Photochromism of a spiropyran and a diarylethene in bile salt aggregates in aqueous solution

Article abstract of DOI:10.1021/la503164e

Bile salt aggregates incorporate aqueous-insoluble photochromic compounds. The photochromism of a spiropyran (1, 1′,3′,3′-trimethyl-6-nitrospiro[2H-1]-benzopyran-2,2′-indoline) and a diarylethene derivative (2, 1,2-bis(2,4-dimethyl-5-phenyl-3-thienyl)-3,3,4,4,5,5-hexafluoro-1-cyclopentene) was quantified in different bile salt aggregates. These aggregates act as efficient hosts to solubilize aqueous insoluble photochromic compounds where either both isomers are nonpolar, for example, 2, or compounds where one isomer is hydrophobic and the other is more polar, for example, 1. Methodology was developed to determine molar absorptivity coefficients for solutions containing both isomers and to determine the photoconversion quantum yields under continuous irradiation. The methods were validated by determining parameters in homogeneous solution, which were the same as previously reported. In the case of the colored isomer of 1, the molar extinction coefficient in ethanol at 537 nm ((3.68 ± 0.03) × 10⁴cm^-1M^-1) was determined with higher precision. The quantum yields for the photoconversion between the isomers of 2 were shown to be the same in cyclohexane and in the aggregates of sodium cholate (NaCh), deoxycholate (NaDC), and taurocholate (NaTC), showing that bile salt aggregates are not sufficiently rigid to affect the equilibrium between the two possible conformers of the colorless form. In contrast, for 1 the quantum yields for the conversion from the colorless to the colored isomer were higher in bile salts than in ethanol, and the quantum yield was highest in the more hydrophobic aggregates of NaDC, followed by NaCh and then NaTC. The structure of the bile salt had no effect on the quantum yield for the conversion of the colored to the colorless isomer of 1, but these values were higher than in ethanol. For all three bile salts, the absorption maximum for the colored form of 1 suggested that this isomer was located in an environment that is more polar than ethanol.

Full text of DOI:10.1021/la503164e

Article
pubs.acs.org/Langmuir
Photochromism of a Spiropyran and a Diarylethene in Bile Salt
Aggregates in Aqueous Solution
Cerize S. Santos,† Allyson C. Miller,† Tamara C. S. Pace,† Kentaro Morimitsu,*,‡ and Cornelia Bohne*,†
†Department of Chemistry, University of Victoria, P.O. Box 3065, Victoria, BC V8W 3V6, Canada
‡Xerox Research Centre of Canada, 2660 Speakman Drive, Mississauga, ON L5K 2L1, Canada
S
* Supporting Information
ABSTRACT: Bile salt aggregates incorporate aqueous-insolu-
ble photochromic compounds. The photochromism of a
spiropyran (1, 1′,3′,3′-trimethyl-6-nitrospiro[2H-1]-benzopyr-
an-2,2′-indoline) and a diarylethene derivative (2, 1,2-bis(2,4-
dimethyl-5-phenyl-3-thienyl)-3,3,4,4,5,5-hexafluoro-1-cyclopen-
tene) was quantified in different bile salt aggregates. These
aggregates act as efficient hosts to solubilize aqueous insoluble
photochromic compounds where either both isomers are
nonpolar, for example, 2, or compounds where one isomer is
hydrophobic and the other is more polar, for example, 1.
Methodology was developed to determine molar absorptivity coefficients for solutions containing both isomers and to determine
the photoconversion quantum yields under continuous irradiation. The methods were validated by determining parameters in
homogeneous solution, which were the same as previously reported. In the case of the colored isomer of 1, the molar extinction
coefficient in ethanol at 537 nm ((3.68 0.03) × 104 cm−1 M−1) was determined with higher precision. The quantum yields for
the photoconversion between the isomers of 2 were shown to be the same in cyclohexane and in the aggregates of sodium
cholate (NaCh), deoxycholate (NaDC), and taurocholate (NaTC), showing that bile salt aggregates are not sufficiently rigid to
affect the equilibrium between the two possible conformers of the colorless form. In contrast, for 1 the quantum yields for the
conversion from the colorless to the colored isomer were higher in bile salts than in ethanol, and the quantum yield was highest
in the more hydrophobic aggregates of NaDC, followed by NaCh and then NaTC. The structure of the bile salt had no effect on
the quantum yield for the conversion of the colored to the colorless isomer of 1, but these values were higher than in ethanol. For
all three bile salts, the absorption maximum for the colored form of 1 suggested that this isomer was located in an environment
that is more polar than ethanol.
colorless open (2a) and colored closed (2b) forms is only
induced photochemically.1,3
INTRODUCTION
■
The changes in the optical and other physicochemical
properties of photochromic molecules induced by the reversible
switching between isomeric forms upon irradiation have been
employed in a number of applications, including sensing,
imaging, data and signal processing, photocontrol of biological
systems, modulation of energy and electron transfer, and
incorporation into organic electronic materials.1−8 Develop-
ment of many applications, especially those involving biological
systems, requires use of aqueous media, and strategies to readily
solubilize photochromic molecules in water are therefore
required.
Spiropyrans and diarylethenes are two commonly studied
families of photochromic molecules. Generally, a spiropyran (1,
1′,3′,3′-trimethyl-6-nitrospiro[2H-1]-benzopyran-2,2′-indoline,
Scheme 1) can be reversibly switched between the colorless
spiro form (1a) and the colored merocyanine form (1b) either
thermally or photochemically.2 In contrast, a typical diary-
lethene derivative (2, 1,2-bis(2,4-dimethyl-5-phenyl-3-thienyl)-
3,3,4,4,5,5-hexafluoro-1-cyclopentene, Scheme 1) exhibits no
thermal isomerization, and reversible switching between the
The spiropyran and diarylethene backbones are not intrinsi-
cally soluble in aqueous solution. Limited soluble spiropyrans
have been synthesized either through peptide9 or aminoalkyl10
substitution on the indoline nitrogen or through replacement of
the nitro group with a carboxylate9,11 or a pyridinium moiety.12
Aqueous solubility for diarylethene derivatives has been
achieved through introduction of cationic ammonium,13
pyridinium groups,14 or bulky water-soluble moieties.15−17
A universal approach that easily allows for solubilization of a
broad selection of photochromic molecules, without the need
for synthetic modification, is the use of water-soluble
supramolecular host systems to bind hydrophobic photo-
chromic guests, affording host−guest complexes in aqueous
solution. Most work in this area has focused on inclusion
complexes where the photochromic molecule, or an appendant
moiety, is included in a cyclodextrin,13,18−22 cavitand,23 or
Received: August 7, 2014
Revised: September 4, 2014
Published: September 9, 2014
© 2014 American Chemical Society
11319
dx.doi.org/10.1021/la503164e | Langmuir 2014, 30, 11319−11328

Langmuir
Article
cosolvent.38 Thus, bile salt aggregates are good candidates for
the solubilization of photochromic molecules with varying
polarities, and the inherent adaptability of these host systems
may be ideal for controlling properties of the photochromic
system, such as position of the photostationary state, or rates of
thermal or photoconversion reactions between the isomers.
We have previously shown in a preliminary communication
that NaCh is an excellent host system for the solubilization of
both 1 and 2 and that the efficiencies of photochromic
conversion were only slightly decreased in comparison to those
in ethanol.39 In addition, the rate constant for the thermal re-
equilibration of 1 was affected by increasing the concentration
of NaCl, in line with expected changes in the polarity of the
guest’s binding environment. In this work, the photochromism
of 1 and 2 in bile salt aggregates was comprehensively
quantified, and the effect of bile salt structure on the
photochromism observed in 1 and 2 was established.
Scheme 1. Photochromic Reactions of 1 and 2 and
Structures of NaCh, NaDC, and NaTC
EXPERIMENTAL METHODS
■
The purity of the materials and their sources, the equipment used,
sample preparation methods, and the continuous irradiation setup
(Supporting Information Figure S1) are described in the Supporting
Information. Specific methods are described below. The complexes
between 1 or 2 and NaCh were stable for at least 24 h for
concentrations of 1 below 120 μM and 2 below 300 μM, when
monitored by UV−vis spectroscopy.39 The current studies were
performed with a maximum concentration of 50 μM.
HPLC Methods. For experiments with 1, solutions were kept in an
ice bath for at least 15 min prior to irradiation, exposed to ambient
light to obtain a solution containing only 1a, then irradiated at 338 nm
for 2 min. Absorption spectra were obtained before and after
irradiation, then the sample was transferred to an HPLC vial and
kept in the dark in an ice bath, and an HPLC chromatogram was
obtained as soon as possible. For the HPLC chromatogram of 1, the
absorption was detected at 318 and 517 nm, and a mixture of
methanol and water (80:20 v/v) was used as the eluent with a flow
rate of 1.2 mL/min. Solutions of 2 were exposed to ambient light to
obtain solutions containing only 2a, then irradiated at 285 nm
(hexane) or 287 nm (bile salts) for 2 min. Absorption spectra were
obtained before and after irradiation, then the sample was transferred
to an HPLC vial and kept in the dark, and an HPLC chromatogram
was obtained. For the HPLC chromatogram of 2, the absorption was
detected at 269 and 562 nm, and a mixture of methanol and
acetonitrile (70:30 v/v) was used as the eluent with a flow rate of 1.5
mL/min.
Quantum Yield Determination. Samples of 1 or 2 were
irradiated continuously at the isosbestic point (for 1, 298 nm in
ethanol, 299 nm in NaCh and NaTC, and 302 nm in NaDC; for 2, 287
nm in cyclohexane and 292 nm in NaCh and NaDC). Absorption was
collected simultaneously at several wavelengths, including the
absorption maxima for 1b or 2b in the visible region. During
irradiation the samples were stirred continuously, and the temperature
of the samples was kept constant at 15 °C to minimize evaporation of
organic solvent and minimize the thermal reaction for 1b. Experiments
in NaDC were carried out at 25 °C to avoid gel formation by this bile
salt. The obtained kinetic curves were numerically fit using the
software Scientist 3.0 (Micromath). The potassium ferrioxalate
actinometer was prepared40 and was used to determine the light flux
incident on the sample. The variation in the light flux was typically
smaller than 5% when measured before and after collecting a kinetic
trace, and the average value was used for the fits.
curcubituril.24,25 As these hosts have a defined shape and size,
the cavity typically incorporates a single guest molecule through
specific interactions; however, these host−guest systems tend
to have limited solubility in water. Conversely, the structure of
a micelle is dynamic, and multiple guests of varying shapes and
sizes can easily be incorporated nonspecifically in the
hydrophobic interior of micelles. Solubilization of 126 and
other spiropyran derivatives27 was achieved in cationic or
nonionic micellar solutions. The small increases observed in
both thermo- and photocolorability in the micelles followed the
trend established in homogeneous solvents due to polarity
changes in the vicinity of the photochromic molecule.27
Bile salt aggregates combine the solubilization properties of
micelles with specific host−guest interactions. The hydro-
phobic convex surfaces of the bile salt molecules interact to
form small primary aggregates, which are incorporated into
larger secondary aggregates at higher bile salt concentrations.
Therefore, bile salt aggregates continuously increase in size with
the increase in the concentration of bile salt monomer or the
increase in ionic strength.28−30 Primary aggregates provide a
restricted environment for bound guests, and the properties of
these guest−aggregate systems are somewhat dependent on the
size and shape of the guest.31−33 Secondary aggregates provide
a more micelle-like environment for bound guests, from which
the guest binding dynamics is faster than when bound to
primary aggregates.34 Bile salt aggregates can be viewed as
adaptable hosts because the guest−aggregate structure depends
on the structure of the guest and the concentrations of the bile
salt and other salts. Therefore, the size of these structures is yet
unknown. Modification to the structure of the bile salt (sodium
cholate (NaCh), sodium deoxycholate (NaDC), or sodium
taurocholate (NaTC), Scheme 1), either in the headgroup or in
the number of hydroxyl groups on the concave face, causes
changes in the structure of the guest−aggregate system.32,35
The properties of these systems can also be easily modulated by
binding cations,36 changing the ionic strength,37 or adding a
Thermal Relaxation Kinetics. Bile salt solutions containing 1
were irradiated at 338 nm for 10 min at the temperature of the
subsequent kinetic experiment to obtain a solution containing a
significant amount of 1b. Immediately following irradiation, the
solutions were placed in the spectrophotometer, and the absorption at
517 nm was collected at 1 min intervals until the absorbance did not
decrease significantly. Because of experimental limitations, no kinetics
11320
dx.doi.org/10.1021/la503164e | Langmuir 2014, 30, 11319−11328

Langmuir
Article
were monitored for more than 30 h. The absorbance of a blank
reference solution, containing all components except 1, was obtained
concurrently in the same measurement. The data from the blank were
subtracted from the experimental data to obtain the kinetic curves,
which were fitted to a sum of exponential functions.
absorptivities for 2b were calculated by dividing the spectrum
for 2b by the concentration of 2b (Figure 1, Table 1).
RESULTS
■
When using absorption techniques to monitor a photochromic
reaction, knowledge of the absorption spectra and molar
absorptivity coefficients (ε) for both isomers is essential.
Solutions of 1 and 2 can be converted to 100% 1a and 2a by
irradiation with visible light. The isomers 1b and 2b are
similarly obtained by irradiation with UV light; however, a
photostationary state is reached where the final concentrations
of a and b are unknown.
Methods to determine the values of ε for thermally stable
isomers, such as 2b, have usually involved the isolation of the
isomer by chromatographic separation, concentration of the
solution, and measurement of the absorption spectra.41,42
These methods require appreciable amounts of isomer to be
isolated and are not suitable for the determination of ε values
for thermally unstable isomers, such as 1b, where after isolation
the thermal back reaction will lead to formation of 1a. These
methods are also not generally applicable to the determination
of ε values in host−guest systems that exist in equilibrium in
solution. Determination of molar absorptivities from absorption
spectra of mixtures at the photostationary state is possible by
measuring the spectra at different irradiation wavelengths.43
This method assumes that the quantum yields for photo-
conversion are the same at different irradiation wavelengths,
and it could not be applied in the current study because for 2
and related derivatives the quantum yields show a wavelength
dependence.44,45 For mixtures of both photochromic isomers,
an overall absorption spectrum is obtained. If the spectrum of
one isomer and the ratio of concentrations are known, then the
spectrum of the other isomer can be calculated. We describe
here an analytical HPLC method to readily carry out such a
determination in various solvents, including host−guest
solutions.
For solutions containing a mixture of 2a and 2b in 80 mM
NaCh/0.2 M NaCl, two peaks were observed in the HPLC
chromatogram at a detection wavelength of 269 nm
(Supporting Information Figure S2); the absorption spectrum
of the peak at 4.3 min was consistent with the spectrum of 2a,
whereas the absorption spectrum of the peak at 5.4 min showed
absorption in the visible region consistent with the absorption
of 2b. At a detection wavelength of 562 nm, only the 5.4 min
peak was observed, consistent with the assignment of this peak
to 2b. Other peaks observed at short retention times were also
observed in the chromatogram of the blank and are due to
impurities in the bile salt, which have been observed in previous
photophysical studies.37
Figure 1. Molar absorptivities of 2a (dashed lines) and 2b (solid lines)
in hexane (black), NaCh (80 mM, 0.2 M NaCl, blue), and NaDC (80
mM, 0.2 M NaCl, red).
Table 1. Molar Absorptivity (ε) of 1b and 2b at the
a b
,
Absorption Maximum in the Visible Region
ε/104 cm−1 M−1
medium
1b
2b
ethanol
hexane
3.68 0.03 (2)
c
c
1.11 0.03 (2)
1.09 0.04 (2)
1.1 0.1 (2)
d
80 mM NaCh/0.2 M NaCl
80 mM NaDC/0.2 M NaCl
80 mM NaTC/0.2 M NaCl
2.84 0.06 (2)
2.77 0.05 (2)
2.76 0.06 (2)
a
Values of ε determined at 537 nm in ethanol and 517 nm in bile salts
b
for 1b, and at 562 nm in hexane and 577 nm in bile salts for 2b. The
numbers in parentheses correspond to the number of independent
c
experiments. The errors are average deviations. Not determined.
d
Not determined because of solubility limitations as solutions in the
presence of 2 were turbid.
The same procedure was used to determine the ε values for
2b in NaDC and hexane (Figure 1, Table 1). The absorption
spectra of 2a were similar in different solvents, whereas a small
bathochromic shift was observed for 2b in the bile salt solutions
as compared to hexane. At 562 nm the molar absorptivity for
2b in hexane was determined to be (1.11 0.03) × 104 cm−1
M−1, which is in good agreement with the reported value of 1.1
× 104 cm−1 M−1.41 It should be noted that for solutions of 2 in
hexane, the peaks for both 2a and 2b in the HPLC
chromatogram exhibited a shoulder at longer times (Supporting
Information Figure S4). The main peak and the shoulder were
found to belong to the same compound based on the
absorption spectra (Supporting Information Figure S5), and
thus the areas of both the main peak and the shoulder were
used in the molar absorptivity calculations.
A similar strategy was used to determine ε values for 1b, with
the addition of keeping the solutions in an ice bath following
irradiation to minimize thermal reversion prior to measurement
of the HPLC chromatogram. For solutions containing a
mixture of 1a and 1b, two peaks were observed in the HPLC
chromatogram at a detection wavelength of 338 nm
(Supporting Information Figure S6); the absorption spectrum
of the broad peak at 3.1 min showed absorption in the visible
region, consistent with the absorption of 1b, whereas the
absorption spectrum of the sharp peak at 12.8 min was
Solutions of known concentrations of 2a in 80 mM NaCh/
0.2 M NaCl were used to construct an HPLC response curve
(Supporting Information Figure S3). A sample of 2 was
irradiated at 285 nm, and an absorption spectrum of the
resulting mixture of 2a and 2b was measured. The HPLC
chromatogram was obtained for this mixture, and the response
curve was used to determine the concentration of 2a. The
corresponding spectrum for 2a was calculated and was
subtracted from the spectrum for the mixture leading to the
spectrum of 2b. The concentration of 2b was known because
the total concentration of 2 was known, and the molar
11321
dx.doi.org/10.1021/la503164e | Langmuir 2014, 30, 11319−11328

Langmuir
Article
consistent with the spectrum of 1a. For the detection at 517 nm
(Supporting Information Figure S6), only the broad 3.1 min
peak was observed, consistent with the assignment of this peak
to 1b. The position (2.8−3.2 min) and shape of the peak for 1b
were found to vary in different experiments. It is possible that a
partial conversion of 1b to 1a inside the column leads to the
broadening of this peak. This is not problematic as all
quantitative information was obtained from the 1a peak; the
shape and position of the 1a peak did not vary, and the 1a and
1b peaks never overlapped.
Solutions of known concentrations of 1a in 80 mM NaTC/
0.2 M NaCl were used to construct an HPLC response curve
for 1a (Supporting Information Figure S7), and the ε values for
1b in 80 mM NaTC/0.2 M NaCl were calculated as described
above (Figure 2, Table 1). Molar absorptivities were also
Quantum yields for the interconversion of isomers a and b in
a photochromic system can be determined by following the
kinetics of the photochemical reaction under continuous
irradiation, where rate equations (eqs 1−4) are defined for
each photochemical (Φab, Φba) or thermal conversion (kab, kba).
v1 = Io(1 − 10−ε [a])Φ
a
(1)
(2)
ab
v2 = kab[a]
v3 = Io(1 − 10−ε [b])Φba
b
(3)
(4)
v4 = kba[b]
For 1, both thermal and photochemical processes occur, and
the differential equation that describes the change in the
concentration of 1a when the system is continuously irradiated
is given by the sum of the rate equations for the individual
processes (eq 5).
d[a]
dt
Φba + kba[b]
= −Io(1 − 10−ε [a])Φ − kab[a] + Io(1 − 10−ε [b]
)
a
b
ab
(5)
where Io is the light flux incident on the sample, [a] and [b] are
the concentrations of 1a and 1b, εa and εb are the molar
absorptivities of 1a and 1b at the wavelength of irradiation, Φab
and Φba are the quantum yields for conversion of 1a to 1b and
1b to 1a, respectively, and kab and kba are the rate constants for
the thermal conversion of 1a to 1b and 1b to 1a, respectively.
As 2 is thermally stable, the rate equations for thermal
conversion can be excluded, and the differential equation that
describes the change in the concentration of 2a when the
system is continuously irradiated can be simplified to eq 6.
Figure 2. Molar absorptivities of 1a (dashed lines) and 1b (solid lines)
in ethanol (black), NaCh (80 mM, 0.2 M NaCl, blue, overlapped with
green trace for 1a), NaDC (80 mM, 0.2 M NaCl, red), and NaTC (80
mM, 0.2 M NaCl, green, overlapped with red trace for 1b).
d[a]
= −Io(1 − 10−ε [a])Φ + Io(1 − 10−ε [b])Φba
a
b
ab
(6)
dt
The photochemical reactions can be monitored at a
wavelength in the visible region where only the b isomer
absorbs (eq 7).
determined for 1b in NaCh, NaDC, and ethanol. The
absorption spectra of 1a were similar in all solvents, whereas
the absorption spectra of 1b were very similar in bile salt
solutions, but a large bathochromic shift was observed in
ethanol. At 537 nm the molar absorptivity for 1b in ethanol was
Abvis = εbvis[b]
(7)
where Avbis is the absorbance of isomer b at time t and εvbis is the
molar absorptivity coefficient of isomer b at the monitoring
wavelength. The concentration of b at time t, [b], can be
expressed as the difference between the total concentration of
photochromic compound, [a]o, which is a known and constant
quantity, and the concentration of a at time t, [a] (eq 8).
determined to be (3.68
0.03) × 104 cm−1 M−1, which is
similar to the estimated value reported of 3.5 × 104 cm−1
M−1.46,47 This literature value is an estimate based on the mean
value of ε assuming that a solution of 1 contains 51−100% 1b
at the photostationary state.47 This value, determined for
ethanol, has been assumed to be valid in polar solvents in
general.46,48−51
[b] = [a]o − [a]
(8)
To determine the extent of thermal reaction during the time
of the HPLC experiment, a solution of 1 in 80 mM NaCh/0.2
M NaCl was irradiated and then kept covered in an ice bath.
The absorption spectrum was collected immediately after
irradiation, and then the HPLC chromatogram was obtained. A
second absorption spectrum and HPLC chromatogram were
then obtained for the same solution. The two absorption
spectra obtained immediately and 13 min after irradiation were
very similar (Supporting Information Figure S8), and the peak
areas for 1a, obtained from the HPLC chromatograms 6 and 22
min after irradiation, were not significantly different (114.87
and 114.97 mAU, Supporting Information Figure S9). From
these results, we can assume that when the solution is kept in
an ice bath following irradiation, no significant thermal
conversion occurs before the HPLC chromatogram is obtained.
The strategy employed in this investigation to determine the
quantum yields used a numerical analysis of the above kinetic
equations in Scientist 3.0 (see models employed in the
Supporting Information). A custom-built spectrometer was
used to continuously measure the sample absorption during
constant irradiation at a constant temperature. During the
irradiation of 2, the absorbance remained constant at the
isosbestic point between 2a and 2b (287 nm) and at a
wavelength where neither isomer absorbed (800 nm), and the
formation of 2b was observed at 562 nm (Figure 3). With this
setup, noise with a periodic oscillation was observed at all
detection wavelengths. The example shown in Figure 3 shows a
worst-case scenario for these signal fluctuations. To ensure that
this noise would not affect the fitting of the kinetic traces, the
11322
dx.doi.org/10.1021/la503164e | Langmuir 2014, 30, 11319−11328

Langmuir
Article
isosbestic point were found to be, within error, the same in all
of the conditions studied.
Table 2. Quantum Yields for the Photoisomerization of 2
a
Irradiated at the Isosbestic Point
medium
cyclohexane (2)
Φab
Φba
0.39 0.02
0.39 0.02
0.36 0.03
0.36 0.03
0.07 0.02
0.06 0.02
0.07 0.02
0.05 0.02
80 mM NaCh/0.2 M NaCl (2)
80 mM NaCh/1 M NaCl (2)
80 mM NaDC/0.2 M NaCl (2)
a
The numbers in parentheses correspond to the number of
independent experiments, ε287 = (2.02
0.05) × 104 cm−1 M−1
determined in hexane was used for the experiments in cyclohexane,
ε292 = (1.6 0.1) × 104 cm−1 M−1 in NaCh, and ε292 = (1.9 0.2) ×
104 cm−1 M−1 in NaDC. T = 15 °C, except for NaDC where T = 25
°C.
The same method was used for the determination of Φab and
Φba for the isomerization of 1, with the additional consideration
of the thermal reactions of 1 (Figure 4). The rates of reaction
Figure 3. Top: Absorbance during irradiation at 287 nm of 2 in
cyclohexane at 15 °C at 287 nm (a, isosbestic point; the dashed red
line is the average value), 562 nm (b, absorption maximum of 2b), and
800 nm (c, baseline). Bottom: Absorbance at 577 nm for 2 in 80 mM
NaCh/0.2 M NaCl during irradiation at 292 nm at 15 °C.
Experimental data (d, black), numerical fit (d, red), and residuals
between the experimental data and the fit (e; the dashed red line
indicates zero).
kinetics were followed for at least one cycle of oscillation after
the photostationary state was reached. The resulting residuals
for fits of the kinetic traces displayed the same periodic
oscillation, indicating that the noise was not included in the fit
(see the Supporting Information for details).
Figure 4. Absorbance at 517 nm for 1 in 80 mM NaCh/0.2 M NaCl
during irradiation at 299 nm at 15 °C. Experimental data (a, black),
numerical fit (a, red), and residuals between the experimental data and
the fit (b, where the dashed red line indicates zero).
Solutions of 2 were irradiated at the isosbestic point, and
absorption was monitored at the maximum of the absorption
peak in the visible region for 2b. For the numerical fit of this
kinetic data, the molar absorptivity coefficient, εa, was
determined from the absorption spectra of a solution of 2a
only. As the irradiation was performed at the isosbestic point,
the value of εb was equal to the value of εa. As described above,
the molar absorptivity coefficients of 2b at other wavelengths
are not trivial to determine, and initially the curves were fit
without fixing εvbis. This resulted in under-determination of the
parameters εvbis, Φab, and Φba. The same under-determination
was observed when the reaction was monitored at several
wavelengths and the curves were fit simultaneously. In
subsequent fits (Figure 3), εbvis was fixed to the value
determined as described above. For 2 in cyclohexane, the εbvis
values for hexane were used because the absorption of 2 at 287
nm in hexane and cyclohexane was the same within 1%. The
resulting quantum yields were determined with a small standard
deviation, and the same values were found irrespective of the
initial guess values. The parameter found to have the greatest
effect on the determination of the quantum yields was εvbis. The
fitting was therefore carried out with both the minimum and
the maximum values for εbvis, as defined by the experimental
error for this parameter, and the resulting quantum yield varied
up to 0.02. Therefore, an uncertainty of 0.02 was assumed
for all quantum yield values except where a larger error was
observed for independent experiments (Table 2). The values of
both Φab and Φba for isomerization of 2 by irradiation at the
for the thermal reactions of 1 were determined in independent
kinetic studies (see below), and these parameters were fixed
when fitting the kinetic curves for the photoconversion. The
resulting quantum yields were determined with a small standard
deviation, and the same values were found independent of the
initial guess values. The fitting was performed considering the
experimental error of εvbis, kab, and kba. The quantum yields from
these fits were assumed to have an error of 0.01 unless a
larger error was obtained for independent experiments (Table
3). The quantum yields for 1 were affected by the medium.
Both Φab and Φba were smaller in ethanol than in bile salts, and
the value of Φab was also seen to depend on the structure of the
bile salt.
The relaxation rate constant for the thermal equilibration
between 1a and 1b can be obtained by monitoring the
relaxation of the system after the formation of an excess of 1a
or 1b; excess 1a or 1b can be formed by irradiation with visible
or UV light, respectively. The kinetics monitored at the
absorption maxima for 1b show a fast growth followed by a
slow decay when 1a is in excess and fast decay followed by a
slow decay when 1b is in excess. The kinetic curves were fit to a
sum of two exponentials (eq 9), where A is the absorbance, ai
are pre-exponential factors, and ki are first-order rate constants.
11323
dx.doi.org/10.1021/la503164e | Langmuir 2014, 30, 11319−11328

Langmuir
Article
kab
kba
Table 3. Quantum Yields for the Photoisomerization of 1
Irradiated at the Isosbestic Point
[1b]
[1a]
K =
=
a
(11)
medium
Φab
Φba
The absorption of 1b at equilibrium in the absence of
decomposition is equal to the difference between the
absorption of the solution at time zero (At=0) and the pre-
exponential factor, a1 (eq 9, where At=0 = a1 + a2). From this
difference, the concentration of 1b at equilibrium is
determined. From the concentration of 1b and the total
concentration of 1, the concentration of 1a is obtained, and the
equilibrium constant can be calculated using eq 11. Equations
10 and 11 can be combined to determine kba (eq 12), and kab
can then be calculated using eq 10.
ethanol (2)
0.15 0.01
0.30 0.01
0.32 0.01
0.51 0.03
0.26 0.01
0.04 0.01
0.14 0.01
0.15 0.02
0.15 0.02
0.15 0.01
80 mM NaCh/0.2 M NaCl (2)
80 mM NaCh/1.0 M NaCl (2)
80 mM NaDC/0.2 M NaCl (2)
80 mM NaTC/0.2 M NaCl (2)
a
The numbers in parentheses correspond to the number of
independent experiments, ε298 = (8.57 0.07) × 103 cm−1 M−1 in
ethanol, ε299 = (7.3 0.3) × 103 cm−1 M−1 in NaCh, ε302 = (7.6
0.1) × 103 cm−1 M−1 in NaDC, and ε299 = (7.3 0.2) × 103 cm−1 M−1
in NaTC. T = 15 °C, except for NaDC where T = 25 °C.
k1
1 + K
A = a1 e−k t + a2 e−k t
1
2
kba
=
(9)
(12)
The fast component (k1) is attributed to the equilibration of
1, whereas the slow component (k2) is attributed to the
decomposition of 1 through hydrolysis, similar to that
previously observed for water-soluble spiropyrans.9 The
decomposition rate constants reported for a water-soluble
spiropyran ((2−4) × 10−3 min−1 at pH 7−8)9 are significantly
higher than those previously reported for 1 in NaCh aggregates
((3−5) × 10−5 min−1),39 suggesting that the aggregates afford 1
protection from hydrolysis.
We have previously reported the relaxation rate constants for
the thermal equilibration between 1a and 1b in aqueous
solution containing 80 mM NaCh and 0.2 or 1 M NaCl.39 In
this previous work, the same relaxation rate constants were
obtained from the kinetics for the conversion of excess 1a to 1b
or the conversion of excess 1b to 1a, as is expected for
relaxation kinetics. All data reported in this current study were
measured only for the relaxation of 1b to 1a.
A fit to the sum of two exponentials was required to get fits
with low and random residuals, and reproducible values for k1.
However, as the values obtained for k2 ((1−8) × 10−5 min−1 at
25 °C) were 2 orders of magnitude smaller than those for k1,
there was not sufficient data on the time scales monitored for
the determination of precise values of k2. No discernible trend
in k2 was obtainable, and the individual values for k2 are
therefore not presented.
Relaxation rate constants, k1, were obtained for 1 in various
bile salt solutions (Supporting Information Figure S10). The
value of k1 varied for the different bile salts (NaDC > NaCh >
NaTC), whereas the opposite trend was observed for the
equilibrium constants (Table 4). The k1 value in NaCh and 0.2
M NaCl was the same as previously reported, whereas this
value at the higher NaCl concentration of 1.0 M was higher
than the previous value of (8.74 0.06) × 10−3 min−1.39 This
discrepancy could be due to the method used for sample
preparation; in the previous work, 1 and 2 were injected after
the temperature annealing cycle for the bile salts, whereas in the
current work, the samples with the guests were heated, which
ensured equilibration of the host−guest system.
Rate constants calculated for the thermal conversion between
1a and 1b (kab) and 1b and 1a (kba) were used in the
determination of the quantum yields for photoconversion (see
above). It is important to note that because the reaction is
unimolecular, the initial absorbance at time zero is not
significant, and the initial absorbance value is determined
solely by the duration of the irradiation to form an excess of 1b.
DISCUSSION
■
Supramolecular systems frequently impart properties to the
system that are not achievable by the individual, isolated
building blocks. In this study, bile salt aggregates were used as
hosts in aqueous solution for insoluble photochromic
compounds, and the desired property for the supramolecular
system was to achieve photochromism in a solvent incompat-
ible with the isolated photochromic molecules. Supramolecular
systems are in equilibrium, and therefore characterization of the
parameters that describe the photochromism of these systems
cannot be studied by isolating individual molecules. This
situation led to the requirement for the development of suitable
methodology, which will be discussed first.
The relaxation rate constant, k1, corresponds to the sum of
the rate constants for the thermal conversion of 1a to 1b (kab)
and 1b to 1a (kba) (eq 10).
k1 = kab + kba
(10)
The equilibrium constant, K, for the thermal equilibrium
between 1a and 1b can be described either as a ratio of rate
constants or as a ratio of the concentrations of 1a and 1b (eq
11).
Table 4. Relaxation Rate Constant, k1, Equilibrium Constants, K, and Rate Constants, kab and kba, for the Thermal Equilibrium
a
between 1a and 1b in Aqueous Solutions Containing Various Bile Salts (80 mM) and 0.2 or 1.0 M NaCl at 25 °C
bile salt
[NaCl]/M
k1/10−3 min−1
K/10−2
kab/10−3 min−1
kba/10−3 min−1
NaCh (3)
NaCh (3)
NaDC (2)
NaTC (2)
0.2
1.0
0.2
0.2
5.6 0.1
11.2 0.1
6.26 0.06
4.34 0.01
10
1
0.51 0.06
0.24 0.04
0.177 0.005
0.48 0.02
5.1 0.2
10.9 0.1
6.08 0.06
3.85 0.01
2.2 0.4
2.9 0.1
12.5 0.2
a
The numbers in parentheses correspond to the number of independent experiments. The errors are standard deviations when the number of
independent experiments is equal to or larger than three or average deviations when only two independent experiments were performed.
11324
dx.doi.org/10.1021/la503164e | Langmuir 2014, 30, 11319−11328

Langmuir
Article
Knowledge of the concentrations of isomers of a photo-
chromic system is required to quantify the quantum yields for
the photoconversion reactions. Isolation of each isomer is not a
viable method, especially in the case where re-equilibration can
occur in the absence of light. In view of these restrictions, we
developed an HPLC method where a solution that contained
both isomers was analyzed. This method is applicable to any
solution and can be used for the characterization of
photochromic compounds in general, as the physical isolation
and manipulation of the isomers is not required. In particular,
the method provided a molar absorptivity value for 1b with a
much higher precision ((3.68 0.03) × 104 cm−1 M−1 at 537
nm in ethanol) than the standard estimated value (3.5 × 104
cm−1 M−1)46,47 used in many studies.
The ε values for 1b and 2b were determined from the overall
absorption spectra of a solution containing both photochromic
isomers, using the known spectrum of one isomer and the
concentration of that isomer in the mixture obtained from
HPLC measurements. This method allows for measurement of
ε values for photochromic compounds in complex solutions,
including host−guest solutions, and for compounds where
thermal stability is an issue. The determination of ε values can
be carried out for thermally unstable isomers, such as 1b, as
long as no significant thermal conversion occurs prior to the
HPLC measurement. As the calculation is based on the peak
area for the stable isomer, thermal conversion of the thermally
unstable isomer on the HPLC column, which may lead to
significant peak broadening, does not affect the determination
of ε values, as long as there is no overlap between the peaks for
the two isomers.
Quantum yields for the photoconversion between photo-
chromic isomers were determined using numerical analysis of
kinetic data obtained during continuous irradiation of the
solution at a single wavelength at a constant temperature. The
simplification of the relevant rate equations by introduction of a
photokinetic factor has frequently been used in the
determination of quantum yields in photochromic systems.52,53
The photokinetic factor should only be used when the
absorption at the irradiation wavelength is less than 0.1, as
above this value systematic errors are introduced. Numerical
analysis avoids this restriction, and higher concentrations can
be employed to obtain better signal-to-noise ratios. Numerical
analysis of kinetic data obtained by irradiation at two
wavelengths has been used to simultaneously recover Φab,
Φba, and ε values in various systems,54,55 but requires the
assumption that the quantum yields are wavelength independ-
ent. However, the quantum yields of some of 2 and some of its
derivatives are known to be wavelength dependent.44,45 The
numerical method described herein allowed recovery of Φab
and Φba from kinetic data obtained by irradiation at a single
wavelength through the use of ε values determined
independently. This method used the determination of the
total photon flux with the well-established ferrioxalate actino-
meter for the determination of quantum yields for photo-
chromic compounds, and not relative quantum yields
determined in comparison to that known for a photochromic
reaction. Measuring relative rates requires an assumption of the
same reaction mechanism; this assumption is often not
appropriate in supramolecular systems,56 and especially not in
bile salt solutions, where the structure of the guest is known to
affect the structure of the bile salt aggregate itself.31−33
(0.17, 25 °C), which was determined from transient absorption
measurements.48,57 Values of Φba are not reported using this
previous method due to experimental limitations; no evidence
was observed for the 1b to 1a reaction on the time scale of the
fast transient absorption measurements (<3 ns) in nonpolar
solvents, such as toluene,58 suggesting that Φba is lower than
Φab. The Φab value for 2 in cyclohexane at 15 °C determined by
our method (0.39 0.02) is similar to the value reported in the
literature (0.46, 22 °C, hexane).41 The Φba value for 2 was
reported to vary with temperature from 0.015 at 22 °C in
hexane to 0.037 at 80 °C,41 and to vary with the excitation
wavelength from 0.015 for irradiation of 2b at 620 nm to 0.027
at 480 nm.45 This wavelength dependence was correlated to the
presence of an energy barrier in the excited-state surface.45 This
effect is consistent with the higher value observed for our
determination of Φba, because an excitation wavelength of 287
nm was employed. The good agreement with literature values
for the Φab values of 1 and 2 in homogeneous solution was
used as the validation of our method, and the discrepancy for
Φba values was not further investigated, as the focus of the
current work was to determine if these quantum yields changed
when the compounds were incorporated into bile salt
aggregates.
Bile salt aggregates are versatile host systems that can
incorporate guests with different structures.31−33 A preliminary
qualitative study showed that bile salt aggregates are suitable
hosts for 1 and 2. Not all bile salt structures were able to
solubilize 2 in aqueous solution; clear solutions could not be
obtained for 2 in NaTC. NaCh and NaDC can both solubilize 2
in aqueous solution, but the different bile salt structures had no
effect on photochromism of 2. The same molar absorptivities
were recovered, within error, for 2b in hexane or in the bile salt
solutions, and the same quantum yields for the photochemical
ring-opening and ring-closing reactions were obtained, within
error, in cyclohexane or bile salt solutions. The open isomer of
a diarylethene, such as 2a, has two conformations, and only the
antiparallel conformation has the correct stereochemistry for
ring closure to occur in the excited state.1 The lifetime of the
excited state of this isomer is too short for interconversion
between the conformers to occur in the excited state, and in
homogeneous solution the maximum quantum yield achievable
has been proposed to be 0.5.1 In constrained environments, the
quantum yield for the formation of the closed isomer can be
affected; an increase relative to homogeneous solution was
observed in Φab for diarylethene derivatives in β-13,21 and γ-
cyclodextrin,21 attributed to an enrichment of the antiparallel
conformation when complexed with the cyclodextrins. The
same value of Φab for 2 in hexane and bile aggregates suggests
that the aggregates are not sufficiently rigid to change the
equilibrium between the two conformers of 2a. This
interpretation is in line with the adaptability of the binding
sites in the bile salt aggregates to the shape and size of the
guest.31−33
All three bile salts investigated were easily able to solubilize 1
in aqueous solution. The position of the absorption maxima for
1b was dependent on the polarity of the solvent (538 nm in
ethanol to 603 nm in toluene),11 and thus 1b in bile salt
solutions (absorption maxima of 517 nm) experienced an
environment somewhat more polar than that of ethanol,
although very little variation was observed between bile salts.
The values of ε determined for 1b in bile salt solutions were
significantly lower than that in ethanol, and revision is needed
The Φab value for 1 in ethanol at 15 °C determined by our
method (0.15
0.01) is consistent with the reported value
11325
dx.doi.org/10.1021/la503164e | Langmuir 2014, 30, 11319−11328

Langmuir
Article
of the assumption commonly made in the literature that the ε
value for 1b is the same in all polar solvents.46,48−51
photochromic parameters. Molar absorptivities were obtained
from HPLC experiments, and, in the case of 1b, a high
Relaxation rate constants for the thermal isomerization of 1
are known to decrease with increasing solvent polarity.46,48,50,59
The k1 values obtained for 1 in bile salt solutions varied in the
different bile salts (NaDC > NaCh > NaTC), but all were lower
than k1 reported in ethanol (2.3 × 10−2 min−1),46 consistent
with 1 in bile salts being located in an environment somewhat
more polar than ethanol. The opposite trend was observed for
the equilibrium constants in these three bile salts (NaTC >
NaCh > NaDC). In homogeneous solution, the equilibrium
between 1a and 1b shifts toward the more polar 1b as the
solvent polarity is increased, resulting in an increase in K.
NaDC, which has only two hydroxyl groups on the bile salt
backbone, is known to be more hydrophobic and to form more
rigid aggregates that better protect molecules from the aqueous
solution than NaCh or NaTC.32,35 The I/III ratio in the pyrene
fluorescence spectrum is known to be sensitive to the polarity
of the environment in which pyrene is located, and this ratio
increases with solvent polarity. Studies of pyrene in various bile
salt solutions found a higher pyrene I/III ratio in NaCh than in
NaDC,60 and an even higher I/III ratio was observed for pyrene
in NaTC aggregates.61 Thus, the trends observed in k1 and K,
which indicated that 1 experienced the most polar environment
in NaTC and the least polar environment in NaDC, are
consistent with the trends observed for other guests included in
bile salt aggregates. It is important to note that the comparison
with the pyrene studies is made with respect to the polarity
changes; no information is available on whether the binding site
for the isomers of 1 is the same as that for pyrene.
A similar dependence on bile salt structure was observed for
the values of Φab obtained for 1 (NaDC > NaCh > NaTC),
whereas Φba for 1 did not vary with bile salt structure. The
values of Φab for 1 are known to decrease with increasing
solvent polarity; however, some variation in this trend occurs
because the photoisomerization can occur by a singlet or a
triplet pathway.62 Both Φab and Φba for 1 were larger in bile salt
solutions than in ethanol, and this trend does not correlate with
the expected changes with polarity of different solvents. This
result could be due to a different partitioning between the
singlet and triplet pathways in the bile salt than in
homogeneous solution. Alternatively, the trend could also be
influenced by a relocation of the less polar isomer 1a from a
hydrophobic binding site in the aggregate to a more polar
binding site when 1b is formed. This relocation dynamics could
play a role in the continuous irradiation experiments used to
determine the quantum yield for the photoconversion of 1a to
1b.
precision value of (3.68
0.03) × 104 cm−1 M−1 was
determined at 537 nm in ethanol. Quantum yields for the
photoconversion between the isomers were obtained from
kinetic experiments using continuous irradiation. The structure
of the bile salt did not alter either quantum yield for 2,
suggesting that changes in the structure of the aggregates for
the different bile salts were not sufficient to change the
equilibrium between the reactive and nonreactive conformers
of 2. In the case of 1, the structure of the bile salt aggregate
influenced the quantum yield for the reaction of 1a to 1b,
whereas the quantum yield for the reverse reaction from 1b to
1a was not affected. The position of the absorption maxima
indicated that 1b was located in an environment that was more
polar than ethanol, and the relaxation kinetic studies suggested
that the polarity for 1b was highest in the aggregates of NaTC,
followed by NaCh and NaDC. These experiments showed that
bile salt aggregates can be used to develop photochromic
systems in aqueous solutions that incorporate photochromic
molecules where both isomers are hydrophobic or photo-
chromic compounds where the polarity of the isomers is
significantly different.
ASSOCIATED CONTENT
* Supporting Information
■
S
Details for experimental procedures, description of continuous
irradiation system, HPLC chromatograms and response curves,
numerical models for quantum yield determination, and
relaxation kinetics for 1b. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: cornelia.bohne@gmail.com.
*E-mail: kentaro.morimitsu@xrcc.xeroxlabs.com.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Xerox Foundation for a University Affairs
Committee grant and the Natural Sciences and Engineering
Research Council of Canada (NSERC) for a Collaborative
Research and Development grant. We thank Luis Netter for his
assistance in the development of the irradiation system.
REFERENCES
■
Finally, it is important to note that the studies here reported
were performed at one concentration of bile salt. Bile salts are
known to form two types of aggregates, small hydrophobic
primary aggregates and larger secondary aggregates, and the
size of the aggregate continuously increases as the concen-
tration of bile salt is raised.28−30 It is possible that changes in
the aggregate size will have an effect on photochromic
properties, and such an effect will be investigated in future.
(1) Irie, M. Diarylethenes for Memories and Switches. Chem. Rev.
2000, 100, 1685−1716.
(2) Bouas-Laurent, H.; Durr, H. Organic Photochromism. Pure Appl.
̈
Chem. 2001, 73, 639−665.
(3) Tian, H.; Wang, S. Photochromic Bisthienylethene as Multi-
Function Switches. Chem. Commun. 2007, 781−792.
(4) Orgiu, E.; Samorì, P. 25th Anniversary Article: Organic
Electronics Marries Photochromism: Generation of Multifunctional
Interfaces, Materials, and Devices. Adv. Mater. 2014, 26, 1827−1845.
́
(5) Gust, D.; Andreasson, J.; Pischel, U.; Moore, T. A.; Moore, A. L.
CONCLUSIONS
■
Data and Signal Processing using Photochromic Molecules. Chem.
Commun. 2012, 48, 1947−1957.
The photochromism of 1 and 2 was quantified in aqueous
solutions of bile salt aggregates with different structures. The
incorporation of 1 and 2 into a supramolecular system required
the use of methods that did not rely on the isolation of the
individual isomers for determination of spectroscopic and
́
(6) Szymanski, W.; Beierle, J. M.; Kistemaker, H. A. V.; Velema, W.
A.; Feringa, B. L. Reversible Photocontrol of Biological Systems by the
Incorporation of Molecular Photoswitches. Chem. Rev. 2013, 113,
6114−6178.
11326
dx.doi.org/10.1021/la503164e | Langmuir 2014, 30, 11319−11328

Langmuir
Article
(7) Raymo, F. M.; Tomasulo, M. Electron and Energy Transfer
Modulation with Photochromic Switches. Chem. Soc. Rev. 2005, 34,
327−336.
(28) Hinze, W. L.; Hu, W.; Quina, F. H.; Mohammadzai, I. U. Bile
Acid/Salt Surfactant Systems: General Properties and Survey of
Analytical Applications. In Organized Assemblies in Chemical Analysis;
Hinze, W. L., Ed.; JAI Press Inc.: Stamford, 2000; Vol. 2: Bile Acid/
Salt Surfactant Systems, pp 1−70.
(29) Kratohvil, J. P. Size of Bile Salt Micelles: Techniques, Problems
and Results. Adv. Colloid Interface Sci. 1986, 26, 131−154.
(30) O’Connor, C. J.; Wallace, R. G. Physico-Chemical Behavior of
Bile Salts. Adv. Colloid Interface Sci. 1985, 22, 1−111.
(31) Amundson, L. L.; Li, R.; Bohne, C. Effect of the Guest Size and
Shape on Its Binding Dynamics with Sodium Cholate Aggregates.
Langmuir 2008, 24, 8491−8500.
(32) Li, R.; Carpentier, E.; Newell, E. D.; Olague, L. M.; Heafey, E.;
Yihwa, C.; Bohne, C. Effect of the Structure of Bile Salt Aggregates on
the Binding of Aromatic Guests and the Accessibility of Anions.
Langmuir 2009, 25, 13800−13808.
(33) Waissbluth, O. L.; Morales, M. C.; Bohne, C. Influence of
Planarity and Size on Guest Binding with Sodium Cholate Aggregates.
Photochem. Photobiol. 2006, 82, 1030−1038.
(8) Klajn, R. Spiropyran-based Dynamic Materials. Chem. Soc. Rev.
2014, 43, 148−184.
(9) Stafforst, T.; Hilvert, D. Kinetic Characterization of Spiropyrans
in Aqueous Media. Chem. Commun. 2009, 287−288.
(10) Hammarson, M.; Nilsson, J. R.; Li, S.; Beke-Somfai, T.;
Andreasson, J. Characterization of the Thermal and Photoinduced
́
Reactions of Photochromic Spiropyrans in Aqueous Solution. J. Phys.
Chem. B 2013, 117, 13561−13571.
(11) Keum, S.-R.; Roh, S.-J.; Ahn, S.-M.; Lim, S.-S.; Kim, S.-H.; Koh,
K. Solvatochromic Behavior of Non-Activated Indolinobenzospiropyr-
an 6-Carboxylates in Aqueous Binary Solvent Mixtures. Part II. Dyes
Pigm. 2007, 74, 343−347.
̈
(12) Kohl-Landgraf, J.; Braun, M.; Ozco̧ ban, C.; Gonca̧ lves, D. P. N.;
Heckel, A.; Wachtveitl, J. Ultrafast Dynamics of a Spiropyran in Water.
J. Am. Chem. Soc. 2012, 134, 14070−14077.
(13) Yamada, M.; Takeshita, M.; Irie, M. Photochromism of
Diarylethene Diammonium Derivative in the Cyclodextrin Cavity.
Mol. Cryst. Liq. Cryst. 2000, 345, 107−112.
(34) Rinco, O.; Nolet, M.-C.; Ovans, R.; Bohne, C. Probing the
Binding Dynamics to Sodium Cholate Aggregates using Naphthalene
Derivatives as Guests. Photochem. Photobiol. Sci. 2003, 2, 1140−1151.
(35) Ju, C.; Bohne, C. Dynamics of Probe Complexation to Bile Salt
Aggregates. J. Phys. Chem. 1996, 100, 3847−3854.
(14) Al-Atar, U.; Fernandes, R.; Johnsen, B.; Baillie, D.; Branda, N. R.
A Photocontrolled Molecular Switch Regulates Paralysis in a Living
Organism. J. Am. Chem. Soc. 2009, 131, 15966−15967.
(36) Pace, T. C. S.; Souza Junior, S. P.; Zhang, H. T.; Bohne, C.
́
(15) Hirose, T.; Matsuda, K.; Irie, M. Self-Assembly of Photochromic
Diarylethenes with Amphiphilic Side Chains: Reversible Thermal and
Photochemical Control. J. Org. Chem. 2006, 71, 7499−7508.
(16) Tong, Z.; Pu, S.; Xiao, Q.; Liu, G.; Cui, S. Synthesis and
Photochromism of a Novel Water-Soluble Diarylethene with
Glucosyltriazolyl Groups. Tetrahedron Lett. 2013, 54, 474−477.
(17) Shoji, Y.; Yagi, A.; Horiuchi, M.; Morimoto, M.; Irie, M.
Photochromic Diarylethene Derivatives Bearing Hydrophilic Sub-
stituents. Isr. J. Chem. 2013, 53, 303−311.
Effect of Terbium(III) on the Binding of Aromatic Guests with
Sodium Taurocholate Aggregates. Photochem. Photobiol. Sci. 2011, 10,
1568−1577.
(37) Fuentealba, D.; Thurber, K.; Bovero, E.; Pace, T. C. S.; Bohne,
C. Effect of Sodium Chloride on the Binding of Polyaromatic
Hydrocarbon Guests with Sodium Cholate Aggregates. Photochem.
Photobiol. Sci. 2011, 10, 1420−1430.
(38) Yihwa, C.; Quina, F. H.; Bohne, C. Modulation with Acetonitrile
of the Dynamics of Guest Binding to the Two Distinct Binding Sites of
Cholate Aggregates. Langmuir 2004, 20, 9983−9991.
(39) Li, R.; Santos, C. S.; Norsten, T. B.; Morimitsu, K.; Bohne, C.
Aqueous Solubilization of Photochromic Compounds by Bile Salt
Aggregates. Chem. Commun. 2010, 46, 1941−1943.
(40) Hatchard, C. G.; Parker, C. A. A New Sensitive Chemical
Actinometer. II. Potassium Ferrioxalate as a Standard Chemical
Actinometer. Proc. R. Soc. London, Ser. A 1956, 235, 518−536.
(41) Irie, M.; Sakemura, K.; Okinaka, M.; Uchida, K. Photochromism
of Dithienylethenes with Electron-Donating Substituents. J. Org. Chem.
1995, 60, 8305−8309.
(42) Shibata, K.; Kobatake, S.; Irie, M. Extraordinarily Low
Cycloreversion Quantum Yields of Photochromic Diarylethenes with
Methoxy Substituents. Chem. Lett. 2001, 30, 618−619.
(43) Fischer, E. The Calculation of Photostationary States in Systems
A.dblarw. B When Only A Is Known. J. Phys. Chem. 1967, 71, 3704−
3706.
(18) Zhou, J.; Sui, Q.; Huang, B. Photoinduced Self-Assembly of the
Supramolecular Photochromic Systems - Photoinduced Dimer
Formation of the Inclusion Complexes of an Indolinospiropyran
with CDs. J. Photochem. Photobiol., A 1998, 117, 129−136.
(19) Sueishi, Y.; Nishimura, T.; Miyakawa, T. A Method of
Evaluating the Rate Constant of the Thermal Isomerization of 6-
‑
SO3 -Spiropyran by using β-Cyclodextrin. Chem. Lett. 1995, 1061−
1062.
(20) Tan, W.; Zhang, D.; Wen, G.; Zhou, Y.; Zhu, D. Tuning the
Energy Transfer Process for the Ensemble of Fluorescein with β-
Cyclodextrin (β-CD) Unit and Spiropyran with Adamantyl (AD)
Unit: A Temperature-Gated Molecular Fluorescence Switch. J.
Photochem. Photobiol., A 2008, 200, 83−89.
(21) Takeshita, M.; Kato, N.; Kawauchi, S.; Imase, T.; Watanabe, J.;
Irie, M. Photochromism of Dithienylethenes Included in Cyclo-
dextrins. J. Org. Chem. 1998, 63, 9306−9313.
(22) Tamaki, T.; Sakuragi, M.; Ichimura, K.; Aoki, K.; Arima, I. The
Photochromism of Nitrospiropyran Included in γ-Cyclodextrin. Polym.
Bull. 1990, 24, 559−564.
(44) Irie, M.; Mohri, M. Thermally Irreversible Photochromic
Systems. Reversible Photocyclization of Diarylethene Derivatives. J.
Org. Chem. 1988, 53, 803−808.
(23) Saitoh, M.; Fukaminato, T.; Irie, M. Photochromism of a
Diarylethene Derivative in Aqueous Solution Capping with a Water-
Soluble Nano-Cavitand. J. Photochem. Photobiol., A 2009, 207, 28−31.
(45) Sumi, T.; Takagi, Y.; Yagi, A.; Morimoto, M.; Irie, M.
Photoirradiation Wavelength Dependence of Cycloreversion Quantum
Yields of Diarylethenes. Chem. Commun. 2014, 50, 3928−3930.
(46) Flannery, J. B., Jr. The Photo- and Thermochromic Transients
from Substituted 1′,3′,3′-Trimethylindolinobenzospiropyrans. J. Am.
Chem. Soc. 1968, 90, 5660−5671.
́
(24) Miskolczy, Z.; Biczok, L. Photochromism in Cucurbit[8]uril
Cavity: Inhibition of Hydrolysis and Modification of the Rate of
Merocyanine-Spiropyran Transformations. J. Phys. Chem. B 2011, 115,
12577−12583.
(47) Heller, C. A.; Fine, D. A.; Henry, R. A. Photochromism. J. Phys.
Chem. 1961, 65, 1908−1909.
́
(25) Miskolczy, Z.; Biczok, L. Photochromism of a Merocyanine Dye
Bound to Sulfonatocalixarenes: Effect of pH and the Size of
Macrocycle on the Kinetics. J. Phys. Chem. B 2013, 117, 648−653.
(26) Ikeda, S.; Saso, Y. Solubilization of Spiropyran in Aqueous NaBr
Solutions of Dodecyltrimethylammonium Bromide. Colloids Surf.
1992, 67, 21−27.
(48) Gorner, H. Photochromism of Nitrospiropyrans: Effects of
̈
Structure, Solvent and Temperature. Phys. Chem. Chem. Phys. 2001, 3,
416−423.
(49) Kawanishi, Y.; Seki, K.; Tamaki, T.; Sakuragi, M.; Suzuki, Y.
Tuning Reverse Ring Closure in the Photochromic and Thermochro-
mic Transformation of 1′,3′,3′-Trimethyl-6-nitrospiro[2H-1-benzopyr-
an-2,2′-indoline] Analogues by Ionic Moieties. J. Photochem. Photobiol.,
A 1997, 109, 237−242.
(27) Favaro, G.; Ortica, F.; Malatesta, V. Photochromism and
Thermochromism of Spiro[indolinoxazines] in Normal and Reversed
Micelles. J. Chem. Soc., Faraday Trans. 1995, 91, 4099−4103.
11327
dx.doi.org/10.1021/la503164e | Langmuir 2014, 30, 11319−11328

Langmuir
Article
(50) Song, X.; Zhou, J.; Li, Y.; Tang, Y. Correlations between
Solvatochromism, Lewis Acid-Base Equilibrium and Photochromism
of an Indoline Spiropyran. J. Photochem. Photobiol., A 1995, 92, 99−
103.
(51) Zhang, S.; Zhang, Q.; Ye, B.; Li, X.; Zhang, X.; Deng, Y.
Photochromism of Spiropyran in Ionic Liquids: Enhanced Fluo-
rescence and Delayed Thermal Reversion. J. Phys. Chem. B 2009, 113,
6012−6019.
(52) Delbaere, S.; Vermeersch, G.; Micheau, J.-C. Quantitative
Analysis of the Dynamic Behaviour of Photochromic Systems. J.
Photochem. Photobiol., C 2011, 12, 74−105.
(53) Ottavi, G.; Ortica, F.; Favaro, G. Photokinetic Methods: A
Mathematical Analysis of the Rate Equations in Photochromic
Systems. Int. J. Chem. Kinet. 1999, 31, 303−313.
(54) Pimienta, V.; Lavabre, D.; Levy, G.; Samat, A.; Guglielmetti, R.;
Micheau, J. C. Kinetic Analysis of Photochromic Systems under
Continuous Irradiation. Application to Spiropyrans. J. Phys. Chem.
1996, 100, 4485−4490.
́
(55) Pimienta, V.; Froute, C.; Deniel, M. H.; Lavabre, D.;
Guglielmetti, R.; Micheau, J. C. Kinetic Modelling of the Photo-
chromism and Photodegradation of a Spiro[indoline-naphthoxazine].
J. Photochem. Photobiol., A 1999, 122, 199−204.
(56) Bohne, C. Supramolecular Dynamics. Chem. Soc. Rev. 2014, 43,
4037−4050.
(57) Gorner, H. Photochemical Ring Opening in Nitrospiropyrans:
̈
Triplet Pathway and the Role of Singlet Molecular Oxygen. Chem.
Phys. Lett. 1998, 282, 381−390.
(58) Wohl, C. J.; Kuciauskas, D. Excited-State Dynamics of
Spiropyran-Derived Merocyanine Isomers. J. Phys. Chem. B 2005,
109, 22186−22191.
(59) Keum, S.-R.; Hur, M.-S.; Kazmaier, P. M.; Buncel, E. Thermo-
and Photochromic Dyes: Indolino-benzospiropyrans. Part 1. UV-Vis
Spectroscopic Studies of 1,3,3,-Spiro(2H-1-benzopyran-2,2′-indolines)
and the Open-Chain Merocyanine Forms; Solvatochromism and
Medium Effects on Spiro Ring Formation. Can. J. Chem. 1991, 69,
1940−1947.
(60) Zana, R.; Guveli, D. Fluorescence Probing Study of the
Association of Bile Salts in Aqueous Solutions. J. Phys. Chem. 1985, 89,
1687−1690.
(61) Hashimoto, S.; Thomas, J. K. Photophysical Studies of Pyrene in
Micellar Sodium Taurocholate at High Salt Concentrations. J. Colloid
Interface Sci. 1984, 102, 152−163.
(62) Chibisov, A. K.; Gorner, H. Singlet Versus Triplet Photo-
̈
processes in Indodicarbocyanine Dyes and Spiropyran-Derived
Merocyanines. J. Photochem. Photobiol., A 1997, 105, 261−267.
11328
dx.doi.org/10.1021/la503164e | Langmuir 2014, 30, 11319−11328