Energy landscape of a hydrogen-bonded non-degenerate molecular shuttle

Article abstract of DOI:10.1039/b926868e

In rotaxane 1, two co-conformations are populated in CDCl₃ at temperatures between 250 and 330 K. The thermodynamic parameters show strong enthalpy-entropy compensation, and a non-negligible heat capacity difference between the two forms. The Royal Society of Chemistry.

Full text of DOI:10.1039/b926868e

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Energy landscape of a hydrogen-bonded non-degenerate
molecular shuttlew
¨
Duygu Deniz Gunba-s, Leszek Zalewskiz and Albert M. Brouwer*
Received (in Cambridge, UK) 21st December 2009, Accepted 30th January 2010
First published as an Advance Article on the web 18th February 2010
DOI: 10.1039/b926868e
In rotaxane 1, two co-conformations are populated in CDCl3 at
temperatures between 250 and 330 K. The thermodynamic
parameters show strong enthalpy–entropy compensation, and a
non-negligible heat capacity difference between the two forms.
and entropy of the two forms are derived. Interestingly,
the van’t Hoff plot is distinctly non-linear, indicating that a
non-negligible heat capacity change is involved. Such detailed
thermodynamic studies of hydrogen bonded molecular
shuttles have hitherto not been reported.
In the previously studied naphthalimide/succinamide
rotaxane 2,7–9 the succ co-conformation predominates very
strongly in the neutral state. Translocation of the macrocycle
from the succ station to the ni-station occurs upon one-
electron reduction of the naphthalimide unit using electro-
chemical or photochemical methods.7–9 In rotaxane 1 we have
inserted an amide unit (glycine) in the thread, in order to
facilitate binding of the ring near the ni-station.
The understanding of the fundamental forces involved in the
non-covalent interactions that govern molecular recognition in
biological and other (macro)molecules is of paramount
importance in contemporary chemistry.1 Molecular architectures
in which components are mechanically bonded are an attractive
research tool for the study of such relatively weak interactions,
because the interlocking allows their study under a wide range
of conditions, including those under which similar non-
interlocked complexes would dissociate.2 In rotaxane molecular
shuttles a macrocyclic ring trapped on a dumbbell-shaped axle
or thread can move between different distinct binding sites
(‘‘stations’’), giving rise to different co-conformations.3–5 In
order to design systems with a predictable distribution of
co-conformers, accurate knowledge of the relative binding
free energies is needed.5,6 In the present work we have studied
rotaxane 1 (Scheme 1) in which hydrogen bonding forms the
main interaction between the ring and the two stations. Using
1H NMR spectroscopy, the populations of the two co-conformers
are obtained, and from the temperature dependence of the
equilibrium constant, the differences in free energy, enthalpy
Comparison of the 1H-NMR spectra of
1 and the
corresponding thread10 3 in CDCl3 (Fig. 1) shows that the
methylene protons Hp and Hr are shielded due to the aromatic
ring current effect of the p-xylylene rings. However, the extent
of shielding is less (1.1 ppm) than what was observed in the
previously studied rotaxane 2 (1.45 ppm).7,8 Moreover, also
the Hd protons are markedly shielded in rotaxane 1 compared
to thread 3, which shows that a significant fraction of rotaxane
1 exists as the ni-co-conformer.
Upon lowering the temperature, the resonances associated
with the Hd protons are less shielded, which indicates that the
population of the ni-co-conformer decreases when the
temperature is decreased. Over the whole temperature range
studied the shuttling remains fast on the NMR time scale, and
the chemical shifts are population weighted averages,11 in
agreement with published results for related systems.12
In order to obtain a measure of the shielding of the Hd
protons in a pure ni-co-conformer, a symmetrical model
Scheme 1 Co-conformational equilibria in rotaxanes 1 and 2 (CDCl3,
298 K).
van ’t Hoff Institute for Molecular Sciences, Nieuwe Achtergracht 129,
1018 WS, Amsterdam, the Netherlands. E-mail: a.m.brouwer@uva.nl;
Fax: +31 205255680; Tel: +31 205255491
w Electronic supplementary information (ESI) available: Details of the
synthesis and characterization, and variable temperature NMR
studies. See DOI: 10.1039/b926868e
Fig. 1 1H-NMR spectra (400 MHz) of rotaxane 1 and thread 3 in
z Present address: Reckitt Benckiser Produktions GmbH, Ludwig-
Bertram-Str. 8-10, 67059 Ludwigshafen, Germany.
CDCl3 at 298 K.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 2061–2063 | 2061

View Article Online
Scheme 2 Structure of model rotaxane 4.
rotaxane 4 was used. This consists of two ni-gly units, linked
with a saturated C12 chain (Scheme 2). In this rotaxane the
macrocycle can only hydrogen bond to one of the ni-gly units,
so that an intrinsic shielding of the CH2 group (1.178 ppm at
298 K) could be derived by comparison with the corresponding
thread 5.10
Fig. 3 DG0, DH0 and ꢁTDS0 vs. temperature for the co-conformational
equilibrium of rotaxane 1.
The population (a) of the ni-co-conformer in 1 was
calculated from eqn (1). The d’s refer to the chemical shifts
of Hd in 1, 3, 4 and 5. Knowing a, the population of the
succ-co-conformer is simply given by 1 ꢁ a, and the
equilibrium constant K can be obtained as K = (1 ꢁ a)/a.
succinamide protons indicates a population of ca. 76% of the
succ-co-conformer. From the shielding of Hd a population
of 20% of the ni-co-conformer is derived. Thus, if a third
co-conformer exists, it is minor (o5%), and its presence
cannot account for the non-linearity of the van’t Hoff plot
(see ESIw for more detailed discussion).
d1 ꢁ d3
a ¼
ð1Þ
2ðd4 ꢁ d5Þ
The interaction between the macrocycle and the stations
primarily involves hydrogen bonding. Because of its enormous
importance in the chemistry of biological molecules as well as
synthetic materials, simple predictive models for the strengths
of hydrogen bonds are very useful.14 Although such models
provide useful guidelines, accurate a priori prediction of
hydrogen bond strengths remains difficult. Ab initio calculations
can in principle give very accurate interaction enthalpies
for model systems,15 but, as the present results underscore,
entropy factors can also be very important, and these are
difficult to calculate. Moreover, even in a relatively simple
system as 1, the conformational space is quite large. The two
translational co-conformers really correspond to low-energy
valleys in the free energy landscape in which several local
minima are populated. In Schemes 1 and 2, the hydrocarbon
chains are drawn in the most favorable extended conformation,
but at room temperature several gauche kinks will be present.
The macrocyclic rings were drawn in Scheme 1 in patterns that
are frequently encountered in molecular dynamics simulations,16
but the rings can adopt several other conformations as well.
Interactions between the two stations may occur in the
threads, and secondary interactions may exist between the
macrocyclic ring and the station which it does not encircle.
Such folded conformers have been detected in IR studies of 2
and its radical anion, but they were of relatively minor
importance.9 Moreover, the NMR shielding parameters are
sensitive primarily to the position of the p-xylylene rings with
respect to the thread protons. Therefore they report on the
position of the macrocycle on the thread, and are not
very sensitive to the other aspects of the three-dimensional
molecular structure. We believe that the co-conformational
equilibrium in 1 will be an attractive target for the development
of force fields and advanced simulation techniques that aim at
free energies with chemical accuracy.17
A van’t Hoff plot of ln K vs. Tꢁ1 is shown in Fig. 2. The
non-linearity of this plot (see also ESI,w Fig. S13) points to a
0 13
non-negligible heat capacity difference DCp . Inclusion of
this in the thermodynamic model leads to eqn (2), which
allows an excellent fit to the data, giving the parameters
DH0 = ꢁ18 ꢂ 1 kcal molꢁ1, DS0 = ꢁ300 ꢂ 30 cal molꢁ1 Kꢁ1
0
and DCp = 52 ꢂ 4 cal molꢁ1 Kꢁ1
.
DH0
T
R ln K ¼ ꢁ
þ DC0p ln T þ DS0 ꢁ DC0p
ð2Þ
From these parameters, the reaction enthalpy DH0, entropy
DS0 and Gibbs free energy DG0 at any temperature can be
calculated, as shown in Fig. 3. The Gibbs free energy for the
equilibrium is only slightly temperature dependent, increasing
from ꢁ1.1 kcal molꢁ1 at 268 K to ꢁ0.7 kcal molꢁ1 at 328 K,
but the contributions from enthalpy and entropy (ꢁTDS0) are
of opposite sign and change strongly with temperature.
Another possible cause of the non-linearity of the van’t Hoff
plot could be that more than two species are present in the
conformational equilibrium. At 298 K, the shielding of the
Fig. 2 Plot of ln K vs. Tꢁ1 for the equilibrium succ–ni in 1 in CDCl3.
Fit of the data according to eqn (2). The error bars were derived
assuming errors in the chemical shifts (eqn (1)) of 0.003 ppm.
Entropy–enthalpy compensation is typically associated with
weak interactions, because optimization of the stabilizing
enthalpic contribution requires some extent of freezing of
ꢀc
This journal is The Royal Society of Chemistry 2010
2062 | Chem. Commun., 2010, 46, 2061–2063

View Article Online
the conformational freedom.18,19 In the succ-co-conformer of
1, the macrocyclic ring must adopt a chair conformation and
the succinamide unit must be in a (planar) trans form for
hydrogen bonding to be optimal. In the ni-co-conformer any
of the CQO groups may point to the inside of the macrocyclic
ring, to make a hydrogen bond with the glycine N–H. One
amide group is free, allowing low-frequency movements.
Large DCp effects have been observed in supramolecular
complexation based on hydrophobic interactions.13,19,20 The
characteristic changes in the enthalpy, entropy, and heat
capacity that accompany protein (un)folding have been
attributed to hydrophobic effects as well, although hydrogen
bonding may also be important.21 Also for the folding of
nucleic acids, heat capacity effects are substantial.22 The role
of heat capacity in the conformational equilibrium of synthetic
B. Windisch, A. Mohry and F. Vogtle, J. prakt. Chem.
¨
(Weinheim, Ger.), 2000, 342, 437–444.
3 E. R. Kay, D. A. Leigh and F. Zerbetto, Angew. Chem., Int. Ed.,
2007, 46, 72–191.
4 V. Balzani, A. Credi, B. Ferrer, S. Silvi and M. Venturi, Top. Curr.
Chem., 2005, 262, 1–27; F. Durola and J. P. Sauvage, Angew.
Chem., Int. Ed., 2007, 46, 3537–3540; A. H. Flood, R. J. A.
Ramirez, W. Q. Deng, R. P. Muller, W. A. Goddard and
J. F. Stoddart, Aust. J. Chem., 2004, 57, 301–322; K. Hirose,
Y. Shiba, K. Ishibashi, Y. Doi and Y. Tobe, Chem.–Eur. J., 2008,
14, 3427–3433; H. Tian and Q. C. Wang, Chem. Soc. Rev., 2006,
35, 361–374; T. Umehara, H. Kawai, K. Fujiwara and T. Suzuki,
J. Am. Chem. Soc., 2008, 130, 13981–13988.
5 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., 2006, 12, 261–279.
6 D. H. Qu, Q. C. Wang and H. Tian, Mol. Cryst. Liq. Cryst., 2005,
430, 59–65.
7 A. Altieri, F. G. Gatti, E. R. Kay, D. A. Leigh, D. Martel,
F. Paolucci, A. M. Z. Slawin and J. K. Y. Wong, J. Am. Chem.
Soc., 2003, 125, 8644–8654.
8 A. M. Brouwer, C. Frochot, F. G. Gatti, D. A. Leigh, L. Mottier,
F. Paolucci, S. Roffia and G. W. H. Wurpel, Science, 2001, 291,
2124–2128.
9 D. C. Jagesar, F. Hartl, W. J. Buma and A. M. Brouwer,
Chem.–Eur. J., 2008, 14, 1935–1946.
10 Structures of compounds 3 and 5 are shown in the ESIw.
11 K. Hirose, in Analytical Methods in Supramolecular Chemistry,
ed. C. A. Schalley, Wiley-VCH, Weinheim, 2006.
12 A. S. Lane, D. A. Leigh and A. Murphy, J. Am. Chem. Soc., 1997,
119, 11092–11093.
13 D. A. Stauffer, R. E. Barrans and D. A. Dougherty, J. Org. Chem.,
1990, 55, 2762–2767.
14 M. H. Abraham and J. A. Platts, J. Org. Chem., 2001, 66,
3484–3491; C. Laurence, K. A. Brameld, J. Graton, J. Y. Le
Questel and E. Renault, J. Med. Chem., 2009, 52, 4073–4086.
molecular shuttles has never been investigated, as far as
0
we know. Our data analysis leads to a value of DCp
E
50 cal molꢁ1
K
ꢁ1. This is a relatively small magnitude
compared to the effects seen in protein chemistry,23 but of
similar magnitude as the effects in host–guest binding.13
Although in terms of free energy the binding of the
macrocyclic ring to the ni-gly station is only slightly less
favorable than the binding to the succ station, the latter is a
much better template for the synthesis of the macrocyclic ring
in the 5-component clipping method.24 The synthetic yield of
rotaxane 4 with only the ni-gly template was only 8%, much
smaller than that of 1 (30%). On the other hand, weaker
binding of the macrocycle to a ni-gly motif should allow faster
switching of molecular shuttles, on
a sub-microsecond
timescale,3,8 and work along these lines is now in progress.
This work was financially supported by the European
Union (Marie Curie Training Site ‘‘Molecular Photonic
Materials’’, contract number HPMT-CT-2001-00311), and
by NanoNed, a national nanotechnology program coordinated
by the Dutch Ministry of Economic Affairs.
15 J. Antony, B. Bruske and S. Grimme, Phys. Chem. Chem. Phys.,
¨
2009, 11, 8440–8447.
16 Preliminary results obtained with Macromodel 9.7, Stochastic
Dynamics, AMBER* force field, GBSA solvent model for CHCl3.
17 H. Meirovitch, Curr. Opin. Struct. Biol., 2007, 17, 181–186;
M. Cecchini, S. V. Krivov, M. Spichty and M. Karplus, J. Phys.
Chem. B, 2009, 113, 9728–9740.
18 H. J. Schneider, Angew. Chem., Int. Ed., 2009, 48, 3924–3977.
19 K. N. Houk, A. G. Leach, S. P. Kim and X. Y. Zhang, Angew.
Chem., Int. Ed., 2003, 42, 4872–4897.
20 M. V. Rekharsky and Y. Inoue, Chem. Rev., 1998, 98, 1875–1917.
21 W. D. Hoff, A. Xie, I. H. M. van Stokkum, X. J. Tang, J. Gural,
A. R. Kroon and K. J. Hellingwerf, Biochemistry, 1999, 38,
1009–1017; J. S. Khan, Y. Imamoto, M. Kataoka, F. Tokunaga
and M. Terazima, J. Am. Chem. Soc., 2006, 128, 1002–1008;
A. Cooper, Biophys. Chem., 2000, 85, 25–39.
22 P. J. Mikulecky and A. L. Feig, Biopolymers, 2006, 82, 38–58.
23 J. D. Fisk, M. A. Schmitt and S. H. Gellman, J. Am. Chem. Soc.,
2006, 128, 7148–7149.
Notes and references
1 D. H. Leung, R. G. Bergman and K. N. Raymond, J. Am. Chem.
Soc., 2008, 130, 2798–2805; G. R. Bishop, J. Ren, B. C. Polander,
B. D. Jeanfreau, J. O. Trent and J. B. Chaires, Biophys. Chem.,
2007, 126, 165–175; M. V. Rekharsky, T. Mori, C. Yang,
Y. H. Ko, N. Selvapalam, H. Kim, D. Sobransingh,
A. E. Kaifer, S. Liu, L. Isaacs, W. Chen, S. Moghaddam,
M. K. Gilson, K. Kim and Y. Inoue, Proc. Natl. Acad. Sci.
U. S. A., 2007, 104, 20737–20742.
2 C. A. Hunter, J. Am. Chem. Soc., 1992, 114, 5303–5311;
D. A. Leigh, P. J. Lusby, A. M. Z. Slawin and D. B. Walker,
Angew. Chem., Int. Ed., 2005, 44, 4557–4564; O. Safarowsky,
24 F. G. Gatti, D. A. Leigh, S. A. Nepogodiev, A. M. Z. Slawin,
S. J. Teat and J. K. Y. Wong, J. Am. Chem. Soc., 2001, 123,
5983–5989.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 2061–2063 | 2063