"Magic rod" rotaxanes: The hydrogen bond-directed synthesis of molecular shuttles under thermodynamic control

Article abstract of DOI:10.1021/ol0344927

(Matrix presented) Peptide [2]- and [3]rotaxanes are assembled in high yields under thermodynamic control using hydrogen bonding interactions and reversible cross olefin metathesis.

Full text of DOI:10.1021/ol0344927

ORGANIC
LETTERS
2003
Vol. 5, No. 11
1907-1910
“Magic Rod” Rotaxanes: The Hydrogen
Bond-Directed Synthesis of Molecular
Shuttles under Thermodynamic Control
Jeffrey S. Hannam, Timothy J. Kidd, David A. Leigh,* and Andrew J. Wilson
School of Chemistry, UniVersity of Edinburgh, The Kings Buildings, West Mains Road,
Edinburgh EH9 3JJ, United Kingdom
daVid.leigh@ed.ac.uk
Received March 21, 2003
ABSTRACT
Peptide [2]- and [3]rotaxanes are assembled in high yields under thermodynamic control using hydrogen bonding interactions and reversible
cross olefin metathesis.
While the use of noncovalent interactions to direct the
synthesis of rotaxanes1 often leads to greatly improved yields
over statistical methods, the final step in the rotaxane-forming
reaction (“clipping” or “capping”, Scheme 1) is generally
the rotaxane is the most energetically favored structure
because of the built in favorable noncovalent interactions
between the components. So-called “slippage” strategies2 are
under thermodynamic control,3 but the necessary elevated
temperatures reduce the strength of noncovalent interactions
and therefore also the efficiency of rotaxane formation.
Accordingly, novel approaches to the efficient synthesis of
rotaxanes are still desirable.
Scheme 1. Common Noncovalent Bond-Directed Strategies for
Rotaxane Synthesisa
Thermodynamic control over mechanical bond formation
is particularly amenable to the principles of dynamic com-
binatorial chemistry (DCC).4 The kinetic lability of metal-
ligand coordination bonds and disulfide exchange reactions
have both been employed to shift equilibria between
interlocked and non-interlocked products.5-7 Imine exchange,
together with post-assembly covalent modification,3 allows
the high-yielding synthesis of rotaxanes by both clipping and
a Clipping and capping steps are usually irreversible, so “errors”
in the synthesis (i.e., noninterlocked byproducts) cannot be cor-
rected.
(2) (a) Raymo, F. M.; Houk, K. N.; Stoddart, J. F. J. Am. Chem. Soc.
1998, 120, 9318-9322. (b) Heim, C.; Affeld, A.; Nieger, M.; Vo¨gtle, F.
HelV. Chim. Acta 1999, 82, 746-759.
(3) (a) Lindsey, J. S. New J. Chem. 1991, 15, 153-180. (b) Philp, D.;
Stoddart, J. F. Angew. Chem., Int. Ed. Engl. 1996, 35, 1154-1196.
(4) (a) Lehn, J.-M. Chem Eur. J. 1999, 5, 2455-2463. (b) Rowan, S. J.;
Cantrill, S. J.; Cousins, G. R. L.; Sanders, J. K. M.; Stoddart, J. F. Angew.
Chem., Int. Ed. 2002, 41, 898-952.
(5) (a) Fujita, M.; Ibukuro, F.; Hagihara, H.; Ogura, K. Nature 1994,
367, 720-723. (b) Fujita, M.; Aoyagi, M.; Ibukuro, F.; Ogura, K.;
Yamaguchi, K. J. Am. Chem. Soc. 1998, 120, 611-612. (c) Try, A. C.;
Harding, M. M.; Hamilton, D. G.; Sanders, J. K. M. Chem. Commun. 1998,
723-724. (d) Padilla-Tosta, M. E.; Fox, O. D.; Drew, M. G. B.; Beer, P.
D. Angew. Chem., Int. Ed. 2001, 40, 4235-4239.
under kinetic control. Consequently, noninterlocked byprod-
ucts formed during the reaction cannot be recycled even if
(1) (a) Schill, G. Catenanes, Rotaxanes and Knots; Academic Press: New
York, 1971. (b) Amabilino, D. B.; Stoddart, J. F. Chem. ReV. 1995, 95,
2725-2828. (c) Sauvage, J.-P., Dieterich-Buchecker, C. O., Eds. Molecular
Catenanes Rotaxanes and Knots; Wiley-VCH: Weinheim, Germany, 1999.
10.1021/ol0344927 CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/30/2003

Scheme 2. Synthesis of [2]- and [3]Rotaxanes 4a-c and 5a-b under Thermodynamic Control Using a Reversible Threading Strategy
Facilitated by Olefin Metathesis
capping8 and has also been used to achieve catenane synthesis
under thermodynamic control.9 Of the remaining synthetic
transformations previously used in DCC, olefin metathesis
is advantageous because removal of a reaction-specific
catalyst (e.g., 1) is all that is required to “lock” the product
distribution. Indeed, olefin metathesis has been used to
synthesize catenanes,9,10 rotaxanes11 and knots12 under kinetic
control and cantenanes and catenates under thermody-
namic9,13 control. In our own laboratories, reversible ring-
opening metathesis combined with the hydrogen bond-
directed assembly of self-complementary macrocycles allowed
the synthesis of kinetically robust catenanes in near quantita-
tive yields.14 Here we report the extension of this concept
to the synthesis of hydrogen bond-assembled rotaxanes under
thermodynamic control.
Schematically, when benzylic amide macrocycle 2 and a
rod 3a-c containing two bulky stoppers, peptide-based
template sites, and a “magic”14 olefin are exposed to Grubbs
first-generation metathesis catalyst 1, the rod is opened and
the macrocycle binds to the template through four-point
hydrogen bonding. Upon reformation of the carbon-carbon
double bond, the macrocycle is trapped on the thread
producing a [2]rotaxane (4a-c, Scheme 2). The [2]rotaxane
(6) For capping, see: (a) Chichak, K.; Walsh, M. C.; Branda, N. R. Chem.
Commun. 2000, 847-848. (b) Baer, A. J.; Macartney, D. H. Inorg. Chem.
2000, 39, 1410-1417. (c) Gunter, M. J.; Bampos, N.; Johnstone, K. D.;
Sanders, J. K. M. New J. Chem. 2001, 25, 166-173. For clipping, see: (d)
Hunter, C. A.; Low, C. M. R.; Packer, M. J.; Spey, S. E.; Vinter, J. G.;
Vysotsky, M. O.; Zonta, C. Angew. Chem., Int. Ed. 2001, 40, 2678-2682.
(e) Chang, S.-Y.; Choi, J. S.; Jeong, K.-S. Chem. Eur. J. 2001, 7, 2687-
2697.
(7) (a) Furusho, Y.; Hasegawa, T.; Tsuboi, A.; Kihara, N.; Takata, T.
Chem. Lett. 2000, 18-19. (b) The kinetically controlled crystallization of
a potentially dynamic system appears to have been used to advantage in a
high-yielding preparation of a [3]rotaxane: Kolchinski, A. G.; Alcock, N.
W.; Roesner, R. A.; Busch, D. H. Chem. Commun. 1998, 1437-1438.
(8) (a) Cantrill, S. J.; Rowan, S. J.; Stoddart, J. F. Org. Lett. 1999, 1,
1363-1366. (b) Rowan, S. J.; Stoddart, J. F. Org. Lett. 1999, 1, 1913-
1916. (c) Glink, P. T.; Oliva, A. I.; Stoddart, J. F.; White, A. J. P.; Williams,
D. J. Angew. Chem., Int. Ed. 2001, 40, 1870-1875.
(11) (a) Wisner, J. A.; Beer, P. D.; Drew, M. G. B.; Sambrook, M. R. J.
Am. Chem. Soc. 2002, 124, 12469-12476. (b) Coumans, R. G. E.; Elemans,
J. A. A. W.; Thordarson, P.; Nolte, R. J. M.; Rowan, A. E. Angew. Chem.,
Int. Ed. 2003, 42, 650-654;
(12) Dietrich-Buchecker, C. O.; Rapenne, G.; Sauvage, J.-P. Chem.
Commun. 1997, 2053-2054.
(13) Hamilton, D. G.; Feeder, N.; Teat, S. J.; Sanders, J. K. M. New J.
Chem. 1998, 1019-1021.
(14) For organic “magic” rings-macrocycles that interlock without any
final net change to their covalent structure, see: Kidd, T. J.; Leigh, D. A.;
Wilson, A. J. J. Am. Chem. Soc. 1999, 121, 1599-1600. For the original
magic rings based on coordination chemistry, see ref 5a.
(9) Leigh, D. A.; Lusby, P. J.; Teat, S. J.; Wilson, A. J. Angew. Chem.,
Int. Ed. 2001, 40, 1538-1543.
(10) (a) Mohr, B.; Weck, M.; Sauvage, J.-P.; Grubbs, R. H. Angew.
Chem., Int. Ed. Engl. 1997, 36, 1308-1310. (b) Weck, M.; Mohr, B.;
Sauvage, J.-P. Grubbs, R. H. J. Org. Chem. 1999, 64, 5463-5471.
1908
Org. Lett., Vol. 5, No. 11, 2003

Table 1. Yields of Magic Rod Rotaxanes 4a-c and 5a-b from Olefin Metathesis of Magic Rods 3a-c
reactants
concn
[2]rotaxane 4a -c
[3]rotaxane 5a -b
interlocked products
1 equiv of 2 and 3a
1 equiv of 2 and 3b
1 equiv of 2 and 3c
5 equiv of 2 and 3b
1 equiv of 2 and 3b
5a
0.2 M
0.2 M
0.2 M
0.2 M
0.05 M
0.05 M
0.0002 M
15%
36%
34%
52%
29%
29%
1%
0%
20%
24%
43%
11%
11%
0%
15%
56%
58%
95%
40%
40%
1%
1 equiv of 2 and 3b
can subsequently become a [3]rotaxane (5a, 5b) through
another round of metathesis (models show that the ruthenium
carbene acts as a stopper in the first round preventing direct
assembly of the [3]rotaxane).
ever, this is probably because the N-terminal amide is
sterically hindered by the stopper and therefore cannot bind
efficiently to the macrocycle. The reactions were shown to
be under true thermodynamic control by the identical product
distributions obtained from the metathesis of different starting
materials containing the same overall proportions of thread
and macrocycle (for example [3]rotaxane 4b and a 1:2
mixture of thread 3b and macrocycle 2).
The synthesis of macrocycle 2 has previously been
reported,14 while the syntheses of threads 3a-c were carried
out in four steps using simple amide and ester bond-forming
reactions and cross olefin metathesis in the final step. The
long C20 threads were used because we found shorter chains
(C10) to react poorly, either because of steric hindrance or
the formation of metal chelates with the Grubbs catalyst as
proposed in the literature15. A triphenylmethine stopper was
used as smaller groups allowed the macrocycle to de-
thread.16,17 Metathesis experiments were carried out in CH2-
Cl2 (the noncompeting solvent maximizes the strength of the
intercomponent hydrogen bonding) using Grubbs catalyst 1.18
As the metathesis reaction is reversible (although care needs
to be taken to ensure this14), the product distribution is
determined only by the relative stabilities of the products;
at high concentrations, predominantly threaded species are
obtained that can, if desired, be converted back to the
uninterlocked components by simply diluting the reaction
mixture (but only in the presence of active catalyst). The
rather spectacular results (overall yields of interlocked
products ranging from 1 to 95% merely by changing the
concentration) of the metathesis reactions of “magic rods”
3a-c at different concentrations are shown in Table 1.
Decomposition of the catalyst (for example, by adding
n-propylamine) or simply sequestering it from the reaction
mixture with poly(divinylbenzene) fixes the product distribu-
tion unless, and until, additional catalyst is added. Trifluo-
roacetylation ((CF3CO)2O) of the amide groups of the
rotaxane removes the intramolecular hydrogen bonding
interactions and allows disassembly of the rotaxane into its
components at any concentration (but only in the presence
of 1).14
Representative 1H NMR spectra of the glycylglycine magic
rod rotaxanes 4b, 5a, and thread 3b in CDCl3 at 50 °C are
shown in Figure 1. The resonances of the thread He and Hg
protons appear at the normal chemical shifts for glycine
residues (Figure 1a). The same signals in the [2]rotaxane
4b (Figure 1b), however, display the characteristic upfield
shifts19 of a threaded species as a result of shielding by the
aromatic rings of the macrocycle (note, each peptide station
is only occupied at most 50% of the time in the [2]rotaxane).
The amide protons, HD, of the macrocycle appear downfield
due to hydrogen bonding. Only one thread amide (Hf)
undergoes an appreciable change in chemical shift and is
shifted upfield, suggesting that the effects of hydrogen
bonding are mostly offset by the shielding effect of the
macrocycle and/or the thread amide groups intramolecularly
hydrogen bonding in 3b. Only one set of signals is seen for
each glycylglycine station, indicating that shuttling of the
macrocycle along the axis of the thread is fast on the NMR
time scale. The [3]rotaxane (Figure 1c) experiences changes
in chemical shift similar to the [2]rotaxane, although these
are somewhat more pronounced (because each peptide station
is occupied nearly 100% of the time in the [3]rotaxane).
Finally, in the case of the [3]rotaxane, an ABX system is
observed for the HE protons of macrocycle 2. This arises
because the faces of the macrocycle experience different
environments in the [3]rotaxane (one points toward the other
macrocycle, the other toward the nearest stopper), while in
the analogous [2]rotaxane, the two faces of the fast-shuttling
macrocycle effectively experience identical environments.
In a manner similar to previously described peptide-based
molecular shuttles,19 this class of rotaxanes exhibit solvent-
dependent translational isomerism; that is, in d6-DMSO, the
hydrogen bonding between the thread and the macrocycle
is disrupted and the macrocycle(s) sit(s) predominantly on
the alkyl chain of the thread shielding it from the polar
solvent.
The yields imply that at least two amides are necessary in
each template for rotaxane assembly to be effective. How-
(15) Fu¨rstner, A.; Langemann, K. J. Am. Chem. Soc. 1997, 119, 9130-
9136.
(16) Experiments (Supporting Information) reveal that macrocycle 2 and
glycylglycine threads bearing diphenyl stoppers form complexes (pseu-
dorotaxanes, Ka ) 300 M-1) only slowly at room temperature. This could
be considered “slow slippage” rotaxane formation.
(17) The trischlorophenyl derivative was readily available in larger and
cheaper quantities than other alternatives.
(18) For a detailed description of the experimental procedure, see
Supporting Information.
(19) Lane, A. S.; Leigh, D. A.; Murphy, A. J. Am. Chem. Soc. 1997,
119, 11092-11093.
Org. Lett., Vol. 5, No. 11, 2003
1909

Figure 1. 1H NMR spectra in CDCl3 at 50 °C of (a) bis-glycylglycine thread 3b, (b) [2]rotaxane 4b, and (c) [3]rotaxane 5a.
In conclusion, the combination of directed hydrogen
bonding and cross olefin metathesis provides a powerful tool
for the assembly (or disassembly) of rotaxanes through built
in recycling of nonpreferred byproducts; a useful quality in
developing “engineering up” approaches to large functional
supermolecules. Significantly, since the reversible reaction
involves breaking a strong CdC bond and only occurs in
the presence of a specific catalyst, molecules assembled in
this way are not inherently labile. Furthermore, this technique
can be carried out at room temperature and below, so the
programmed noncovalent interactions can be maximized. The
result is the best possible kind of synthetic strategy: a
thermodynamically controlled, noncovalent bond-directed
assembly of a kinetically robust final superstructure.
Acknowledgment. The authors would like to thank the
EPSRC and the TMR initiative of the European Union
through Contracts FMRX-CT96-0059 and FMRX-CT97-
0097 for support of this research. D.A.L. is an EPSRC
Advanced Research Fellow.
Supporting Information Available: Full experimental
procedures. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL0344927
1910
Org. Lett., Vol. 5, No. 11, 2003