Photochemical Molecular Devices

Encyclopedia of Nanoscience and Nanotechnology

www.aspbs.com/enn

Photochemical Molecular Devices Alberto Credi Università degli Studi di Bologna, Bologna, Italy

CONTENTS 1. Introduction 2. Pseudorotaxanes, Rotaxanes, and Catenanes 3. Artificial Photosynthetic Devices 4. Molecular-Level Machines 5. Molecular Devices for Information Processing 6. Conclusion Glossary References

1. INTRODUCTION In everyday life we make extensive use of macroscopic devices, which are assemblies of components designed to achieve a specific function. Each component of the device performs a simple act, while the entire device performs a more complex function, characteristic of the assembly. For example, the function performed by a hair dryer (production of hot wind) is the result of acts performed by a switch, a heater, and a fan, suitably connected by electric wires and assembled in an appropriate framework (Fig. 1a). The concept of a device can be extended to the molecular level [1–4]. A molecular-level device can be defined as an assembly of a discrete number of molecular components (i.e., a supramolecular structure) designed to achieve a specific function. Each molecular component performs a single act, while the entire assembly performs a more complex function, which results from the cooperation of the various molecular components (Fig. 1b). The extension of the concept of a device to the molecular level is of interest for the growth of nanoscience and the development of nanotechnology. Indeed, the miniaturization of components for the construction of useful ∗

This chapter first appeared in Handbook of Photochemistry and Photobiology, Volume 3: Supramolecular Photochemistry; Edited by H. S. Nalwa, © 2003, American Scientific Publishers.

ISBN: 1-58883-064-0/$35.00 Copyright © 2004 by American Scientific Publishers All rights of reproduction in any form reserved.

devices, which is an essential feature of modern technology, is currently pursued by the large-downward (top-down) approach [5]. This approach, however, which leads physicists and engineers to manipulate progressively smaller pieces of matter, has intrinsic limitations [6]. An alternative and promising strategy is offered by the small-upward (bottomup) approach [7]. Chemists, by the nature of their discipline, are already at the bottom, since they are able to manipulate molecules (i.e., the smallest entities with distinct shapes and properties) and are therefore in the ideal position to develop bottom-up strategies for the construction of nanoscale devices.

1.1. Role of Light in Molecular-Level Devices Light is made of photons, and photons are at the same time quanta of energy and bits of information [8]. This is evident in nature, where light constitutes an energy source and is consumed (converted) in massive amounts in photosynthesis, while it functions as a signal in vision-related processes, the energy used to run the operation being biologic in nature. Therefore, taking one of these two extreme views, one can think of molecular-level devices that use light as an energy supply to perform energy-expensive functions, for example, conversion into chemical energy. On the other hand, light could be used by a molecular device as an input/output signal to be processed and eventually stored and retrieved. In general, molecular-level devices that perform lightinduced functions, that is, in which photons act as an energy supply and/or input/output signals, can be termed photochemical molecular devices (PMDs) [2, 4, 9]. The role of light in reference to the relevant features of molecular-level devices will be discussed in more detail in the following section.

1.2. Characteristics of Molecular-Level Devices Molecular-level devices operate via electronic and/or nuclear rearrangements, that is, through some kind of chemical reaction. Like their macroscopic counterpart, they are Encyclopedia of Nanoscience and Nanotechnology Edited by H. S. Nalwa Volume 8: Pages (527–548)

528

Photochemical Molecular Devices a

b molecular components

SIMPLE ACTS

supramolecular assembly

COMPLEX FUNCTION

Figure 1. Schematic representation of the assembly of (a) a macroscopic device and (b) a supramolecular system capable of performing as a molecular-level device.

characterized by (a) the kind of energy input supplied to make them work, (b) the way in which their operation can be monitored, (c) the possibility of repeating the operation at will (cyclic process), (d) the time scale needed to complete a cycle, and (e) the performed function. As far as point (a) is concerned, the most obvious way to supply energy to a chemical system is through an exergonic chemical reaction. In his address to the American Physical Society, R. P. Feynman observed [10]: “An internal combustion engine of molecular size is impossible. Other chemical reactions, liberating energy when cold, can be used instead.” This is exactly what happens in our body, where the chemical energy supplied by food is used in long series of slightly exergonic reactions to power the biological machines that sustain life. If a molecular-level device has to work by inputs of chemical energy, it will need addition of fresh reactants (“fuel”) at any step of its working cycle, with the concomitant formation of waste products [11]. Accumulation of such waste products, however, will compromise the cyclic operation of the device unless they are removed from the system, as it happens in our body as well as in macroscopic internal combustion engines. The need to remove waste products introduces noticeable limitations in the design and construction of artificial molecular-level devices based on “chemical fuel” inputs. In any case, since a molecular device has to work by repeating cycles [point (c)], a fundamental requirement is that any chemical process taking place in the system has to be reversible. Chemical fuel, however, is not the only means by which energy can be supplied to operate molecular-level devices. As recalled in the previous section, nature shows that in green plants the energy needed to sustain the machinery of life is supplied by sunlight. Photochemical energy inputs can indeed cause the occurrence of endergonic chemical reactions, which can make a device work without formation of waste products. Currently there is an increasing interest in

the development of photon-powered molecular-level devices, taking advantage of the recent, outstanding progress made by supramolecular photochemistry [2, 4]. Photochemical inputs offer other advantages compared to chemical inputs. For example, they can be switched on and off easily and rapidly. It should also be noted that lasers provide the opportunity of working in very small spaces and very short time domains. To control and monitor the device operation [point (b)], the electronic and/or nuclear rearrangements of the component parts should cause readable changes in some chemical or physical property of the system. In this regard, photochemical and electrochemical methods are very useful since both photons [2, 4] and electrons (or holes) [12, 13] can play the dual role of “writing” (i.e., causing a change in the system) and “reading” (i.e., reporting the state of the system). Luminescence spectroscopy [14], in particular, is a valuable technique since it is easily accessible and offers good sensitivity and selectivity, along with the possibility of time-resolved studies. The operation time scale of molecular-level devices [point (d)] can range from less than picoseconds to seconds, depending on the type of rearrangement (electronic or nuclear) and the nature of the components involved. Finally, as far as point (e) is concerned, the functions performed by molecular-level devices can be various and, to a large extent, still unpredictable; some specific examples will be discussed in this chapter. In the past 10 years, a noticeable number of photochemical molecular devices based on threaded and interlocked supramolecular architectures, such as pseudorotaxanes, rotaxanes, and catenanes, have been developed. From the viewpoint of the functions performed [point (e)], PMDs of this kind can be subdivided into three categories, namely, artificial photosynthetic systems, molecularlevel mechanical machines, and devices for information processing. It is worth noting that, although the energycarrying and information-carrying aspects of light cannot be separated sharply, artificial photosynthetic devices and molecular-level mechanical machines can be regarded as “light-powered” PMDs, while the third kind of system represents “light-processing” devices. Before illustrating specific examples belonging to these categories, it is useful to recall some important features of pseudorotaxanes, rotaxanes, and catenanes.

2. PSEUDOROTAXANES, ROTAXANES, AND CATENANES Research on pseudorotaxanes, rotaxanes, and catenanes has grown exponentially during the past few years [15–27]. Figure 2 shows schematically the structure of pseudorotaxanes, rotaxanes, and catenanes. The names of these compounds derive from the Latin words rota and axis for wheel and axle and catena for chain. In general, these species are referred to as [n]name, where n is the number of molecular components; therefore, the diagrams in Figure 2 represent [2]pseudorotaxane, [2]rotaxane, and [2]catenane.

529

Photochemical Molecular Devices

N+

O

O

O

O

O

O

12+

O

O

S

S

S

S

O

O

O

O

OH

+N

+

O

O

3

OH

N

+

44+

O

+O O

O

+ O O

+

N

[1⊂2]2+

N+

N

O

O

2

O

O O + N O

O

+ +N

a

O

O

O

b

+

N

OH

N S

S

S

S

N

O N

+

+

O

HO

O O

[3⊂4]4+

Figure 3. Two examples of self-assembling of pseudorotaxanes based on -electron donor/acceptor interactions: (a) from [28] and (b) from [29].

Figure 2. Schematic representation of a [2]pseudorotaxane, a [2]rotaxane, and a [2]catenane. The number in brackets indicates the number of molecular components.

2.1. Pseudorotaxanes Pseudorotaxanes are a particularly interesting family of host– guest systems. As in the case of any host–guest complex, formation of a pseudorotaxane occurs via a molecular recognition process between two “instructed” components [1]. It is a thermodynamically controlled self-assembly process that may occur as a result of a variety of interactions deriving from the size, shape, and electronic properties of the partners. The most important types of interactions are those involving electron donor/acceptor ability, hydrogen bonding, hydrophobic/hydrophylic character, – stacking, coulombic forces, and, on the side of the strong interaction limit, metal–ligand bonding. Usually, it is not difficult to understand which is the predominant interaction in a given system. It should be noted, however, that in some cases the strongest interaction as far as association is concerned is not that causing the most relevant changes in the properties of the system [e.g., ultraviolet–visible (UV–vis) absorption, fluorescence and NMR spectra, electrochemical behavior] on going from the separated components to the pseudorotaxane structure. A number of pseudorotaxanes have been obtained by threading a wire-type molecule containing -electron acceptor units into a macrocycle comprising -electron donor units or, vice versa, a wire-type component containing electron donor units into a macrocycle that comprises electron acceptor units. Two examples are shown in Figure 3: (a) the 1,1 -dibenzyl-4,4 -bipyridinium electron acceptor dication 12+ threaded into the 1,5-dinaphtho38-crown-10 (1/5DN38C10) electron donor macrocycle 2 [28] and (b) the acyclic polyether 3 containing a tetrathiafulvalene electron donor unit threaded into the electron acceptor cyclophane 44+ [29]. Species composed of a macrocycle covalently connected to a wirelike moiety, giving rise to self-threaded pseudorotaxane structures, are also

known [30]. Although in these cases a large contribution to the association driving force comes from the electron donor/acceptor interaction, other kinds of interactions, particularly hydrogen bonding, can play an important role, as clearly shown in the cases of [2]pseudorotaxanes composed of 4,4 -bipyridinium [31] or 1,2-bis(pyridinium)ethane [32] threads and crown ethers. Several pseudorotaxanes have been obtained by threading wire-type and macrocyclic components carrying complementary hydrogen-bonding functions. Figure 4a shows, as an example, the threading of a wire-type component containing a secondary ammonium function, such as the N ,N dibenzylammonium ion 5H+ , into a suitable crown ether like dibenzo-24-crown-8 (DB24C8), 6 (Fig. 4a) [33]. Besides hydrogen bonding, other types of interactions (e.g., ion– dipole, -stacking) can contribute to stabilize the pseudorotaxane structure. Pseudorotaxanes can also be obtained as a consequence of simple hydrophobic/hydrophilic interactions. This is the case for the since long-known species in which a wire-type component (72+ ) threads the cavity of -cyclodextrin 8 (Fig. 4b) [34]. A different approach is the metal-based template method [15, 23], for example, starting from a wire (9) and a macrocycle (10), both containing a phenanthroline unit, and using Cu+ as a template (Fig. 4c) [15]. Species of this kind are essentially metal complexes and may have very large formation constants. The examples reported above refer to pseudorotaxane species with a 1 : 1 thread/macrocycle ratio ([2]pseudorotaxanes). It is also possible to obtain species exhibiting different stoichiometric ratios, as illustrated by the fascinating [5]pseudorotaxane formed from four cyclophane 44+ components and a Zn(II)-phthalocyanine with appended four polyether substituents carrying -electron-rich hydroquinone units [35]. Examples of pseudorotaxanes are known in which a macrocycle is threaded by two linear components [36] and a linear component threads two macrocycles [37]. Long linear arrays can be obtained by using homoditopic threads and macrocycles [38]; homogeneous or

530

Photochemical Molecular Devices a O

+

+NH2

O

O

O

O

O

O

O

O

O

5H+

O

stopper [40], a bipyridinium-type wire (124+ ) with a photoand redox-active [Ru(bpy)3 ]2+ (bpy = 2,2 -bipyridine) moiety [41], a crown ether (13) incorporating a binaphthyl unit [42], and a cyclophane (146+ ) containing a Ru complex and two bipyridinium units [43].

O O

+

N H2 O

O O

[5H⊂6]+

6

2.2. Rotaxanes and Catenanes

b +

N

N

+

(CH2)8 N

N

+

72+

+

+

N

N

N

N

[7⊂8]2+

8

c O HO

O N

+

N

+

N

+

Cu

N

O

S

S

S

S

N

S

S

N

I

Cu

N

S

S

S

S

S

S S

S O

O

10

9

N

S

S HO

O

HO

O

HO

O

[Cu••(9⊂10)]+

Figure 4. Examples of self-assembling of pseudorotaxanes based on (a) [N+ —H· · ·O] hydrogen bonding [33], (b) hydrophobic/hydrophilic interactions [34], and (c) coordination around a metal ion [15].

heterogeneous pseudopolyrotaxanes [16] have also been prepared, as well as dendritic ones [39]. Pseudorotaxane species whose components incorporate units capable of exhibiting specific photochemical and/or redox properties are of particular interest for the purpose of constructing photochemical molecular devices. Some examples are reported in Figure 5, namely, a thread (11H+ ) with an ammonium function and a photoactive anthracene as a

N N+ H2

N

N

+

+

N

N

RuII N

N

N

124+

11H+ N N

N RuII N

N O O

O

O

O

O

13

O

+N

Rotaxanes and catenanes are supramolecular (multicomponent) species [1, 15] strictly related to, but also very different from, pseudorotaxanes (Fig. 2). Whereas pseudorotaxanes can undergo dissociation into their wirelike and macrocyclic components, rotaxanes and catenanes are interlocked species, whose dissociation requires breaking of a covalent bond. The general strategy to prepare rotaxanes and catenanes with high yields is based on the template effect [44], which relies on the presence of molecular recognition sites in the components to be assembled. Figure 6 shows the routes by which components bearing suitable recognition sites lead to the formation of rotaxanes: (i) threading of a molecule through a preformed ring, followed by capping the end(s) of the thread; (ii) slipping of a preformed ring over the stoppers of a preformed dumbbellshaped component into a thermodynamically favorable site on the rod part of the dumbbell; and (iii) clipping of a preformed dumbbell with a suitable U-type component that is subsequently cyclized. Figure 7 shows the synthesis of a rotaxane based on donor/acceptor interaction by the threading or the clipping approach [45]. In the first case (Fig. 7a), threading of the electron acceptor tetracationic cyclophane by a thread containing a dioxybenzene electron donor unit yields a pseudorotaxane; then, reaction of the terminal hydroxy groups of the thread with triisopropylsilyl triflate leads to the rotaxane. In the clipping approach (Fig. 7b), the rotaxane is obtained by constructing the electron acceptor cyclophane around the preformed dumbbellshaped component. In the case of catenanes, the most rationale synthetic strategy is the clipping of a macrocycle onto a preformed one (Fig. 8). A double-clipping procedure can also be used [46]. Figure 9 shows the synthesis of a catenane templated by electron donor/acceptor interactions [45]. Reaction of a

N N+

146+ +N

N+

Figure 5. Some pseudorotaxane components containing units capable of exhibiting specific properties. The wirelike molecule 11H+ , with an ammonium center and a photoactive anthracene unit as a stopper [40]; the bipyridinium-containing wire 124+ , with a photo- and redox-active [Ru(bpy)3 ]2+ moiety [41]; crown ether 13, incorporating a binaphthyl unit [42]; cyclophane 146+ , which contains a [Ru(bpy)3 ]2+ complex and two bipyridinium units [43].

Figure 6. Routes for the synthesis of [2]rotaxanes.

531

Photochemical Molecular Devices

+ +

N+

N

O

O

O

N

HO

N+

O

O

N+

N

N

Br

Br

O

a

+

+

N

N

Si O

+

N

N

O

+

Br

+

O

O

O

O

O N+

N

N +

O

O

O

O

O

O

O

O

O

O

O Si

O

O

O

O

O

O

O

templated

by

N+

+N

N+

Figure 9. Synthesis of a [2]catenane donor/acceptor interactions [45].

Si O

O

+N

O

+

N

O

O

O

+ N

O

O

N+ O

O

N

Br

+N

O

O Si

O

O

O

+

O

Si OTf

O

O

OH

+ +

N

O

O

electron

O

Br

b

Figure 7. Synthesis of a [2]rotaxane by the threading (a) and clipping (b) approaches [45].

Figure 8. Routes for the synthesis of [2]catenanes.

dication with a dibromide gives a tricationic intermediate, which interacts with bis-p-phenylene-34-crown-10 to afford a pseudorotaxane-like, or precatenane, structure. The subsequent cyclization, as a result of nucleophilic displacement of a bromide ion, gives the corresponding catenane. In some cases, catenanes are formed from relatively small molecules by one-pot synthesis in a more or less serendipitous manner [47–49]. Figure 10 shows the one-pot synthesis of an amide-based [2]catenane from very simple compounds [48]. This synthesis is thought to involve the perpendicular preorganization of the components caused by three templating effects: (i) steric complementarity, (ii) hydrogen bonding between carbonyl oxygen atoms and amide protons, and (iii) – interactions between the benzene rings of host and guest units. Several related amide-based rotaxanes have also been prepared [18, 21, 27]. Besides simple rotaxanes and catenanes, a great variety of more complex systems have been synthesized, including branched [n]rotaxanes [50], rotaxanes bearing dendritic stoppers [51], catenanes composed of three, five, and seven interlocked macrocycles [52], polyrotaxanes and polycatenanes [16–18, 24], catenanes with very special shapes [53], rotacatenanes [54], pretzelanes [55], and knots [15, 23, 27]. Some of these compounds are shown in Figure 11.

532

Photochemical Molecular Devices

Cl

Cl O

O

+

H2N

NH2

a

O

CH2Cl2

O

N(C2H5)3

O

O O O

O

O NH

NH

HN

O

HN

O

O

N O

NH O

NH

HN

O

N

O

O N

O

O

O O

O

O

O

O

N

O

O

O

O

O

N

O

O O O

O O

O

HN

O

N

O

O

O

O O

Figure 10. One-pot synthesis of an amide-based [2]catenane [48].

O O

3. ARTIFICIAL PHOTOSYNTHETIC DEVICES An intelligent approach toward the design of artificial systems for solar energy conversion (artificial photosynthetic devices) is to take the natural energy conversion sequence as a model and see whether some of the processes involved can be replaced by more convenient routes and/or the natural components can be replaced by artificial ones. In nature, a large number of light-absorbing species (pigments) deliver the excitation energy to a small number of energy-converting centers where the collected electronic energy is used to obtain charge separation [56]. This choice, made by nature, is very suitable and should therefore also be adopted in artificial systems. Generally speaking, an artificial photosynthetic device should be made of a light-harvesting molecular array (i.e., an antenna system), a charge separation supramolecular species (a reaction center), and some kind of device capable of making use of the separated charges to produce stable, high-energy chemical species (fuels) or electricity [57–60]. Production of electricity by artificial photoelectrochromic devices has already been achieved [61], whereas photogeneration of fuels by artificial systems is a much more difficult task [62].

b

O

O

O

Generally speaking, an antenna for light harvesting is an organized multicomponent system in which several chromophoric molecular species absorb the incident light and channel the excitation energy to a common acceptor component. Artificial antenna systems based on porphyrin arrays [63], multichromophoric cyclodextrins [64], polynuclear metal complexes [65, 66], dendrimers [66–68], and polymers [69, 70] have been reported. In all these systems, the chromophoric units are connected by means of covalent

O

N

O N O

N

O N

O N

O

N

O

N

O

O

O

O

O

N

O

S

S

S

CH3S

S

S

S

O

O

O Si

N O

O

O

O CH3S

O

O

N

O

N

O

N O

O

c O

O

N

O

O

O

O

O

O

O

O

O

N

ON

O

N

d

3.1. Antenna Systems

O

O

N

S

S

S

S

O Si

O

O

O N

O

S S O

N O

S

S

N O

O

S

S

SCH3

O

S

S

SCH3

O

Figure 11. Examples of peculiar rotaxane and catenane structures: (a) a branched [4]rotaxane [50], (b) a [5]catenane named “olympiadane” [52], (c) a rotacatenane [54], and (d) a [2]catenane composed of three macrocyclic rings [53].

533

Photochemical Molecular Devices

bonds, in contrast with natural photosynthetic systems where the molecular components are held together by noncovalent interactions. Artificial antenna systems in which several chromophoric units are mechanically linked together in a polyrotaxane structure have recently been reported [71, 72]. Such species are composed of many -cyclodextrin units, each bearing one or more naphthyl chromophoric groups, threaded by a poly(ethyleneglycol) chain and stoppered by bulky groups. The polyrotaxane 15, schematized in Figure 12 [71], contains about 15 -cyclodextrins, each bearing two naphthalenesulfonate units and an axle composed of an average of 45 [OCH2 CH2 ] units, stoppered by two adamantane groups. When the dansyl-modified -cyclodextrin 16 is added to 15 in aqueous solution at 298 K, a photoinduced energy transfer process takes place from the naphthyl chromophoric groups of 15 to the dansyl unit of 16, which binds to the adamantane stoppers by virtue of hydrophobic interactions (Fig. 12). Such a process, evidenced by the quenching of naphthalene fluorescence and sensitization of the dansyl fluorescence, can be switched off by adding 1-adamantanol to the solution, which competes with the adamantane extremities of 15 for the cavity of 16. In a successive related work [72], anthracene units have been employed as stoppers for a series of polyrotaxanes; indeed, an efficient energy transfer process from the naphthalene units of the -cyclodextrin rings to the anthracene stoppers takes place within the polyrotaxanes. Fluorescence anisotropy measurements also indicate excitation energy migration between naphthalene chromophoric units. The rotaxane architecture, in conjunction with the threading procedure (Fig. 6) employed for the synthesis of this series of compounds, gives the interesting opportunity of adjusting the number of chromophores simply by changing the ratio between naphthalene-decorated and plain -cyclodextrin units during the preparation of the polyrotaxanes.

3.2. Charge Separation Devices The simplest supramolecular species capable of performing a charge separation process are those containing two components (diads) [73]. A possible scheme for a diad is

represented in Figure 13. Light excitation of the electron donor (D) (process 1) leads to an electron transfer (process 2), which produces a charge separation and is in competition with the deactivation of the excited state of D (process 3). However, the recombination of the charges in such systems (process 4) is usually very fast. To increase the lifetime of charge separation, which is of fundamental importance for utilizing the energy transiently stored in the charge separation process, more complex systems have to be considered. In particular, an additional charge stabilization step may be introduced, in which the charges are moved far apart from each other. Supramolecular species of this type are called triads [73]. Two possible schemes for charge-separating triads are shown in Figure 14. Although the scheme in Figure 14b is reminiscent of the natural photosynthetic reaction center, that of Figure 14a seems to be more popular in the field of artificial triads. In both cases, excitation of a chromophoric component (process 1) is followed by a primary photoinduced electron transfer to a primary acceptor (process 2). This is followed by a secondary thermal electron transfer (process 3) from a donor component to the oxidized chromophore (Fig. 14a) or from the primary acceptor to a secondary acceptor component (Fig. 14b). The primary process competes with excited-state deactivation (process 4), while the secondary one competes with primary charge recombination (process 5). Finally, charge recombination between remote molecular components (process 6) leads the triad back to its initial state. Porphyrins, owing to their outstanding photophysical and redox properties [74], are extensively used for the construction of photochemical molecular devices to achieve charge separation over nanometric distance [75]. Porphyrin units have been successfully incorporated into rotaxane structures [76, 77], first playing the role of stoppers. In CH3 CN solution at room temperature, [2]rotaxane 172+ (Fig. 15) [78, 79] undergoes a very fast (ca. 2 ps) electron transfer process from the Zn(II) to the Au(III) porphyrin upon excitation of the Zn(II) porphyrin moiety (process 1 in Fig. 15). Although direct back electron transfer from the Au(II) to the Zn(III) porphyrin indeed occurs (530 ps), the initial state is restored mainly via oxidation/reduction of the central copper unit, that is, through an electron transfer from ∆E

hν′ NaO3S

O

O

O

O

O

O

HN

O

NaO3S

O

+

O

SO2 HN O

n = 15

15

NaO3S

SO2 HN

(H3C)2N

O NH

hν

(H3C)2N

NH

O

O

O

O

NaO3S

16

O

HN

O

O

n = 15

[15⊂16]

Figure 12. The light-harvesting polyrotaxane 15 [71] is composed of a poly(ethyleneglycol) axle encircled by an average of 15 -cyclodextrin units, each bearing approximately two naphthalenesulfonate groups; therefore, 15 contains about 30 naphthyl chromophoric groups held together mechanically. In aqueous solution, the dansyl-modified -cyclodextrin 16 is able to bind to the adamantane stoppers of 15; under such conditions, a photoinduced energy transfer process from the excited naphthalenesulfonate moieties of 15 to the dansyl group of 16 is observed.

534

Photochemical Molecular Devices hν

e– 2

N

O

O

N II

Zn N

N

O N

O

Cu

N

I

N

1

N

e–

O

Figure 13. Cartoon illustration and working scheme, based on an orbital representation, of a supramolecular diad for photoinduced charge separation.

the Cu(I) unit to the oxidized Zn porphyrin (process 2; 20 ps) followed by another electron transfer from the oxidized copper site to the Au porphyrin (process 3; 2.5 ns). Interestingly, if the copper ion is removed, the so-obtained demetallated [2]rotaxane exhibits electron transfer properties similar to those of the free bis-porphyrin dumbbell components [80]. However, the very short lifetime of the primary charge-separated state and the complexity of the deactivation pattern constitute major limitations to the performance of 172+ as a charge separation system. Rotaxane 182+ (Fig. 16) [81] is composed of an axle bearing one Cu(I)bis(phenanthroline) complex and two Zn(II) porphyrins as stoppers and a macrocyclic component that incorporates one Au(III) porphyrin. In such a compound, excitation of the Zn(II) porphyrin units (CH3 CN; room temperature) leads to energy transfer to the central copper unit (process 1), which, in turn, transfers an electron to the Au(III) porphyrin (process 2). Since the oxidized Cu(II) moiety can be reduced by the Zn(II) porphyrin, a charge shift occurs (process 3), leading to the final charge-separated state, which lasts for 5 ns before back electron transfer from the Au(II) to the Zn(III) units (process 4) takes place [81]. Therefore, this assembly operates according to the scheme depicted in Figure 14a, except for the fact that light is absorbed by porphyrins and not by the central copper unit; however, an efficient energy transfer process takes place from the zinc porphyrin to the copper complex.

O

N III N Au N N

O 3 e

–

172+

Figure 15. Photoinduced electron transfer processes that take place in [2]rotaxane 172+ [78, 79].

Among the systems proposed as models for the photosynthetic reaction center, supramolecular assemblies in which Ru(II)-polypyridine complexes and 4,4 -bipyridinium units are held together noncovalently in threaded and interlocked structures have been extensively studied [43, 82–88]. In such assemblies, connections between the molecular components rely on charge transfer interactions between the electron acceptor bipyridinium units and aromatic electron donor groups (Fig. 3). For instance, in the various pseudorotaxanes formed in acetonitrile solution at 298 K by the threading of cyclophane 44+ by the dioxybenzene-containing tethers of 192+ (Fig. 17) [84], an efficient photoinduced electron R N N Zn R hν

R

R

3

e– 2 O

N

N N

R

N N ZnII N

1

II N

R e–

R = C6H13

∆E

Cu

I

O N

N

N

N

N N

O

O

N R

4

e–

R

182+ Figure 14. Cartoon illustrations and working schemes, based on an orbital representation, of two types of supramolecular triads for photoinduced charge separation.

Figure 16. Photoinduced energy and electron transfer processes occurring in [2]rotaxane 182+ [81], which behaves as a triad for photoinduced charge separation according to the scheme illustrated in Figure 14a. For more details, see text.

535

Photochemical Molecular Devices H3CO

OCH3 O

N

O O 2

O O

N

2

O O O

+ N

O

O

2

O

O

N

O

N RuII N

N

3

O

N

3

O

3

O

O

2

O

O

O O

O

O

N+

O

O +N

2

192+

O N+

O N+

206+ O +N

H3CO

O

O

O

E°in = –0.45 V

OCH3

+N

O

2

O

O 3

O

O

O

N

N

N

3

3

H3CO

N

E°out = –0.32 V

O O

N Ru II

OCH3

N+

44+

Figure 17. Branched compound 192+ [84], which contains a [Ru(bpy)3 ]2+ core and six pendant threadlike substituents, forms pseudorotaxanes with macrocycle 44+ . In such pseudorotaxane species, efficient photoinduced electron transfer takes place from the excited state of the ruthenium complex of 192+ to the bipyridinium units of 44+ .

transfer process takes place from the excited state of the [Ru(bpy)3 ]2+ moiety of 192+ to a 4,4 -bipyridinium unit of 44+ . The back electron transfer occurs on a time scale (ca. 1 s) considerably longer than that found for covalently linked [Ru(bpy)3 ]2+ /bipyridinium assemblies (nanosecond time domain), a behavior that has been attributed to increased spatial separation of the photogenerated redox pair due to repulsive electrostatic interactions between the positively charged macrocycle and Ru(II) complex [84]. It has also been found that recombination takes place within the threaded supramolecular assemblies, indicating that the pseudorotaxanes do not break apart upon light irradiation. However, in the light of the large variety of threaded species that can be formed, as well as of the relatively small stability constants of the resulting pseudorotaxanes, well-defined interlocked (catenane or rotaxane) structures might be preferred [85–88]. One intriguing aspect of the bacterial photosynthetic reaction center is the redox asymmetry of the cofactors; electron transfer proceeds exclusively along one branch of an almost symmetrical pair of reagents [89, 90]. The catenane structure can be exploited to reproduce such an asymmetry with artificial systems; in catenane 206+ (Fig. 18) [86], the two bipyridinium electron acceptors, linked to a [Ru(bpy)3 ]2+ type primary electron donor, possess different reduction potentials because one of them is encircled by an electron donor crown ether that both attenuates its electron acceptor affinity via charge transfer interactions and hinders access to solvent molecules. Indeed, in CH3 CN solution at 293 K, a very fast k = 5 9 × 1010 s−1  electron transfer takes place from the photoexcited [Ru(bpy)3 ]2+ moiety to a bipyridinium unit. Although it can be anticipated with thermodynamic arguments that the redox asymmetry of the two acceptor branches in 206+ should result in a fivefold difference in rate constants [86], there is no means to distinguish between the two electron transfer paths by time-resolved

Figure 18. In [2]catenane 206+ [86], upon excitation of its [Ru(bpy)3 ]2+ moiety, a very fast electron transfer process to a bipyridinium unit occurs. Owing to the catenane structure, the two bipyridinium units do not possess the same reduction potential (half-wave potential values versus SCE for the “inside” and “outside” units are indicated); such a redox asymmetry could mimic that of the cofactors in the bacterial photosynthetic reaction center.

spectroscopic techniques. In such a system, back electron transfer is also extremely rapid k = 2 4 × 1010 s−1 , owing to the close proximity of the reactants, despite the fact that it should fall in the Marcus “inverted” region. The photoinduced processes taking place in triad 216+ (Fig. 19), made of a Zn(II) porphyrin linked to a [Ru(bpy)3 ]2+ -crown ether component catenated with one 44+ macrocycle have been studied recently (H2 O, phosphate buffer, 293 K) [88]. Upon excitation of the Zn(II) porphyrin moiety, direct electron transfer occurs from the photoexcited porphyrin to the tetracationic macrocycle. The same charge-separated state can be reached upon excitation of the Ru(II) complex followed by two electron transfer steps (Fig. 19), according to the scheme reported in Figure 14a. Two semisynthetic systems, obtained by incorporation of 216+ into two different proteins, namely, cyctochrome b562 and myoglobin, were also studied [88]. Interestingly, the lifetime values for the photoinduced charge-separated state increase substantially when the heme moiety of the triad is embedded in the protein matrix: from 300 ns found for free 216+ to 600–900 ns of cytochrome-b562 -(216+ ) and 1.1–18 s of myoglobin-(216+ ). This study not only demonstrates that protein matrix effects play a crucial role in electron transfer processes of biological importance, but also that hybrid e– hν

e–

O N N

Zn II

N

COOH H N

N

N O

N

O

Ru N

O

+

O

N

O

O N

N

+

II N N N

O O

N

+

O

O

+

O

216+

Figure 19. Photoinduced processes taking place in the catenane triad 216+ [88].

536 natural/artificial supramolecular assemblies are promising candidates for the realization of molecular devices to be used in artificial photosynthetic systems.

4. MOLECULAR-LEVEL MACHINES A molecular-level machine can be defined [91–93] as an assembly of a distinct number of molecular components that are designed to perform mechanical movements as a result of an appropriate external stimulation. Although there are many chemical compounds whose structure and/or shape can be modified by an external stimulus (see, e.g., photoisomerizable species), the term molecular machine is used only for systems showing large-amplitude movements of the molecular components. Chemical, photochemical, or electrochemical stimuli can be used to feed molecular-level machines [91–93]; however, for the reasons discussed in the Introduction, light is the most convenient form of energy to make molecular machines work [11]. Two kinds of lightinduced processes have been used so far for this purpose: photoisomerization reactions [94] and photoinduced electron transfer processes [95]. It is very important that such molecular-level motions are accompanied by changes of some chemical or physical property of the system, resulting in a readout signal that can be used to monitor the operation of the machine. The reversibility of the movement, that is, the possibility to restore the initial situation by means of an opposite stimulus, is an essential feature of a molecular machine. Since the induced motions usually correspond to transitions between two stable structures (states), systems of this kind behave according to a binary logic and could also prove useful for information processing [3]. Threaded and interlocked compounds, owing to their peculiar structure [15–27], are attractive candidates for the construction of molecular-level machines; therefore, it is not surprising that most of the recently designed molecularlevel machines are based on pseudorotaxanes, rotaxanes, and catenanes. Figure 20 shows pictorially some of the movements that can be imagined for these systems, namely, threading/dethreading of the wire and ring components of pseudorotaxanes (Fig. 20a), shuttling of the macrocyclic

Photochemical Molecular Devices

component along the axle in rotaxanes (Fig. 20b), and rotation of one molecular ring with respect to the other in catenanes (Fig. 20c).

4.1. Piston/Cylinder Systems Dethreading/rethreading of the wire and ring components of a pseudorotaxane resembles the movement of a piston in a cylinder (Fig. 20a) [91]. The first attempts at designing a photochemically driven molecular machine of this type were carried out on pseudorotaxanes stabilized by electron donor/acceptor interactions (Fig. 3). In such systems, the donor/acceptor interactions introduce low-energy charge transfer (CT) excited states responsible for absorption bands in the visible region. Light excitation in these CT absorption bands leads formally to the transfer of an electron from the donor to the acceptor component, as illustrated in Figure 21 for the pseudorotaxane formed in H2 O or CH3 CN solution by the electron donor thread 22 and the electron acceptor macrocycle 44+ . As a consequence, particularly when this process leads to formation of charges of the same sign in the two components, one can expect destabilization of the pseudorotaxane structure followed by dethreading. In practice, however, this simple approach does not work because the back electron transfer process is much faster than the separation of the molecular components, a process that requires extended nuclear motions and solvent rearrangement. In some particular cases [96, 97], laser flash photolysis experiments have suggested that a small fraction of the irradiated pseudorotaxane may undergo dissociation. To really achieve photoinduced dethreading, a different approach has been devised [98, 99], based on the use of an external electron transfer photosensitizer (P) and a sacrificial reductant (Red), as illustrated in Figure 22. The photosensitizer must be able to (i) absorb light efficiently and (ii) have a sufficiently long-lived and reductant excited state, so that its excitation (process 1) in the presence of the pseudorotaxane will lead (process 2) to the transfer of an electron to a bipyridinium unit of the cyclophane. The relatively fast back electron transfer from the reduced cyclophane component to the oxidized photosensitizer is prevented by the sacrificial reductant, which, if N

O

O

O

N

+ HO

+ O

O

dethreading

hν +

N

N

+ +

+

+

+

O

O O

OH +

Figure 20. Pictorial representation of machine-like movements that can be obtained with pseudorotaxanes, rotaxanes, and catenanes: (a) dethreading/rethreading of the molecular components in a [2]pseudorotaxane, (b) shuttling of the macrocyclic component along the axle in a [2]rotaxane, and (c) ring rotation in a [2]catenane.

N

N

+

[22⊂4]4+

O

N+

N

N

threading

N +

4

4+

O

OH

+ N

22·+

+

O

+

O HO

O

back electron transfer

4

back electron transfer

HO O

N

O

+

O O

·3+

N

O

OH

O

OH N +

N+

N

O O

O O

O

HO

22

Figure 21. Photochemical processes associated with [2]pseudorotaxane [22⊂4]4+ upon excitation in its charge transfer absorption band [98, 99]. The processes indicated by dashed arrows are unlikely to occur.

537

Photochemical Molecular Devices +

N O

N

O

O N

O

4 +

N+

+

N

N

O O

HO

O

+

Oxidant (O2)

2

5

*P

P

1

products

+

+

N

hν

N

+N HO

O

O

O

4·3+

P+

3 products

O

OH

+

O

OH

reductant (triethanolamine)

N+

N

dethreading

HO

O

O

(re)threading

O

O

O N

O

OH

+

6

O

OH

+N

N

+

N+

N+

O O

O O

HO

O

+

[22⊂4]4+

22

44+

Figure 22. Photochemically induced dethreading of [2]pseudorotaxane [22⊂4]4+ , based on the use of an external photosensitizer (P) and a reductant scavenger. Rethreading occurs upon oxygenation of the previously deaerated solution [98, 99].

present in a sufficient amount, intercepts the oxidized photosensitizer and regenerates (process 3) its original redox state. Good candidates for the role of photosensitizer are 9anthracenecarboxylic acid [100] and metal complexes such as [Ru(bpy)3 ]2+ [101], while efficient reductant scavengers are triethanolamine and polycarboxylate (e.g., oxalate) anions [102]. Under these conditions, the persistent reduction of a bipyridinium unit of 44+ is achieved and the pseudorotaxane dethreads (process 4), as evidenced by absorption spectral changes and, more important, by the increase in the intensity of the dioxynaphthalene fluorescence, which can only originate from free 22. Oxygenation of the solution, from which O2 was initially removed, reoxidizes the macro-

cycle back to the tetracationic form (process 5), thereby promoting rethreading with 22 (process 6), as also shown by the absorption and luminescence spectra. Very recently, it has been shown that this same system can be operated in unconventional environments such as the interior pores of a sol–gel silica framework [103]. The possibility of mounting such a molecular-level machine on a solid support, by trapping the wirelike molecule onto the surface of a sol–gel film, has also been demonstrated [103]. This strategy has been extended recently to secondgeneration pseudorotaxanes [23⊂2]4+ and [22⊂24]4+ in which the metal-complex photosensitizer (the “light-fueled” motor [104]) has been incorporated either into the thread (Fig. 23a) [41] or into the ring (Fig. 23b) [105] component. The successful operation of these pseudorotaxanes as molecular machines is the result of (i) the appropriate choice of the functional units and (ii) their covalent linking into the thread and ring components in order to achieve the correct integration of the needed functions (e.g., receptor ability, redox features, photophysical properties, etc.), the right sequence of processes, and the lack of interference between the active units. As in the case of the molecular machine shown in Figure 22, the dethreading and rethreading motions of the pseudorotaxanes represented in Figure 23 can be triggered by visible-light irradiation and oxygenation of the solution, respectively. The motions can also be easily monitored by means of UV–vis absorption and luminescence spectroscopy. The most important readout signal is the intensity of the dioxynaphthalene fluorescence associated with the free ring 2 (Fig. 23a) or free thread 22 (Fig. 23b)

products e–

hν

O

O

N N N

O

O + N O

N Ru II N

a

reductant (triethanolamine)

O

N

+ O

O [23⊂2]4+

N II Ru N

N

N

N

O

O

N

N

N

O

O

N

O

O

+

O

+ O

oxidant (O2)

O O

23·3+

products

O

O 2

hν OC

e–

CO

Re I

OC N

N

O

+N

products

Cl

O

O N+

OH

OC

CO

OC Re I Cl N

reductant (triethanolamine)

N

b N+

N

+ +N

HO

O

O

N+ oxidant (O2)

O [22⊂24]4+

products

+N

N+ 24·3+

O

OH

O O

O O

HO

O

22

Figure 23. Photocontrollable molecular machines based on [2]pseudorotaxanes. In these second-generation systems, the “light-fueled” motor (i.e., the photosensitizer) is part of the threadlike and of the macrocyclic components of [23⊂2]4+ and [22⊂24]4+ , respectively [41, 105]. As for the system described in Figure 22, a reductant scavenger is employed, and rethreading is performed by allowing oxygen to enter the irradiated solution.

538

Photochemical Molecular Devices

components. It is worth noting that many [deoxygenation– irradiation (dethreading)/oxygenation (rethreading)] cycles can be performed on the same solution without any appreciable loss of signal until most of the reductant scavenger is consumed. It should also be stressed that systems that rely on this photosensitizer–scavenger strategy utilize, in addition to light energy, the irreversible decomposition of a reductant scavenger that produces “waste” species. In this regard, the search for efficient molecular machines exploiting “clean,” reversible photochemical reactions (in other words, machines that use only light as an energy supply) is of fundamental importance. Photoisomerization reactions, particularly the well-known [94] reversible cis/trans photoisomerization of the azobenzene group, have long been used to exert photochemical control on chemical systems [106]. Since the early 1980s, azobenzene-containing compounds have been employed both as photocontrollable hosts [107, 108] and guests [109]. The first example of a pseudorotaxane in which dethreading/rethreading is based on such a principle and is powered exclusively by light energy, without generation of any waste products, has been reported only recently (Fig. 24) [110]. The threadlike species trans-25, which contains an electronrich azobiphenoxy unit, and the electron acceptor macrocycle 264+ self-assemble very efficiently in acetonitrile solution to give a pseudorotaxane, stabilized by donor/acceptor interactions. In the pseudorotaxane structure, the intense fluorescence characteristic of free 264+ is completely quenched by the donor/acceptor interaction. Irradiation with 365 nm light of a solution containing trans-25 and 264+ , in which the majority of the species are assembled to give the pseudorotaxane, causes trans → cis photoisomerization of 25. Since the affinity of the macrocycle for cis-25 is much lower than that for trans-25, photoexcitation causes a dethreading process (Fig. 24), as indicated by a substantial increase in the fluorescence intensity of free 264+ . On irradiation at 436 nm or by warming the solution in the dark, the trans isomer of 25 can be reformed and, as a result, it rethreads inside the OH

O O

+N N

N

+N

O

K ca. 1.5 × 105 M-1

+

O HO

N+

O

N+

+ N

N

O

+N N N

HO

O

O

cis-25

HO

N+

O2N

N

+

O

+

N

N

NO2

+

trans-274+

K ca. 1 × 104 M-1

hν (365 nm)

+ O N

+ N O

N +

hν′ (430 nm)

O

N N N+

N

+

N

O

N

HO

+ +N

O2N

O2N

hν′ (440 nm) or ∆

O O

O

N +

OH O

In rotaxanes containing at least two (or more) different recognition sites (“stations”) in the axle component, it is possible to switch the position of the macrocycle between the two stations by an external stimulus [91]. The [2]rotaxane trans-274+ (Fig. 25) [111] incorporates an -cyclodextrin torus and a trans-azobiphenoxy-containing axle. Initially, the cyclodextrin ring resides exclusively around the transazobiphenoxy recognition site. Upon irradiation at 365 nm in H2 O at 278 K, the azobiphenoxy unit isomerizes [94] from trans to cis, “pushing” the -cyclodextrin component away to encircle one of the [(CH2 )2 O] chains. Further irradiation at 430 nm causes back isomerization of the azobiphenoxy group from the cis to the trans form, leading to the shuttling of the cyclodextrin component back to the transazobiphenoxy recognition site (Fig. 25). The photoisomerization/shuttling processes are accompanied by changes in the circular dichroism signals induced by the chiral cavity of the cyclodextrin on the azobiphenoxy – ∗ transitions. The [2]rotaxane 28+ (Fig. 26) [112] has a phenanthroline and a terpyridine unit in its axle component, and it incorporates a Cu(I) ion coordinated tetrahedrally by the phenanthroline ligand of the axle together with the phenanthroline ligand of the macrocycle. Oxidation of the tetracoordinated Cu(I) center produces a tetracoordinated Cu(II) complex; in response to the preference of Cu(II) for a pentacoordination geometry, the macrocycle shuttles away from the bidentate phenanthroline ligand of the axle and places itself on the terdentate terpyridine site. Subsequent reduction of the Cu(II) center to Cu(I) causes an opposite shuttling process, affording a return to the original structure. These movements can of course, be, performed electrochemically [113], but can also be induced photochemically [114], as illustrated in Figure 26. Upon irradiation at 464 nm of an acetonitrile solution of the [2]rotaxane (process 1), the Cu(I)-based

[trans-25 26]4+

hν (360 nm)

OH

4.2. Molecular Shuttles

N

O

+N

264+

trans-25

+ N

OH

macrocycle. Owing to the full reversibility of the photoisomerization process, the light-driven dethreading/rethreading cycle can be repeated at will (Fig. 24). Another relevant feature of this system is that it exhibits profound changes of a strong fluorescence signal.

N

O2N

N +

[cis-25 26]4+

Figure 24. Dethreading/rethreading of [2]pseudorotaxane [25⊂26]4+ as a consequence of the cis/trans photoisomerization of the azobenzenetype unit contained in the threadlike component 25 [110].

O2N

N +

N

+

O

N

O2N O

N

+

N

+

NO2

cis-274+

Figure 25. Light-controlled reversible shuttling of the macrocyclic component of 274+ along its axle, based on the cis/trans photoisomerization of the azobenzene unit [111].

539

Photochemical Molecular Devices

hν

O

O

O

28+ O

O

1

O O

N

N

N

CuI N O

N

N

O

shuttling N

N

5

O

O

O

N

N

N

CuI N N

O

N

O

products

2 4

products

reductant (ascorbic acid) O

O

O

O

O O

N

N

CuII N O

O O

oxidant (p-NO2C6H5CH2Br)

O

O

O

N

N

N

O

N

O

O

O

shuttling N

3

O

O

O

N

O O

N

N

CuII N N

N

O

282+

Figure 26. Photoinduced shuttling, based on the use of an oxidant scavenger (p-nitrobenzylbromide), of the macrocyclic component in the coppercontaining [2]rotaxane 28+ [114].

chromophoric unit is excited to a metal-to-ligand charge transfer (MLCT) state. In the presence of a suitable oxidant (p-nitrobenzylbromide), electron transfer from the photoexcited rotaxane to the oxidant follows (process 2), which generates a tetracoordinated Cu(II) center and products deriving from reduction of p-nitrobenzylbromide. Such a reaction competes with the intrinsic deactivation of the MLCT excited state, which includes luminescence; therefore, emission spectroscopy is a useful tool to evaluate whether photooxidation takes place. Indeed, light excitation of the rotaxane causes spectral changes indicating the disappearance of the Cu(I) chromophore and the concomitant formation of the tetracoordinated Cu(II) center [114]; spectroscopic measurements showed that subsequent transformation of tetracoordinated Cu(II) into the more stable pentacoordinated Cu(II) species, that is, shuttling of the macrocyclic component (process 3), was slowly occurring. The cycle can be completed upon addition of an excess of ascorbic acid, which reduces the Cu(II) center back to Cu(I) (process 4), affording the shuttling of the macrocyclic ring back to its initial position (process 5). It is worth noticing that in this case the back reduction step cannot be induced by light since in Cu(II) complexes the lowest excited state decays to the ground state very rapidly and therefore cannot be involved in a bimolecular reaction with a reductant. The design principles that form the basis of the lightdriven molecular machines shown in Figure 23 have been employed to obtain the [2]rotaxane 296+ (Fig. 27) [115], specifically designed to achieve photoinduced ring shuttling. This compound is made of the electron donor macrocycle R and a dumbbell-shaped component that contains (i) [Ru(bpy)3 ]2+ (P) as one of its stoppers, (ii) a 4,4 bipyridinium unit (A1 ) and a 3,3 -dimethyl-4,4 -bipyridinium unit (A2 ) as electron-accepting stations, (iii) a p-terphenyltype ring system as a rigid spacer (S), and (iv) a tetraarylmethane group as the second stopper (T). The stable translational isomer of rotaxane 296+ is the one in which the R component encircles the A1 unit, in keeping with the fact

that this station is a better electron acceptor than the other one. Two strategies have been devised in order to obtain the photoinduced abacus-like movement of the R macrocycle between the two stations A1 and A2 : One is based on processes involving only the rotaxane components (intramolecular mechanism), while the other one requires the use of external reactants (sacrificial mechanism). The intramolecular mechanism, illustrated in the left part of Figure 27, is based on the following four operations [115]: (a) Destabilization of the stable translational isomer. Light excitation of the photoactive unit P (process 1) is followed by the transfer of an electron from the excited state to the A1 station, which is encircled by the ring R (process 2), with the consequent “deactivation” of this station; such a photoinduced electron transfer process has to compete with the intrinsic decay of ∗ P (process 3). (b) Ring displacement. The ring moves from the reduced A1 station to A2 (process 4), a step that has to compete with the back electron transfer process from A− 1 (still encircled by R) to the oxidized photoactive unit, P+ (process 5). This is the most difficult requirement to meet in the intramolecular mechanism. (c) Electronic reset. A back electron transfer process from + the “free” A− 1 station to P (process 6) restores the electron acceptor power to the A1 station. (d) Nuclear reset. As a consequence of the electronic reset, back movement of the ring from A2 to A1 takes place (process 7). The results obtained (CH3 CN solution, 298 K) [115] seem to indicate that the back electron transfer (process 5) is faster than the ring displacement (process 4). It is worthwhile noticing that in a system that behaves according to the intramolecular mechanism shown in Figure 27 (left) each light input causes the occurrence of a forward and back ring movement (i.e., a full cycle) without generation of any

540

Photochemical Molecular Devices

Figure 27. [2]Rotaxane 296+ and schematic representation of the intramolecular (left) and sacrificial (right) mechanisms for the photoinduced shuttling movement of macrocycle R between the two stations A1 and A2 located on the axle component.

waste product. In some way, it can be considered as a “fourstroke” cyclic linear motor powered by light. The alternative, less demanding mechanism is based on the use of external sacrificial reactants (a reductant like triethanolamine and an oxidant like dioxygen) that operate as illustrated in the right part of Figure 27: (a) Destabilization of the stable translational isomer, as in the previous mechanism. (b ) Ring displacement after scavenging of the oxidized photoactive unit. Since the solution contains a suitable sacrificial reductant, a fast reaction of such species with P+ (process 8) competes successfully with the back electron transfer reaction (process 5); therefore, the originally occupied A1 station remains in its reduced state, A− 1 , and the displacement of the ring R to A2 (process 4), even if it is slow, does take place. (c ) Electronic reset. After an appropriate time, restoration of the electron acceptor power of the A1 station is obtained by oxidizing A− 1 with a suitable oxidant, such as O2 (process 9).

(d) Nuclear reset, as in the previous mechanism (process 7). Such a sacrificial mechanism, although fully successful, is less appealing than the intramolecular one because it leads to the formation of waste products. However, instead of using a sacrificial reductant, that is, an electron donor molecule that undergoes a fast decomposition reaction after electron transfer has taken place, a “reversible” reductant, giving rise to a stable oxidized form, may be successfully employed, provided that the back electron transfer process can be slowed down by a wise choice of the partners. A light-driven molecular shuttle that relies on this strategy has been reported [116]. The [2]rotaxane 30 (Fig. 28) consists of a benzylic amide macrocycle that surrounds an axle featuring two hydrogen-bonding stations, namely, a succinamide unit and a naphthalimide unit, separated by a long alkyl chain. Initially, the macrocycle resides on the succinamide station because the naphthalimide unit is a much poorer hydrogen-bonding recognition site. Light excitation at 355 nm (process 1) in acetonitrile at 298 K

541

Photochemical Molecular Devices

O O

N H

1

HN

O

shuttling

O

NH

O

N H

[CH2]10

N

6 O

HN

O N H

N O

NH

N H O N

[CH2]10

O H N

30

O

O

4.3. Ring Rotation in Catenanes

O

hν

O NH

HN

O

O

DABCO

DABCO

DABCO

2

5

3

DABCO+

DABCO+

DABCO+

O O

O NH

O

N H

HN

NH O

O

O

O _

N H

[CH2]10

HN O

shuttling N

4 O

O N H

N O

NH

[CH2]10 ·_

30

N H O _

N O H N

HN

O

O

Figure 28. Light-induced reversible shuttling of the macrocyclic component in the hydrogen-bonded [2]rotaxane 30 [117]. The operation of this system relies on the use of a reductant (DABCO). For more details, see the text.

generates the singlet excited state of the naphthalimide unit, which then undergoes high-yield intersystem crossing to the triplet excited state. Such a triplet state can be reduced in bimolecular encounters by an electron donor (1,4-diazabicyclo[2.2.2]octane; DABCO) added to the solution in a sufficiently large amount (process 2). Because the back electron transfer process (process 3) is spin forbidden and thus slow, the photogenerated ion pair can efficiently dissociate; as a matter of fact, the naphthalimide radical anion survives for hundreds of microseconds before it decays by bimolecular charge recombination with a DABCO radical cation. Since the naphthalimide anion is a much stronger hydrogen-bonding station compared to the succinamide, upon reduction of the naphthalimide unit the macrocycle is expected to shuttle from the latter to the former station (process 4); this has been demonstrated by cyclic voltammetric experiments. Laser flash photolysis studies have demonstrated that this is indeed the case; the time required for ring shuttling (ca. 1 s) is much shorter than the lifetime of the naphthalimide radical anion (ca. 100 s). After charge recombination (process 5), the macrocycle moves back to its original position (process 6). This [2]rotaxane constitutes an outstanding example of a linear molecular motor driven exclusively by light, although its operation still relies on the presence of external reactants, which, however, are not consumed. The device can be cycled at a frequency depending on the charge recombination rate of the rotaxane radical anion. It can be estimated that if the shuttle is pumped by a laser at the frequency of its “recovery stroke” (process 5), that is, 104 s−1 , this molecular-level machine generates approximately 10−15 W of mechanical power per molecule [116]. In a very recent work [117], the complex fluorescence behavior exhibited by a similar peptide-based [2]rotaxane bearing an anthracene unit as one of the stoppers has been interpreted in terms of a very fast (subnanosecond), short-amplitude translation of the macrocycle upon light excitation of the anthracene subunit. However, it seems that alternative explanations, such as the formation of intercomponent exciplexes [118], cannot be ruled out.

There are several examples of catenanes where ring movements can be induced by external stimulations like simple chemical reactions or homogeneous or heterogeneous electron transfer processes [91–93], but only very few cases are reported in which the stimulus employed is light. It has been shown that in azobenzene-containing [2]catenanes like 314+ (Fig. 29) it is possible to control the rate of thermally activated rotation of the macrocyclic components by photoisomerization of the azobenzene moiety [119, 120]. Such systems can be viewed as molecular-level brakes operated by light. To date, only one case has been reported in which the ring motions are induced photochemically, using the same strategy adopted to operate molecular shuttle 28+ (Fig. 26). Irradiation at 464 nm of the [2]catenane 32+ (Fig. 30) [121] in CH3 CN solution at room temperature in the presence of p-nitrobenzylbromide causes an electron transfer process from the photoexcited [2]catenane to p-nitrobenzylbromide, thus generating a Cu(II) center. Owing to the preference of the Cu(II) ion for a pentacoordination geometry, the terpyridine-containing macrocycle rotates through the cavity of the other, affording a pentacoordinated Cu(II) center. Upon addition of ascorbic acid, the Cu(II) ion is reduced to Cu(I); in response to the preference of Cu(I) for tetracoordination, the terpyridine-containing macrocycle rotates again through the cavity of the other, restoring the original structure (Fig. 30).

5. MOLECULAR DEVICES FOR INFORMATION PROCESSING As mentioned in Section 1.1, photons represent not only quanta of energy, but also information bits, and can be used by a PMD as input/output signals, to be processed and eventually even stored and retrieved. Apart from futuristic applications related, for instance, to the construction of a chemical computer [122], the design and realization of a molecular-level electronic set—that is, a set of molecular-level systems capable of playing functions that mimic those performed by macroscopic components in electronic devices—is

N

+N

O

N N N

O

+N

O

O N+

O

N+ O

O

O

trans-314+

O

O

hν (350 nm)

+N

hν′ (440 nm)

+N

O

O O N+

O

O

O

O

N+ O

O

O

cis-314+

Figure 29. Photocontrollable molecular-level brake. The thermally activated circumrotation of the macrocyclic polyether component of [2]catenane 314+ can be modulated reversibly by cis/trans photoisomerization of the azobenzene unit incorporated into the tetracationic macrocycle [119, 120].

542

Photochemical Molecular Devices hν O

O

O

O

O

O O

N

N

N

CuI

N

N

ring rotation

N O

N N

N

N

O

N CuI

N

N

O

N

O O

O

O

O

32+

O

O

oxidant (p-NO2C6H5CH2Br)

O

5.2. Plug/Socket and Extension Systems

products

A macroscopic plug/socket system is characterized by the following features: (i) the possibility to connect/disconnect the two components in a reversible way and (ii) the occurrence of energy flow from the socket to the plug when the two components are connected. Pseudorotaxane-type supramolecular systems have been designed that may be considered as molecular-level plug/socket devices. In the system illustrated in Figure 32 [42], the plug-in function is related to the threading, driven by the formation of strong N+ —H· · ·O hydrogen bonds in dichloromethane solution [33], of (±)binaphthocrown ether 33 by a (9anthracenyl)benzylammonium ion, obtained by protonation of the corresponding amine (34). The association process can be reversed quantitatively (plug-out) by addition of a suitable base, like tributylamine, which deprotonates the ammonium ion. In the plug-in state, which corresponds to a pseudorotaxane structure, light excitation of the binaphthyl unit of the crown causes the sensitized fluorescence of the anthracenyl unit of the thread, showing that an efficient electronic energy transfer process has occurred between the two chromophoric groups. Addition of a stoichiometric amount of base to the pseudorotaxane structure causes the revival of the binaphthyl fluorescence and the disappearance of the anthracenyl fluorescence upon excitation in the binaphthyl bands, demonstrating that plug-out has happened. The plug/socket concept at the molecular level can

reductant (ascorbic acid)

products O

O

O

O O

O O

N

N

N

N

N

CuII

N O N

ring rotation

N N

N N

O

N CuII

N

O

N O

O O

O

O O

O

is then polymerized to form a conjugated polypseudorotaxane, which can then be stoppered at both ends to give a polyrotaxane (Fig. 31) [139]. It remains to establish, though, whether the encirclement of the molecular wire with macrocyclic rings in the polyrotaxane will result in improvement of the performance of photonic molecular wires.

O

322+

Figure 30. Photoinduced rotation, based on the use of an oxidant scavenger (p-nitrobenzylbromide), of the terpyridine-containing macrocycle in the copper-containing [2]catenane 32+ [121]. The system is brought back to the initial structure through another ring rotation, induced chemically by reduction with ascorbic acid.

of great scientific interest because it introduces new concepts into the field of chemistry and stimulates the ingenuity of researchers engaged in the bottom-up approach to nanotechnology. In the past few years, many systems that could prove useful for information processing at the molecular level [3, 4, 6–9, 123] (e.g., wires [6, 124–127], antennas [63–70], switches [127, 128], rectifiers [129, 130], plug/socket devices [42], memories [127, 128, 131], logic gates [3, 127, 128, 132–134]) have been constructed and studied. It has already been pointed out that suitably designed pseudorotaxanes, rotaxanes, and catenanes could prove useful for information processing because they can be interconverted at will between two (or more) stable states that can be used to represent information bits on the molecular scale. In some particular cases, the relationship between the input and output signals in such molecular devices corresponds to a logic operation, opening the way to the construction of molecular-level logic gates based on pseudorotaxanes, rotaxanes, and catenanes.

5.1. Molecular Wires Very attractive candidates to play the role of wire-type compounds for the vectorial transfer of energy or electrons are linear oligo- or poly-p-phenylenes [135]. Such rigid rodlike conjugated molecules have been employed as spacers between chromophoric units in systems designed to achieve long-range photoinduced energy or electron transfer (photonic molecular wires) [136, 137]. It has been shown that the luminescence, stability, and processability of polymers of the poly-p-phenylene type can be enhanced by threading them through macrocycles to form conjugated polypseudorotaxanes and polyrotaxanes that could be viewed as “insulated molecular wires” [138, 139]. Hydrophobic binding holds the monomer inside the cavity of a macrocycle, for example, a cyclodextrin; such a [2]pseudorotaxane

Figure 31. Synthesis, directed by hydrophobic binding, of an “insulated molecular wire” [139].

543

Photochemical Molecular Devices

DB24C8 crown ether, which plays the role of a socket. The wire-type component 36H3+ is also made of two moieties: an ammonium unit, which, driven by hydrogen-bonding interactions, threads as a plug into the DB24C8 socket, and a bipyridinium unit, which, driven by donor/acceptor interaction, threads as a plug into the third component 2, a 1/5DN38C10 crown-ether socket. In CH2 Cl2 /CH3 CN (98 : 2 v/v) solution, reversible connection/disconnection of the two plug/socket functions can be controlled independently by acid/base and red/ox stimulation, respectively. In the fully connected 352+ ⊃36H3+ ⊂2 triad, light excitation of the [Ru(bpy)3 ]2+ unit of component 352+ is followed by electron transfer to the remote bipyridinium unit of component 36H3+ , which is plugged into component 2. Possible schemes to improve the system have also been discussed.

Figure 32. Acid/base-controlled plug-in/plug-out of (9-anthracenyl) benzylammonium ion 34H+ , obtained by protonation of the corresponding amine 34, with (±)-binaphthocrown ether 33 [42]. The occurrence of photoinduced energy transfer in the plug-in state is schematized.

be extended straightforwardly to the construction of systems where (i) light excitation induces an electron flow instead of an energy flow and (ii) the plug-in/plug-out function is stereoselective (the enantiomeric recognition of chiral ammonium ions by chiral crown ethers is well known). The plug/socket concept has recently been used to design and construct a self-assembling [3]pseudorotaxane that mimics at the molecular level the function played by a macroscopic extension. The system is made (Fig. 33) [140] of three components, 352+ , 36H3+ , and 2. Component 352+ consists of two moieties—a [Ru(bpy)3 ]2+ unit, which plays the role of electron donor under light excitation, and a

O

hν'

hν

5.3. Logic Gates Computers are based on semiconductor logic gates that perform binary algebraic operations [141]. Logic gates are switches whose output state (0 or 1) depends on the input conditions (0 or 1). YES and NOT single-input gates are the simplest logic devices. A YES gate passes the input bits to the output without changes (input: 1, output: 1; input: 0, output: 0), while a NOT gate inverts any input data (input: 1, output: 0; input: 0, output: 1). Molecular systems that can perform simple YES and NOT logic operations are very common, and luminescence

O

O +

O

352+⊃36H3+⊂2

O

O H2N

O

O

O

O

_

+

H

N

+

+H

O

N

N RuII N N

N

e

O

O

O

O

O

HN

O

N

N

O O

+

O

+O

N

N O

352+

N

reduction

N

oxidation

O O

O N

N

N RuII N N

N

O

O O

O

O

O O

N

O O

+

N

N

N

O

O O

+

O

N RuII N

O

O

O

2

O

O

O

O

O

+

O

O

O

O H2N

O

O+

O

O

352+⊃36H·2+

N

N RuII N

oxidation

O

O

O N

+O

N

362+⊂2

O

O

_

reduction

O

O+ N O O

HN

36·+

O

+

O

+

¥

N

+

N

O

_ O

O O

H+

+ H+

O O

O O

352+

O

2

Figure 33. Supramolecular system that mimics the function played by a macroscopic extension [140]. The two pseudorotaxane-type connections between the three molecular components can be controlled independently by acid/base and red/ox stimulation, respectively. In the fully assembled [3]pseudorotaxane, a photoinduced electron transfer process occurs from the excited state of the [Ru(bpy)3 ]2+ moiety of 352+ to the bipyridinium unit of 36H3+ , which, in turn, is plugged into crown ether 2.

544

Photochemical Molecular Devices

is a particularly useful signal to monitor such operations [132, 142]. To perform more complex logic operations, however, carefully designed multicomponent chemical systems are needed. Figure 34 shows schematically the changes that have to occur in a chemical system in order to perform the AND, OR, and XOR fundamental logic operations under the action of two chemical inputs (X and Y). For illustration purposes, the equivalent (from a logic viewpoint) electric circuits are also shown. The equivalent circuit of the AND gate has switches connected in series (Fig. 34a), whereas that of the OR gate has switches connected in parallel (Fig. 34b). An XOR (eXclusive OR) gate is a much more complex device, as one can understand from the fact that its equivalent circuit contains two bipolar switches (Fig. 34c). The truth table of the XOR operation is the same as that of the OR operation except that the output is 0 if both the inputs are 1. In today’s processors, addition is performed with an AND gate, which gives the carry digit, and an XOR gate, which gives the sum digit. It is also important to notice that the XOR gate is actually a comparator because it can establish whether the two inputs have the same value. Interesting Output

AND

a +X

P, X

Output state "0 "

P

0

I1

P, X, Y

Output state "0 "

+Y

+Y

Output state "1 "

+X

0 1 0 1

OR

b +X

P, X

Output state "1 "

P

+Y

+Y

Output state "1 "

P, Y

Output state "1 "

I1

0

I2

0 1

+X

P, X

Output state "1 "

P

+Y

+Y

I1 Output state "0 "

P, Y

Output state "1 "

0 1 1 1

Output

P, X, Y

Output state "0 "

1

0 1 0 1

XOR

c

0 0 0 1

X I 1 Y I 2 Output 0 0 1 1

+X

1

Output

P, X, Y

Output state "0 "

I2

X I 1 Y I 2 Output 0 0 1 1

P, Y

Output state "0 "

0

1

examples of supramolecular systems capable of performing the AND, OR, and XOR logic operations, as well as more complex ones, and even integrating two such functions in a single supramolecular device, have been reported [132, 142– 147]. In all these systems, binding of different chemical species (input) results in changes in luminescence intensity (output). The first example of a chemical system capable of performing the XOR logic operation was based on the pseudorotaxane [37⊂38]2+ illustrated in Figure 35 [148]. It results from self-assembly, in CH2 Cl2 /CH3 CN (90 : 10 v/v) solution, of the electron-accepting 2,7-dibenzyldiazapyrenium dication 372+ with the crown ether 38, which contains two 2,3-dioxynaphthalene electron-donating units. In the pseudorotaxane structure, the electron-deficient diazapyrenium unit is sandwiched between the electron-rich 2,3-dioxynaphthalene units of 38. Because of the electron donor/acceptor interaction, a low-energy charge transfer excited state is formed that is responsible, inter alia, for the disappearance of the strong fluorescence exhibited by 38 ( max = 343 nm). Upon addition of tributylamine (B, Fig. 35), a 1 : 2 adduct, [37 (B)2 ]2+ , is formed between the 2,7-dibenzyldiazapyrenium dication 372+ and the amine, with the consequent dethreading of 38 (process 1 in Fig. 35). This process causes large spectral changes, including the recovery of the fluorescence of free crown ether 38. Subsequent addition of a stoichiometric amount (relative to the previously added amine) of trifluoromethanesulfonic acid unlocks 372+ from the [37 (B)2 ]2+ adduct (process 2 in Fig. 35) and allows rethreading between dication 372+ and macrocycle 38 to give back the original pseudorotaxane. This process is accompanied by spectral changes opposite to those observed upon addition of amine. Processes 1 and 2 can be repeated on the same solution by repeating the addition of amine and acid; of course, the ammonium ion originated from protonation of amine B is also produced at the end of each cycle. As shown in Figure 35, the dethreading/rethreading cycle can also be performed by reverting the order of the two inputs. Both processes 1 and 3 cause a strong increase in emission intensity at 343 nm, which is cancelled by processes 2 and 4, respectively. Therefore, we can conclude that the chemical system described above shows the input/output

+X

0

1

1

0

I2

X I 1 Y I 2 Output 0 0 1 1

0 1 0 1

0 1 1 0

Figure 34. Schematic representation of a chemical system (P) that performs the AND (a), OR (b), and XOR (c) logic operations under the action of two chemical inputs (X and Y). The truth tables of such operations are also shown, along with their representations based on electric circuit schemes.

Figure 35. Schematic representation of the threading/dethreading pattern of [2]pseudorotaxane [37⊂38]2+ , which corresponds to an XOR logic function.

545

Photochemical Molecular Devices

relationships indicated by the truth table of the XOR logic gate (Fig. 34c): The strong fluorescent signal at 343 nm is present (output: 1) only when either amine or H+ (inputs X and Y in the truth table) are added (i.e., X: 1 and Y: 0 or vice versa); conversely, the fluorescent signal is absent (output: 0) when none or both of the inputs are present (i.e., X = Y: 0 or X = Y: 1).

6. CONCLUSION The progress made in the fields of supramolecular chemistry and photochemistry, and in particular the investigations performed in the last few years on pseudorotaxanes, rotaxanes, and catenanes, has led to the design and construction of molecular-level devices capable of performing a variety of light-induced functions (photochemical molecular devices). Light stimulation is particularly convenient for operating molecular-level devices. For example, it can be switched on and off easily and rapidly and, by employing lasers, it provides the opportunity of working in small space regions and very short time domains. A further advantage offered by the use of photochemical techniques is that photons, besides supplying the energy needed to make a device work, can also be useful in reading the state of the system and thus controlling and monitoring its operation. It should be noted, however, that the molecular-level devices described in this chapter operate in solution, that is, in an incoherent fashion. For most kinds of applications, they need to be interfaced with the macroscopic world by ordering them in some way, for example, at an interface or on a surface [103, 149], so that they can behave coherently, either in parallel or in series. Research on this topic is developing at a fast-growing rate [6, 7]. Furthermore, addressing a single molecular-scale device by instruments working at the nanometer level is no longer a dream [150, 151]. The extension of the concept of a device to the molecular level is of interest not only for the development of nanotechnology, but also for the growth of basic research. Looking at supramolecular chemistry from the viewpoint of functions with reference to devices of the macroscopic world is indeed a very interesting exercise that introduces novel concepts into chemistry as a scientific discipline.

GLOSSARY Antenna for light harversing (molecular-level) A recognized array of molecular components (i.e., a supramolecular species) capable of absorbing light and delivering the resulting electronic energy to a predetermined component of the array; this function is often called antenna effect. Catenane A supramolecular species consisting of two or more interlocked macrocyclic components. Charge separation A chemical reaction, usually photoinduced, involving the transfer of an electron in a supramolecular species from a neutral component to another neutral component, leading to a species made of a positively and a negatively charged moieties. Logic gate (molecular-level) A molecular or supramolecular species capable of performing a logic operation.

Molecular device An assembly of a discrete number of molecular components (i.e., a supramolecular species) designed to perform a specific function. Molecular machine A particular type of molecular device in which the component parts display changes in their relative position as a result of some external stimulus. Molecular wire A common name to indicate a long, highly conjugated molecule. Pseudorotaxane An inclusion complex in which a molecular thread is encircled by a macrocyclic component; see also rotaxane. Rotaxane A supramolecular species consisting of a dumbbell-shaped component and a macrocyclic component which surrounds the linear portion of the dumbbell and is trapped mechanically by bulky stoppers; when at least one of the stoppers is absent, the macrocyclic component can dethread and the supramolecular species is called pseudorotaxane. Supramolecular chemistry Branch of chemistry bearing on organized entities of higher complexity that result from the association of two or more molecular components.

ACKNOWLEDGMENTS I am indebted to Professors Vincenzo Balzani and Margherita Venturi for their help in the preparation of the manuscript. This work was supported by the University of Bologna and the Ministero dell’Istruzione, dell’Università e della Ricerca.

REFERENCES 1. J.-M. Lehn, “Supramolecular Chemistry.” VCH, Weinheim, 1995. 2. V. Balzani and F. Scandola, “Supramolecular Photochemistry.” Ellis Horwood, Chichester, 1991. 3. V. Balzani, A. Credi, and M. Venturi, in “Supramolecular Science: Where It Is and Where It Is Going” (R. Ungaro and E. Dalcanale, Eds.), p. 1. Kluwer Academic, Dordrecht, 1999. 4. V. Balzani and F. Scandola, in “Comprehensive Supramolecular Chemistry” (J. L. Atwood, J. E. D. Davies, D. D. MacNicol, and F. Vögtle, Eds.), Vol. 10, p. 687. Pergamon, Oxford, 1996. 5. G. M. Wallraff and W. D. Hinsberg, Chem. Rev. 99, 1801 (1999). 6. J. M. Tour, Acc. Chem. Res. 33, 791 (2000). 7. Acc. Chem. Res. 32 (1999). 8. V. Balzani, A. Credi, and F. Scandola, in “Transition Metals in Supramolecular Chemistry” (L. Fabbrizzi and A. Poggi, Eds.), p. 1. Kluwer Academic, Dordrecht, 1994. 9. V. Balzani, L. Moggi, and F. Scandola, in “Supramolecular Photochemistry” (V. Balzani, Ed.), p. 1. Reidel, Dordrecht, 1987. 10. R. P. Feynman, Eng. Sci. 23, 22 (1960). 11. R. Ballardini, V. Balzani, A. Credi, M. T. Gandolfi, and M. Venturi, Acc. Chem. Res. 34, 445 (2001). 12. P. L. Boulas, M. Gómez-Kaifer, and L. Echegoyen, Angew. Chem., Int. Ed. 37, 216 (1998). 13. A. E. Kaifer and M. Gómez Kaifer, “Supramolecular Electrochemistry.” Wiley–VCH, Weinheim, 1999. 14. J. R. Lakowicz, “Principles of Fluorescence Spectroscopy,” 2nd ed. Kluwer Academic, New York, 1999. 15. J.-C. Chambron, C. O. Dietrich-Buchecker, and J.-P. Sauvage, Top. Curr. Chem. 165, 131 (1993). 16. H. W. Gibson, M. C. Bheda, and P. T. Engen, Prog. Polym. Sci. 19, 843 (1994).

546 17. D. B. Amabilino and J. F. Stoddart, Chem. Rev. 95, 2725 (1995). 18. R. Jäger and F. Vögtle, Angew. Chem., Int. Ed. Engl. 36, 930 (1997). 19. S. A. Nepogodiev and J. F. Stoddart, Chem. Rev. 98, 1959 (1998). 20. M. Fujita, Acc. Chem. Res. 32, 53 (1999). 21. D. A. Leigh and A. Murphy, Chem. Ind. 178 (1999). 22. G. A. Breault, C. A. Hunter, and P. C. Mayers, Tetrahedron 55, 5265 (1999). 23. J.-P. Sauvage and C. O. Dietrich-Buchecker, Eds., “Molecular Catenanes, Rotaxanes and Knots.” Wiley–VCH, Weinheim, 1999. 24. F. M. Raymo and J. F. Stoddart, Chem. Rev. 99, 1643 (1999). 25. T. J. Hubin, A. G. Kolchinski, A. L. Vance, and D. L. Busch, Adv. Supramol. Chem. 5, 237 (1999). 26. T. J. Rubin and D. L. Busch, Coord. Chem. Rev. 200–202, 5 (2000). 27. C. Reuter, R. Schmieder, and F. Vögtle, Pure Appl. Chem. 72, 2233 (2000). 28. A. Credi, M. Montalti, V. Balzani, S. J. Langford, F. M. Raymo, and J. F. Stoddart, New J. Chem. 22, 1061 (1998). 29. M. Asakawa, P. R. Ashton, V. Balzani, A. Credi, G. Mattersteig, O. A. Matthews, M. Montalti, N. Spencer, J. F. Stoddart, and M. Venturi, Chem.—Eur. J. 3, 1992 (1997). 30. V. Balzani, P. Ceroni, A. Credi, M. Gómez-López, C. Hamers, J. F. Stoddart, and R. Wolf, New J. Chem. 25, 25 (2001) and references therein. 31. K. N. Houk, S. Menzer, S. P. Newton, F. M. Raymo, J. F. Stoddart, and D. J. Williams, J. Am. Chem. Soc. 121, 1479 (1999). 32. S. J. Loeb and J. A. Wisner, Angew. Chem., Int. Ed. 37, 2838 (1998). 33. P. R. Ashton, P. J. Campbell, E. J. T. Chrystal, P. T. Glink, S. Menzer, D. Philp, N. Spencer, J. F. Stoddart, P. A. Tasker, and D. J. Williams, Angew. Chem., Int. Ed. Engl. 34, 1865 (1995). 34. R. S. Wylie and D. H. Macartney, J. Am. Chem. Soc. 114, 3136 (1992). 35. M. Kimura, Y. Misawa, Y. Yamaguchi, K. Hanabusa, and H. Shirai, Chem. Commun. 2785 (1996). 36. P. R. Ashton, R. Ballardini, V. Balzani, M. C. T. Fyfe, M. T. Gandolfi, M. V. Martínez-Díaz, M. Morosini, C. Schiavo, K. Shibata, J. F. Stoddart, A. J. P. White, and D. J. Williams, Chem.—Eur. J. 4, 2332 (1998). 37. P. R. Ashton, D. Philp, N. Spencer, and J. F. Stoddart, J. Chem. Soc., Chem. Commun. 1677 (1991). 38. N. Yamaguchi and H. W. Gibson, Angew. Chem., Int. Ed. 38, 143 (1998). 39. A. E. Kaifer, Acc. Chem. Res. 32, 62 (1999). 40. M. Montalti, R. Ballardini, L. Prodi, and V. Balzani, Chem. Commun. 2011 (1996). 41. P. R. Ashton, R. Ballardini, V. Balzani, E. C. Constable, A. Credi, O. Kocian, S. J. Langford, J. A. Preece, L. Prodi, E. R. Schofield, N. Spencer, J. F. Stoddart, and S. Wenger, Chem.—Eur. J. 4, 2411 (1998). 42. E. Ishow, A. Credi, V. Balzani, F. Spadola, and L. Mandolini, Chem.—Eur. J. 5, 984 (1999). 43. P. R. Ashton, V. Balzani, A. Credi, O. Kocian, D. Pasini, L. Prodi, N. Spencer, J. F. Stoddart, M. S. Tolley, M. Venturi, A. J. P. White, and D. J. Williams, Chem.—Eur. J. 4, 590 (1998). 44. D. H. Busch, A. L. Vance, and A. G. Kolchinski, in “Comprehensive Supramolecular Chemistry” (J. L. Atwood, J. E. D. Davies, D. D. MacNicol, and F. Vögtle, Eds.), Vol. 9, p. 1. Pergamon, Oxford, 1996. 45. P. L. Anelli, P. R. Ashton, R. Ballardini, V. Balzani, M. Delgado, M. T. Gandolfi, T. T. Goodnow, A. E. Kaifer, D. Philp, M. Pietraszkiewicz, L. Prodi, M. V. Reddington, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, C. Vicent, and D. J. Williams, J. Am. Chem. Soc. 114, 193 (1992). 46. D. B. Amabilino, F. M. Raymo, and J. F. Stoddart, in “Comprehensive Supramolecular Chemistry” (J. L. Atwood, J. E. D. Davies,

Photochemical Molecular Devices

47. 48. 49. 50.

51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63.

64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77.

D. D. MacNicol, and F. Vögtle, Eds.), Vol. 9, p. 85. Pergamon, Oxford, 1996. C. A. Hunter, J. Am. Chem. Soc. 114, 5003 (1992). A. G. Johnston, D. A. Leigh, R. J. Pritchard, and M. D. Degan, Angew. Chem., Int. Ed. Engl. 34, 1209 (1995). F. Vögtle, T. Dunnwald, and T. Schmidt, Acc. Chem. Res. 29, 451 (1996). D. B. Amabilino, M. Asakawa, P. R. Ashton, R. Ballardini, V. Balzani, M. Behloradsky, A. Credi, M. Higuchi, F. M. Raymo, T. Shimizu, J. F. Stoddart, M. Venturi, and K. Yase, New J. Chem. 22, 959 (1998). D. B. Amabilino, P. R. Ashton, V. Balzani, C. L. Brown, A. Credi, J. M. J. Fréchet, J. W. Leon, F. M. Raymo, N. Spencer, J. F. Stoddart, and M. Venturi, J. Am. Chem. Soc. 118, 12012 (1996). D. B. Amabilino, P. R. Ashton, V. Balzani, S. E. Boyd, A. Credi, J. Y. Lee, S. Menzer, J. F. Stoddart, M. Venturi, and D. J. Williams, J. Am. Chem. Soc. 120, 4295 (1998). Z.-T. Li, P. C. Stein, J. Becher, D. Jensen, P. Mørk, and N. Svenstrup, Chem.—Eur. J. 2, 624 (1996). D. B. Amabilino, P. R. Ashton, J. A. Bravo, F. M. Raymo, J. F. Stoddart, A. J. P. White, and D. J. Williams, Eur. J. Org. Chem. 1295 (1999). C. Yamamoto, Y. Okamoto, T. Schmidt, R. Jager, and F. Vögtle, J. Am. Chem. Soc. 119, 10547 (1997). D. P. Häder and M. Tevini, “General Photobiology.” Pergamon, Oxford, 1987. V. Balzani, A. Credi, and M. Venturi, Curr. Opin. Chem. Biol. 1, 506 (1997). D. Gust, T. A. Moore, and A. L. Moore, Acc. Chem. Res. 34, 40 (2001). L. Sun, L. Hammarström, B. Åkermark, and S. Styring, Chem. Soc. Rev. 30, 36 (2001). D. Kuciauskas, P. A. Liddell, S. Lin, T. E. Johnson, S. J. Weghorn, J. S. Lindsey, A. L. Moore, T. A. Moore, and D. Gust, J. Am. Chem. Soc. 121, 8604 (1999). A. Hagfeldt and M. Grätzel, Acc. Chem. Res. 33, 269 (2000). G. Steinberg-Yfrach, P. A. Liddell, S.-C. Hung, A. L. Moore, D. Gust, and T. A. Moore, Nature 385, 239 (1997). J.-S. Hsiao, B. P. Krueger, R. W. Wagner, T. E Johnson, J. K. Delaney, D. C. Mauzerall, G. R. Fleming, J. S. Lindsey, D. F. Bocian, and R. J. Donohoe, J. Am. Chem. Soc. 118, 11181 (1996) and references therein. M. N. Berberan-Santos, P. Choppinet, A. Fedorov, L. Jullien, and B. Valeur, J. Am. Chem. Soc. 122, 11876 (2000) and references therein. V. Balzani, A. Juris, M. Venturi, S. Campagna, and S. Serroni, Chem. Rev. 96, 759 (1996). V. Balzani, P. Ceroni, A. Juris, M. Venturi, S. Campagna, F. Puntoriero, and S. Serroni, Coord. Chem. Rev. 219, 545 (2001). A. Adronov and J. M. J. Fréchet, Chem. Commun. 1701 (2000). M. Maus, R. De, M. Lor, T. Weil, S. Mitra, U.-M. Wiesler, A. Herrmann, J. Hofkens, T. Vosch, K. Müllen, and F. C. De Schryver, J. Am. Chem. Soc. 123, 7668 (2001). J. E. Guillet, Trends Polym. Sci. 4, 41 (1996). G. M. Stewart and M. A. Fox, J. Am. Chem. Soc. 118, 4354 (1996). M. Tamura and A. Ueno, Bull. Chem. Soc. Jpn. 73, 147 (2000). M. Tamura, D. Gao, and A. Ueno, Chem.—Eur. J. 7, 1390 (2001). V. Balzani, Ed., “Electron Transfer in Chemistry,” Vol. 3, Part II. Wiley–VCH, Weinheim, 2001. K. Kalyanasundaram, “Photochemistry of Polypyridine and Porphyrin Complexes.” Academic Press, London, 1992. L. Flamigni, Pure Appl. Chem. 73, 421 (2001). J.-C. Chambron, J.-P. Collin, J.-O. Dalbavie, C. O. DietrichBuchecker, V. Heitz, F. Odobel, N. Solladié, and J.-P. Sauvage, Coord. Chem. Rev. 178–180, 1299 (1998). M.-J. Blanco, M. C. Jiménez, J.-C. Chambron, V. Heitz, M. Linke, and J.-P. Sauvage, Chem. Soc. Rev. 28, 293 (1999).

Photochemical Molecular Devices 78. A. M. Brun, S. J. Atherton, A. Harriman, V. Heitz, and J.-P. Sauvage, J. Am. Chem. Soc. 114, 4632 (1992). 79. J.-C. Chambron, A. Harriman, V. Heitz, and J.-P. Sauvage, J. Am. Chem. Soc. 115, 6109 (1993). 80. J.-C. Chambron, A. Harriman, V. Heitz, and J.-P. Sauvage, J. Am. Chem. Soc. 115, 7419 (1993). 81. M. Linke, J.-C. Chambron, V. Heitz, J.-P. Sauvage, S. Encinas, F. Barigelletti, and L. Flamigni, J. Am. Chem. Soc. 122, 11824 (2000). 82. M. Seiler, H. Dürr, I. Willner, E. Joselevich, A. Doron, and J. F. Stoddart, J. Am. Chem. Soc. 116, 3399 (1994). 83. M. Kropf, E. Joselevich, H. Dürr, and I. Willner, J. Am. Chem. Soc. 118, 655 (1996). 84. E. David, R. Born, E. Kaganer, E. Joselevich, H. Dürr, and I. Willner, J. Am. Chem. Soc. 119, 7778 (1997). 85. Y.-Z. Hu, D. van Loyen, O. Schwarz, S. Bossmann, H. Dürr, V. Huch, and M. Veith, J. Am. Chem. Soc. 120, 5822 (1998). 86. A. C. Benniston, P. R. Mackie, and A. Harriman, Angew. Chem., Int. Ed. 37, 354 (1998). 87. Y.-Z. Hu, S. H. Bossmann, D. Van Loyen, O. Schwarz, and H. Dürr, Chem.—Eur. J. 5, 1267 (1999). 88. Y.-Z. Hu, H. Takashima, S. Tsukiji, S. Shinkai, T. Nagamune, S. Oishi, and I. Hamachi, Chem.—Eur. J. 6, 1907 (2000). 89. J. Deisenhofer and H. Michel, Angew. Chem., Int. Ed. Engl. 28, 829 (1989). 90. R. Huber, Angew. Chem., Int. Ed. Engl. 28, 848 (1989). 91. V. Balzani, A. Credi, F. M. Raymo, and J. F. Stoddart, Angew. Chem., Int. Ed. 39, 3348 (2000). 92. Acc. Chem. Res. 34 (2001). 93. Struct. Bonding 99 (2001). 94. H. Dürr and H. Bouas-Laurent, Eds., “Photochromism: Molecules and Systems.” Elsevier, Amsterdam, 1990. 95. V. Balzani, Ed., “Electron Transfer in Chemistry.” Wiley–VCH, Weinheim, 2001. 96. A. C. Benniston, A. Harriman, D. Philp, and J. F. Stoddart, J. Am. Chem. Soc. 115, 5298 (1993). 97. A. C. Benniston, A. Harriman, and D. S. Yufit, Angew. Chem., Int. Ed. Engl. 36, 2356 (1997). 98. R. Ballardini, V. Balzani, M. T. Gandolfi, L. Prodi, M. Venturi, D. Philp, H. G. Ricketts, and J. F. Stoddart, Angew. Chem., Int. Ed. Engl. 32, 1301 (1993). 99. P. R. Ashton, R. Ballardini, V. Balzani, S. E. Boyd, A. Credi, M. T. Gandolfi, M. Gómez-López, S. Iqbal, D. Philp, J. A. Preece, L. Prodi, H. G. Ricketts, J. F. Stoddart, M. S. Tolley, M. Venturi, A. J. P. White, and D. J. Williams, Chem.—Eur. J. 3, 152 (1997). 100. O. Johansen, A. W. H. Mau, and W. H. F. Sasse, Chem. Phys. Lett. 94, 107 (1983). 101. A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, and A. von Zelewsky, Coord. Chem. Rev. 84, 85 (1988). 102. E. Amouyal, Sol. Energy Mater. Sol. Cells 38, 249 (1995). 103. S. Chia, J. Cao, J. F. Stoddart, and J. I. Zink, Angew. Chem., Int. Ed. 40, 2447 (2001). 104. M. Freemantle, Chem. Eng. News 76, 37 (1998). 105. P. R. Ashton, V. Balzani, O. Kocian, L. Prodi, N. Spencer, and J. F. Stoddart, J. Am. Chem. Soc. 120, 11190 (1998). 106. S. Shinkai and O. Manabe, Top. Curr. Chem. 121, 76 (1984). 107. A. Ueno, H. Yoshimura, R. Saka, and T. Osa, J. Am. Chem. Soc. 101, 2779 (1979). 108. S. Shinkai, T. Nakaji, T. Ogawa, K. Shigematsu, and O. Manabe, J. Am. Chem. Soc. 103, 111 (1981). 109. A. Ueno, K. Takahashi, and T. Osa, J. Chem. Soc., Chem. Commun. 837 (1980). 110. V. Balzani, A. Credi, F. Marchioni, and J. F. Stoddart, Chem. Commun. 1861 (2001). 111. H. Murakami, A. Kawabuchi, K. Kotoo, M. Kunitake, and N. Nakashima, J. Am. Chem. Soc. 119, 7605 (1997). 112. P. Gaviña and J.-P. Sauvage, Tetrahedron Lett. 38, 3521 (1997).

547 113. J.-P. Collin, P. Gaviña, and J.-P. Sauvage, New J. Chem. 21, 525 (1999). 114. N. Armaroli, V. Balzani, J.-P. Collin, P. Gaviña, J.-P. Sauvage, and B. Ventura, J. Am. Chem. Soc. 121, 4397 (1999). 115. P. R. Ashton, R. Ballardini, V. Balzani, A. Credi, K. R. Dress, E. Ishow, C. J. Kleverlaan, O. Kocian, J. A. Preece, N. Spencer, J. F. Stoddart, M. Venturi, and S. Wenger, Chem.—Eur. J. 6, 3558 (2000). 116. A. M. Brouwer, C. Frochot, F. G. Gatti, D. A. Leigh, L. Mottier, F. Paolucci, S. Roffia, and G. W. H Wurpel, Science 291, 2124 (2001). 117. G. W. H. Wurpel, A. M. Brouwer, I. H. M. van Stokkum, A. Farran, and D. A. Leigh, J. Am. Chem. Soc. 123, 11327 (2001). 118. M. J. MacLachlan, A. Rose, and T. M. Swager, J. Am. Chem. Soc. 123, 9180 (2001). 119. F. Vögtle, W. M. Müller, U. Müller, M. Bauer, and K. Rissanen, Angew. Chem., Int. Ed. Engl. 32, 1295 (1993). 120. M. Bauer, W. M. Müller, U. Müller, K. Rissanen, and F. Vögtle, Liebigs Ann. 649 (1995). 121. A. Livoreil, J.-P. Sauvage, N. Armaroli, V. Balzani, L. Flamigni, and B. Ventura, J. Am. Chem. Soc. 119, 12114 (1999). 122. D. Rouvray, Chem. Br. 34, 26 (1998). 123. J. Jortner and M. A. Ratner, Eds., “Molecular Electronics.” Blackwell, Oxford, 1997. 124. A. Harriman and R. Ziessel, Coord. Chem. Rev. 171, 331 (1998). 125. H. L. Anderson, Chem. Commun. 2323 (1999). 126. F. Barigelletti and L. Flamigni, Chem. Soc. Rev. 29, 1 (2000). 127. M. D. Ward, J. Chem. Educ. 78, 1021 (2001). 128. B. L. Feringa, Ed., “Molecular Switches.” Wiley–VCH, Weinheim, 2001. 129. R. M. Metzger, Acc. Chem. Res. 32, 950 (1999). 130. C. A. Brady and J. R. Sambles, in “Electron Transfer in Chemistry” (V. Balzani, Ed.), Vol. 5, Part I, Ch. 4. Wiley–VCH, Weinheim, 2001. 131. Chem. Rev. 100 (2000). 132. A. P. de Silva, D. B. Fox, T. S. Moody, and S. M. Weir, Pure Appl. Chem. 73, 503 (2001). 133. F. M. Raymo and S. Giordani, J. Am. Chem. Soc. 123, 4651 (2001). 134. F. Remacle, S. Speiser, and R. D. Levine, J. Phys. Chem. B 105, 5589 (2001). 135. A. J. Berresheim, M. Muller, and K. Mullen, Chem. Rev. 99, 1747 (1999). 136. B. Schlicke, P. Belser, L. De Cola, E. Sabbioni, and V. Balzani, J. Am. Chem. Soc. 121, 4207 (1999). 137. B. Schlicke, L. De Cola, P. Belser, and V. Balzani, Coord. Chem. Rev. 208, 267 (2000). 138. S. Anderson and H. L. Anderson, Angew. Chem., Int. Ed. Engl. 35, 1956 (1996). 139. P. N. Taylor, M. J. O’Connell, L. A. McNeill, M. J. Hall, R. T. Aplin, and H. L. Anderson, Angew. Chem., Int. Ed. 39, 3456 (2000). 140. R. Ballardini, V. Balzani, M. Clemente-León, A. Credi, M. T. Gandolfi, E. Ishow, J. Perkins, J. F. Stoddart, H.-R. Tseng, and S. Wenger, J. Chem. Soc., in press. 141. C. H. Roth, Jr., “Fundamentals of Logic Design.” PWS, Boston, 1995. 142. A. P. de Silva, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J. M. Huxley, C. P. McCoy, J. T. Rademacher, and T. E. Rice, Chem. Rev. 97, 1515 (1997). 143. A. P. de Silva, H. Q. N. Gunaratne, and C. P. McCoy, Nature 364, 42 (1993). 144. A. P. de Silva, H. Q. N. Gunaratne, and C. P. McCoy, J. Am. Chem. Soc. 119, 7891 (1997). 145. H.-G. Ji, R. Dabestani, and G. M. Brown, J. Am. Chem. Soc. 122, 9306 (2000).

548 146. T. Gunnlaugsson, D. A. Mac Dónail, and D. Parker, Chem. Commun. 93 (2000). 147. A. P. de Silva and N. D. McClenaghan, J. Am. Chem. Soc. 122, 3965 (2000). 148. V. Balzani, A. Credi, S. J. Langford, and J. F. Stoddart, J. Am. Chem. Soc. 119, 2769 (1999).

Photochemical Molecular Devices 149. C. P. Collier, G. Mattersteig, E. W. Wong, Y. Luo, K. Beverly, J. Sampaio, F. M. Raymo, J. F. Stoddart, and J. R. Heath, Science 289, 1172 (2000). 150. J. K. Gimzewski and C. Joachim, Science 283, 1683 (1999). 151. H. Shigekawa, K. Miyake, J. Sumaoka, A. Harada, and M. Komiyama, J. Am. Chem. Soc. 122, 5411 (2000).