sulfate and filtered. pD values were
measured for the aqueous portion (prior to mixing with
CH3CN cosolvent) using a Fisher Accumet 915 pH meter
and were not corrected for the deuterium isotope effect. So-
lutions were unbuffered with pD values adjusted with either
NaOD (pD > 7) or D2SO4 (pD < 7).
1
(28) mp 172.5–173.5 °C). H NMR (400 MHz, CDCl3
– 10% CD3OD) δ: 2.74 (s, 3H, CH3), 6.90 (d, J = 7.6 Hz,
1H, Ar-H), 7.40 (m, 1H, Ar-H), 7.47 (m, 1H, Ar-H), 7.92 (d,
J = 7.1 Hz, 1H, Ar-H), 8.14 (d, J = 8.6 Hz, 1H, Ar-H), 8.49
(d, J = 8.6 Hz, 1H, Ar-H).
A solution of 16 (1.60 g, 8.60 mmol) in EtOH (59 mL)
was stirred at 0 °C and sodium borohydride (0.57 g,
15.1 mmol) was added. After 10 min, the solution was al-
lowed to warm to room temperature and stirred for 1.5 h,
then concentrated to ~10 mL. Unreacted borohydride was
destroyed by dropwise addition of glacial acetic acid and the
solution was diluted with water and extracted with ethyl ace-
tate. The organic phase was washed with brine, dried and
evaporated under reduced pressure. The residue was purified
by flash chromatography (EtOAc–hexanes (3:7)) to give the
title alcohol 4 as beige crystals. Yield: 0.89 g (55%); mp
Materials
5-(1-Hydroxyethyl)-1-naphthol (4)
A solution of 5-bromo-1-naphthoic acid 11 (12) (4.5 g,
17.9 mmol) and concentrated sulfuric acid (1.35 mL) in
methanol (65 mL) was heated at reflux for 8 h, cooled, di-
luted with diethyl ether and washed with aqueous sodium
hydrogen carbonate, water and brine, dried and evaporated
under reduced pressure. The residue was crystallized from
methanol to give methyl 5-bromo-1-naphthoate 12. Yield:
3.5 g (74%); mp 66 to 67 °C (lit. (12) mp 66 to 67 °C).
A mixture of 12 (6.0 g, 22.6 mmol), CuBr (0.93 g,
6.5 mmol), ethyl acetate (1.77 mL, 18.1 mmol) and 25%
sodium methoxide in methanol (Aldrich; 30.8 mL,
142.6 mmol) was heated under reflux for 3.5 h under nitro-
gen. The copper salts were filtered and the solid was washed
with methanol. The combined filtrate was concentrated to
1
167.5–169 °C (from EtOAc–hexanes). H NMR (500 MHz,
CDCl3 – 10% CD3OD) δ: 1.63 (d, J = 6.0 Hz, 3H, Me), 5.62
(q, J = 6.0 Hz, 1H, CHOH), 6.85 (d, J = 7.7 Hz, 1H, Ar-H),
7.33 (m, 1H, Ar-H), 7.45 (t, J = 7.7 Hz, 1H, Ar-H), 7.58 (d,
J = 8.5 Hz, 1H, Ar-H), 7.68 (d, J = 6.8 Hz, 1H, Ar-H), 8.20
(d, J = 8.6 Hz, 1H, Ar-H). HR-MS calcd. for C12H12O2:
188.0837; observed: 188.0838.
© 2004 NRC Canada

Lukeman et al.
251
CD3OD) δ: 2.42 (s, 3H, CH3), 5.66 (s, 1H, CHN2), 7.23–
8.40 (m, 6H, Ar-H).
4-(1-Hydroxyethyl)-1-naphthol (5)
1-Acetyl-4-methoxynaphthalene 17 (29) was demethyl-
ated as described above for the 5-isomer 15 to give 1-acetyl-
4-hydroxynaphthalene 18 as yellow crystals (from aqueous
methanol), mp 198–199.5 °C (lit. (30) mp 197 °C).
A solution of 23 (1.43 g, 5.36 mmol) in a mixture of THF
and methanol (1:1; 28.6 mL) was treated with concentrated
aqueous ammonia (4.77 mL). After 2 h at room temperature,
the solution was diluted with ice water and adjusted to
pH 8.0 with saturated aqueous oxalic acid. The aqueous
layer was extracted with ethyl acetate and the organic extract
was washed with water, dried and evaporated under reduced
pressure. The residue was flash chromatographed (ethyl ace-
tate – hexanes (3:7)) to give 2-diazo-1-(5-hydroxy-1-
naphthalenyl)ethanone 24 as a yellow solid. Yield: 0.56 g
(47%). 1H NMR (90 MHz, CDCl3 – 10% CD3OD) δ: 5.73 (s,
1H, CHN2), 6.80–8.44 (m, 6H, Ar-H). This material (0.47 g,
2.17 mmol) was added in portions over ~3 min to a stirred
mixture of concentrated hydrochloric acid (0.44 mL) and
glacial acetic acid (4.4 mL). After 0.5 h, the solution was di-
luted with water (22 mL) and stirred for 10 min. The precip-
itated solid was filtered, washed with cold water (3 mL), and
dried under vacuum to give crude 1-chloroacetyl-5-
hydroxynaphthalene 25 as a yellow powder. Yield: 0.39 g
A solution of 18 (2.0 g, 10.75 mmol) and TBDMS chlo-
ride (2.89 g, 19.2 mmol) in anhyd DMF (9.3 mL) was
treated with imidazole (2.72 g, 40.0 mmol) and the mixture
was stirred at room temperature for 2 h, then diluted with di-
ethyl ether. The diluted solution was washed with 0.1 mol L–1
aqueous sodium carbonate and brine, dried and evaporated
under reduced pressure to give crude 4-tert-
butyldimethylsilyloxy-1-acetylnaphthalene 19 as a beige
solid. Yield: 2.58 g (80%). This material was used without
further purification. Thus, a solution of 19 (3.33 g,
11.1 mmol) in methanol (86 mL) was stirred at 10 °C and
treated with sodium borohydride (0.42 g, 11.1 mmol) in
three portions. The solution was stirred for 10 min at 10 °C
and for 45 min at room temperature, then concentrated to
~30 mL and treated dropwise with glacial acetic acid
(1.5 mL). The solution was diluted with water and extracted
with ethyl acetate. The organic phase was washed with brine,
dried and evaporated under reduced pressure to give crude 1-
(4-tert-butyldimethylsilyloxy-1-naphthalenyl)ethanol 20 as a
reddish brown viscous gum. Yield: 3.0 g (89%). This mate-
rial was dissolved in THF (30.5 mL) and treated with 1 mol
L–1 tetrabutylammonium fluoride in THF (10.93 mL). After
20 min at room temperature the solution was diluted with di-
ethyl ether and washed with water and brine, dried and evap-
orated under reduced pressure. The residue was flash
chromatographed (EtOAc-hexanes (3:7)) to give the title al-
cohol 5 as a brown solid. Yield: 1.64 g (88%). Crystalliza-
tion from EtOAc–hexanes was accompanied by some
decomposition in the mother liquor but gave the pure alco-
1
(82%). H NMR (90 MHz, CDCl3 – 5% CD3OD) δ: 4.76 (s,
2H, CH2Cl), 6.82–8.55 (m, 6H, Ar-H).
A stirred solution of 25 (0.39 g, 1.77 mmol) and glacial
acetic acid (0.132 mL, 2.3 mmol) in dry toluene (23 mL)
was cooled in ice and treated with DBU (0.344 mL,
2.3 mmol). The solution was stirred in the ice bath for 1 h,
then at room temperature overnight. The solution was di-
luted with water, the aqueous layer was adjusted to pH 2
with 1 mol L–1 hydrochloric acid, and the mixture was ex-
tracted with diethyl ether. The organic layer was washed
with water and brine, dried and evaporated under reduced
pressure. The residue was dissolved in ethanol (7 mL),
thiourea (0.15 g) was added and the solution was heated at
reflux for 0.5 h to remove residual 25, then concentrated un-
der reduced pressure and partitioned between ethyl acetate
and water. The organic phase was washed with water and
brine, dried and evaporated and the residue was flash
chromatographed (ethyl acetate – hexanes (3:7)). The recov-
ered material was crystallized from toluene to give the title
compound 9 as yellow plates. Yield: 0.29 g (50%); mp 99.5–
1
hol, mp 126–127.5 °C. H NMR (500 MHz, CDCl3 – 10%
CD3OD) *: 1.63 (d, J = 6.5 Hz, 3H, CH3), 5.57 (q, J =
6.5 Hz, 1H, CHOH), 6.83 (d, J = 7.7 Hz, 1H, Ar-H), 7.44–
7.47 (m, 2H, Ar-H), 7.49–7.53 (m, 1H, Ar-H), 8.07 (d, J =
8.4 Hz, 1H, Ar-H), 8.27–8.29 (m, 1H, Ar-H). HR-MS calcd.
for C12H12O2: 188.0837; observed: 188.0837.
1
101 °C. H NMR (400 MHz) δ: 2.25 (s, 3H, CH3), 5.31 (s,
2-(5-Hydroxy-1-naphthalenyl)-2-oxoethyl acetate (9)
2H, CH2), 6.86 (d, J = 7.5 Hz, 1H, Ar-H), 7.39 (dd, J = 7.8
and 8.4 Hz, Ar-H), 7.48 (dd, J = 7.4 and 8.3 Hz, Ar-H), 7.83
(d, J = 7.0 Hz, Ar-H), 8.14 (d, J = 8.6 Hz, Ar-H), 8.44 (d,
J = 8.4 Hz, Ar-H). Anal. calcd. for C14H12O4: C 68.85, H
4.95; found: C 68.57, H 4.89.
A stirred solution of 5-acetoxy-1-naphthoic acid 21 (31)
(1.75 g, 7.6 mmol) in dry dichloromethane (30 mL) was
treated at room temperature with oxalyl chloride (1.3 mL,
14.9 mmol) and dry DMF (65 µL). After 1 h, the solution
was evaporated under reduced pressure and kept overnight
under vacuum to leave the crude acid chloride 22. Yield:
1.95 g (100%). This material (1.89 g, 7.6 mmol) was dis-
solved in a mixture of anhyd THF and acetonitrile (1:1;
15 mL) and treated with triethylamine (2.11 mL,
15.1 mmol). The solution was cooled in an ice bath and a
2 mol L–1 solution of trimethylsilyldiazomethane in hexane
(Aldrich; 7.6 mL, 15.2 mmol) was added over 15 min. After
0.5 h, the solution was allowed to warm to room temperature
and stirred overnight, then diluted with diethyl ether and
washed with sat NaHCO3, water, and brine, dried and evapo-
rated under reduced pressure. The residue was triturated
with diethyl ether to give crude diazoketone 23 as a brown
2-(3-Hydroxyphenyl)-2-oxoethyl acetate (10)
A stirred solution of 3-hydroxyphenacyl bromide (32)
(4.0 g, 18.6 mmol) and glacial acetic acid (2.45 mL,
42.8 mmol) in acetonitrile (63 mL) was cooled in an ice bath
and treated dropwise with triethylamine (5.44 mL,
39.1 mmol). The solution was allowed to warm to room tem-
perature and stirred for 2 h, then diluted with brine and ex-
tracted with diethyl ether. The extract was washed with
1 mol L–1 aq. HCl and brine, dried and evaporated under re-
duced pressure to give a brown gum (3.3 g) that was dis-
solved in boiling water, decolorized with charcoal, and
extracted with diethyl ether. The extract was dried and evap-
orated and the residue was crystallized from ethyl acetate –
1
solid. Yield: 1.43g (74%). H NMR (90 MHz, CDCl3 – 5%
© 2004 NRC Canada

252
Can. J. Chem. Vol. 82, 2004
hexanes to give the title compound 10. Yield: 0.9 g (25%);
300 MHz 1H NMR.
decomposition was observed when NaHCO3 was not added
to the solution.
A
small amount of thermal
1
mp 94 °C. H NMR (400 MHz, CDCl3 – 10% CD3OD) δ:
2.21 (s, 3H, CH3), 5.30 (s, 2H, CH2), 6.94–7.46 (m, 4H, Ar-
H). Anal. calcd. for C10H10O4: C 61.85, H 5.19; found: C
61.56, H 5.16.
Photolysis of 9 and 10
In a typical procedure, 2–5 mg of substrate was dissolved
in ~2 mL of D2O–CD3CN (1:1). The solution was trans-
ferred to an NMR tube and deoxygenated by purging with
argon for 10 min. The tube was then placed in the middle of
the Rayonet reactor (16 × 300 nm lamps) and irradiated for
Product studies
Preparative photolyses for 1, 2, 4, and 5 were carried out
with samples (10–20 mg) dissolved in the appropriate sol-
vent (40–100 mL) and transferred to a quartz tube. The solu-
tion was irradiated in a RPR-100 photochemical reactor
fitted with 254 or 300 nm lamps with continuous cooling (by
a cold finger) and purging by a stream of argon for approxi-
mately 10 min before and continuously during irradiation
(via a long stainless steel needle). Photolysis times ranged
from 3 to 30 min, depending on the conversion desired, the
efficiency of the reaction, and the size of the sample. After
photolysis, aqueous samples were worked up by extraction
with CH2Cl2 after addition of aq. NaCl. Direct evaporation
of the organic solvent was used when samples were irradi-
ated in wholly organic solvents. Typical experiments are de-
scribed below. Analytical scale photolyses were employed
for 9 and 10, using NMR tubes, as described below.
1
up to 30 min. H NMR spectra were taken before and after
photolysis without workup. Both 9 and 10 were found to be
photochemically inert for up to 30 min photolysis. Under
these conditions, 6 would have resulted in quantitative con-
version to 8 in about 10 min of photolysis.
Quantum yields
The quantum yield for total deuterium incorporation in 1
has been previously reported (8) to be ΦH-D = 0.11 in 99%
D2O, pD 3. Our results demonstrate that the exchange effi-
ciency is the same at pD 7 and pD 3, and that the exchange
efficiency appears to plateau around 1:1 H2O–CH3CN.
Given these data, it is reasonable to assume that ΦH-D in neu-
tral 1:1 D2O–CH3CN solution should be approximately
equal to 0.11 as previously measured. The quantum yield for
the photosubstitution reaction of 4 was measured relative to
the photomethanolysis reaction of o-hydroxybenzhydrol in
H2O–CH3OH (1:1) (Φ = 0.46) (5). Preparative scale
photoreactions (10–3 mol L–1, 50 mL H2O–CH3OH (1:1), Ar
purged, 2 × 254 nm lamps, 1–3 min) were performed for 4
and the reference compound, and the extent of reaction was
Photolysis of 1
Photolysis of 1 (20 mg) in 1:1 D2O–CH3CN (v/v)
(40 mL) with eight lamps (300 nm) for 20 min followed by
standard workup gave starting material with deuterium in-
corporation at the 5-position (55%) and 8-position (19%)
1
with no side products, as determined by 500 MHz H NMR.
MS analysis of the same sample indicated that it consisted
of 37% non-deuterated, 52% monodeuterated, and 11%
dideuterated substrate, giving the equivalent overall deute-
rium incorporation as measured by NMR.
1
measured by 300 MHz H NMR. The reported value of Φ =
0.84 0.04 is the average of three such runs.
Fluorescence measurements
Steady-state fluorescence measurements were conducted
on a Photon Technology International A-1010 (PTI) Quanta-
Photolysis of 2
Photolysis of 2 (20 mg) in 1:1 D2O–CH3CN (v/v)
(40 mL) with 16 lamps (300 nm) for 1 h followed by stan-
dard workup gave only starting material with no deuterium
Master
luminescence
spectrometer.
All
solutions
(absorbance ≈ 0.1 at λex 300 nm) were purged with argon
prior to excitation. Fluorescence lifetimes were measured on
a PTI LS-1 instrument using time-correlated single-photon-
counting techniques (10 000 counts, λex 300 nm, λem
350 nm).
1
incorporation at any position, as determined by 500 MHz H
NMR. MS analysis of the same sample also indicated that
no additional deuterium was incorporated in the sample fol-
lowing photolysis.
Photolysis of 4
Laser flash photolysis (LFP)
Photolysis of 4 (50 mg) in 1:1 H2O–CH3OH (v/v)
(100 mL) with four lamps (300 nm) for 1 min followed by
standard workup gave methyl ether 29 in 48% conversion.
Photolysis product 29 was chromatographically isolated on
A Spectra-Physics excimer laser (308 nm, ~10 ns, <30
mJ/pulse) was used for excitation and signals were digitized
with a Tektronix TDS 520 recorder. Samples of OD ~ 0.3 at
308 nm were prepared and irradiated in quartz cells. Static
cells were used for the kinetic studies of samples, otherwise
flow cells were used for all other studies. Samples in flow
cells were continuously purged with a stream of N2 or O2.
Samples in static cells were purged with N2 or O2 then
sealed prior to photolysis.
1
silica gel (10% EtOAc in CH2Cl2). H NMR (300 MHz,
CDCl3) δ: 1.58 (d, 3H, CH3), 3.30 (br s, 3H, OCH3), 5.04 (q,
1H, ArC-H), 5.80 (br s, 1H, OH), 6.82 (d, 1H, Ar-H), 7.33
(t, 1H, Ar-H), 7.48 (t, 1H, Ar-H), 7.59 (d, 1H, Ar-H), 7.70
(d, 1H, Ar-H) 8.16 (d, 1H, Ar-H). HR-MS calcd. for
C13H14O2: 202.0994; observed: 202.0993.
Photolysis of 5
Acknowledgments
Photolysis of 5 (20 mg) in 1:1 H2O–CH3OH (v/v)
(100 mL) containing NaHCO3 (25 mg) with 16 lamps
(300 nm) for 3 min followed by standard workup cleanly re-
generated starting material with no reaction observable by
We thank the Natural Sciences and Engineering Research
Council of Canada (NSERC) for continued support of this
research, and for a post-graduate scholarship to one us
© 2004 NRC Canada

Lukeman et al.
253
15. J.B. Birks. Photophysics of aromatic molecules. Wiley-
Interscience, New York. 1983. p. 123.
(ML). We thank Ms. Kai Zhang and Dr. Darryl W. Brous-
miche for some initial experiments.
16. (a) H. Shizuka and S. Tobita. J. Am. Chem. Soc. 104, 6919
(1982); (b) H. Shizuka. Acc. Chem. Res. 17, 141 (1985).
17. R. Pollard, S. Wu, G. Zhang, and P. Wan. J. Org. Chem. 58,
2605 (1993).
18. P. Wan and G. Zhang. Res. Chem. Intermed. 19, 119 (1993).
19. (a) R.A. McClelland. Tetrahedron, 52, 6823 (1996); (b) N.
Mathivanan, F. Cozens, R.A. McClelland, and S. Steenken. J.
Am. Chem. Soc. 114, 2198 (1992).
References
1. P. Wan, B. Barker, L. Diao, M. Fischer, Y. Shi, and C. Yang.
Can. J. Chem. 71, 465 (1996).
2. P. Wan, D.W. Brousmiche, C.Z. Chen, J. Cole, M. Lukeman,
and M. Xu. Pure Appl. Chem. 73, 529 (2001).
3. R.W. Van de Water and T.R.R. Pettus. Tetrahedron, 58, 5367
(2002).
4. M. Fischer and P. Wan. J. Am. Chem. Soc. 121, 4555 (1999).
5. L. Diao, C. Yang, and P. Wan. J. Am. Chem. Soc. 117, 5369
(1995).
20. D.N. Kevill and S.W. Anderson. J. Org. Chem. 56, 1845
(1991).
21. Y. Chiang, A.J. Kresge, B. Hellrung, P. Schünemann, and J.
Wirz. Helv. Chim. Acta, 80, 1106 (1997).
22. H. Shizuka, H. Hagiwara, and M. Fukushima. J. Am. Chem.
Soc. 107, 7816 (1985).
23. D.W. Brousmiche and P. Wan. J. Photochem. Photobiol. A,
6. Y. Shi and P. Wan. J. Chem. Soc. Chem. Commun. 273 (1997).
7. P. Seiler and J. Wirz. Helv. Chim. Acta, 55, 2693 (1972).
8. S.P. Webb, L.A. Philips, S.W. Yeh, L.M. Tolbert, and J.H.
Clark. J. Phys. Chem. 90, 5154 (1986).
149, 71 (2002).
24. M. Lukeman and P. Wan. J. Am. Chem. Soc. 125, 1164
(2003).
25. H.J.E. Lowenthal and L. Gottlieb. J. Org. Chem. 57, 2631
(1992).
26. A.I. Myers and K. Higashiyama. J. Org. Chem. 52, 4592
(1987).
27. M. Sawada, T. Fujii, Y. Tairaka, and Y. Yukawa. Bull. Chem.
9. (a) M. Lukeman and P. Wan. J. Chem. Soc. Chem. Commun.
1004 (2001); (b) M. Lukeman and P. Wan. J. Am. Chem. Soc.
124, 9458 (2002).
10. (a) R.S. Givens, A. Jung, C.-H. Park, J. Weber, and W. Bart-
lett. J. Am. Chem. Soc. 119, 8369 (1997); (b) R.S. Givens, P.S.
Athey, B. Matuszewski, L.W. Kueper III, J.-y. Xue, and T.
Fister. J. Am. Chem. Soc. 115, 6001 (1993); (c) P.G. Conrad
II, R.S. Givens, B. Hellrung, C.S. Rajesh, M. Ramseier, and J.
Wirz. J. Am. Chem. Soc. 122, 9346 (2000).
Soc. Jpn. 48, 3356 (1975).
28. T. Bisanz and M. Bukowska. Rocz. Chem. 48, 777 (1974).
29. N.P. Buu-Hoi and P. Cagniant. Rec. Trav. Chim. Pays–Bas, 64,
214 (1945).
11. K. Zhang, J.E.T. Corrie, V.R.N. Munasinghe, and P. Wan. J.
Am. Chem. Soc. 121, 5625 (1999).
30. W.H. Hunter, R.M. Quinton, P.H. Sherman, C.R. Worthing,
and R.J. Boscott. J. Med. Chem. 7, 167 (1964).
31. J. Cason. J. Am. Chem. Soc. 63, 828 (1941).
32. S.J. Pasaribu and L.R. Williams. Aust. J. Chem. 26, 1327
(1973).
12. W.F. Short and H. Wang. J. Chem. Soc. 991 (1950).
13. S. Lee, S.P. Frescas, and D.E. Nichols. Synth. Commun. 25,
2275 (1995).
14. M.K. Nayak and A.K. Chakraborti. Tetrahedron Lett. 38, 8749
(1997).
© 2004 NRC Canada