Photophysical and electrochemical investigations of the fluorescent probe, 4,4′-Bis(2-benzoxazolyl)stilbene

Article abstract of DOI:10.1021/jp312598c

In solution, 4,4′-bis(2-benzoxazolyl)stilbene (BBS) was found to exhibit consistently high absolute fluorescence quantum yields (Φ_fl ≥ 0.88) and a monoexponential lifetime, both independent of BBS concentration. The BBS steady-state and time-resolved photophysics were investigated by different techniques to understand the various deactivation pathways. Nonradiative deactivation of BBS singlet excited state by intersystem crossing was found to be negligible. Other than fluorescence, the excited state of BBS was found to be deactivated by trans-cis photoisomerization. At low concentrations (≈5 μg/mL), UV spectroscopy and laser flash photolysis showed concordant results that the photoinduced cis isomer gradually replaced the original absorption spectrum of the pure trans isomer. However, at high concentrations (≈0.2 mg/mL), ¹H NMR and DOSY measurements confirmed that irradiating BBS at 350 nm induced a conversion from the trans-BBS into its cis isomer by photoisomerization. It was further found that the stilbene moiety of both isomers was photocleaved. The resulting photoproduct was an aldehyde that was oxidized under ambient conditions to its corresponding carboxylic acid, i.e., 4-(1,3-benzoxazol-2-yl)benzoic acid. The structure of the photoproduct was unequivocally confirmed by X-ray diffraction. Spectroscopic investigation of BBS showed a limited photoisomerization after irradiation at 350 nm of a trans solution. The BBS electrochemistry showed irreversible oxidation, resulting in an unstable and highly reactive radical cation. Similarly, the cathodic process was also found to be irreversible, giving rise to a radical anion and showing its n-doping character.

Full text of DOI:10.1021/jp312598c

Article
pubs.acs.org/JPCA
Photophysical and Electrochemical Investigations of the Fluorescent
Probe, 4,4′-Bis(2-benzoxazolyl)stilbene
M. Amine Fourati, W. G. Skene,* C. Geraldine Bazuin,* and Robert E. Prud’homme*
́
́ ́ ́ ́ ́
Departement de Chimie, Universite de Montreal, CP 6128, Succ. Centre-Ville, Montreal, Quebec, Canada, H3C 3J7
S
* Supporting Information
ABSTRACT: In solution, 4,4′-bis(2-benzoxazolyl)stilbene (BBS)
was found to exhibit consistently high absolute fluorescence
quantum yields (Φfl ≥ 0.88) and a monoexponential lifetime, both
independent of BBS concentration. The BBS steady-state and
time-resolved photophysics were investigated by different
techniques to understand the various deactivation pathways.
Nonradiative deactivation of BBS singlet excited state by
intersystem crossing was found to be negligible. Other than
fluorescence, the excited state of BBS was found to be deactivated
by trans−cis photoisomerization. At low concentrations (≈5 μg/
mL), UV spectroscopy and laser flash photolysis showed concordant results that the photoinduced cis isomer gradually replaced
the original absorption spectrum of the pure trans isomer. However, at high concentrations (≈0.2 mg/mL), 1H NMR and DOSY
measurements confirmed that irradiating BBS at 350 nm induced a conversion from the trans-BBS into its cis isomer by
photoisomerization. It was further found that the stilbene moiety of both isomers was photocleaved. The resulting photoproduct
was an aldehyde that was oxidized under ambient conditions to its corresponding carboxylic acid, i.e., 4-(1,3-benzoxazol-2-
yl)benzoic acid. The structure of the photoproduct was unequivocally confirmed by X-ray diffraction. Spectroscopic investigation
of BBS showed a limited photoisomerization after irradiation at 350 nm of a trans solution. The BBS electrochemistry showed
irreversible oxidation, resulting in an unstable and highly reactive radical cation. Similarly, the cathodic process was also found to
be irreversible, giving rise to a radical anion and showing its n-doping character.
beneficial to determine whether BBS can be photoisomerized
and under what conditions. To the best of our knowledge, the
BBS photoisomerization dissipating energy processes have not
been previously studied.
1. INTRODUCTION
An extended π-conjugation gives 4,4′-bis(2-benzoxazolyl)-
stilbene (BBS) optical properties that can lead to its use as
photoactive switches1,2 and as a fluorescent brightener in
textiles,3 detergents, and other materials.3,4 Furthermore, BBS’s
capacity to form excimers, leading to a change of its color under
UV illumination at high concentrations in polymer films, has
opened the door for applications in host−guest systems as a
molecular probe of deformation and temperature.5−7 Its
anisotropic flat shape could also be used to produce polarized
light in oriented macromolecular matrices.8 In addition, BBS
exhibits resistance to solvent extraction as well as high melting
and degradation temperatures, above 300 °C, that satisfy the
regulations of the U.S. Food and Drug Administration (FDA),
thus enabling its use as an additive for food and consumer
packaging materials.9
Although BBS has been used for different applications
because of its optical and thermal properties, no extended opto-
electrochemical studies of this fluorophore have been pursued.
In particular, it would be interesting to probe the BBS
photophysics to get a clear idea about the different energy
dissipating processes that may be responsible of the lowering of
its fluorescence quantum yield, which remains its greatest asset.
Such studies are pivotal for determining the fluorescence BBS
limitations and for its use as a universal fluorophore. Our
purpose is to report the BBS photophysics and electrochemistry
to try to better exploit its advantages.
The spectroscopic properties of BBS are in part courtesy of
the central stilbene. This provides a conjugated framework
across which the electron rich benzoxazoyl termini delocaliza-
tion can occur, resulting in a high degree of conjugation. The
rigid framework is additionally responsible for its intense
emission. Similar to other stilbene derivatives, BBS potentially
can photoisomerize between its trans and cis isomers.10−13 This
competing excited-state deactivation process could affect BBS’
fluorescence properties, and in turn, limit its use in applications
that require intense emission.14−16 For this reason, it would be
2. EXPERIMENTAL SECTION
Materials. 4,4′-Bis(2-benzoxazolyl)stilbene (97%, melting
point >300 °C) was purchased from Aldrich Chemicals. The
compound was purified by sonicating and heating the material
in the given solvent. The undissolved solids were subsequently
Received: August 8, 2012
Revised: January 8, 2013
Published: January 10, 2013
© 2013 American Chemical Society
836
dx.doi.org/10.1021/jp312598c | J. Phys. Chem. A 2013, 117, 836−844

The Journal of Physical Chemistry A
Article
removed by filtering upon cooling the solution and the filtered
solution was then used for the measurements.
data processing. Chemical shifts were found relative to TMS,
and they were derived from spectral simulations using both
ChemDraw Ultra 8.0 and DFT calculations. Diffusion experi-
ments were performed at 298 K on a Bruker AVII-400 MHz
spectrometer, which was equipped with a diff60 probe. A
standard stimulated-echo pulse sequence was used with a
gradient duration (δ) of 1 ms, a diffusion time (Δ) of 40 ms, a
repetition time of 5 s and 16 gradients values. An average of
around 128 scans were accumulated.
Crystallographic Study. A needle yellow crystal (0.20 ×
0.03 × 0.03 mm3), suitable for X-ray diffraction analysis, was
obtained by partial slow evaporation from a TCE solution after
UV irradiation at 350 nm, triggering an increase of the
concentration of 0.2 mg/mL to about 0.8 mg/mL. The single
crystal study was carried out at 100 K using a Bruker APEX II
diffractometer equipped with an Incoatec microsource
generator producing Cu Kα radiation, a three-circle platform
goniometer and an APEX2 CCD detector.
The structure was solved by direct methods using SHELXS
97.20 All atoms from the main molecule were refined with
anisotropic thermal displacement parameters. Hydrogen atoms
were positioned geometrically and refined with a riding model,
with C−H distances of 0.95 Å and Uiso (H) values constrained
to be 1.2 times Ueq of the carrier atom. The structure contained
an included solvent molecule that was found highly disordered.
Several electron density peaks were located that did not
assemble to a recognizable fragment of TCE molecule. So the
solvent contribution was taken on using the SQUEEZE
option21 implemented in the PLATON software.22
Spectroscopic Measurements. Absorbance measure-
ments were done on a Varian Cary-500 UV−visible
spectrometer, and the solvent absorption spectra were
subtracted from the spectra of the analyzed samples, having a
concentration of 5 μg/mL. Fluorescence emission spectra were
recorded at ambient temperature in 10 mm cuvettes, excited at
the corresponding absorption maximum, using an Edinburgh
Instruments FLS-920 combined steady-state and time-resolved
fluorometer. The fluorescence lifetimes were measured
according to standard time-correlated single photon counting
methods, with the FLS-920 apparatus, by fitting the data with a
monoexponential decay function. The calculated values have an
experimental error of approximately 10%. Absolute fluores-
cence quantum yields were measured by using an integrating
sphere and used to quantify the efficiency of the emission
process by taking the ratio of photons absorbed to photons
emitted through fluorescence at a particular wavelength in the
same period of time.17,18
The HOMO and LUMO frontier molecular orbitals were
calculated semiempirically using DFT methods available in
Spartan 06 with the 6-31g* basis set. The bond angles,
distances, torsions, and other parameters used for the
theoretical calculations were experimentally derived from the
X-ray data for BBS.19
Laser Flash Photolysis (LFP). The triplet−triplet
absorption spectra were measured in anhydrous 1,1,2,2-
tetrachloroethane (TCE) (from Aldrich Chemicals) at a BBS
concentration of 5 μg/mL using a Luzchem mini-LFP system,
excited at 355 nm from the third harmonic of a Continuum
YAG:Nd Sure-lite laser, using a 175 W xenon lamp. An average
of 15 shots per wavelength was used for generating the
transient absorption spectrum. The samples were dissolved in
anhydrous TCE in static quartz cuvettes and had an absorbance
of less than 0.4. All samples were purged with nitrogen for
more than 20 min before LFP analyses.
Photoirradiation. The sample for UV measurement was
degassed in a 20 mL vial by bubbling nitrogen directly into the
solution for about 30 min before sealing the container to avoid
contamination from air. The sample was then irradiated with
350 nm lamps in a Luzchem photoreactor. A small volume of
the irradiated solution was removed from the reactor and
transferred into a quartz 10 × 10 mm cell at prescribed times,
and the absorbance spectrum was measured. For the NMR
studies, the sample was introduced into a J-Young 7 mm tube at
high concentration (≈0.2 mg/mL). The tube was heated and
placed in an ultrasound bath flowed by three successive freeze−
pump−thaw cycles before it was sealed and irradiated at 350
nm in the same photoreactor. The NMR tube was then taken
out at different times for measuring NMR spectra. For the
crystallographic studies, a degassed 0.2 mg/mL BBS solution in
TCE was irradiated for 10 h by using 350 nm lamps in the
Luzchem photoreactor.
3. RESULTS AND DISCUSSION
The photophysical properties of BBS were measured in TCE at
a concentration of 5 μg/mL, where it exhibits an absorption
maximum at 374 nm, an emission maximum at 434 nm and a
molar absorption coefficient of 65 000 M−1 cm−1. Chlorinated
solvents (TCE, CHCl3, and CDCl3) were chosen for evaluating
the physical properties of BBS owing to the high solubility of
BBS in these solvents.23 Though chlorinated solvents are
assumed not to be ideal solvents for photophysical studies
because of their propensity to produce radicals upon direct
photoirradiation, these undesired byproducts can be avoided by
irradiating at wavelengths >350 nm. Photophysical studies of
BBS were therefore done by irradiating at wavelengths outside
the absorbance windows of the solvents, being at wavelengths
>254 nm. Irradiations were done in Pyrex instead of quartz
glassware because of its optical transmission cutoff of <350 nm.
Using Pyrex glassware therefore ensures the removal of UV
wavelengths from the broad emitting lamps that could
potentially induce radical formation upon direct excitation of
the chlorinated solvents.
The BBS solution emission has a unique lifetime on the
order of 1 ns. The monoexponential emission confirms the
absence of bimolecular species. The consistent unimolecular
lifetime was observed even when the BBS concentration was
increased. Given that previously reported BBS emission yields
were done by relative actinometry, which is an imprecise
method,24 we measured its absolute fluorescence quantum
yields (Φfl) more accurately with an integrating sphere.25 BBS
was found to fluoresce strongly in solution with Φfl = 0.88 in
TCE. The measured value is higher than that of trans-stilbene
(0.80 at 25 °C).26
1H NMR Spectroscopy. A typical sample was prepared by
dissolving a 0.2 mg/mL solution of BBS in CDCl3 or in
deuterated TCE (both from Aldrich Chemicals) in a J-Young 7
mm tube by alternating between heating and placing the sample
in an ultrasound bath to avoid precipitation and to accelerate
solubility. The tube was sealed and it was subjected to three
successive freeze−pump−thaw cycles. Spectra were run on a
Bruker AV-400 MHz spectrometer for all compounds dissolved
in CDCl3. MestRe-C and TopSpin software was used for NMR
The measured absolute Φfl confirms that BBS is highly
fluorescent. However, the lower than unity Φfl indicates that
837
dx.doi.org/10.1021/jp312598c | J. Phys. Chem. A 2013, 117, 836−844

The Journal of Physical Chemistry A
Article
deactivation processes other than fluorescence occur. The
nonradiative deactivation processes were subsequently inves-
tigated, notably its photoisomerization. The BBS trans−cis
photoisomerization was investigated by both steady-state and
time-resolved techniques according to UV spectroscopy and
laser flash photolysis (LFP), respectively. Figure 1 shows that
Figure 2. BBS average transient absorbance response at 308 nm (red
line), specific to cis-BBS, and 374 nm (black line), characteristic of
trans-BBS, in TCE measured 0.3 μs after the laser pulse at 355 nm.
Inset: transient absorbance spectra of BBS recorded in TCE at 9.79
(black square) and 27.09 μs (red square), after the laser pulse at 355
nm.
new absorbance at 450 nm is assigned to the BBS triplet. This
was based on its monoexponential decay and its lifetime (8 μs)
that is consistent with triplets.27 Unequivocal proof for the
triplet was obtained by quenching the transient with 1,3-
cyclohexadiene, which is a well-known triplet deactivator.28
The transient absorbance kinetics were measured at both 308
and 374 nm to confirm the identity of the photoinduced
products. These wavelengths were chosen because they
correspond to the absorbance maximum measured at steady
state. As seen in Figure 2, the absorbance at 308 nm grows
according to first-order kinetics with a lifetime of 3.5 μs. It is
further obvious that the transient signal does not decay within
the time resolution of the LFP instrument. The transient
kinetic at 374 nm is the mirror image of that observed at 308
nm. The negative signal is a result of the ground state being
depleted. Similar to what was observed at 308 nm, the signal at
374 nm does not recover during the course of the
measurement. This implies that the absorbance at 374 nm is
irreversibly converted into a photoproduct whereas the
absorbance at 308 nm is the formation of a photoproduct
that is stable within the time resolution of the instrument.
Though the exact photoisomerization mechanism and the
lifetimes of the discrete processes cannot be determined, the
overall trans → cis photoconversion is 2.5 μs, according to the
LFP measurements. The similar microsecond kinetics and the
mirror image of the two absorbances confirm the interdepend-
ence of the signals at 308 and 374 nm. The signal of the trans-
BBS at 374 nm is therefore converted into the cis isomer, which
is observed at 308 nm. The results are consistent with the
steady-state measurements. From the plateau regions of the
kinetics, the molar absorption coefficient (ε) of the cis isomer
can be approximated to be 38 500 M−1 cm−1 at 308 nm.
The efficiency of photoisomerization cannot accurately be
quantified from either the time-resolved or steady-state
analyses. However, taking into account the absolute ϕfl and
assuming a minor triplet contribution as per the absorbance at
450 nm, the trans → cis photoisomerization efficiency is
approximately 10%.
Figure 1. BBS trans−cis photoisomerization in TCE measured by
absorbance spectroscopy at room temperature at 0 (black box), 10
(red box), 30 (blue box), 40 (green box), 60 (black circle), 80 (red
circle), 100 (blue circle), 120 (green circle), 160 (black triangle), 190
(red triangle), 220 (blue triangle), 250 (green triangle), and 320 min
(open box) after irradiation at 350 nm with a 16 W light source.
the absorbance spectrum of trans-BBS exhibits three peaks at
358, 374, and 395 nm. Upon irradiating at 350 nm, at room
temperature, these three peaks decrease considerably in
intensity over a period of 5 h, without disappearing completely.
This is accompanied by a concomitant increase in absorbance
at 308 nm, which is assigned to the cis-BBS. Figure 1 also shows
the existence of an isosbestic point at 330 nm, at which point
the trans and cis isomers have the same extinction coefficient.
This isosbestic point indicates that only two species that vary in
concentration contribute to the absorbance. It further confirms
the interdependence of the two species, being the inter-
conversion of the trans and cis isomers. It should be noted that
at longer irradiation times at 350 nm, the isosbestic point
vanishes (data not shown) and the intensity of both isomers
eventually disappears. This implies that both isomers are
converted into a new product with extended irradiation (vide
infra).
In a second step, BBS was investigated by LFP to confirm the
trans−cis photoisomerization. The advantage of the time-
resolved method versus steady-state irradiation is that the
transients are instantaneously produced within the short laser
pulse (ca. 10 ns). The intense laser power further produces
sufficient transients within its pulse for them to immediately be
detected. This is in contrast to steady irradiation that requires
extended irradiation times to detect any light induced species.
The inset of Figure 2 shows the transient absorbance spectra
of BBS measured in TCE at 9.5 and 18.5 μs after the laser
pulse. The negative Δ absorbance at 350 nm corresponds in
part to the ground-state bleaching of the trans-BBS and the
signal at 300 nm is ascribed to the cis isomer. Meanwhile, the
838
dx.doi.org/10.1021/jp312598c | J. Phys. Chem. A 2013, 117, 836−844

The Journal of Physical Chemistry A
Article
Chart 1. BBS trans-cis Photoisomerization and Conversion of Both Isomers in BBA
1
Figure 3. H NMR spectrum between 6.5 and 10.3 ppm, of trans-BBS in CDCl3, after 0 (a), 4 (b), 10 (c), and 20 h (d) of irradiation at 350 nm
(light power, P = 16 W).
NMR spectroscopy was used to both investigate the presence
of additional photoinduced processes and to confirm the trans−
cis photoisomerization. The advantage of NMR spectroscopy is
that the cis and trans isomers of stilbene derivatives can readily
confirm the accuracy of the predicted chemical shifts of cis-BBS,
the H NMR spectrum of trans-BBS was also calculated. The
1
predicted chemical shifts for the vinylene protons were 6.56
and 7.26 ppm, respectively, for the cis- and trans- BBS isomers.
The calculated chemical shift for the trans-vinylene proton
(located at position f in Chart 1) is in accordance with the
experimental value of 7.32 ppm, as per Figure 3a.
Figure 3a shows the NMR spectrum of trans-BBS, prior to
irradiating at 350 nm. The singlet at 7.32 ppm is assigned
proton f (Chart 1). It should be noted that there is no signal at
around 6.5 ppm that can be ascribed to the cis isomer. Figure
3
be distinguished by their distinctive JHH coupling con-
stants.29,30 However, this was not possible because of BBS’
high degree of symmetry that gives rise to a singlet for the
vinylene protons instead of a doublet. Therefore, the chemical
shift of the cis-vinylene protons was estimated by theoretical
means using commercial software. This is particularly important
given the absence of any reported NMR data for cis-BBS. To
839
dx.doi.org/10.1021/jp312598c | J. Phys. Chem. A 2013, 117, 836−844

The Journal of Physical Chemistry A
Article
Figure 4. DOSY NMR results for BBS, after 4 (a) and 10 h (b) of UV irradiation. All spectra were recorded in CDCl3 at room temperature at 400
1
MHz. The x-axis represents the standard H dimension, from 6 to 10.5 ppm, and the y-axis represents the diffusion dimension.
3b displays the BBS NMR spectrum after 4 h of UV irradiation.
It is obvious that a new singlet appears at 6.76 ppm, belonging
to cis-BBS, in addition to the singlet of trans-BBS. The
formation of the cis isomer is accompanied by a change of the
positions of the closest peaks to the central vinylic bond.
Indeed, the intensity of the peaks corresponding to protons at
positions d and e, located at 8.33 and 7.74 ppm of trans-BBS,
decreased progressively, while new peaks corresponding to the
same positions appear at 8.22 and 7.59 ppm for cis-BBS,
respectively. It should be noted that the remaining peaks,
belonging to the protons at positions a, b, and c remain
unchanged. Figure 3c shows the NMR spectrum after 10 h of
UV irradiation. The appearance of a well-defined singlet at
10.15 ppm is consistent with a photoproduct having an
aldehyde. The new resonance is accompanied by the emergence
of new peaks at 8.48 and 8.08 ppm, corresponding to the
protons at positions d and e, respectively. Figure 3d shows the
NMR spectrum after 20 h of UV irradiation, displaying the
disappearance of the cis isomer, the increase of the photo-
product singlet at 10.15 ppm and the decrease of the trans-BBS
singlet at 7.32 ppm. When the BBS solution is irradiated for
longer times, the singlet at 7.32 ppm, specific to trans-BBS,
disappears while the peaks specific to the photoproduct become
more intense. The emergence of new peaks specific to the
protons at positions d and e, in addition to the appearance of a
new well-defined singlet at 10.15 ppm, which has been
attributed to an aldehyde, confirm the presence of another
conversion process.
By integrating the cis isomer signal at 6.76 ppm relative to the
trans peak at 8.33 ppm, it was found that the cis integral
gradually increases before reaching a plateau of 18% relative to
the sum of the trans and cis isomers after 3 h of UV irradiation
in CDCl3. After which time, the signal decreases until becoming
null after around 14 h of UV irradiation. These results show
that another process occurs, leading to a simultaneous
conversion of both isomers into a photoproduct, thus
suppressing the BBS trans−cis photoisomerization. Because
UV spectroscopy shows a greater conversion from trans to cis,
even when the difference of the molar absorption coefficients of
the two isomers was taken into account, a series of absorbance
measurements were made with a concentration similar to that
used for the NMR study with an initial concentration of ≈0.2
mg/mL. After irradiation, an aliquot was removed and it was
diluted to 0.005 mg/mL, similar to the concentration used in
Figure 1. This was required to not saturate the detector of the
UV spectrometer and to get an absorbance below unity for each
measurement. This series of measurements led to a trans−cis
conversion similar to that found by NMR in Figure 3. It appears
that a high trans−cis conversion is obtained upon irradiating at
low concentrations (Figure 1). In contrast, a low conversion is
obtained upon irradiating at high concentrations (Figure 3).
This is probably due to an inner-filter effect and/or self-
quenching at high concentrations that favor continuous energy
transfer between BBS molecules. On the other hand, the
aldehyde signal at 10.15 ppm continuously increases relative to
the trans peak at 8.33 ppm. Also, the chemical shifts of the a, b,
and c protons remain unchanged. These data collectively
suggest that the second process does not affect the benzoxazol
part of the molecule, but only the stilbene moiety. The higher
than normal downfield chemical shift of the aldehyde confirms
that it is highly conjugated. This is only possible when it is
conjugated with the benzoxazol unit. The photoproduct
therefore is most likely from stilbene photooxidation in the
presence of trace amounts of residual oxygen. The photo-
cleavage resulting in aldehyde photoproducts is a proven
degradation mechanism of stilbenes31−33 and it makes the
trans−cis photoisomerization irreversible.
To gain more information about the presence of a secondary
process, diffusion ordered spectroscopy (DOSY) experiments
were conducted. DOSY enables us to differentiate between
blend components because, in a solution of a given viscosity,
molecules of various sizes diffuse at different rates.34−38 The
diffusion coefficients (Ddiff) of the individual species can be
determined by the DOSY experiments and related to their
molecular hydrodynamic volume. These experiments were
conducted at identical concentrations and temperatures in
CDCl3.
Figure 4 shows the diffusion coefficient (Ddiff) as a function
of the chemical shift between 6 and 10.5 ppm after 4 (a) and 10
h (b) of UV irradiation. Figure 4a shows that the trans-BBS
original signals at 8.33, 7.74, and 7.32 ppm, as well as the cis
isomer signals at 8.22, 7.59, and 6.76 ppm, corresponding to
protons d, e, and f, have the same diffusion coefficient of 0.8 ×
10−9 m2·s−1. Figure 4b shows that, after 10 h of UV irradiation,
840
dx.doi.org/10.1021/jp312598c | J. Phys. Chem. A 2013, 117, 836−844

The Journal of Physical Chemistry A
Article
Figure 5. Atomic displacement ellipsoid plot of BBA. The second molecule (atoms labeled with suffix A) are generated by the symmetry operation
−x − 1, −y − 1, −z. Hydrogen bonds are shown as dotted lines. Ellipsoids are drawn at the 50% probability level and hydrogen atoms are
represented by sphere of arbitrary size.
Figure 6. View of the π-stacking interactions between two hydrogen-bonded dimers. Hydrogen bonds and the shortest distances between ring
centroids involved in π-stacking interactions are indicated by dotted lines.
the emerging peaks at 8.48 and 8.08 ppm, corresponding to
protons d and e, respectively, have higher diffusion coefficients
of the order of (1.2−1.3) × 10−9 m2·s−1, as well as the new
singlet, located at 10.15 ppm. These results show that these
three peaks do not belong to the original BBS, but to a new
product. Therefore, the Ddiff values are so different, before and
after a prolonged UV irradiation, that a standard deviation of up
to 10% cannot affect the understanding of the conversion. The
presence of two singlets at 7.02 and 7.54 ppm are evident in
Figure 4. These are assigned to impurities because they are
always present. They additionally diffuse rapidly, which
confirms that they are smaller than BBS. According to Figure
3A, these impurities are minor and they are present in less than
2 mol %. Although the commercial source of BBS is 97% pure,
it was purified by alternating between heating and sonication,
followed by filtering. This leads to purer BBS than the original
commercial source. Moreover, these impurities do not interfere
with the isomerization studies because they do not contain a
vinyl group and they do not absorb at the wavelengths used for
photoirradiation, owing to their limited degree of conjugation.
On the other hand, the solvent (CDCl3) diffusion coefficient is
the same, within uncertainty, for the different solutions,
showing similar viscosities before and after irradiation. The
Stokes−Einstein equation, applicable for spherical molecules
and describing diffusion phenomena, is
eq 1, the hydrodynamic radius of the photoproduct is estimated
to be 1.6 times smaller than BBS.
To gain a deeper understanding of the products formed from
UV irradiation and, more specifically, to unequivocally identify
both the photoproduct and cis-BBS, crystallographic studies
were conducted. For this, a 0.2 mg/mL solution of BBS in
TCE, similar in concentration to that used for NMR
measurements, was irradiated at 350 nm for 10 h. The resulting
solution contained both BBS isomers and the photoproduct.
This was confirmed by NMR (Figure 3c). The solvent was then
partially evaporated, as described in the Experimental Section.
Unlike what was expected, the single crystal that formed upon
partial solvent evaporation was neither the cis nor the trans
isomer. In fact, the X-ray diffraction of the single crystal
confirmed that the photoproduct was a carboxylic acid, 4-(1,3-
benzoxazol-2-yl)benzoic acid (BBA) (Figure 5). The latter is a
result of photocleavage of the stilbene moiety, leading to two
half-molecules with an acid terminal function for each.
Although an aldehyde photoproduct was observed in NMR
studies, this intermediate is readily oxidized to BBA under the
aerobic crystallization conditions, according to known
means.39,40 Irradiation of BBS at 350 nm therefore leads to
trans−cis photoisomerization and the two isomers are photo-
oxidized with residual oxygen. The resulting aldehyde photo-
product is further oxidized to BBA. The conversion of both
photoisomerization and photocleavage is shown in Chart 1.
The photocleavage is concentration-dependent and is slow at
concentrations ≤5 μg/mol and faster at concentrations ≥0.2
mg/mL.
kBT
6πηr
D
=
diff
(1)
BBA was found to crystallize in the triclinic space group P1
with one independent molecule in the asymmetric unit. It is
almost planar, and the angle between the benzene ring and the
benzoxazolyl group is only 1.50(14)°, which is smaller than that
found in BBS (2.30(16)°).19 The different BBA molecules were
stable due to intermolecular interactions. Indeed, Figure 5
shows that two BBA molecules form a dimer, linked by two
hydrogen bonds involving the carboxylic acid groups according
to the classical graph set motif R22(8).41
̅
where Ddiff is the diffusion coefficient, r is the hydrodynamic
radius of the molecule, kB is the Boltzmann constant, T is the
temperature and η is the viscosity. Ddiff is inversely proportional
to the hydrodynamic radius. By assuming that the molecules are
spherical, the higher value of Ddiff after 10 h of UV irradiation
proves that the BBS molecule is converted into a molecule of
lower hydrodynamic radius. Meanwhile, the new well-defined
singlet at 10.5 ppm confirms it has an aldehyde. In contrast, a
broad resonance would be observed for a carboxylic acid, whose
chemical shift would be concentration-dependent. According to
Figure 6 shows that π-stacking interactions between the
benzene and the oxazole rings enable a packing with the
841
dx.doi.org/10.1021/jp312598c | J. Phys. Chem. A 2013, 117, 836−844

The Journal of Physical Chemistry A
Article
shortest centroid to centroid distances of 3.58 and 3.68 Å. The
chains of dimers are arranged parallel to the a-axis and the
molecules from these chains are connected to a neighboring
chain via C−H···O interaction (d(C−O) = 3.26 Å) to form
layers in the (a, b) plane. It should be noted that the disordered
solvent molecules occupy the remaining space between the
layers (35% of the unit-cell volume).
The most interesting property of BBS is its fluorescence.
Though extended characterization of the trans isomer is
possible, the cis isomer has not previously been characterized.
This is primarily owing to the difficulty in isolating the sterically
hindered isomer. It would nonetheless be beneficial to
characterize the fluorescence of the cis isomer for gaining
valuable insight into both the photoisomerization process and
to determine the limit of usefulness of BBS as a fluorescence
probe. For these reasons, we extended our fluorescence studies
to include the cis and trans isomers. However, given the
challenges in isolating the pure cis isomer, spectral character-
izations were done indirectly on mixed samples having both cis
and trans isomers. According to Figure 7, three well-defined
shoulders at 384 and 474 nm, similar to analogous crescent-like
conjugated stilbenes.42,43 These shoulders are specific to cis-
and trans-BBS, respectively, whereas the emission maximum at
422 nm results from an overlap between the emissions of the
two BBS isomers. The emission of the cis isomer is then
hypsochromically shifted relative to the trans spectrum and the
presence of both isomers even after prolonged irradiation
suggests a photostationary cis−trans conversion. This hampers
accurate determination of the spectroscopic data specific to cis-
BBS. Because the cis isomer was impossible to isolate, the
solution obtained after 5 h of UV irradiation was used to
measure the fluorescence quantum yield and the lifetime of the
cis- and trans-BBS mixture relative to the pure trans isomer.
Only qualitative information is possible given that both isomers
have different molar absorption coefficients and they do not
absorb the same number of photons. Both the mixed isomer
and pure trans isomer solutions had similar fluorescence
lifetimes, being ca. 1 ns. In contrast, the fluorescence quantum
yield decreased from 0.88 for pure trans-BBS to 0.67 for the
solution containing a mixture of both isomers. A rough estimate
of the energy gap (Eg), i.e., the absolute energy difference
between the ground (HOMO) and excited (LUMO) states,
was calculated from the intercept of the normalized absorption
and emission spectra (Figure 7). The approximated values are
3.1 eV for trans-BBS and 3.4 eV for the irradiated solution (for
5 h) containing trans and cis isomers.
The energy gaps not only can be obtained from the intercept
of the normalized absorbance and fluorescence spectra, as
shown in the previous section, but also can be obtained by
cyclic voltammetry. According to Figure 8, trans-BBS under-
Figure 7. Normalized absorbance and emission spectra in TCE of BBS
before UV irradiation: absorbance (red box) and emission (red circle);
and after 5 h of irradiation at 350 nm: absorbance (black box) and
fluorescence (black circle).
peaks, located at 412, 434, and 464 nm, are seen in the trans-
BBS fluorescence spectrum. These are assigned to the 00,
01, and 02 radiative transitions, respectively, and they are
characteristic of BBS monomers. In comparison, the
absorbance spectrum of the irradiated sample exhibits three
ill-defined peaks at 358, 374, and 395 nm. Both the absorbance
and fluorescenc spectra of the cis isomer are hypsochromically
shifted further bathochromically shifted relative to trans isomer
because of its reduced degree of conjugation. No spectral
changes were observed for the trans absorbance spectrum with
increasing the BBS concentration.
Figure 7 also shows the spectroscopic features of a solution
of trans-BBS irradiated for 5 h at 350 nm. The absorbance
spectrum exhibits an intense peak at 308 nm and a shoulder at
375 nm. These are assigned to the absorbance maximum of the
cis and trans isomers, respectively (vide supra). Similarly, the
emission spectrum shows a maximum at 422 nm and two
Figure 8. Cyclic voltammogram of trans-BBS recorded in TCE at 100
mV/s against SCE with Bu4NPF6 as the supporting electrolyte.
goes a one-electron oxidation process, demonstrating its p-
doping-type behavior. Similarly, BBS undergoes a one-electron
reduction, at low potential, confirming its n-doping character.
Both the oxidation and reduction processes observed were
irreversible, regardless of the scan rate, indicating highly
reactive radical ion intermediates.
From the cyclic voltammetry data, we can calculate E
ox
onset
,
which is the oxidation potential onset versus the SCE electrode,
and Eroendset, which is the reduction potential onset. Their values,
taken from Figure 8, are given in Table 1. The HOMO energy
value can be calculated from the ionization potential (IP),
ox
onset
according to the well accepted approximation of IP = E
+
842
dx.doi.org/10.1021/jp312598c | J. Phys. Chem. A 2013, 117, 836−844

The Journal of Physical Chemistry A
Article
a
be too slow to be detected at low concentrations. The limited
photoisomerization at high concentrations was proved by a
combined absorbance and emission spectra of both trans and cis
species. Electrochemistry of trans-BBS exhibits both irreversible
oxidation and reduction.
Table 1. trans-BBS Electrochemical Properties
b
c
d
e
f
Eox /V
Ered /V
IP /eV
EA /eV
Eegl /eV
1.66
−1.79
5.72
2.87
2.9
a
b
The electrochemical data are reported against SCE. Oxidation
potential. Reduction potential. Ionization potential. Electron
c
d
e
f
affinity. Electrochemical energy-gap.
ASSOCIATED CONTENT
* Supporting Information
■
S
4.4.44−47 Similarly, the LUMO energy level is derived from the
Magnification of the 1H NMR spectrum, from 6.5 to 10.3 ppm,
of trans-BBS in CDCl3, after 0, 4, 10, and 20 h of irradiation at
350 nm (light power, P = 16 W). 1H NMR spectrum, from 6.5
to 10.2 ppm, of trans-BBS in d-TCE, before irradiation at 350
nm (light power, P = 32 W). This information is available free
of charge via the Internet at http://pubs.acs.org.
red
onset
electron affinity (EA), according to EA = E
+ 4.4. The IP
and EA values of BBS are also reported in Table 1. The
difference between the HOMO and LUMO energy levels gives
the electrochemical energy-gap, Eegl (2.9 eV in Table 1) that is
similar to the spectroscopic energy gap, Eg (3.1 eV from Figure
7). The high degree of conjugation of the BBS molecule is
confirmed, first, by both the electrochemical and spectroscopic
data and, second, by the theoretically calculated molecular
orbitals that are depicted in Figure 9. The latter show the
AUTHOR INFORMATION
Corresponding Author
*Correspondence email: re.prudhomme@umontreal.ca,
geraldine.bazuin@umontreal.ca and w.skene@umontreal.ca.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors gratefully acknowledge financial support from the
Natural Sciences and Engineering Research Council of Canada,
■
the Fonds Queb
Technologies, the Canada Foundation for Innovation, Nano-
Quebec and the Centre for Self-Assembled Chemical
Structures. They also thank Thierry Maris and Cedric Malveau,
from Universite de Montreal, for their help in X-ray diffraction
and diffusion ordered spectroscopy experiments, respectively.
W.G.S. thanks both the Humboldt Foundation and the Royal
Society of Chemistry for fellowships allowing the manuscript to
be drafted. M.A.F. thanks the Tunisian Government for a
scholarship of excellence.
́ ́
ecois de la Recherche sur la Nature et les
́
́
́
́
Figure 9. BBS LUMO (top) and HOMO (bottom) features calculated
by DFT, using the 6-31g* basis set and the X-ray crystal data.19
delocalization of the HOMO along the rigid backbone of BBS,
confirming its high degree of conjugation. Conversely, the
LUMO is located exclusively on the central stilbene moiety.
REFERENCES
■
(1) Irie, M. Chem. Rev. 2000, 100, 1685−1716.
(2) Momotake, A.; Arai, T. J. Photochem. Photobiol. C Photochem. Rev.
2004, 5, 1−25.
4. CONCLUSIONS
In solution, BBS fluoresces with high quantum yields (Φfl ≈
0.88), regardless of its concentration. The steady-state and
time-resolved photophysics of this stilbene-derivative fluoro-
phore have been investigated for the first time in this study by
absorbance, laser flash photolysis, and NMR spectroscopy, to
unravel its quenching mechanism, whereas its fluorescent and
electrochemical features were studied by fluorescence spectros-
copy and cyclic voltammetry, respectively.
Although the primary deactivation mode of the BBS singlet
excited state is by fluorescence, it also deactivates by trans−cis
photoisomerization. Intersystem crossing to the triplet state was
found to be negligible and excimer formation was nonexistent,
proved by a concentration-independent monoexponential
lifetime, of the order of 1 ns.
At low concentrations (≈5 μg/mL), irradiating BBS at 350
nm induces a trans−cis photoisomerization. However, at higher
concentrations (≥0.2 mg/mL), this process is accompanied by
formation of an oxidized photoproduct, occurring by photo-
cleavage at the stilbene moiety. The resulting aldehyde is
oxidized to its corresponding carboxylic acid under ambient
conditions. X-ray crystallography confirmed that the oxidized
photoproduct was 4-(1,3-benzoxazol-2-yl)benzoic acid and that
it arranged as dimers in the solid state. The secondary process
kinetics was found to be concentration-dependent and seems to
(3) Bischoff, P.; Hutter, C.; Puebla, C. In Ciba Specialty Chemicals
Holding Inc.: Basel, Switzerland, 2001; Vol. EP 1294846.
(4) Bur, A. J.; Roth, S. C. Polym. Eng. Sci. 2004, 44, 898−908.
(5) Pucci, A.; Bertoldo, M.; Bronco, S. Macromol. Rapid Commun.
2005, 26, 1043−1048.
(6) Pucci, A.; Ruggeri, G.; Bronco, S.; Bertoldo, M.; Cappelli, C.;
Ciardelli, F. Prog. Org. Coat. 2007, 58, 105−116.
(7) Sing, C. E.; Kunzelman, J.; Weder, C. J. Mater. Chem. 2009, 19,
104−110.
(8) Pucci, A.; Cappelli, C.; Bronco, S.; Ruggeri, G. J. Phys. Chem. B
2006, 110, 3127−3134.
(9) Jervis, D. A. Plast. Addit. Compd. 2003, 5, 42−46.
(10) Seydack, M.; Bendig, J. J. Fluoresc. 2000, 10, 291−294.
(11) Saltiel, J.; Megarity, E. D. J. Am. Chem. Soc. 1972, 94, 2742−
2749.
(12) Saltiel, J.; Marinari, A.; Chang, D. W. L.; Mitchener, J. C.;
Megarity, E. D. J. Am. Chem. Soc. 1979, 101, 2982−2996.
(13) Saltiel, J.; Sun, Y. P. J. Phys. Chem. 1989, 93, 6246−6250.
(14) Arai, T.; Tokumaru, K. Chem. Rev. 1993, 93, 23−39.
(15) Meier, H. Angew. Chem., Int. Ed. 1992, 31, 1399−1420.
(16) Whitten, D. G. Acc. Chem. Res. 1993, 26, 502−509.
(17) Johnson, A. R.; Lee, S.-J.; Klein, J.; Kanicki, J. Rev. Sci. Instrum.
2007, 78, 096101−096103.
(18) Porres
C.; Beeby, A. J. Fluores. 2006, 16, 267−273.
̀
̊
, L.; Holland, A.; Palsson, L.-O.; Monkman, A. P.; Kemp,
843
dx.doi.org/10.1021/jp312598c | J. Phys. Chem. A 2013, 117, 836−844

The Journal of Physical Chemistry A
Article
(19) Fourati, M. A.; Maris, T.; Bazuin, C. G.; Prud’homme, R. E. Acta
Crystallogr. 2010, C66, o11−o14.
(20) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(21) van der Sluis, P.; Spek, A. L. Acta Crystallogr. 1990, A46, 194−
201.
(22) Spek, A. L. PLATON, A Multipurpose Crystallographic Tool;
Utrecht University: Utrecht, The Netherlands, 2010.
(23) Pucci, A.; Bertoldo, M.; Bronco, S. Macromol. Rapid Commun.
2005, 26, 1043−1048.
(24) Karstens, T.; Kobs, K. J. Phys. Chem. 1980, 84, 1871−1872.
(25) Bolduc, A.; Dufresne, S.; Hanan, G. S.; Skene, W. G. Can. J.
Chem. 2010, 88, 236−246.
(26) Birch, D. J. S.; Birks, J. B. Chem. Phys. Lett. 1976, 38, 432−436.
(27) Carmichael, I.; Hug, G. L. J. Phys. Chem. Ref. Data 1986, 15, 1−
250.
(28) Scaiano, J. C. CRC Handbook of Organic Photochemistry; CRC
Press: Boca Raton, FL, 1989.
(29) Introduction to NMR spectroscopy, 2nd ed.; Abraham, R. J.,
Fisher, J., Loftus, P., Eds.; Wiley: Chichester, U.K., 1988.
(30) Lambert, J. B.; Shurvell, H. F.; Lightner, D. A.; Cooks, R. C.
Organic Structural Spectroscopy; Prentice Hall: Upper Saddle River, NJ,
1998.
(31) Seiber, R. P.; Needles, H. L. Textile Res. J. 1972, 42, 261−262.
(32) Murthy, R. S.; Bio, M.; You, Y. Tetrahedron Lett. 2009, 50,
1041−1044.
(33) Baucherel, X.; Uziel, J.; Juge,
4510.
́
S. J. Org. Chem. 2001, 66, 4504−
(34) Yin, P.; Wu, P.; Xiao, Z.; Li, D.; Bitterlich, E.; Zhang, J.; Cheng,
P.; Vezenov, D. V.; Liu, T.; Wei, Y. Angew. Chem., Int. Ed. 2011, 50,
2521−2525.
(35) Floquet, S.; Lemonnier, J. F.; Kachmar, A.; Rohmer, M. M.;
Benard, M.; Marrot, J.; Terazzi, E.; Piguet, C.; Cadot, E. Dalton trans.
2007, 3043−3054.
(36) Parac-Vogt, T. N.; Van Lokeren, L.; Cartuyvels, E.; Absillis, G.;
Willem, R. Chem. Commun. 2008, 2774−2776.
(37) Haupt, E. T. K.; Schaffer, C.; Bogge, H.; Merca, A.; Weinstock,
I. A.; Rehder, D.; Muller, A. Angew. Chem., Int. Ed. 2009, 48, 8051−
8056.
(38) Groves, P.; Rasmussen, M. O.; Molero, M. D.; Samain, E.;
Canada, F. J.; Driguez, H.; Jimenez-Barbero, J. Glycobiology 2004, 14,
451−456.
(39) Bhalerao, U. T.; Sridhar, M. Tetrahedron Lett. 1993, 34, 4341−
4342.
(40) Hirashima, S.-i.; Kudo, Y.; Nobuta, T.; Tada, N.; Itoh, A.
Tetrahedron Lett. 2009, 50, 4328−4330.
(41) Bernstein, J.; Davis, R. E.; Shimoni, L.; Chang, N. L. Angew.
Chem., Int. Ed. 1995, 34, 1555−1573.
(42) Coluccini, C.; Sharma, A. K.; Caricato, M.; Sironi, A.; Cariati, E.;
Righetto, S.; Tordin, E.; Botta, C.; Forni, A.; Pasini, D. Phys. Chem.
Chem. Phys. 2013, 15, 1666−1674.
(43) Cariati, E.; Lanzeni, V.; Tordin, E.; Ugo, R.; Botta, C.;
Giacometti Schieroni, A.; Sironi, A.; Pasini, D. Phys. Chem. Chem. Phys.
2011, 13, 18005−18014.
(44) Pommerehne, J.; Vestweber, H.; Guss, W.; Mahrt, R. F.; Bassler,
̈
H.; Porsch, M.; Daub, J. Adv. Mater. 1995, 7, 551−554.
(45) Koepp, H. M.; Wendt, H.; Stkehlow, H. Z. Elektrochem. 1960,
64, 483−491.
(46) Sun, Q.; Wang, H.; Yang, C.; Li, Y. J. Mater. Chem. 2003, 13,
800−806.
(47) Lu, Y.; Xiao, Z.; Yuan, Y.; Wu, H.; An, Z.; Hou, Y.; Gao, C.;
Huang, J. J. Mater. Chem. C 2013, DOI: 10.1039/C1032TC00327A.
844
dx.doi.org/10.1021/jp312598c | J. Phys. Chem. A 2013, 117, 836−844