Autocatalytic fluorescence photoactivation

Article abstract of DOI:10.1021/ja5068383

We designed an autocatalytic photochemical reaction based on the photoinduced cleavage of an α-diketone bridge from the central phenylene ring of a fluorescent anthracene derivative. The product of this photochemical transformation sensitizes its own formation from the reactant, under illumination at a wavelength capable of exciting both species. Specifically, the initial and direct excitation of the reactant generates the product in the ground state. The subsequent excitation of the latter species results in the transfer of energy to another molecule of the former to establish an autocatalytic loop. Comparison of the behavior of this photoactivatable fluorophore with that of a model system and the influence of dilution on the reaction progress demonstrates that the spectral overlap between the emission of the product and the absorption of the reactant together with their physical separation govern autocatalysis. Indeed, both parameters control the efficiency of the resonant transfer of energy that is responsible for establishing the autocatalytic loop. Furthermore, the proximity of silver nanoparticles to reactant and product increases the energy-transfer efficiency with a concomitant acceleration of the autocatalytic process. Thus, this particular mechanism to establish sensitization offers the opportunity to exploit the plasmonic effects associated with metallic nanostructures to boost photochemical autocatalysis.

Full text of DOI:10.1021/ja5068383

Article
pubs.acs.org/JACS
Autocatalytic Fluorescence Photoactivation
Ek Raj Thapaliya, Subramani Swaminathan, Burjor Captain, and Franci̧ sco M. Raymo*
Laboratory for Molecular Photonics, Department of Chemistry, University of Miami, 1301 Memorial Drive, Coral Gables, Florida
33146-0431, United States
S
* Supporting Information
ABSTRACT: We designed an autocatalytic photochemical reaction based on the photo-
induced cleavage of an α-diketone bridge from the central phenylene ring of a fluorescent
anthracene derivative. The product of this photochemical transformation sensitizes its own
formation from the reactant, under illumination at a wavelength capable of exciting both
species. Specifically, the initial and direct excitation of the reactant generates the product in the
ground state. The subsequent excitation of the latter species results in the transfer of energy to
another molecule of the former to establish an autocatalytic loop. Comparison of the behavior
of this photoactivatable fluorophore with that of a model system and the influence of dilution
on the reaction progress demonstrates that the spectral overlap between the emission of the
product and the absorption of the reactant together with their physical separation govern
autocatalysis. Indeed, both parameters control the efficiency of the resonant transfer of energy
that is responsible for establishing the autocatalytic loop. Furthermore, the proximity of silver
nanoparticles to reactant and product increases the energy-transfer efficiency with a
concomitant acceleration of the autocatalytic process. Thus, this particular mechanism to establish sensitization offers the
opportunity to exploit the plasmonic effects associated with metallic nanostructures to boost photochemical autocatalysis.
such a spatiotemporal control permits the monitoring of
dynamic events in real time13 and the visualization of samples
with subdiffraction resolution.14 Indeed, photoactivatable
fluorophores are becoming invaluable probes for the inves-
tigation of the dynamic and structural properties of a diversity
of specimens.
INTRODUCTION
■
Replicating molecules catalyze their own formation from
appropriate precursors.1 In most instances, supramolecular
contacts between product and reactants template the assembly
of an identical copy of the former from the latter species. Under
these conditions, the reaction rate increases, during the course
of the chemical transformation, to impose a sigmoidal temporal
dependence on the concentration of the replicating product.2
Such a kinetic amplification is responsible for a diversity of
biochemical processes and is believed to have played a
fundamental role in the origin of life.3 In fact, a number of
artificial counterparts to replicating biomolecules have been
developed already with the ultimate goal of elucidating the basic
factors responsible for autocatalysis.1 Photochemical analogues
of supramolecular replicators remain, instead, limited to a few
remarkable examples, mostly aimed at signaling molecular
recognition with changes in fluorescence intensity.4−10 Indeed,
the unique kinetic profile of autocatalytic reactions can be
exploited to impose the amplification on fluorescence signaling
that is particularly convenient for chemical sensing.11
Photoactivatable fluorophores switch from a nonemissive to
an emissive state under irradiation at an appropriate activation
wavelength (λAc).12 Illumination of the resulting product at a
given excitation wavelength (λEx) then produces significant
fluorescence. The concatenation of a photochemical reaction
(activation) with a photophysical process (fluorescence) is
therefore responsible for the operating principles of these
photoresponsive compounds. Their unique behavior, in
combination with the interplay of beams illuminating at λAc
and λEx, can be exploited to switch fluorescence on within a
defined region of space at a particular interval of time. In turn,
In the wake of our research efforts directed at the
identification of viable structural designs to photoactivate
fluorescence,12−14 we realized that the photochemical and
photophysical events responsible for switching emission on can
be manipulated to engineer an autocatalytic transformation.
Specifically, the structures of the nonemissive reactant and
fluorescent product can be selected to permit excitation of both
at the same wavelength (λAc = λEx) as well as to ensure a
significant overlap between the absorption spectrum of the
former and the emission spectrum of the latter. Under these
conditions, irradiation would excite the reactant to generate the
product first and then excite the product to sensitize the
excitation of another reactant, establishing an autocatalytic
loop. Indeed, either the resonant transfer of the excitation
energy of the product to a proximal reactant or the
reabsorption of its fluorescence by a distal reactant would
translate into sensitization. These operating principles are
reminiscent of the energy cascades governing quantum
amplification in certain photochemical processes.15−20 In such
chain reactions, the excitation of the reactant eventually
generates the product in an excited state. The resulting species
can then transfer energy to another reactant molecule, ensuring
Received: July 7, 2014
Published: September 10, 2014
© 2014 American Chemical Society
13798
dx.doi.org/10.1021/ja5068383 | J. Am. Chem. Soc. 2014, 136, 13798−13804

Journal of the American Chemical Society
Article
the propagation of the chain reaction, and decay to the ground
state. The overall result is that the absorption of a single photon
by the reactant can produce multiple copies of the product,
leading to quantum amplification. In these processes, however,
the product sensitizes its own formation prior to populating its
ground state, and therefore, the concentration of its ground
state has no influence on the rate of the overall transformation.
According to our design logic, instead, the product is formed in
the ground state first and then it is excited directly to sensitize
its own formation. Thus, while the absorption of multiple
photons is still required to form multiple copies of the product,
the increasing concentration of the latter in the ground state,
during the course of the reaction, can accelerate the
photochemical conversion and lead to autocatalysis. Further-
more, such a mechanism for the realization of a photochemical
replicator could be accelerated even further with the aid of
plasmonic effects.21 In fact, the enhancement of the electro-
magnetic field in close proximity to the surface of illuminated
metallic nanostructures can increase the efficiencies of photo-
chemical reactions22 and photophysical processes.23 Therefore,
the conversion of reactant into product and the emission of the
latter could both be promoted with metallic nanostructures to
boost the overall autocatalytic process.
The introduction of an α-diketone bridge across positions 9
and 10 of anthracene isolates electronically the two peripheral
phenylene rings of the oligoacene platform.24 The absorption
spectrum of the resulting adduct shows a broad and weak band
in the visible region for a n → π* transition of the α-diketone
chromophore.25 Illumination at wavelengths within this band
results in the cleavage of the α-diketone bridge with the release
of two molecules of carbon monoxide and the regeneration of
anthracene.26 In fact, the photoinduced cleavage of similar α-
diketone adducts has been employed already to prepare
photochemically several oligoacenes27 and, in a few instances,
to activate fluorescence.28,29 Furthermore, these photochemical
reactions can be performed efficiently within rigid matrices to
avoid the dimerization and oxidation of the resulting
oligoacenes that often accompany these transformations in
solution and, hence, permit the isolation of otherwise elusive
polycyclic aromatic hydrocarbons.30,31 These literature prece-
dents suggest that an anthracene derivative, capable of emitting
in the same spectral region where the α-diketone bridge
absorbs, should sensitize its own formation from the
corresponding adduct upon excitation. On the basis of such
considerations, we designed an α-diketone adduct able to
satisfy these stringent spectral requirements as well as identified
a control system with minimal spectral overlap instead. In this
article, we report the synthesis of the former compound, the
structural characterization of both adducts together with the
investigation of their photochemical and photophysical proper-
ties and the influence of silver nanoparticles on the behavior of
the former species.
Figure 1. Photoinduced decarbonylation of 1 and 2 to produce 3 and
4, respectively.
Adduct 1 can be prepared from 3 in three synthetic steps,
following a literature protocol.32 The very same protocol can be
adapted to convert 4 into 2 (Figure 2). Specifically, treatment
of 4 with vinylene carbonate results in the bridging of positions
9 and 10 of the anthracene core with the formation of 5. Basic
hydrolysis of 5 and then oxidation of 6 convert the
corresponding bridging unit into a photocleavable α-diketone
in the shape of 2. The overall yield for the three consecutive
synthetic steps is 26%.
The structural identities of 2, 5 and 6 were confirmed by
electrospray ionization mass spectrometry (ESIMS) together
with 1H and 13C nuclear magnetic resonance (NMR)
spectroscopies (Figures S1−S3, Supporting Information (SI)).
In addition, single crystals of 1 and 2 suitable for X-ray
diffraction analysis (Table S1 (SI)) were obtained after
diffusion of hexane vapors into benzene solutions of the
compounds. The resulting structures (Figure 3) clearly reveal
that the α-diketone bridge, between positions 9 and 10, forces
the three fused rings out of planarity and interrupts electronic
conjugation across them.
Photochemical and Photophysical Properties. The
absorption spectra of 1 and 2 (a and b in Figure 4) reveal a
broad and weak band centered at 462 nm (λAb in Table 1) for a
n → π* transition of the α-diketone bridge. Upon visible
illumination, the characteristic vibronic structure of the
corresponding anthracene chromophores appear between 300
and 450 nm (Figure 4c,d). These absorptions are identical to
those observed in the spectra of 3 and 4 (Figure S4a,b (SI))
and demonstrate that the α-diketone bridge of 1 and 2 cleaves
under irradiation to restore the anthracene chromophore, in
agreement with literature precedents.24−31 Similar changes are
also evident in the corresponding emission spectra (e → g and f
→ h in Figure 4). Intense bands appear between 350 and 600
nm (λEm in Table 1) only after visible illumination of the
sample. Once again, the photogenerated bands are identical to
those observed in the spectra of 3 and 4 (Figure S4c,d (SI)).
Thus, the photoinduced conversion of nonemissive reactants 1
and 2 into emissive products 3 and 4, respectively, translates
into efficient fluorescence activation.
The photoinduced conversion of 1 and 2 into 3 and 4
respectively occurs in poly(butyl methacrylate) (PBMA) matrix
as well as in acetonitrile solution with essentially identical
spectral changes (Figures 4 and S5 (SI)). The polymer matrix,
however, is relatively rigid and prevents the diffusion of the
entrapped species. Fluorescence images (Figure S6a,b (SI)) of a
polymer film doped with 2 reveal significant emission only after
visible illumination. Specifically, the irradiation of a circular area
at the center of the imaging field with a laser operating at 458
nm switches 2 to 4 exclusively in the illuminated area with the
concomitant appearance of intense fluorescence. The very same
RESULTS AND DISCUSSION
■
Design, Synthesis and Structural Characterization.
Adducts 1 and 2 differ in the nature of the group (R in Figure
1) on their two ortho-phenylene rings and, upon illumination,
are supposed to generate 3 and 4 respectively. This particular
group is expected to have a negligible influence on the n → π*
absorption of the α-diketone bridge of 1 and 2, but a
pronounced effect on the emission of 3 and 4 and, hence, on
the spectral overlap between reactant and product of the two
photochemical reactions.
13799
dx.doi.org/10.1021/ja5068383 | J. Am. Chem. Soc. 2014, 136, 13798−13804

Journal of the American Chemical Society
Article
Figure 2. Three-step synthesis of 2 from 4.
Figure 3. ORTEP representations of the crystal structures of 1 and 2
(30 and 50% thermal ellipsoid probability, respectively).
fluorescent spot is also evident in an image (Figure S6c (SI))
recorded after as many as 10 min, confirming the lack of any
significant diffusion.
The inability of reactant and product to diffuse within the
polymer matrix offers the opportunity to induce and probe the
transformation of one into the other in a portion of the film
with the sole aid of the excitation source of an emission
spectrometer. In particular, the sequential acquisition of
emission spectra (Figure 5) of a PBMA film doped with 2
shows the gradual growth of the characteristic bands of 4.
These spectra were recorded under excitation at a wavelength
(390 nm) that matches the 0 → 2 absorption band (Figure 4d)
of the photogenerated anthracene chromophore. At this
wavelength, the absorbance of 2 (Figure 4b) is relatively
small, yet it is sufficient to initiate the photochemical generation
of 4. Once formed, 4 can absorb part of the incoming photons
and emit as a result. In turn, its emission can induce the
conversion of another molecule of 2 into a new molecule of 4
and establish an autocatalytic loop. Indeed, a plot (Figure 6a) of
the emission intensity, detected at 453 nm in the sequence of
spectra, shows the sigmoidal temporal dependence typical of
autocatalysis.2 The reaction accelerates significantly with an
increase in the product concentration. By contrast, the same
experiment performed with 1 does not reveal any change in
fluorescence over the same temporal scale (Figure 6b).33
Indeed, comparison of the emission spectra of 3 and 4 (Figure
4g,h) reveals optimal spectral overlap with the α-diketone
absorption of the corresponding adduct (Figure 4a,b) only for
the latter. In particular, the overlap integral (J in Table 1) for
the 2/4 pair is 22.4 × 10−16 M−1 cm−1, while it is only 8.3 ×
Figure 4. Absorption (a−d) and emission (e−h) spectra of PBMA
films, doped (8% w/w) with 1 (λEx = 350 nm) or 2 (λEx = 390 nm)
and spin coated on quartz slides, before (a, b, e and f) and after (c, d, g
and h) irradiation (420 nm, 2.3 mW cm−2, 300 s for 1 and 60 s for 2).
10−16 M−1 cm−1 for the 1/3 counterpart. These values
correspond to Forster distances (R in Table 1) of 26 and 18
̈
0
Å respectively and energy-transfer efficiencies (E in Table 1) of
25.8 and 0.3% respectively at a dopant concentration of 8% w/
w relative to the polymer.34 Thus, only one of the two
photochemical products can transfer efficiently its excitation
energy to the corresponding reactant and sensitize its own
formation in full agreement with the pronounced difference
between the photokinetic profiles of the two reactions (Figure
6a,b).35
The role of resonant energy transfer in imposing the
sigmoidal profile on the photoinduced conversion of 2 into 4 is
further confirmed by the effect of dilution on the reaction
course.36 Specifically, a 10-fold dilution of the sample increases
the time required to convert 50% of the reactant from 1450 to
3200 s (Figure 7a,b). Indeed, the average distance between
molecules elongates sufficiently with dilution to lower the
energy-transfer efficiency to only 0.01%. Consistently, fittings
(Figures S7 and S8 (SI)) of the corresponding sigmoidal plots
indicate a 35-fold decrease in the quantum yield for the
13800
dx.doi.org/10.1021/ja5068383 | J. Am. Chem. Soc. 2014, 136, 13798−13804

Journal of the American Chemical Society
Article
Table 1. Photochemical and Photophysical Parameters
a
d
d
e
λAb
(nm) λEm (nm) ϕA
J
(10−16 R0
E
a
b
c
ϕF
M−1 cm3) (Å)
(%)
1 → 3
2 → 4
462 381, 403,
421, 445
0.20
0.27
8.3
18
26
0.3
462 430, 454,
479
0.51
0.85
22.4
25.8
a
Wavelengths at the absorption (λAb) and emission (λEm) maxima of
b
reactant and product, respectively, in PBMA at 25 °C. The activation
quantum yield (ϕA) is the quantum yield for the photochemical
conversion of reactant into product. This parameter was determined
by illuminating aerated MeCN solutions of the reactant within the
chamber of a photoreactor (420 nm) and monitoring periodically the
formation of the product by absorption spectroscopy. The irradiation
power per unit area (2.3 mW cm−2) was measured with a potassium
ferrioxalate actinometer, and this value was used to estimate ϕA from
the corresponding absorbance evolution during photolysis, according
to an established procedure (ref 41). The error in the determination of
c
ϕA is ca. 15%. Literature values for the fluorescence quantum yield
Figure 7. Evolution of the emission intensity (λEx = 390 nm, λEm = 453
nm) of PBMA films, doped with 2 at a concentration of 8 (a) or 0.8%
w/w (b), relative to the polymer, in the absence of silver nanoparticles
or at a concentration of 8% w/w (c) in their presence and spin coated
on quartz slides, during the sequential acquisition of spectra over the
course of 3000 s (scan rate = 10 nm s−1) at 25 °C.
d
(ϕF) of the product in aerated MeCN at 25 °C (ref 40). Overlap
integral (J) and Forster distance (R ) estimated from the spectra
̈
e0
recorded in aerated MeCN at 25 °C. Efficiency (E) of energy transfer
from product to reactant estimated from the value of R0 and the
average distance between molecules in PBMA at a concentration of 8%
w/w relative to the polymer.
sensitized formation of 4, under illumination at 390 nm, with
dilution.37
In these experiments, the irradiation wavelength of 390 nm is
essentially both λAc as well as λEx. The initial activation of the
reactant and the subsequent excitation of the product are both a
consequence of absorption at this particular wavelength, which
happens to fall within the spectral range associated with the
surface-plasmon band of silver nanoparticles (Figure S10
(SI)).38 As a result of the local enhancement in electromagnetic
field,21−23 these metallic nanostructures can therefore promote
the transformation of 2 into 4 in their proximity, under such
illumination conditions. Indeed, the sequential acquisition of
emission spectra (Figure S11 (SI)) of a PBMA film, doped with
2 and spin coated on silver nanoparticles deposited on a quartz
slide, reveals the developing bands of 4 with, yet again, a
sigmoidal temporal dependence (Figure 7c) of the emission
intensity. However, the time required to convert 50% of the
reactant is only 550 s in the presence of the silver nanoparticles,
while it is 1450 s in their absence. Fittings (Figures S8 and S9
(SI)) of the corresponding sigmoidal plots indicate a 1.6-fold
increase in the quantum yield for the sensitized formation of 4
under the influence of the metallic nanostructures.37
Figure 5. Emission spectra (λEx = 390 nm) of a PBMA film, doped
with 2 (8% w/w) and spin coated on a quartz slide, recorded
consecutively over the course of 3000 s (scan rate = 10 nm s−1) at 25
°C.
The effect of the nanoparticles on the photoinduced
conversion of 2 into 4 can be a result of their ability to
facilitate the direct excitation of the reactant and the conversion
of the resulting excited state into the product. Alternatively, the
nanostructures can promote the excitation of the product,
enhance its emission and, hence, encourage sensitization.
Comparison of the emission spectra (Figure S12a,b (SI)) of
4, recorded without and with silver nanoparticles, suggests that
the latter mechanism is mostly responsible for accelerating the
autocatalytic process. Indeed, the fluorescence of 4 increases
significantly in the presence of the nanoparticles. Specifically,
the ratio between the integrated emission intensities measured
with the nanoparticles and that recorded without is 2.7. Such an
Figure 6. Evolution of the emission intensity (λEx = 350 nm) of PBMA
films, doped (8% w/w) with 2 (a, λEm = 453 nm) or 1 (b, λEm = 400
nm) and spin coated on quartz slides, during the sequential acquisition
of spectra over the course of 9000 s (scan rate = 10 nm s−1) at 25 °C.
enhancement in the fluorescence of 4 elongates the Forster
̈
distance to 31 Å and increases the energy transfer to 48%.
Interestingly, the ratio between the energy transfer efficiencies,
estimated with and without the silver nanoparticles, is 1.8. This
13801
dx.doi.org/10.1021/ja5068383 | J. Am. Chem. Soc. 2014, 136, 13798−13804

Journal of the American Chemical Society
Article
CH2Cl2 (50 mL), the mixture was washed with aqueous HCl (1 M, 20
mL), H2O (20 mL) and brine (20 mL). The organic layer was dried
over Na2SO4 and then the solvent was distilled off under reduced
pressure. The residue was purified by column chromatography
[SiO2:AcOEt/hexanes (2:3, v/v)] to give 2 (90 mg, 50%) as a yellow
solid: ESIMS m/z = 457.1202 [M + Na]+ (m/z calcd. for C32H18O2Na
=457.1204); 1H NMR (CDCl3) δ = 5.04 (1H, s), 6.22 (1H, s), 7.14−
7.19 (4H, m), 7.29−7.34 (2H, m), 7.37−7.40 (2H, m), 7.43−7.48
(6H, m), 7.59−7.61 (2H, d, 8 Hz); 13C NMR (CDCl3) δ = 56.4, 60.4,
85.7, 95.0, 122.5, 122.7, 126.4, 128.8, 129.1, 129.6, 132.3, 133.1, 135.8,
136.3, 183.0, 184.0.
value is remarkably close to the enhancement (cf. 1.6) in
photoactivation efficiency determined from the sigmoidal plots.
Thus, the role of the metallic nanostructures in accelerating the
autocatalytic conversion of 2 into 4 appears to be
predominantly a consequence of their ability to promote
energy transfer from the product to the reactant.
CONCLUSIONS
■
Photochemical autocatalysis can be implemented on the basis
of fluorescence activation. The fluorescent product of such a
photochemical transformation must be designed to emit in the
same range of wavelengths where the nonemissive reactant
absorbs. If both species can be excited at the same wavelength
and if they are maintained in close proximity, then resonant
energy transfer from the product to the reactant can sensitize
the formation of the latter from the former to establish an
autocatalytic loop. Furthermore, the local enhancement in
electromagnetic field, associated with the illumination of
metallic nanoparticles, can be exploited to enhance the
energy-transfer efficiency and accelerate the overall photo-
chemical transformation. Ultimately, such operating principles
for autocatalysis with plasmonic boost translate into fluo-
rescence amplification. Thus, our representative example of
photochemical autocatalysis might eventually evolve into a
general design logic for the realization of plasmonic systems
capable of signal amplification.
5. A solution of 4 (360 mg, 0.9 mmol) and vinylene carbonate (1
mL, 16 mmol) in m-xylene was heated at 180 °C in sealed tube for 24
h. The reaction mixture was cooled down to ambient temperature and
diluted with dry MeOH (50 mL). The resulting precipitate was filtered
off, washed with MeOH (100 mL) and dried to give 5 (345 mg, 78%)
as a white solid: ESIMS m/z = 487.1294 [M + Na]+ (m/z calcd. for
1
C33H20O3Na = 487.1309); H NMR [(CD3)2CO] δ = 5.05 (1H, s),
5.13−5.18 (1H, m), 5.21−5.26 (1H, m), 5.95 (1H, s), 7.21−7.30 (4H,
m), 7.31−7.44 (4H, m), 7.46−7.59 (8H, m); 13C NMR (CDCl3) δ =
43.9, 48.5, 76.3, 76.4, 86.2, 86.5, 93.9, 94.3, 121.6, 122.5, 122.9, 123.3,
125.8, 127.0, 127.9, 128.0, 128.7, 128.8, 129.0, 131.7, 131.9, 132.2,
136.9, 137.8, 138.5, 138.9, 154.4.
6. A mixture of 5 (320 mg, 0.7 mmol) and KOH (160 mg, 3 mmol)
in H2O (2 mL) and EtOH (10 mL) was heated at 80 °C for 3 h. The
hot reaction mixture was filtered and the solid residue was purified by
column chromatography [SiO2:AcOEt/hexanes (1:1, v/v)] to give 6
(200 mg, 66%) as a white solid: ESIMS m/z = 461.1504 [M + Na]+
1
(m/z calcd. for C32H22O2Na =461.1517); H NMR [(CD3)2CO] δ =
4.05 (2H, s), 4.16 (1H, s), 4.42 (1H, s), 4.54 (1H, s), 5.65 (1H, s),
7.16−7.30 (6H, m), 7.32−7.52 (10H, m); 13C NMR [(CD3)2CO] δ =
47.7, 52.0, 59.6, 67.0, 67.3, 86.8, 87.5, 91.8, 92.3, 119.4, 121.0, 123.0,
123.4, 125.0, 125.8, 126.5, 126.7, 128.3, 128.5, 128.5, 129.4, 129.8,
131.5, 131.6, 140.3, 141.0, 141.5.
EXPERIMENTAL SECTION
■
Materials and Methods. Chemicals were purchased from
commercial sources and used as received with the exception of
CH2Cl2 and MeCN, which were distilled over CaH2, and H2O, which
was purified with a Barnstead International NANOpure Diamond
Analytical system. Compounds 1 and 4 were prepared according to
literature procedures.32,39 EISMS was performed with a Bruker
micrOTO-Q II spectrometer. NMR spectra were recorded with a
Bruker Avance 400 spectrometer. Absorption and emission spectra
were recorded with Varian Cary 100 Bio and Varian Cary Eclipse
spectrometers, respectively. Measurements were performed either in
aerated MeCN solutions, using quartz cells with a path length of 1.0
cm, or in PBMA matrices, using quartz slides mounted on custom-
built sample holders. The values of ϕF listed in Table 1 are literature
data.40 Those of ϕA listed in the same table are the quantum yields for
the photochemical conversions of 1 into 3 and of 2 into 4 in aerated
acetonitrile at 25 °C. The values of ϕA were determined by monitoring
the evolving absorbance of the photochemical product under
illumination at 420 nm, using a potassium ferrioxalate actinometer
to measure the irradiation power per unit area (2.3 mW cm−2) with an
established procedure.41 Samples were illuminated with a Luzchem
Research LZC-4V photoreactor (420 nm, 2.3 mW cm−2) for the
experiments in Figures 4 and S5 (SI) and with the excitation source
(350 or 390 nm) of the emission spectrometer for the experiments in
Figures 6, 7, and S7. A Chemat Technologies KW-4A spin coater was
used to prepare the polymer films. A Tencor Instruments 10−00090
surface profilometer was used to measure the thickness of the polymer
films. Fluorescence images were recorded with a Leica SP5 confocal
laser-scanning microscope.
Crystallographic Analysis. The data crystals of 1 and 2 were
glued onto the end of a thin glass fiber. X-ray intensity data were
measured with a Bruker SMART APEX2 CCD-based diffractometer,
using Mo Kα radiation (λ = 0.71073 Å).42 The raw data frames were
integrated with the SAINT+ program by using a narrow-frame
integration algorithm. Corrections for Lorentz and polarization effects
were also applied with SAINT+. An empirical absorption correction
based on the multiple measurement of equivalent reflections was
applied using the program SADABS. The structures were solved by a
combination of direct methods and difference Fourier syntheses and
refined by full-matrix least-squares on F2 with the SHELXTL software
package.43 All non-hydrogen atoms were refined with anisotropic
displacement parameters. Hydrogen atoms were placed in geometri-
cally idealized positions and included as standard riding atoms during
the least-squares refinements. Crystal data, data collection parameters
and results of the analysis are listed in Table S1 (SI).
Yellow single crystals were obtained after diffusion of hexane vapors
into benzene solutions of 1 and 2. Compound 1 crystallized in the
monoclinic crystal system. The systematic absences in the intensity
data identified the unique space group P21/c. Compound 2 crystallized
in the triclinic crystal system, and the space group P1 was assumed and
̅
confirmed by the successful refinement of the structure. Two
molecules are present in the asymmetric crystal unit.
Silver Nanoparticles. Aqueous NaOH (1.2 M, 0.1 mL) was added
to aqueous AgNO3 (0.22 g, 26 mL) under vigorous stirring. A dark-
brown precipitate formed immediately. Aqueous NH4OH (7.3 M, 1
mL) was added dropwise to dissolve the precipitate. The resulting
clear solution was cooled down to 5 °C. Quartz slides were submerged
in the cooled solution and aqueous D-glucose (0.35 g, 4 mL) was
added. The mixture was stirred for 2 min at 5 °C, allowed to warm up
to ambient temperature, heated to 40 °C and stirred for a further 10
min at this temperature. In the process, the yellow-green solution
turned brown and a greenish coating deposited on the slides. The
slides were removed from the solution, washed with H2O, sonicated in
H2O for 1 min at ambient temperature, washed again with H2O, dried
in air for 2 h and coated with the polymer films.
2. Trifluoroacetic anhydride (1 mL, 7.0 mmol) was added dropwise
over 10 min to a mixture of dry DMSO (1 mL) and CH2Cl2 (4 mL)
maintained at −60 °C under Ar. The resulting solution was stirred
under these conditions for a further 10 min and then a solution of 6
(180 mg, 0.4 mmol) in a mixture of dry DMSO (1 mL) and dry
CH2Cl2 (2 mL) was added dropwise over 10 min. The resulting
solution was stirred for a further 60 min under the same conditions
and then i-Pr2EtN (2.7 mL, 16 mmol) was added dropwise over 5 min.
The resulting solution was stirred for a further 60 min and then it was
allowed to warm up to ambient temperature. After dilution with
13802
dx.doi.org/10.1021/ja5068383 | J. Am. Chem. Soc. 2014, 136, 13798−13804

Journal of the American Chemical Society
Article
Polymer Films. A solution of PBMA (MW = 337 × 103) and either
1 or 2 (0.8, 2 or 8% w/w relative to PBMA) was deposited dropwise
on either a glass or a quartz slide. The substrate was spun at 1000 rpm
for 20 s and then again at 1000 rpm for a further 60 s. The coated
slides were stored under reduced pressure for 6 h prior to any imaging
(glass) and spectroscopic (quartz) experiments. The same protocol
was employed to deposit polymer films on quartz slides precoated with
silver nanoparticles.
Soc. 1997, 46, 11202−11210. (c) Saltiel, J.; Wang, S.; Ko, D.-H.;
Groming, D. A. J. Phys. Chem. A 1998, 102, 5383−5392. (d) Saltiel, J.;
Crowder, J. M.; Wang, S. J. J. Am. Chem. Soc. 1999, 121, 895−902.
(e) Saltiel, J.; Dmitrenko, O.; Reischl, W.; Bach, R. D. J. Phys. Chem. A
2001, 105, 3934−3939. (f) Saltiel, J.; Dmitrenko, O.; Pillai, Z. S.;
Klima, R.; Wang, S.; Wharton, T.; Huang, Z.-N.; van de Burgt, L. J.;
Arranz, J. Photochem. Photobiol. Sci. 2008, 7, 566−577.
(17) (a) Lechtken, P.; Yekta, A.; Turro, N. J. J. Am. Chem. Soc. 1973,
95, 3027−3028. (b) Turro, N. J.; Schore, N. E.; Steinmet, H. C.; Yekta,
A. J. Am. Chem. Soc. 1974, 96, 1936−1938. (c) Renner, C. A.; Katz, T.
J.; Pouliquen, J.; Turro, N. J.; Waddell, W. H. J. Am. Chem. Soc. 1975,
97, 2568−2570. (d) Turro, N. J.; Waddell, W. H. Tetrahedron Lett.
1975, 2069−2072. (e) Lechtken, P.; Yekta, A.; Shore, N. E.;
Steinmetzer, H. C.; Waddell, W. H.; Turro, N. J. Z. Phys. Chem.
1976, 101, 79−90. (f) Turro, N. J.; Ramamurthy, V.; Katz, T. J. Nouv.
J. Chim. 1977, 1, 363−365.
(18) (a) Okamoto, H.; Arai, T.; Sakuragi, H.; Tokumaru, K.;
Kawanishi, Y. Bull. Chem. Soc. Jpn. 1991, 64, 216−220. (b) Karatsu, T.;
Hiresaki, T.; Arai, T.; Sakuragi, H.; Tokumaru, K.; Wirz, J. Bull. Chem.
Soc. Jpn. 1991, 64, 3355−3362. (c) Lednev, I. K.; Alfimov, M. V.;
Kuriyama, Y.; Arai, T.; Sakuragi, H.; Tokumaru, K. J. Photochem.
Photobiol. 1992, 63, 201−209. (d) Arai, T.; Furuya, Y.; Furuuchi, H.;
Tokumaru, K. Chem. Phys. Lett. 1993, 212, 597−603. (e) Karatsu, T.;
Tsuchiya, M.; Arai, T.; Sakuragi, H.; Tokumaru, K. Bull. Chem. Soc. Jpn.
1994, 67, 3030−3039. (f) Arai, T.; Takahashi, O.; Asano, T.;
Tokumaru, K. Chem. Lett. 1994, 205−208. (g) Tokumaru, K.; Arai,
T. Bull. Chem. Soc. Jpn. 1995, 68, 1065−1087.
ASSOCIATED CONTENT
* Supporting Information
■
S
1H and 13C NMR spectra of 2, 5 and 6; crystallographic data
(CIF) for 1 and 2; absorption and emission spectra of 1−4 in
MeCN; fluorescence images of 2 in PBMA before and after
activation; kinetic model and fittings of the sigmoidal plots;
absorption and emission spectra of 2 and silver nanoparticles in
PBMA; emission spectra of 4 with and without silver
nanoparticles in PBMA. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
fraymo@miami.edu
■
Notes
The authors declare no competing financial interest.
(19) (a) Merkel, P. B.; Roh, Y.; Dinnocenzo, J. P.; Robello, D. R.;
Farid, S. J. Phys. Chem. A 2007, 111, 1188−1199. (b) Ferrar, L.; Mis,
M.; Dinnocenzo, J. P.; Farid, S.; Merkel, P. B.; Robello, D. R. J. Org.
Chem. 2008, 73, 5683−5692.
(20) (a) Kuzmanich, G.; Natarajan, A.; Chin, K.; Veerman, M.;
Mortko, C.; Garcia-Garibay, M. A. J. Am. Chem. Soc. 2008, 130, 1140−
1141. (b) Kuzmanich, G.; Gard, M.; Garcia-Garibay, M. A. J. Am.
Chem. Soc. 2009, 131, 11606−11614. (c) Nielsen, A.; Kuzmanich, G.;
Garcia-Garibay, M. A. J. Phys. Chem. A 2014, 118, 1858−1863.
ACKNOWLEDGMENTS
■
The National Science Foundation (CAREER Award CHE-
0237578, CHE-0749840 and CHE-1049860) is acknowledged
for financial support.
REFERENCES
■
(1) Bissette, A. J.; Fletcher, S. P. Angew. Chem., Int. Ed. 2013, 52,
12800−12826.
(2) von Kiedrowski, G. Bioorg. Chem. Front. 1993, 3, 113−146.
(3) Ruiz-Mirazo, K.; Briones, C.; de la Escosura, A. Chem. Rev. 2014,
114, 285−366.
(21) Alvarez-Puebla, R. A.; Liz-Marzan
Phys. Chem. Lett. 2010, 1, 2428−2434.
́
, L. M.; García de Abajo, F. J. J.
(22) Scaiano, J. C.; Stamplecoskie, K. J. Phys. Chem. Lett. 2013, 4,
1177−1188.
(4) Hong, J. I.; Feng, Q.; Rotello, V.; Rebek, J. Science 1992, 255,
848−850.
(5) Garrido Montalban, A.; Meunier, H. G.; Ostler, R. B.; Barrett, A.
G. M.; Hoffman, B. M.; Rumbles, G. J. Phys. Chem. A 1999, 103,
4352−4358.
(6) (a) Dadon, Z.; Samiappan, M.; Safranchik, E. Y.; Ashkenasy, G.
Chem.Eur. J. 2010, 16, 12096−12099. (b) Samiappan, M.; Dadon,
Z.; Ashkenasy, G. Chem. Commun. 2011, 47, 710−712. (c) Dadon, Z.;
Samiappan, M.; Wagner, N.; Ashkenasy, G. Chem. Commun. 2012, 48,
1419−1421.
(7) (a) Kottani, R. R.; Majjigapu, J. R. R.; Kurchan, A. N.; Majjigapu,
K.; Gustafson, T. P.; Kutateladze, A. G. J. Am. Chem. Soc. 2006, 128,
14794−14795. (b) Gustafson, T. P.; Metzel, G. A.; Kutateladze, A. G.
Org. Biomol. Chem. 2011, 9, 4752−4755. (c) Gustafson, T. P.; Metzel,
G. A.; Kutateladze, A. G. Photochem. Photobiol. Sci. 2012, 11, 564−577.
(8) Sella, E.; Lubelski, A.; Klafter, J.; Shabat, D. J. Am. Chem. Soc.
2010, 132, 3945−3952.
(9) Vignoni, M.; Cabrerizo, F. M.; Lorente, C.; Claparols, C.;
Oliveros, E.; Thomas, A. Org. Biomol. Chem. 2010, 8, 800−810.
(10) Dutta, S.; Mokhir, A. Chem. Commun. 2011, 47, 1243−1245.
(11) Thomas, S. W., III; Joly, G. D.; Swager, T. M. Chem. Rev. 2007,
107, 1339−1386.
(12) Raymo, F. M. Phys. Chem. Chem. Phys. 2013, 15, 14840−14850.
(13) Raymo, F. M. ISRN Phys. Chem. 2012, 619251−1−15.
(14) Raymo, F. M. J. Phys. Chem. Lett. 2012, 3, 2379−2385.
(15) Hyndman, H. L.; Monroe, B. M.; Hammond, G. S. J. Am. Chem.
Soc. 1969, 91, 2852−2859.
(16) (a) Saltiel, J.; Townsend, D. E.; Skyes, A. J. Am. Chem. Soc. 1973,
95, 5968−5973. (b) Saltiel, J.; Zhang, Y. X.; Sears, D. F. J. Am. Chem.
(23) Nabika, H.; Takase, M.; Nagasawa, F.; Murakoshi, K. J. Phys.
Chem. Lett. 2010, 1, 2470−2487.
(24) Strating, J.; Zwanenburg, B.; Wagenaar, A.; Udding, A. C.
Tetrahedron Lett. 1969, 10, 125−128.
(25) Mondal, R.; Shah, B. K.; Neckers, D. C. J. Am. Chem. Soc. 2006,
128, 9612−9613.
(26) (a) Mondal, R.; Okhrimenko, A. N.; Shah, B. K.; Neckers, D. C.
J. Phys. Chem. B 2008, 112, 11−15. (b) Bettinger, H. F.; Mondal, R.;
Krasowska, M.; Neckers, D. C. J. Org. Chem. 2013, 78, 1851−1857.
(27) Suzuki, M.; Aotake, T.; Yamaguchi, Y.; Noguchi, N.; Nakano,
H.; Nakayama, K.; Yamada, H. J. Photochem. Photobiol., C 2014, 18,
50−70.
(28) (a) Aotake, T.; Tanimoto, H.; Hotta, H.; Kuzuhara, D.;
Okujima, T.; Uno, H.; Yamada, H. Chem. Commun. 2013, 49, 3661−
3663. (b) Aotake, T.; Yamashita, Y.; Okujima, T.; Shirasawa, N.; Jo, Y.;
Fujimori, S.; Uno, H.; Ono, N.; Yamada, H. Tetrahedron Lett. 2013, 54,
1790−1793.
(29) Peng, P.; Wang, C. M.; Shi, Z.; Johns, V. K.; Ma, L. Y.; Oyer, J.;
Copik, A.; Igarashi, R.; Liao, Y. Org. Biomol. Chem. 2013, 11, 6671−
6674.
(30) (a) Mondal, R.; Adhikari, R. M.; Shah, B. K.; Neckers, D. C. Org.
Lett. 2007, 9, 2505−2508. (b) Zhao, Y.; Mondal, R.; Neckers, D. C. J.
Org. Chem. 2008, 73, 5506−5513. (c) Mondal, R.; Tonshoff, C.; Khon,
̈
D.; Neckers, D. C.; Bettinger, H. F. J. Am. Chem. Soc. 2009, 131,
14281−14289.
(31) Tonshoff, C.; Bettinger, H. F. Angew. Chem., Int. Ed. 2010, 48,
̈
4125−4128.
(32) Wright, M. W.; Welker, M. E. J. Org. Chem. 1996, 61, 133−141.
13803
dx.doi.org/10.1021/ja5068383 | J. Am. Chem. Soc. 2014, 136, 13798−13804

Journal of the American Chemical Society
Article
(33) Only one of the two anthracene chromophores absorbs at 390
nm. Instead, both compounds have similar molar absorption
coefficients at 350 nm. As a result, the behavior of the two diketones
under illumination was compared using the latter wavelength (Figure
6a,b). Under these conditions, the spectral changes observed for 2 are
similar to those detected with an illumination wavelength of 390 nm
(Figure 5).
(34) The average distance between the many molecules immobilized
within the rigid polymer matrix is 32 Å at this particular dopant
loading. This distance value was estimated from the molar
concentration of the reactant (1 or 2) in PBMA. The latter parameter
was determined from the corresponding absorbance at 462 nm (Figure
4a,b), the molar absorption coefficient determined in MeCN (Figure
S5a,b (SI)) and the film thickness (2.5 μm) measured with a surface
profilometer.
(35) A literature precedent (ref 20c) demonstrates that Dexter
energy transfer governs the cascade responsible for quantum
amplification in the photoinduced decarbonylation of a diary-
lcyclopropanone. In principle, such a mechanism could also contribute
to the ability of 4 to sensitize its own formation from 2. Considering
the exponential dependence of Dexter energy transfer on the donor−
acceptor separation, however, its efficiency should be negligible at the
relatively long distance (32 Å) between 2 and 4 ensured by the
selected experimental conditions (ref 34).
(36) In principle, 2 can reabsorb part of the photons emitted by 4
and undergo decarbonylation as a result. However, the contribution of
this radiative process, if at all occurring, to the sensitized formation of
4 is most likely negligible under the conditions selected for these
experiments. Indeed, the absorbance of the n → π* band of 2 is less
than 0.01 in all instances and such small values ensure insignificant
reabsorption.
(37) The mathematical model employed to fit the photokinetic
profiles is reported in the Supporting Information.
(38) Lakowicz, J. R.; Shen, Y.; Auria, S. D.; Malicka, J.; Fang, J.;
Gryczynski, Z.; Gryczynski, I. Anal. Biochem. 2002, 301, 261−277.
(39) Goichi, M.; Segawa, K.; Suzuki, S.; Toyota, S. Synthesis 2005,
2116−2118.
(40) Thapaliya, E. R.; Captain, B.; Raymo, F. M. J. Org. Chem. 2014,
79, 3973−3981.
(41) Scaiano, J. C., Ed.; Handbook of Organic Photochemistry; CRC
Press: Boca Raton, FL, 1989.
(42) Apex2 Version 2.2−0 and SAINT+ Version 7.46A; Bruker
Analytical X-ray System, Inc.: Madison, WI, 2007.
(43) (a) Sheldrick, G. M. SHELXTL Version 6.1; Bruker Analytical X-
ray Systems, Inc.: Madison, WI, 2000. (b) Sheldrick, G. M. Acta
Crystallogr. 2008, A64, 112−122.
13804
dx.doi.org/10.1021/ja5068383 | J. Am. Chem. Soc. 2014, 136, 13798−13804