Stabilization of cucurbituril/guest assemblies via long-range coulombic and ch···o interactions

Article abstract of DOI:10.1021/ja4092165

Cucurbit[n]urils (CB[n], n = 6-8) interact strongly with metal-bound 4-substituted terpyridine ligands (M = Fe(II) and Ir(III)) via CH···O hydrogen bonding, despite significant separation between the positive metallic cation and the carbonylated rim of CB

Full text of DOI:10.1021/ja4092165

Article
pubs.acs.org/JACS
Stabilization of Cucurbituril/Guest Assemblies via Long-Range
Coulombic and CH···O Interactions
Roymon Joseph,† Anna Nkrumah,† Ronald J. Clark,‡ and Eric Masson*,†
†Department of Chemistry and Biochemistry, Ohio University, Athens, Ohio 45701, United States
‡Department of Chemistry and Biochemistry, Florida State University, Tallahassee, Florida 32306, United States
S
* Supporting Information
ABSTRACT: Cucurbit[n]urils (CB[n], n = 6−8) interact
strongly with metal-bound 4′-substituted terpyridine ligands
(M = Fe(II) and Ir(III)) via CH···O hydrogen bonding,
despite significant separation between the positive metallic
cation and the carbonylated rim of CB[n], and the location of
the latter in the second coordination sphere of the metal ion.
While water has been shown to mediate interactions between
cations and CB[n]s in some assemblies, mediation by organic ligands is unprecedented. The recognition process is driven by the
contrasted combination of extremely favorable binding enthalpies (up to 20.2 kcal/mol) and very unfavorable entropic
components (as low as −10.2 kcal/mol). Dynamic oligomers were prepared in the presence of CB[8], which acts as a “soft”,
noncovalent linker between metal/terpyridine complexes, and interconnects two 4′-substituents inside its cavity. Social self-
sorting between CB[8] and metal/terpyridine complexes bearing 4′-(2-naphthyl) and 4′-(2,3,5,6-tetrafluorophenyl) substituents
was also observed, and could afford well-organized oligomers with alternating Fe(II) and Ir(III) cations.
Coulombic interactions between positive units and the
portals of CB[n]s can be caused by direct contact between a
metallic or organic cation and the carbonyl laced portal of the
macrocycle,5 or by water mediation between the two units as it
is often the case with transition metals and clusters in the solid
state.5d,6 Here, we report for the first time that Coulombic
interactions between a metal and CB[n]s can be mediated by
an organic ligand, and we show that even in the second
coordination sphere, CB[n]s form remarkably strong CH···O
hydrogen bonds with the ligand surrounding the metallic core.
INTRODUCTION
■
Some three decades ago, Mock and co-workers showed that
Cucurbit[6]uril (CB[6])1 formed tight complexes with
positively charged amphiphilic guests. Complex formation was
attributed to the interaction between the positive charges and
the carbonylated portals of CB[6], and between the hydro-
phobic moiety of the guests and the hydrophobic cavity of the
macrocycle. While this assessment remains generally valid, the
pioneering work by Kaifer, Isaacs, Gilson, Kim and Inoue,2 as
well as Nau and co-workers3 afforded a much more detailed
and subtle description of the recognition process. Briefly, the
extreme binding affinities often observed with CB[n]s are
caused by (1) the ability of the guests, in particular positively
charged ones, to return as many hydration water molecules as
possible to the bulk upon binding (a process that is both
enthalpically and entropically favorable); (2) as mentioned
above, favorable ion-dipole interactions between positively
charged substituents and the CB[n] rims, and other Coulombic
contributions such as hydrogen bonding; however, the
magnitude of the interaction is severely hampered by a loss
of solvation upon encapsulation; (3) the rigidity of the
macrocycles and the limited degrees of rotational freedom of
some guests (an entropic advantage); and (4) van der Waals
interactions between the inner wall of CB[n]s and the surface
of the guests; the magnitude of this contribution is a subject of
controversy: while attractive dispersive interactions reach a
maximum when the separation between two atoms is
approximately 3 Å4 (a distance that is likely to be present
between several pairs of host and guest atoms), Nau showed
that the cavity of CB[7] is overall highly unpolarizable.3
RESULTS AND DISCUSSION
■
A set of terpyridine ligands 1−3, bearing 4-tolyl, 2-naphthyl,
and 2,3,5,6-tetrafluorophenyl substituents at position 4′,
respectively, were prepared, followed by their complexes with
Fe(II) and Ir(III) cations (chloride and nitrate as their
counteranions, respectively; see Scheme in Figure 1).7 The
interaction of the six metal/ligand complexes with CB[7] was
1
then monitored by H nuclear magnetic resonance spectrosco-
py (NMR). The 4-tolyl group of complexes [Fe·12]2+ and [Ir·
12]3+ was found to sit inside the cavity of the macrocycle, with
strong upfield shifts of hydrogen nuclei located at positions 2
and 3 of the 4-tolyl unit (up to 1.04 ppm; noted 7 and 8 in
Figure 1), and of the methyl hydrogens at its 4-position (up to
0.52 ppm) being measured upon complexation. The
encapsulation of the 2-naphthyl and 2,3,5,6-tetrafluorophenyl
groups in complexes [Fe·22]2+, [Fe·32]2+, [Ir·22]3+, and [Ir·32]3+
was also confirmed by upfield shifts of the relevant units. As
Received: September 12, 2013
Published: April 16, 2014
© 2014 American Chemical Society
6602
dx.doi.org/10.1021/ja4092165 | J. Am. Chem. Soc. 2014, 136, 6602−6607

Journal of the American Chemical Society
Article
shifts of the CH donor hydrogens are indeed symptomatic of
such interactions.8 Chemical shifts observed upon interaction of
complex [Ir·12]3+ with CB[6] are also consistent with this
statement. A 0.82 ppm downfield shift was observed for H(3′)
hydrogens this time (see Figure 1, spectrum d), while H(3) and
H(3″) nuclei underwent weaker shifts than when sitting at the
portal of CB[7] (0.57 vs 0.95 ppm). Hydrogen bonds between
H(3′) atoms and the closest CB[6] oxygens are consistent with
a shorter and tighter penetration of the tolyl head of the metal−
ligand complex into CB[6] due to the reduced diameter of the
latter.
No CB[n]-containing assembly afforded crystals suitable for
quality X-ray diffraction. To circumvent this problem, we
carried out two-dimensional nuclear Overhauser effect spec-
troscopy (NOESY) experiments using assembly [Ir·12]3+·
(CB[7])2. This technique has been used on several occasions
to quantify through-space distances in solution, by gradually
increasing mixing times and monitoring the growth of the cross
peaks attributed to the interactions between hydrogen nuclei.9
In this case, we monitored the interaction of hydrogens at
position 3 and 3″ of the terpyridine ligands with the methylene
hydrogens of CB[7] that point toward the metallic core. Also in
parallel, we optimized the structure of assembly [Ir·12]3+·
(CB[7])2 at the TPSS-D3(BJ)/def2-SVP level4a,10 with the
COSMO solvation model11 (see Figure 2; our choice of
Figure 2. Optimized structure of complex [Ir·12]3+·(CB[7])2,
calculated at the TPSS-D3(BJ)/def2-SVP level with the COSMO
solvation model. Distances between host and guest atoms are given in
angstroms (Å).
Figure 1. Preparation of metal−ligand complexes. 1H NMR spectra of
complex [Fe·12]2+ (a) in the absence of CB[7], and after addition of
1
(b) 1.0 equiv and (c) 2.0 equiv of CB[7]; H NMR spectra of (d)
assembly [Ir·12]3+·(CB[6])2, (e) free metal−ligand complex [Ir·12]3+,
and (f) its inclusion complex with two CB[7] macrocycles.
functional was based on its high accuracy in the treatment of
both transition metal complexes12 and supramolecular
systems).4b The r−6-weighted average distance (see Supporting
Information for details) between H(3) or H(3″) atoms and the
hydrogens at the CB[7] portal was found to be 3.92 ( 0.02) Å
when determined by NOESY experiments and 3.87 Å in the
calculated structure. Consequently, the excellent agreement
between both methods allowed us to use the structure obtained
in silico to evaluate distances between H(3) atoms and the
nearest CB[7] oxygens, as well as C(3)−H(3)−O angles.
Distances were 2.08 ( 0.02) Å and CHO angles 132−161°,
well within the range deemed acceptable for hydrogen bonding
(distances shorter than 2.5 Å and angles narrower than 110°).13
The binding affinities of the metal−ligand complexes toward
CB[7] were then determined by isothermal titration calorim-
etry (ITC). They not only surprised us by their magnitude
(ranging from 3.4 × 106 to 3.8 × 107 M−1, see Supporting
expected, saturation was reached upon addition of 2.0 equiv of
CB[7], until the terpyridine 4′-substituents were fully
encapsulated. Unexpectedly, exchanges between the metal−
ligand complexes and CB[7] were slow on the NMR time scale
(see Figure 1, spectrum b); this is in stark contrast with the p-
tolylmethylammonium cation (4) which undergoes a faster
intermediate exchange with CB[7], despite a direct contact
between the positive ammonium group and the carbonylated
rim of CB[7]. Also, exceptionally strong downfield shifts of
hydrogens located at positions 3 and 3″ of the terpyridine
ligands (between 0.65 and 1.12 ppm among the 6 complexes),
and significant downfield shifts of the opposite 6 and 6″-
hydrogens (0.26 to 0.37 ppm) were measured. These
singularities led us to consider possible CH···O hydrogen
bonds between these hydrogen atoms and the oxygens of the
carbonylated portals of CB[7]; unusually strong downfield
6603
dx.doi.org/10.1021/ja4092165 | J. Am. Chem. Soc. 2014, 136, 6602−6607

Journal of the American Chemical Society
Article
Information for details), but also by their lack of dependence
on the nature of the metal, and their very unfavorable entropic
component (as low as −10.2 kcal/mol at 25 °C, the lowest
parameter ever measured with CB[7], see Figure 3). As a
derivatives bearing neutral and positively charged substituents
(see Figure 3, guests 5−7). These authors showed that the gain
in binding affinities when positively charged groups interact
with the rim of CB[7] is solely an entropic effect, at least in
these cases. Since the solvation of positive species is more
entropically penalizing compared to neutral structures,15 their
partial desolvation upon CB[7] binding (i.e., the return of
water molecules to the bulk) must therefore be more
entropically favorable. As shown in Figure 3 (see pink dots),
an overview of published thermodynamic parameters reveals
that all neutral guests with significant affinities toward CB[7]
(>103 M−1) suffer from an entropic impediment upon binding.
The constant binding enthalpies along the series of guests 5−7
are likely caused by a perfect balance between host−guest
electrostatic interactions (which increase dramatically when
positive charges are present) and unfavorable enthalpies of
desolvation upon binding, that follow the same trend.2
For those same reasons, the binding entropy measured with
guest 4 is positive, since CB[7] becomes part of the first
coordination sphere, and returns constrained water molecules
to the bulk. In the case of our metal−ligand complexes, CB[7]
is located in the second coordination sphere upon binding, and
replaces water molecules whose arrangement around the
charged complex is much less ordered, hence a marked
decrease in binding entropy. We attribute the more favorable
enthalpic component of the interaction, compared to guest 4,
to the three strongly positive points of contact presented to
CB[7] (H(3), H(3″) on the terpyridine ligand, as well as one
H(6) atom of its opposite neighbor; see Figure 2). Figure 4a
Figure 4. Electrostatic potential maps of complexes (a) [Ir·12]3+ and
(b) IrCl3·1, superimposed on the isodensity surface of the structures
(isovalue 0.002) and probed with a positive point charge. Calculations
carried out at the TPSS-D3(BJ)/def2-SVP level; energies in kilocalorie
per mole (kcal/mol). Color-coding: dark blue, light blue, green,
yellow, and red correspond to +200, +100, 0, −100, and −200 kcal/
mol, respectively.
Figure 3. Enthalpy−entropy compensation plot for the interaction of
CB[7] with various guests,3b,16 including the metal−ligand complexes
described in this study. Light blue and pink dots represent positive and
neutral3b,16a−c guests, respectively.
shows an electrostatic potential map of guest [Ir·12]3+; the most
electropositive regions of the potential surface are located at the
top of both pairs of H(3) and H(6) atoms (+172 and +181
kcal/mol, respectively; as a reference, the electrostatic potential
surrounding the hydrogens of the ammonium cation in guest 4
reaches +165 kcal/mol).
To stress the importance of the radiating positive charge
emanating from the metallic cation, we compared the binding
affinities of CB[7] toward complex [Ir·12]3+ and its analogue
IrCl3·1, that bears only one terpyridine ligand; the comparison
was carried out in dimethyl sulfoxide-d6 (DMSO) since
complex IrCl3·1 is insoluble in aqueous medium. The binding
affinity of complex [Ir·12]3+ still reached 3.7 × 102 M−1, despite
DMSO being a much better hydrogen bond acceptor than
water (and as a result, a better candidate than water when
competing for solute interactions against CB[7]);17 the ejection
of aprotic DMSO from the cavity of CB[7] to the bulk upon
reference, we also determined the binding affinity of CB[7]
toward the p-tolylmethylammonium cation (4; 6.4 × 106 M−1).
Although its binding affinity is similar to the metal−ligand
complexes, its binding entropy is much more favorable (+1.3
kcal/mol at 25 °C), but is compensated by a weaker enthalpic
contribution (see Figure 3). Titrations were suitably fitted with
a 1:1 binding model, thereby indicating that the presence of
one CB[7] unit on the metal−ligand complex does not
significantly affect the affinity of the second opposite macro-
cycle. This observation is in stark contrast with the behavior of
disubstituted ammonium cations, which can accommodate only
one CB unit per ammonium group.1c,14
To rationalize these thermodynamic parameters, one needs
to reexamine the landmark studies by Kaifer, Isaacs, Gilson,
Kim, Inoue and co-workers,2 related to the CB[7] encapsula-
tion of ferrocene, adamantane and bicyclo[2.2.2]octane
6604
dx.doi.org/10.1021/ja4092165 | J. Am. Chem. Soc. 2014, 136, 6602−6607

Journal of the American Chemical Society
Article
binding is also likely less favorable than the ejection of water
(water−water interactions are less than optimal inside CB[7],
hence the energetically favorable ejection). In complex IrCl3·1,
however, the partially negative chlorine ligands dramatically
decrease the positive potential at the surface of H(3) and
H(3″) (from +172 to +29.6 kcal/mol, see Figure 4b), resulting
in a complete loss of affinity toward CB[7]. These results
suggest that C−H···O hydrogen bonding between the metal−
ligand complex and the carbonylated rims of CB[n]s is only
enabled in the presence of a neighboring positive charge. The
ligand can then be seen as an area of low relative permittivity
that allows the propagation of the charge toward the periphery
of the complex and the CB[n] portals. We also attribute the
favorable binding enthalpy of our metal−ligand complexes to a
less penalizing enthalpy of desolvation upon CB[7] binding,
due to the longer separation between second coordination
sphere water molecules and the metallic cation; a similar effect
has been observed when comparing the solvation enthalpies of
ammonium and tetramethylammonium cations (−78 vs −51
kcal/mol, respectively).15
The recognition properties of the metal−ligand complexes
were extended to CB[8] (4′-substituents in terpyridines 2 and
3 were specifically chosen to interact with this macrocycle). As
shown by Zhang and co-workers,18 CB[8] can encapsulate two
naphthyl units into its cavity when they are flanked by
positively charged groups. In our case, Coulombic and CH···O
hydrogen bonding on both portals of CB[8] led to the
formation of a dynamic supramolecular oligomer in the
presence of stoichiometric amounts of CB[8] and complexes
[Fe·22]2+ and [Ir·22]3+ (see Figure 5). This is supported by the
free CB[8] above 1.0 equiv. Diffusion-ordered NMR spectros-
copy experiments (DOSY) yielded diffusion coefficients D =
2.8 × 10−10 and 7.1 × 10−11 m2/s for complex [Fe·22]2+ and its
CB[8]-containing oligomer, respectively, at a 2.0 mM
concentration. The molecular weight (M) dependence on the
diffusion coefficients follows the power law D ∝ M−m, where m
ranges from approximately 0.3 to 0.6 depending on the nature
of the polymer.19 After calibration with assemblies of known
molecular weights and diffusion coefficients (see Supporting
Information section), we extracted an approximate value for m
(0.47).
The approximate average molecular weight of oligomer ([Fe·
22]2+·CB[8])n is thus 2.4 × 104 g/mol, which corresponds to 11
monomers held together by CB[8]. A heavier average
molecular weight (1.2 × 105 g/mol, corresponding to 54
monomeric units) was found in the case of oligomer ([Ir·22]3+·
CB[8])n at a 1.0 mM concentration. We do stress, however,
that molecular weights extrapolated from DOSY experiments
suffer from significant uncertainty, which may reach factors of
approximately 3 at 105 g/mol.
We note a peculiar feature of the 1H NMR spectrum of iron
and iridium complexes of ligands 2 and 3 in the presence of
increasing, yet substoichiometric amounts of CB[8] (see the
case of complex [Fe·22]2+ in Figure 5): a symmetrical multiplet
belonging to the CB[8]-encapsulated 2-naphthyl substituent
first appears at 6.57 ppm (see black circles in Figure 5), and
upon further addition of the macrocycle, dissipates as another
naphthyl signal of oligomer ([Fe·22]2+·CB[8])n forms (6.70
ppm; black dots). DOSY experiments yielded a diffusion
coefficient of 1.7 × 10−10 m2/s and a 2.8 × 103 g/mol molecular
weight for the assembly at low CB[8] concentration; therefore,
we assign its structure as ([Fe·22]2+)2·CB[8] (M = 2877 g/
mol). The existence of two sets of discrete signals for this short
dimer and the longer oligomers indicates a slightly different
organization of the pair of naphthyl units inside CB[8] in these
two cases. We suggest that the connection of an additional [Fe·
22]2+·CB[8] fragment to the short dimer ([Fe·22]2+)2·CB[8]
benefits from positive cooperativity originating from the iron
cation located two metals away: electrostatic potential maps
indicate that the positive surface area of dimer ([Fe·22]2+)2·
CB[8] available for subsequent [Fe·22]2+·CB[8] binding is
significantly more electropositive (+155 and +152 kcal/mol
above position H(3) and H(6), respectively, see Figure 6a)
than the same area in free metal−ligand complex [Fe·22]2+
(+134 and +133 kcal/mol above H(3) and H(6), see Figure
6b).
CB[8] also encapsulates two 2,3,5,6-tetrafluorophenyl units,
and dynamic oligomers of complexes [Fe·32]2+ and [Ir·32]3+ are
formed with a stoichiometric amount of the macrocycle (see
Figure 7; M = 5.8 × 104 and 3.5 × 104 g/mol, corresponding to
27 and 15 monomeric units, respectively, at 2.0 mM). However,
the appearance of a new signal at 7.1 ppm in the presence of
excess CB[8] (see crossed circles in spectra c and f) suggests
that binary complexes between CB[8] and the tetrafluor-
ophenyl units (affording assemblies [M·32]n+·(CB[8])2),
compete with the ternary complexes responsible for the
formation of the oligomers (black dots).
Finally, we tested whether CB[8] would opt for narcissistic
or social self-sorting in the presence of equimolar amounts of
homo-oligomers ([Fe·22]2+·CB[8])n and ([Ir·32]3+·CB[8])n.
The disappearance of the signature signal for the encapsulated
naphthyl dimer at 6.70 ppm (see Figure 8, spectra a and c, black
dots and the fact that iridium−terpyridine complexes were
Figure 5. 1H NMR spectra of complex [Fe·22]2+ (2.0 mM) (a) in the
absence of CB[8], and (b−f) with increasing amounts of CB[8] (up to
2.0 mM). 1H NMR spectra of complex [Ir·22]3+ (g) in the absence of
CB[8], and (h) in the presence of a stoichiometric amount of the
macrocycle. Black circles and dots highlight the signature signals of the
two modes of naphthyl/naphthyl interaction inside CB[8].
very strong upfield shifts of the naphthyl hydrogen nuclei upon
interaction with CB[8] (approximately 1.1 ppm, see Figure 5,
spectra a−f), as well as the observed signal broadening
indicating an increase in correlation time, shorter transverse
relaxation times and hence a large increase in the molecular
weight of the assemblies. In addition, the aromatic region of the
1H NMR spectra is not altered in the presence of an excess
amount of CB[8], thereby indicating that the latter is present as
6605
dx.doi.org/10.1021/ja4092165 | J. Am. Chem. Soc. 2014, 136, 6602−6607

Journal of the American Chemical Society
Article
Figure 8. 1H NMR spectra of (a) homo-oligomer ([Fe·22]2+·CB[8])n,
(b) homo-oligomer ([Ir·32]3+·CB[8])n, (c) hetero-oligomer ([Fe·
22]2+·CB[8]·[Ir·32]3+·CB[8])n, (d) homo-oligomer ([Ir·22]3+·CB[8])n,
(e) homo-oligomer ([Fe·32]2+·CB[8])n, and (f) hetero-oligomer ([Ir·
22]3+·CB[8]·[Fe·32]2+·CB[8])n. Black dots and dashed circles highlight
the disappearance of the signature signal for naphthyl/naphthyl
interactions, and black squares highlight the formation of new
characteristic signals of the hetero-oligomers.
Figure 6. Electrostatic potential maps of complexes (a) ([Fe·22]2+)2·
CB[8] and (b) [Fe·22]2+, superimposed on the isodensity surface of
the structures and probed with a positive point charge. See Figure 4 for
calculation method and color-coding.
stoichiometric amounts of CB[8], which acts as a “soft”,
noncovalent connector between metal/terpyridine complexes.
ASSOCIATED CONTENT
* Supporting Information
■
S
Preparation and characterization of ligands 1−3, as well as their
Fe(II) and Ir(III) complexes; titration of metal−ligand
complexes with CB[7] and CB[8]; ITC results; DOSY
calibration; distance measurements by 2D NOESY experi-
ments; Cartesian coordinates of the optimized structure of
assemblies [Ir·12]3+·(CB[7])2 and ([Fe·22]2+)2·CB[8]. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Figure 7. 1H NMR spectra of complex [Fe·32]2+ (a) in the absence of
CB[8], and in the presence of (b) a stoichiometric amount of CB[8],
and (c) 2.5 equiv of CB[8]. Spectra (d−f) were obtained similarly with
complex [Ir·32]3+.
AUTHOR INFORMATION
■
Corresponding Author
masson@ohio.edu
Notes
The authors declare no competing financial interest.
found not to be labile indicate by transposition the quantitative
formation of the alternating “social” hetero-oligomer ([Fe·
22]2+·CB[8]·[Ir·32]3+·CB[8])n!20
ACKNOWLEDGMENTS
■
This work was supported by the American Chemical Society
Petroleum Research Fund (PRF No. 51053−ND4), the
Department of Chemistry and Biochemistry, the College of
Arts and Sciences and the Vice President for Research at Ohio
University. We thank the Ohio Supercomputer Center (OSC)
in Columbus for its generous allocation of computing time.
Signature signals of the latter are highlighted with black
squares in Figure 8. The clear preference for naphthyl/
tetrafluorophenyl encapsulation inside CB[8] to the expense of
the corresponding homodimers is likely due to favorable
quadrupole−quadrupole interactions in the former case (this is
reminiscent of the iconic interaction between benzene and
perfluorobenzene).21 Similar observations were made upon
combination of assemblies ([Ir·22]3+·CB[8])n and ([Fe·32]2+·
CB[8])n (see Figure 8, spectra d−f).
In conclusion, we have shown that (1) CB[n] macrocycles
can form tight complexes with guests bearing a positive core,
even when the former occupy the second coordination sphere;
(2) like water in the solid state, organic ligands such as
functionalized terpyridines can act as mediators between the
positive charge and the carbonylated portal of CB[n]s, resulting
in the formation of favorable CH···O interactions; and (3) well-
organized dynamic oligomers can be formed in the presence of
REFERENCES
■
(1) (a) Mock, W. L.; Shih, N. Y. J. Org. Chem. 1986, 51, 4440−4446.
For two reviews covering the recognition properties and applications
of CB[n] (n = 5− 8, 10), see: (b) Lagona, J.; Mukhopadhyay, P.;
Chakrabarti, S.; Isaacs, L. Angew. Chem., Int. Ed. 2005, 44, 4844−4870.
(c) Masson, E.; Ling, X.; Joseph, R.; Kyeremeh-Mensah, L.; Lu, X.
RSC Adv. 2012, 2, 1213−1247.
(2) (a) Rekharsky, M. V.; Mori, T.; Yang, C.; Ko, Y. H.; Selvapalam,
N.; Kim, H.; Sobransingh, D.; Kaifer, A. E.; Liu, S.; Isaacs, L.; Chen,
W.; Moghaddam, S.; Gilson, M. K.; Kim, K.; Inoue, Y. Proc. Natl. Acad.
Sci. U. S. A. 2007, 104, 20737−20742. (b) Moghaddam, S.; Yang, C.;
6606
dx.doi.org/10.1021/ja4092165 | J. Am. Chem. Soc. 2014, 136, 6602−6607

Journal of the American Chemical Society
Article
Rekharsky, M.; Ko, Y. H.; Kim, K.; Inoue, Y.; Gilson, M. K. J. Am.
Chem. Soc. 2011, 133, 3570−3581. (c) Ko, Y. H.; Hwang, I.; Lee, D.
W.; Kim, K. Isr. J. Chem. 2011, 51, 506−14.
Biomol. Chem. 2013, 11, 3116−3127. (f) Yu, J. S.; Wu, F. G.; Zhou, Y.;
Zheng, Y. Z.; Yu, Z. W. Phys. Chem. Chem. Phys. 2012, 14, 8506−8510
as well as ref 1c for the exhaustive list of all references that have
contributed to this plot.
(3) (a) Florea, M.; Nau, W. M. Angew. Chem., Int. Ed. 2011, 50,
9338−9342. (b) Biedermann, F.; Uzunova, V. D.; Scherman, O. A.;
Nau, W. M.; Simone, A. D. J. Am. Chem. Soc. 2012, 134, 15318−
15323. (c) Nau, W. M.; Florea, M.; Assaf, K. I. Isr. J. Chem. 2011, 51,
559−577. (d) Marquez, C.; Nau, W. M. Angew. Chem., Int. Ed. 2001,
40, 4387−4390. (e) Mohanty, J.; Nau, W. M. Angew. Chem., Int. Ed.
2005, 44, 3750−3754. (f) Koner, A. L.; Nau, W. M. Supramol. Chem.
2007, 19, 55−66.
(17) For an excellent review that describes the use of parameters for
hydrogen bonding to predict host−guest recognition patterns, see:
Hunter, C. A. Angew. Chem., Int. Ed. 2004, 43, 5310−5324.
(18) (a) Fang, R.; Liu, Y.; Wang, Z.; Zhang, X. Polym. Chem. 2013, 4,
900−903. (b) Liu, Y.; Fang, R.; Tan, X.; Wang, Z.; Zhang, X. Chem.
Eur. J. 2012, 18, 15650−15654. (c) Liu, Y.; Yang, H.; Wang, Z.; Zhang,
X. Chem.Asian J. 2013, 8, 1626−1632. (d) Liu, Y.; Huang, Z.; Tan,
X.; Wang, Z.; Zhang, X. Chem. Commun. 2013, 49, 5766−5768.
(19) It is 0.33 in the case of spherical species (based on the Stokes−
Einstein equation) and a linear correlation between volumes and
molecular weights; 0.50 according to the Zimm model; see Shimada,
K.; Kato, H.; Saito, T.; Matsuyama, S.; Kinugasa, S. J. Chem. Phys.
2005, 122, 244914−244920 for details.
(20) Subscript “n” is used indiscriminately throughout the study,
although the length of each oligomer varies depending on its
composition.
(21) (a) Patrick, C. R.; Prosser, G. S. Nature 1960, 187, 1021.
(b) Williams, J. H. Acc. Chem. Res. 1993, 26, 593−598. (c) Tsuzuki, S.;
Uchimaru, T.; Mikami, M. J. Phys. Chem. A 2006, 110, 2027−2033.
(4) (a) Grimme, S.; Ehrlich, S.; Goerigk, L. J. Comput. Chem. 2011,
32, 1456−1465. (b) Grimme, S. Chem.Eur. J. 2012, 18, 9955−9964.
(5) (a) Gerasko, O. A.; Samsonenko, D. G.; Fedin, V. P. Russ. Chem.
Rev. 2002, 71, 741−760. (b) Whang, D.; Heo, J.; Park, J. H.; Kim, K.
Angew. Chem., Int. Ed. 1998, 37, 78−80. (c) Jeon, Y. M.; Kim, J.;
Whang, D.; Kim, K. J. Am. Chem. Soc. 1996, 118, 9790−9791.
(d) Abramov, P. A.; Adonin, S. A.; Peresypkina, E. V.; Sokolov, M. N.;
Fedin, V. P. J. Struct. Chem. 2010, 51, 731−736.
(6) (a) Gerasko, O. A.; Sokolov, M. N.; Fedin, V. P. Pure Appl. Chem.
2004, 76, 1633−1646. (b) Samsonenko, D. G.; Sokolov, M. N.;
Virovets, A. V.; Pervukhina, N. V.; Fedin, V. P. Eur. J. Inorg. Chem.
2001, 167−172. (c) Algarra, A. G.; Sokolov, M. N.; Gonzal
́
ez-Platas, J.;
Fernandez-Trujillo, M. J.; Basallote, M. G.; Hernandez-Molina, R.
́
́
Inorg. Chem. 2009, 48, 3639−3649. (d) Bernal, I.; Mukhopadhyay, U.;
Virovets, A. V.; Fedin, V. P.; Clegg, W. Chem. Commun. 2005, 3791−
3792.
(7) (a) Collin, J. P.; Dixon, I. M.; Sauvage, J. P.; Williams, J. A. G.;
Barigelletti, F.; Flamigni, L. J. Am. Chem. Soc. 1999, 121, 5009−5016.
(b) Flamigni, L.; Collin, J. P.; Sauvage, J. P. Acc. Chem. Res. 2008, 41,
857−871. (c) Tang, B.; Yu, F.; Li, P.; Tong, L.; Duan, X.; Xie, T.;
Wang, X. J. Am. Chem. Soc. 2009, 131, 3016−3023. (d) Wild, A.;
Winter, A.; Hager, M. D.; Gorls, H.; Schubert, U. S. Macromol. Rapid
̈
Commun. 2012, 33, 517−521. (e) Newkome, G. R.; Cho, T. J.;
Moorefield, C. N.; Mohapatra, P. P.; Godínez, L. A. Chem.Eur. J.
2004, 10, 1493−1500.
(8) (a) Kar, T.; Scheiner, S. J. Phys. Chem. A 2004, 108, 9161−9168.
(b) Scheiner, S.; Kar, T. J. Phys. Chem. A 2008, 112, 11854−11860.
(9) (a) Baleja, J. D.; Moult, J.; Sykes, B. D. J. Magn. Reson. 1990, 87,
375−384. (b) Mugridge, J. S.; Zahl, A.; van Eldik, R.; Bergman, R. G.;
Raymond, K. N. J. Am. Chem. Soc. 2013, 135, 4299−4306.
(10) (a) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Phys.
Rev. Lett. 2003, 91, 146401−146404. (b) Grimme, S.; Antony, J.;
Ehrlich, S.; Krieg, H. J. Chem. Phys. 2010, 132, 154104−154119.
(11) (a) Klamt, A.; Schuurmann, G. J. Chem. Soc. Perkin Trans. 2
̈
̈
1993, 5, 799−805. (b) Klamt, A.; Jonas, V. J. Chem. Phys. 1996, 105,
9972−9980. The COSMO model uses a hypothetical solvent of
infinite relative permittivity; the error caused by this approximation is
insignificant for highly polar solvents like water.
(12) (a) Buhl, M.; Reimann, C.; Pantazis, D. A.; Bredow, T.; Neese,
̈
F. J. Chem. Theory Comput. 2008, 4, 1449−1459. (b) Vícha, J.;
Patzschke, M.; Marek, R. Phys. Chem. Chem. Phys. 2013, 15, 7740−
7754.
(13) (a) McDonald, I. K.; Thornton, J. M. J. Mol. Biol. 1994, 238,
777−793. (b) Arunan, E.; Desiraju, G. R.; Klein, R. A.; Sadlej, J.;
Scheiner, S.; Alkorta, I.; Clary, D. C.; Crabtree, R. H.; Dannenberg, J.
J.; Hobza, P.; Kjaergaard, H. G.; Legon, A. C.; Mennucci, B.; Nesbitt,
D. J. Pure Appl. Chem. 2011, 83, 1637−1641.
(14) Rekharsky, M. V.; Yamamura, H.; Kawai, M.; Osaka, I.; Arakawa,
R.; Sato, A.; Ko, Y. H.; Selvapalam, N.; Kim, K.; Inoue, Y. Org. Lett.
2006, 8, 815−818.
(15) Marcus, Y. Biophys. Chem. 1994, 51, 111−127.
(16) See (a) Gupta, S.; Choudhury, R.; Krois, D.; Brinker, U. H.;
Ramamurthy, V. J. Org. Chem. 2012, 77, 5155−5160. (b) Choudhury,
R.; Gupta, S.; Da Silva, J. P.; Ramamurthy, V. J. Org. Chem. 2013, 78,
1824−1832. (c) Yu, J. S.; Wu, F. G.; Tao, L. F.; Luo, J. J.; Yu, Z. W.
Phys. Chem. Chem. Phys. 2011, 13, 3638−3641. (d) Lee, S. J. C.; Lee, J.
W.; Lee, H. H.; Seo, J.; Noh, D. H.; Ko, Y. H.; Kim, K.; Kim, H. I. J.
Phys. Chem. B 2013, 117, 8855−8864. (e) Joseph, R.; Masson, E. Org.
6607
dx.doi.org/10.1021/ja4092165 | J. Am. Chem. Soc. 2014, 136, 6602−6607