Rapidly accessible “ click ” rotaxanes utilizing a single amide hydrogen bond templating motif †

The synthesis of hydrogen bond templated rotaxanes using the CuAAC click reaction has been achieved in yields of up to 47%, employing near stoichiometric equivalents of macrocycle and readily prepared azide and alkyne half-axle components. Interlocked structure formation has been confirmed by NMR spectroscopy and mass spectrometry. Density functional theory calculations support 1H NMR spectroscopic analysis that the macrocycle resides over the amide of the axle component, rather than the newly formed triazole, as a result of more favourable hydrogen bond interactions.


Introduction
Rotaxanes, 1 interlocked molecules consisting of macrocyclic ring(s) trapped on stoppered axle(s), have been applied to a number of nanotechnological applications, 2 some of which rely on the relative motion of their ring and axle components.
Alongside metal cations 4 and π-π donor-acceptor interactions, 5 hydrogen bonding, 6 sometimes augmented with additional cationic 7 or anionic 8 assistance, has proven a popular choice of template to generate rotaxanes.While a large number of protocols for rotaxane synthesis have now been disclosed, new synthetic routes that allow rapid access to rotaxanes in good yield and using only stoichiometric amounts of precursor components -that minimises waste -are still of considerable interest.Little used to date has been an elegantly simple hydrogen bond templation system reported by Philp and co-workers in 2008. 9This relies upon the hydrogen bond templated threading of an amide containing component threading through a 2,6-bis-amide pyridyl macrocycle, followed by covalent capture of the resulting pseudo-rotaxane by either maleimide/nitrone 1,3-dipolar cycloaddition 9−11 or aza-Wittig reaction.
12 Inspired by this work, we chose to investigate and are able to report the isolation of rotaxanes in good yields by use of the straightforward and versatile CuAAC click reaction 13 to stopper an azide half-axle, that contains the critical amide functional group required for the pre-requisite pseudo-rotaxane formation with a polyether-isophthalamide macrocycle (Figure 1).Interlocked structure formation was confirmed by 1 H NMR spectroscopy and mass spectrometry.Density functional theory calculations support 1 H NMR spectroscopic analysis which suggests that in the prepared rotaxanes the macrocyclic ring remains over the amide of the axle component rather than translating to the newly formed triazole.

Synthesis of rotaxane precursors
The structures of all rotaxane precursors used in this study, along with the syntheses of novel compounds, are presented

"Click" synthesis and spectral characterisation of rotaxanes
Synthesis of rotaxanes 6 and 7 were attempted as according to Scheme 2. In each case, to a solution of macrocycle 1 in dichloromethane, 1.1 equivalents of azide 5 and 1.1 equivalents of alkyne (2 or 3) were added.These were followed by catalytic Cu(CH 3 CN) 4 BF 4 and TBTA, and 1.2 equivalents of DIPEA.The reactions were stirred overnight at room temperature under an inert atmosphere, and then submitted to aqueous workup, before purification on silica.By analogous reactions (in the absence of macrocycle 1), noninterlocked axles 8 and 9 were also prepared (in 88% and 66% yield respectively) to aid 1 H NMR spectroscopic interpretation.It was not possible to separate rotaxane 6 and unthreaded macrocycle 1 by silica gel column chromatography.Purification using prep TLC, allowed for isolation of pure rotaxane 6 in a reasonable 32% yield.Pleasingly, rotaxane 7 could be purified by careful column chromatography, being isolated in a yield of 47%.Please do not adjust margins Please do not adjust margins The 1 H NMR spectrum of rotaxane 6, along with that of non-interlocked macrocycle 1 and axle 8 for comparison, is shown in Figure 2. 16 The upfield shift and splitting of protons f and g in rotaxane 6 compared to macrocycle 1 is consistent with intercalation of the axle component between the aromatic rings of the macrocycle.The downfield shift of protons c and d in the rotaxane are indicative of interactions with a hydrogen bond acceptor on the axle.The functional group of the axle residing within the macrocyclic ring in rotaxane 6 is believed to be the templating amide, and not the newly formed triazole.Evidence for this conclusion is provided by the large upfield shifts in axle protons m and n and in particular the lack of a significant shift in proton r.Another notable feature of the 1 H NMR spectrum of rotaxane 6 is the splitting of protons e, h, i and j, which arise from the two faces of the macrocycle no longer being equivalent in the rotaxane due to the directionality of the threaded axle.Comparison of the 1 H NMR spectra of macrocycle 1, rotaxane 7 and axle 9 reveals similar trends, indicative of the macrocycle also residing over the axle amide in rotaxane 7, rather than the triazole or ester (see ESI †). 16 The interlocked natures of rotaxanes 6 and 7 are further supported by the appearance of through-space correlations being observed in the

Computational modelling of rotaxanes
While reasonably confident in our deduction that the macrocycle is residing over the axle amide, and not the triazole functionality, in both rotaxanes, we sought further evidence to confirm the co-conformational behaviour of the rotaxanes.Unable to grow single crystals for x-ray structure determination, we therefore turned our attention to computational modelling.Density-functional theory (DFT) calculations were undertaken on both 6 and 7. 17 The calculations were performed using the Gaussian 09 program, 18 using the B3LYP exchange-correlation functional 19 together with the 6-31G* basis set for all atoms.Solvent effects (CHCl 3 ) were modelled using a polarisable continuum model, using the default solvent cavity parameters as defined in Gaussian 09.
For each of the rotaxanes, geometry optimisations were run with the macrocycle starting over (i) the amide and (ii) the triazole functional groups of the axle component. 20Distinct optimised structures were obtained in each case (see Figure 3 and ESI †), but for both rotaxanes there was an energetic preference for the macrocycle being over the amide "station" (41 kJ mol −1 for 6 and 76 kJ mol −1 for 7).This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Inspection of the optimised structures suggests that this preference is due to more favourable hydrogen bonding interactions being present when the amide "station" is occupied by the macrocycle. 21In this case the isophthalamide of the macrocycle adopts a syn-syn conformation and presents three hydrogen bond donors to the O atom of the amide "station".If the macrocycle resides at the triazole "station", the isophthalamide adopts a syn-anti conformation, resulting in only one N-H hydrogen bond donating to the N atom of the triazole. 22Also, in the preferred amide "station" coconformation, the triazole C-H hydrogen bonds to a polyether oxygen on the macrocyclic ring, further stabilising this conformation.A structural consequence of this additional hydrogen bonding interaction is to bend the axle, such that protons o and p are partially within the macrocycle cavity, which is consistent with the significant upfield shifts of these protons in the rotaxane 1 H NMR spectra compared to the spectra of the free axles (see Figure 2 and ESI †).Finally, there is a substantially greater difference in energies for the two coconformations of rotaxane 7 (compared to 6).We attribute this to the disruption in the planarity of the ester and bulky aromatic stopper group in the co-conformation of 7 when the macrocycle occupies the neighbouring triazole "station" (Figure 3b).

Conclusions
This study has demonstrated it is possible to generate rotaxanes by short, straightforward and highly yielding reaction sequences, employing CuAAC click chemistry to stopper Philp-style self-assembled pseudo-rotaxanes of an amide half-axle threaded through an aryl bis-amide macrocycle.Density functional theory calculations provide strong supporting evidence for the co-conformational behaviour of the prepared rotaxanes, with the macrocycle remaining over the templating amide of the axle, rather than over the newly formed triazole functionality.We are confident that the synthetic ease in generating these rotaxanes will not only be appealing to those already preparing rotaxanes, but will encourage other chemists to begin using rotaxanes in their own research.We ourselves are undertaking further study of this class of rotaxane with respect to synthetic methodology development and molecular motion and host-guest applications.

Experimental General information
Commercially available solvents and chemicals were used without further purification unless stated.Dry solvents, NEt 3 and DIPEA were purchased dry and stored under an inert atmosphere.Cu(CH 3 CN) 4 BF 4 was stored in a desiccator over P 4 O 10 .Deionised water was used in all cases.Previously reported macrocycle 1 14 has now been prepared in higher yield by use of a template method, details of which may be found in the ESI †.Alkyne 2 was prepared as according to the previously reported procedure. 15 Silica gel with a 60Å particle size was used as the stationary phase for column chromatography.Analytical TLC was used to monitor the progress of column chromatography and analytical TLC plates were typically examined under short wavelength (λ = 254 nm) UV light.If required, ceric ammonium molybdate or potassium permanganate stains were used to develop the analytical TLC plates.Preparatory TLC was carried out on silica gel possessing a fluorescent indicator to allow for examination with short wavelength UV light.
IR spectra were recorded on an Agilent Technologies Cary 630 FTIR spectrometer.NMR spectra were recorded on a Bruker Ultrashield 400 Plus spectrometer at 298 K, with the NMR data reported below assigned according to the atom labels to be found in the Schemes 1 and 2. Mass spectra were recorded on a Waters Xevo G2-S or Thermofisher LTQ Orbitrap XL at the EPSRC UK National Mass Spectrometry Facility at Swansea University, UK, or a Shimadzu LCMS IT ToF instrument at Lancaster University, UK.Melting points were recorded on a Gallenkamp capillary melting point apparatus and are uncorrected.

Computational methodology
The Gaussian 09 program was used to conduct all density functional theory calculations. 18In each case, geometry optimisations using the B3LYP/6-31G* model chemistry were undertaken, 19 with CHCl 3 solvent effects modelled using the default polarisable continuum model (and default solvent cavity parameters) as defined in Gaussian 09.To obtain reasonable starting structures, the structures were preoptimised using a combination of molecular mechanics and semi-empirical PM6 methods.Several 'starting points' were considered for the optimisations, to minimise bias in the final structures.
Phys., 1993, 98, 1372-1377; (d) C. Lee, W. Yang, and R. G. Parr, Phys.Rev. B, 1998, 37, 785-789.20 In the case of rotaxane 7 similar geometry optimisations were also run with the macrocycle starting over the ester functional group, but repeatedly the macrocycle would translate along the axle over the course of the optimisation, to one of the other "stations".This result is as would be expected as the ester group is unable to act as a hydrogen bond donor.21 See ESI † for a full tabulated summary of energies and hydrogen bond distances in the simulated rotaxane structures.22 As detailed in the ESI † for both rotaxanes, when the macrocycle resides at the amide "station", the axle N-H•••O polyether hydrogen bond is significantly longer than the equivalent triazole C-H•••O hydrogen bond when the macrocycle occupies the triazole "station".However, we hypothesise that this difference (which would favour macrocycle occupancy of the triazole "station") is outweighed by the interactions that favour occupancy of the amide "station" detailed in the main text.

Figure 1 :
Figure 1: Schematic representation of the rapid synthesis of rotaxanes utilizing a single amide hydrogen bond templating motif and the CuAAC click reaction.

1 H- 1 H
ROESY NMR spectra between signals arising from protons in the macrocyclic and axle components (see ESI †).In addition, molecular ion peaks for the rotaxanes are observed in the positive-ion electrospray mass spectra (see ESI †).

Figure 3 :
Figure 3: Minimum energy structures of rotaxane 7 with macrocycle occupying (a) amide and (b) triazole "stations".Hydrogen bonds are represented by green dashed lines.Hydrogen atoms not participating in hydrogen bonding are omitted for clarity.