Self-assembled phenylethynylene bis-urea macrocycles facilitate the selective photodimerization of coumarin

Article abstract of DOI:10.1021/ja110779h

There is much interest in designing molecular sized containers that influence and facilitate chemical reactions within their nanocavities. On top of the advantages of improved yield and selectivity, the studies of reactions in confinement also give import

Full text of DOI:10.1021/ja110779h

ARTICLE
pubs.acs.org/JACS
Self-Assembled Phenylethynylene Bis-urea Macrocycles Facilitate the
Selective Photodimerization of Coumarin
Sandipan Dawn,† Mahender B. Dewal,† David Sobransingh,† Monissa C. Paderes,† Arief C. Wibowo,†
Mark D. Smith,† Jeanette A. Krause,‡ Perry J. Pellechia,† and Linda S. Shimizu*,†
†Department of Chemistry and Biochemistry, University of South Carolina, Columbia, South Carolina 29208, United States
‡The Richard C. Elder X-ray Crystallographic Facility, Department of Chemistry, University of Cincinnati, Cincinnati, Ohio 45221,
United States
S Supporting Information
b
ABSTRACT:
There is much interest in designing molecular sized containers that influence and facilitate chemical reactions within their
nanocavities. On top of the advantages of improved yield and selectivity, the studies of reactions in confinement also give important
clues that extend our basic understanding of chemical processes. We report here, the synthesis and self-assembly of an expanded bis-
urea macrocycle to give crystals with columnar channels. Constructed from two C-shaped phenylethynylene units and two urea
groups, the macrocycle affords a large pore with a diameter of ∼9 Å. Despite its increased size, the macrocycles assemble into
columns with high fidelity to afford porous crystals. The porosity and accessibility of these channels have been demonstrated by gas
adsorption studies and by the uptake of coumarin to afford solid inclusion complexes. Upon UV-irradiation, these inclusion
complexes facilitate the conversion of coumarin to its anti-head-to-head (HH) photodimer with high selectivity. This is contrary to
what is observed upon the solid-state irradiation of coumarin, which affords photodimers with low selectivity and conversion.
’ INTRODUCTION
inclusion complexes (Figure 1), and show their utility as confined
environments for a selective photochemical reaction of coumarin.
Our group is interested in synthesizing porous materials with
homogeneous channels for use as confined reaction environ-
ments. Specifically, we have used macrocycles that contain two
rigid spacers and two urea groups, which self-assemble into
straw-like structures guided by the formation of 3-centered
ureaꢀurea hydrogen bonds between the neighboring macro-
cycles as well as by the formation of favorable aryl stacking
interactions.8 The size of the open cylindrical channel is con-
trolled by the size and shape of the macrocyclic building blocks.
Thus far, we have used benzophenone and phenylether as
spacers to generate macrocycles. These spacers are very similar
in size and generate porous materials with channels of ∼6 Å that
can be used to facilitate photochemical reactions.9 Herein, we
report the use of phenylethynylene spacers that are nearly double
the length of the prior systems (Figure 2). We investigated the
ability of two ureas to guide the assembly of these larger macrocycles
Chemists do not normally think twice about the flask when
setting up a new reaction. Yet the container can dramatically
influence molecular processes. Nanocapsules,1 zeolites,2 and
solꢀgels3 have been used as molecular sized flasks to influence
the properties, stability, and even the reactivity of absorbed
or encapsulated molecules. Encapsulation has been shown
to enhance the stability of unstable molecules.4 For example,
molecular sized containers stabilized carbocations and enabled
chemists to probe their structure.5 Encapsulation can be used
to modulate reactivity and dramatically change product distri-
butions.6 Zeolites in particular can be used to selectively afford
products that are not observed in solution.7 On top of the
advantages of improved yield and selectivity, the studies of
reactions in confinement also give important clues that extend
our basic understanding of chemical processes. We report herein,
the synthesis and self-assembly of a new molecular container,
a phenylethynylene bis-urea macrocycle 1. This macrocycle
assembles into columnar structures to yield porous crystals with
channels with diameters of ∼9 Å. We demonstrate the accessibility
of these channels by gas adsorption studies, characterize their
Received: December 1, 2010
Published: April 19, 2011
r
2011 American Chemical Society
7025
dx.doi.org/10.1021/ja110779h J. Am. Chem. Soc. 2011, 133, 7025–7032
|

Journal of the American Chemical Society
ARTICLE
Scheme 1. Synthesis of Macrocycle 1a
a Reagents and conditions: (a) Pd(PPh3)2Cl2/ CuI/piperidine, 0 °C,
3 h, 97%; (b) NBS/PPh3 in THF, 0 °C, 12 h, 87%; (c) triazinanone,
NaH/THF, reflux 48 h, 22%; (d) 1:1 MeOH:(20% NH(CH2CH2OH)2
in H2O, pH ∼2), reflux 12 h, 96%.
Figure 1. Bis-urea macrocycles were designed to self-assemble into
columns affording accessible channels. Uptake of coumarin from solution
yields a solid hostꢀguest complex. Broad-band UV-irradiation of the
complex selectively affords the anti-head-to-head photodimer.
macrocyclization with triazinanone under basic conditions, fol-
lowed by deprotection, afforded the bis-urea macrocycle as a
white precipitate. Microneedles of 1 (∼20 μm ꢁ 3 μm) were
regularly obtained from hot DMSO (50 mg/10 mL) at 120 °C
solution by slow cooling (1 °C/h) and were used for all the
photochemical experiments. Larger pale yellow needles (0.20 ꢁ
0.05 ꢁ 0.04 mm3) of similar morphology were obtained by slow
cooling from 2:8 DMSO/nitrobenzene (30 mg/12 mL) and
were subjected to a synchrotron X-ray diffraction study.
The X-ray crystal structure revealed the desired macrocycle and
encapsulated but disordered nitrobenzene solvent (Figure S26
Supporting Information), which was omitted for clarity in
Figure 3a. Despite their greater size, the phenylethynylene macro-
cycles assembled into columns very similar in structure to the
smaller macrocyclic bis-ureas.8,9 The individual macrocycles were
organized into columns through the characteristic three centered
Figure 2. Comparison of the relative length of rigid spacers used to
construct macrocyclic bis-ureas.
with high fidelity into columnar structures from two different
solvents and characterized the solid-state assembled structures.
The porosity of these materials was evaluated by gas adsorption
and guest uptake studies. Finally, the utility of these larger
channels was demonstrated by their ability to facilitate the
photodimerization of coumarin under UV-irradiation.
urea hydrogen bonds, which displayed (N)H O distances of
3 3 3
2.06(3) and 2.20(3) Å respectively (Figure 3b). Additional stabi-
lization was provided by edge to face aryl stacking and by stacking
interactions between the alkyne and the phenyl on the neighboring
macrocycle. The three phenyl rings in one-half of the macrocycle
are independent by symmetry. One of the rings (ring z, Figure 3b)
is disordered in a 50/50 ratio over two orientations (one orientation
omitted for clarity). Two adjacent z and z0 rings provide stabilizing
edge-to-face aryl interaction. Additional interactions between the
alkyne and neighboring phenyl groups further stabilize the as-
sembled columns. One of the triple bonds is packed closely to
phenyl rings on adjacent macrocycles with distances of 3.49 and
3.38 Å, respectively (Figure 3c). Figure 3d highlights the view down
the crystallographic b axis of seven hydrogen bonded tubes with the
guests removed. Without the entrapped nitrobenzene, the calcu-
lated void volume is 491.1 Å3 per unit cell (21% of the total unit cell
volume). Individual columns are close packed together with
benzene rings alternating tilt direction in the neighboring rows of
tubes (Figure 3d).
’ RESULTS AND DISCUSSION
The phenylethynylene spacers are of interest due to their
larger length and conjugated nature. Substituted poly para-
phenylethynylenes have demonstrated applications in optical/
fluorescence based sensing of substrates of biological interest10
including DNA.11 Oligomers of meta-phenylethynylenes fold
into helical structures12 or can be cyclized to give macrocycles
such as those reported by Moore,13 Hoger,14 and others.15
Moore’s shape persistent arylene ethynylene macrocycles as-
sembled into fibrils that displayed polarized emission parallel to
the stacking direction.16 Such emission is indicative of inter-
molecular delocalization of the electrons in the pi clouds. Thus,
columnar assembled phenylethynylene derivatives may have appli-
cations as sensors, semiconductors, and photovoltaic devices.
Our long-term goal is to evaluate if the urea assembly motif could
augment that of phenylethynylenes to afford a larger macrocyclic
cavity that upon assembly would give porous materials that
combine molecular recognition properties with the electronic
characteristics of the component phenylethynylenes.
Thermogravimetric analysis (TGA) was used to investigate
whether the nitrobenzene solvent could be removed from the
crystals. Host 1 nitrobenzene displays a one-step desorption
3
curve with a weight loss of 13.9% between 70 and 140 °C
(Figure 4a). From the percent weight loss, we calculated the
macrocycle:guest stoichiometry as 1:0.89 (calculated weight loss,
Macrocycle 1 was synthesized in four steps from commercial
1,3-diethynylbenzene using a SonogashiraꢀHagihara coupling
with excess 4-iodobenzyl alcohol to yield the diol spacer, which
was subsequently converted to the dibromide (Scheme 1). A
14.1% for a 1:0.94 complex). In comparison, the host 1 DMSO
3
crystals displays a two-step desorption curve from 30 to 170 °C
with weight loss of 18.38% (Figure 4a). The two-step desorption
7026
dx.doi.org/10.1021/ja110779h |J. Am. Chem. Soc. 2011, 133, 7025–7032

Journal of the American Chemical Society
ARTICLE
Figure 5. (a) CO2 adsorption isotherm of host 1 at 273 K. (b) PXRD
pattern of host 1 before (bottom) and after coumarin (3) absorption (top).
structure using POUDRIX,17 suggesting that the bulk powder was
similar in structure to the crystals. The guest was removed by TGA
and the powder re-examined by PXRD (Figure 4b, ii). The PXRD
shows a similar sharp and intense PXRD pattern, indicating that
the empty host maintains long-range crystalline order after
removal of the solvent. Next, the microcrystals of host 1 DMSO
3
were ground to a powder and examined by PXRD (Figure 4b, iii).
The sharp and intense peaks in the powder pattern of host
1 DMSO closely match those observed for host 1 nitrobenzene,
3
3
which suggests that it also has similar highly ordered crystalline
structure (Figure 4b, i and iii). The DMSO guest was removed by
heating. The empty structure displays peak positions and inten-
sities that are in good agreement with the empty host generated
from the nitrobenzene derived crystals (Figure 4b, ii and iv)
suggesting that the empty frameworks assembled from the two
different solvents have similar structures.
Figure 3. X-ray structure of 1 C50H36N4O2 (C6H5NO2)0.94: (a) space
3
filling view of a single macrocycle with intramolecular distances from Ha
to Ha0 of ∼8.7 Å and Hb to Hb0 of ∼8.4 Å; (b) view alongside a single
column that was organized via NH O hydrogen bonds that run along
3 3 3
the crystallographic b axis (nitrobenzene guests and hydrogens omitted);
Single crystal data on host 1 nitrobenzene suggests that the
3
(c) expansion highlighting the alkyne π centroid ring distances; (d) view
3 3 3
solvent-free structure could have open channels for binding
guests. PXRD studies further support the hypothesis that host
1 adopts similar structures when assembled from different
solvents (DMSO vs nitrobenzene). We next turned to gas
adsorption to evaluate the porosity of the evacuated hosts. The
solvent was removed from the abundant microcrystals obtained
from DMSO by heating and the empty host (47 mg) was
subjected to gas adsorption measurements. Host 1 shows a
6.6 mL/g CO2 uptake at relatively low pressure (<0.03 atm).
The adsorption isotherm reveals a steep rise at relatively low
pressure, type I behavior that is typical for microporous materials
(Figure 5a). The BrunauerꢀEmmettꢀTeller (BET) method
was applied to the isotherm to calculate surface area.18 The
isotherm gave an apparent surface area of 349 m2/g at 273 K
between 0.010 and 0.028 relative pressures. The total pore
volume was estimated as 0.025 cm3/g at 273 K for pores smaller
than 11.3 Å in diameter at P/P0 = 0.027.
down the b axis of seven H-bonded tubes (nitrobenzene guests omitted).
Figure 4. (a) TGA desorption curves for the host 1 nitrobenzene
3
(dashed line) and host 1 DMSO (solid line). (b) PXRD comparison of
3
crystals obtained from different solvents before and after heating:
(i) from nitrobenzene before heating, (ii) from nitrobenzene after
heating, (iii) from DMSO before heating, (iv) from DMSO after heating.
curve suggests that the DMSO is absorbed in two different
environments within the crystals. From the measured total
weight loss, we calculated a 1:2 host/guest stoichiometry. The
average total weight loss for 12 different sizes and batches of
crystals was 18.3%, with an 8.2% weight loss observed between
30 and 80 °C and 10.1% between 80 and 130 °C. The
reproducible nature of the DMSO desorption from different
sizes and batches of crystals experimentally suggests that the
DMSO solvent is incorporated into the crystals and not adsorbed
on the surface, much like nitrobenzene.
We next sought to load guests into these porous crystals by
soaking the solid host in a solution containing the guest.19 The
guest was removed from the host 1 DMSO crystals by heating
3
and the free host (30 mg) was soaked in a coumarin 3 solution
(0.1 mM in CH3CN) for 1ꢀ12 h. The depletion of coumarin
from solution was monitored by UVꢀvis (273 nm) and reached
an equilibrium within 3 h (Supporting Information). A Lambertꢀ
Beer plot with variable concentration (0.01ꢀ0.1 mM) of coumarin
in acetonitrile was used to calculate the ratio of host 1 to coumarin
in the treated samples. The host/guest ratio was calculated
as 1:1.4 and was found to be reproducible for different sizes
and batches of host 1 crystals. The reproducible uptake of
coumarin suggests that the coumarin guests are absorbed in
the crystals.
We turned to powder X-ray diffraction (PXRD) to compare the
structure of these two crystalline forms of host 1 (1 nitrobenzene
3
versus 1 DMSO) and their solvent free forms. Host 1 nitroben-
3
3
zene crystals were ground to powder and examined by PXRD
(Figure 4b, i). The experimentally observed PXRD pattern is
nearly identical to the PXRD pattern simulated from the crystal
To further examine the structure of the coumarin treated sample,
the crystals were pressed gently onto the slide and examined by
7027
dx.doi.org/10.1021/ja110779h |J. Am. Chem. Soc. 2011, 133, 7025–7032

Journal of the American Chemical Society
ARTICLE
Figure 7. (a) Solid-state UVꢀvisibile absorption spectra of host 1,
Figure 6. Solid-state cross polarized magic angle spinning 13C {1H}
host 1 coumarin, and coumarin. (b) Comparison of the solid-state
3
CP-MAS (125.79 MHz) NMR spectra for host 1 (bottom) and host 1 3
fluorescence spectra of host 1 (λexc =358 nm), 3 (λexc =351 nm),
3
complex (top). Both spectra were acquired using a double resonant
Doty Scientific XC 4 mm magic angle spinning (MAS) probe with
TPPM modulated dipolar decoupling with a 61 kHz field strength.
host 1 3 complex (λexc =347 nm).
3
wavelength was used to generate the emission spectrum. Host 1, 3,
and host 1 3 were excited at 358, 351, and 347 nm, respectively
3
PXRD (Figure 5b, ii). The host 1 3 displays a distinct PXRD
pattern from host 1 (Figure 5b, i) and shows sharp and intense
peaks consistent with a well ordered crystalline structure. Compar-
(Figure 7b). The solid-state fluorescence spectra of the host 1 3
3
3
complex (λexc = 347 nm) shows the expected bands at 408 nm (3)
and 426 nm (host 1), which suggest that coumarin has been
indeed loaded within the host; however, once again, no new bands
were observed.
ison of the PXRD patterns of the free host and the host 1 3
3
complex indicates that, in presence of the guest, there are changes in
the long-range order of the host, but the complex remains highly
crystalline.
We established that the host 1 crystals absorbed coumarin
from solution to give a highly ordered solid complex (host 1 3).
3
We turned to solid-state NMR to compare the host and its
coumarin complex. Solid-state cross polarized magic angle spin-
ning 13C {1H} CP-MAS (125.79 MHz) NMR spectra were taken
on the host before soaking and on the coumarin-treated host
1 (Figure 6). The treated host shows new peaks in the carbonyl
(δ = 159.9, 153.5 ppm) and in the aromatic region that are
consistent with coumarin, further supporting the hypothesis that
coumarin was absorbed by the host. In addition, the cross-
polarization build-up behavior of the coumarin is very similar
to that of the resonances of the tube. This suggests that the
mobility of the coumarin guest is similar to the columnar
framework. These solid-state NMR and PXRD studies indicate
that 3 is incorporated in the crystal lattice of host 1.
Unfortunately, it is currently difficult to predict the molecular
structure of the complex from the PXRD pattern. To gain further
insight into the structure of the inclusion complex, we turned to
molecular modeling using Spartan.20 Molecular modeling studies
allow us to probe the structure of the host 1 3 inclusion complex
3
to examine the fit of coumarin inside the channel of 1 and to
assess if hostꢀguest interactions limit the number of accessible
orientations a guest can sample inside the columnar structure.
Specifically, we were interested in identifying any (1) interactions
between the guests and the host framework and (2) interactions
between neighboring coumarin molecules. We generated the
host structure in Spartan using the atomic coordinates from the
single crystal structure of host 1 nitrobenzene and deleting
3
We further examined the host 1 3 complex by optical methods.
coordinates of the nitrobenzene guests. The columnar frame-
work was limited to four macrocycles, due to the increased size of
this macrocycle as compared with our previous systems.9b The
tetrameric host model was ‘frozen’ so that no movement was
allowed in the column framework. A coumarin guest was added
and allowed to move to energy minima in the static framework.
Guests were added sequentially and minimized until no addi-
tional guest molecules could be accommodated. The host/guest
ratio predicted by modeling studies was 1:1, which was slightly
lower than the experimentally observed ratio. This might be due
to edge effects given the small length of our model assembly as
well as the fact that we are considering only an isolated column
when in reality these are arrays of columns. In general, such
calculations can suffer from such edge effect due to the small size
of the modeled structure.21
3
The host is only soluble in DMSO, an aggressive solvent that
precludes complex formation. UVꢀvisible absorption studies of a
1:1 mixture of host 1/3 in solution (2.5 ꢁ 10ꢀ5 M in DMSO)
show only the expected bands for host 1 (λmax = 286 and 305 nm)
and coumarin 3 (λmax = 276 and 321 nm) and no extra bands are
apparent (see Figure S14 Supporting Information). UVꢀvisible
absorption studies of host 1, 3, and the host 1 3 complex were also
3
examined in the solid-state using a Perkin-Elmer Lambda 35 UV/
Visible scanning spectrophotometer equipped with an integrating
sphere. Figure 7a shows that the host and the guest absorb in a
similar range and have absorption maxima around 350 nm. The
solid hostꢀguest complex is a simple compilation of the two
UVꢀvisible absorbance spectra and no new bands are apparent.
We used fluorescence spectroscopy to further probe the inter-
action between the host and the guest. In solution, host 1 (2.5 ꢁ
10ꢀ6 M in DMSO) excited at λexc = 292 nm displays a broad
fluorescence band with a maxima at 348 nm. Coumarin 3 does not
fluoresce in DMSO solution and no changes are observed in the
emission spectra of the host in the 1:1 mixture of host 1/3 in
DMSO (Figure S25, Supporting Information). Excitation and
Monte Carlo searching of the conformer distributions at
ground state with Molecular Mechanics (MMFF) was performed
to investigate the low energy conformers of the host 1 coumarin
3
complex. The Monte Carlo minimization program used a
simulated annealing method to generate conformers of the host
1 3 inclusion complex.22 We produced 450 conformers during
3
emission spectra of ground solid samples of host 1, 3, and host 1 3
the calculation, and the stabilization energy of the four lowest
energy conformers fell within 0.5 kcal/mol. Each of these lowest
energy structures showed the coumarin molecules organized
3
were recorded at room temperature using a Perkin-Elmer LS 55
Fluorescence Spectrometer. In each case, the excitation maximum
7028
dx.doi.org/10.1021/ja110779h |J. Am. Chem. Soc. 2011, 133, 7025–7032

Journal of the American Chemical Society
ARTICLE
Figure 8. Models of the host 1 3 complex were constructed using Spartan. (a) Monte Carlo searching of the conformer distributions at ground state
3
with molecular mechanics (MMFF) using Spartan generated this lowest energy structure. Host 1 was rendered in wire structures and 3 in space filling
view. (b) View of the upper pair of coumarin molecules from part a; (c) view of the lower pair of coumarin molecules from part a; (d) schematic
presentation of host 1 3 complex. (e) X° represents angle between the carbonyl groups of two adjacent coumarin units within the host channel. The
3
planes formed by the carbonyl groups of the adjacent coumarins were not always parallel. X° varied significantly from 21 to 48° suggesting there is
significant freedom of movement for the included guests.
Scheme 2. Photolysis of Coumarin Affords Four Possible
Products
between 3.6 and 4.2 Å.25 Theoretically, UV-irradiation of coumarin
may afford four possible dimers (Scheme 2): syn-headꢀhead (HH)
4, syn-headꢀtail (HT) 5, anti-HH 6, and anti-HT 7 in both
solution and in the solid state. Solvent has a significant impact on
both the percent conversion and selectivity of this photodimer-
ization. High selectivity for the anti-HH 6 (91%) was observed
for the reaction in benzene, albeit with low conversion (9%).26
This product was also favored in the presence of triplet photo-
sensitizers.27 In contrast, photodimerization in polar solvents
such as 1,2-ethanediol afforded higher conversion (39%) but
lower selectivity (4/5/6 = 59:19:22).33 Photoreactions in the
solid-state, however, typically show both low selectivity and limited
conversion (<5%) due to their reversibility.28 This reversibility is
not true for the thermal reaction. A nice example of the thermal
dimerization of coumarin from Wen et al. in a crystalline inclusion
complex gives high selectivity (>95%) for the anti HH 6 at 30%
conversion.29
pairwise within the inclusion complex. The lowest energy con-
former is shown in Figure 8a, which depicts the host framework
in a wire view and guest molecules in a space filling view for
clarity. It is notable that numerous edge to face aryl stacking
interactions are predicted. These interactions hold the phenyl
ring of 3 close to the phenyl rings in the wall of the host
framework. These edge to face interactions range between 2.9
and 3.2 Å (from the center of the coumarin’s benzene ring to the
aryl hydrogen of the host framework). It is also notable that the
packed coumarins interact not only with the column walls, but
also with each other. Offset aryl stacking interactions stabilize the
individual pairs of coumarins with the center-to-center distance
between coumarin aryl rings ranging from 3.2 to 3.3 Å, typical
distances for such stacking interactions.23 Minimizations using
different starting hostꢀguest complex structures produced very
similar sets of inclusion complex structures. Thus, our hypothesis
is that coumarin can move within the host columns and that it is
able to access a variety of geometries.
Confinement within a solid host may limit molecular motion
and change the selectivity of the photoreaction.28b For example,
Ramamurthy found that 4 is favored in D2O solution at 90%
selectivity using a Fujita Pd nanocage.32 Therefore, we analyzed
the orientation of the two reactive alkenes in each of the coumarin
pairs in the five lowest energy structures. The lowest energy
conformer (Figure 8a) shows two pairs of coumarins (top and
bottom) each held together by offset aryl stacking interaction. In
the top pair, the reactive alkenes (Figure 8b) are oriented close
together (average distance = 4.27 Å) but not exactly parallel to
each other. Their carbonyl groups are in the same side, which
would give rise to the syn-HH 4 photodimer after UV irradiation.
The orientation of two coumarin molecules with respect to each
other differs significantly in the bottom pair (Figure 8c). They are
also held together by offset aryl stacking interactions; however, their
carbonyl groups are located on opposite sides with the reactive
alkenes relatively close in space (average distance = 4.33 Å). Thus,
we would predict this orientation to form the anti-HT product 7.
The photoreactions of coumarin are well-studied in both
solution24 and in the solid state, and are thought to require
precise orientation and separation of the two reacting alkenes
The four lowest energy conformers of the host 1 3 complex are
very close energetically, within 0.5 kcal/mol. All 10 coumarin pairs
are organized by offset aryl stacking interactions and position their
3
7029
dx.doi.org/10.1021/ja110779h |J. Am. Chem. Soc. 2011, 133, 7025–7032

Journal of the American Chemical Society
ARTICLE
Table 1. Comparison of Solid-State Photodimerization of
Coumarin in the Presence or Absence of Host 1a
continued irradiation.25 Direct dissolution of the samples in
DMSO-d6 gave similar results, indicating that the products were
efficiently removed from the crystals by washing. Longer reaction
times did not change the selectivity of the product but increased
the conversion. Photoirradiation for 24 and 96 h afforded the
same photodimer 6 with 97ꢀ98% selectivity in 9% and 18%
conversion, respectively. This selective solid-state dimerization
of coumarin was quite surprising and led us to further probe if
this molecular cage might be facilitating dimer formation, slow-
ing the reverse reaction of the dimer back to coumarin or
alternatively displaying a thermodynamic preference for the
bound product 6.
The atmosphere is known to affect the conversion of coumarin
to its photodimers. Oxygen can quench the excited state of
coumarin and limit conversion.31 Our gas absorption data demon-
strated that the host channels were accessible to CO2 (g), while
molecular modeling studies suggested that the channels should
be large enough to accommodate a number of different gases
including O2 (g), N2 (g), or Ar (g). To test whether coumarin was
accessible to the atmosphere in the solid complex, we investigated
the reaction under inert atmospheres (N2 (g) and Ar (g)). Sur-
prisingly, reaction under a N2 (g) atmosphere gave similarly high
selectivity (98% of 6) with slightly higher conversion at 12 h
(Table 1, entry 7). Longer irradiation times of 24 and 96 h afford
increased conversion (16% and 37%) with 97ꢀ98% selectivity for
photodimer 6. The conversion could be further enhanced under
Ar (g) to yield 55% conversion at 96 h (Table 1, entries 10ꢀ12)
with no decrease in selectivity (97%). The observed increase in
conversion under inert gases suggests that the coumarin absorbed
in the solid complex is still accessible to quenching by oxygen
in the air atmosphere. The 55% conversion limit may be due to
lack of uniform irradiation of the crystals or inefficient light
penetration. We are currently investigating methods for grinding
the hostꢀguest complex and irradiating the powder or alterna-
tively, irradiating a suspension of the powder in a deoxygenated
solvent.
a Photodimerization in liquid media or in other hosts are shown in
entries 13ꢀ15 for comparison. b In Pd-nanocage host, coumarin guest
molecule was introduced from a solution of host and guest.32 c In
solution at a [3] = 0.5 mol/dm3.26 d In solution [3] = 0.2 mol/dm3.24,26
reacting alkenes from 3.89 to 4.39 Å apart, favorable distances for
the [2 þ 2] reaction; however, the planes formed by the carbonyl
groups of the adjacent coumarins were not always parallel, the
favored alignment for reaction. However, the Ramamurthy group
observed that photodimerization of coumarin is possible through
[2 þ 2] cycloaddition reaction in solid-state even when the two
double bonds are not in parallel orientation.30 A schematic pre-
sentation of coumarin pairs predicted by our models in the host 1 3
3
complex is shown in Figure 8d. If we define an angle (X°) between
the planes of the included coumarins, we find this angle varies
significantly from 21 to 48° (Figure 8e). Theoretically, the pairs
could access three different photodimers (syn-HH 4, anti-HH 6,
and anti-HT 7). Overall, the molecular modeling analysis suggested
that pairs of 3 would be packed close enough to undergo [2 þ 2]
photodimerization reactions; however, their exact orientation might
not be controlled enough to give high selectivity.
We observed unexpected and unusually high conversion and
selectivity for the anti-HH photodimer 6 in the solid state. This
anti-HH photodimer is observed in solution via the triplet excited
state in presence of photosensitizer like benzophenone. This
leads us to the question of whether our host is acting as a triplet
photosensitizer in the solid state or if this product is a result of
confinement effects. Absorption and emission spectra of the solid
Given these predictions, we set out to test if coumarin in the
complex would react under UV-irradiation with particular em-
phasis on the reaction conversion and on its selectivity. The host
host 1 3 complex do not support triplet sensitization; however, it
3
is unfortunately very difficult in the solid-state to probe if the
fluorescence lifetimes are altered by the hostꢀguest complexa-
tion. To address this issue, we turned to an experimental
approach with the assumption that if the host were able to act
as a photosensitizer this effect should also be observed in
solution. Coumarin (25 mM) was UV-irradiated in the presence
and absence of host 1 (1.5 equiv) in d6-DMSO, where host 1 is
soluble as well as in d6-benzene, where host 1 is insoluble. The
coumarin (25 mM in benzene) was similarly UV-irradiated in the
presence and absence of the more soluble host 2 (Scheme 1),
which contains protected ureas. Host 2 is soluble in benzene,
which is generally a better solvent for this reaction. As additional
controls, coumarin was photoirradiated under the same condi-
tion in the presence and absence of benzophenone, a known
triplet sensitizer. No conversion of coumarin was observed in
either solvent in the presence of either host 1 or its soluble
analogue 2, while benzophenone facilitates the conversion of
coumarin 17% (after 12 h) to the anti-HH photodimer, similar to
1 3 complex (30 mg) in Norell S-5-500-7 NMR tubes with 100%
3
transmittance up to 400 nm was irradiated at room temperature
in air using a Hanovia 450 W medium pressure mercury arc lamp
cooled in a quartz immersion well that allows broad spectrum UV
transmittance (200ꢀ400 nm). Samples (5 mg) were removed
after 12, 24, and 96 h for analysis. The photoproducts were
separated from the host by extraction with CDCl3 and analyzed
1
by H NMR spectroscopy. The conversion was estimated by
comparison of the starting material (at 6.42 ppm) to the
cyclobutyl CH’s (δ 3.5ꢀ4.4 ppm). For the 12 h sample, we
observe only one photodimer, which is the anti-HH 6 in 8%
conversion with 97% selectivity (Table 1, entry 4). The exclusive
production of anti-HH photodimer in the solid state was both
unexpected and unprecedented. In general, the solid-state photo-
irradiation of solid coumarin provides dimers with no selectivity
(Table 1, entries 1ꢀ3) and low conversion (<5%), a fact that
is usually attributed to the reversibility of this reaction under
7030
dx.doi.org/10.1021/ja110779h |J. Am. Chem. Soc. 2011, 133, 7025–7032

Journal of the American Chemical Society
ARTICLE
prior reports.33 In addition, neither host 1 or 2 facilitates the
photoisomerization of trans-β-methyl styrene (Supporting In-
formation), a reaction known to require a low energy triplet
sensitizer.34 These experiments along with the solid-state emis-
sion spectra suggest that the production of the anti-HH photo-
dimer is not due to triplet sensitization by a single host molecule
and may be due to encapsulation within the larger assembly.
In summary, we report the synthesis and self-assembly of a
macrocyclic bis-urea that contains a much larger phenylethyny-
lene spacer unit. Despite its increased size, this bis-urea macro-
cycle still assembles in high fidelity to give columnar structures
that pack into porous crystalline materials. The porosity of these
new materials was demonstrated by gas adsorption studies and by
the absorption of coumarin guests from solution. Modeling
studies including Monte Carlo searching of the conformer distribu-
tions at ground state with molecular mechanics (MMFF) using
Spartan did not afford good predictions about the selectivity of the
subsequent photoreaction; however, they did suggest that the
coumarin binds in pairs through advantageous aryl stacking interac-
tions with the interior of the columns and that the pairs of coumarins
have room to move within the channel. We demonstrated the utility
of this new porous framework as a confined environment for
reactions with the solid-state photodimerization of coumarin within
Energy, Office of Energy Sciences Materials Sciences Division,
under contract DE-AC02-05CH11231.
’ REFERENCES
(1) (a) Vriezema, D. M.; Aragones, M. C.; Elemans, J. A. A. W.;
Cornelissen, J. J. L. M.; Rowan, A. E.; Nolte, R. J. M. Chem. Rev. 2005,
105, 1445–1489. (b) Koblenz, T. S.; Wassenaar, J.; Reek, J. N. H. Chem.
Soc. Rev. 2008, 37, 247–262.
(2) (a) Tung, C. H.; Wu, L. Z.; Zhang, L. P.; Chen, B. Acc. Chem. Res.
2003, 36, 39–47. (b) Yang, Q.; Han, D.; Yang, H.; Li, C. Chem.ꢀAsian
J. 2008, 3, 1214–1229.
(3) (a) Fukuda, T.; Yamauchi, S.; Honda, Z. Opt. Mater. 2009, 32,
207–211. (b) Crescimbeni, M. C.; Nolan, V.; Clop, P. D.; Marin, G. N.;
Perillo, M. A. Colloids Surf., A 2010, 76, 387–396. (c) Kin, E.; Fukuda, T.;
Yamauchi, S.; Honda, Z.; Ohara, H.; Yokoo, T.; Kijima, N.; Kamata, N.
J. Alloys Compd. 2009, 480, 908–911. (d) Gill, I.; Ballesteros, A. J. Am.
Chem. Soc. 1998, 120, 8587–8598.
(4) (a) Murase, F.; Horiuchi, S.; Fujita, M. J. Am. Chem. Soc. 2010,
132, 2866–2867. (b) Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew.
Chem., Int. Ed. 2009, 48, 3418–3438. (c) Iwasawa, T.; Hooley, R. J.;
Rebek, J., Jr. Science 2007, 317, 493–496. (d) Ziegler, M.; Brumaghim,
J. L.; Raymond, K. N. Angew. Chem., Int. Ed. 2000, 39, 4119–4121.
(e) Iwasawa, T.; Mann, E.; Rebek, J. J. Am. Chem. Soc. 2006, 128,
9308–9309. (f) Fiedler, D.; Bergman, R. G.; Raymond, K. N. Angew.
Chem., Int. Ed. 2004, 43, 6748–6751. (g) Yoshizawa, M.; Kusukawa, T.;
Fujita, M.; Yamaguchi, K. J. Am. Chem. Soc. 2000, 129, 6311–6312.
(h) Chen, J.; Rebek, J., Jr. Org. Lett. 2002, 4, 327–329. (i) Mal, P.;
Breiner, B.; Rissanen, K.; Nitschke, J. R. Science 2009, 324, 1697–1699.
(j) Butterfield, S. M.; Rebek, J., Jr. Chem. Commun. 2007, 1605–1607.
(k) Sato, S.; Iida, J.; Suzuki, K.; Kawano, M.; Ozeki, T.; Fujita, M. Science
2006, 313, 1273–1275. (l) Cram, D. J.; Tanner, M. E.; Thomas, R.
Angew. Chem., Int. Ed. Engl. 1991, 30, 1024–1027. (m) Yoshizawa, M.;
Fujita, M. Pure Appl. Chem. 2005, 77, 1107–1112.
(5) (a) Lakshminarasimhan, P.; Thomas, K. J.; Brancaleon, L.;
Wood, P. D.; Johnston, L. J.; Ramamurthy, V. J. Phys. Chem. B 1999,
103, 9247–9254. (b) Jayathirtha, R. V.; Prevost, N.; Ramamurthy, V.;
Kojima, M.; Johnston, L. J. Chem. Commun. 1997, 22, 2209–2210.
(6) (a) Yoshizawa, M.; Tamura, M.; Fujita, M. J. Am. Chem. Soc.
2004, 126, 6846–6847. (b) Pluth, M. D.; Bergman, R. G.; Raymond,
K. N. Science 2007, 316, 85–87. (c) Sato, S.; Iida, J.; Suzuki, K.; Ozeki, T.;
Fujita, M. Science 2007, 313, 1273–1276. (d) Murase, T.; Sato, S.; Fujita,
M. Angew. Chem., Int. Ed. 2007, 46, 1083–1085. (e) Kawano, M.;
Kobayashi, Y.; Ozeki, T.; Fujita, M. J. Am. Chem. Soc. 2006, 128,
6558–6559. (f) Dong, V. M.; Fiedler, D.; Carl, B.; Bergman, R. G.;
Raymond, K. N. J. Am. Chem. Soc. 2006, 128, 14464–14465. (g) Rebek,
J., Jr.; Ajami, D. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 16000–16003.
(h) Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Science 2007,
316, 85–88. (i) Pluth, M. D.; Fiedler, D.; Mugridge, J. S.; Bergman, R. G.;
Raymond, K. N. Proc. Natl. Acad. Sci. U.S.A. 2009, 26, 10438–10443.
(j) Balomenou, I.; Pistolis, G. J. Phys. Chem. B 2010, 114 (2), 780–785.
(k) Yoshizawa, M.; Kusukawa, T.; Fujita, M.; Yamaguchi, K. J. Am. Chem. Soc.
2000, 122, 6311–6312.
the solid hostꢀguest complex. UV-irradiation of host 1 coumarin
3
afforded the anti-HH coumarin photodimer with unusually high
selectivity. Longer UV-irradiation times resulted in higher conversion,
in contrast to what is typically observed in solid-state photoreaction of
coumarin, which gives limited conversion due to the photoreversion
of the photodimer to coumarin. We are currently exploring the uptake
and subsequent photoreaction of larger coumarin and stilbene
derivatives that may be better matched to the size and shape of the
channel. We are also investigating the crystallization of host 1 in the
presence of coumarin to see if we can grow crystalline inclusion
complexes suitable for X-ray crystallography. Such structural char-
acterization will help to evaluate the accuracy of our molecular
models. X-ray analysis of such inclusion crystals after intervals of
UV-irradiation might reveal the course of events between light
absorption and guest reaction and help to determine the origin of
the observed selectivity.
’ ASSOCIATED CONTENT
S
Supporting Information. Synthesis and characterization
b
of host 1 and its inclusion complexes; photochemical reaction of
coumarin in the presence and absence of host 1, and X-ray
crystallographic files in CIF format for 1. This material is available
free of charge via the Internet at http://pubs.acs.org.
(7) (a) Chretien,M.N.Pure Appl. Chem. 2007,79,1–20. (b) Hashimoto,
S. J. Photochem. Photobiol., C 2003,4,19–49.(b)Garcia,H.;Roth,H.D.Chem.
Rev. 2002, 102, 3947–4007. (c) Yoon, K. B. Chem. Rev. 1993, 93, 321–339.
(8) Shimizu, L. S.; Smith, M. D.; Hughes, A. D.; Shimizu, K. D. Chem.
Commun. 2001, 1592–1593.
’ AUTHOR INFORMATION
Corresponding Author
shimizul@chem.sc.edu
(9) (a) Yang, J.; Dewal, M. B.; Shimizu, L. S. J. Am. Chem. Soc. 2006,
128, 8122–8123. (b) Yang, J.; Dewal, M. B.; Profeta, S.; Smith, M. D.; Li,
Y.; Shimizu, L. S. J. Am. Chem. Soc. 2008, 130, 612–621. (c) Dewal, M. B.;
Xu, Y.; Yang, J.; Mohammed, F.; Smith, M. D.; Shimizu, L. S. Chem.
Commun. 2008, 3909–3911.
(10) (a) Feng, X.; Liu, L.; Wang, S.; Zhu, D. Chem. Soc. Rev. 2010,
39, 241102419. (b) Dore, K.; Leclerc, M.; Boudreau, D. J. Fluoresc. 2006,
16, 259–265.
’ ACKNOWLEDGMENT
The authors gratefully acknowledge support for this work
from the NSF (CHE-0718171 and CHE-1012298). Synchrotron
data were collected through the SCrALS (Service Crystallogra-
phy at Advanced Light Source) program at Beamline 11.3.1 at the
Advanced Light Source (ALS), Lawrence Berkeley National
Laboratory. The ALS is supported by the U.S. Department of
(11) (a) Thomas, S. W.; Joly, G. D.; Swager, T. M. Chem. Rev. 2007,
107, 1339–1386. (b) Pecher, J.; Mecking, S. Chem. Rev. 2010,
7031
dx.doi.org/10.1021/ja110779h |J. Am. Chem. Soc. 2011, 133, 7025–7032

Journal of the American Chemical Society
ARTICLE
110, 6260–6279. (c) McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem.
Rev. 2000, 100, 2537–2574.
(30) (a) Murthy, G. S.; Arjunan, P.; Venkatesan, K.; Ramamurthy, V.
Tetrahedron 1987, 43, 1225–40. (b) Bhadbhade, M. M.; Murthy, G. S.;
Venkatesan, K.; Ramamurthy, V. Chem. Phys. Lett. 1984, 109, 259–263.
(31) (a) Pemberton, B. C.; Barooah, N.; Srivatsava, D. K.; Sivaguru,
J. Chem. Commun 2010, 46, 225–227.(b) Turro, N. J. . In Modern
Molecular Photochemistry; University Science Books: Sausalito, CA,
1991; pp 462ꢀ464. (c) Lewis, F. D.; Barancyk, S. V. J. Am. Chem. Soc.
1989, 111, 8653–8661. (d) Lewis, F. D.; Howard, D. K.; Oxman, J. D.
J. Am. Chem. Soc. 1983, 105, 3344–3345. (e) Moorthy, J. N.; Venkatesan,
K.; Weiss, R. G. J. Org. Chem. 1992, 57, 3292–3297. (f) Muthuramu, K.;
Ramnath, N.; Ramamurthy, V J. Org. Chem. 1983, 48, 1872–1876.
(h) Wolff, T.; G€orner, H. J. Photochem. Photobiol., A 2010, 209, 219–223.
(32) Karthikeyan, S.; Ramamurthy, V. J. Org. Chem. 2006,
71, 6409–6413.
(12) (a) Mangel, T.; Eberhardt, A.; Scherf, U.; Bunz, U. H. F.;
Mullen, K. Macromol. Rapid Commun. 1995, 16, 571–580. (b) Nelson,
J. C.; Saven, J. G.; Moore, J. S.; Wolynes, P. G. Science 1997,
277, 1793–1796. (c) Brunsveld, L.; Meijer, E. W.; Prince, R. B.; Moore,
J. S. J. Am. Chem. Soc. 2001, 123, 7978–7984.
(13) (a) Zhao, D.; Moore, J. S. Chem. Commun. 2003, 807–818.
(b) Zhu, A.; Bharathi, P.; Drickamer, H. G.; Moore, J. S. J. Polym. Sci.,
Part A: Polym. Chem. 2001, 39, 2859–2865. (c) Zhu, A.; Bharathi, P.;
White, J. O.; Drickamer, H. G.; Moore, J. S. Macromolecules 2001,
34, 4606–4609. (d) Swallen, S. F.; Kopelman, R.; Moore, J. S.; Devadoss,
C. J. Mol. Struct. 1999, 485, 585–597. (e) Buchko, C. J.; Wilson, P. M.;
Xu, Z.; Zhang, J.; Moore, J. S.; Martin, D. C. Polymer 1995,
36, 1817–1825.
(33) Wolff, T.; Gorner, H. Phys. Chem. Chem. Phys. 2004, 6, 368–376
(14) (a) Hoger, S.; Morrison, D. L.; Enkelmann, V. J. Am. Chem. Soc.
2002, 124, 6734–6736. (b) Hoeger, S. J. Polym. Sci., Part A: Polym. Chem.
1999, 37, 2685–2698.
(15) (a) Luo, J.; Yan, Q.; Zhou, Y.; Li, T.; Zhu, N.; Bai, C.; Cao, Y.;
Wang, J.; Pei, J.; Zhao, D. Chem. Commun. 2010, 46, 5725–5727.
(b) Yamaguchi, Y.; Yoshida, Z. Chem.—Eur. J. 2003, 9, 5430–5440.
(c) Grave, C.; Schluter, A. D. Eur. J. Org. Chem. 2002, 3075–3098.
(16) Zang, L.; Che, Y. K.; Moore, J. S. Acc. Chem. Res. 2008, 41,
1596–1608.
and references therein.
(34) (a) Arai, T.; Sakuragi, H.; Tokumaru, K. Bull. Chem. Soc. Jpn.
1982, 55, 2204. (b) Rockley, M. G.; Salisbury, K. J. Chem. Soc. Perkin
Trans. 2 1973, 1582.
(17) Laugier, J.; Bochu, B. POUDRIX, a suite of programs for the
interpretation of X-ray experiments, ENSP/LMGP, BP 46, 38042 Sain
Martin d’Hꢀeres, France, 2000, http://www.inpg.fr/LMPG and http://
www.ccp14.ac.uk/tutorial/lmgp/poudrix.htm.
(18) (a) Barton, T. J.; Bull, L. M.; Klemperer, W. G.; Loy, D. A.;
McEnaney, B.; Misono, M.; Monson, P. A.; Pez, G.; Schere, G. W.;
Varuli, J. C.; Yaghi, O. M. Chem. Mater. 1999, 11, 2633–2656. (b) Groen,
J. C.; Peffer, L.; Perez-Ramirez, J. A. Microporous Mesoporous Mater.
2003, 60, 1–17.
(19) Wang, R.; Bardelang, D.; Waite, M.; Udachin, K. A.; Leek, D. M;
Yu, K.; Ratcliffe, C. I.; Ripmeester, J. A. Org. Biomol. Chem. 2009, 7,
2435–2439.
(20) Spartan 04 for Macintosh, V. 1.1.1.;Wavefunction, Inc.: Irvine,
CA, 2007.
(21) Sieval, B. A.; Hout, B. V. D.; Zulihof, H.; Sudholter, E. J. R.
Langmuir 2001, 17, 2172–2181.
(22) (a) Kzos, J. S.; Sommer, J. U. Macromolecules 2009, 42, 4878–4886.
(b) Meirovitch, H.; Vasquez, M. J. Mol. Struct. 1997, 398ꢀ399, 517–522.
(23) (a) Meyer, E. A.; Castellano, R. K.; Diederich, F. Angew. Chem.,
Int. Ed. 2003, 42, 1210–1250. (b) Coates, G. W.; Dunn, A. R.; Henling,
L. M.; Dougherty, D. A.; Grubbs, R. A. Angew. Chem., Int. Ed. Engl. 1997,
36, 248–251.
(24) (a) Krauch, H.; Farid, S.; Schenck, G. O. Chem. Ber. 1966,
99, 625–631. (b) Hoffman, R.; Wells, P.; Morrision, H. J. Am. Chem. Soc.
1971, 36, 102–107.
(25) (a) Vishnumurthy, K.; Guru Row, T. N.; Venkatesan, K.
J. Chem. Soc., Perkin Trans. 2 1997, 615–619. (b) Vishnumurthy, K.;
Guru Row, T. N.; Venkatesan, K. J. Chem. Soc., Perkin Trans. 2 1996,
1475–1478. (c) Craig, D. P.; Mallet, C. P. Chem. Phys. 1982,
65, 129–142. (d) Craig, D. P.; Lindsay, R. N.; Mallet, C. P. Chem. Phys.
1984, 89, 187–197. (e) Bhadbhade, M. M.; Murthy, G. S.; Venkatesan,
K.; Ramamurthy, V. Chem. Phys. Lett. 1984, 109, 259–263. (f) Gnanaguru,
K.; Murthy, G. S.; Venkatesan, K.; Ramamurthy, V. Chem. Phys. Lett.
1984, 109, 255–258.
(26) Yu, X.; Scheller, D.; Rademacher, O.; Wolff, T. J. Org. Chem.
2003, 68, 7386–7399.
(27) Morrison, H.; Curtis, H.; McDowell, T. J. Am. Chem. Soc. 1966,
88, 5415–5419.
(28) (a) Gnanaguru, K.;Ramasubbu, N.;Venkatesan, K.;Ramamurthy,
V. J. Org. Chem. 1985, 50, 2337–2346. (b) Tanaka, K.; Toda, F. Chem.
Rev. 2000, 100, 1025–1074. (c) Barooah, N.; Pemberton, B. C.;
Sivaguru, J. Org. Lett. 2008, 10, 3339–3342.
(29) Wen, Y.; Song, Y.; Zhao, D.; Ding, K.; Bian, J.; Zhang, X.; Wang,
J.; Liu, Y.; Jiang, L.; Zhu, D. Chem. Commun. 2005, 2732–2734.
7032
dx.doi.org/10.1021/ja110779h |J. Am. Chem. Soc. 2011, 133, 7025–7032