appropriate second station when designing a molecular
switch.6 The search for new threading guests for macro-
cycles with reasonable affinity in solution is never straight-
forward, thereby limiting the development of new molec-
ular switches. Theoretically, weakly binding guest units
would serve as perfectly acceptable primary recognition
sites in molecular switches if an even weaker recognition
unit for the macrocycle were present in the dumbbell-
shaped component. Using this concept, many guest units
that do not, when isolated, exhibit particularly high binding
affinities for macrocycles have the potential to act as
primary or secondary recognition sites in interlocked mo-
lecular switches, potentially increasing the number of re-
cognition units available for the construction of new
molecular switches. Two questions arise: (i) How do we
assemble rotaxanes efficiently from weakly associated host/
guest components? (ii) How do we choose which combina-
tions of recognition units are appropriate for constructing
molecular switches? Herein, we report a new molecular
switch constructed using the concept of weak recognition.
We constructed this [2]rotaxane first by using a metal ion
to template the formation of a pseudorotaxane from bis-p-
xylyl[26]crown-6 (BPX26C6)7 and a diphenylurea deriva-
tive and then modifying the pseudorotaxane post-assembly
to form another weakly interacting tertiary ammonium ion
recognition site for the crown ether component, the migra-
tion of which between the tertiary ammonium center and
the diphenylurea station we could control using in situ base/
acid treatment or metal ion addition/removal.
Scheme 1. Synthesis of the Molecular Switch 4-H PF6
3
To facilitate the synthesis of the [2]rotaxane, we sought a
host/guest system that associates with the requirement of a
template because the binding strength would be weakened
significantly simply by removing the template after inter-
locking the components. Previously, we reported that a
diphenylurea unit can thread through the macrocycle
BPX26C6 in the presence of a templating Naþ ion;8 we
suspected that the ethylene glycol units of BPX26C6 would
interact weakly with the NH units of the diphenylurea
station in the resulting rotaxane after removal of the
template. For the stoppering reaction, we chose reductive
amination9 because of the potential to covalently modify
the resulting secondary amine to form a tertiary ammo-
nium center that would interact significantly weaker with
the crown ether component.10 Thus, after removal of the
Naþ ion that facilitated the synthesis, we expected the
tertiary ammonium center and the diphenylurea unit to
serve as suitable stations for the translocation of the
BPX26C6 component.
We isolated the [2]rotaxane 1-H PF6 in 25% yield after
3
(i) mixing BPX26C6, the aldehyde 2, the amine 3, and
NaClO4 in CH2Cl2, (ii) reducing the resulting imine, and
(iii) acidification (Scheme 1). Two triplets (at δ 2.75ꢀ2.90)
for the methylene protons adjacent to the NH2þ center ap-
peared in the 1H NMR spectrum of 1-H PF6 (Figure 1a);
3
strong cross-signals for these protons to the aromatic and
ethylene glycol protons of BPX26C6 appeared in the 2D
NOESY spectrum, suggesting that the macrocyclic unit
resided about the NH2þ center in the ground state of this
[2]rotaxane. The very weak interaction between BPX26C6
and the diphenylurea station, relative to that between the
macrocyclic unit and the DBAþ station, was confirmed by
our inability to transfer the macrocyclic component to the
diphenylurea station upon the addition of Et3N in CDCl3,
even when we added more than 50 equiv of the base.
We methylated the DBAþ center of the [2]rotaxane
(6) (a) Chiu, S.-H.; Rowan, S. J.; Cantrill, S. J.; Stoddart, J. F.; White,
A. J. P.; Williams, D. J. Chem.;Eur. J. 2002, 8, 5170–5183. (b)
Nakazono, K.; Takata, T. Chem.;Eur. J. 2010, 16, 13783–13794.
(7) (a) Cheng, P.-N.; Huang, P.-Y.; Li, W.-S.; Ueng, S.-H.; Hung,
W.-C.; Liu, Y.-H.; Lai, C.-C.; Peng, S.-M.; Chao, I.; Chiu, S.-H. J. Org.
Chem. 2006, 71, 2373–2383. (b) Chen, N.-C.; Huang, P.-Y.; Lai, C.-C.;
Liu, Y.-H.; Peng, S.-M.; Chiu, S.-H. Chem. Commun. 2007, 4122–4124.
(c) You, Y.-C.; Tzeng, M.-C.; Lai, C.-C.; Chiu, S.-H. Org. Lett. 2012, 14,
1046–1049.
1-H PF6 through treatment with formaldehyde and for-
3
mic acid,10b providing the [2]rotaxane 4 after washing the
resulting solution with aqueous NaOH and purifying the
residue through column chromatography; acidification
(8) Lin, Y.-H.; Lai, C.-C.; Liu, Y.-H.; Peng, S.-M.; Chiu, S.-H.
Angew. Chem., Int. Ed. 2013, 52, 10231–10236.
(9) For more examples of using this method to synthesize rotaxanes,
see: (a) Cantrill, S. J.; Rowan, S. J.; Stoddart, J. F. Org. Lett. 1999, 1,
1363–1366. (b) Eelkema, R.; Maeda, K.; Odell, B.; Anderson, H. L.
J. Am. Chem. Soc. 2007, 129, 12384–12385.
(10) (a) Nakazono, K.; Kuwata, S.; Takata, T. Tetrahedron Lett.
2008, 49, 2397–2401. (b) Suzuki, S.; Nakazono, K.; Takata, T. Org. Lett.
2010, 12, 712–715.
and anion exchange gave the [2]rotaxane 4-H PF6. We
3
recorded 2D COSY and NOESY spectra to identify the
signals of the protons of the two methylated [2]rotaxanes
(see the Supporting Information) and the positions of their
interlocked BPX26C6 components. The observation of
Org. Lett., Vol. 15, No. 22, 2013
5743