Photophysics Applied to Cavitands and Capsules

Abstract

The use of light as a stimulus to control functional materials or nano-devices is appealing as it provides convenient control of triggering events where and when they are desired without introducing extra components to the system. Many photophysical and photochemical processes are extremely fast, giving rise to nearly instantaneous onset of events. However, these fast processes can be challenging to engineer into chemical systems. Supramolecular chemistry offers a convenient way to study and control photoprocesses. Given the reversible and self-programmed nature of modern host-guest systems, a modular approach can be considered in which different photoprocesses are coupled to obtain complex functions that emerge and are controlled solely by light inputs. In this review, we highlight recent examples of photoswitching and photophysics applied in the context of supramolecular host-guest systems, with a particular emphasis on resorcinarene based cavitands and hydrogen bonded capsules.

“…more moving parts than a silo full of Swiss watches” – Neal Stephenson

1. Introduction

Photo-physical and chemical processes can occur on the femtosecond timescale and these ultra-fast processes can be difficult to study and control. The use of constricting environments to alter photoprocesses and obtain new and useful functions has long captured the imagination of chemists. Supramolecular host-guest chemistry is particularly amenable to enable this concept; many different host molecules—micelles, zeolites, cucurbiturils, metal/ligand assemblies and cyclodextrins—have been used to study photophysics in confined spaces. A large body of work has been devoted to this subject and many excellent reviews and books detailing the studies within different host systems have been written.^{[1, 2]} Several photoprocesses have been studied inside host structures including sensing, photoswitching, photochemistry and photophysics. A detailed report covering all these studies is beyond the scope of this review. Rather, we limit this review to systems exhibiting encapsulation in discrete cavities. We begin by discussing photochemical control of encapsulation with a historical perspective and highlight early photophysical studies inside cyclodextrin hosts. Recent studies pertaining to photoswitching and photophysics inside cavitands and capsules are subsequently presented. Applying photophysics to host systems is an exciting route to molecular devices and machines; the reader is directed to excellent books and reviews on the subject.^[3] Ultimately, we are interested in preparing devices where at least two photoprocesses are integrated into single host systems. The focus of this review is on the photophysics of and within cavitands and hydrogen bonded capsules and our own recent contributions to this field will be highlighted.

2. Photochemical Control of Encapsulation

Early in the studies of encapsulation complexes it was apparent that controlling the assembly process after mixing the components together is challenging, especially if it is to be done reversibly. Initial examples by Shinkai and co-workers showed that photocontrolled ion binding was possible with azo-crown ethers.^{[4, 5]} However, controlling self-assembly processes remains a challenge. As recently as 2008 Jerry Atwood commented on the problem: “In particular, the challenge is in controlling the assembly-disassembly-reassembly process of the capsules efficiently and in the encapsulation of a non-solvent molecular guest in solution without the practical inconvenience of changing the solvent(s) or undertaking chemical transformations.”^[6] Using light inputs for the control of reversible encapsulation is one convenient way to achieve this goal, and two different strategies can be used for this purpose: changing the encapsulating ability of the host by incorporating photo-isomerizable groups into the host structure (Figure 1, top), which alter the host’s geometry or inner space upon irradiation or using isomerizable “dummy” guests which preoccupy the internal cavities of the host and can be expelled upon photo-isomerization (Figure 1, bottom). Ueno provided examples of both possibilities by controlling the dynamic encapsulation properties of cyclodextrins with azobenzenes as light switchable units, either appended to the cyclodextrin rim^[7] or as a guest molecule.^[8] Specifically, the outer rim of β-cyclodextrin (β-CD) was covalently capped with azobenzene-4,4′-dicarbonyl.^[7] The azobenzene can be photoisomerized between the cis and trans configuration—occupying different amounts of the inner cavity. Photoisomerization of the azobenzene cap allowed for modulation of the guest binding within the β-CD. The cis isomer has an enlarged cavity and in all but one case binds two guest molecules. The trans isomer has a smaller cavity preferring to bind one guest molecule at a time. 4,4′-bipyridine is too large to fit in the trans isomer and thus only exhibits binding in the cis isomer.

Figure 1.

Cartoon representation of two strategies for light controlled encapsulation. Photo-isomerizable unit (red) attached to the host (top) and photo-isomerizable ‘dummy’ guest (bottom).

Photoswitchable guest binding was also achieved by using a photoisomerizable guest molecule.^[8] In the trans configuration potassium p-(phenyl-azo)benzoate binds to β-CD. This guest molecule can be photoismerized to the cis configuration with a photostationary state of 55-65% cis isomer. The nonlinear geometry of the cis isomer causes it to bind weakly to β-CD. Photocontrol of guest binding with azo capped β-CD^[9] and azo inhibitor^[8] were both used to regulate the β-CD catalyzed hydrolysis of p-nitrophenyl acetate. When the trans azo inhibitor was present in solution the β-CD catalyzed hydrolysis was 2 times slower. In a similar manner the dimerization of cyclodextrins can be controlled with light.^[10]

These early studies were adapted by Reinhoudt and co-workers to manipulate the host properties of cyclodextrin dimers photochemically. To this end, dithienylethenes were used as switchable linker units between two β-CDs to control the encapsulation of large porphyrin guests, a 3 5-fold binding difference was found between the open and closed form of the cyclodextrin dimer.^[11] Similarly, the cooperativity of guest binding can be controlled by photoirradiation.^[11,12] Such systems could be of use in light controlled drug delivery as well as in phototherapy.

Mattay and co-workers studied single molecule switching with AFM by attaching photoresponsive cavitands to gold surfaces.^[13] Zhang and co-workers used amphiphilic azobenzenes, which can form micelles in both their trans- and their cis-conformation. On addition of α-cyclodextrin the micelles of the trans-azobenzenes are destroyed and the corresponding cyclodextrin host-guest complexes are formed. Photoswitching to the cis-conformation disrupts the cyclodextrin-azobenzene interaction and restores the micelles.^[14] Fujita and co-workers presented an intriguing example of altering the host property of a self-assembled capsule by irradiation. In this case, azobenzene units are attached to the inside of the host structure and photochemical trans- to cis-isomerization increases the size of the internal cavity to allow binding of a pyrene guest.^[15]

In a recent example, Shionoya and co-workers used the light-triggered removal of an azobenzene guest from a metal-organic cage to control the crystallization behavior of that host (Figure 2).^[16] In its cis-conformation the bis-sulfonate azobenzene is a formidable guest for the cage structure. After photoisomerization to the trans form it is expelled from the cavity and triggers crystallization by joining the host cages via their Pd(II) metal centers. In this way a light triggered phase change was achieved, which might be useful for nano-construction or lithographic surface manipulation.

Figure 2.

Light triggered removal of an azobenzene guest induces crystallization of the metal-organic host (grey spheres).

From the above examples a clear progression from simply controlling the host-guest binding with light, to using that control to manipulate properties—such as vesicle or crystal formation and functions such as fluorescence (see Section 5)—can be anticipated. Smart materials that can be addressed by light stimuli are becoming increasingly complex and their applications are more advanced. So far, azobenzenes are used in the great majority of systems but other photoswitchable units will become important in the future, especially if multiple inputs are to be used simultaneously to achieve ever more sophisticated functions.

3. Photophysics in Cavitands and Capsules

To create and study new devices we are interested in integrating multiple photoprocesses into single host molecules. Besides controlling the recognition event with light, photochemical and photophysical processes are two valuable subjects to study within host structures. Entrapping photo-reactive molecules is one strategy to alter photochemical processes and there have been many exciting examples recently published.^{[2, 17]} This section will not address photochemical studies, but rather, highlight photophysical studies within cavitands and capsules.

One early photophysical study with cyclodextrins comes from Turro and co-workers where the effect of encapsulation on halonaphthalene phosphorescence was studied.^[18] Upon inclusion in β-CD the triplet lifetime and intensity of phosphorescence was increased. Furthermore, the encapsulated halonaphthalene was not quenched by addition of NaNO₂. Early investigations such as this, paved the way for many more studies ranging from sensing to catalysis. The following sections will highlight recent examples of photophysical properties altered by host/guest assemblies.

3.1. Controlling Excimer Formation by Encapsulation

The water-soluble dimeric capsule 1₂ produced by Gibb and co-workers has offered many unique opportunities to study photophysics inside structurally rigid confines. In a recent example, hydrophobic guests such as naphthalene, anthracene and tetracene readily are taken up by the capsule.^[19] NMR spectroscopy revealed that the smaller naphthalene and intermediate size anthracene form 2:1 complexes with the host whereas only one larger tetracene molecule fits within the host.

The photophysics of these three guests inside the capsule are remarkably different. Models suggest that two naphthalene molecules have enough room within the capsule to adopt multiple arrangements. As a result, a solution of naphthalene and capsule 1₂ in aqueous borate buffer exhibited both monomer and excimer emission, with lifetimes of 62 ns and 74 ns respectively. Even with the increase in effective concentration within the capsule, the small naphthalene molecules have room to dissociate producing both monomer and excimer emissions.

Only one of the larger tetracene molecules fits within capsule 1₂. Once encapsulated, the tetracene monomer is isolated from other tetracene molecules and the only emission produced is from the monomer.

The photoexcitation of anthracene typically produces the anthracene dimer with predominantly monomer emission and very weak excimer emission (530 nm, with a lifetime less than 2 ns). In contrast, the photophysics of anthracene excitation in this system highlights the importance of the small space. ¹H NMR spectroscopy revealed that two anthracene molecules are encapsulated and held out of register to one another. After encapsulation, the monomer fluorescence is replaced with broad emission corresponding to the excimer (Figure 3). The long lifetime of this emission (τ = 263 ns) and matching absorption and excitation spectra suggest the formation of a π-stacked “sandwich excimer.” Detention in the capsule stabilizes this unique excitation and no evidence of dimerization is observed after 10 hours of irradiation. These studies show that encapsulation can significantly alter normal photophysical processes. The end results are dictated by the size and shape of the host cavity.

Figure 3.

Octaacid monomer 1 (top). Emission of free anthracene (a) and encapsulated in dimeric capsule 1₂ (b).

3.2. Encapsulating Fluorescent Protein Chromophores

Fluorescent proteins (FP) have attracted substantial interest for their biological and emissive properties. The discovery and development of these molecules led to the 2008 Nobel Prize in chemistry. The fluorescent properties arise from the chromophores that are entrapped within the β-barrel of the protein. To further study the effects of encapsulation, researchers have prepared fluorescent protein model chromophores such as benzylidene-3-methylimidazolidinones (BMI’s).^[20] In borate buffered solution, BMI’s are encapsulated in dimer 1₂. When BMI’s are free in solution they undergo facile cis/trans photoisomerization reaching a photostationary state favoring the cis isomer. However, encapsulation affords the trans isomer; packing of this isomer is more efficient in the capsule. A BMI with an ortho methyl substituent showed enhanced emission upon encapsulation. This result was rationalized by inhibition of single bond rotation inside the dimeric capsule 1₂. These results nicely parallel those found with FPs that showed internal conversion and loss of fine structure comes primarily from low frequency motions. Further studies have elaborated that stereoelectronic effects contribute to the turn-on emission of FP chromophores inside capsule 1₂.^[21] These studies suggest that control of emission properties by encapsulation could be a general approach to molecular probes.

3.3. A Cage that Stabilizes Azoporphine Fluorescence

Fujita and co-workers have recently realized many exciting photophysical and photochemical studies inside porous coordination cages.^{[2, 22]} Of particular interest is a recent report that details the sequestration of azoporphine (TAP, Figure 4) inside a Pt coordination cage (2, Figure 4) built from pyrazine and tripyridyl triazine. TAP is a porphyrin derivative with appealing red-fluorescent emission that suffers from poor water solubility and quenching from self-aggregation or solvent molecules.

Figure 4.

Coordination cage 2 (left) and red emitting azoporphine (right).

Coordination cage 2 contains a hydrophobic cavity with two electron-deficient panels suitably spaced for intercalation.^[23] Upon intercalation, aromatic guests typically lose their emission through energy transfer to the low-lying LUMO of the triazine ligands. In contrast, the red fluorescence of TAP persists when bound in 2 with a quantum yield ϕ_f = 0.17 and an increase in emission lifetime to 5.9 ns. Once encapsulated, TAP becomes highly water-soluble and the red emission is impervious to “self” and to known external quenchers such as DMF and DMSO.

The red emission of the encapsulated complex can be turned off by addition of a weak base such as triethylamine (TEA). Through a partial or full disassociation process the interior TAP protons are deprotonated by TEA and the anion is stabilized inside the host structure. The resulting anion forms new charge-transfer bands with the host at 450 and 675 nm. The red fluorescence can be regenerated by addition of nitric acid. These results suggest that coordination cages could be used to create new materials with red-emissive properties or for radiative decay engineering with supramolecular interactions.

4. Deep Cavitands with an Azobenzene Wall

Some of the earliest examples photophysical studies in host molecules pertain to photocontrol of guest binding.^{[5, 7]} We strove to achieve spatial and temporal control of guest recognition in cavitands with the ultimate aim of controlling the properties of the system with light. To create a photoswitchable cavitand we took an alternative approach by incorporating the photoismerizable azobenzene molecule as one of the walls of the cavitand (Figure 5).^[24]

Figure 5.

Cavitands 3-6 functionalized with an azobenzene-wall.

Azo substituted cavitands are synthesized from mononitro substituted cavitands in one step by reducing the nitro group with Pd/C followed by immediate condensation with aromatic nitroso compounds. The compounds are isolated as orange solids as a mixture of cis and trans isomers. By varying the aliphatic substitution meta to the azo functionality we determined the importance of weak secondary interactions in this system.

The thermal and photoisomerization of azo cavitands 3 and 6 was analyzed in d₁₂-mesitylene. Heating the cavitands to 164 °C for 5 minutes completely converts the azobenzenes to their trans configuration. Under ambient light the cavitands reach a photostationary state ranging from 92 to 95% trans. In the trans configuration azo cavitands 3-6 adopt a folded vase conformation suitable for guest binding. This conformation is revealed by ¹H NMR spectroscopy where the methine protons are shifted downfield to >6 ppm. The cis configuration is achieved by irradiating with 365 nm light for 15 minutes reaching a photostationary state of >95%. The conformation that the cavitand adopts is dependent on the substitution meta to the azo functionality. For instance, only when the meta substitution appropriately fills the interior volume of the cavitands (5 and 6) do we observe 1H NMR signals corresponding to an introverted conformation.^[25]

Next the guest binding behaviour of trans azo cavitands 3 and 6 was studied. Both azo cavitands bind neutral adamantane guests with association constants ranging from 5 to 700 mol⁻¹. The strongest binding is observed for guests that can attractively interact with the amide functionality on the upper rim of the cavitand, such as 2-adamantanone and 1-adamantane carbonitrile.

When a guest molecule is present, photoisomerization of the azo wall has a dramatically different affect depending on the substitution on the azo wall. Photoisomerization of azo cavitand 3—where no substitution is present—does not influence the concentration of bound guest molecules. This is in striking contrast to tert-butyl-azo cavitand 6 where photoisomerization of the azo wall completely removes the bound guest from the cavitand. The bound guest is replaced with an introverted tert-butyl. Control of the guest association is completely reversible and can be achieved by heating the sample to 164 °C for 5 minutes. Alternatively, the sample can be irradiated with 440 nm light for 20 minutes reaching a photostationary state of 71% trans.

Deep cavitands that can be manipulated with light present exciting opportunities for altering chemical properties. With control of the guest binding demonstrated, this system is poised for elaboration to responsive functional molecules. Our current efforts are directed at using light to control organocatalysis with azo cavitands and will be reported elsewhere.

5. Photophysics in H-Bonded Capsules

Self-assembled hydrogen bonded capsular assemblies have been studied in our lab for nearly two decades and many unique properties have emerged from the interactions of guest molecules with these hosts. Our entry into the field of photochemistry started in 2000 with the application of Fluorescence Resonance Energy Transfer (FRET) to study mechanistic aspects of multi-component assemblies in real time.^[26] In this initial study the dimerization of calix[4]arenes into capsular assemblies was studied: the components were functionalized with ureas on their wider rims and outfitted with FRET donors and acceptors on their narrow rims and the kinetics and thermodynamics of assembly were revealed. Additionally, the FRET signal was shown to be a useful tool for the detection of guest molecules in the same study. Since then, FRET has been used to study the self-assembly process for a variety of capsular assemblies such as cylindrical dimers,^[27] resorcinarene hexamers,^[28] or hybrid capsular assemblies.^[29]

We expected that the unique shape of the inner space of our host capsule 7·7 (nonspherical) as well as its chemical nature, i.e. electron rich aromatic surfaces at the end and electron poor aromatic surfaces of the panels might give rise to unprecedented effects regarding the photophysical or photochemical behavior of guest molecules. That expectation was met in the initial observation that encapsulation of tetrathiofulvalene (TTF) in the hydrogen bonded capsule 7·7 leads to a charge transfer complex between the electron rich guest and the electron poor capsule walls. This interaction causes a significantly broadened absorption spectrum with a shifted maximum wavelength of 467 nm.^[30]

In a follow-up study, the effect of stilbene encapsulation on the fluorescence properties of the resulting solution was investigated. Drawing analogies to the significant fluorescence enhancement observed for inclusion of stilbenes into antibody “hosts”^[31] we expected a similar behavior in our capsule. To our surprise the opposite performance was observed: the moderate fluorescence of 4-methyl-4′-ethylstilbene (8) in solution was completely quenched after encapsulation. The reason for the different behavior can be traced to the intricacies of the inner space of capsule 7·7. Its geometry can be visualized as a square pyramid—the resorcinarene at the end of the capsule—and a square prism—consisting of the four capsule walls. The square prisms are twisted by 45 ° when the two capsule halves come together, which gives rise to a contorted inner space. To maximize interactions with the host structure the stilbene guest molecule 8 traces that twisted conformation once it is encapsulated, which reduces its conjugation leading to efficient fluorescence quenching (Figure 7).^[32]

Figure 7.

a) Cavitand 7 forms the dimeric capsule 7·7 upon encapsulation of 4-Methyl-4′-ethylstilbene (8). b) trans-8 adopts a twisted conformation in 7·7. Its fluorescence is efficiently quenched.

Quite different behavior was found when 4,4′-dimethylbenzil is used as a guest molecule for capsule 7·7.^[33] Its emission upon excitation with 318 nm light is very low in solution, due to its preferred cis-skewed conformation. Encapsulation in 7·7 forces the benzil to adopt a different trans-conformation to fit into the host. This conformational change leads to a remarkable increased emission centered at 560 nm, which is characteristic for benzil phosphorescence from a trans-planar triplet state. Again, the unique inner space of host 7·7 leads to conformational restrictions of the guest molecule that affects its excited state characteristics. Such emission changes upon encapsulation are especially useful in the context of sensing applications for small organic molecules.

If we are to craft devices, which are based on encapsulation phenomena that can be operated by light to exert complex functions, the process of encapsulation has to be controlled by light inputs. In doing so, the guest molecule properties can be switched on or off depending on whether the guest molecule is taken up by the host or resides in solution. As outlined in Section 2, there are two different ways of achieving photocontrol over host-guest binding and Ueno provided early examples of both using cyclodextrins as the host.^{[7, 8]} However, capsule 7·7 is very different from a cyclodextrin: the capsule more or less completely surrounds its guests and shields them from solvent access, unlike the open-ended cyclodextrin structures. The hydrogen bonding that holds the capsule halves together gives rise to kinetically stable assemblies and slow guest exchange with half-lives of the encapsulated species reaching many hours to days. In cyclodextrins, guest exchange is fast, typically with half-lives in the microsecond range.^[34] Therefore, isomerization of an azobenzene dummy guest can take place outside of the cyclodextrin host. Once the guest is isomerized and free in solution its return to the cavity is inhibited. In agreement with this mechanism, it has been shown that restriction of azobenzene guests to the inside of cyclodextrin hosts inhibits their isomerization.^[35] Because of the greater stability of our host-guest capsular assemblies it was uncertain if we could apply a photoswitchable dummy guest to control the uptake of a second guest, and what the actual mechanism for such a process would be.

Initially, we used 4,4′-dimethylazobenzene (9) as the photoswitchable guest in the presence of the mediocre guest tridecane for binding to the capsule 7·7. Brief heating of the components facilitates assembly and ensures all azobenzene 9 is isomerized into its trans-conformation; it is the only guest encapsulated. That situation changes after irradiating the sample with 365 nm light for 50 minutes. Now tridecane is encapsulated and azobenzene 9 is found in its cis-conformation, outside of the capsule. Heating the solution to 160 °C for 2 minutes returns the system to the starting state as shown in Figure 8, with isomerization to trans-azobenzene and its uptake in the capsule.^[36] The heating-irradiation-heating cycle can be repeated many times without loss of performance of the system. To unravel the mechanism of guest exchange under irradiation conditions, we compared the rates of guest exchange under irradiation and in the dark. Under irradiation the exchange of azobenzene with a 30-fold excess of tridecane was finished after 50 min but in the dark the guest exchange reached equilibrium only after about 1600 min, reflecting the slow guest exchange of well fitting guests (i.e. trans-4,4′-dimethylazobenzene (9) in this case) in capsule 7·7. Thus, the normal guest exchange of trans-9 and its subsequent isomerization in solution is not the likely mechanism for guest exchange under irradiation. Rather the azobenzene has to actively “break out” of the capsule during irradiation to achieve the considerably faster exchange rates we observed.

Figure 8.

Photochemical control of reversible encapsulation of tridecane in hydrogen bonded capsule 7·7 by switchable 4,4′-dimethylazobenzene (9). ¹H NMR traces 1 and 3 show encapsulated trans-9 after heating of the solution, ¹H NMR trace 2 shows encapsulated n-tridecane after irradiation with 365 nm light.

It is also possible to return the system to its starting point by irradiation of the cis-azobenzene (9) with 450 nm light at 20 °C. The reason we used heat instead of 450 nm light is solely convenience, since the guest exchange kinetics are much slower at 20 °C and the rate determining step in this case is not the light driven back-isomerization of azobenzene 9 into its trans-form but the subsequent slow kinetics of guest exchange. Using the 4,4′-dimethylazobenzene switch, a variety of guest molecules such as 4, 4′-dimethylbenzil or hydrogen bonded dimers of benzoic acid or benzamide can be transferred in and out of capsule 7·7 by light and heat. If comparably good guests are used, more equivalents of the azobenzene are required to suppress their binding after the heating step.

Photochemical control of encapsulation in the extended capsular assembly 7·10₄·7 was next investigated. The extended assembly is formed by insertion of a belt of 4 soluble glycoluril spacer units (10) between the two capsule halves. Longer guests had to be used for this extended capsule and the asymmetric 4-methyl-4′-hexylazobenzene (11) was determined to be the best fitting among a series of elongated symmetrical (4,4′-dibutyl- or dipropyl- or di-tert-butyl) and asymmetrical (4-methyl-4′-hexyl or pentyl and 4-ethyl-4′-hexyl or pentyl) substituted azobenzenes. With this guest the encapsulation of two molecules of p-cymene or hydrogen bonded dimers of 4-ethylbenzamide in 7·10₄·7 can be controlled in a reversible way by light and heat as shown in Figure 9.

Figure 9.

a) Chemical structures of glycoluril 10 and long azobenzene trans-11. b) Photocontrol of reversible guest exchange of 4-ethylbenzamide in the extended capsule 7·10₄·7 using trans-11 as dummy-guest. ¹H NMR traces 1 and 3 show encapsulated trans-11 after heating of the solution, ¹H NMR trace 2 shows encapsulated hydrogen bonded dimer of 4-ethylbenzamide after irradiation with 365 nm light.

Having found a convenient way of controlling the encapsulation state of a guest by light stimuli, we now turned our attention to the assembly itself. Would it be possible to control not only the guest but at the same time the nature of the assembly photochemically? The switchable azobenzene guests for capsule 7·7 and extended capsule 7·10₄·7 made for a straightforward attempt to interconvert these two assemblies. For this, we used 4-dodecanephenyl glycoluril 12 as a spacer unit with lower solubility in d₁₂-mesitylene. Despite its low solubility 12 is readily incorporated into an extended assembly with the long 4-methyl-4′-hexylazobenzene (trans-11) as guest.

We then added 4,4′-dibromobenzil (13) as a mediocre guest for the smaller capsule 7·7, and after heating, the extended assembly with encapsulated trans-11 is the only complex seen. On irradiation however, the only assembly found is encapsulated benzil 13 in capsule 7·7. A slight turbidity of the solution indicates that glycoluril 12 is expelled from the assembly. Another heating step recovers the extended assembly and the system can be cycled many times without loss of performance. Thus, we established not only control over the encapsulation state of a guest but at the same time control over the nature of its surrounding host using light and heat (Figure 10).

Figure 10.

a) Chemical structures of glycoluril 12 and 4,4′-dibromobenzil 13. b) Photoswitching between the extended assembly 7·12₄·7 and the smaller capsule 7·7 and concomitant guest exchange. ¹H NMR traces 1 and 3 show encapsulated trans-11 in 7·12₄·7 after heating of the solution, ¹H NMR trace 2 shows encapsulated 13 in 7·7 after irradiation with 365 nm light.

We have now established essential photochemical processes and phenomena to start planning their combination and integration into simple photonic devices. In a first example, the fluorescence quenching associated with encapsulation of 4-methyl-4′-ethylstilbene (8) was combined with the photochemical control of guest exchange by azobenzenes to obtain a supramolecular fluorescence switch operated by irradiation and heat inputs.^[37] To characterize the system accurately, we correlated fluorescence measurements of the solution with ¹H NMR spectroscopy of that very same solution to be able to assign changes in the supramolecular state to changes in fluorescence. When briefly heating 2 equivalents of 4,4′-dimethylazobenzene 9, 1 equivalent of capsule 7·7 and 1 equivalent of trans-4-methyl-4′-ethylstilbene (trans-8) the trans-azobenzene 9 is found encapsulated, stilbene 8 resides in solution and high fluorescence is measured for that solution. After irradiation at 365 nm, the ¹H NMR spectrum showed all stilbene encapsulated in 7·7 and no residual stilbene trans-8 in solution. The fluorescence of the solution is quenched by a factor of 15 to 20. Heating that solution to 160 °C for 2 min reverts the system to the starting point with trans-azobenzene 9 encapsulated, free trans-stilbene 8 in solution and high fluorescence upon excitation with 318 nm light. That system can be cycled many times and establishes a light responsive, reversible fluorescence switch (Figure 11).

Figure 11.

Photoswitching between unbound and bound stilbene trans-8 (top) and associated changes in stilbene fluorescence at 380 nm upon excitation with 318 nm light (bottom). Odd experiment numbers indicate heating steps, even numbers indicate irradiation with 365 nm light.

Although there is very good correlation of ¹H NMR data with fluorescence data, we still had to ascertain that the azobenzene-triggered control over stilbene encapsulation gives rise to the reversible behavior of the system. To this end, a series of control experiments were performed in which the contribution of each component of the system to the overall fluorescence could be deciphered. Again, fluorescence data were correlated to ¹H NMR data of the same solution, each after a heating step, an irradiation step, and another heating step. It was found that the stilbene fluorescence is quenched little by the capsule and somewhat more severely by the azobenzene. Interestingly, a combination of stilbene trans-8 and azobenzene 9 alone gives rise to fluorescence changes upon irradiation with 365 nm light and heat. The reason for this behavior is the higher quenching efficiency of azobenzenes in their trans-conformation as compared to their cis-conformers. However, that fluorescence change is only 2-fold compared to the 15-fold change in the full system and, more importantly, it is in the opposite direction. After irradiation an increase in fluorescence is observed instead of the quenching in the full system. Thus, in the full system the intrinsic response of stilbene and azobenzene combinations to irradiation with 365 nm light and heat is overwritten by the supramolecular effects of encapsulation. The fluorescence changes we observe after irradiation and heat are indeed under supramolecular control.

6. Summary and Outlook

The widespread activity in photophysical studies of supramolecular host/guest complexes shows it is an exciting field with many opportunities. Early systems showed that encapsulation could be used to influence guest photophysics and light could be used as a tool to control host/guest association. In future studies, controlling the encapsulation process will remain a challenge. The unique kinetics and thermodynamic factors of each system will likely preclude a universal photo-method for controlling guest encapsulation. Up to the present, the encapsulation process has been controlled with light by incorporating a photoswitch into the host or employing photoisomerizable guest molecules; new techniques for regulating encapsulation will surely surface.

A second important aspect is the advent of unique photophysical properties emerging upon encapsulation of photoactive molecules within confined spaces of host-guest complexes. Such studies have been used to understand the intricacies of these fleeting processes enabling us to manipulate and ultimately design photophysical properties. Clearly these examples show that future endeavours will move toward more sophisticated systems where encapsulation and light inputs work in tandem to create functional devices with a wide range of properties.

Figure 6.

CAChe AM1 molecular models depicting the open conformation of trans-6 (left) and the closed introverted conformation of cis-6 (right).

Biography

Following a vigorous upbringing in Homer, Alaska, Orion B. Berryman received his Ph. D. from the University of Oregon in 2008. Under the guidance of prof. Darren W . Johnson he investigated weak attractive inter actions between anions and electr on-deficient aromatic rings. Orion is currently a Ruth L. Kirschstein NIH fellow working at The Scripps Research Institute, La Jolla. As a postdoctoral fellow working with Professor Julius Rebek Jr., his primary project deals with light responsive cavitands.

Henry Dube was born in Germany and finished his Ph. D. in chemistry on synthetic models for heme proteins in the group of Prof. François Diederich at the ETH Zurich, Switzerland in 2008. Since then he is a Postdoctoral Fellow with Prof. Julius Rebek, Jr. at the Scripps Research Institute, La Jolla where his current interest lies in self-assembly and the photochemistry of encapsulation complexes.

Julius Rebek, Jr. is the Director of The Skaggs Institute for Chemical Biology at The Scripps Research Institute. He was born in Hungary and educated at the University of Kansas and the Massachusetts Institute of Technology. He held professorships at UCLA, the University of Pittsburgh and MIT before moving to La Jolla in 1996. His research interests include synthetic, self-replicating molecules, self-assembling systems, recognition phenomena and molecular behavior in small spaces.