A 3-fold butterfly valve in command of the encapsulation's kinetic stability. Molecular baskets at work

Article abstract of DOI:10.1021/ja8041977

Molecular basket 1, composed of a semirigid tris-norbornadiene framework and three revolving pyridine-based gates at the rim, has been built to dynamically enclose space and as such regulate molecular encapsulation. The gates were shown to fold via intramolecular hydrogen bonding and thereby form a C_3v symmetrical receptor: the ¹H NMR resonance for the amide N-H protons of the pyridine gates appeared downfield (δ = 10.98 ppm), and the N-H vibrational stretch (IR) was observed at 3176 cm^-1. Accordingly, density functional theory (DFT, B3LYP) investigations revealed for the closed conformers of 1 to be energetically the most stable and dominant. The gearing of the pyridine gates , about their axis, led to the interconversion of two dynamic enantiomers 1_A and 1_B comprising the clockwise and counterclockwise seam of intramolecular hydrogen bonds. Dynamic ¹H NMR spectroscopic measurements and line-shape simulations suggested that the energy barrier of 10.0 kcal/mol (ΔG^?_A/B, 298 K) is required for the 1_A/B interconversion, when CCl₄ occupies the cavity of 1. Likewise, the activation free energy for CCl₄ departing the basket was found to be 13.1 kcal/mol (ΔG?, 298 K), whereas the thermodynamic stability of 1:CCl₄ complex was -2.7 kcal/mol (ΔG°, 298 K). In view of that, CCl₄ (but also (CH ₃)₃CBr) was proposed to escape from, and a molecule of solvent to enter, the basket when the gates rotate about their axis: the exit of CCl₄ requires the activation energy of 12.7 kcal/mol (ΔG ^?_A/B + ΔG°), similar to the experimentally found 13.1 kcal/mol (ΔG^?).

Full text of DOI:10.1021/ja8041977

Published on Web 10/21/2008
A 3-fold “Butterfly Valve” in Command of the Encapsulation’s
Kinetic Stability. Molecular Baskets at Work
Bao-Yu Wang, Xiaoguang Bao, Zhiqing Yan, Veselin Maslak,
Christopher M. Hadad, and Jovica D. Badjic´*
Department of Chemistry, The Ohio State UniVersity, 100 West 18th AVenue,
Columbus, Ohio 43210
Received June 10, 2008; E-mail: badjic@chemistry.ohio-state.edu
Abstract: Molecular basket 1, composed of a semirigid tris-norbornadiene framework and three revolving
pyridine-based gates at the rim, has been built to “dynamically” enclose space and as such regulate
molecular encapsulation. The gates were shown to fold via intramolecular hydrogen bonding and thereby
form a C3v symmetrical receptor: the 1H NMR resonance for the amide N-H protons of the pyridine gates
appeared downfield (δ ) 10.98 ppm), and the N-H vibrational stretch (IR) was observed at 3176 cm-1
.
Accordingly, density functional theory (DFT, B3LYP) investigations revealed for the closed conformers of
1 to be energetically the most stable and dominant. The gearing of the pyridine “gates”, about their axis,
led to the interconversion of two dynamic enantiomers 1A and 1B comprising the clockwise and
counterclockwise seam of intramolecular hydrogen bonds. Dynamic 1H NMR spectroscopic measurements
and line-shape simulations suggested that the energy barrier of 10.0 kcal/mol (∆GqA/B, 298 K) is required
for the 1A/B interconversion, when CCl4 occupies the cavity of 1. Likewise, the activation free energy for
CCl4 departing the basket was found to be 13.1 kcal/mol (∆Gq, 298 K), whereas the thermodynamic stability
of 1:CCl4 complex was -2.7 kcal/mol (∆G°, 298 K). In view of that, CCl4 (but also (CH3)3CBr) was proposed
to escape from, and a molecule of solvent to enter, the basket when the gates rotate about their axis: the
exit of CCl4 requires the activation energy of 12.7 kcal/mol (∆GqA/B + ∆G°), similar to the experimentally
found 13.1 kcal/mol (∆Gq).
Introduction
function of its size. The substrate binding rate for smaller guests
(<2.4 Å radius), thereby, becomes modestly reduced, whereas
The molecular trafficking in natural systems is governed via
programmed and dynamic chemical processes.1 Complex opera-
tions, such as selective transport of molecules2 and catalysis,3
are thus repeatedly executed in a precise manner, albeit
challenging to duplicate into functional artificial systems.4
Acetylcholinesterase, for instance, regulates its specificity
without sacrificing efficiency through a mechanism of confor-
mation gating:5 stochastically controlled entry, inside a deep
gorge leading to the enzyme active site, opens for a small
fraction of time to modulate the guest’s incoming rate as a
bulkier substrates (>2.8 Å radius) experience a considerable
decrease in the rate of access to the active site; the fluctuating
constriction of the entrance brings a required selectivity without
sacrificing the catalytic efficiency! In unnatural systems,6
however, conformationally dynamic groups at the skin of
hemicarcerands7 have afforded a measurable gain in kinetic
stability of the complexes via gating access to the host’s interior.
Notably, investigating novel ways for directing molecular
encapsulation8 and constrictive binding9 will provide insight into
understanding the molecular translocation in unnatural environ-
(1) Essays in Biochemistry: Molecular Trafficking; Bernstein, P., Ed.;
Portland Press: London, 2000.
(2) Pallotta, B. S. Ann. N.Y. Acad. Sci. 1997, 812, 133–140.
(3) Swift, R. V.; McCammon, J. A. Biochemistry 2008, 47, 4102–4111.
(4) For selected references, see: (a) Badjic´, J. D.; Ronconi, C. M.; Stoddart,
J. F.; Balzani, V.; Silvi, S.; Credi, A. J. Am. Chem. Soc. 2006, 128,
1489–1499. (b) Pei, R.; Taylor, S. K.; Stefanovic, D.; Rudchenko, S.;
Mitchell, T. E.; Stojanovic, M. N. J. Am. Chem. Soc. 2006, 128,
12693–12699. (c) Wintjes, N.; Bonifazi, D.; Cheng, F.; Kiebele, A.;
Stoehr, M.; Jung, T.; Spillmann, H.; Diederich, F. Angew. Chem., Int.
Ed. 2007, 46, 4089–4092. (d) Hidalgo, R. P.; Coumans, R. G. E.;
Deutman, A. B. C.; Smits, J. M. M.; De Gelder, R.; Elemans,
J. A. A. W.; Nolte, R. J. M.; Rowan, A. E. J. Am. Chem. Soc. 2007,
129, 5699–5702. (e) Percec, V.; Rudick, J. G.; Peterca, M.; Heiney,
P. A. J. Am. Chem. Soc. 2008, 130, 7503–7508. (f) Angelos, S.; Yang,
Y.-W.; Patel, K.; Stoddart, J. F.; Zink, J. I. Angew. Chem., Int. Ed.
2008, 47, 2222–2226. (g) Stock, C.; Heureux, N.; Browne, W. R.;
Feringa, B. L. Chem.-Eur. J. 2008, 14, 3146–3153.
(6) (a) Talukdar, P.; Bollot, G.; Mareda, J.; Sakai, N.; Matile, S. Chem.-
Eur. J. 2005, 11, 6525–6532. (b) Gokel, G. W.; Leevy, W. M.; Weber,
M. E. Macrocycl. Chem. 2005, 253–265.
(7) (a) Houk, K. N.; Nakamura, K.; Sheu, C.; Keating, A. E. Science 1996,
273, 627–629. (b) Sheu, C.; Houk, K. N. J. Am. Chem. Soc. 1996,
118, 8056–8070.
(8) For selected examples, see: (a) Rudkevich, D. M.; Hilmersson, G.;
Rebek, J., Jr. J. Am. Chem. Soc. 1998, 120, 12216–12225. (b)
Vysotsky, M. O.; Thondorf, I.; Bohmer, V. Angew. Chem., Int. Ed.
2000, 39, 1264–1267. (c) Zuccaccia, D.; Pirondini, L.; Pinalli, R.;
Dalcane, E.; Macchioni, A. J. Am. Chem. Soc. 2005, 127, 7025–7032.
(d) Hooley, R. J.; Anda, H. J. V.; Rebek, J., Jr. J. Am. Chem. Soc.
2006, 128, 3894–3895. (e) Mukhopadhyay, P.; Zavalij, P. Y.; Isaacs,
L. J. Am. Chem. Soc. 2006, 128, 14093–14102. (f) Gottschalk, T.;
Jaun, B.; Diederich, F. Angew. Chem., Int. Ed. 2007, 46, 260–264.
(g) Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Angew. Chem.,
Int. Ed. 2007, 46, 8587–8589.
(5) Zhou, H.-X.; Wlodek, S. T.; McCammon, J. A. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 9280–9283.
(9) Liu, Y.; Warmuth, R. Org. Lett. 2007, 9, 2883–2886.
9
10.1021/ja8041977 CCC: $40.75
2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 15127–15133 15127

A R T I C L E S
Wang et al.
Figure 1. Chemical structures and synthesis of molecular basket 1 and model compound 2. Top and side views of energy-minimized (DFT, B3LYP)
structure of folded 1.
ments. Furthermore, these endeavors will facilitate the rational
design of newly formulated catalysts and molecular devices.10
In that vein, the present study has been directed toward
developing a functional and modular abiotic receptor, molecular
basket, for regulating the kinetics (kon/off) of molecular encap-
sulation. The ambiguity of synchronizing the host dynamics with
the course of the guest exchange11 has been addressed in this
study by developing an approach so that one can learn about
controlling the gating of molecules in artificial systems. Mo-
lecular baskets (Figure 1) were originally made12 to incorporate
a concave-shaped C3V symmetrical tris-norbornadiene frame-
work. Three aromatic rings, containing transition-metal chelat-
ing13 or hydrogen-bonding12a sites, have been installed at the
rim to act as “gates” in: (a) controlling the basket’s folding, (b)
the formation of its “dynamic” interior, and (c) the in/out
trafficking of the guest(s). The conformational dynamics of the
hydrogen-bonded host12a was, in particular, revealed to have
an adverse effect on the thermodynamics of the guest encap-
sulation. A notion that more “restricted” gates that hold to each
other at two “points” could be incorporated in the basket’s
design was exercised in constructing compound 1 (Figure 1).
This receptor was therefore envisaged to contain three pyridine
gates, each with an acetylated amide unit, predisposed to form
intramolecular hydrogen bonds (HBs) and thereby physically
enfold space (Figure 1). In particular, learning about the
relationship between the folding characteristics of the basket,
its conformational dynamics (opening/closing), and the guest
exchange was of great interest. The present study, for the first
time, reveals such a dependence and, concomitantly, offers these
dynamic receptors as convenient models for investigating the
fundamentals of molecular trafficking.
Results and Discussion
Synthesis. Molecular basket 1 was obtained (63% yield) by
condensation of tris-anhydride 3 with acetylated (aminometh-
yl)pyridine 4 (Figure 1). The reaction was completed in dimethyl
sulfoxide, when promoted with a catalytic amount of pyridine.
Likewise, model compound 2 was synthesized from phthalic
anhydride (Figure 1).
1H NMR Spectroscopic Studies - Intramolecular versus
1
Intermolecular Contacts. The H NMR spectrum (400 MHz,
298 K) of 1 in CDCl3 has been assigned to a C3 symmetrical
molecule (Figure 2B). The signal for the N-H proton appeared
downfield (δ ) 10.98 ppm) and stayed constant as the CDCl3
(10) (a) Davis, A. V.; Yeh, R. M.; Raymond, K. N. Proc. Natl. Acad. Sci.
U.S.A. 2002, 99, 4793–4796. (b) Vriezema, D. M.; Aragones, M. C.;
Elemans, J. A. A. W.; Cornelissen, J. J. L. M.; Rowan, A. E.; Nolte,
R. J. M. Chem. ReV. 2005, 105, 1445–1489. (c) Yoshizawa, M.;
Tamura, M.; Fujita, M. Science 2006, 312, 251–254. (d) Yu, J.;
RajanBabu, T. V.; Parquette, J. R. J. Am. Chem. Soc. 2008, 130, 7845–
7847.
1
solution was diluted from 20 to 0.5 mM. Respectively, the H
NMR resonance for the N-H proton in model compound 2
was found at a higher field (δ ) 7.30 ppm, Figure 2A). The
resonance underwent an upfield shift (∆δ ) 0.3 ppm) with
solvent dilution from 13.0 to 0.1 mM. The spectroscopic results,
unequivocally, support the occurrence of stable intramolecular
(11) See, for instance: Palmer, L. C.; Rebek, J., Jr. Org. Biomol. Chem.
2004, 2, 3051–3059.
(12) (a) Maslak, V.; Yan, Z.; Xia, S.; Gallucci, J.; Hadad, C. M.; Badjic´,
J. D. J. Am. Chem. Soc. 2006, 128, 5887–5894. (b) Yan, Z.;
McCracken, T.; Xia, S.; Maslak, V.; Gallucci, J.; Hadad, C. M.; Badjic´,
J. D. J. Org. Chem. 2008, 73, 355–363.
1
HBs in 1. The observed H NMR responsiveness of the N-H
resonance in model 2 with the dilution is, however, in agreement
with this compound’s propensity toward forming intermolecular
HBs. Furthermore, the temperature-dependent 1H NMR chemi-
cal shift coefficient, ∆δ/∆T, for the N-H resonance in 1, was
(13) (a) Yan, Z.; Xia, S.; Gardlik, M.; Seo, W.; Maslak, V.; Gallucci, J.;
Hadad, C. M.; Badjic´, J. D. Org. Lett. 2007, 9, 2301–2304. (b) Rieth,
S.; Yan, Z.; Xia, S.; Gardlik, M.; Chow, A.; Fraenkel, G.; Hadad,
C. M.; Badjic´, J. D. J. Org. Chem. 2008, 73, 5100–5109.
9
15128 J. AM. CHEM. SOC. VOL. 130, NO. 45, 2008

Molecular Baskets
A R T I C L E S
Figure 2. 1H NMR spectra (400 MHz, 298 K) of (A) model compound 2 (13.0 mM) and (B) molecular basket 1 (1.87 mM) in CDCl3. Insets: segments of
1H NMR spectra showing the N-H resonance shift of 2 (top) and 1 (bottom), recorded in dilution experiments.
found to be 6.1 ppb/K (298 to 197 K).14 The low value is in
Table 1. Computed Relative Thermodynamic Parameters (DFT,
B3LYP) for the Stereoisomers of 1 (kcal/mol), and Their
agreement with the absence of complex equilibria and the N-H
protons residing in alternative environments at variable tem-
peratures. Molecular basket 1 was also shown to stay predomi-
nantly monomeric and incapable of intermolecular aggregation
in solution. 1H NMR DOSY measurements (500 MHz, 298 K,
CDCl3) of 1 revealed that the apparent diffusion coefficients
Dapp are practically invariant (5.6 and 5.9 × 10-10 m2 s-1) at
concentrations of 13.0 and 0.7 mM, respectively.14 Additionally,
MALDI mass spectrometry measurements of 1 revealed only
the monomeric form.14
Boltzmann Distribution (Population) in the Gas Phase (298 K)
B3LYP/6-31G(d)a
B3LYP/6-311+G(d,p)a
molecular basket 1
∆H°
∆G°b
∆H°
∆G°b
population (%)
[Z3]
0
0
0
0
97
1
0
0
2
[Z2E]
[ZE2]
[E3]
3.56
6.61
9.10
3.97
3.41
7.57
11.52
2.64
4.03
7.51
10.54
3.97
3.92
8.45
12.96
2.65
[Z3-HB]
a In kcal/mol, using the optimized B3LYP/6-31G(d) geometry.
b Corrected for the statistical degeneracy.
Conformational Studies: Infrared Spectroscopy and Theo-
retical Calculations. The design algorithm incorporated in the
structure of 1 was to allow for the exclusive formation of a
seam of intramolecular N-H---N HBs (Figure 1) on the rim of
the basket. The desired outcome required additional scrutiny
of the observed 1H NMR spectra, illustrating the intramolecular
interactions of 1 in detail. The FT-IR spectrum of model
compound 2 revealed a sharp band at 3431 cm-1 (CHCl3),
corresponding to the N-H stretching vibration that is free of
intermolecular interactions.14 Molecular basket 1 (CDCl3),
however, showed the corresponding vibration14 as a concentra-
stable (Table 1), with other structures ([Z2E], [ZE2], [E3])
populating the equilibrium to a much lesser degree (Table 1).14
Importantly, further theoretical studies revealed the existence
of a conformer of 1 (Figure 1), lacking one hydrogen bond ([Z3-
HB], Table 1),14 and this conformer was computed to populate
the equilibrium to about 2% (Table 1). The IR result can thus
be qualitatively put in agreement with the basket being folded
via the engineered N-H---N seam of hydrogen bonds (Figure
1) at the rim of the basket, albeit with a population distribution
indicating an existence of other isomeric structures.17
tion-independent (20.0 to 1.0 mM) broadband at 3176 cm-1
,
Conformational Stereoisomerism: 2D and Dynamic 1H
in good agreement with strong intramolecular HBs. In the same
region of the IR spectrum of 1 (3100-3150 cm-1), however,
there were apparent satellite signals14 suggesting a potential
conformational distribution.
NMR Spectroscopic Measurements. The results of two-
dimensional 1H-1H ROESY spectroscopic measurements of 1
gave further support to the projected folding pattern and
indicated the existence of a dynamic equilibrium 1A/1B (Figure
3).14 The magnetization transfer (ROE) through space was
detected for Ha, but not Hc, and N-H resonances of the gates
(Figure 3A). The result indicated a restricted rotation of the
Secondary amides, such as 1, can exhibit E/Z isomerization,15
leading to different intramolecular HB patterns. Energy mini-
mizations (density functional theory, B3LYP)16 of 1 demon-
strated the folded [Z3] isomer as thermodynamically the most
(14) See Supporting Information for more details.
(15) Ilieva, S.; Hadjieva, B.; Galabov, B. J. Mol. Spectrosc. 1999, 508,
73–80.
(17) Presently, we are investigating the conformational characteristics of
1 using various theoretical models for the solvation. These procedures
will afford a more “realistic” conformational energy landscape and,
perhaps, allow us to reproduce the experimentally observed infrared
spectra.
(16) (a) Perdew, J. P. Phys. ReV. B 1986, 33, 8822. (b) Becke, A. D. Phys.
ReV. A 1998, 38, 3098.
9
J. AM. CHEM. SOC. VOL. 130, NO. 45, 2008 15129

A R T I C L E S
Wang et al.
Figure 3. (A) Clockwise and counterclockwise arrangements of the seam of hydrogen bonds in dynamic enantiomers 1A and 1B, respectively. (B) Two
conformers of 2 obtained by 180° rotation of the appended amide about the Ar-NH σ bond. The observed ROE and NOE signals in 1 and 2 (CDCl3, 400
MHz, 300 K), respectively, are shown in red.
appended amide(s) about the Ar-NH bond(s) and is a corollary
of the constraint imposed by the seam of the intramolecular
N-H---N HBs in 1 (Figure 3A). On the contrary, model
compound 2 cannot engage in such intramolecular interactions,
and the rotation about its Ar-NH bond was indeed found to be
unhindered: the NOE cross peaks were detected for both Ha/c
and the N-H (Figure 3B)!
The projected seam of HBs in 1 is dynamic and displayed in
either clockwise (+, 1A) or counterclockwise (-, 1B) direction
of the N-H---N connections (Figure 3A). The gearing of the
pyridine “gates”, about their axis, should thus permit the
interconversion of the dynamic enantiomers 1A and 1B. This
1
effect, as monitored by H NMR spectroscopy, would lead to
the chemical exchange of the “hinge” Hd/e protons; the exchange
process would distinguish these protons as being diastereotopic
at low and enantiotopic at high temperatures. Indeed, the
1
dynamic H NMR studies of 1 revealed the Hd/e resonance(s)
appear as a four-line AB-type quartet at low and a singlet at
high temperatures (Figure 4), verifying the 1A/B equilibrium!
The Inner Phase Solvation and the Host’s Dynamics. Inter-
estingly, the Hd/e signal decoalescence was observed in CDCl3
at a higher temperature (∼225 K), while in CD2Cl2 at a
considerably lower temperature (<183 K).14 At first, this came
as a surprise because the conformational 1A/B interconversion
must entail the rupture of the intramolecular HBs, for which
the stability should be almost invariant in these two, noncoor-
dinating and low-polarity, solvents.18 Accordingly, a perception
that more sizable CHCl3 (75 Å3) fits better inside of 1 (∼221
Å3)14 than smaller CD2Cl2 (61 Å3) arose. In fact, the compu-
tational studies (PM3, CAChe) suggested that the egress of
CH2Cl2, along a reaction coordinate directed through the center
of an aperture of the folded host,7 necessitates a lower activation
energy (∼7 kcal/mol) than the expulsion of CHCl3 (∼14 kcal/
mol).14 Dichloromethane was, for that reason, concluded to be
Figure 4. Simulated and experimental (WINDNMR-Pro)23 resonances for
the Hd/e protons in 1 (CDCl3, 1.17 mM). Apparent first-order rate constants
k1/k-1 for the 1A/1B interconversion at the examined temperatures.
less competitive for occupying the inner space of the basket,9
and hence more suitable for investigating its host-like behav-
ior!19
Recognition Studies: The Thermodynamics of CCl4 and
CHCl3 Entrapment. A gradual addition of CCl4 to a solution
of 1 (2.98 mM) in CD2Cl2 caused considerable changes in its
1H NMR spectrum.14 The N-H resonance shifted to a lower
field (∆δ ) 0.5 ppm), as the concentration of CCl4 was
gradually increased. The nonlinear curve fitting of the binding
isotherm,20 to a 1:1 stoichiometric model, yielded an apparent
298
association constant of Ka ) 109 ( 1 M-1 14
In a similar
.
titration experiment, chloroform exhibited a lower affinity
toward 1: the downfield shift of the N-H resonance was
noticeable, yet insufficient to quantify the binding.14 Variable-
(18) (a) Taft, R. W.; Gurka, D.; Joris, L.; Schleyer, P. v. R.; Rashys, J. W.
J. Am. Chem. Soc. 1969, 91, 4801–4808. (b) Beeson, C.; Pham, N.;
Shipps, G.; Dix, T. A. J. Am. Chem. Soc. 1993, 115, 6803–6812.
(19) Chapman, K. T.; Still, W. C. J. Am. Chem. Soc. 1989, 111, 3075–
3077.
(20) Mecozzi, S.; Rebek, J., Jr. Chem.-Eur. J. 1998, 4, 1016–1022.
9
15130 J. AM. CHEM. SOC. VOL. 130, NO. 45, 2008

Molecular Baskets
A R T I C L E S
1
Figure 5. (A) Selected regions of variable-temperature (VT) H NMR spectra (400 MHz, CD2Cl2) of a solution of 1 (2.4 mM) and CCl4 (3.7 mM). (B) A
selected region of VT 13C NMR spectra (100 MHz) of a solution of 1 (13.1 mM) and 13CCl4 (19.6 mM). (C) van’t Hoff plot and thermodynamic parameters
for the 1(solvent)/1(CCl4) equilibrium.
Table 2. Activation Parameters for the 1A/1B Interconversion in
temperature (VT) 1H NMR spectra (400 MHz) of a solution of
CD2Cl2, CDCl3, and CD2Cl2 Containing 50.0 mol equiv of CCl4
(Eyring Plots Were Used To Generate the Data)14
1 (2.4 mM, CD2Cl2), containing CCl4 (3.7 mM), revealed
decoalescence of signals at ca. 260 K (Figure 5A). Two sets of
peaks developed, with the intensity ratio changing with tem-
perature. Each group of signals has been assigned to a C3
symmetrical basket: one occupied with the solvent (1(solvent)) and
another with the added guest (1(CCl4), Figure 5C). Indeed, when
13CCl4 was employed in the experiment (Figure 5B), the
decoalescence in VT 13C NMR spectra was observed at
practically the same temperature (ca. 260 K). The resolved
resonances at 96.0 and 93.0 ppm corresponded to free and
encapsulated molecules of 13CCl4, respectively. The integration
of the N-H signals (Figure 5A), at each temperature, afforded
association constants and thermodynamic parameters for the
conversion of 1(solvent) into 1(CCl4) (Figure 5C). In essence, the
incorporation of carbon tetrachloride into the basket is driven
by enthalpy (∆H° ) -3.1 ( 0.2 kcal/mol). This, for the most
part, can be viewed as a result of the favorable host/guest van
der Waals contacts that CCl4 affords on the account of its size
(Table 1);20 in fact, the apparent lower tenacity of smaller
CH2Cl2 for occupying 1 can be justified on the same basis.20
Along with this guest occupation, there is concomitant change
of the intramolecular HBs of the occupied host, as indicated by
the observed lower-field 1H NMR N-H signals of 1(CCl4) versus
solvent system
ET(30)
volumea (Å3)
∆Hq (kcal/mol)
∆S‡ (cal/mol·K)
CD2Cl2
41
35
41
61
75
89
N/A
11.5 ( 0.2
10.4 ( 0.3
N/A
3.9 ( 0.1
1.4 ( 0.1
b
CDCl3
c
CD2Cl2/CCl4
a Calculated with Spartan Software. b T
180-223 K.
) )
202-240 K. c T
spins (Figure 4), was subjected to a total band-shape analysis22
to get the interconversion apparent first-order rate constants k1/
k-1 (Figure 4).23 Subsequently, the activation parameters for
the interconversion in (a) neat chloroform with CHCl3, and (b)
CH2Cl2 with CCl4 as guests, were extracted from Eyring plots
(Table 2).14 Remarkably, the rates for the basket opening/closing
(1A/1B) were nearly identical when bigger molecules CHCl3 (75
Å3) or CCl4 (89 Å3) occupied its cavity (221 Å3). The smaller
CH2Cl2 (61 Å3), however, did not “impose” on the basket’s
1
conformational dynamics (Table 2), as the H NMR spectro-
scopic results were invariant with temperature (T ) 179-298
(21) (a) Wilcox, C. In Frontiers in Supramolecular Organic Chemistry and
Photochemistry; Schneider, H.-J., Durr, H., Eds.; Wiley-VCH: Wein-
heim, 1991; pp 123-143. (b) Connors, K. A. Binding Constants, The
Measurement of Molecular Complex Stability; Wiley-Interscience: New
York, 1987; pp 24-28.
(22) (a) Kaplan, J. J.; Fraenkel, G. NMR of Chemically Exchanging Systems;
Academic Press: New York, 1980. (b) Pons, M.; Millet, O. Prog. Nucl.
Magn. Reson. Spectrosc. 2001, 38, 267–324. (c) Bain, A. D.; Rex,
D. M.; Smith, R. N. Magn. Reson. Chem. 2001, 39, 122–126. (d) Bain,
A. D. Prog. Nucl. Magn. Reson. Spectrosc. 2003, 43, 63–103.
(23) Reich, H. J. WinDNMR: Dynamic NMR Spectra for Windows; J. Chem.
Ed.: Software, Series D, 1996.
1(solvent) (Figure 5A). The observed loss in entropy (∆S° ) -1.2
( 0.8 eu) is negligible but, to a first approximation, congruent
with the exchange of a solvent molecule, CD2Cl2, with a
molecule of CCl4.
Dynamic 1H NMR Studies: The Kinetics of Molecular
Folding and Translocation. The basket 1A/1B conformational
equilibrium (Figure 3), manifested in the exchange of the Hc/d
9
J. AM. CHEM. SOC. VOL. 130, NO. 45, 2008 15131

A R T I C L E S
Wang et al.
Figure 6. Eyring plot and the activation parameters for CCl4 departing 1(CCl4), obtained from dynamic 1H NMR measurements in CD2Cl2.14
Figure 7. Energy diagram showing the partitioned standard (∆G°) and activation free energies (∆Gq), at 298 K, for synchronized gated transfer of CCl4
in/out of molecular basket 1.
K).14 As stated earlier, dichloromethane slippage24 through the
aperture of 1 is energetically easier (∼7 kcal/mol) and, evidently,
is not correlated with the motion of the gates. In contrast, bigger
guests (CHCl3 and CCl4) require more room to traffic into or
out of the host 1, thus imposing on the 1A/1B interconversion.
Notably, this observation points to the correlation between the
guest exchange and the gates’ rotary motion and needs to be
taken into account for elucidating the operational mechanism
of the basket (see below).
We, furthermore, examined the apparent rate constant kb for
CCl4 departing 1(CCl4) at various temperatures (Figure 6), by
simulating the line shapes of the two exchanging and unequally
populated N-H spins21-23 (Figure 5A): one corresponding to
1(solvent) and another to 1(CCl4) basket. From the Eyring plot
(Figure 6), the activation enthalpy for CCl4 leaving the basket
was found to be positive (∆Hq ) 8.6 ( 0.5 kcal/mol), while
the activation entropy was negative (∆Sq ) -15.0 ( 0.8 eu).
On the Encapsulation Mechanism. With the acquired kinetic
and thermodynamic information (Figure 7), the following
question was posed: does the opening/closing of the basket,
expressed via 1A/1B interconversion, restrict (regulate) the
kinetics of the egress/ingress of CCl4 molecules? Carbon
tetrachloride is, in the context of 1, a sizable molecule. Its
slippage through the aperture of the folded basket demands a
high activation energy (>14 kcal/mol, PM3);14 importantly, the
calculation does not treat a more realistic guest swapping that
requires a greater expansion of the host. Alternatively, the escape
of the guest from the basket could be occurring by way of an
opening that forms when the gates rotate about their axis (1A/
1B, Figure 3). In fact, the revolving of gates takes a portion of
the host’s inner space occupied by CCl4, presumably forcing
the guest’s departure to be correlated with motion of the host.
Moreover, the rotary motion of the gates would not only be
expected to guide the guest out but also to create two additional
apertures for an incoming molecule to enter the host with almost
no additional expense in energy. The conformational 1A/1B
(24) See, for instance: Raymo, F. M.; Houk, K. N.; Stoddart, J. F. J. Am.
Chem. Soc. 1998, 120, 9318–9322.
9
15132 J. AM. CHEM. SOC. VOL. 130, NO. 45, 2008

Molecular Baskets
A R T I C L E S
interconversion bears 10.0 kcal/mol of the activation energy
(∆Gq) at 298 K (Table 2). Furthermore, a molecule of CCl4 is
occupying the host with the complexation free energy (∆G°)
of 2.7 kcal/mol at 298 K (Figure 5C). When the above two
numerals are added (∆Gq + ∆G°), an approximate activation
energy for the exit of CCl4 is expected on the order of 12.7
kcal/mol (Figure 7).7 Remarkably, the experimental activation
free energy (∆Gq) for CCl4 departing the basket was found to
be 13.1 kcal/mol at 298 K (Figure 6). In fact, when the kinetics
and thermodynamics for the entrapment of bigger (CH3)3CBr
(107 Å3) were examined, the energetics appeared to be additive
as well (Figures S19-S24, ∆Gq(experimental) ) 11.8 kcal/mol vs
∆Gq(calculated) ) 12.4 kcal/mol).14 The evaluated free energies
(∆Gq(experimental) and ∆Gq(calculated)) indeed alter with temperature,
but the proposed additive relationship25 holds reasonably well
at the investigated 200-300 K. The apparent absence of the
additivity for the corresponding enthalpies and entropies,
however, is indicative of the enthalpy/entropy compensation.26
Likewise, the uncertainty in the experimental enthalpy and
entropy27,28 perhaps contributes to the apparent discrepancy. One
can additionally anticipate that the exchange of CCl4/(CH3)3CBr
(or any other guest) may be dominated by the energetics being
additive, provided that the guest occupies a sufficiently small
portion of the host’s cavity. Rebek and co-workers20 have
suggested that 55% of the inner space is an upper bound. We
compute that CCl4 occupies 40% while (CH3)3CBr 48% of the
host inner space, thus leading to the observed additive energet-
ics.29 A correspondence between the estimated and the experi-
mental activation barriers (Figure 7) suggests a synchronized
gating mechanism, operating in the action of 1! The gates must
revolve, each about its axis, to form a corridor for molecules to
exchange in a way akin to the operation of a butterfly valve in
regulating the flow of liquids. Importantly, the rotation of the
gates may sweep a part of the basket’s inner space, occupied
by a guest, to impose on the exchange process.30 The discovery
indicates a new mechanism31 for regulating the encapsulation’s
kinetic stability and adds to the scope of the french/sliding door
formalism6 and other known pathways31 for the exchange of
guests to and from synthetic hosts.
Conclusions
Maneuvering the kinetic stability of a host-guest complex
necessitates an ability to synchronize multiple dynamic structural
elements within the host. This study exemplifies the design of
a molecular basket capable of regulating the trafficking of guests
using a set of functionalized aromatics - gates. The gates
assemble via intramolecular hydrogen bonding to enfold space.
The rotary motion of the gates, each about its longitudinal axis,
forms a dynamic corridor to impose on guests to exchange to
and from the interior. Such a new mechanism for the dynamic
regulation of molecular encapsulation creates a unique op-
portunity to explore the interdependence of kinetic stability and
chemical reactivity, as well as the controlled translocation of
molecules in artificial environments.
Acknowledgment. We thank Professor Gideon Fraenkel of the
Ohio State University for useful suggestions. This work was
financially supported with funds obtained from the Ohio State
University, and the National Science Foundation under CHE-
0716355. Generous computational resources were provided by the
Ohio Supercomputer Center.
Supporting Information Available: Additional experimental
and computational results. This material is available free of
charge via the Internet at http://pubs.acs.org.
JA8041977
(31) Other scenarios acting in the exchange of guests cannot be ignored;
for example, another mechanistic possibility is the rotation of one
pyridyl arm versus the motion of all three arms. Each postulated
mechanism may be a function of the guest characteristics and will
certainly benefit from additional experimental scrutiny. In this study,
however, IR spectroscopy showed that the N-H hydrogen-bonding
pattern was dominated by intramolecular HBs, and little evidence was
seen for a free pyridyl arm. Calculations are in support of this
observation. Moreover, preliminary results of our molecular dynamics
(MD) and steered molecular dynamics simulations with the Amber
suite of programs indicate the rupture of all three N-H---N hydrogen
bonds occurs in the removal of CCl4 from the interior of 1.
Experimental and computational studies are in progress to reveal
mechanistic details. See also: Pluth, M. D.; Raymond, K. N. Chem.
Soc. ReV. 2007, 36, 161–171.
(25) Dill, K. A. J. Biol. Chem. 1997, 272, 701–704.
(26) Liu, L.; Guo, Q.-X. Chem. ReV. 2001, 101, 673–695.
(27) Exner, O. Nature 1970, 227, 366–367.
(28) (a) Alper, J. S.; Gelb, R. I. J. Phys. Org. Chem. 1993, 6, 273–280. (b)
Gelb, R. I.; Alper, J. S. J. Phys. Org. Chem. 1995, 8, 825–832.
(29) The hypothesis is currently under investigation with differently sized
guests, and the results will be reported in due course.
(30) (a) Jiang, X.; Bollinger, J. C.; Lee, D. J. Am. Chem. Soc. 2006, 128,
11732–11733. (b) Riddle, J. A.; Jiang, X.; Huffman, J.; Lee, D. Angew.
Chem., Int. Ed. 2007, 46, 7019–7022. (c) Riddle, J. A.; Jiang, X.;
Lee, D. Analyst 2008, 133, 417–422.
9
J. AM. CHEM. SOC. VOL. 130, NO. 45, 2008 15133