www.rsc.org/nanoscale
Volume 5 | Number 16 | 21 August 2013 | Pages 7077–7640
ISSN 2040-3364
FEATURE ARTICLE de Juan and Pérez Getting tubed: mechanical bond in endohedral derivatives of carbon nanotubes?
2040-3364(2013)5:16;1-9
Nanoscale FEATURE ARTICLE
Cite this: Nanoscale, 2013, 5, 7141
Getting tubed: mechanical bond in endohedral derivatives of carbon nanotubes?† ´rez* Alberto de Juan and Emilio M. Pe
Received 4th April 2013 Accepted 28th April 2013
We present a brief overview of some of the most prominent examples of encapsulation of molecules inside carbon nanotubes. We then relate them to mechanically interlocked molecules, and in particular rotaxanes, by examining the most prominent features of the mechanical bond (topology, dynamic properties, and
DOI: 10.1039/c3nr01683h
stability) and comparing them to those of endohedral derivatives of nanotubes. Our analysis shows that
www.rsc.org/nanoscale
there is a surprisingly clear link between these two apparently disparate species.
Introduction A great deal of research into carbon nanotubes has concentrated on their extraordinary physical properties.1 As a result of these investigations, it has been established that carbon nanotubes show extremely high tensile strength and thermal conductivity.2 Metallic carbon nanotubes have been shown to transport electric current ballistically, without dissipation of heat, and, uniquely, without the need for any dopant.3
IMDEA Nanoscience, C/Faraday 9, Ciudad Universitaria de Cantoblanco, 28049, Madrid, Spain. E-mail:
[email protected]; Fax: +34 912998730; Tel: +34 912998852
Semiconducting nanotubes have been utilized as the active semiconductor in a variety of nanoscale devices.4 In order to exploit their amazing properties, it became immediately apparent that it was necessary to pretreat nanotubes so that they could be manipulated and investigated by standard analytical methods. This was the main driving force behind the rst insights into the functionalization of single wall carbon nanotubes (SWNTs), and in fact, their chemical manipulation has been shown to improve solubility and processability.5,6 Intensive research has been carried out on both the covalent7 and noncovalent modications of the outer surface of SWNTs,8 but to date the toolbox of reactions applicable to SWNTs is not as abundant as would be desirable. In this sense, a hot topic is the endohedral modication of SWNTs, that is, the
† Dedicated to Prof. David A. Leigh with the occasion of his 50th birthday.
Alberto de Juan Garrudo (right) was born in Madrid, Spain, in 1988. He obtained his degree in chemistry in 2011 at Universidad Complutense de Madrid. He started his research in organic chemistry at the same University with an IMDEA Nanociencia studentship. In 2012 he started his PhD studies with Emilio M. P´erez in IMDEA-Nanociencia. His research focuses on the noncovalent modication of carbon nanotubes.Emilio M. P´erez (le)
This journal is ª The Royal Society of Chemistry 2013
obtained his BSc (2000) and MSc (2001) from the Universidad de Salamanca (Spain). He then joined the group of Prof. David A. Leigh at the University of Edinburgh (UK) where he obtained his PhD in 2005. His PhD work was recognized with the 1st Prize at the 2004 Society of Chemical Industry Symposium on Novel Organic Chemistry and the 2006 IUPAC Prize for Young Chemists. He joined the group of Prof. Nazario Mart´ın at the Universidad Complutense de Madrid in 2005 as a Juan de la Cierva postdoctoral fellow. In December 2008 he joined IMDEA Nanoscience as a Ram´ on y Cajal researcher. During his stay in Madrid, he received the 2009 Real Sociedad Espa~ nola de Qu´ımica Prize for Novel Researchers and the 2010 Universidad Complutense de Madrid Foundation Prize for Science and Technology. In 2011– 2012 he received support from both Spanish (MINECO) and European sources (ETSF, ERC Starting Independent Research Grant) to establish his own research group at IMDEA. His main research interests concern the development of unconventional methods for the modication of carbon nanotubes, molecular recognition, the self-assembly of functional materials and the construction of molecular machinery.
Nanoscale, 2013, 5, 7141–7148 | 7141
Nanoscale introduction of molecular entities inside the cavity of the nanotubes. In this feature article we present some of the most prominent examples of endohedral modication of carbon nanotubes, which will serve to reect on the nature of the interaction between the guest molecule(s) and the SWNT container, and indicate its relationship to mechanically interlocked molecules in general and rotaxanes in particular.
Fullerene “peapods” The lling of multiwall carbon nanotubes (MWNTs) was realized as early as 1993,9 only two years aer the discovery of carbon nanotubes.10 Meanwhile, the debut of SWNTs as nanocontainers was the accidental observation of a row of C60 molecules inside as-synthesized nanotubes by Luzzi and coworkers,11 and the intentional inclusion of Ru crystals reported by Sloan et al., both in 1998.12 The nanotubes were synthesized by pulsed laser vaporization, a method which produces fullerenes as side products. Although most fullerenes were removed during the purication procedure, some of the nanotubes with adequate diameters (1.3–1.4 nm) contained C60 molecules, as evidenced by high resolution transmission electron microscopy (HRTEM, Fig. 1). Two years later, the same group reported a method to improve the yield of “peapods”. They rst oxidized the sample containing empty SWNTs and fullerenes with HNO3 to create structural defects in the carbon nanotubes, particularly at the ends, and then annealed the mixture at high temperatures (100–1200 C) and reduced pressure (20–40 mPa) to allow the fullerenes to diffuse into the SWNT cavity.13 The encapsulation of C60 inside a (10,10) SWNT has been calculated to be exergonic, with an energy gain of 0.51 eV (11.8 kcal mol1).14 This relatively high noncovalent interaction arises from the adequate shape and size complementarity between the convex fullerene
Fig. 1 Top: HRTEM image of a fullerene “peapod”. The scale bar represents 2.0 nm. Reproduced from ref. 11 with permission from Nature Publishing Group. Bottom: molecular model (MM+) showing the structure of a C60 peapod. A (10,10) SWNT has been arbitrarily chosen, considering its diameter (1.36 nm).
7142 | Nanoscale, 2013, 5, 7141–7148
Feature Article guest and the concave cavity of the SWNT, which maximizes dispersion forces.15 As a consequence, C60 can also be included into SWNTs from solution, at temperatures as low as 70 C.16 Initially, most C60 molecules form linear arrays with a fullerene–fullerene spacing of approximately 0.3 nm, equal to the interplanar van der Waals distance in graphite. Nevertheless, some pairs of molecules appear to have coalesced, as those marked with a red circle in Fig. 1. By either irradiating the sample with electrons under TEM conditions or thermal stimulation, the fullerene “peas” can be forced to merge, to form corrugated stable nanotubules of 0.5–0.7 nm diameter inside the SWNT.17 Remarkably, both the translational motion along the SWNT and a corkscrew-like rotational motion of these unique fused fullerene chains have been observed in real time through HRTEM.18 Other types of fullerenes, including C70, azafullerenes, and endohedral fullerenes have also been included into SWNTs.19,20 A particularly interesting consequence of the positive interaction between fullerenes and SWNTs is that the former can be utilized to “tow” other moieties that would not be spontaneously encapsulated into the cavity of the nanotube.21 For example, a C60-derived fulleropyrrolidine bearing an N-oxide (1 in Fig. 2) has been introduced into SWNTs with the help of supercritical CO2.22 With this paramagnetic probe, the peapod structure could be investigated through electron paramagnetic resonance (EPR). The anisotropic rotational freedom of the radical is limited when inside the SWNTs, but its EPR spectral characteristics remain intact. This observation, together with Raman spectroscopy, demonstrated that it is only the fullerene cage that interacts strongly with the SWNT cavity. Following the same strategy, transition metal complexes have also been encapsulated inside SWNTs (2 in Fig. 2).23 This is particularly relevant as the electronic properties of the SWNT can be modulated through interaction with the different oxidation states of the metal centres. Nakamura and co-workers relied on C60 to anchor simple alkenyl (3 in Fig. 2) and alkyl (4 in Fig. 2) chains inside SWNTs.24 In this case, the objective was to observe the conformational and translational motion of the hydrocarbon chains inside the SWNT by single-molecule real-time TEM (SMRT-TEM).25 Surprisingly, the chain in 3 adopts only two conformations, changing between them over a 0.5 s exposure time at 293 K (Fig. 3a and b). Besides this conformational motion, the alkenyl chain of 3 was seen
Fig. 2 Chemical structures of the fullerene derivatives introduced into SWNTs by Campestrini et al.22 (1), Fan et al.23 (2) and Koshino et al.24 (3 and 4).
This journal is ª The Royal Society of Chemistry 2013
Feature Article
Nanoscale
Fig. 4
Fig. 3 (a and b) Conformational motion of the alkenyl chain of 3 inside a SWNT. SMRT-TEM images are shown on the left and the corresponding molecular models on the right. (c) Extrusion of the alkyl chain of 4 through a hole in the nanotube's wall. Reproduced from ref. 24 with permission from Nature Publishing Group.
protruding from the nanotube through a hole in its wall and returning back in. This extrusion from the SWNT was further investigated at 4 K in the case of the alkyl derivative 4 (Fig. 3c). The alkyl chain can also exit the nanotube through a structural defect by moving along its molecular axis (that is, in extended conformation). Remarkably, the speed of the molecular motion of 3 at 293 K and 4 at 4 K is comparable, and rather slow. The temperature independency of the movement of molecules conned inside SWNTs and under high vacuum conditions is not as counterintuitive as it initially seems, considering the absence of particles with which to collide.25 In all previous examples, the fullerene “peas” were encapsulated in SWNT “pods” of adequate diameter for their tight encapsulation. Fullerenes can also be included into SWNTs or DWNTs of larger diameter, where they form unique crystalline phases,26 and their motion27 and coalescence to form nanotubules28 can be controlled with the electron beam of the TEM. Further exohedral functionalization is still possible once the SWNT has been endohedrally functionalized with fullerenes. For example, C60@SWNT peapods have been functionalized with aryldiazonium salts generated in situ under microwave irradiation.29 With a similar initial step, Imahori and co-workers have decorated C60 peapods with porphyrins, and demonstrated that electron transfer from the porphyrin donor to the C60@SWNT acceptor occurs upon photoirradiation, a process that does not take place in the equivalent compound with an empty SWNT.30 The complexity of the fullerene peapod systems makes it difficult to investigate the inclusion process in detail. The synthesis of discrete SWNT fragments31 reduces the problem to the association of a fullerene guest by a macrocyclic host.32 The shortest possible C60 peapod was recently reported by Yamago and coworkers, featuring [10]cycloparaphenylene ([10]CPP, 5 in Fig. 4) as a SWNT fragment. The binding event was investigated through UV-vis and uorescence titrations, which afforded binding constants of log Ka ¼ 3.8 and 6.4 in This journal is ª The Royal Society of Chemistry 2013
Molecular structures of [10]CPP and [4]CC.34
1,2-dichlorobenzene and toluene, respectively, both at room temperature. This represents a stabilization energy of 9.1 kcal mol1 for the C60@[10]CPP complex in toluene. Remarkably, C60 is encapsulated selectively inside [10]CPP, and not in larger or smaller CPPs, demonstrating that an excellent match between the diameter of the cavity and that of the fullerene guest is crucial for association.33 More recently, Isobe et al. have described the association of C60 by [4]cyclochrysenylene ([4]CC, 6 in Fig. 4),34 a zigzag SWNT fragment featuring a more rigid cavity with sixteen aromatic rings.35 As a result the binding constant at room temperature of [4]CC towards C60 increases to an astonishing log Ka ¼ 9.6 and 11.5 in 1,2 dichlorobenzene and toluene, respectively. In energetic terms, that is equivalent to 13.1 and 15.7 kcal mol1. Note that these values are in good agreement with those calculated for the formation of C60@(10,10) SWNT “peapods”.14 The data for these model compounds provide a denitive explanation for the inclusion of C60 inside SWNTs, and the thermodynamic stability of the corresponding “peapods”.36
Endohedral functionalization with other organic molecules Although fullerenes are the most frequent llers for SWNTs, other organic molecules can also be included into their cavities, through a variety of synthetic methods.37,38 Takenobu et al. reported the p- and n-type doping of SWNTs by encapsulation of organic molecules of adequate electronic properties (7–14 in Fig. 5 and C60).39 This collection includes electron donors (top row in Fig. 5), acceptors (bottom row in Fig. 5 plus C60) and
Fig. 5 Chemical structures of the organic molecules inserted into SWNTs by Takenobu et al.39 C60 (not shown) was also included.
Nanoscale, 2013, 5, 7141–7148 | 7143
Nanoscale polycyclic aromatic hydrocarbons (PAHs, middle row in Fig. 5). The molecules were inserted into the SWNT by vapour diffusion, by heating a mixture of SWNTs with sufficient defects for the entrance of the guests and the corresponding molecule in a sealed glass tube at reduced pressure, just above the sublimation temperature of the organic molecule. From X-ray diffraction data, the authors calculated that one molecule was inserted for every 100–150 SWNT carbon atoms, depending on the guest molecule. The doping effect was investigated through absorption and Raman spectroscopy, and corroborated by constructing eld effect transistor (FET) devices. In agreement with their electronic properties, no doping occurs with PAHs, n-type doping was observed with the electron donors 7, 8 and 9, while the electron acceptors 13 and 14 produced p-type doping. For instance, 13@SWNT showed 0.021 holes and 9@SWNT 0.009 electrons per SWNT carbon atom. Interestingly, when C60, a renowned electron acceptor, was used as the guest molecule, the authors observed no signs of doping. Recently, Khlobystov and co-workers have described the synthesis of graphene nanoribbons terminated with sulfuratoms (S-GNRs, Fig. 6) from TTF@SWNTs, mixtures of (C60:TTF) @SWNTs and other sulfur-containing fullerene derivatives.40 To produce the GNRs, the authors either irradiated the guest molecules with the electron beam or heated the samples to high temperatures (250–1000 C). At the beginning of the TEM observations, sulfur atoms appear evenly spaced, at a distance of 0.29 nm, which corresponds to weak S–S bonding (bond order < 1, Fig. 6a and c). Upon further irradiation, however, they were observed at alternating distances of 0.21 and 0.34 nm, a fact which was explained through the formation of very stable dithiolium cations (Fig. 6b and d). The authors propose that, rather counterintuitively, the electron beam acts as a chemical oxidant, removing valence electrons from the S-GNR. The introduction of coronene (15 in Fig. 7a) into SWNTs has been investigated by several groups, with intriguingly different results. Okazaki et al. described the formation of columns of stacked coronene molecules inside SWNTs.41 The authors utilized the vapour diffusion method by heating a mixture of coronene and opened SWNTs at 450 C in a sealed tube for approximately 24 h. However, Talyzin and co-workers observed
Feature Article
Fig. 7 (a) Chemical structures of coronene (15), perylene (16), and diamantane4,9-dicarboxylic acid (17); (b) HRTEM of the diamond nanowire formed from 17@DWNTs; (c) simulated TEM image and model of the diamond nanowire@DWNT. Adapted from ref. 44 with permission from John Wiley & Sons.
the formation of GNRs from either coronene or perylene (16 in Fig. 7a), following a practically identical experimental procedure.42 This discrepancy can most likely be attributed to differences in the SWNT starting material,43 and/or inhomogeneity of the samples. This last example highlights the importance of using a variety of characterization techniques to probe functionalized SWNTs rather than relying solely on HRTEM imaging. Very recently, Shinohara and co-workers have reported the formation of diamond nanowires inside DWNTs.44 Firstly, the authors encapsulated diamantane-4,9-dicarboxylic acid (17 in Fig. 7a) into DWNTs by vapour diffusion. When 17@DWNTs were annealed at 600 C for 12 h under a ow of hydrogen, HRTEM images showed the formation of diamond-type nanowires inside the nanotubes (Fig. 7b). Many other organic and inorganic species including porphyrins, carboranes, metallocenes, polymeric iodine chains, metals and metal oxides, etc. have been included into SWNTs.38 For the purposes of this review, we have just presented a few selected examples that will help us illustrate the relationship between endohedral derivatives of SWNTs and mechanically interlocked molecules.
Mechanical bond in endohedral SWNTs? Fig. 6 (a) TEM images of S-GNRs at the beginning of the exposure to the e-beam and (b) after prolonged irradiation. (c) and (d) proposed the chemical structures of the S-GNRs shown in (a) and (b). Reproduced from ref. 40 with permission from Nature Publishing Group.
7144 | Nanoscale, 2013, 5, 7141–7148
We have seen that a great variety of molecular species can be encapsulated into SWNTs. In the following, we will compare the nature of the interaction between the guest and the nanotube host in X@SWNTs to mechanically interlocked species, and in particular rotaxanes.
This journal is ª The Royal Society of Chemistry 2013
Feature Article
Nanoscale
Fig. 8 Cartoons showing the main structural features of a [2]catenane and a [2] rotaxane. The numbers in brackets refer to the number of molecular components.
Mechanically interlocked molecules (MIMs) consist of two or more separate components which are not connected by chemical (i.e. covalent) bonds.45 Examples of MIMs are rotaxanes, where one or more macrocycles are trapped onto a linear component (thread) by bulky substituents at its ends (stoppers) that prevent dissociation, and catenanes, where two or more macrocycles are interlocked like links in a chain (Fig. 8). These structures are molecular entities, as each component is intrinsically linked to the other through a mechanical bond, which prevents dissociation without cleavage of one or more covalent bonds. They are, therefore, fundamentally different from classic supramolecular species, where an equilibrium between bound and unbound states exists. To investigate the connection between X@SWNTs and MIMs, we will compare the most distinctive characteristics of the mechanical bond, namely topology, dynamic properties, and stability, to those of endohedrally modied nanotubes.
Fig. 9 (a) A mug and a doughnut are topologically identical, since they can be transformed into each other through stretching and bending (and, in this case, sugar glazing). (b) A [2]catenane and the corresponding two macrocycles are topologically different, since tearing and gluing would be required to transform one into the other. (c) Endohedrally modified SWNTs and [2]rotaxanes share the same topology.
several knots,54 from the simplest trefoil knot described in 1989 by Dietrich-Buchecker and Sauvage55 to a ve-fold knot recently reported by Leigh and co-workers.56 Moreover, the topology of MIMs gives rise to their remarkable dynamic properties. On dynamics
On topology Topology is a branch of mathematics which investigates the most basic properties of space, such as connectivity. Objects that can be transformed into each other through continuous deformations, for instance, stretching and bending, without tearing or gluing, are topologically identical.46 As a consequence, topology is oen described as “rubber sheet geometry”. From this point of view, a doughnut and a coffee mug are topologically indistinguishable—note that the mug's handle is crucial, since it provides the hole in the doughnut. On the other hand, two macrocycles and the corresponding catenane have different topologies, for it would require breaking one of the macrocycles apart, threading it through the other and then gluing it back together to construct the catenane. With these concepts in mind, it is immediately clear that a [2]rotaxane and any given X@SWNT share the same topology, as long as we consider an isolated guest molecule inside the nanotube (Fig. 9).47 Particularly evident examples are those that involve 1-D guests, like the diamond nanowire reported by Shinohara44 or the GNRs described by Talyzin and co-workers.42,48 The fact that X@SWNTs and rotaxanes are topologically indistinguishable is signicant. Topology is one of the most distinctive features of MIMs, to the point that the challenge of synthesizing topologically complex molecules is one of the main driving forces behind research into MIMs.49–51 Some remarkable achievements in this eld include the synthesis of intricate structures such as Borromean rings,52 [8]catenanes,53 and
This journal is ª The Royal Society of Chemistry 2013
The particular topology of rotaxanes, where the components are not covalently but mechanically connected, allows their components to move relative to each other in unique ways. The thread serves as a “track” that restricts the degrees of freedom along which the macrocyclic component can move, carrying out the same function that microtubules do in the case of kinesins.57,58 There are two main kinds of large-amplitude motion in rotaxanes: “pirouetting”, in which one ring rotates around the thread, and “shuttling”, in which the macrocycle moves along the thread. These dynamic properties have made rotaxanes one of the most attractive candidates for the construction of synthetic molecular machinery.59 For instance, molecular shuttles—rotaxanes in which the macrocycle can be selectively displaced between two or more different parts of the thread60— have been proven to be able to generate mechanical work,61 and to store information.62 Although shuttling in rotaxanes is usually described as if the thread remained static while the macrocycle(s) move along it, it is obvious that this is just a convention, and one could just as easily dene it as the movement of the thread through the macrocycle. In the case of X@SWNTs, this would translate into the movement of the X guest through the SWNT host. We have already seen that this type of motion is indeed possible in endohedral SWNTs,27 but we will illustrate it further with two recent examples. In 2010, Lindsay, Nuckolls and co-workers described the translocation of single DNA strands through SWNTs.63 They
Nanoscale, 2013, 5, 7141–7148 | 7145
Nanoscale
Fig. 10 (a) SEM image of the nanofluidic device constructed by Nuckolls and coworkers.63 Red arrows show a 2 mm barrier before removal of the SWNTs exposed to plasma. (b) Optical micrograph of the input (IR) and output (OR) reservoirs. (c) Current flow through a SWNT. Approximately 20% of the devices pass unexpectedly large currents, some of which also allowed translocation of DNA strands (in red). Reproduced from ref. 63 with permission from the American Association for the Advancement of Science.
constructed a nanouidic device with two reservoirs connected through a single SWNT of diameter between 1 and 2 nm. With a mild plasma treatment, the authors opened the ends of the SWNTs and proved that only the devices that were connected through open SWNTs showed ion conductance, and in, in some cases, translocation of DNA strands (Fig. 10). More recently, Lindsay and co-workers have detected the passing of single molecules through SWNT monitoring them simultaneously by measurement of the ion current and singlemolecule uorescence spectroscopy.64 To this end, the authors utilized both positively (Rhodamine) and negatively (Alexa 546) charged uorophores. Besides translocation/shuttling, other types of movement of the molecules encapsulated into SWNTs have also been observed.25 These cases demonstrate that large-amplitude motion of one of the components of X@SWNTs with respect to the other can be obtained, in parallelism with rotaxane-based molecular shuttles. On stability Finally, the blueprint for the existence of a mechanical bond between several components is the stability of the nal assembly. For example, to separate the macrocycle and thread of a [2] rotaxane requires breaking at least one covalent bond. The kinetic stability of MIMs is the main difference between MIMs and classic supramolecular host–guest systems, where, no matter how large the association constant (i.e. the thermodynamic stability), there is always equilibrium between bound and unbound states. Whereas for catenanes this is a clear-cut situation,65 this is not the case for rotaxanes, where the distinction between interlocked species (rotaxanes) and supramolecular complexes (pseudorotaxanes) is oen more blurred.66 According to the IUPAC recommended nomenclature a rotaxane is a “molecular arrangement comprising at least one molecule with a linear section threaded through at least one macrocyclic part of another or the same molecule and having end-groups large enough to prevent dethreading” whereas a pseudorotaxane is a “rotaxane-like molecular assembly in which the threading component(s) has (have) ends small enough to permit threading or dethreading of the macrocyclic molecule(s)”.67 Unfortunately, the situation is not that simple. Schalley and co-workers demonstrated that several factors affect the kinetic stability of rotaxanes/pseudorotaxanes,
7146 | Nanoscale, 2013, 5, 7141–7148
Feature Article besides the size complementarity between stoppers and macrocycle. For example the shape and exibility of the rotaxanes' components, to the point that in some cases, larger stoppers can lead to a higher rate of disassembly. Noncovalent forces between the thread and the macrocycle can also help prevent de-threading. Finally, solvent effects also play a role.68 A reasonable criterion would be to choose the covalent bond as a benchmark to discriminate between rotaxanes (MIMs) and pseudorotaxanes (supramolecular complexes). In this way, rotaxanes would disassemble only aer the breaking of at least one covalent bond, to yield structures with different connectivity than their original components. On the other hand, pseudorotaxanes would be rotaxane-like species in which disassembly to yield their intact components is feasible, even if only under extreme conditions.69 Let us examine endohedrally functionalised SWNTs using these parameters. A typical procedure for the inclusion of guest molecules inside the cavity of SWNTs includes the following steps: (1) opening of the tubes, unless they are naturally opened as a result of the synthetic method; (2) lling of the tubes in either gas (vapour diffusion) or liquid states (molten or saturated solution of the guest) and (3) purication to remove guest molecules adsorbed on the sidewalls of the nanotube.38 A typical purication procedure consists of washing the tubes with a solvent in which the guest molecules are soluble, to wash away those not inside the nanotube, which implies that there is no exchange between encapsulated and non-encapsulated molecules. Moreover, the investigations that we have seen above indicate that X@SWNTs withstand harsh conditions including annealing at high temperatures and irradiation with the electron beam of a TEM without dissociation. Other highly energetic processes, such as the reaction between guest molecules, even remarkably stable ones like the fullerenes, are clearly favoured. Indeed, partial escape of the guest has only been observed when the covalent structure of the nanotube is defective.25 Although the only system for which we have experimental thermodynamic data, the fullerene peapods, shows a very strong noncovalent interaction between SWNT and C60 of up to 15.7 kcal mol1, the energy of a C–C single bond is much greater, at about 80 kcal mol1. Therefore, the extraordinary stability of X@SWNTs is most likely a consequence of a large kinetic barrier for escape of the guest molecules from the nanotubes, in part due to the large number of guests. In any case, the experimental observations indicate that the stability of Xn@SWNT is notably larger than most classical supramolecular complexes. In a conservative approximation, they can be considered remarkably stable pseudorotaxane-type molecules.
Conclusions and outlook We have reviewed some notable examples of endohedral derivatives of carbon nanotubes, with emphasis on the cases that have served to establish a connection between X@SWNTs and MIMs, by comparing their main characteristics. In particular, we have shown that the topology of X@SWNTs is identical to that of rotaxanes, since they both consist of a hollow component – tubular or circular, respectively – and another This journal is ª The Royal Society of Chemistry 2013
Feature Article element which resides inside it and from which it cannot be (easily) detached. This unique topology allows for the largeamplitude motion of the guest molecule(s) inside SWNTs, in analogy with the shuttling motion of macrocycles along the thread in rotaxane-based molecular shuttles. Finally, we have also analysed the stability of X@SWNTs, nding that experimental data point to a very remarkable kinetic stability, with dissociation energies much higher than most other supramolecular associates. The intriguing connection between these two apparently disparate species—even if conjectural at this stage—poses a great opportunity for research, given the remarkable properties of both SWNTs and MIMs.70 There are many questions to be explored: for instance, can fully interlocked species based on X@SWNTs be synthesized? Can their dynamic properties be exploited in the construction of synthetic molecular machinery? In this sense, we have already seen that large-amplitude intercomponent motion is possible in X@SWNTs, but can it be made unidirectional? Can it be ratcheted71,72 and used to construct Brownian motors?73 We have just embarked on the journey to address these, and other questions.
Acknowledgements We are grateful to Dr. D. B. M. Walker for critical reading of the manuscript, to the European Research Council (ERC StG 307609) and the Ministerio de Econom´ıa y Competitividad of Spain (CTQ2011-25714 and Ram´ on y Cajal Fellowship awarded to EMP) for nancial support.
Notes and references 1 J. M. Schnorr and T. M. Swager, Chem. Mater., 2011, 23, 646– 657. 2 R. H. Baughman, A. A. Zakhidov and W. A. de Heer, Science, 2002, 297, 787–792. 3 W. Liang, M. Bockrath, D. Bozovic, J. H. Hafner, M. Tinkham and H. Park, Nature, 2001, 411, 665–669. 4 Ph. Avouris, Z. Chen and V. Perebeinos, Nat. Nanotechnol., 2007, 2, 605–615. 5 J. Liu, A. G. Rinzler, H. Dai, J. H. Hafner, R. K. Bradley, P. J. Boul, A. Lu, T. Iverson, K. Shelimov, C. B. Huffman, F. Rodr´ıguez-Mac´ıas, Y.-S. Shon, T. R. Lee, D. T. Colbert and R. E. Smalley, Science, 1998, 280, 1253–1256. 6 M. A. Hamon, J. Chen, H. Hu, Y. Chen, M. E. Itkis, A. M. Rao, P. C. Eklund and R. C. Haddon, Adv. Mater., 1999, 11, 834– 840. 7 P. Singh, S. Campidelli, S. Giordani, D. Bonifazi, A. Bianco and M. Prato, Chem. Soc. Rev., 2009, 38, 2214–2230. 8 Y.-L. Zhao and J. F. Stoddart, Acc. Chem. Res., 2009, 42, 1161– 1171. 9 P. M. Ajayan and S. Iijima, Nature, 1993, 361, 333–334. 10 S. Iijima, Nature, 1991, 354, 56–58. 11 B. W. Smith, M. Monthioux and D. E. Luzzi, Nature, 1998, 396, 323–324. 12 J. Sloan, J. Hammer, M. Zwiea-Sibley and M. L. H. Green, Chem. Commun., 1998, 347–348.
This journal is ª The Royal Society of Chemistry 2013
Nanoscale 13 D. E. Luzzi and B. W. Smith, Carbon, 2000, 38, 1751–1756. 14 S. Okada, S. Saito and A. Oshiyama, Phys. Rev. Lett., 2001, 86, 3835–3838. 15 E. M. P´ erez and N. Mart´ın, Chem. Soc. Rev., 2008, 37, 1512– 1519. 16 F. Simon, H. Kuzmany, H. Rauf, T. Pichler, J. Bernardi, H. Peterlik, L. Korecz, F. Fulop and A. Janossy, Chem. Phys. Lett., 2004, 383, 362–367. 17 E. Hern´ andez, V. Meunier, B. W. Smith, R. Rurali, H. Terrones, M. Buongiorno Nardelli, M. Terrones, D. E. Luzzi and J. C. Charlier, Nano Lett., 2003, 3, 1037– 1042. 18 J. H. Warner, Y. Ito, M. Zaka, L. Ge, T. Akachi, H. Okimoto, K. Porfyrakis, A. A. R. Watt, H. Shinohara and G. A. D. Briggs, Nano Lett., 2008, 8, 2328–2335. 19 M. Vizuete, M. Barrej´ on, M. J. G´ omez-Escalonilla and F. Langa, Nanoscale, 2012, 4, 4370–4381. 20 O. Vostrowsky and A. Hirsch, Angew. Chem., Int. Ed., 2004, 43, 2326–2329. 21 D. A. Britz, A. N. Khlobystov, J. Wang, A. S. O'Neil, M. Poliakoff, A. Ardavan and G. A. D. Briggs, Chem. Commun., 2004, 176–177. 22 S. Campestrini, C. Corvaja, M. De Nardi, C. Ducati, L. Franco, M. Maggini, M. Meneghetti, E. Menna and G. Ruaro, Small, 2008, 4, 350–356. 23 J. Fan, T. W. Chamberlain, Y. Wang, S. Yang, A. J. Blake, M. Schroeder and A. N. Khlobystov, Chem. Commun., 2011, 47, 5696–5698. 24 M. Koshino, N. Solin, T. Tanaka, H. Isobe and E. Nakamura, Nat. Nanotechnol., 2008, 3, 595–597. 25 E. Nakamura, Angew. Chem., Int. Ed., 2013, 52, 236–252. 26 A. N. Khlobystov, D. A. Britz, A. Ardavan and G. A. D. Briggs, Phys. Rev. Lett., 2004, 92, 245507. 27 K. Ran, J.-M. Zuo, Q. Chen and Z. Shi, ACS Nano, 2011, 5, 3367–3372. 28 C. S. Allen, Y. Ito, A. W. Robertson, H. Shinohara and J. H. Warner, ACS Nano, 2011, 5, 10084–10089. 29 N. Karousis, S. P. Economopoulos, Y. Iizumi, T. Okazaki, Z. Liu, K. Suenaga and N. Tagmatarchis, Chem. Commun., 2010, 46, 9110–9112. 30 T. Umeyama, J. Mihara, H. Hayashi, N. Kadota, V. Chukharev, N. V. Tkachenko, H. Lemmtyinen, K. Yoshida, S. Isoda and H. Imahori, Chem. Commun., 2011, 47, 11781–11783. 31 Fragments of Fullerenes and Carbon Nanotubes: Designed Synthesis, Unusual Reactions, and Coordination Chemistry, ed. M. A. Petrukhina and L. T. Scott, John Wiley & Sons, 2012. 32 D. Canevet, E. M. P´ erez and N. Mart´ın, Angew. Chem., Int. Ed., 2011, 50, 9248–9259. 33 D. Canevet, M. Gallego, H. Isla, A. de Juan, E. M. P´ erez and N. Mart´ın, J. Am. Chem. Soc., 2011, 133, 3184–3190. 34 The [4]CC reported by Isobe and co-workers was synthesized with solubilising hexane chains, as a mixture of isomers (see ref. 35). [4]CC is shown in Fig. 4 without substituents, for simplicity. 35 H. Isobe, S. Hitosugi, T. Yamasaki and R. Iizuka, Chem. Sci., 2013, 4, 1293–1297.
Nanoscale, 2013, 5, 7141–7148 | 7147
Nanoscale 36 For a comprehensive review on fullerene “peapods”, see: F. Simon and M. Monthioux, Fullerenes inside Carbon Nanotubes: The Peapods, in Carbon Meta-nanotubes, John Wiley & Sons, 2011, pp. 273–321. 37 A. N. Khlobystov, D. A. Britz and G. A. D. Briggs, Acc. Chem. Res., 2005, 38, 901–909. 38 J. Sloan and M. Monthioux, Filled Carbon Nanotubes: (X@CNTs), in Carbon Meta-Nanotubes, John Wiley & Sons, 2011, pp. 225–271, and references therein. 39 T. Takenobu, T. Takano, M. Shiraishi, Y. Murakami, M. Ata, H. Kataura, Y. Achiba and Y. Iwasa, Nat. Mater., 2003, 2, 683– 688. 40 A. Chuvilin, E. Bichoutskaia, M. C. Gim´ enez-L´ opez, T. W. Chamberlain, G. A. Rance, N. Kuganathan, J. Biskupek, U. Kaiser and A. N. Khlobystov, Nat. Mater., 2011, 10, 687–692. 41 T. Okazaki, Y. Iizumi, S. Okubo, H. Kataura, Z. Liu, K. Suenaga, Y. Tahara, M. Yudasaka, S. Okada and S. Iijima, Angew. Chem., Int. Ed., 2011, 50, 4853–4857. 42 A. V. Talyzin, I. V. Anoshkin, A. V. Krasheninnikov, R. M. Nieminen, A. G. Nasibulin, H. Jiang and E. I. Kauppinen, Nano Lett., 2011, 11, 4352–4356. 43 Okazaki et al. utilized SWNT-APJ, purchased from Meijo Nanocarbon Co., Ltd. and puried following the procedure reported in S. Okubo, T. Okazaki, N. Kishi, S.-K. Joung, T. Nakanishi, S. Okada and S. Iijima, J. Phys. Chem. C, 2008, 113, 571–575. Talyzin et al. used CVD-synthesized SWNTs produced in-house. 44 J. Zhang, Z. Zhu, Y. Feng, H. Ishiwata, Y. Miyata, R. Kitaura, J. E. P. Dahl, R. M. K. Carlson, N. A. Fokina, P. R. Schreiner, D. Tom´ anek and H. Shinohara, Angew. Chem., Int. Ed., 2013, 52, 3717–3721. 45 J. F. Stoddart, Chem. Soc. Rev., 2009, 38, 1802–1820. 46 G. A. Breault, C. A. Hunter and P. C. Mayers, Tetrahedron, 1999, 55, 5265–5293. 47 The topology of an endohedral derivative with two guest molecules would be equivalent to a [3]rotaxane consisting of two threads and one macrocycle. For an example of the synthesis of such species, see: H. M. Cheng, D. A. Leigh, F. Maffei, P. R. McGonigal, A. M. Z. Slawin and J. Wu, J. Am. Chem. Soc., 2011, 133, 12298–12303. 48 Please note that the stoppers can be “made” by stretching the ends of the thread, so there is no topological difference between rotaxanes and pseudorotaxanes. Their distinction comes from their stability. We will consider this later in the text. 49 Molecular Catenanes, Rotaxanes and Knots: A Journey Through the World of Molecular Topology, ed. J. P. Sauvage and C. Dietrich-Buchecker, Wiley-VCH, Weinheim, 1999. 50 J.-P. Sauvage, From Chemical Topology to Molecular Machines, in Supramolecular Polymer Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA, 2011, pp. 347–359. 51 D. B. Amabilino and L. P´ erez-Garc´ıa, Chem. Soc. Rev., 2009, 38, 1562–1571. 52 K. S. Chichak, S. J. Cantrill, A. R. Pease, S.-H. Chiu, G. W. V. Cave, J. L. Atwood and J. F. Stoddart, Science, 2004, 304, 1308–1312.
7148 | Nanoscale, 2013, 5, 7141–7148
Feature Article 53 L. Wang, M. O. Vysotsky, A. Bogdan, M. Bolte and V. B¨ ohmer, Science, 2004, 304, 1312–1314. 54 J.-F. Ayme, J. E. Beves, C. J. Campbell and D. A. Leigh, Chem. Soc. Rev., 2013, 42, 1700–1712. 55 C. O. Dietrich-Buchecker and J.-P. Sauvage, Angew. Chem., Int. Ed. Engl., 1989, 28, 189–192. 56 J.-F. Ayme, J. E. Beves, D. A. Leigh, R. T. McBurney, K. Rissanen and D. Schultz, Nat. Chem., 2012, 4, 15–20. 57 M. Schliwa, Molecular Motors, Wiley-VCH, Weinheim, 2003. 58 E. M. P´ erez, Angew. Chem., Int. Ed., 2011, 50, 3359–3361. 59 E. R. Kay, D. A. Leigh and F. Zerbetto, Angew. Chem., Int. Ed., 2007, 46, 72–191. 60 J. Bern´ a, G. Bottari, D. A. Leigh and E. M. P´ erez, Pure Appl. Chem., 2007, 79, 39–54. 61 J. Bern´ a, D. A. Leigh, M. Lubomska, S. M. Mendoza, E. M. P´ erez, P. Rudolf, G. Teobaldi and F. Zerbetto, Nat. Mater., 2005, 4, 704–710. 62 J. E. Green, J. Wook Choi, A. Boukai, Y. Bunimovich, E. Johnston-Halperin, E. DeIonno, Y. Luo, B. A. Sheriff, K. Xu, Y. Shik Shin, H.-R. Tseng, J. F. Stoddart and J. R. Heath, Nature, 2007, 445, 414–417. 63 H. Liu, J. He, J. Tang, H. Liu, P. Pang, D. Cao, P. Krstic, S. Joseph, S. Lindsay and C. Nuckolls, Science, 2010, 327, 64–67. 64 W. Song, P. Pang, J. He and S. Lindsay, ACS Nano, 2012, 7, 689–694. 65 With some notable exceptions, for instance Fujita's assembly of [2]catenanes, which occurs under thermodynamic equilibrium. See: M. Fujita, F. Ibukuro, H. Hagihara and K. Ogura, Nature, 1994, 367, 720– 723. 66 P. R. Ashton, I. Baxter, M. C. T. Fyfe, F. M. Raymo, N. Spencer, J. F. Stoddart, A. J. P. White and D. J. Williams, J. Am. Chem. Soc., 1998, 120, 2297–2307. 67 A. Yerin, E. S. Wilks, G. P. Moss, A. Harada, R. M. Hartshorn, T. Damhus, J. Brecher, K. Degtyarenko, S. R. Heller, K. H. Hellwich, P. Hodge, A. T. Hutton, G. J. Leigh, J. Wilson, J. Kahovec, R. G. Jones, A. Lawson, E. Norlander, J. Nyitrai, W. H. Powell, F. L. Ansari, Y. Do, I. L. Dukov, M. A. Hashem, L. H. J. Lajunen, H. Ogino, J. Reedijk, D. Schomburg, W. V. Metanomski, H. A. Favre and G. Eller, Pure Appl. Chem., 2008, 80, 2041–2068. 68 A. Affeld, G. M. Hubner, C. Seel and C. A. Schalley, Eur. J. Org. Chem., 2001, 2877–2890. 69 These denitions would still leave room for argument in the case of reversible covalent bonds. 70 MIMs containing other nanoforms of carbon, namely fullerenes, have been extensively investigated. For reviews, see: (a) A. Mateo-Alonso, Chem. Commun., 2010, 46, 9089– 9099; (b) A. Mateo-Alonso, D. M. Guldi, F. Paolucci and M. Prato, Angew. Chem., Int. Ed., 2007, 46, 8120–8126. 71 M. N. Chatterjee, E. R. Kay and D. A. Leigh, J. Am. Chem. Soc., 2006, 128, 4058–4073. 72 V. Serreli, C.-F. Lee, E. R. Kay and D. A. Leigh, Nature, 2007, 445, 523–527. 73 R. D. Astumian, Science, 1997, 276, 917–922.
This journal is ª The Royal Society of Chemistry 2013
