Excited state intramolecular proton transfer in 1-hydroxypyrene

Article abstract of DOI:10.1139/V11-010

1-Hydroxypyrene (1) shows unusual acid'base chemistry in its singlet excited state. Whereas most hydroxyarenes experience a marked enhancement in their acidity when excited, and rapidly deprotonate to give the corresponding phenolate anion, this is not an important pathway for 1, despite theoretical predictions that 1 should experience enhanced acidity as well. In this work, we demonstrate that 1 undergoes a competing excited state intramolecular proton transfer from the OH to carbon atoms at the 3, 6, and 8 positions of the pyrene ring to give quinone methide intermediates. When the reaction is carried out in D2O, reversion of these quinone methides to starting material results in replacement of the ring hydrogens with deuterium, providing a convenient handle to follow the reaction with NMR spectroscopy and mass spectrometry. The quantum yield for the reaction is 0.025 and appears to not be strongly dependent on the water content when aqueous acetonitrile solutions are used. 1-(2-Hydroxyphenyl) pyrene (19) was prepared and studied and shows similar reactivity to 1.

Full text of DOI:10.1139/V11-010

4
33
Excited state intramolecular proton transfer in 1-
hydroxypyrene
Matthew Lukeman, Misty-Dawn Burns, and Peter Wan
Abstract: 1-Hydroxypyrene (1) shows unusual acid–base chemistry in its singlet excited state. Whereas most hydroxy-
arenes experience a marked enhancement in their acidity when excited, and rapidly deprotonate to give the corresponding
phenolate anion, this is not an important pathway for 1, despite theoretical predictions that 1 should experience enhanced
acidity as well. In this work, we demonstrate that 1 undergoes a competing excited state intramolecular proton transfer
from the OH to carbon atoms at the 3, 6, and 8 positions of the pyrene ring to give quinone methide intermediates. When
the reaction is carried out in D2O, reversion of these quinone methides to starting material results in replacement of the
ring hydrogens with deuterium, providing a convenient handle to follow the reaction with NMR spectroscopy and mass
spectrometry. The quantum yield for the reaction is 0.025 and appears to not be strongly dependent on the water content
when aqueous acetonitrile solutions are used. 1-(2-Hydroxyphenyl)pyrene (19) was prepared and studied and shows similar
reactivity to 1.
Key words: 1-pyrenol, 1-hydroxypyrene, excited state intramolecular proton transfer, quinone methide, photoprotonation,
photochemistry.
R e´ sum e´ : Dans son e´ tat excit e´ singulet, le 1-hydroxypyr e` ne (1) pr e´ sente une chimie acide–base inhabituelle. La plupart
des hydroxyar e` nes pr e´ sentent une augmentation marqu e´ e de leur acidit e´ dans leur e´ tat excit e´ et ils se d e´ protonent rapide-
ment pour conduire a` l’anion ph e´ nolate correspondant; ce n’est pas une voie importante pour 1 m eˆ me si les pr e´ dictions
th e´ oriques indiquent que 1 devrait aussi voir son acidit e´ augment e´ e. Dans ce travail, on d e´ montre que dans son e´ tat excit e´
1
subit aussi une r e´ action comp e´ titive de transfert intramol e´ culaire de proton a` partir du groupe OH vers les atomes de
carbone dans les positions 3, 6 et 8 du noyau pyr e` ne qui conduisent a` des interm e´ diaires quinone-m e´ thides. Quand on ef-
fectue la r e´ action dans du D2O, la r e´ version de ces quinone-m e´ thides vers le produit de d e´ part conduit au remplacement
des atomes d’hydrog e` ne du cycle par du deut e´ rium qui fournit un moyen utile de suivre la r e´ action a` l’aide de la spectros-
copie RMN et de la spectrom e´ trie de masse. Le rendement quantique pour la r e´ action est de 0.025 et il ne semble pas d e´ -
pendre beaucoup du contenu d’eau lorsqu’on utilise des solutions aqueuses d’ac e´ tonitrile. On a aussi pr e´ par e´ et e´ tudi e´ la
r e´ action du 1-(2-hydroxyph e´ nyl)pyr e` ne (19) et il pr e´ sente la m eˆ me r e´ activit e´ que le 1.
Mots-cl e´ s : 1-pyr e´ nol, 1-hydroxypyr e` ne, transfert intramol e´ culaire de proton dans l’ e´ tat excit e´ , quinone-m e´ thides, photo-
protonation, photochimie.
[
Traduit par la R e´ daction]
Introduction
It is well-known that a large variety of aromatic alcohols
experience enhanced acidity when excited to S . For exam-
a source of interest because unlike most other phenols, it is
difficult to discern the existence of ESPT from fluorescence
spectroscopy. Milosavljevic and Thomas used F o¨ rster cycle
3
1
1
calculations to predict that 1 should possess a pK (S ) value
a
1
ple, the hydroxyl group of 2-naphthol increases in acidity by
nearly 7 orders of magnitude on excitation from its ground
state (pK (S ) & 9.5) to its first excited state (pK (S ) & 2.8).
of 4.5, indicating that ESPT in neutral water should be ther-
modynamically favourable. They then explored the photo-
physics of 1 in water and concluded that ‘‘no discernable
equilibrium exists between the excited state [of 1] and its
excited ionic forms’’, which is at variance with the F o¨ rster
cycle predictions. This prompted the authors to question the
validity of F o¨ rster cycle calculations, and indeed the ‘‘con-
cepts of acidity of excited states’’. By comparing steady-
state emission spectraof 1 in pH 4–7 unbuffered water with
those in low pH (<2) solution, Kombarova and Il’ichev4
a
0
a
1
The enhancement of acidity on excitation is even more pro-
nounced for 1-naphthol, whose pK changes from 9.2 in the
ground state to 0.4 in S . Excited phenols will usually deprot-
onate adiabatically within picoseconds in water via a process
known as excited state proton transfer (ESPT), and observa-
tion of fluorescence emission from the resultant excited-state
phenolate is generally taken as a diagnostic sign of ESPT.
The excited-state acidity of 1-hydroxypyrene (1) has been
a
2
1
Received 31 May 2010. Accepted 28 November 2010. Published on the NRC Research Press Web site at canjchem.nrc.ca on 5 March
011.
2
This article is part of a Special Issue dedicated to Professor J. C. Scaiano.
1
M. Lukeman and M.-D. Burns. Department of Chemistry, 6 University Avenue, Acadia University, NS B4P 2R6, Canada.
P. Wan. Department of Chemistry, Box 3065, University of Victoria, BC V8W 3V6, Canada.
1Corresponding author (e-mail: matthew.lukeman@acadiau.ca).
Can. J. Chem. 89: 433–440 (2011)
doi:10.1139/V11-010
Published by NRC Research Press

4
34
Can. J. Chem. Vol. 89, 2011
were able to detect emission from the pyrenolate arising
from ESPT, although only a small fraction of the excited
phenol appears to dissociate. Using time-resolved fluores-
cence data, the authors are able to confirm that the pyreno-
late emission indeed arises from excited-state dissociation of
the pyrenol hydroxyl group, and they estimate a pK (S ) of
½2ꢁ
a
1
3
.8 for 1, which is in good agreement with the value ob-
tained by the F o¨ rster cycle calculations. However, the calcu-
6
–1
lated rate constant for ESPT was only 4 ꢀ 10 s , which is
exceptionally slow for a proton transfer reaction. For com-
parison, pyranine, the 3,6,8-trisulfonated derivative of 1,
undergoes ESPT to water with a rate constant >1010 s .
½
3ꢁ
4ꢁ
–1 5
½
Over the past decade, we have shown that several simple ar-
omatic alcohols can undergo efficient excited state intramo-
lecular proton transfer (ESIPT) from the OH group to
aromatic carbon atoms to yield quinone methide (QM) inter-
mediates. For example, ESIPT in 1-naphthol (2) from the hy-
droxyl group to the 5 or 8 position leads to formation of QMs 3
or 4, which then undergo reverse proton transfer to refurnish
Experimental
General
All fine chemicals were purchased from Sigma-Aldrich
Canada and used as received. Solvents were purchased as
spec. grade or better, acetonitrile was dried over CaH and
2
6
starting material (eq. [1]). Evidence for the ESIPT pathway
was provided by monitoring deuterium incorporation at the 5
distilled prior to use, and water was distilled and deionized.
1
H NMR spectra were recorded on a Bruker AVANCE
and 8 positions (when the solvent contained D O), and by
500 MHz instrument. MS were recorded on a Kratos Con-
cept H spectrometer (EI). UV–vis were recorded on a Var-
ian CARY 50 or CARY 100 spectrometer.
2
nanosecond laser flash photolysis (LFP) studies that allowed
direct observation of QM 3. Whereas ESIPT in 2 requires
water to mediate the proton transfer, an intrinsic or ‘‘direct’’
ESIPT that does not require water was discovered to be avail-
Materials
7
able for 2-phenylphenol (5) to give QM 6 (eq. [2]). Since
1
-(2’-Hydroxyphenyl)pyrene (19)
these discoveries, we have also found that similar ESIPT proc-
8
1-Bromopyrene (1.00 g, 3.56 mmol), o-methoxyphenyl-
boronic acid (0.684 g, 4.5 mmol), K CO (4 g), Pd(PPh )
esses can occur in naphthylphenols and hydroxyphenylan-
thracenes,9 and in these cases, products in addition to
deuterium-labeled starting material were obtained. Irradiation
of 7–9 resulted in ESIPT from the OH to the 7’ carbon atom on
the attached naphthalene ring to give quinone methides 10–12,
which underwent electrocyclic ring closure to give benzo-
2 3 3 4
(
0.20 g), and DME (50 mL, N purged prior to addition)
2
were combined and refluxed under N for 20 h and allowed
to cool. Sixty millilitres of 1 mol/L NaOH were added to the
reaction mixture, which was then extracted with two por-
2
8
tions of CH Cl totaling 70 mL. The combined organics
pyran products 13–15 (eq. [3]). ESIPT from the OH to the 10
2 2
were dried with MgSO , filtered, and the solvent was
position of 16 gave QM 17, which reacted with water to give
4
9
removed by rotary evaporation. The crude 1-(2’-methoxy-
phenyl)pyrene was redissolved in 40 mL of diethyl ether,
which was passed through a 1 cm plug of silica gel and a
the isolable hydration product 18 (eq. [4]). Discovery of these
elementary photochemical pathways in these relatively simple
aromatic alcohols piqued our curiosity as to whether a similar
process might be available for 1, which would shed some light
on its apparently anomalous ESPT behavior and may help ex-
2
cm plug of celite to remove catalyst. The silica and celite
were flushed with an additional 100 mL of diethyl ether, and
solvent was removed from the combined ethers by rotary
evaporation. The resulting material was redissolved in
–
plain why ESPT to water (to yield 1 ) is inefficient. In addition,
biaryl derivative 19 was prepared and examined for ESIPT re-
activity via possible deuterium exchange, photocyclization, or
photohydration. We present results that indicate that ESIPT can
occur in 1 and 19 from the OH to the 3, 6, and 8 positions of the
pyrene ring in both compounds to give rise to QM intermedi-
ates, which undergo exclusive reverse proton transfer to give
5
0 mL of CH Cl and placed in an ice bath. To this cooled
2 2
solution was added a solution of BBr (10 mmol) in CH Cl
3
2
2
(
60 mL) via a dropping funnel over a period of 30 min, after
which the mixture was allowed to stir for 1 h. The reaction
was quenched by the addition of 100 mL of H O, 10 mL of
2
1
0% HCl, and 10 mL of brine. The organic layer was re-
moved and the aqueous layer was washed with an additional
0 mL of CH Cl . The organics were combined, dried with
starting material (labeled with deuterium when D O is used).
2
3
2
2
MgSO , filtered, and the solvent was removed by rotary
4
½1ꢁ
evaporation. The resulting solid was recrystallized from
1
methanol to yield 19 as white crystals. H NMR (500 MHz,
Published by NRC Research Press

Lukeman et al.
435
CDCl , ppm) d: 4.842 (1H, s, OH), 7.118 (1H, ddd, J = 7.4,
where F and F are the quantum yields of the deuterium
3
1
5
7
.4, 1.2 Hz), 7.128 (1H, dd, J = 8.4, 1.2 Hz), 7.387 (1H, dd,
J = 7.4, 1.6 Hz), 7.423 (1H, ddd, J = 8.4, 7.4, 1.6 Hz), 7.894
1H, d, J = 9.2 Hz), 7.984 (1H, d, J = 7.8 Hz), 8.032 (1H, t,
J = 7.6 Hz), 8.053 (1H, d, J = 9.2 Hz), 8.108 (1H, d, J =
incorporation of 1 and 5, respectively, and %conv(1) and
%conv(5) are the percents of conversion as measured by H
NMR. The reported quantum yields are the average of the
four runs.
1
(
8
0
7
1
1
1
.9 Hz), 8.135 (1H, d, J = 8.9 Hz), 8.190 (1H, dd, J = 7.6,
.9 Hz), 8.224 (1H, d, J = 7.6, 0.9 Hz), 8.267 (1H, d, J =
Fluorescence and laser flash photolysis measurements
Solutions for fluorescence spectroscopy were purged with
argon for 5 min before analysis. Spectra were obtained using
a Photon Technology International (PTI) A-1010 Quanta-
Master luminescence spectrometer. All transient absorption
spectra and lifetimes were obtained using a Nd:YAG laser
.8 Hz). 13C NMR (125 MHz, CDCl ) d: 115.68, 120.72,
3
24.67, 124.69, 125.05, 125.21, 125.38, 125.52, 126.29,
26.81, 127.27, 128.02, 128.26, 128.37, 129.37, 129.66,
30.96, 131.03, 131.33, 131.42, 131.60, 153.25.
1
-Methoxypyrene
To a 10 mL flask was added 1 (35 mg, 0.16 mmol),
(
Spectra Physics Quanta-Ray, GCR, <30 mJ/pulse) with a
pulse duration of ~10 ns and an excitation wavelength of
55 nm. Flow cells were used (7 mm path length) and solu-
DMSO (10 mL), crushed KOH (1 g), and methyl iodide
2.85 g, 20 mmol). The mixture was stirred at room temper-
3
(
tions were purged with nitrogen or oxygen for 10 min prior to
excitation. Solution concentrations were adjusted to achieve
an optical density of 0.3 at the excitation wavelength.
ature for 16 h and then diluted by the addition of 50 mL of
water to dissolve remaining KOH. The resulting mixture
was extracted with 50 mL of CH Cl , and the organic layer
2
2
was washed with 1 mol/L NaOH (50 mL) and water (2 ꢀ
Results and discussion
1
50 mL), and then dried over MgSO , and the solvent was
4
removed by rotary evaporation. The crude reaction mixture
was purified by preparatory TLC (1000 mm silica gel) using
Our primary evidence for the ESIPT pathway in 2 came
from the observation of H–D exchange at the 5 and 8 posi-
tions when the substrate was irradiated in D O–CH CN sol-
vent mixtures. The same strategy was employed for 1 in an
attempt to observe the deuterated products that would arise
if a similar ESIPT process as that of 2 was available for 1.
1
5:85 v/v ether–hexanes as the elutant, and yielded 20 as a
2
3
1
6
white solid (90% yield). H NMR (500 MHz, acetone-d ,
6
ppm) d: 4.189 (3H, s, OCH ), 7.706 (1H, d, J = 8.4 Hz),
3
7
.935 (1H, d, J = 9.0 Hz), 7.990 (1H, t, J = 7.6 Hz), 8.025
1H, d, J = 9.0 Hz), 8.097 (1H, d, J = 9.2 Hz), 8.152 (1H,
dd, J = 7.6, 0.9 Hz), 8.173 (1H, dd, J = 7.6, 0.9 Hz), 8.208
–
4
(
Irradiation of 1 in D O–CH CN ( 1:1, v/v) (~5 ꢀ 10 mol/L,
2
3
16 ꢀ 350 nm lamps, argon saturated, 2 h) gave starting ma-
terial with deuterium incorporated at the 3 position (7%),
and to the 6 and 8 positions (34% for both positions com-
bined) to give 1-3D, 1-6D, and 1-8D, respectively (eq. [6]).
1
(
1H, d, J = 8.4 Hz), 8.425 (1H, d, J = 9.2 Hz). the H NMR
1
0 13
datamatches a previous report.
C NMR (125 MHz,
acetone-d ) d: 56.67, 109.43, 120.90, 121.97, 125.12,
6
1
1
1
25.27, 125.79, 126.16, 126.57, 126.84, 127.21, 127.24,
28.30, 132.69, 132.82, 154.70.
The deuterium exchange is evident in the H NMR spectrum
(Fig. 1) where irradiation led to a decrease in the areas of
the peaks at 8.10, 8.12, and 8.14 ppm assigned to the
protons at the 3, 6, and 8 positions, respectively.11 A broad
doublet is observed to grow in under the doublet of doublets
at 7.96 ppm assigned to the proton at the 7 position. This
reduced coupling is consistent with deuterium incorporation
at one of the adjacent sites. Similarly, a small broad singlet
has grown in under the doublet at 7.62 ppm assigned to the
Product studies
Solutions were prepared in a quartz reaction vessel and
purged with argon or oxygen for 10 min prior to and contin-
uously during irradiation in a Rayonet RPR 100 or Luz-
Chem LZC-Org photochemical reactor using 350 nm lamps.
The photolysate was extracted with CH Cl , dried with an-
2
2
2
position, consistent with a lesser amount of deuterium in-
hyd MgSO , and solvent was removed under reduced pres-
4
1
corporation at the 3 position. The overall extent of exchange
was confirmed by MS, which showed a distribution of 65.8%
undeuterated, 29.8% monodeuterated, and 4.2% dideuterated
starting material, giving an overall deuterium exchange of
sure, and the resulting material analyzed by H NMR
spectroscopy and MS. For quantum yields, four such runs
were carried out for each of 1 and 5 (both in D O–CH CN
2
3
(
1:9) solution), and the total amount of deuterium incorpora-
3
9% (NMR measurements indicate a total of 41%). No other
tion was determined based on peak integrations in the re-
spective H NMR spectra. Quantum yields were calculated
1
photoproducts were observed in the photolysate. Irradiation
of 1-methoxypyrene, the methyl ether of 1, under the same
conditions led to no observable deuterium incorporation by
NMR, confirming that the reaction requires the OH func-
tional group, consistent with an ESIPT mechanism.
according to eq. [5]:
%
convð1Þ
convð5Þ
½
5ꢁ
6ꢁ
F ¼ F5
1
%
½
Published by NRC Research Press

4
36
Can. J. Chem. Vol. 89, 2011
1
Fig. 1. H NMR spectrum (500 MHz) of 1 before (bottom) and after (top) irradiation in D2O–CH3CN (1:1; 2 h), showing decreases in the
areas of the peaks at 8.14, 8.12, and 7.99 ppm assigned to ring protons at the 8, 6, and 3 positions, respectively.
ture.6 Only residual reaction was observed at D O
Fig. 2. Plot of the sum of the % deuterium exchange at the 3, 6,
and 8 positions of 1 as a function of [D2O] (in CH3CN) as deter-
mined by MS.
2
concentrations less than ~2 mol/L. Also, less water was ap-
parently required for ESIPT to the 8 position than for ESIPT
to the 5 position, which we rationalized to be a consequence
of the different distances between the acidic OH and aro-
matic ring sites; the closer the phenol OH is to the site that
receives the proton, the less water is required to mediate the
proton transfer. To probe the dependence on water of the
deuterium exchange reaction of 1, irradiations were carried
out in which the concentration of D O (in CH CN) was var-
2
3
ied. The photolysate (after evaporation in vacuo) was exam-
1
ined by H NMR and MS for deuterium incorporation at
aromatic ring positions, and the total deuterium exchange at
the three positions (as determined by MS) is presented in
Fig. 2. Very different behaviour is observed for this deriva-
tive than was observed for 2 when the D O content is low-
2
ered: the total deuterium exchange did not decrease
significantly until very small D O concentrations were
2
reached, suggesting that very little water (or possibly none
at all) is required to mediate the process. An intrinsic ESIPT
7
(
such as that observed for 5) seems unlikely, since the hy-
By irradiating solutions of 2 in CH CN containing vary-
droxyl functionality is located a significant distance from the
sites of deuterium exchange, particularly the 6 and 8 posi-
tions, preventing the ground-state hydrogen bonding interac-
tion that is thought to be a prerequisite.
3
1
ing amounts of D O and analyzing the photolysate by H
2
NMR, it was found that the ESIPT efficiency for this com-
pound relied heavily on the composition of the solvent mix-
Published by NRC Research Press

Lukeman et al.
437
1
Fig. 3. H NMR spectra (500 MHz), 7.8–8.3 ppm region, of 19 before (bottom) and after (top) irradiation in D2O–CH3CN (1:1; 30 min),
highlighting the changes in peaks at 7.98 and 8.03 ppm that are consistent with loss of vicinal coupling with adjacent hydrogens because of
deuterium exchange. Resonances owing to the phenolic ring appear between 7.0 and 7.5 ppm and are not included in this figure.
The relative extents of deuterium exchange at the 3, 6,
and 8 positions were monitored by H NMR at the different
be more than twice as efficient in CH CN–H O (9:1, v/v),
3 2
assuming a primary isotope effect of 2 or greater.
1
D O concentrations. If the deuterium exchange reaction is
the result of a water-mediated ESIPT, deuterium incorpora-
Irradiation of 19 was carried out in a similar manner as
2
–4
for 1 in D O–CH CN (1:1, v/v) (~5 ꢀ 10 mol/L, 16 ꢀ
2
3
tion at the 3 position should require far less water (or D O)
350 nm lamps, argon saturated, 30 min), and the photolysate
2
than at the 6 and 8 positions, because of the smaller distance
between the 3 position and the OH moiety that the water
would have to span. The efficiency of exchange at the 6
and 8 positions was expected to show a much stronger de-
1
was examined by H NMR spectroscopy (peak assignments
were made with the aid of COSY and NOESY NMR spec-
tra, included in the Supplementary data). Deuterium ex-
change was observed at the 3, 6, and 8 positions of the
pendence on D O concentration with significantly larger
1
2
pyrene ring, the same positions as observed for 1. The H
D O concentrations needed. The ratio of deuterium ex-
2
NMR spectra of 19 before and after irradiation are shown
in Fig. 3. The areas of the peaks at 8.27, 8.22, and
.19 ppm are reduced following irradiation, consistent with
change at the 3 position to the 6 and 8 positions was com-
pletely unchanged as the D O content of the solution was
2
8
reduced. This result argues against a water-mediated ESIPT
replacement of the hydrogens at these positions with deute-
rium. A broad singlet is observed to have grown beneath the
doublet at 7.98, which is assigned to the proton on carbon 2
of the pyrene ring, which is consistent with loss of coupling
to the adjacent proton (Fig. 3). Similarly, a doublet has
grown beneath the triplet at 8.03 ppm, consistent with loss
of coupling to an adjacent proton at position 6 or 8. The
overall quantum yield for exchange of these three sites is
similar to that of 1. No other products, such as photocycliza-
tion or photohydration products, were observed, indicating
that this derivative does not show chemistry similar to 7, 8,
9, or 16.
6
such as that observed for 2, indicating water is likely not
acting as a ‘‘bridge’’ to mediate the proton (deuteron) trans-
fer.
The quantum yield of deuterium exchange for 1 was
1
measured by H NMR in CH CN–D O (9:1, v:v) solutions
3
2
using the deuterium exchange reaction of 5 as a secondary
7
b
reference standard. The quantum yield for exchange was
measured to be 0.021 at the 6 or 8 positions and 0.004 for
the 3 position, for a total quantum yield of deuterium incor-
poration at any one of these sites of 0.025. This value is
somewhat lower than the other examples of ESIPT to aro-
6
–9
matic carbon reported by us.
The ESIPT is expected to
Published by NRC Research Press

4
38
Can. J. Chem. Vol. 89, 2011
Scheme 1.
Fluorescence measurements
Fluorescence spectroscopy was used to study 1 to confirm
the lack of significant observable emission from the pyreno-
–
late (1 ) that would result from an adiabatic ESPT to water.
Excitation of 1 (lex = 345 nm) in neat CH CN or in neat
3
water gave strong and structured emission with maxima at
3
1
l
83, 405, and 427 nm. Excitation of an authentic sample of
(formed by dissolving 1 in 0.01 mol/L NaOH solution,
–
3
= 400 nm) gave strong, broad, and unstructured fluores-
previously reported in methanol. This transient completely
decayed within ~3 ms to leave a long-lived transient with
lmax = 440 nm, whose absorption profile and lifetime
ex
cence emission at 450 nm. The fluorescence yield of 1 was
not noticeably diminished on moving from CH CN to water,
3
ꢂ
and no phenolate emission was observed under neutral con-
ditions, confirming that ESPT to water is at best a very mi-
strongly resemble that of PyO reported in methanol solu-
3
ꢂ
tion. PyO can presumably be readily generated by depro-
ꢂ+
nor pathway for
1
under these conditions. These
tonation of 1 , which in turn can be generated by
observations mirror previous fluorescence studies carried
photoejection of an electron. In our experiments, the sol-
vated electron was not visible, so laser power dependence
studies were not carried out. No additional transients were
observed, so direct evidence for the formation of QMs 20–
out for 1 by others.3
,4
Solutions of 1 in water with the same concentration as
4
those employed in the product studies (~5 ꢀ 10 mol/L)
2
2 in this solvent could not be obtained by this method.
were also examined by fluorescence spectroscopy to monitor
for excimer formation. No long wavelength band assignable
to excimer emission was observed at these concentrations,
indicating that excimer formation does not contribute to the
observed photochemistry.
Our inability to detect these species likely stems from their
low quantum yield of formation, although it is also possible
that they are too short-lived for the time resolution of the in-
strumentation (i.e., t < 20 ns).
Nanosecond laser Fflash photolysis (LFP)
4
Milosavljevic and Thomas studied the photophysics of 1
½
7ꢁ
using LFP in cyclohexane, methanol, and methanol–NaOH.
In cyclohexane and methanol solutions, three transients
were observed and assigned to the triplet excited state of 1
3
*
ꢂ+
(
1 , l
= 430 nm), the radical cation of 1 (1 , l
=
=
max
max
ꢂ
15 nm), and the pyrenyloxyl radical (PyO , l
max
5
4
30 nm). In methanol, the solvated electron that accompa-
Mechanism
ꢂ+
nies formation of 1 is also observed, and its yield was ob-
served to vary with the square of the laser power, indicating
that the solvated electron (and the accompanying radicals
The relevant photophysical and photochemical processes
for 1 in aqueous solution are summarized in Scheme 1. Ex-
citation of 1 produces the singlet excited state ( 1 ), which
has a lifetime of approximately 16 ns. The dominant relax-
ation pathway for this excited state is fluorescence emission
1
*
ꢂ+
ꢂ
1
and PyO ) are formed via a two-photon process and are
4
therefore not likely to be important species at low light in-
tensities such as those used in the product studies in this
work. The pyrenolate anion (generated by deprotonating 1
12
3
(
F * 0.66) at 380 nm. A small fraction of the singlet
F
excited state undergoes ESPT of the phenolic OH to the sol-
vent (water) adiabatically to give the pyrenolate anion in its
singlet excited state (1 ), which has a lifetime of 3 ns, and
returns to the ground-state pyrenolate anion (1 ) via fluores-
cence emission with lmax 450 nm. The triplet state of 1 ( 1 )
3
*
with NaOH in methanol) was also studied and gave 1 ,
ꢂ
PyO , and solvated electrons (via a one-photon process),
–*
4
ꢂ+
but no 1 . No data was reported in aqueous solution, which
–
is the relevant solvent for the deuterium exchange reaction
that we report in the present work. An ESIPT reaction to
the 3, 6, and 8 positions of 1 would give QM intermediates
3
*
3
is observed in LFP experiments and absorbs at 430 nm. The
ꢂ+
ꢂ
radical cation (1 ) and pyrenoxyl radicals (PyO ) are also
observed in LFP experiments and are proposed to arise via
two-photon processes that are only relevant under conditions
of high photon flux (i.e., under laser irradiation), as was the
case for these transients previously reported in cyclohexane
and methanol.4
2
0–22, respectively (eq. [7]). QMs 20 and 22 are non-
Kekule structures, and thus were expected to have short life-
1
2
times that would be inaccessible by LFP. QM 21 was an-
ticipated to have a much longer lifetime, and so we made an
attempt to detect it with LFP. LFP of a solution of 1 in
CH CN–H O (1:1), nitrogen purged, gave rise to a transient
3
2
absorption with maxima at 440 and 510 nm. The signal at
5
In addition to these photophysical properties that have
been previously reported for 1 in water and in other solvents,
ꢂ+
10 nm was assigned to the radical cation (1 ), which was
Published by NRC Research Press

Lukeman et al.
439
we propose an additional deactivation pathway that is avail-
the electronics of the 1-substituted pyrene ring in S and that
the 3, 6, and 8 positions are the sites with the highest elec-
1
1
*
able to 1 in water that is an ESIPT from the OH to carbon
atoms on the pyrene ring to give quinone methides 20–22,
which revert to 1 via reverse proton transfer (resulting in
tron density in S . This is an interesting result, given that
1
when the 3, 6, and 8 positions of 1 are substituted with the
deuterium labeling when the solvent is D O). Classification
electron-withdrawing sulfonate groups (to make pyranine),
2
5
of the observed deuterium exchange reaction as an ESIPT
process is consistent with recent literature on related com-
pounds6 and is supported by the absence of deuterium in-
corporation observed for the methyl ether analogue under
the same conditions. An unexpected dependence on water
content was observed with very little or no water required
to mediate the proton transfer between the OH and the 3, 6,
and 8 positions. Such behaviour was only previously ob-
served for 2-phenylphenol and 2,2’-bisphenol that have an
intrinsic ESIPT mechanism available to them because of the
close proximity of the phenolic OH to the site of exchange.7
An intrinsic ESIPT from the OH to the 6 or 8 positions
would be difficult to envisage for 1 because of their consid-
erable distance from the OH. Also of interest is that the ratio
of exchange efficiencies for the three sites remains constant
ESPT to water becomes rapid (t = 87 ps) and the acidity
14
in S becomes much greater (pK (S ) = 0.4–1.4). The sub-
1
a
1
–9
stitution pattern of the sulfonate groups in pyranine is ideal
for promoting photoacidity considering the localization elec-
2a
tron density on the 3, 6, and 8 carbons in S . Tolbert et al.
1
successfully employed the strategy of substituting aromatic
carbon–hydrogen bonds of phenols with electron-withdraw-
ing groups at positions that are electron-rich in S to create
1
phenols with very low values of pK (S ), a class of com-
a
1
pounds known as ‘‘super photoacids’’. The lack of photocyc-
lization or photohydration products for 19 (or 1 in fact)
similar to those observed on irradiation of 7–9 and 16 is
probably a consequence of the high resonance energy asso-
ciated with the cyclic array of 14 p electrons that make up
the periphery of the pyrene ring,15 since photocyclization or
photohydration would necessarily disrupt this aromaticity.
over wide variations of D O concentration, suggesting the
2
formation of 20–22 from the excited state of 1 is independ-
ent of the degree of aqueous solvation of the excited state. A
possible mechanism to account for these observations involves
the existence of an intermediate prior to QM formation that
is common to all three QMs that are formed (20–22). One
such possibility involves an initial proton transfer occurring
from the OH to the pyrene p system to form a ‘‘protonated
p complex’’ such as 23. This proton transfer might occur
without solvent assistance or with the help of a small
amount of a hydroxylic solvent. The acidic proton in 23
may then rearrange to form 20–22 via a solvent-independent
process. The relative efficiencies for protonation at these
three sites would simply be a function of the electron den-
sity at each position and the mobility of the proton. It would
be expected that the most favourable destination of the pro-
ton in an intermediate such as 23 would be the oxygen
atom, an overall process that would not lead to observable
deuterium incorporation. If this is the case, the quantum
yield of this initial ESIPT would be considerably higher
than the quantum yield of deuterium exchange (F = 0.025)
ESIPT from the OH to one of the central (ipso) carbons
might also be possible, and would give rise to QM inter-
mediates 24 or 25, which could undergo reverse proton
transfer to regenerate starting material (without deuterium
labeling) or rearrange to QMs 20–22. The exact nature, and
indeed the existence of an intermediate prior to QM forma-
tion, remains speculative at the moment, and the relatively
low efficiency of these processes complicates their study
via time-resolved absorption techniques, especially in light
of the strong signals produced by the other absorbing transi-
ent species.
The discovery of this ESIPT pathway provides additional
evidence that 1 becomes more strongly acidic in S , support-
1
4
ing earlier findings based on fluorescence spectroscopy and
3
Forster calculations. However, the ESIPT is not particularly
efficient (F = 0.025), especially when compared with fluo-
12
rescence (F = 0.66), the dominant photophysical process
F
from S . The ESIPT pathway has a similar efficiency to
1
ESPT to water, indicating that it is a major photoprototropic
reaction of 1 and provides a competitive pathway that would
–*
reduce the yield of 1 . Discovery of this ESIPT reaction
and the related process for 19 provides further evidence that
ESIPT is a very general reaction for hydroxyaromatic com-
pounds and that this process must be considered in any seri-
ous investigation of the photophysics of hydroxyarenes.
Supplementary data
Supplementary data are available with this article through
the journal Web site (www.nrcresearchpress.com/cjc).
References
(1) (a) Weller, A.; In Progress in Reaction Kinetics; Pergamon
Press: London, 1961; Vol. 1, p 187.; (b) Martynov, I. Yu.;
Demyashkevich, A. B.; Uzhinov, B. M.; Kuz’min, M. G.
Usp. Khim. 1977 46, 3.
(
2) (a) Tolbert, L. M.; Solntsev, K. M. Acc. Chem. Res. 2002 35
1), 19. doi:10.1021/ar990109f.; (b) Webb, S. P.; Philips, L.
A.; Yeh, S. W.; Tolbert, L. M.; Clark, J. H. J. Phys. Chem.
986 90 (21), 5154. doi:10.1021/j100412a053.
The deuterium exchange reaction of 19 is proposed to
proceed via a related pathway — ESIPT from the phenol to
the pyrene ring gives QMs that undergo reverse proton
transfer to regenerate starting material, labeled with deute-
(
1
(
3) Milosavljevic, B. H.; Thomas, J. K. Photochem. Photobiol.
rium when D O is used in place of H O as the solvent. The
2
2
Sci. 2002 1 (2), 100. doi:10.1039/b107925e.
(4) Kombarova, S. V.; Il’ichev, Y. V. Langmuir 2004 20 (15),
same positions in 1 and 19 undergo exchange, indicating
that the regiochemistry of the ESIPT is controlled solely by
6158. doi:10.1021/la049585r.
Published by NRC Research Press

4
40
Can. J. Chem. Vol. 89, 2011
(
(
(
5) Tran-Thi, T.-H.; Gustavsson, T.; Prayer, C.; Pommeret, S.;
Hynes, J. T. Chem. Phys. Lett. 2000 329 (5-6), 421. doi:10.
016/S0009-2614(00)01037-X.
(11) Peak assignments were made using COSY and NOESY
spectra of 1 (available in the Supplementary data) and agree
with an earlier report. Cerniglia, C. E.; Kelly, D. W.; Freeman,
J. P.; Miller, D. W. Chem. Biol. Interact. 1986 57 (2), 203.
doi:10.1016/0009-2797(86)90038-4.
(12) These expectations are based on our past experience with
Kekule and non-Kekule QM intermediates. For example,
QM 3 is readily detectable with ns LFP, however, non-
Kekule QM 4 was not (ref. 6).
1
6) Lukeman, M.; Veale, D.; Wan, P.; Munasinghe, V. R. N.;
Corrie, J. E. T. Can. J. Chem. 2004 82 (2), 240. doi:10.
1
139/v03-184.
7) (a) Lukeman, M.; Wan, P. Chem. Commun. (Camb.) 2001
11): 1004. doi:10.1039/b100817j.; (b) Lukeman, M.; Wan,
(
P. J. Am. Chem. Soc. 2002 124 (32), 9458. doi:10.1021/
ja0267831.
(13) Ivanec-Goranina, R.; Kulys, J. Cent. Eur. J. Biol. 2008 3 (3),
224. doi:10.2478/s11535-008-0021-x.
(
8) (a) Lukeman, M.; Wan, P. J. Am. Chem. Soc. 2003 125 (5),
1
164. doi:10.1021/ja029376y.; (b) Flegel, M.; Lukeman, M.;
Wan, P. Can. J. Chem. 2008 86 (2), 161. doi:10.1139/V07-
43.
(14) (a) Valeur, B.; Weber, G. Photochem. Photobiol. 1977 25
(5), 441. doi:10.1111/j.1751-1097.1977.tb09168.x.; (b) Pines,
E.; Huppert, D.; Agmon, N. J. Chem. Phys. 1988 88 (9),
5620. doi:10.1063/1.454572.
(15) Wu, J. I.; Dobrowolski, M. A.; Cyra n´ ski, M. K.; Merner, B.
L.; Bodwell, G. J.; Mo, Y.; Schleyer, P. R. Mol. Phys. 2009
107 (8), 1177. doi:10.1080/00268970902784918.
1
(
9) Flegel, M.; Lukeman, M.; Huck, L.; Wan, P. J. Am. Chem.
Soc. 2004 126 (25), 7890. doi:10.1021/ja039078g.
10) Wunder, T.; Marr, J.; Kremer, S.; Sterner, O.; Anke, H.
Arch. Microbiol. 1997 167 (5), 310. doi:10.1007/
s002030050449.
(
Published by NRC Research Press