Structural control at the organic–solid interface

Abstract

The structure–function relationships of a series of bistable [2]rotaxane and [2]pseudorotaxane-based devices have been evaluated across different length scales. The switching characteristics of bistable [2]rotaxanes and self-assembled [2]pseudorotaxanes, which can be controlled chemically, electrochemically, or photochemically, enable them to function as prototypes of molecular machines. The switching processes are operative, not only in solution, but also in a wide variety of condensed phases. The universality of the switching mechanism demonstrates that these functional organic materials can be incorporated onto solid metallic and inorganic supports for device applications, despite the fact that interactions at the organic substrate interface can influence molecular structure and function. Through iterative design–analysis feedback loops that focus upon fine-tuning device performance, based on molecular structures and molecule-substrate interactions, the fabrication of functioning micro-actuators, nanovalves and light-harvesting devices has been achieved.

---

Adam B. Braunschweig

Adam B. Braunschweig received his B.A. degree in Chemistry from Cornell University, Ithaca, New York, in 2001 while working with Professor D. Y. Sogah on the synthesis of amphiphilic block copolymers. In 2001, he joined the laboratory of Professor J. F. Stoddart in the Department of Chemistry and Biochemistry at the University of California, Los Angeles where he is conducting research on the synthesis and characterisation of molecular and supramolecular materials.

Brian H. Northrop

Brian H. Northrop received his B.A. degree in Chemistry from Middlebury College in Vermont in 2001 while working with Professor J. Byers on the synthesis of indole derivatives via radical aromatic substitution. In 2002 he joined the laboratories of Professors J. F. Stoddart and K. N. Houk in the Department of Chemistry and Biochemistry at the University of California, Los Angeles. His research includes the synthesis and computational modelling of interlocked molecular bundles, single-molecule force spectroscopy of [2]rotaxanes, and the study of charge–charge interactions in mechanically interlocked molecules.

J. Fraser Stoddart

J. Fraser Stoddart received all his degrees (B.Sc., Ph.D., D.Sc.) from the University of Edinburgh. He presently holds the Fred Kavli Chair of Nanosciences at the University of California, Los Angeles, where he is also the Director of the California NanoSystems Institute. His research interests relate to the nature of the mechanical bond and the properties and functions of mechanically interlocked molecular compounds.

---

1. Introduction

The unique properties of noncovalently associated and mechanically interlocked molecules, such as pseudorotaxanes^1,2 and switchable, bistable rotaxanes^3,4 and catenanes,^4,5 respectively, have shown considerable promise as prototypes of functional nanosystems,^6,7 such as sensors, actuators, valves, and amplifiers, on account of their ability to undergo controlled mechanical motions—for example, switching—with an appropriate external stimulus. [2]Rotaxanes are comprised of a ring component, encircling a dumbbell-shaped component (Fig. 1a), while [2]catenanes are composed of two mutually interlocking rings (Fig. 1b). The presence of two different recognition sites on the dumbbell of a [2]rotaxane, or on one of the two rings of a [2]catenane, enables these molecules to function as bistable, molecular switches.^3,5 Alternatively, removal of the large end groups of [2]rotaxanes—the stoppers—gives rise to pseudorotaxanes^1,2 (Fig. 1c), where a ring is able to move on and off a rod component with the equilibrium ratio of the pseudorotaxane to the free ring and rod being determined by the strength of the binding.

Fig. 1 Schematic structural representations of (a) a bistable [2]rotaxane³ and (b) a bistable [2]catenane⁵. The ring component in both the bistable [2]rotaxane and bistable [2]catenane reside predominantly around the more favourable recognition site at equilibrium. (c) Dynamic equilibrium for threading and dethreading of a [2]pseudorotaxane.¹

The host–guest system based on the π-electron poor cyclophane host, cyclobis(paraquat-p-phenylene) (CBPQT⁴⁺), and π-electron rich guests has been used extensively by our group and others to make surface-bound functional supramolecular organic materials. CBPQT⁴⁺ binds such π-electron rich guests as neurotransmitters,⁸ hydroquinones,^3,5,9 dioxynaphthalenes^2,3,5,9 and tetrathiafulvalenes^3,5,10 through complementary [C–H⋯O], [π⋯π], [C–H⋯π], and hydrophobic interactions (Fig. 2). Factors that increase the strength of complexation with CBPQT⁴⁺ include increasing the π-surface overlap¹⁰ between host and guest and the addition of glycol chains such as di- or triethylene glycol (DEG and TEG),⁹ which increase binding as a result of the ability of their ether oxygen atoms to form [C–H⋯O] bonds with the more acidic hydrogen atoms on the cyclophane. Host–guest complexation involving the tetracationic cyclophane, as its tetrakis(hexafluorophosphate) salt (CBPQT·4PF₆), occurs in a variety of organic solvents—and in water, as the corresponding tetrachloride salt CBPQT·4Cl.

Fig. 2 X-Ray crystal superstructure showing the complexation of DEG-DNP by CBPQT⁴⁺. Complexation is stabilised by noncovalent [C–H⋯O], [C–H⋯π], and [π⋯π] interactions, as indicated by dashed lines. Hydrogen atoms not participating in [C–H⋯O] or [C–H⋯π] interactions have been removed for clarity.

The process of chemically, electrochemically, or photochemically controlled mechanical switching of CBPQT⁴⁺-based systems has been studied extensively in the solution phase with a range of analytical techniques,^3,5,11 such as ¹H NMR spectroscopy, cyclic voltammetry (CV), and UV–vis spectroscopy. The solution phase is, however, an incoherent environment where molecular conformations and orientations, as well as relative molecular motions, are far from uniform. Different solvents affect the strength of binding between recognition units and may also affect intermolecular interactions and molecular conformations. In addition, the relative orientations of switchable molecules in solution are random, precluding the possibility of achieving directed motion. Thus, the solution phase is not suitable for the majority of device applications. Attaching molecules to surfaces provides a potential solution to the problems of orientation, coherence, conformation and coordination. It is important, however, to work in a regime where bistability and not surface effects determine performance, otherwise the rich solution chemistry of bistable, switchable molecules is negated. Therefore, thorough studies of how these switchable molecules self-assemble, pack, align, and function on surfaces is essential to device design. With this knowledge and by working at the organic–solid interface, device properties are controlled through an iterative feedback loop incorporating experimental results and a qualitative appreciation of the influence of structural modification of the organic components upon molecular behaviour and their method of surface attachment.

2. Modular synthesis

The design of functional organic materials from host–guest complexes of CBPQT⁴⁺ involves the close interplay between synthetic and structural considerations so that the structures and surfaces can work in unison to reveal the desired molecular function(s). The synthesis of host–guest systems involving the π-electron poor cyclophane CBPQT⁴⁺ and π-electron rich guests has evolved to the point where a modular molecular toolkit (Fig. 3) is available^3,5,8–10 for the preparation of any number of related compounds or complexes (Scheme 1) such that each molecule or supermolecule can be customized for a specific purpose.

Fig. 3 Molecular structures and schematic representations of a collection of (a) cyclophanes, (b) recognition units, and (c) stoppers used for the construction of a myriad of [2]rotaxanes for a wide range of nanoscale applications.

Scheme 1 Schematic representation of a modular synthesis of a bistable [2]rotaxane. A variety of different molecular components may be incorporated into the synthesis to arrive at a diverse structural collection of bistable [2]rotaxanes.

Functional bistable compounds and host–guest complexes can be tailored by changing the nature of the recognition sites, the spacers between the sites, the stoppers (or lack thereof), and the method by which the molecules or supermolecules interact with the surface. To form mechanically interlocked [2]rotaxanes and [2]catenanes, the cyclophane is either clipped,¹² slipped,¹³ or stoppered¹⁴ onto the dumbbell (Scheme 2). A molecule that is stoppered, such that the CBPQT⁴⁺ ring cannot be removed from the linear dumbbell portion under normal conditions, is a [2]rotaxane, and if the ring is free to move on or off the linear guest, then the resulting supermolecule is a [2]pseudorotaxane (Fig. 1a and 1c, respectively). In the case of [2]pseudorotaxanes, complexation is favoured on account of a decrease in free energy arising from the formation of noncovalent bonding interactions between the ring and recognition units on the rod component. The molecular structures of the CBPQT⁴⁺-based [2]rotaxanes have been developed to include bistable molecular shuttles, wherein the positions of the ring can be partitioned between two different recognition sites, depending on the relative strengths with which they interact with the CBPQT⁴⁺ ring. In the case of a [2]rotaxane composed of a CBPQT⁴⁺ ring and a dumbbell containing both tetrathiafulvalene (TTF) and 1,5-dioxynaphthalene (DNP) components, redox-controlled mechanical switching can be achieved^3,5 either chemically or electrochemically (Fig. 4). At equilibrium, the CBPQT⁴⁺ ring interacts preferentially with the TTF recognition unit, resulting in the ground state co-conformation (GSCC) of the molecule. Oxidation of the TTF to its dication (TTF²⁺) causes the CBPQT⁴⁺ ring to move to the DNP recognition unit because of charge–charge repulsion that leads to the disappearance of the energy minimum associated, at the onset, with the stabilising interactions between CBPQT⁴⁺ and TTF. After reduction of the TTF²⁺ dication back to its neutral state (TTF), the CBPQT⁴⁺ ring continues to encircle the less energetically stable DNP site, in what can now be referred to as the metastable state co-conformation (MSCC) whose lifetime is determined by the activation barrier between the MSCC and the GSCC. As a result of thermal relaxation, the molecule will return, in the course of time, to the GSCC, thus completing the switching cycle (Fig. 4).

Fig. 4 The redox-controlled switching process of a bistable [2]rotaxane.³ In the ground state co-conformation (GSCC), the CBPQT⁴⁺ ring (blue) preferentially encircles the TTF recognition unit (green). A two-electron oxidation causes the ring to switch to the DNP recognition unit (red) as a result of charge–charge repulsion between the tetracationic cyclophane and the dicationic TTF²⁺ (yellow). After reduction of TTF²⁺ to its neutral state, the CBPQT⁴⁺ ring continues to encircle the DNP unit in the so-called metastable state co-conformation (MSCC). Thermal relaxation allows switching of the CBPQT⁴⁺ back to the GSCC.

Scheme 2 Three different methods for the construction of [2]rotaxanes. Clipping¹³ (a) involves the template-directed self-assembly of a CBPQT⁴⁺ ring onto the recognition unit of a dumbbell. The slipping¹⁴ process (b) requires the cyclophane to, upon thermal activation, pass over the bulky stopper at one end of the dumbbell. Alternatively, the formation of a [2]pseudorotaxane, followed by the addition of bulky stoppering units, may also be used to form [2]rotaxanes in a process (c) referred to as threading-followed-by-stoppering.¹⁵

There are two considerations in designing organic supermolecules and molecules to work in tandem with inorganic and metallic surfaces in such a way that the behaviour of the supermolecule or molecule remains at least qualitatively the same after attachment. They are (1) the appropriate choice of condensed phase and (2) the means by which the supermolecules or molecules are associated with or tethered to the surface. Surfaces that have been used in conjunction with the CBPQT⁴⁺ recognition system include Si,¹⁵ Ti,¹⁵ SiO₂,^16,17 Au,¹⁸ TiO₂¹⁹ and carbon nanotubes,²⁰ each of which relies upon a different means of chemical attachment to the organic species. Molecules can be bound to surfaces either by chemisorption,²¹ where the molecule and surface are covalently linked, or by physisorption,²² where noncovalent interactions hold the surface and molecule together—or even by some regime that is intermediate between the two.¹⁸ Chemisorption involves robust surface–molecule bonds that are made under kinetic control and so may require harsh reaction conditions in order to be formed. Physisorption is characterised by much milder reaction conditions, such as thermodynamically-controlled self-assembled monolayer (SAM) formation, involving weak bonding between the surface and the adsorbed species. The weaker bonding involved in physisorption is dynamic and leads to the most thermodynamically stable monolayer in the course of time. Covalent attachment of alkylsilylethers to SiO₂ provides an example of chemisorption (ΔH°_C–Si = 83 kcal mol⁻¹), while the binding between pyrenes and single-walled carbon nanotubes (SWNTs)²³ represents a form of physisorption (ΔH°_π-π = 3–7 kcal mol⁻¹). SAMs formed by the adsorption of thiols on gold^18,24 (ΔH°_Au–S = 44 kcal mol⁻¹) lie in that regime between chemisorption and physisorption. The range of possible surfaces, the structural modularity of organic compounds, and the means of attachment between the two provide a chemical toolkit that allows for the rational design of surface morphology under either kinetic or thermodynamic control.

3. Thermodynamic control of switching

The equilibrium ratio of MSCC to GSCC (N_MSCC/N_GSCC) in bistable [2]rotaxanes (1) in solution, (2) in viscous polymer matrices and (3) on surfaces is determined by the difference in the binding strengths between the two recognition sites on the dumbbell and the ring.¹¹ The thermodynamic binding parameters (ΔH°, ΔS°, ΔG°) measured for the formation of [2]pseudorotaxanes can be used to predict approximately the ratios of MSCC to GSCC (N_MSCC/N_GSCC) in the corresponding bistable [2]rotaxanes made up of these individually studied components. The equilibrium ratio of these two states (K_eq) depends on the difference in binding free energy (ΔΔG°) between the two recognition sites.

K_eq = e^{−(ΔΔG°)/RT} (1)

ΔΔG° is estimated from the difference in binding (K_a) of the components (i.e., TTF–DEG or DNP–DEG) with CBPQT⁴⁺, upon formation of the [2]pseudorotaxane. For example, a comparison of [2]pseudorotaxanes involving the CBPQT⁴⁺ ring host and either (1) a TTF derivative with diethylene glycol substituents (TTF–DEG) or (2) a DNP derivative with diethylene glycol substituents (DNP–DEG) reveals that the binding of TTF–DEG by the CBPQT⁴⁺ ring is in excess of 2 kcal mol⁻¹ more favourable than is the binding with DNP–DEG. Hence, it follows that, in a [2]rotaxane containing both TTF and DNP recognition units, the CBPQT⁴⁺ ring will encircle the TTF unit in more than 95% of the molecules, i.e., one of the translational isomers is greatly preferred over the other one.

The variation in K_eq with temperature across all three environments in bistable [2]rotaxanes can be controlled rationally by choosing recognition sites based on guests with appropriate thermodynamic binding parameters for host–guest complexes ([2]pseudorotxanes), which have been determined in solution by a variety of techniques, including isothermal titration microcalorimetry (ITC), the UV–vis dilution and the single-point ¹H NMR spectroscopic methods. To approximate the values of ΔΔG°, ΔΔH° and ΔΔS° for a bistable [2]rotaxane, the values from the different model complexes are subtracted from each other. The K_eq is determined from eqn. (1) and the temperature dependence of K_eq is determined from eqn (2).

ΔΔH°/T − ΔΔS° = −RlnK_eq (2)

Notably, a large ΔΔH° between the two recognition sites will result in a significant change in K_eq when the temperature (T) is changed, whereas if ΔΔH° is small, almost no change in N_MSCC/N_GSCC is observed as T varies. Experimental results demonstrate¹¹ that, in solution, in a polymer matrix and also in a SAM, a large change in K_eq as the temperature is varied is a consequence of a large ΔΔH° between the two recognition sites on the dumbbell, whereas almost no change in K_eq is observed when ΔΔH° is small. Because of this consistency across phases between molecular structure and the behaviour of K_eq, the properties of bistable [2]rotaxanes on surfaces can be predicted by studying the complexation of model [2]pseudorotaxanes in solution.

4. Conformations of thin films

Our initial motivation for studying the surface properties of bistable, mechanically interlocked molecules was to establish, after some controversy arose,²⁵ that the mechanism of switching was a result of the mechanical movement of the CBPQT⁴⁺ ring in bistable [2]rotaxanes and bistable [2]catenanes. The information gained from Langmuir–Blodgett (LB) studies was used to understand and improve both the manufacturing process and the switching behaviour of molecular switch tunnel junctions^16,17,20 (MSTJs) employed in molecular electronic devices. In MSTJs incorporating bistable [2]rotaxanes, a monolayer of [2]rotaxane molecules is deposited on an array of Si nanowires, followed by the electron-beam deposition of a perpendicular set of Ti/Al electrodes on top of the one-molecule thick layer. Upon oxidation of the bistable [2]rotaxane, the current between the MSTJ increases as a result of a co-conformational change from the GSCC, the lower conductance state, to the MSCC, the higher conductance state. Factors, such as how the bistable molecules pack, the intermolecular interactions between them and the mean molecular area, all influence device performance. These data have led to some improvements in switching speed, better N_MSCC/N_GSCC (on/off) current ratios and hence manufacture. The current situation has been discussed extensively in numerous reviews.²⁶ Most importantly, the information gained from the fundamental surface-based chemistry has led to a series of other binary devices based on switchable, bistable, mechanically interlocked molecules.

The properties of SAMs at the air–water interface can be employed to rationalise the molecular behaviour on a variety of surfaces; by changing molecular structure methodically and understanding the effects of these modifications on single-molecule thick superstructures, specific monolayer properties can be obtained rationally. To determine the effects of structural modifications on monolayer stability, viscosity, packing area, conformation and thickness, LB monolayers of a series of dumbbells and bistable [2]rotaxanes with the DNP and TTF recognition sites for the CBPQT⁴⁺ ring that are most commonly used to make functional devices, were prepared.²⁷ The hydrophilic tail length and end groups of the stoppers, the relative positions of the TTF and DNP recognition sites along the dumbbell component, and the presence or absence of a CBPQT⁴⁺ ring on the dumbbell were all varied (Fig. 5). The consequences of these structural variations were evaluated. The effects of modifying experimental parameters, such as subphase temperature, monolayer compression rates, spreading solvent, and salt concentration were also all investigated.

Fig. 5 Molecular structures of the structurally diverse dumbbells and [2]rotaxanes used in the Langmuir–Blodgett studies.²⁸

The amphiphilic bistable [2]rotaxanes and the dumbbell compounds which serve as the templates for their synthesis are shown in Fig. 5. They are expected to have contorted conformations at the air–water interface as a result of the variable hydrophilicities and hydrophobicities of their constituent parts. TTF, di- and tetraethylene glycol, and CBPQT⁴⁺ are all hydrophilic, while the large aromatic surfaces of DNP and the tetraarylmethane stoppers are hydrophobic. The presence or absence of the CBPQT⁴⁺ ring gave rise to the most pronounced changes in Langmuir isotherms. Collapse pressures of bistable [2]rotaxane films averaged ∼63 mN m⁻¹, whereas films of the respective dumbbell compounds showed average collapse pressures of ∼45 mN m⁻¹, indicating increased stability of the [2]rotaxane monolayers over the dumbbell monolayers. The results suggest that the presence of the CBPQT⁴⁺ ring contributes to hydrophilic anchoring at the air–water interface in the liquid-expanded phase (Fig. 6) of the monolayers. The added structural integrity of [2]rotaxane films may also be the result of dispersive interactions between CBPQT⁴⁺ rings in adjacent molecules. Altering the length of the polyether tails attached to the hydrophilic stopper from diethylene glycol (DEG) to tetraethylene glycol (TEG) decreases viscosity and increases compressibility at high surface pressures. The longer TEG tails appear to be spread out along the air–water interface and so interact with the TEG tails of adjacent molecules much more than do the DEG tails, which only interact when they are sufficiently compressed. It is likely that the CBPQT⁴⁺ ring is either of the same size or slightly larger than the DEG tails, indicating that the stoppers with the DEG tails sit more upright on the surface. The more upright conformation of the shorter DEG tails explains their smaller mean molecular area, greater viscosity, stronger intermolecular interactions, and decreased compressibility when compared to the longer, more flexible TEG tails. In the case of both DEG and TEG tails, the effect of changing from OH to OMe end groups is the same: greater hydrophilicity and viscosity are observed with OH end groups, while OMe end groups result in a larger molecular footprint. In addition to recording Langmuir isotherms, ellipsometry and scanning force microscopy measurements were also performed on all monolayer films. Tapping mode scanning force microscopy showed all the films are smooth with very few multilayer domains. Film thicknesses, as measured by ellipsometry (21–39 Å), were significantly less than the lengths of elongated [2]rotaxanes and dumbbells (65–75 Å), lending further evidence to the notion that the molecules adopt folded conformations. Otherwise, they must lie at quite acute angles to the surface. This wealth of experimentally derived information allows us to design molecules more rationally that form monolayers with robust mechanical properties by understanding the role of the stopper, the CBPQT⁴⁺ ring and the influence of different co-conformations.

Fig. 6 Schematic representation of amphiphilic bistable [2]rotaxanes at the air–water interface of a Langmuir–Blodgett film.²⁸ In the liquid-expanded phase, the bistable [2]rotaxanes adopt an elongated conformation where the hydrophilic CBPQT⁴⁺ ring of each [2]rotaxane is able to come in contact with the water. An increase in surface pressure causes a π-transition to the liquid-condensed phase where the conformations of the [2]rotaxanes are more upright and elongated.

The effects of varying the subphase temperature, the compression rate, the spreading solvent and the subphase salt concentration on the LB films were also investigated. While the subphase temperature and compression rate were shown to affect the stabilities of the monolayers, the spreading solvent and subphase salt concentration did not. Increasing the subphase temperature during film preparation accelerated relaxation of the films, resulting in more stable monolayers. An increase in the compression rate, however, produced less stable films since there is less time for molecular equilibration to occur, i.e., films prepared at slower compression rates were more stable. Notably, the changes in experimental parameters only affected film stability and did not influence molecular area, conformation, and viscosity. These results indicate that molecular structure, not experimental processing, has the greatest bearing on how amphiphilic bistable [2]rotaxane molecules arrange themselves on surfaces. Overall, LB studies of the range of structurally different [2]rotaxanes and dumbbells indicate that [2]rotaxanes with DEG tails and OH end groups display the greatest film stabilities in the liquid-expanded phase and are most likely to be compressed further into relatively stable liquid-condensed phases. We reiterate that the experimental data, obtained on changing the experimental conditions of the monolayers, suggest that the structure of the molecules, and not the experimental conditions employed in the formation of the monolayers, has the greatest influence on the structure of the resulting film.

Computational simulations of [2]rotaxane monolayers provide insight into the specific orientations and packing of the molecules in the monolayers, and the alterations in packing and energy that occur within the monolayer as the molecules switch between the GSCC and the MSCC, leading to changes in the macroscopic properties of the monolayers. Goddard et al.²⁸ have performed fully atomistic molecular dynamics simulations on a simulated Au(111) surface at 300 K on a bistable [2]rotaxane, containing a disulfide tether at one terminus and a hydrophobic stopper at the other. Cells of 16 [2]rotaxane molecules were constructed, with dimensions ranging from over-packed (65 Ų molecule⁻¹) to under-packed (353 Ų molecule⁻¹) representing the extremes of the liquid-condensed and liquid-expanded phases. Both the GSCC and MSCC of the bistable [2]rotaxane were modelled, and the relative stabilities, surface tensions and stress distributions, were calculated within each SAM. Molecular dynamics simulations showed that the most stable packing for both the GSCC and MSCC of the bistable [2]rotaxane corresponds to a mean molecular area of 115 Ų molecule⁻¹ when confined to a 4 × 4 grid of Au atoms, and experimentally, the amphiphilic bistable [2]rotaxanes studied by Frank et al.²⁷ had mean molecular areas of 120–180 Ų molecule⁻¹. The surface tensions of the most stable GSCC and MSCC monolayers were calculated to be 45 and 65 dyn cm⁻¹, respectively. The lower surface tension is a result of the [2]rotaxanes being in the more energetically favourable co-conformation, and the changes in packing and energy that occur within the monolayer as the molecules switch between the GSCC and the MSCC lead to an increase in surface tension. These calculations explain why SAMs composed of bistable [2]rotaxanes in the GSCC are more hydrophobic and thus have larger contact angles than bistable [2]rotaxanes in the MSCC, an observation which is consistent with experimental measurements performed on bistable [2]rotaxanes that are sterically locked in either the GSCC or MSCC. For all cell dimensions, the GSCC was shown to be more stable than the MSCC by 14 kcal mol⁻¹ for the 16 molecule ensemble, a result which is accounted for by the per molecule energy difference of approximately 0.9 kcal. These calculations demonstrate that the macroscopic surface properties of a monolayer of bistable [2]rotaxanes are altered by the co-conformational change as the molecules switch from the GSCC to the MSCC.

Computational simulations also provided insight into the conformations and orientations of the bistable [2]rotaxanes on Au(111) surfaces. With large cell dimensions, representing liquid-expanded phases, the CBPQT⁴⁺ ring lies parallel to the surface. As surface pressure is increased, the tilt angle of the CBPQT⁴⁺ ring and the dumbbell relative to the Au(111) surface increases. In the most stable 4 × 4 grid, tilt angles of the CBPQT⁴⁺ ring in the GSCC and MSCC are 39° and 61°, respectively, while the tilt angle of their respective dumbbells are 41° and 46°. Computations predict²⁸ the optimal thickness of the most stable SAMs to be 40.5 Å for the GSCC and 40.0 Å for the MSCC, distances which are in reasonable agreement with experimental ellipsometry measurements²⁷ of 21–39 Å. These computational and experimental studies provide feedback on the affects of structural modifications upon the superstructure of the bistable [2]rotaxane SAMs.

Building upon the insight gained from LB studies as to what structural factors lead to stable [2]rotaxane monolayer films, it was then essential to establish that mechanical switching is still possible in tightly packed LB monolayers, where individual molecules have fewer degrees of freedom compared with when they are in the solution phase. Using X-ray photoelectron spectroscopy (XPS),²⁹ we were able to observe the directional movement of the CBPQT⁴⁺ ring up and down the dumbbell components of two constitutionally isomeric bistable [2]rotaxanes. Quantitative spectroscopic analysis shows that the CBPQT⁴⁺ ring moves from one recognition site to the other as a result of the introduction of the oxidizing agent Fe(ClO₄)₃ into the aqueous subphase in the Langmuir trough. The evidence for redox-controlled mechanical switching of the bistable [2]rotaxanes in highly packed Langmuir films is derived from the fact that XPS photoemission intensity is exponentially related to the depth of an atom in the film, and therefore the depths for each atom within a film can be differentiated. By comparing photoemission in LB films prepared with and without Fe(ClO₄)₃ in the aqueous subphase, the mechanical switching process could be observed, statically, by differences in N^1s intensity. XPS results for LB films prepared in the presence of oxidant showed an increase in N^1s intensity—corresponding to a ring movement of 1.4 nm—over those films that were prepared with no oxidant in the aqueous subphase. The changes in the intensity of only N^1s support directly the hypothesis that the ring switches between the upper and lower recognition sites as a result of redox control in closely packed films. These static experiments, however, account for the position of the CBPQT⁴⁺ ring before or after redox-controlled switching, but do not track the switching process itself. Recent work using X-ray reflectometry has provided direct evidence³⁰ of dynamic redox-controlled switching of amphiphilic, bistable [2]rotaxanes at the air–water interface in real time. Thus, extensive studies of bistable [2]rotaxanes provide insight into the nuances of molecular structure, manufacture parameters and operation at the interface between solid and organic phases.

5. Solid-state devices

The thorough investigation into the behaviour of CBPQT⁴⁺-based bistable [2]catenanes, and [2]rotaxanes, as well as [2]pseudorotaxanes, in solutions, in polymer gels, and on monolayers has provided a foundation to design molecules rationally for a variety of solid-state devices. This underpinning work has provided the basis for bistable molecules that (1) behave as supramolecular and molecular nanovalves, (2) undergo mechanical movement in response to chemical input, and (3) harness photochemical energy to power a supramolecular machine. In each case, the final molecular design has been the result of an iterative feedback process in which the bistable molecules are studied at surfaces and the experimental results are used to redesign and optimise the desired properties of the device.

6. Supramolecular and molecular nanovalves

The potential for using bistable systems, either supramolecular or molecular, for constructing nanovalves³¹ on a hexagonal array of 2 nm cylindrical pores of a sol–gel follows from the ability of the CBPQT⁴⁺ ring (1) to close these pores sterically—and, with an external stimulus, move away from the surface in either a [2]pseudorotaxane or a [2]rotaxane—thus (2) opening up the nanopores.¹⁷ Initially, a redox-controlled [2]pseudorotaxane^17a was used to gate-keep the nanopores by orchestrating the movement of luminescent molecules from within mesostructured silica by the deslipping of the ring from the thread component, which was covalently linked to the sol–gel surface. In our search for a more robust and reusable design, the [2]pseudorotaxane was replaced by a bistable [2]rotaxane^17b as the gatekeeper presiding over the opening and closing of the nanopores.

[2]Pseudorotaxanes have been shown to operate (i.e., thread and dethread in response to external stimuli) in solution, when trapped within a silica matrix,¹⁶ and on silicon surfaces.¹⁷ For example, photochemical reduction and oxidation by O₂ have been shown¹⁶ to induce dethreading and threading, respectively, of [2]pseudorotaxanes composed of a CBPQT⁴⁺ ring and DEG–DNP encased in a silica framework. Irradiation at 365 nm of the trapped [2]pseudorotaxanes in the presence of a photosensitizer and a photoactive reductant resulted in photo-induced reduction of the CBPQT⁴⁺ ring and, after a period of one hour, dethreading of the [2]pseudorotaxane. Exposure of the silica matrix to air resulted in oxidation of the CBPQT³⁺˙ ring and subsequent rethreading. Alternatively, surface functionalisation of silica can be achieved³² by utilizing the silanol groups at the surfaces of sol–gel films. The sol–gels react with (substituted) triethoxysilanes, such as isocyanatopropyltriethoxysilane and 3-aminopropyltriethoxysilane, to form monolayers of isocyanate or amine tethers on films of silica. These surface-bound organic functional groups may then react with, for example, primary alcohols such that various organic compounds can easily be attached to silica. This procedure was used to append DNP-containing threads to silica (Fig. 7a). These threads were then complexed with CBPQT⁴⁺. Threading and dethreading of the CBPQT⁴⁺ ring at the surface was achieved both photochemically and chemically using NaCNBH₃.

Fig. 7 Operational supramolecular nanovalves. (a) A [2]pseudorotaxane nanovalve^18a where the complexation of CBPQT⁴⁺ by DEG-DNP covers SiO₂ nanopores and traps Ir(PPy)₃ dye molecules inside. Reduction of the CBPQT⁴⁺ ring induces dethreading of the [2]pseudorotaxane and releases the dye. (b) Reversible [2]rotaxane nanovalves^18b can be loaded with dye molecules while in their ground state co-conformation (GSCC). Chemical oxidation traps the dye molecules within the nanoporous SiO₂ as a result of the CBPQT⁴⁺ ring blocking their openings. Reduction with ascorbic acid restores the GSCC and releases the dye.

[2]Pseudorotaxanes have been employed^17a to fabricate nanoscale valves that can be opened and closed as a result of the decomplexation and complexation of the CBPQT⁴⁺ ring. The DNP-containing threads were covalently linked to the surface of mesostructured silica, which contains a hexagonal array of cylindrical nanopores with diameters of ∼2 nm. Immersing the derivatised mesostructured silica in a 1.0 mM PhMe solution of the 1 nm diameter luminescent metal complex, tris(2,2′-phenylpyridyl)iridium(III), or Ir(PPy)₃, loaded the porous silica by diffusion. Subsequent immersion of the loaded derivatised silica in a 1.0 mM aqueous solution of CBPQT·4Cl resulted in the formation of [2]pseudorotaxanes, sterically sealing off the nanopores and trapping the Ir(PPy)₃ molecules inside. Chemical reduction of the CBPQT⁴⁺ ring turns off the recognition between the ring and thread components, opening the valve, and releasing the Ir(PPy)₃ molecules into the solution. The escape of the Ir(PPy)₃ molecules from the nanopores was observed by following the increase in the luminescence intensity of the solution.

Perhaps the most important structural factor governing nanovalve function is the size complementarity between the CBPQT⁴⁺ ring, the Ir(PPy)₃ molecules and the porous nanostructured silica. The ring is large enough to cover the pore openings without being so large as to hinder [2]pseudorotaxane formation as a result of intermolecular steric repulsion. DNP was an effective thread component on account of its strong complexation (∼3.9 kcal mol⁻¹) with CBPQT⁴⁺. Although this system works as an effective gatekeeper to the nanopores and is reusable, for each new cycle, additional CBPQT⁴⁺ has to be added to the system. Leakage is also an issue because of the spontaneous dethreading of the CBPQT⁴⁺ ring from the [2]pseudorotaxane under equilibrium conditions.

Encouraged by results from both LB²⁷ and XPS²⁹ studies, the nanovalve was redesigned to incorporate a redox-active bistable [2]rotaxane^17a to control the opening and closing of the pores (Fig. 7b). The bistable [2]rotaxane attached to the sol–gel surface was designed such that the DNP unit was closest to the nanopores, and the TTF unit was furthest from the nanopores. The porous silica can be loaded with Ir(PPy₃) molecules when the surface bound [2]rotaxanes are in their GSCC, in which the CBPQT⁴⁺ ring is sufficiently far away from the nanopore openings to allow the dye molecules to go inside. In the presence of an oxidant, however, the CBPQT⁴⁺ ring moves to the DNP unit, sitting significantly close to the nanopores so that the dye cannot escape because of blockage of the nanopore entrances by the CBPQT⁴⁺ ring. According to LB studies, the mean molecular area of the bistable [2]rotaxane is controlled by the CBPQT⁴⁺ ring and is greater than 1 nm², effectively closing the nanopores. Therefore, the Ir(PPy)₃ cannot escape from the nanopores while the ring sits close to the orifices. Upon reduction of the TTF unit, the ring undergoes a mechanical movement away from the openings of the nanopores, and the release of the dye occurs. In contrast to the [2]pseudorotaxane-based nanovalve, the nanopores controlled by [2]rotaxane-based gatekeepers can be opened and closed reversibly without the readditon of CBPQT⁴⁺, a structural modification which marks an improvement over the previous design.

7. An artificial molecular actuator

The redox-controlled mechanical motions of bistable [2]rotaxanes enables them to behave as functional nanoscale actuators as these mechanical motions can be harnessed to perform work. Specifically, doubly bistable [3]rotaxanes—comprised of two CBPQT⁴⁺ rings encircling a palindromic dumbbell containing two TTF and two naphthalene (NP) recognition units³³—have been designed to function as linear artificial molecular muscles (Fig. 8).³⁴ SAMs of these particular motor-molecules have been formed on Au-coated cantilevers via disulfide tethers covalently attached to the CBPQT⁴⁺ rings. A four-electron oxidation of the [3]rotaxane causes the CBPQT⁴⁺ rings to move from the peripheral TTF recognition units, where the rings are as far apart as 4.2 nm, to the interior NP recognition units, where the rings are separated by only 1.4 nm. The ring contraction of 2.8 nm represents a mechanical strain of 67% along the palindromic dumbbell and is reminiscent of a linear motor or muscle. Approximately 10⁸ of these [3]rotaxane molecules were self-assembled onto a Au-coated 500 × 100 × 1 µm microcantilever array using the disulfide tethers on the CPBQT⁴⁺ rings, and the device was placed in a transparent fluid cell for evaluation. A continuous flow of aqueous solutions of oxidant and reductant through the fluid cell caused the CBPQT⁴⁺ ring to move reversibly from the GSCC to the MSCC. This mechanical movement resulted in an upward deflection of ∼35 nm of the 500 × 100 × 1 µm cantilevers because of molecular actuations. A series of control compounds were synthesized to show that the addition of disulfide tethers to the mobile parts of the [3]rotaxanes—the CBPQT⁴⁺ rings—was necessary for the bending of the microcantilevers. Attachment of the same disulfide tethers to the stoppers of the dumbbell moiety resulted in no upward bending of the microcantilevers.

Fig. 8 (a) Molecular structure of a palindromic, doubly bistable [3]rotaxane linear motor-molecule where disulfide tethers have been attached covalently to the mobile CBPQT⁴⁺ rings to facilitate self-assembly onto gold. (b) Schematic representation of the reversible, redox-controlled bending of a gold-coated microcantilever by a [3]rotaxane linear motor (not to scale). Upon oxidation of the TTF units to TTF²⁺, the mechanical movement of the [3]rotaxanes' ring components is transferred to the microcantilever, inducing upward bending. Reduction of the TTF²⁺ units back to their neutral state releases the microcantilever as the tethered CBPQT⁴⁺ rings move to their ground state co-conformations.

Induction of micro-scale mechanical changes by cumulative nanoscale movements was achieved through the careful structural design of the palindromic doubly bistable [3]rotaxanes. The palindromic nature of the structure imparts a high degree of symmetry upon the molecule, a feature which assists in the synthesis and doubles the amount of actuation per molecule. A 1,4-diethynylbenzene central portion was incorporated into the molecule as a spacer between the two NP units in order to provided a large degree of structural rigidity to the [3]rotaxanes, thus increasing their effectiveness as linear motor-molecules. The weak Au–S bond allows for the formation of the most closely packed and thermodynamically stable monolayer.^18,24 The tightly packed SAMs maximise the number of molecules on the surface, and thus optimise the total force applied to the cantilever.

8. Light-harvesting molecular triad

The miniaturization of functional devices down to the nanoscale requires the development of nanoscale power supplies to drive them. Toward this aim, we have recently reported³⁵ the design, fabrication and characterisation of a nanoscale light-harvesting molecular triad, capable of generating ∼1.4 µA cm⁻² upon irradiation with 413 nm light. Building upon (1) previous studies of SAMs of [2]rotaxanes, (2) surface studies and (3) rational design based on previous molecular dyad and triad systems,³⁶ a photochemically active device was fabricated that powers a supramolecular machine.

Light harvesting molecular dyads and triads³⁶ have been utilized to convert light into electrical energy in order to transport ions across artificial lipid membranes. The structure of a typical molecular triad is a donor–chromophore–acceptor motif. Physical properties of molecular triads are governed by the electronic and photophysical properties of their constituent parts. For this reason, the structural design of molecular triads is critical to their performance in molecular device settings. Our design (Fig. 9) involves a C₆₀–porphyrin–TTF triad self-assembled on a Au surface. C₆₀ exhibits very low reorganisational energy and sustains long charge-separated state lifetimes as a result of its spherical, three-dimensional structure. Porphyrins are efficient chromophores that typically have a sharp, well-defined absorption maximum (λ_max). Electron donors based on TTF have displayed microsecond-long charge-separated states in molecular triad systems, resulting from the low oxidation potential of TTF derivatives. Finally, a physisorbed disulfide tether was used to form a thermodynamically stable molecular monolayer of triads on a Au surface. CVs of the molecular triad on Au revealed oxidation and reduction peaks that were a linear combination of TTF, C₆₀ and the meso-porphyrin. In addition, the UV absorption spectrum of the molecular triad in solution (CH₂Cl₂) was also a linear combination of the UV–vis absorption spectra of its free components. It is also important to note that, of the triad's nine absorption peaks in the UV, the peak at 420 nm corresponding to the λ_max of the porphyrin, is both more intense (by 3–50 times) than the other eight peaks and well isolated from them. This λ_max of the triad can be exploited to photoexcite the porphyrin selectively to its singlet-excited state, such that it will then behave as an electron donor. These experiments indicated that there are no electronic interactions between the donor–chromophore–acceptor units in the ground state and highlighted the modularity of the triad’s design: the different donor, chromophore, and acceptor units may be modified and fine-tuned according to the properties of the constituent parts with experimental structure–property feedback loops.

Fig. 9 Schematic representation of how a light harvesting molecular triad can be used to power a supramolecular machine. Photoinduced electron transfer from the porphyrin unit (Porph.) to C₆₀ is achieved upon irradiation with 413 nm light. A charge shift from TTF to the porphyrin establishes a charge-separated state. The TTF⁺˙ is neutralized by the transfer of an electron from the Au electrode. Consequently, an electron is transferred from C₆₀ to the [2]pseudorotaxane, which reduces the CBPQT⁴⁺ to its radical anion and induces dethreading. Electron transfer from CBPQT³⁺˙ to the Pt counter electrode leads to threading of the [2]pseudorotaxane and completes the electrochemical circuit.

A photoelectrochemical cell was fabricated from a Pt counter electrode, a Ag reference electrode, a 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF₆) electrolyte solution, and a SAM of the triad molecules on a Au working electrode in MeCN. Irradiation at 413 nm with a Kr ion laser resulted in the production of ∼1.4 µA cm⁻² of current at zero applied voltage with an efficiency of 1%. A linear relationship between the magnitude of photocurrent and power of the laser, indicated that a larger input of photons generates proportionately a greater output of electrons. The mechanism of photocurrent generation involves photoexcitation of the porphyrin to its singlet-excited state, which then transfers an electron to the C₆₀ acceptor. Subsequent electron transfer from the donor TTF to porphyrin establishes the donor˙⁺–chromophore–acceptor˙⁻ charge-separated state. Electronic reset is achieved by a unidirectional electron flow from the Au working electrode, through the triad, to the Pt counter electrode (Fig. 9) favoured by long charge-separated states or, alternately, by back-electron-transfer.

The photoelectrochemical cell was used to power a nanoscale supramolecular machine consisting of a CBPQT⁴⁺ host and a DNP–DEG guest. Strong complexation between CBPQT⁴⁺ and DEG–DNP results^1a in the formation of a [2]pseudorotaxane in solution. Reduction of CBPQT⁴⁺ causes dethreading of the pseudorotaxane and, as in the case of the supramolecular nanovalve, the process can be monitored by an increase in the fluorescence intensity of DEG–DNP. The radical anion of C₆₀, generated in the charge-separated state of the molecular triad, has a reduction potential of E_red = −500 mV, making it capable of transferring an electron to CBPQT⁴⁺, which has a reduction potential of E_red = −300 mV. Thus, a photoelectrochemical cell consisting of a Au electrode functionalised with triad molecules and an electrolyte solution containing the CBPQT⊂DEG–DNP pseudorotaxanes, constitutes a light-driven supramolecular machine. Irradiation of the photoelectrochemical cell at 413 nm generated a current of 1.1 µA that was sustained for a period of ∼50 min and resulted in a gradual increase in the DEG–DNP fluorescence intensity even when E_applied = 0 V. The number of electrons generated by an average current of 1.1 µA over a period of 50 min can be calculated to be 3.3 × 10⁻⁸ mol of electrons which, if 100% efficient, would be capable of reducing 7% of the [2]pseudorotaxanes in solution. The experimentally observed increase in DEG–DNP fluorescence was 6.7%, which is in excellent agreement with the theoretical maximum. These results demonstrate how a surface-mounted light harvesting molecular triad may be used as a nanoscale power supply for the purpose of driving a supramolecular machine. The modular synthesis and observation that the triad's physical properties are a linear combination of its constituent parts, even when mounted on a surface, shows the immense potential of how structure–function feedback loops can be used to optimise device performance.

9. Conclusions

In summary, a series of binary molecular machines—nanoswitches, nanovalves, motor-molecules, and light-harvesting triads—have been constructed using the donor–acceptor recognition motif. The modular, template-directed synthesis of bistable [2]rotaxanes and [2]catenanes, as well as [2]pseudorotaxanes, aid and abet the construction of functional molecules and supermolecules for a range of different applications. Device properties can be optimised by knowing the effects of structure and superstructure on surface and interfacial functions and using an iterative feedback loop to improve device design.

Acknowledgements

We are grateful to the National Science Foundation for financial support through a GK-12 program grant DGE 02-31988 fellowship to A. B. B. and an IGERT (MCTP)-DGE0114443 fellowship to B. H. N.

References

1. (a) P.-L. Anelli, P. R. Ashton, N. Spencer, A. M. Z. Slawin, J. F. Stoddart and D. J. Williams, Angew. Chem., Int. Ed. Engl., 1991, 30, 1036; (b) 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., 1993, 32, 1301; (c) P. R. Ashton, P. T. Glink, M.-V. Martinez-Díaz, J. F. Stoddart, A. J. P. White and D. J. Williams, Angew. Chem., Int. Ed. Engl., 1996, 35, 1930.
2. (a) S. J. Loeb and J. A. Wisner, Angew. Chem., Int. Ed., 1998, 37, 2838; (b) G. Doddi, G. Ercolani, S. Franconeri and P. Mencarelli, J. Org. Chem., 2001, 66, 4950; (c) J. A. Wisner, P. D. Beer and M. G. B. Drew, Angew. Chem., Int. Ed., 2001, 40, 3606; (d) J. A. Wisner, P. D. Beer, N. G. Berry and B. Tomapatanget, Proc. Natl. Acad. Sci. USA, 2002, 99, 4983; (e) H.-J. Kim, W. S. Heon, Y. H. Ko and K. Kim, Proc. Natl. Acad. Sci. USA, 2002, 99, 5007; (f) H.-S. Kim and J.-I. Hong, Chem. Lett., 2002, 336; (g) L. Avram and Y. Cohen, J. Org. Chem., 2002, 67, 2639; (h) D. Tuncel and J. H. G. Steinke, Chem. Commun., 2002, 496; (i) J. Li, X. Ni and K. Leong, Angew. Chem., Int. Ed., 2003, 42, 69; (j) A. Arduini, F. Calzavacca, A. Pochini and A. Secchi, Chem.—Eur. J., 2003, 9, 793; (k) K. Miyake, S. Yasuda, A. Harada, J. Sumaoka, M. Komiyama and H. Shigekawa, J. Am. Chem. Soc., 2003, 125, 5080; (l) S. Kawano, N. Fujita and S. Shinkai, Chem. Commun., 2003, 1352; (m) K.-S. Leong, K.-J. Chang and Y. J. An, Chem. Commun., 2003, 1450; (n) H. Ritter, O. Sadowski and E. Tepper, Angew. Chem., Int. Ed., 2003, 42, 3171; (o) M. Clemente-León, F. Marchioni, S. Silvi and A. Credi, Synth. Met., 2003, 139, 773; (p) E. Coronado, A. Forment-Aliaga, P. Gaviña and F. M. Romero, Inorg. Chem., 2003, 42, 6959; (q) S.-H. Chiu, K.-S. Liao and J.-K. Su, Tetrahedron Lett., 2004, 45, 213; (r) K. Moon and A. E. Kaifer, Org. Lett., 2004, 6, 185.
3. (a) M. Asakawa, P. R. Ashton, C. L. Brown, M. C. T. Fyfe, S. Menzer, D. Pasini, C. Scheuer, N. Spencer, J. F. Stoddart, A. J. P. White and D. J. Williams, Chem.—Eur. J., 1997, 3, 1136; (b) 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., 2000, 6, 3558; (c) J. O. Jeppesen, K. A. Nielsen, J. Perkins, S. A. Vignon, A. Di Fabio, R. Ballardini, M. T. Gandolfi, M. Venturi, V. Balzani, J. Becher and J. F. Stoddart, Chem.—Eur. J., 2003, 9, 2982; (d) J. O. Jeppesen, S. A. Vignon and J. F. Stoddart, Chem.—Eur. J., 2003, 9, 4611; (e) H.-R. Tseng, S. A. Vignon, P. C. Celestre, J. Perkins, J. O. Jeppesen, A. Di Fabio, R. Ballardini, M. T. Gandolfi, M. Venturi, V. Balzani and J. F. Stoddart, Chem.—Eur. J., 2004, 10, 155; (f) B. W. Laursen, S. Nygaard, J. O. Jeppesen and J. F. Stoddart, Org. Lett., 2004, 6, 4167; (g) T. Iijima, S. A. Vignon, H.-R. Tseng, T. Jarroson, J. K. M. Sanders, F. Marchoni, M. Venture, E. Apstoli, V. Balzani and J. F. Stoddart, Chem.—Eur. J., 2004, 10, 6375; (h) J. O. Jeppesen, S. Nygaard, S. A. Vignon and J. F. Stoddart, Eur. J. Org. Chem., 2005, 196.
4. (a) G. Schill, Catenanes, Rotaxanes and Knots, Academic Press, New York, 1971; (b) D. M. Walba, Tetrahedon, 1985, 41, 3161; (c) C. O. Dietrich-Buchecker and J.-P. Sauvage, Chem. Rev., 1987, 87, 795; (d) H. W. Gibson and H. Marand, Adv. Mater., 1993, 5, 11; (e) A. Livoreil, C. O. Dietrich-Buchecker and J.-P. Sauvage, J. Am. Chem. Soc., 1994, 116, 9399; (f) D. B. Amabilino and J. F. Stoddart, Chem. Rev., 1995, 95, 2725; (g) R. Jäger and F. Vögtle, Angew. Chem., Int. Ed. Engl., 1997, 36, 930; (h) G. A. Breault, C. A. Hunter and P. C. Mayers, Tetrahedron, 1999, 55, 5265; (i) Molecular Catenanes, Rotaxanes and Knots, ed. J.-P. Sauvage and C. O. Dietrich-Buchecker, Wiley-VCH, Weinheim, 1999; (j) L. Raehm, D. G. Hamilton and J. K. M. Sanders, SYNLETT, 2002, 1743.
5. M. Asakawa, P. R. Ashton, V. Balzani, A. Credi, C. Hamers, G. Mattersteig, M. Montalti, A. N. Shipway, N. Spencer, J. F. Stoddart, M. S. Tolley, M. Venturi, A. J. P. White and D. J. Williams, Angew. Chem., Int. Ed., 1998, 37, 333.
6. (a) J. F. Stoddart, Chem. Aust., 1992, 59, 576; (b) M. Gómez-López, J. A. Preece and J. F. Stoddart, Nanotechnology, 1996, 7, 183; (c) V. Balzani, M. Gómez-López and J. F. Stoddart, Acc. Chem. Res., 1998, 31, 405; (d) V. Balzani, A. Credi, F. M. Raymo and J. F. Stoddart, Angew. Chem., Int. Ed., 2000, 39, 3348; (e) F. M. Raymo and J. F. Stoddart, in Supramolecular Organization and Materials Design, ed. W. Jones and C. N. R. Rao, Cambridge University Press, Cambridge, 2002, pp. 332–362; (f) V. Balzani, A. Credi and M. Venturi, Molecular Devices and Machines—A Journey into the Nano World, Wiley-VCH, Weinheim, 2003; (g) P. M. Mendes, A. H. Flood and J. F. Stoddart, Appl. Phys. A: Solid Surf., 2005, 80, 1197; (h) A. H. Flood, R. J. A. Ramirez, W.-Q. Deng, R. P. Muller, W. A. Goddard, III and J. F. Stoddart, Aust. J. Chem., 2004, 57, 301.
7. (a) C. Dietrich-Buchecker, M. C. Jiménez-Molero, V. Sartor and J.-P. Sauvage, Pure Appl. Chem., 2003, 75, 1383; (b) M. Barboiu and J.-M. Lehn, Proc. Natl. Acad. Sci. USA, 2002, 99, 5201; (c) R. A. van Delden, M. K. J. ter Wiel, H. de Jong, A. Meetsma and B. L. Feringa, Org. Biomol. Chem., 2004, 2, 1531; (d) J. V. Hernandez, E. R. Kay and D. A. Leigh, Science, 2004, 306, 1532.
8. A. R. Bernardo, J. F. Stoddart and A. E. Kaifer, J. Am. Chem. Soc., 1992, 114, 10624.
9. R. Castro, K. R. Nixon, J. D. Evanseck and A. E. Kaifer, J. Org. Chem., 1996, 61, 7298.
10. M. Asakawa, W. Dehaen, G. L'abbé, S. Menzer, J. Nouwen, F. M. Raymo, J. F. Stoddart and D. J. Williams, J. Org. Chem., 1996, 61, 9591.
11. (a) D. W. Steuerman, H.-R. Tseng, A. J. Peters, A. H. Flood, J. O. Jeppesen, K. A. Nielsen, J. F. Stoddart and J. R. Heath, Angew. Chem., Int. Ed., 2004, 43, 6486; (b) A. H. Flood, A. J. Peters, S. A. Vignon, D. W. Steuerman, H.-R. Tseng, S. Kang, J. R. Heath and J. F. Stoddart, Chem.—Eur. J., 2004, 10, 6558; (c) J.-W. Choi, A. H. Flood, D. W. Steuerman, S. Nygaard, A. B. Braunschweig, N. N. P. Moonen, B. W. Laursen, Y. Luo, E. DeIonno, A. J. Peters, J. O. Jeppesen, K. Xu, J. F. Stoddart and J. R. Heath, Chem.—Eur. J., 2005 DOI:10.1002/chem.200500934.
12. (a) B. Odell, M. V. Reddington, A. M. Z. Slawin, N. Spencer, J. F. Stoddart and D. J. Williams, Angew. Chem., Int. Ed. Engl., 1988, 27, 1547; (b) P. R. Ashton, B. Odell, M. V. Reddington, A. M. Z. Slawin, J. F. Stoddart and D. J. Williams, Angew. Chem., Int. Ed. Engl., 1988, 27, 1550; (c) G. Doddi, G. Ercolani, S. Franconeri and P. Mencarelli, J. Org. Chem., 2001, 66, 4950; (d) G. Ercolani and P. Mencarelli, J. Org. Chem., 2003, 68, 6470; (e) C. D'Acerno, G. Doddi, G. Ercolani, S. Franconeri, P. Mencarelli and A. Piermattei, J. Org. Chem., 2004, 69, 1393.
13. (a) I. T. Harrison, J. Chem. Soc., Chem. Commun., 1972, 231; (b) G. Schill, W. Beckmann, N. Schweikert and H. Fritz, Chem. Ber., 1986, 119, 2647; (c) P. R. Ashton, M. Belohradsky, D. Philp and J. F. Stoddart, J. Chem. Soc., Chem. Commun., 1993, 1269; (d) M. Händel, M. Plevoets, S. Gestermann and F. Vögtle, Angew. Chem., Int. Ed. Engl., 1997, 36, 1199; (e) F. M. Raymo, K. N. Houk and J. F. Stoddart, J. Am. Chem. Soc., 1998, 120, 9318.
14. (a) S. J. Rowan, S. J. Cantrill and J. F. Stoddart, Org. Lett., 1999, 1, 129; (b) S. J. Rowan and J. F. Stoddart, Org. Lett., 1999, 1, 1913; (c) S. J. Rowan, S. J. Cantrill, J. F. Stoddart, A. J. P. White and D. J. Williams, Org. Lett., 2000, 2, 759.
15. (a) E. W. Wong, C. P. Collier, M. Belohradsky, F. M. Raymo, J. F. Stoddart and J. R. Heath, J. Am. Chem. Soc., 2000, 122, 5831; (b) C. P. Collier, J. O. Jeppesen, Y. Luo, J. Perkins, E. W. Wong, J. R. Heath and J. F. Stoddart, J. Am. Chem. Soc., 2001, 123, 12632; (c) D. R. Stewart, D. A. A. Ohlberg, P. A. Beck, Y. Chen, R. S. Williams, J. O. Jeppesen, K. A. Nielsen and J. F. Stoddart, Nano Lett., 2004, 4, 133.
16. S. Chia, J. Cao, J. F. Stoddart and J. I. Zink, Angew. Chem., Int. Ed., 2001, 40, 2447.
17. (a) R. Hernandez, H.-R. Tseng, J. W. Wong, J. F. Stoddart and J. I. Zink, J. Am. Chem. Soc., 2004, 126, 3370; (b) T. D. Nguyen, H.-R. Tseng, P. C. Celestre, A. H. Flood, Y. Liu, J. F. Stoddart and J. I. Zink, Proc. Natl. Acad. Sci. USA, 2005, 102, 10029.
18. (a) H.-R. Tseng, D. Wu, N. X. Fang and J. F. Stoddart, ChemPhysChem, 2004, 5, 111; (b) E. Katz, O. Lioubashevsky and I. Willner, J. Am. Chem. Soc., 2004, 126, 15520.
19. (a) A. Y. Fadeev and T. J. McCarthy, J. Am. Chem. Soc., 1999, 121, 12184; (b) R. Helmy and A. Y. Fadeev, Langmuir, 2002, 18, 8924; (c) H. Ding, M. K. Ram and C. Nicolini, J. Mater. Chem., 2002, 12, 3585; (d) B. Long, K. Nikitin and D. Fitzmaurice, J. Am. Chem. Soc., 2003, 125, 5152; (e) B. Long, K. Nikitin and D. Fitzmaurice, J. Am. Chem. Soc., 2003, 125, 15490; (f) M. P. Danahy, M. J. Avaltroni, K. S. Midwood, J. E. Schwarzbauer and J. Schwatrz, Langmuir, 2004, 20, 5333; (g) G. K. R. Senadeera, T. Kitamura, Y. Wada and S. Yanagida, J. Photochem. Photobiol., A, 2004, 164, 61.
20. (a) M. R. Diehl, D. W. Steuerman, H.-R. Tseng, S. A. Vignon, A. Star, P. C. Celestre, J. F. Stoddart and J. R. Heath, ChemPhysChem, 2003, 4, 1335; (b) A. Star, Y. Liu, K. Grant, L. Ridvan, J. F. Stoddart, D. W. Steuerman, M. R. Diehl, A. Boukai and J. R. Heath, Macromolecules, 2003, 36, 553.
21. E. Shustorovich, Acc. Chem. Res., 1988, 21, 183.
22. A. Ulman, An Introduction to Ultrathin Organic Films, Academic Press, San Diego, 1991.
23. R. J. Chen, Y. Zhang, D. Wang and H. Dai, J. Am. Chem. Soc., 2001, 123, 3838.
24. (a) E. Delamarche, B. Michel, H. A. Biebuyck and C. Gerber, Adv. Mater., 1996, 8, 719; (b) D. Witt, R. Klajn, P. Barski and B. A. Gryzbowshi, Curr. Org. Chem., 2004, 8, 1763; (c) J. C. Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo and G. M. Whitesides, Chem. Rev., 2005, 105, 1103.
25. (a) R. F. Service, Science, 2001, 294, 2442; (b) R. F. Service, Science, 2003, 302, 556; (c) J. R. Heath and J. F. Stoddart, Science, 2003, 303, 1136.
26. (a) A. R. Pease, J. O. Jeppesen, J. F. Stoddart, Y. Luo, C. P. Collier and J. R. Heath, Acc. Chem. Res., 2001, 34, 433; (b) Y. Luo, C. P. Collier, J. O. Jeppesen, K. A. Nielsen, E. DeIonno, G. Ho, J. Perkins, H.-R. Tseng, T. Yamamoto, J. F. Stoddart and J. R. Heath, ChemPhysChem, 2002, 3, 519; (c) A. H. Flood, J. F. Stoddart, D. W. Steuerman and J. R. Heath, Science, 2004, 306, 2055.
27. I. C. Lee, C. W. Frank, T. Yamamoto, H.-R. Tseng, A. H. Flood, J. F. Stoddart and J. O. Jeppesen, Langmuir, 2004, 20, 5809.
28. (a) S. S. Jang, Y. H. Jang, Y.-H. Kim, W. A. Goddard, III, A. H. Flood, B. W. Laursen, H.-R. Tseng, J. F. Stoddart, J. O. Jeppesen, J. W. Choi, D. W. Steuerman, E. DeIonno and J. R. Heath, J. Am. Chem. Soc., 2005, 127, 1563; (b) Y. H. Jang, S. S. Jang and W. A. Goddard, III, J. Am. Chem. Soc., 2005, 127, 4959.
29. T. J. Huang, H.-R. Tseng, L. Sha, W. Lu, B. Brough, A. H. Flood, B.-D. Yu, P. C. Celestre, J. P. Chang, J. F. Stoddart and C.-M. Ho, Nano Lett., 2004, 4, 2065.
30. K. Nørgaard, B. W. Laursen, S. Nygaard, K. Kjaer, H.-R. Tseng, A. H. Flood, J. F. Stoddart and T. Bjørnholm, Angew. Chem., Int. Ed., 2005, 44, 1.
31. (a) N. K. Mal, M. Fujiwara and Y. Tanaka, Nature, 2003, 421, 350; (b) N. K. Mal, M. Fujiwara, Y. Tanaka, T. Taguchi and M. Matsukata, Chem. Mater., 2003, 15, 3385; (c) N. Lui, D. R. Dunphy, P. Atanassov, S. D. Bunge, Z. Chen, G. P. López, T. J. Boyle and J. C. Brinker, Nano Lett., 2004, 4, 551.
32. I. Haller, J. Am. Chem. Soc., 1978, 100, 8050.
33. (a) T. J. Huang, B. Brough, C.-M. Ho, Y. Liu, A. H. Flood, P. A. Bonvallet, H.-R. Tseng, J. F. Stoddart, M. Baller and S. Magonov, Appl. Phys. Lett., 2004, 85, 5391; (b) Y. Liu, A. H. Flood, P. A. Bonvallet, S. A. Vignon, B. H. Northrop, H.-R. Tseng, J. O. Jeppesen, T. J. Huang, B. Brough, M. Baller, S. Magonov, S. D. Solares, W. A. Goddard, III, C.-M. Ho and J. F. Stoddart, J. Am. Chem. Soc., 2005, 127, 9745.
34. (a) T. F. Otero, H. Grande and J. Rodríguez, J. Phys. Org. Chem., 1996, 9, 381; (b) M. C. Jiménez-Molero, C. Dietrich-Buchecker and J.-P. Sauvage, Angew. Chem., Int. Ed., 2000, 39, 3284; (c) J.-P. Collin, C. Dietrich-Buchecker, P. Gaviña, M. C. Jiménez-Molero and J.-P. Sauvage, Acc. Chem. Res., 2001, 34, 477.
35. S. Saha, E. Johansson, A. H. Flood, H.-R. Tseng, J. I. Zink and J. F. Stoddart, Chem.—Eur. J., 2005 DOI:10.1002/chem.200500371.
36. (a) G. Steinberg-Yfrach, P. A. Liddell, S.-C. Hung, A. L. Moore, D. Gust and T. A. Moore, Nature, 1997, 385, 239; (b) D. Gust, T. A. Moore and A. L. Moore, Acc. Chem. Res., 2001, 34, 40; (c) S. D. Straight, J. Andréasson, G. Kodis, A. L. Moore, T. A. Moore and D. Gust, J. Am. Chem. Soc., 2005, 127, 2717; (d) S. D. Straight, J. Andréasson, G. Kodis, S. Bandyopadhyay, R. H. Mitchell, T. A. Moore, A. L. Moore and D. Gust, J. Am. Chem. Soc., 2005, 127, 9403.

---