Fast and stable photochromic oxazines for fluorescence switching

Article abstract of DOI:10.1021/la201062h

The stringent limitations imposed by diffraction on the spatial resolution of fluorescence microscopes demand the identification of viable strategies to switch fluorescence under optical control. In this context, the photoinduced and reversible transforma

Full text of DOI:10.1021/la201062h

INVITED FEATURE ARTICLE
pubs.acs.org/Langmuir
Fast and Stable Photochromic Oxazines for Fluorescence Switching
Erhan Deniz,† Massimiliano Tomasulo,† Janet Cusido,† Salvatore Sortino,*,‡ and Franc-isco M. Raymo*,†
†Laboratory for Molecular Photonics, Department of Chemistry, University of Miami, 1301 Memorial Drive, Coral Gables,
Florida 33146-0431, United States
‡Laboratory of Photochemistry, Department of Drug Sciences, University of Catania, viale Andrea Doria 6, Catania, I-95125, Italy
S Supporting Information
b
ABSTRACT: The stringent limitations imposed by diffraction on the spatial
resolution of fluorescence microscopes demand the identification of viable strategies
to switch fluorescence under optical control. In this context, the photoinduced
and reversible transformations of photochromic compounds are particularly
valuable. In fact, these molecules can be engineered to regulate the emission
intensities of complementary fluorophores in response to optical stimulations.
On the basis of this general design logic, we assembled a functional molecular
construct consisting of a borondipyrromethene fluorophore and a nitrospiropyran photochrome and demonstrated that the
emission of the former can be modulated with the interconversion of the latter. This fluorophoreꢀphotochrome dyad, however,
has a slow switching speed and poor fatigue resistance. To improve both parameters, we developed a new family of photochromic
switches based on the photoinduced opening and thermal closing of an oxazine ring. These compounds switch back and forth
between ring-closed and -open isomers on nanosecondꢀmicrosecond timescales and tolerate thousands of switching cycles with
no sign of degradation. In addition, the attachment of appropriate chromophoric fragments to their switchable oxazine ring can be
exploited to either deactivate or activate fluorescence reversibly in response to illumination with a pair of exciting beams.
Specifically, we assembled three dyads, each based on either a borondipyrromethene or a coumarin fluorophore and an oxazine
photochrome, and modulated their fluorescence in a few microseconds with outstanding fatigue resistance. The unique
photochemical and photophysical properties of our fluorophoreꢀphotochrome dyads can facilitate the development of
switchable fluorophores for superresolution imaging and, ultimately, provide valuable molecular probes for the visualization
of biological samples on the nanometer level.
operating principles to avoid diffraction can extend the resolving
power of fluorescence imaging down to the nanoscale. None-
theless, these strategies require viable mechanisms to switch
fluorescence under optical control to allow a transition from
microscopy to nanoscopy.
1. INTRODUCTION
The introduction of fluorescent labels1 within a biological
sample offers the opportunity to reconstruct noninvasively an
image of the specimen after the excitation of the probes and the
collection of their emission with the aid of a microscope.2 Indeed,
fluorescence microscopy is currently the method of choice in the
biomedical laboratory for the visualization of cells and tissues.3
Nonetheless, the phenomenon of diffraction4 limits the resolu-
tion of conventional fluorescence microscopes to hundreds of
nanometers in both the horizontal plane and vertical direction.2
Specifically, fluorescent probes separated by a few nanometers
cannot be distinguished in a conventional fluorescence image.
Therefore, this convenient technique cannot appreciate the
structural details that govern biological processes on the
molecular level.
Time can be exploited to resolve what cannot be distinguished
in space. In particular, the stringent limitations imposed by
diffraction on the spatial resolution of conventional fluorescence
microscopes can be overcome with a combination of switchable
probes and multiphoton illumination schemes.5ꢀ12 Indeed,
labels designed to turn their fluorescence from on to off, or vice
versa, in response to optical stimulations permit the temporal
resolution of spatially indistinguishable objects and the sequen-
tial reconstruction of subdiffraction images. In fact, these clever
Photochromic compounds13ꢀ18 switch reversibly between
states with distinct absorption spectra in the visible region. In
some instances, their photoinduced changes in absorption are
also accompanied by changes in emission.18ꢀ20 Specifically, only
one of the two interconverting states of a photochromic com-
pound is often fluorescent.21ꢀ25 Under these conditions, the
photoinduced interconversion of the two states translates into
fluorescence switching. Alternatively, the pronounced structural
and electronic modifications associated with a photochromic
transformation can be engineered to switch the emission of a
complementary fluorophore. In particular, fluorescent and
photochromic components can be integrated within the same
molecular skeleton and the emission of the former can be
switched with the photoinduced interconversion of the latter.
Indeed, the transformation of one state of the photochromic
Received: March 21, 2011
Revised:
May 4, 2011
Published: May 17, 2011
r
2011 American Chemical Society
11773
dx.doi.org/10.1021/la201062h Langmuir 2011, 27, 11773–11783
|

Langmuir
INVITED FEATURE ARTICLE
Figure 2. UV irradiation of 2a converts its nitrospiropyran com-
ponent to the corresponding merocyanine. In resulting isomer 2b, the
transfer of either an electron or energy from the excited borondipyrro-
methene to the merocyanine fragment quenches the fluorescence of the
former.
the two aromatic chromophores in the ground state (S0). As a
consequence, electronic transitions from S0 to the first singlet
excited state (S1) of these compounds are generally centered in
the ultraviolet region. Upon illumination within this range of
wavelengths, however, nitrospiropyrans switch to merocyanine
isomers (e.g., 1c in Figure 1) with good quantum yields. The
extended chromophoric system within the photogenerated spe-
cies absorbs in the visible region, and hence these photoinduced
processes result in the appearance of color. In most instances, the
photogenerated isomer reverts thermally to the original one with
first-order kinetics over the course of hundreds of seconds in
organic solvents. Indeed, the photoinduced coloration and
thermal decoloration of materials doped with nitrospiropyrans
can easily be achieved simply by turning an ultraviolet illumina-
tion source on and off.
Figure 1. Ultraviolet (UV) illumination of nitrospiropyran 1a results in
the cleavage of the [CꢀO] bond at the spirocenter to form ring-open
intermediate 1b and the cis f trans isomerization of the adjacent
[CdC] bond to generate final merocyanine 1c. In turn, photogenerated
isomer 1c thermally reverts back to original species 1a through the same
intermediate, 1b, after trans f cis isomerization and ring-closing steps.
component into the other can either activate or prevent an
intercomponent quenching pathway. Under these conditions,
the interconversion of the two states of the photochromic
component controls the excitation dynamics of the fluore-
scent partner and modulates its emission intensity. In most
fluorophoreꢀphotochrome constructs, either electron26ꢀ29 or
energy30ꢀ35 transfer is responsible for quenching.36 In one
instance, an electron is transferred either to or from the excited
fluorophore from or to, respectively, only one of the two states of
the photochrome. This mechanism requires either the oxidation
or the reduction potential, respectively, of the photochrome to
change significantly with the photochromic transformation. In
the other instance, energy is transferred from the excited
fluorophore to only one of the two states of the photochrome.
This strategy demands the overlap between the emission band of
the former and the absorption band of the latter to change
significantly with the photochromic conversion. When at least
one of these conditions is satisfied, the photochromic transfor-
mation translates into fluorescence switching.
Diverse structural designs have been explored over the past six
decades to implement photochromic transformations, and nu-
merous families of photochromic molecules have emerged as a
result.13ꢀ18 In particular, the photochromism of spiropyrans has
been extensively investigated and these compounds are now
routinely integrated within a wealth of photoresponsive molec-
ular and supramolecular constructs.37ꢀ48 Synthetic accessibility
and reliable photochemical response are the main reasons behind
the success of these versatile photoresponsive building blocks. In
particular, spiropyrans with a nitro group on their benzopyran
fragment (e.g., 1a in Figure 1) can be prepared from 2-hydroxy-5-
nitrobenzaldehyde in a single synthetic step with good yields.
Their spirocenter holds the two heterocycles in an orthogonal
arrangement and prevents electronic communication between
2. FLUORESCENCE SWITCHING WITH A FLUOROPHOREꢀ
PHOTOCHROME DYAD
The significant absorbance changes in the visible region,
associated with the photoinduced transformation of a nitrospiro-
pyran to the corresponding merocyanine, can be exploited to
activate an energy-transfer pathway and control the emission of a
fluorescent partner. On the basis of these considerations, we
envisaged the possibility of pairing a borondipyrromethene
fluorophore and a nitrospiropyran photochrome within the same
molecular construct in the form of compound 2a (Figure 2).49
We prepared this molecule in a single synthetic step from
preformed borondipyrromethene and nitrospiropyran precur-
sors with pendant carboxylic acid and 2-hydroxyethyl groups,
respectively. In particular, the condensation of these two species,
under the assistance of 4-dimethylaminopyridine and N,N0-
dicyclohexylcarbodiimide, produced fluorophoreꢀphotochrome
dyad 2a with a yield of 54%.
The steady-state absorption spectrum of 2a (a in Figure 3) is
essentially the sum of those of its two separate chromophoric com-
ponents.49a These observations are indicative of negligible
electronic interactions between the fluorescent and photochro-
mic fragments of 2a in S0. Specifically, the spectrum of 2a shows
a band at 523 nm for the borondipyrromethene fluorophore
and one at 347 nm for the nitrospiropyran photochrome. The
selective excitation of the fluorescent fragment at 480 nm
results in the appearance of its characteristic fluorescence at
11774
dx.doi.org/10.1021/la201062h |Langmuir 2011, 27, 11773–11783

Langmuir
INVITED FEATURE ARTICLE
Figure 4. Illumination of a poly(methyl methacrylate) film doped with
fluorescent species 2a, with a laser producing a doughnut-shaped pattern
at 351 nm, generating nonfluorescent state 2b in the illuminated area. As
a result, the subsequent excitation at 480 nm reveals a doughnut-shaped
nonfluorescent pattern in the corresponding image (scale bar = 1 mm).
of 2b back to 2a within the polymer film, however, is relatively slow,
and the imprinted pattern fades only after tens of minutes.
Furthermore, the emission intensity of the doped film drops to
ca. 60% of the initial value after only five switching cycles because of
the gradual photodegradation of the dopant. In fact, the slow
switching speed and poor fatigue resistance of this fluorophoreꢀ
photochrome dyad prevent the reconstruction of fluorescence
images with subdiffraction resolution, even though the operating
principles engineered into this functional molecular construct
permit the spatial confinement of fluorescence. Thus, both the
speed and the stability of this particular system are in need of
significant improvements.
Figure 3. Absorption spectra (0.1 mM, MeCN, 25 °C) of a solution of
2a recorded (a) before and (b) after ultraviolet illumination (254 nm, 0.5
mW cmꢀ2, 5 min), revealing the appearance of a band at 570 nm for the
merocyanine component of photogenerated isomer 2b. The corre-
sponding emission spectra (0.01 mM, MeCN, 25 °C, 480 nm), recorded
(c) before and (d) after irradiation, show a decrease in the borondipyrro-
methene fluorescence at 539 nm with the photoinduced isomerization.
539 nm in the emission spectrum (c in Figure 3). Indeed, the
lack of any overlap between the emission of the fluorophore and
the absorption of the photochrome prevents the transfer of
energy from the former to the latter and permits the radiative
deactivation of the excited borondipyrromethene. Upon ultra-
violet illumination, however, the nitrospiropyran switches to
the corresponding merocyanine to form 2b (Figure 2) with the
concomitant appearance of its characteristic absorption (b in
Figure 3) in the same range of wavelengths where the fluor-
ophore emits. The significant overlap between the absorption
of one and the emission of the other permits the transfer of
energy from the fluorescent to the photochromic fragment. In
addition, the oxidation and reduction potentials of the separate
components indicate that the transfer of one electron from the
excited fluorophore to the photochrome becomes exergonic
with the conversion of the nitrospiropyran to the correspond-
ing merocyanine. Thus, the photochromic transformation
activates both electron- and energy-transfer pathways that
ultimately encourage the nonradiative deactivation of the
excited fluorophore. In fact, the emission intensity recorded
for the photostationary state (d in Figure 3) is significantly
lower than that measured before ultraviolet illumination (c in
Figure 3). Photogenerated state 2b, however, has a lifetime of
270 s and eventually reverts back to original isomer 2a. As a
result, the initial absorption and emission spectra are fully
restored on a timescale of minutes.
The photochemical and photophysical properties of 2a offer
the opportunity to confine fluorescence spatially with patterned
illumination.49b Indeed, the spin coating of a dichloromethane
solution of poly(methyl methacrylate) and 2a on the surface of a
glass slide results in the entrapment of the fluorophoreꢀ
photochrome dyad within a micrometer-thick polymer film.
The illumination of the doped film with a laser, designed to
generate a doughnut-shaped pattern with a wavelength of
351 nm, encourages the conversion of 2a to 2b exclusively within
the irradiated area. Under these conditions, the subsequent
illumination at the excitation wavelength of the borondipyrro-
methene component reveals a dark doughnut-shaped pattern
(Figure 4) within the fluorescent film. The thermal reisomerization
3. IMPROVING THE SPEED AND STABILITY OF THE
PHOTOCHROMIC COMPONENT
The nitro group on the benzopyran fragment of nitrospiro-
pyrans promotes intersystem crossing and facilitates their
photoinduced isomerization.50 In particular, the ultraviolet illumi-
nation of nitrospiropyran 1a eventually populates its first triplet
state (T1) and encourages the cleavage of the [CꢀO] bond at the
spirocenter. This process results in the formation of ring-open
intermediate 1b (Figure 1) in T1 on a picosecond timescale.51ꢀ55
This species can first undergo a cis f trans isomerization along the
potential energy surface of T1 and then decay down to S0 with the
formation of final merocyanine 1c on a microsecond timescale.
Alternatively, it can first decay to S0 and then undergo a cis f trans
isomerization along the potential energy surface of S0 to form 1c,
once again, in microseconds. Thus, the photoinduced coloration of
a nitrospiropyran involves essentially two main chemical steps,
occurring on picosecond (ring opening) and microsecond (cis f
trans isomerization) timescales respectively.
The participation of T1 in the isomerization of nitrospiropyr-
ans facilitates the formation of the corresponding merocyanines
but also tends to encourage degradation.56 In fact, the photo-
induced conversion of 1a to 1c is accompanied by the sensitiza-
tion of significant amounts of singlet oxygen (1Δg).50f,g This
species is a strong oxidant, and its production encourages the
gradual oxidative degradation of the photochromic system with a
depressive effect on its fatigue resistance. Consistently, the
fatigue resistances of nitrospiropyrans increase considerably in
the presence of singlet oxygen scavengers.57,58 In principle, the
generation of singlet oxygen could be avoided by either drasti-
cally reducing the lifetime of T1 or avoiding its participation in
the isomerization process altogether.
Photogenerated merocyanine 1c spontaneously reverts
back to nitrospiropyran 1a along the potential energy surface
of S0, once again, after two main chemical steps.50c,e,f First, the
11775
dx.doi.org/10.1021/la201062h |Langmuir 2011, 27, 11773–11783

Langmuir
INVITED FEATURE ARTICLE
Figure 5. Treatment of the bromide salt of 3H-indolium cation 3 with
potassium hydroxide to generate 2H,4H,5H-oxazole 5 instead of ex-
pected tautomer 4.
Figure 7. Steady-state absorption spectra (0.1 mM, MeCN, 25 °C) of
(a) 6a and (b) 4-nitroanisole showing essentially the same band for the
S0 absorption of their 4-nitrophenoxy chromophore. (c) The time-
resolved absorption spectrum of a solution (0.1 mM, MeCN, 25 °C) of
6a, recorded 30 ns after illumination with a pulsed laser (355 nm, 6 ns, 8
mJ), and (d) the steady-state absorption spectrum (0.1 mM, MeCN, 25
°C) of t-butylammonium 4-nitrophenolate reveal essentially the same
band for the S0 absorption of their 4-nitrophenolate chromophore.
2H,4H,5H-oxazole ring. In light of the structural implications in
the isomerization mechanism of nitrospiropyrans, we realized
that the introduction of an appropriate chromophore onto the
rim of the oxazole ring of 5 would offer the opportunity to
implement photochromic transformations based exclusively on
fast ring-opening and -closing steps. On the basis of these
considerations and the availability of appropriate precursors,
we designed 2H,4H-[1,3]oxazine 6a (Figure 6) and synthesized
this compound in two steps, starting from phenyl hydrazine and
i-propylphenyl ketone.61 In particular, the condensation of these
two commercial precursors under acidic conditions gave 3H-
indole 7 in a yield of 76%. The reaction of 7 with 2-chloromethyl-
4-nitrophenol produced target molecule 6a in a yield of 58%.
The steady-state absorption spectrum of 6a (a in Figure 7)
shows a band centered at 308 nm (λa in Table 1). This band
resembles the absorption of 4-nitroanisole (b in Figure 7) and
can be assigned to the 4-nitrophenoxy chromophore of 6a.
Illumination in the tail of this absorption with a pulsed laser
operating at 355 nm results in the cleavage of the [CꢀO] bond at
the junction of the two heterocycles and the opening of the
oxazine ring with a quantum yield of 0.10 (φ in Table 1). This
process generates zwitterionic isomer 6b (Figure 8) within the
laser pulse (6 ns) and is accompanied by the appearance of an
absorption band centered at 440 nm (λb in Table 1 and c in
Figure 7). This band resembles the S0 absorption of t-butylam-
monium 4-nitrophenolate (d in Figure 7) and can be assigned to
the anionic chromophore of 6b. This transient absorption decays
monoexponentially on a nanosecond timescale with the sponta-
neous reisomerization of 6b back to 6a.62 Curve fitting of the
temporal absorbance evolution (Figure 9) indicates the lifetime
of the photogenerated species to be 22 ns (τ in Table 1).63 In fact,
a full switching cycle, from 6a to 6b and back, can be completed
in a few tens of nanoseconds. Furthermore, the photoisomeriza-
tion of 6a to 6b is not accompanied by the sensitization of singlet
oxygen, as in the case of nitrospiropyrans. As a result, this
photochromic system can be switched back and forth between
its two states thousands of times without decomposing, even in
the presence of molecular oxygen. Thus, the transition from
Figure 6. Condensation of phenylhydrazine and i-propylphenylketone
to generate 3H-indole 7 and the reaction of this species with 2-chloro-
methyl-4-nitrophenol to produce oxazine 6a.
trans f cis isomerization of the [CdC] double bond joining the
cationic and anionic fragments of 1c produces the intermediate
1b. Then, the [CꢀO] bond at the spirocenter reforms to close
the benzopyran ring and regenerate 1a. The first of these two
steps, however, is relatively slow and determines the decoloration
rate.59 Specifically, it imposes a lifetime of several hundreds of
seconds on 1c in organic solvents.50c,e,f Thus, both the photo-
induced coloration and the thermal decoloration of these photo-
chromic compounds are slowed down by the need to change the
configuration of a [CdC] bond from cis to trans and trans to cis.
In principle, however, the cleavage and reformation of the
[CꢀO] bond at the spirocenter should be sufficient to ensure
the pronounced absorbance changes in the visible region ex-
pected from photochromic transformations. Therefore, photo-
chromic switches faster than conventional nitrospiropyrans
could simply be realized by preventing the slow cis/trans
isomerization steps and instead relying exclusively on the in-
herently fast ring-opening and -closing steps.
To synthesize a nitrospiropyran with a 2-hydroxyethyl appen-
dage on its 2H,3H-indole nitrogen atom, we intended to convert
the bromide salt of 3H-indolium cation 3 (Figure 5) to 3H-indole
4 and condense this compound with 2-hydroxy-5-nitrobenzal-
dehyde.60 The treatment of 3 with potassium hydroxide, how-
ever, did not produce expected compound 4. Instead, it resulted
in the formation of its tautomer, 5, after the closing of a
11776
dx.doi.org/10.1021/la201062h |Langmuir 2011, 27, 11773–11783

Langmuir
INVITED FEATURE ARTICLE
Table 1. Photochemical and Photophysical Parameters As-
sociated with Oxazines 6a, 9a, 14a, 15a, 17aꢀ20a, 22a, 24a,
and 25aa
λa (nm)
λb (nm)
φ
τ
ref
6a
308
338
304
310
288
330
305
318
305
305
412
440
0.10
0.11
0.11
0.01
0.08
0.28
0.07
0.03
0.05
0.02
0.02
22 ns
29 ns
21 ns
10 μs
38 ns
140 ns
2 μs
61
9a
510
64c
14a
15a
17a
18a
19a
20a
22a
24a
25a
440
65d
440
65d
420
65aꢀ65c
65b,65c
65d
430
440 and 550
440
22 ns
1 μs
61b,61c
72
430 and 560
430 and 560
570
Figure 9. S0 absorption of the 4-nitrophenolate chromophore of 6b,
detected after the illumination of a solution (0.1 mM, MeCN, 25 °C) of
6a with a pulsed laser (355 nm, 6 ns, 8 mJ), decaying monoexponentially
with the reisomerization of 6b back to 6a on a nanosecond timescale.
3 μs
72
0.2 μs
73
a The absorption wavelengths for the ring-closed (λa) and -open (λb)
isomers, the quantum yield (φ) of the photoinduced ring-opening
process, and the lifetime (τ) of the ring-open isomer were measured
by steady-state and time-resolved absorption spectroscopies in MeCN at
20ꢀ25 °C. The time-resolved spectra were recorded after excitation at
355 nm with a pulsed Nd:YAG laser (6 ns, 8ꢀ12 mJ). The structure of
20a is shown below.
Figure 10. Substituent R1 on the phenoxy fragment of the oxazine core
can be modified, in the form of compounds 8aꢀ12a, to extend the
conjugation of this chromophore and, in principle, to regulate the color
of the ring-open isomer.
the fluorescent partner integrated within the fluorophoreꢀ
Figure 8. UV illumination of 6a opens the oxazine ring to form
zwitterionic isomer 6b. In turn, photogenerated isomer 6b thermally
reverts back to the original species, 6a, after the closing of the
oxazine ring.
photochrome dyad.
4. REGULATING THE COLOR OF THE PHOTOCHROMIC
COMPONENT
nitrospiropyran 1a to oxazine 6a translates into a decrease in the
time required to complete a full switching cycle from hundreds of
seconds to only tens of nanoseconds and an increase in the
number of switching cycles tolerated by the photochromic
system from a few tens to several thousands. The significant
improvements in switching speed and fatigue resistance are
mostly a consequence of the fact that the photoinduced colora-
tion of 6a and the thermal decoloration of 6b do not require cis/
trans isomerization steps and are instead exclusively based on
ring opening and closing. Thus, the replacement of the nitros-
piropyran component of fluorophoreꢀphotochrome dyad 2a
with an oxazine similar to 6a can translate into the needed
enhancements in speed and stability. Nonetheless, the absorp-
tion properties of the photogenerated isomer of the oxazine
must first be optimized to ensure the transfer of energy from
The 4-nitrophenolate anion of zwitterionic isomer 6b absorbs
in the visible region and is responsible for the photoinduced
coloration of this particular photochromic system.61 In acetoni-
trile, its absorption band is centered at 440 nm (λb in Table 1)
and, in principle, can be shifted bathochromically by extending
the conjugation of this chromophoric fragment. Specifically, the
insertion of π systems between the phenoxy ring of 6a and its
nitro group can offer the opportunity to regulate the color of the
photogenerated isomer. On the basis of these considerations, we
synthesized oxazines 8aꢀ12a (Figure 10), differing in the nature
of the substituent (R1) in the para position relative to the
phenoxy oxygen atom.64 In particular, we isolated 8aꢀ12a in
yields ranging from 21 to 51% after the reaction of 7 (Figure 6)
with the corresponding 2-bromomethyl-4-R1-phenol. Instead,
we synthesized 11a and 12a in yields of 81 and 16% by coupling
11777
dx.doi.org/10.1021/la201062h |Langmuir 2011, 27, 11773–11783

Langmuir
INVITED FEATURE ARTICLE
4-nitrostyrene and (4-nitrophenyl)acetylene, respectively, with a
preformed oxazine having a bromine substituent in the para
position relative to the phenoxy oxygen atom.
The steady-state absorption spectra of 8aꢀ12a (Figure S1)
show bands for their phenoxy chromophores centered at wave-
lengths significantly longer than that of parent oxazine 6a.
Specifically, the bathochromic shift relative to 6a, designed into
these chromophores, varies from 30 nm for 9a to up to 63 nm for
8a. Excitation within these bands under experimental conditions
identical to those employed for 6a, however, opens the oxazine
ring to form the corresponding zwitterionic isomer in S0 only for
9a. In particular, 9a switches to 9b upon ultraviolet illumination
with a quantum yield of 0.11 (φ in Table 1). Consistently, a band
for the S0 absorption of the phenolate anion of 9b appears at
510 nm (λb in Table 1) in the spectrum recorded after excitation.
The photogenerated isomer has a lifetime of 29 ns (τ in Table 1)
and switches back to the original species with first-order kinetics
and the monoexponential decay of its absorbance in the visible
region. In fact, the photochemical response of this photochromic
system is almost identical to that of the parent one, with the
exception of the bathochromic shift engineered into the phenoxy
and phenolate chromophores of the two interconverting isomers.
The excitation dynamics of 8a and 10aꢀ12a are, instead,
significantly different from those of 6a and 9a. Indeed, the time-
resolved spectrum of 8a (a in Figure S2) shows the bleaching of
its S0 absorption upon excitation with the concomitant appear-
ance of a weaker band at shorter wavelengths. These transient
absorptions persist for hundreds of microseconds and eventually
disappear on millisecondꢀsecond timescales. This behavior is
consistent with the photoinduced trans f cis isomerization of
the 4-nitrophenylazophenoxy chromophore and its subsequent
thermal cis f trans reisomerization. In contrast to the behavior
of 8a, the time-resolved absorption spectra of 10a and 11a (b and
c in Figure S2) reveal the appearance of absorptions upon
excitation that decay on a microsecond timescale. The lifetimes
of these transient species, however, decrease significantly in the
presence of molecular oxygen. Furthermore, their bands resem-
ble those observed upon the excitation of model phenoxy
chromophores and can be assigned to T1 absorptions of 10a
and 11a. The time-resolved absorption spectrum of 12a (d in
Figure S2) also shows the formation of a transient species upon
excitation with a lifetime dependent on the concentration
of molecular oxygen. Its band, however, resembles the one
observed upon excitation of a model phenolate equivalent to
that incorporated within ring-open isomer 12b and can be
assigned to an absorption in the triplet manifold of this species.
Thus, the trans f cis isomerization of the phenoxy chromophore
of 8a and the intersystem crossing of the phenoxy chromophores
of 10aꢀ12a dominate the excitation dynamics of these systems
and, with the exception of 12a, prevent the opening of the
oxazine ring.
Figure 11. Substituent R2 on the chiral center of the oxazine ring can be
modified, in the form of compounds 13aꢀ19a, to extend the conjuga-
tion of the 3H-indolium cation generated upon ring opening and, in
principle, regulate the color of the ring-open isomer.
in the nature of the substituent (R2) on the chiral center.65 By
analogy to the synthetic procedure devised for the preparation of
6a (Figure 6), we isolated 13aꢀ15a in overall yields ranging from
26 to 37% after the condensation of phenylhydrazine with the
corresponding i-propyl-R2-ketone and the reaction of the result-
ing 3H-indole with 2-chloromethyl-4-nitrophenol. Instead, we
prepared 16aꢀ19a in a single step with yields ranging from 40 to
76%, starting from preformed oxazine 20a (Table 1) and the
aldehyde of the corresponding chromophoric appendage.66,67
This versatile synthetic approach offers the opportunity to
condense a switchable oxazine to virtually any chromophore
with a formyl substituent and is definitely a convenient protocol
for the preparation of functional molecular constructs.
The steady-state absorption spectra of 13aꢀ19a show a band
for the 4-nitrophenoxy chromophore at ca. 310 nm (a in
Figure 12).65 Upon excitation within this band, the oxazine ring
of these compounds, with the exception of 13a and 16a, opens to
form the corresponding zwitterionic isomers with quantum
yields ranging from 0.01 to 0.28 (φ in Table 1).68 Consistently,
the S0 absorption of the 4-nitrophenolate anion of the ring-open
isomers appears at ca. 440 nm (λb in Table 1) in the spectra
recorded after excitation. Instead, the time-resolved spectra of
13a and 16a do not reveal any significant change upon illumina-
tion. In fact, the redox potentials of model fragments suggest that
the photoinduced transfer of an electron from R2 to the nitro
group of 13a and 16a is exergonic, and presumably, this
competitive process prevents the opening of their oxazine
ring.65b
The complications associated with the excitation dynamics of
8a and 10aꢀ12a encouraged us to explore an alternative strategy
to regulate the color of the photogenerated isomer. In particular,
we realized that, in addition to the anionic fragment, the cationic
component of this zwitterionic species also can be designed to
absorb in the visible region with the introduction of appropriate
substituents. Indeed, the opening of the oxazine ring can extend
the conjugation of a chromophore attached to the chiral center at
the junction of the two heterocyclic fragments and shift its
absorption band bathochromically. On the basis of these con-
siderations, we designed oxazines 13aꢀ19a (Figure 11), differing
The photoinduced ring opening of 14a, 15a, and 17aꢀ19a
brings R2 into conjugation with the 3H-indolium cation of their
ring-open isomers. This structural transformation bathochromi-
cally shifts the absorption of R2 and, with the exception of the
conversion of 14a into 14b,69 results in the appearance of an
additional band in the visible region. In 15b, 17b, and 18b, this
11778
dx.doi.org/10.1021/la201062h |Langmuir 2011, 27, 11773–11783

Langmuir
INVITED FEATURE ARTICLE
Figure 12. Steady-state absorption spectra (0.05 mM, MeCN, 25 °C)
of (a) 19a and (b) 4-nitroanisole showing essentially the same band for
their 4-nitrophenoxy chromophore. (c) The time-resolved absorption
spectrum of a solution (0.01 mM, MeCN, 25 °C) of 19a, recorded 0.1 μs
after illumination with a pulsed laser (355 nm, 6 ns, 12 mJ), and the
steady-state absorption spectra (0.1 mM, MeCN, 25 °C) of (d) t-
butylammonium 4-nitrophenolate and (e) the hexafluorophosphate salt
of 21 reveal essentially the same band for the S0 absorptions of the
4-nitrophenolate and 3H-indolium chromophores.
Figure 13. Dyads 22aꢀ24a incorporating a borondipyrromethene
fluorophore and an oxazine photochrome and differing in the nature
of the covalent linkage bridging the two functional components.
absorption overlaps that of their 4-nitrophenolate anion. In fact,
the photoinduced formation of bichromophoric isomers 17b and
18b translates into 2- and 7-fold enhancements in the coloration
efficiency at 430 nm, respectively, relative to that of parent
monochromophoric system 6b.70,71 In both instances, the tran-
sient absorption decays monoexponentially with the reisomeri-
zation of the ring-open to the ring-closed isomer. The lifetime of
15b, however, is 10 μs (τ in Table 1), whereas those of all of the
other ring-open isomers are in the nanosecond domain. Pre-
sumably, the ability of the dimethylamino group to donate
electrons results in the stabilization of the 3H-indolium cation
of 15b with a concomitant delay of the reisomerization kinetics,
relative to parent system 6b.
In 19b, the cationic and anionic fragments absorb at distinct
wavelengths, and consistently, a pair of bands at 440 and 550 nm
(λb in Table 1) appear in the absorption spectrum (c in
Figure 12) recorded upon illumination.65d These transient bands
resemble the steady-state absorptions (d and e in Figure 12) of t-
butylammonium 4-nitrophenolate and the hexafluorophosphate
salt of model 3H-indolium cation 21, respectively, confirming
their assignment. Furthermore, they decay with identical kinetics
(a and b in Figure S3), demonstrating that they are both
associated with the same species. Specifically, monoexponential
fittings of both absorbance evolutions indicate the lifetime of 19b
to be 2 μs (τ in Table 1). Once again, the ability of the
dimethylamino group to donate electrons and stabilize the 3H-
indolium cation of 19b is presumably responsible for the slower
reisomerization kinetics relative to parent system 6b. In any case,
only a few microseconds are sufficient to complete a full switch-
ing cycle with this bichromophoric photochrome. Furthermore,
this system tolerates hundreds of cycles without decomposing,
and the intense absorption of the 3H-indolium chromophore
within its photogenerated isomer can be exploited to ensure the
transfer of energy from a complementary fluorophore. Thus, this
particular oxazine is a viable alternative to the nitrospiropyran
embedded in 2a for the realization of fast and stable
fluorophoreꢀphotochrome dyads.
5. FAST AND STABLE FLUOROPHOREꢀ
PHOTOCHROME DYADS
The slow switching speeds and poor fatigue resistance of
fluorophoreꢀphotochrome dyad 2a can be improved with the
introduction of an oxazine, similar to 19a, within this molecular
construct in place of the nitrospiropyran component.72 Indeed,
the 3H-indolium cation of ring-open isomer 19b absorbs in the
same range of wavelengths where the borondipyrromethene
fluorophore of 2a emits. As a result, this particular chromophore can
accept the excitation energy of the borondipyrromethene
fragment. In addition, the oxidation potential of this fluoro-
phore and the reduction potential of the hexafluorophosphate
salt of model 3H-indolium cation 21 indicate that the transfer of
one electron from the former to the latter upon excitation is
exergonic with a free-energy change of ꢀ0.5 eV. Thus, both
electron- and energy-transfer processes can quench the bor-
ondipyrromethene fluorescence, after the photoinduced open-
ing of the oxazine ring, if the two components are integrated
within the same molecular skeleton. On the basis of these
considerations, we designed fluorophoreꢀphotochrome dyads
22aꢀ24a (Figure 14) and synthesized these compounds in
three to eight steps from known precursors. These molecules
incorporate essentially the same fluorescent and photochromic
fragments but differ in the covalent bridge linking the two
functional components.
11779
dx.doi.org/10.1021/la201062h |Langmuir 2011, 27, 11773–11783

Langmuir
INVITED FEATURE ARTICLE
Figure 15. UV illumination of 25a, which opens the oxazine ring and
brings the coumarin fluorophore into conjugation with the 3H-indolium
cation of photogenerated isomer 25b.
Figure 14. Decrease in the emission intensity (0.01 mM, MeCN, 20 °C,
532 nm) of a solution of 23a when a UV source (355 nm, 6 ns, 12 mJ) is
turned on. The emission intensity returns to its original value when the
irradiation source is turned off.
By analogy to the spectroscopic response of 2a, the steady-
state absorption spectra of 22aꢀ24a are approximately the sum
of those of their separate fluorescent and photochromic
fragments.72 Once again, these observations are indicative of
the lack of electronic interactions between the fluorescent and
photochromic components within each dyad in S0. In all
instances, the 4-nitrophenoxy chromophore of the photochro-
mic fragment absorbs at 305 nm (λa in Table 1). Excitation
within this band opens the oxazine rings of 22a and 24a but not
that of 23a, with quantum yields of 0.05 and 0.02, respectively
(φ in Table 1). The process is accompanied by the appearance of
the S0 absorptions of the 4-nitrophenolate and 3H-indolium
chromophores of ring-open isomers 22b and 24b at 430 and
560 nm, respectively (λb in Table 1). Both bands decay mono-
exponentially with the reisomerization of these species back to
the original ring-closed isomers. Curve fittings of the temporal
absorbance evolutions indicate the lifetimes of 22b and 24b to be
1 and 3 μs, respectively (τ in Table 1). In fact, the behavior of
both fluorophoreꢀphotochrome dyads upon ultraviolet excita-
tion is almost identical to that of the photochrome 19a. Thus, the
photochromism of the oxazine is essentially unaffected by the
covalent attachment of a borondipyrromethene appendage in
these particular dyads. In contrast to the behavior of 22a and 24a,
however, the other dyad, 23a, does not open its oxazine ring in
response to ultraviolet excitation, under otherwise identical
conditions. In this instance, the corresponding time-resolved
absorption spectrum does not reveal any significant absorbance
change in the range of wavelengths where the anionic and
cationic fragments of zwitterionic isomer 23b are expected to
absorb.
On the basis of the spectral overlap and redox potentials, the
3H-indolium chromophore of photogenerated isomers 22b and
24b can quench the fluorescence of the borondipyrromethene
component in both dyads.72 Indeed, the emission intensity of the
two fluorophoreꢀphotochrome assemblies decreases with the
photoinduced ring opening of the photochromic fragment and
reverts to the original value in a few microseconds with its
thermal ring closing. In fact, their fluorescence can be modulated
for hundreds of switching cycles with no sign of degradation
simply by turning a beam at 355 nm on and off to operate the
photochromic component while illuminating the sample at
532 nm to excite the fluorescent component (Figure 14). None-
theless, the limited quantum efficiency for the photochromic
Figure 16. Steady-state (a, 2.5 μM) and time-resolved (b, 0.01 mM,
355 nm, 6 ns, 10 mJ, 0.03 μs) absorption spectra of solutions (MeCN,
20 °C) of 25a showing S0 absorptions at 412 and 570 nm respectively for
the coumarin fluorophore. The emission spectra of solutions (MeCN,
20 °C) of 25a (c, 0.01 mM), recorded upon simultaneous illumination
at 355 nm (6 ns, 10 mJ) and 532 nm (6 ns, 30 mJ), and of the
hexafluorophosphate salt of model 3H-indolium cation 26 (d, 2.5 μM),
recorded upon excitation at 593 nm, show essentially the same band for
the coumarin fluorophore.
transformation translates into modest contrast ratios at moderate
illumination intensities for both systems.
In search of strategies to enhance the contrast ratio of our
photoswitchable fluorophores, we realized that the mecha-
nism designed into 13aꢀ19a to control color can be adapted
to regulate fluorescence as well. Specifically, the ability of a
fluorescent fragment, attached to the chiral center at the junction
of the two heterocycles, to absorb exciting radiation in the visible
region can be controlled by opening and closing the oxazine ring.
In turn, the associated change in absorbance at the excitation
wavelength of the fluorophore translates into a change in
emission intensity. On the basis of these considerations, we
designed fluorophoreꢀphotochrome dyad 25a (Figure 15).73
This compound combines a coumarin fluorophore and an
oxazine photochrome into its molecular backbone and can be
isolated with a yield of 40% after the condensation of 20a with a
formylated coumarin, under the assistance of trifluoroacetic acid.
The steady-state absorption spectrum (a in Figure 16) of 25a
shows a band at 412 nm (λa in Table 1) for the coumarin
appendage attached to the oxazine ring.73 Upon ultraviolet
illumination, 25a switches to 25b (Figure 15) with a quantum
yield of 0.02 (φ in Table 1) and a concomitant bathochromic
11780
dx.doi.org/10.1021/la201062h |Langmuir 2011, 27, 11773–11783

Langmuir
INVITED FEATURE ARTICLE
shift of the absorption of the coumarin fluorophore. Indeed, the
ring-opening process brings the fluorescent appendage into
conjugation with the 3H-indolium cation and shifts its band to
570 nm (λb in Table 1 and b in Figure 16). Photogenerated
isomer 25b reverts to ring-closed species 25a with first-order
kinetics on a microsecond timescale. As a result, the absorbance
at 570 nm decays monoexponentially with the reisomerization
process (Figure S4). Curve fitting of the absorbance evolution
indicates the lifetime of 25b to be 0.2 μs (τ in Table 1).
The pronounced bathochromic shift that accompanies the
photoinduced transformation of ring-closed isomer 25a to ring-
open species 25b can be exploited to activate fluorescence with a
virtually infinite contrast ratio.73 Indeed, the illumination of 25a
at 532 nm does not result in any significant fluorescence, simply
because the coumarin appendage of the ring-closed isomer
cannot absorb at this particular wavelength. After ring opening,
however, the coumarin band shifts sufficiently to permit the
absorption of the exciting radiation at 532 nm. Consistently, the
emission spectrum (c in Figure 16), recorded upon simultaneous
irradiation of the sample at 355 nm to open the oxazine ring and
at 532 nm to excite the coumarin fluorophore, shows a band at
650 nm. This band resembles that detected in the steady-state
emission spectrum (d in Figure 16) of the hexafluorophosphate
salt of model 3H-indolium cation 26 and corresponds to the
fluorescence of the coumarin fragment of 25b. In fact, the
fluorescence of this particular fluorophoreꢀphotochrome dyad
can be switched on and off for hundreds of cycles with no sign of
degradation simply by turning an irradiation source at 355 nm on
and off while illuminating the sample at 532 nm.
microsecond timescale for hundreds of switching cycles simply
by opening and closing the oxazine ring under optical control.
However, the ratio between the emission intensities of the two
interconverting states, achieved at moderate illumination inten-
sities, is relatively small because of the modest quantum efficiency
of the photochromic transformation. The structural design of the
fluorophoreꢀphotochrome dyad can be adjusted, yet again, to
photoactivate, rather than photodeactivate, fluorescence and
ensure a virtually infinite contrast ratio. In particular, the photo-
induced generation of a chromophoric fragment that is able to
absorb in the visible region can also be exploited to activate
fluorescence, if the very same fragment is designed to be
fluorescent. Specifically, the covalent connection of a coumarin
fluorophore to an oxazine ring allows the photoinduced activa-
tion of fluorescence with a pair of beams designed to switch the
photochromic component and excite the fluorescent one. Under
these conditions, fluorescence turns on against a dark back-
ground and can be modulated on a microsecond timescale for
hundreds of switching cycles. In principle, this unique behavior
can be exploited to overcome diffraction with appropriate multi-
photon illumination protocols and temporally resolve spatially
indistinguishable fluorophores. Thus, the photochemical and
photophysical properties of our compounds, coupled to their
synthetic accessibility, can ultimately lead to the development of
an entire family of photoswitchable fluorophores for super-
resolution imaging. However, a further understanding of their
excitation dynamics is essential to identifying viable strategies to
enhance the quantum efficiency of the photoinduced ring-open-
ing process, the brightness of the fluorescent state, and the ratio
between the emission intensities of the two interconverting
isomers. In this context, ultrafast spectroscopic analyses and
electronic structure calculations will contribute to the elucidation
of the stereoelectronic factors governing the properties of our
molecular switches and guide our future structural designs.
6. CONCLUSIONS
The photoinduced conversion of a nitrospiropyran to the
corresponding merocyanine can be exploited to switch off the
fluorescence of a covalently connected borondipyrromethene
and then turn it back on after thermal reisomerization. Indeed,
the photoinduced transformation of the photochromic compo-
nent activates electron- and energy-transfer pathways and
results in the effective quenching of the excited fluorophore.
The emission photodeactivation in such a fluorophoreꢀ
photochrome dyad permits the spatial confinement of fluores-
cence with patterned illumination and, in principle, offers the
opportunity to reconstruct images with subdiffraction resolution.
Nonetheless, the slow reisomerization kinetics of the photo-
chromic component and its tendency to degrade under illumina-
tion impose a poor switching speed and limited fatigue resistance.
Both parameters can be improved dramatically simply by elim-
inating the need for the cis/trans isomerization steps that
accompany the nitrospiropyran/merocyanine interconversion
and relying instead solely on the intrinsically fast opening and
closing of an oxazine ring. In fact, these structural modifications
decrease the time required to complete a full switching cycle from
hundreds of seconds to a few nanoseconds and increase the
number of switching cycles, tolerated by the photochromic
system, from a few tens to several thousands. Furthermore, the
attachment of an appropriate chromophoric appendage to the
oxazine heterocycle translates into the appearance of an intense
absorption band in the visible region upon photoinduced ring
opening. This absorption can be engineered to overlap the
emission of a covalently connected borondipyrromethene and
encourage the transfer of energy upon excitation. As a result, this
structural design allows the modulation of fluorescence on the
’ ASSOCIATED CONTENT
S
Supporting Information. Steady-state absorption spec-
b
tra of 8aꢀ12a. Time-resolved absorption spectra of 8a and
10aꢀ12a. Temporal absorbance evolutions of 19a and 25a upon
excitation. This material is available free of charge via the Internet
at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: ssortino@unict.it; fraymo@miami.edu.
’ ACKNOWLEDGMENT
We thank the National Science Foundation (CAREER Award
CHE-0237578 and CHE-0749840) and the University of Miami
for financial support.
’ REFERENCES
(1) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Springer:
New York, 2006.
(2) Murphy, D. B. Fundamentals of Light Microscopy and Electronic
Imaging; Wiley-Liss: New York, 2001.
(3) Pawley, J. B., Ed.; Handbook of Biological Confocal Microscopy;
Springer: New York, 2006.
11781
dx.doi.org/10.1021/la201062h |Langmuir 2011, 27, 11773–11783

Langmuir
INVITED FEATURE ARTICLE
(4) Born, M.; Wolf, E. Principles of Optics; Cambridge University
Press: Cambridge, U.K., 2002.
(5) Fernꢀandez-Suꢀarez, M.; Ting, A. Y. Nat. Rev. Mol. Cell. Biol. 2008,
9, 929–943.
(6) Huang, B.; Bates, M.; Zhuang, X. Annu. Rev. Biochem. 2009,
78, 993–1016.
(37) (a) Bertelson, R. C. In Photochromism; Brown, G. H., Ed.;
Wiley: New York, 1971; pp 45ꢀ431. (b) Bertelson, R. C. In Organic
Photochromic and Thermochromic Compounds; Crano, J. C., Guglielmetti,
R., Eds.; Plenum Press: New York, 1999; Vol. 1, pp 11ꢀ83.
(38) Kholmanskii, A. S.; Dyumanev, K. M. Russ. Chem. Rev. 1987,
56, 136–151.
(7) Hell, S. W. Nat. Methods 2009, 6, 24–32.
(8) Heilemann, M.; Dedecker, P.; Hofkens, J.; Sauer, M. Laser Photon
Rev. 2009, 3, 180–202.
(9) Toomre, D.; Bewersdorf, J. Annu. Rev. Cell Dev. Biol. 2010,
26, 285–314.
(10) Vogelsang, J.; Steinhauer, C.; Forthmann, C.; Stein, I. H.;
Person-Skergo, B.; Cordes, T.; Tinnefeld, P. ChemPhysChem 2010,
11, 2475–2490.
(11) Thompson, M. A.; Biteen, J. S.; Lord, S. J.; Conley, N.;
Moerner, W. E. Methods Enzymol. 2010, 475, 27–59.
(12) Cusido, J.; Impellizzeri, S.; Raymo, F. M. Nanoscale 2011,
3, 59–70.
(13) Dorion, G. H.; Wiebe, A. F. Photochromism; Focal Press:
New York, 1970.
(14) Brown, G. H., Ed. Photochromism; Wiley: New York, 1971.
(15) El’tsov, A. V., Ed. Organic Photochromes; Consultants Bureau:
New York, 1990.
(39) Guglielmetti, R. In Photochromism: Molecules and Systems; D€urr,
H., Bouas-Laurent, H., Eds.; Elsevier: Amsterdam, 1990; pp 314ꢀ466
and 855ꢀ878.
(40) (a) Ichimura, K. In Photochromism: Molecules and Systems; D€urr,
H., Bouas-Laurent, H., Eds.; Elsevier: Amsterdam, 1990; pp 903ꢀ918.
(b) Ichimura, K.; Seki, T.; Kawamishi, Y.; Suzuki, Y.; Sakuragi, M.;
Tamaki, T. In Photo-Refractive Materials for Ultrathin Density Optical
Memory; Irie, M., Ed.; Elsevier: Amsterdam, 1994; pp 55ꢀ83.
(c) Ichimura, K. In Photochromic Polymers; Crano, J. C., Guglielmetti, R.,
Eds.; Plenum Press: New York, 1999; Vol. 2, pp 9ꢀ63. (d) Ichimura,
K. Chem. Rev. 2000, 100, 1847–1873.
(41) Crano, J. C.; Kwak, W. S.; Welch, C. N. In Applied Photochromic
Polymer Systems; McArdle, C. B., Ed.; Blackie: Glasgow, 1992; pp
31ꢀ79.
(42) (a) Krongauz, V. In Applied Photochromic Polymer Systems;
McArdle, C. B., Ed.; Blackie: Glasgow, 1992; pp 121ꢀ173. (b) Berkovic,
G.; Krongauz, V.; Weiss, V. Chem. Rev. 2000, 100, 1741–1754.
(43) Irie, M. In Applied Photochromic Polymer Systems; McArdle,
C. B., Ed.; Blackie: Glasgow, 1992; pp 174ꢀ206.
(16) Bouas-Laurent, H., D€urr, H., Eds. Photochromism: Molecules and
Systems; Elsevier: Amsterdam, 1990.
(17) Crano, J. C., Guglielmetti, R., Eds. Organic Photochromic and
Thermochromic Compounds; Plenum Press: New York, 1999.
(18) Irie, M. Chem. Rev. 2000, 100, 1683–1890.
(19) Kuz’min, M. G.; Koz’menko, M. V. In Organic Photochromes;
El’tsov, A. V., Ed.; Consultants Bureau: New York, 1990; pp 245ꢀ265.
(20) (a) Raymo, F. M.; Tomasulo, M. Chem. Soc. Rev. 2005,
34, 327–336. (b) Raymo, F. M.; Tomasulo, M. J. Phys. Chem. A 2005,
109, 7343–7352. (c) Cusido, J.; Deniz, E.; Raymo, F. M. Eur. J. Org.
Chem. 2009, 2031–2045. (d) Yildiz, I.; Deniz, E.; Raymo, F. M. Chem.
Soc. Rev. 2009, 38, 1859–1867.
(21) Gross, H.; D€urr, H.; Retting, W. J. Photochem. 1984, 26,
165–178.
(22) G€orner, H.; Fischer, C.; Gierisch, S.; Daub, J. J. Phys. Chem.
1993, 97, 4110–4117.
(23) Weber, C.; Rustemeyer, F.; D€urr, H. Adv. Mater. 1998, 10,
1348–1351.
(24) Sheepwash, M.; Mitchell, R. H.; Bohne, C. J. Am. Chem. Soc.
2002, 124, 4693–4700.
(25) Liang, Y.; Dvornikov, A. S.; Rentzepis, P. M. J. Mater. Chem.
2003, 13, 286–290.
(26) Myles, A. J.; Branda, N. R. J. Am. Chem. Soc. 2001, 123,
177–178.
(27) Straight, S. D.; Andrꢀeasson, J.; Kodis, G.; Bandyopadhyay, S.;
Mitchell, R. H.; Moore, T. A.; Moore, A. L.; Gust, D. J. Am. Chem. Soc.
2005, 127, 9403–9409.
(44) Hibino, J.; Hashida, T.; Suzuki, M. In Photo-Refractive Materials
for Ultrathin Density Optical Memory; Irie, M., Ed.; Elsevier: Amsterdam,
1994; pp 25ꢀ53.
(45) Kawata, S.; Kawata, Y. Chem. Rev. 2000, 100, 1777–1788.
(46) Delaire, J. A.; Nakatani, K. Chem. Rev. 2000, 100, 1817–1845.
(47) Tamai, N.; Miyasaka, H. Chem. Rev. 2000, 100, 1875–1890.
(48) Minkin, V. I. Chem. Rev. 2004, 104, 2751–2776.
(49) (a) Tomasulo, M.; Deniz, E.; Alvarado, R. J.; Raymo, F. M.
J. Phys. Chem. C 2008, 112, 8038–8045. (b) Deniz, E.; Tomasulo, M.;
DeFazio, R. A.; Watson, B. D.; Raymo, F. M. Phys. Chem. Chem. Phys.
2010, 12, 11630–11634.
(50) (a) Atabekyan, L. S.; Chibisov, A. K. Zh. Prikl. Spektrosk. 1987,
46, 651–654. (b) G€orner, H.; Atabekyan, L. S.; Chibisov, A. K. Chem.
Phys. Lett. 1996, 260, 59–64. (c) G€orner, H. Chem. Phys. 1997,
222, 315–329. (d) Chibisov, A. K.; G€orner, H. J. Photochem. Photobiol.,
A 1997, 105, 261–267. (e) Chibisov, A. K.; G€orner, H. J. Phys. Chem. A
1997, 101, 4305–4312. (f) G€orner, H. Chem. Phys. Lett. 1998,
282, 381–390. (g) G€orner, H. Phys. Chem. Chem. Phys. 2001, 3, 416–423.
(51) (a) Krysanov, S. A.; Alfimov, M. V. Chem. Phys. Lett. 1982,
91, 77–80. (b) Krysanov, S. A.; Alfimov, M. V. Laser Chem. 1984, 4,
129–138.
(52) Kalisky, Y.; Orlowski, T. E.; Williams, D. J. J. Phys. Chem. 1983,
87, 5333–5338.
(53) (a) Ito, T.; Hiramatsu, M.; Hirano, I.; Ohtani, H. Macromole-
cules 1990, 23, 4528–4532. (b) Ito, T.; Hiramatsu, M.; Hosoda, M.;
Tsuchiya, Y. Rev. Sci. Instrum. 1991, 62, 1415–1419.
(28) Liu, L.-H.; Nakatani, K.; Pansu, R.; Vachon, J.-J.; Tauc, P.;
Ishow, E. Adv. Mater. 2007, 19, 433–436.
(54) Ernsting, N. P.; Arthen-Engeland, T. J. Phys. Chem. 1991,
(29) Berberich, M.; Krause, A.-M.; Orlandi, M.; Scandola, F.;
W€urthner, F. Angew. Chem., Int. Ed. 2008, 47, 6616–6619.
(30) Bahr, J. L.; Kodis, G.; de la Garza, L.; Lin, S.; Moore, A. L.;
Moore, T. A.; Gust, D. J. Am. Chem. Soc. 2001, 123, 7124–7133.
(31) Myles, A.; Branda, N. R. Adv. Funct. Mater. 2002, 12, 167–173.
(32) Giordano, L.; Jovin, T. M.; Irie, M.; Jares-Erijman, E. A. J. Am.
Chem. Soc. 2002, 124, 7481–7489.
(33) Harbron, E. J.; Vicente, D. A.; Hoyt, M. T. J. Phys. Chem. B
2004, 108, 18789–18792.
(34) Bossi, M.; Belov, V.; Polyakova, S.; Hell, S. W. Angew. Chem.,
Int. Ed. 2006, 45, 7462–7465.
(35) Jukes, R. T.; Adamo, V.; Hartl, F.; Belser, P.; De Cola, L. Inorg.
Chem. 2004, 43, 2779–2792.
(36) The operating principles for fluorescence switching with
photochromic compounds are reviewed in ref 20. Representative
examples of these mechanisms in action are reported in refs 21ꢀ35.
95, 5502–5509.
(55) Quenching and sensitization experiments, coupled to the
investigation of the photoinduced reisomerization of the colored
isomers, demonstrate the fundamental role of T1 in the photoisomeriza-
tion of nitrospiropyrans.50 Earlier reports, however, suggest alternative
mechanisms for these photoinduced transformations. See refs
51ꢀ53 and (a) Irie, M.; Menju, A.; Hayashi, K. Macromolecules 1979,
12, 1176–1180. (b) Krongauz, V.; Kiwi, J.; Gr€atzel, M. J. Photochem.
1980, 13, 89–97. (c) Kalisky, Y.; Williams, D. J. Macromolecules 1984,
17, 292–296. (d) Lenoble, C.; Becker, R. S. J. Phys. Chem. 1986, 90,
62–65. (e) Tamaki, T.; Sakuragi, M.; Ichimura, K.; Aoki, K. Chem. Phys.
Lett. 1989, 161, 23–26.
(56) Malkin, Ya. N.; Krasieva, T. B.; Kuzmin, V. A. J. Photochem.
Photobiol., A 1989, 49, 75–88.
(57) Sakuragi, M.; Aoki, K.; Tamaki, T.; Ichimura, K. Bull. Chem. Soc.
Jpn. 1990, 63, 74–79.
11782
dx.doi.org/10.1021/la201062h |Langmuir 2011, 27, 11773–11783

Langmuir
INVITED FEATURE ARTICLE
(58) Salemi, C.; Giusti, G.; Guglielmetti, R. J. Photochem. Photobiol.,
A 1995, 86, 247–252.
(59) Sheng, Y.; Leszcynski, J.; Garcia, A. A.; Rosario, R.; Gust, D.;
Springer, J. J. Phys. Chem. B 2004, 108, 16233–16243.
(60) (a) Raymo, F. M.; Giordani, S. J. Am. Chem. Soc. 2001,
123, 4651–4652. (b) Raymo, F. M.; Giordani, S.; White, A. J. P.;
Williams, D. J. J. Org. Chem. 2003, 68, 4158–4169.
(61) (a) Tomasulo, M.; Sortino, S.; Raymo, F. M. Org. Lett. 2005,
7, 1109–1112. (b) Tomasulo, M.; Sortino, S.; White, A. J. P.; Raymo,
F. M. J. Org. Chem. 2005, 70, 8180–8189. (c) Tomasulo, M.; Sortino, S.;
Raymo, F. M. Asian Chem. Lett. 2007, 11, 219–222.
(71) The attachment of the 2-(4-(2-phenylethynyl)phenyl)ethynyl
substituent of 18a to the para position, relative to the nitrogen atom, of
the 2H,3H-indole fragment rather than to the chiral center at the
junction of the two heterocycles prevents ring opening.65b Instead of
the S0 absorption of the 4-nitrophenolate anion of the ring-open isomer,
the time-resolved absorption spectrum of the resulting oxazine reveals a
band at 510 nm. This band resembles the T1 absorption of phenylvinyl-
stilbene (Hara, M.; Samori, S.; Xichen, C.; Fujitsuka, M.; Majima, T.
J. Org. Chem. 2005, 70, 4370ꢀ4374) and decays monoexponentially on a
microsecond timescale. Thus, intersystem crossing dominates the
excitation dynamics of this particular system.
(62) In principle, the nitro group of 6a can facilitate intersystem
crossing, as observed for 1a. Nonetheless, the lifetime of T1 associated
with the 4-nitrophenoxy chromophore of 6a, if at all formed, is
presumably too short to sensitize significant amounts of singlet oxygen.
Indeed, T1 of 4-nitroanisole has a subnanosecond lifetime in acetonitrile
(Mir, M.; Jansen, L. M. G.; Wilkinson, F.; Bourdelande, J. L.; Marquet, J.
J. Photochem. Photobiol., A 1998, 113, 113–117).
(72) (a) Deniz, E.; Sortino, S.; Raymo, F. M. J. Phys. Chem. Lett.
2010, 1, 1690–1693. (b) Deniz, E.; Ray, S.; Tomasulo, M.; Impellizzeri,
S.; Sortino, S.; Raymo, F. M. J. Phys. Chem. A 2010, 114, 11567–11575.
(73) Deniz, E.; Sortino, S.; Raymo, F. M. J. Phys. Chem. Lett. 2010,
1, 3506–3509.
(63) Photochromic oxazine 6a can be trapped within poly(methyl
methacrylate) films61b and attached covalently to hydrophobic
(Tomasulo, M.; Deniz, E.; Benelli, T.; Sortino, S.; Raymo, F. M. Adv.
Funct. Mater. 2009, 19, 3956ꢀ3961) or hydrophilic (Tomasulo, M.;
Deniz, E.; Sortino, S.; Raymo, F. M. Photochem. Photobiol. Sci. 2010, 9,
136ꢀ140) polymers. The photochemical properties of the resulting
materials are similar to those observed for 6a in acetonitrile. However,
the S0 absorption, associated with the 4-nitrophenolate chromophore of
the photogenerated species, decays biexponentially on a microsecond
timescale under these conditions.
(64) (a) Tomasulo, M.; Raymo, F. M. Org. Lett. 2005, 7, 4633–4636.
(b) Tomasulo, M.; Sortino, S.; White, A. J. P.; Raymo, F. M. J. Org. Chem.
2006, 71, 744–753. (c) Åxman Petersen, M.; Deniz, E.; Brøndsted Nielsen,
M.; Sortino, S.; Raymo, F. M. Eur. J. Org. Chem. 2009, 4333–4339.
(65) (a) Tomasulo, M.; Sortino, S.; Raymo, F. M. Adv. Mater. 2008,
20, 832–835. (b) Tomasulo, M.; Sortino, S.; Raymo, F. M. J. Org. Chem.
2008, 73, 118–126. (c) Tomasulo, M.; Sortino, S.; Raymo, F. M.
J. Photochem. Photobiol., A 2008, 200, 44–49. (d) Deniz, E.; Tomasulo,
M.; Sortino, S.; Raymo, F. M. J. Phys. Chem. C 2009, 113, 8491–8497.
(66) Oxazine 20a switches to corresponding zwitterionic isomer
20b upon ultraviolet illumination.61b,61c The quantum yield for the
photoinduced ring-opening process is 0.03 (φ in Table 1), and the
lifetime of the photogenerated isomer is 22 ns (τ in Table 1). The
photochemical properties of this compound are also reported in
Barkauskas, M.; Martynaitis, V.; Shachkus, A. A.; Rotomskis, R.;
Sirutkaitis, V.; Vengris, M. Lith. J. Phys. 2008, 48, 231–242.
(67) The synthesis of 20a was originally reported in (a) Shachkus,
A. A.; Degutis, J.; Jezerskaite, A. In Chemistry of Heterocyclic Compounds;
Kovac, J., Zalupsky, P., Eds.; Elsevier: Amsterdam, 1987; Vol. 35,
pp 518ꢀ520. (b) Shachkus, A. A.; Degutis, J.; Urbonavichyus, A. G.
Khim. Geterotskil. Soed. 1989, 5, 672–676. (c) Chem. Heterocycl. Compd.
1989, 562–565.
(68) The introduction of an electron-withdrawing fluorine substi-
tuent in the para position, relative to the nitrogen atom, of the 2H,3H-
indole fragment of 6a65d and 19a (Deniz, E.; Impellizzeri, S.; Sortino, S.;
Raymo, F. M. Can. J. Chem. 2011, 89, 110ꢀ116) enhances the quantum
efficiency of the photoinduced ring-opening process. By contrast, the
attachment of an electron-donating methoxy group in the very same
position prevents the photochemical transformation. These observa-
tions indicate that the quantum efficiency of the photoinduced ring-
opening process is sensitive to the electron density on the nitrogen atom
of the 2H,3H-indole fragment. Presumably, the transfer of one electron
from this atom to the nitro group on the phenoxy chromophore upon
excitation competes with the cleavage of the [CꢀO] bond required to
open the oxazine ring.
(69) The S0 f S1 absorption of the 3H-indolium cation of 14b is
centered at ca. 340 nm.65d
(70) The coloration efficiency of 15b is lower than that of parent
system 6b because of the poor quantum yield for its photoinduced
formation.65d
11783
dx.doi.org/10.1021/la201062h |Langmuir 2011, 27, 11773–11783