1.

05 The Supramolecular Alkali Metal Cation–p Interaction

GW Gokel and S Negin, University of Missouri, St. Louis, MO, United States
Ó 2017 Elsevier Ltd. All rights reserved.

1.05.1 Introduction 103

1.05.2 Structural Features of Alkali Metal Cations 105
1.05.3 The Cation–p Interaction 105
1.05.4 Early Crystal Structure Evidence for Cation–p Interactions 108
1.05.5 Molecular Receptors Used to Systematically Probe Cation–p Interactions 108
1.05.5.1 Solution Evidence for Cation–p Interactions 112
1.05.5.2 Imidazole as a Sigma- or p Donor 112
1.05.5.3 Control Studies for the Diaza-18-crown-6 Lariat Ether Complexes 113
1.05.6 Double and Triple Bonds as p-Donor Groups 113
1.05.6.1 Receptors and Other Experimental Studies 113
1.05.6.2 The Cation–p Interaction in Synthesis 118
1.05.7 Literature Reviews 118
1.05.8 Conclusion 118
References 118

1.05.1 Introduction

There are many weak or feeble forces in chemistry that have considerable cumulative strength. These individual forces contribute in
a variety of ways to the overall formation of a supramolecular assembly. Among these are the formation of hydrogen bonds, dipole–
dipole interactions, ion–dipole interactions, ion–ion interactions, van der Waals contacts, and solvation effects.
It is probable that the best known of all of these weak forces are hydrogen bond interactions. Historically, hydrogen bonding has
been the most studied supramolecular force in biochemistry. A great deal is known about their strengths and orientations, especially
from experimental observations in peptide and protein chemistry. Two principles seem to dominate hydrogen bond interactions.
These are that the strongest interactions among three bonded species (XeH/Y) are linear and the largest electronegativities of X and
Y will afford the strongest bonds.
During the past several decades, the occurrence and importance of cation–p interactions have been recognized. Cation–p inter-
actions often occur in concert with hydrogen bonds in complexes or enzymatic active sites. A number of studies discussed later have
attempted to determine whether H bonding or cation–p interactions contributes more stability to a supramolecular structure. It is,
therefore, worth briefly recounting the basic understanding of hydrogen bond interactions.
The most common hydrogen bond interactions occur between hydroxyl groups and oxygen (OeH/:O), or (OeH/:N) or
nitrogen-hydrogen (NeH) bonds and oxygen (NeH/:O) or nitrogen (NeH/:N). Generally speaking, the hydrogen bond may
be characterized simply as XeH/:Y, where X and Y are electronegative. When nitrogen is positively charged, that is, (NþeH),
the hydrogen bond from it to an electronegative element is typically stronger and shorter than when the nitrogen is uncharged. Jef-
frey and Sanger state in their 1991 monograph 1 titled, Hydrogen Bonding in Biological Structures the following. “[T]here is no evidence
of a quantitative relationship between the OeH and H/O bond lengths and O/O bond distances. There is no direct experimental
relationship between hydrogen bond lengths and hydrogen bond strengths.” Notwithstanding these assertions, where direct
comparisons can be made, it is typically observed that the strongest hydrogen bond interactions are the shortest and most linear
ones.
Hydrogen bonds involving OeH or NeH donors to electronegative elements, especially in peptide and protein structures, have
been understood for the better part of a century. In contrast, the so-called carbon-hydrogen, hydrogen bonds were discredited. In
1962, Suter wrote “. over the past 10 years, numerous examples of short intermolecular distances between a carbon atom with at
least one hydrogen atom attached to it and an oxygen atom have appeared in papers . In practically all cases, the significance of
these interactions has not been noted.” 2 Unfortunately, others in the scientific community expressed skepticism about these obser-
vations and exploration of this weak force languished.
Two decades later, Desiraju formulated rules 3 to adequately describe CeH hydrogen bonding interactions.4 As might have been
expected, he noted that CeH/:O hydrogen bond is most significant when the distance between the heavy atoms is < 4 Å and the
three atom bond angle is > 150 degrees. These rules are generally applicable to more conventional hydrogen bonds as well.
Among the genetically encoded amino acids, virtually all functional groups are represented. Serine and threonine are OH
hydrogen bond donors. Cysteine is a sulfur (SeH) hydrogen bond donor. Arginine and lysine possess guanidine and amino groups,
respectively, that are protonated at physiologic pH and that can serve as hydrogen bond donors. Aspartic acid and glutamic acid
have carboxyl groups within the pendant chains. Typically, however, these will be ionized at physiologic pH and will serve as
H-bond acceptors rather than donors. All of these residues can participate in one way or another in hydrogen bonding interactions.

Comprehensive Supramolecular Chemistry II, Volume 1 http://dx.doi.org/10.1016/B978-0-12-409547-2.12483-7 103

104 The Supramolecular Alkali Metal Cation–p Interaction

Perhaps the most important of these interactions is salt bridge formation. At physiologic pH, the carboxylate anion serves as the
base and the protonated amine is the H-bond donor. In any event, two hydrogen atoms will be associated with nitrogen and the
carboxylate oxygen atoms. This is illustrated in Fig. 1 for aspartic acid and lysine.
Four of the common 20 amino acids possess aromatic side chains. These are phenylalanine, tyrosine, tryptophan, and histidine.
The first three of these contain electron-rich aromatic residues: benzene, indole, and phenol. It is these that are the common Lewis
bases that are involved in cation–p interactions. Two of the arenes possess additional functionality. Tyrosine contains a phenolic
hydroxyl group that can serve as a hydrogen bond donor. Tryptophan contains an NeH group in the five-membered portion of the
indole ring. This can, in principle, serve as a hydrogen bond donor as well. Phenylalanine lacks additional functionality and usually
interacts only as a p donor (see Fig. 2).
Histidine differs from the other three aromatic amino acids because its imidazole residue is relatively electron poor. Further, it
contains two nitrogen atoms within its imidazole heterocyclic residue. One of these has a hydrogen atom attached to nitrogen that
can serve as the hydrogen bond donor. The other nitrogen atom can serve as a hydrogen bond acceptor. Because of imidazole’s
structure, histidine is a critical component of the charge relay system found in the serine protease family of enzymes. The arene,
however, is not an important Lewis base donor for cation–p interactions.
Electron-rich aromatics are Lewis bases and their p systems can serve as hydrogen bond acceptors. Arenes can also associate with
other arenes. When two arenes associate face to face, the phenomenon is referred to as p–p stacking. Typical arrangements for p–p
stacking are the so-called T-shaped arrangement or an offset arrangement. When direct concentric stacking is observed, it is likely
that the interaction is fostered by the exclusion of solvent rather than a p–p affinity ( Fig. 3).

Figure 1 Salt bridge formation involving the side chains of lysine and arginine.

Figure 2 Structures phenylalanine (Phe, F), tyrosine (Tyr, Y), tryptophan (Trp, W), and histidine (His, H).

Figure 3 Modes of p–p stacking illustrated with two benzene molecules.

 The Supramolecular Alkali Metal Cation–p Interaction 105

Figure 4 Structures of 18-crown-6$Kþ and [2.2.2]cryptand$Kþ.

Ion–dipole interactions can be considered as Lewis acid-base interactions. The ion can be either positive or negative and the
dipole must then be the counterpart. Such interactions have been extensively studied for cations and oxygen or nitrogen donors
in the context of crown ether and cryptand chemistry. Sodium and potassium cations are readily bound by receptors such as
18-crown-6 and [2.2.2]cryptand ( Fig. 4). The interactions involved here are donation of electron pairs from the heteroatoms to
the positively charged cation. It is also possible to bind anions with these receptors if the nitrogen atoms within the heterocycles
are protonated. In such a case, stabilization results from hydrogen bond donor interactions with the bound anion, but this is
beyond the scope of the present discussion.

1.05.2 Structural Features of Alkali Metal Cations

The complexation of transition and coinage metals by a variety of ligands has been known for more than 100 years. 5 Complexation
of alkali metal cations such as sodium and potassium has proved to be a greater challenge because they lacked the complex orbital
structures of the metals more to the center of the periodic table. Essentially, the alkali metal cations are featureless spheres.
Complexation of alkali metal cations by crowns6 and cryptands7 was, therefore, a breakthrough and engendered the field that
has become supramolecular chemistry.8
Supramolecular interactions tend to be relatively weak compared to covalent bond interactions. Covalent bonds strengths typi-
cally range from  50 kcal mol 1 ( 200 kJ mol 1) to more than 200 kcal mol 1 ( 800 kJ mol 1). Hydrogen bond interactions
are often quoted as ranging in strength from  3 kcal mol 1 ( 12 kJ mol 1) to  10 kcal mol 1 ( 40 kJ mol 1). Ion–dipole inter-
actions, such as those stabilizing crown ether complexes and cation–p interactions, are typically quoted to be in the
0.5–3 kcal mol 1 (2 to  12 kJ mol 1) range.
In supramolecular chemistry, we are often constrained in our thinking by considerations of enthalpy. Chemists are well aware of
the fundamental relationship DG ¼ DH  TDS. In chemistry laboratories, we typically measure some parameters related to DG.
When we think about chemical interactions, however, we typically contemplate the enthalpy (DH). What largely eludes us is the
role played by entropy in the TDS term. Indeed, the questions of entropy and solvent organization have proved especially chal-
lenging for those attempting to understand drug interactions and pharmacodynamics. 9
Numerous interactions are normally at play in the active site of an enzyme. These are hydrogen bonds, polar contacts, salt bridge
formation, cation–p interactions, and the inclusion, bridging, or exclusion of solvent. None of these interactions is a powerful indi-
vidual force, but taken together they ensure the order needed for receptor-ligand interactions to occur or for enzymatic catalysis to
proceed.

1.05.3 The Cation–p Interaction

In its most generalized form, we can consider a cation as a point charge and a p system as a negative dipole in the form of an isolated
double bond. Of course, the p system can be an isolated double bond, an isolated triple bond, a conjugated bond array, an arene, or
a combination of these. Fig. 5 illustrates how p bonds can interact with an alkali metal cation.
In aqueous solution, cations such as sodium and potassium are extensively solvated. In the gas phase, however, the sodium
cation is free to function as a Lewis acid. A reaction between a sodium cation as a Lewis acid and benzene could be detected by
mass spectrometry.
Naþ þ C6 H6 /½Na,C6 H6 þ
As early as the 1980s, Kebarle and coworkers 10 undertook mass spectrometric experiments to determine the strengths of cation–
p interactions. In particular, they focused their attention on sodium and potassium cations interacting with benzene. Later studies
by Castleman and coworkers11 added to our understanding of these gas phase interactions. Data for the solvent-free interactions are
recorded in Table 1.
Additional energetic data were obtained by mass spectrometric methods for the Naþ$phenylalanine complex. 12 Enthalpies were
determined for the sodium complexes of alanine and phenylalanine. Their binding enthalpies were 38 and 45 kcal mol 1,
106 The Supramolecular Alkali Metal Cation–p Interaction

Figure 5 Interactions of an alkali metal cation with a double or triple bond with an arene represented by benzene.

Table 1 Gas phase energetics for cation–p interactions

Interaction DH (kcal mol 1) References

D
Na D C6H6 28.0  0.1 Castleman et al. 11
NaD D H2O 24 Castleman et al.
NaD D HOCH3 26.6 Castleman et al.
KD D C6H6 18.3 Kebarle et al. 10
KD D H2O 17.9 Kebarle et al.

respectively, suggesting that the arene-cation interaction that is possible in phenylalanine afforded  7 kcal mol 1 of stabilization in
this context.
In 1985, Meot-Ner and Deakyne 13 demonstrated by mass spectrometry that arenes can interact with ammonium cations just as
they do with alkali metal cations. At about the same time, Burley and Petsko14 surveyed high-resolution protein structures and
found that an unusually large percentage of amine groups were located near the aromatic rings of aromatic amino acids. This like-
wise suggested (Fig. 6) an ammonium ion-arene interaction. While ammonium ions are not the focus of this article, the similarities
to alkali metal interactions are significant and, furthermore, ammonium ion cation–p interactions are important in biological
systems.
The mass spectrometric work conducted by these pioneers laid the groundwork for our understanding of cation–p interactions
as a supramolecular force. Unfortunately, mass spectrometry does not provide structural details that would help supramolecular
chemists think more clearly about how to visualize and to utilize cation–p interactions. It was obviously important, however, to
know that such interactions existed and could be measured so that chemists could design receptors and biologists could search
for such interactions in the complex milieu of protein chemistry.
Rhyzhov and Dunbar 15 assessed the interactions of Naþ and Kþ with phenylalanine (Phe, F), tyrosine (Tyr, Y), and tryptophan
(Trp, W). The work was done mass spectrometrically using the kinetic method (8 kV beam). The binding strengths for either cation
were found to be in the order Phe  Tyr < Trp. These experimental studies were augmented by quantum mechanical (density func-
tional theory or DFT) calculations that were found to be in general agreement with the experimentally observed gas phase results.
The calculations provided models that showed multiple interactions between Naþ or Kþ with nitrogen and oxygen sigma donors in
Mþ$Phe or Mþ$Trp complexes.
Dunbar et al. 16 recently reported a mass spectrometric study of a “cation–pi cage” comprised of the ion and the dipeptide phe-
nylalanyl phenylalanine (PhePhe). They evaluated complexes of Liþ, Naþ, Kþ, Rbþ, Csþ, Ca2 þ, Ba2 þ, and Agþ. Extensive mass spec-
trometric, spectroscopic, and computational analyses led to the conclusion that an amide oxygen and an arene were always
important donors for the alkali metal cations. They noted that the alkaline earth metal ions were bound by both available arenes
in a very compact cage structure.
Castleman and coworkers 17 reported competition experiments involving Naþ and benzene in competition with water, acetoni-
trile, acetone, and several other solvents. An important conclusion reached in this work was that when water and benzene were
compared, the affinity of Naþ for benzene exceeded that of a water molecule. More recently, Rao et al.18 reported a study to deter-
mine the effect of solvent on cation–p interactions involving Liþ, Kþ, and Mg2 þ with benzene. The authors describe a number of
solvent-mediated effects such as cation–p enhancement of arene proton hydrogen bond donicities and variations in cation to arene
centroid distance with changes in solvent number.
Considerable computational study of alkali metal cation–p interactions has been reported. One of the earliest studies was re-
ported by Kumpf and Dougherty 19 who calculated interaction energies for Liþ, Naþ, Kþ, and Rbþ interacting with benzene.20 At
that time, no solid-state structure21 was available to help understand the selectivity of ion channels such as the KcsA voltage-

Figure 6 Schematic representation of an arene-ammonium ion interaction.

 The Supramolecular Alkali Metal Cation–p Interaction 107

gated protein channel. This channel was selective for potassium, but the mechanism by which this selectivity was manifested was
not understood. They proposed that cation–p interactions were responsible for the selectivity. Based on the computations, this
seemed a plausible explanation. However, shortly thereafter, MacKinnon and coworkers experimentally examined this proposal.22
By using site-directed mutagenesis (Tyr / Phe), they showed that the hydroxyl groups rather than the arenes per se were critical to
selectivity. Ultimately, the solvation-stripping selectivity mechanism became well understood, when MacKinnon and coworkers
revealed the solid-state structure of this channel and its mechanisms of action.23
Extensive theoretical work has been reported in the intervening years. Wouters explored the Naþ–Trp interaction using several
computational approaches. 24 The focus was the thermophilic bacterium Bacillus stearothermophilus, the high-resolution solid-state
structure which showed a Naþ–Trp contact.25 A computational (MP2/6-311G**) study reported by Tsuzuki et al.26 showed that
induction and electrostatic interactions were the major source of attraction, at least for the Liþ, Naþ, and Kþ cations studied
with benzene, toluene, ethylbenzene, and t-butylbenzene.
Biot et al. used quantum mechanical calculations (Meller–Plesset (MP2)) and a comparison of X-ray structures to assess the
strengths of cation–p and ammonium-p interactions applicable to proteins. 27 In particular, they focused on adenine bases and
arginine, lysine, asparagine, and glutamine as a means to assess stabilizing interactions at a protein-DNA interface. At about the
same time, Mo et al. used theoretical methods to probe the importance of cation–p interactions in opioid receptors.28 They
concluded that electrostatics, although not the only stabilizing interaction, played the major role in cation–p binding in the
delta-opioid receptor.
Crowley and Golovin surveyed a range of solid-state structures available in the Protein Data Bank (PDB) 29 or the protein quater-
nary structure server.30 Computational analysis of protein–protein interfaces was then undertaken based on a range of parameters.
The authors noted that the most common cation–p interactions involved the arginine-tyrosine pair. Rodgers and coworkers31 con-
ducted a theoretical study (B3LYP/6-31G*) to gain insight into “nature’s preference for engaging tryptophan over phenylalanine
and tyrosine in cation–pi interactions in biological systems.” In particular, they reported structures, vibrational frequencies, and
rotational constants for complexes involving the five alkali metal cations. A molecular dynamics simulation conducted by Khande-
lia and Kaznessis32 showed that cation–p interactions involving tryptophan stabilize the antimicrobial peptide indolicidin (Fig. 7).
Several studies have been reported recently that address arene-substituent effects and geometric variations in p-donor geometry.
Wheeler and Houk 33 showed that the electrostatic potentials at benzene centroids result from a through-space effect of arene
substituents. Lewis and coworkers34 examined the effect of substituents on arenes that form alkali metal-p complexes. They ob-
tained energies for cation-single arene complexes and for the corresponding Mþ(arene)2 structures. Although the 2[Mþ(arene)]
energies did not perfectly correlate to those of Mþ(arene)2, substituents on the arenes generally correlated with their Hammett
constants.
Computational studies have revealed that cation–p interactions are strongest when the cation is oriented perpendicular to the
arene. The computed interaction energy for sodium interacting with benzene along the perpendicular centroid was 22 kcal mol 1.
This value diminished by half when the orientation moved to 60 degrees off the normal. 35
A number of theoretical studies focused on ammonium ion-p interactions. Gallivan and Dougherty 36 compared cation–p inter-
action energetics with those of salt bridge formation. Their key conclusion was that the “cation–pi interaction shows a well depth of
5.5 kcal mol 1 in water, significantly larger than the 2.2 kcal mol 1 seen for the salt bridge.” They report substantiation of this
finding by surveying protein structures. Ansorg et al.37 used perturbation theory to assess the interaction of NHþ 4 with benzene.
They found that the electrostatic and induction energy components contribute almost equally to stabilization. A quantum mechan-
ical study focused on cation–p interactions involving methylammonium ions was directed in understanding their role in proteins

Figure 7 Structure of the antimicrobial drug, indolicidin.

108 The Supramolecular Alkali Metal Cation–p Interaction

such as histones.38 The methylation state of the ammonium ion and its geometric relationship to an arene were both found to be
factors in overall stability. The computational results were compared to the structural data available in the PDB.

1.05.4 Early Crystal Structure Evidence for Cation–p Interactions

A number of crystal structures appeared in the 1970s through the 1990s that made clear the potential for cation–p interactions. A
structure published in 1962 showed two benzene rings from the electron-rich anion tetraphenylborate interacting with a rubidium
cation. 39 A related potassium complex was published in 197440 and there is also a sodium complex that is recorded in the Cam-
bridge structural database (CSD) as ZZZUPI10. The structures of these complexes are shown in Fig. 8.
Lithium cation is a highly charged dense ion and its complexation with arenes has been documented in a number of crystal
structures. Two of these are shown in Fig. 9. In one, lithium is complexed by tetramethylethylenediamine and an aromatic ring
of naphthalene.41 A structure reported in 1998 shows two lithium cations bound between two stacked fluorene molecules.42
Two structures reported by Corbelin and coworkers in 1991, 43 show arene complexation along with more traditional amine
ligation of the cation (Fig. 10).
In 1981, Hrncir et al. 44 reported a complex of potassium with dibenzo-18-crown-6. Simple complexes of this macrocycle were
already well known for potassium halide salts. What distinguishes this particular structure is that the apical position is occupied by
a molecule of benzene and the halide ion is not in the solvation sphere of the cation. The relationship of the benzene ring to the
cation was perpendicular to the plane of the six-membered ring and exactly to its centroid (Fig. 11).
Also in 1998, Wouters examined the CSD and closely analyzed the solid-state structure of lysozyme. 25 He focused in particular
on a tryptophan indole residue (Trp123) that was in contact with a Naþ cation. In a survey of 44 cation–p interactions reported, the
arene to Naþ distance ranged from 2.83 to 4.92 Å. He concluded that the average distance from the aromatic plane to the Naþ cation
was  4 Å.
Schultz and coworkers used staphylococcal nuclease (Snase) as a vehicle to assess cation–p interactions in a biological context. 45
The 149 amino acid protein has a hydrophobic pocket within it that is lined by two phenylalanines and a tyrosine. In the wild-type
protein, an alanine resides within the pocket. To probe cation–p interactions, S-methylmethionine, which has a pendant
CH2CH2Sþ(CH3)2 side chain, was studied. In a series of experiments, the authors concluded that  DG ¼ 2.6 kcal mol 1. The
authors also note that other workers have reported values for  DG ranging from 0.446 to 2.5 kcal mol 1 depending on the context
in which they were determined.

1.05.5 Molecular Receptors Used to Systematically Probe Cation–p Interactions

In 1980, the author’s lab was developing compounds that had been named lariat ethers. 47 The original concept of lariat ethers was
that the macrocycle would bind a cation in the well-understood macrocyclic fashion, but that sidearms containing donor groups
could provide additional apical solvation. The complexes that resulted would be pseudo-cryptands. The value of such compounds
was that they were more dynamic than cryptands, although their binding strengths were lower. The structures of [2.2.2]cryptand and
a lariat ether bearing structural similarity to it are shown as potassium complexes in Fig. 12.
During the development of the lariat ether compounds, a number of diaza-18-crown-6 molecules were available with a variety of
sidearms. Among these sidearms were benzyl and allyl. The benzyl and the allyl group both contain potential p donors that, in
principle, could have functioned as donor groups in the apical positions of a ring-bound cation complex. The computational study
noted earlier, had it been available at the time, would have made clear that the orientation of the p-donor residues were at too great
an angle to prove effective.
Notwithstanding, lariat ethers were prepared as a series that contained benzyl, propargyl, allyl, and propyl sidearms. Potassium
cation binding strengths 48 were measured for these compounds in methanol solution.49 The binding strengths increased from alkyl

Figure 8 Solid-state structures of apparent p interactions of cations with the arenes of tetrafluoroborate anions.
 The Supramolecular Alkali Metal Cation–p Interaction 109

Figure 9 Lithium cation complexes by aromatic anions.

Figure 10 Sodium cations ligated and interacting with aromatic anions.

Figure 11 Dibenzo-18-crown-6 complex of Kþ in which a single molecule of benzene provides apical cation–p stabilization.

Figure 12 Schematic for cation complexation by a single-armed lariat ether.

110 The Supramolecular Alkali Metal Cation–p Interaction

(propyl) to unsaturated (allyl, double bond) to propargyl (triple bond). This suggested a strengthening cation–p effect. As a control
compound, the crown having cyanomethyl sidearms was prepared. The cyano triple bond was expected to behave differently than
the carbon–carbon triple bond in the propargyl analog. In fact, binding was about the same for the two triple bond side-armed
compounds. The postulated cation–p interactions for a cation and a double-bond-containing sidearm is shown in Fig. 13.
The complexation constants for these unsaturated lariat ethers ( Fig. 14) were measured at several different temperatures and
Van’t Hoff plots were obtained. From these, thermodynamic parameters could be adduced. The binding constant profile obviously
reflected DG. However, the enthalpy contributions, which should have increased linearly with an increase in p-bond donicity, failed
to do so. In addition, the solid-state structures that were obtained, showed no evidence of cation–p interactions.
Nearly 20 years later, the topic came under discussion again in the author’s laboratory. A consideration of hydrogen bond inter-
actions and the requirement of appropriate spatial orientation suggested that perhaps the lariat ether model system could succeed if
the arenes were more appropriately positioned. It was noted that nature separated its arenes from amino acid nitrogen atoms by two
carbons rather than one. A series of diaza-18-crown-6 lariat ethers was then prepared, which had benzene, phenol, and indole side-
arms linked to the macroring ethylene groups. The compounds prepared are shown in the structures of Fig. 15.
Complexes between these two-armed or bibracchial lariat ethers were formed with potassium cation when the sidearms were
terminated by benzene, phenol, or indole. In all three cases (shown in Fig. 16), the ring-bound potassium cation was solvated
by the arene. In the phenylalanine analog, the arene was oriented so that the centroid was perpendicular to the cation. A similar
orientation was observed for the tyrosine analog. The orientation of the indole sidearm in the tryptophan analog corresponded,
but it was the five membered, rather than the six-membered, ring that served as the p donor. In all three complexes, the iodide anion
was excluded from the solvation sphere. When the p donor was the aromatic ring of tyrosine, a hydrogen bond was apparent
between the phenolic hydroxyl group and the anion. Likewise, the amino hydrogen in indole was oriented toward the anion.
The phenylalanine analog lacks an obvious hydrogen donor residue, but the anion was positioned similarly, even though the
hydrogen bond interaction was absent.
The fact that all three solid-state structures were so similar despite differences in the abilities to solvate the cation suggested that
perhaps the apparent cation–p interactions were artifacts of crystal packing forces. As a control, a phenylalanine analog was
prepared in which the arenes were perfluorinated. The presence of five fluorines in each aromatic ring reverses the arene from Lewis
base to Lewis acid character. When the solid potassium iodide complex crystallized and its structure was obtained, the cation was
bound in the macrocycle as expected. However, both of the aromatic sidearms were turned away from the ring-bound cation, which
interacted directly with iodide anion.
As noted earlier, the orientation of indole as a p donor in these lariat ether complexes was somewhat surprising. Computational
studies showed that indole’s six-membered ring is more electron rich than the five-membered, nitrogen-containing ring. A new lariat
ether was prepared in which the indole was attached at the 5-position rather than the 3-position. This would permit the benzene
ring portion of indole to more readily orient itself over the ring-bound cation. In fact, the solid-state crystal structure showed exactly
that type of complex. This was an important finding because it suggested that the pure electrostatic interaction is not always the
defining element in the complexation orientation. In this case, the poorer donor was better oriented for complexation, suggesting
that the indole residue of tryptophan is a very versatile donor in the biological context.
The space-filling metaphor shown in Fig. 16 clearly illustrates the enveloping nature of these complexes. Fig. 17 shows the struc-
ture of the bis(indole) lariat in the stick format, except that the cation and anion are shown in their full volumes. The interactions of
the macroring heteroatoms in a plane with the cation are apparent. The H-bond interaction between indole’s NeH and the iodide
anion are also obvious. The indoles are positioned above and below the cation such that the five-membered ring serves as the p
donor. Computational studies have shown that the six-membered ring should be the more electron-rich portion of the aromatic

Figure 13 4,13-Diaza-18-crown-6 lariat ether having allyl sidearms, shown in the unrealized cation–p complexation conformation.

Figure 14 4,13-Diaza-18-crown-6 derivatives (lariat ethers) having various unsaturated sidearms: R ¼ CH2CH2CH3, CH2CH]CH2, CH2C^CH,
CH2C^N, CH2Ph.
 The Supramolecular Alkali Metal Cation–p Interaction 111

Figure 15 Two-armed lariat ethers prepared to evaluate cation–p interactions.

Figure 16 Potassium cation complexes of diaza-lariats having indole (left), phenol (center), and phenyl (right) apical p-donor groups.

Figure 17 Structure of 3-indolylCH2CH2<N18N>CH2CH2-3-indolyl$NaI shown as sticks for the framework and full volume for the cation and anion.
112 The Supramolecular Alkali Metal Cation–p Interaction

system. It, therefore, is apparent that distance and angular restrictions that are potentially unfavorable allow for a less energetically
favorable cation-p contact, rather than foregoing it entirely. This suggests that, to the extent tryptophan’s indole is involved in cation
solvation in a protein’s receptor site, it is a versatile donor.24

1.05.5.1 Solution Evidence for Cation–p Interactions

The indolylethyl side-armed compound (3-indolylCH2CH2< N18N>CH2CH2-3-indolyl) provided a vehicle to determine if
cation–p interactions were apparent in solution as well as in the solid state. Thus, the receptor was titrated with NaI in acetone-
d6 and changes in the nuclear magnetic resonance or NMR spectra were recorded. The nuclear Overhauser effects (NOEs) initially
observed suggested an extended (uncomplexed) conformation for the receptor. When an equivalent of NaI was present, the NOEs
observed were in concert with the complexed structure shown in Fig. 17.50
The observation of complexation by the indolylethyl receptor using its less electron-rich ring as the p donor, raised the question
of whether a single arene could afford a p complex. The two lariat ether receptors shown in Fig. 18 were prepared to address this
question.
Complexation was then attempted with KI, KB(C6H4Cl)4, and NaBPh4. The phenylethyl side-armed macrocycle formed
complexes with Kþ essentially as expected, fitting well in the 18-membered ring. 51 In the receptor having two phenylethyl side chains,
both apical positions in the Kþ complex were occupied by the arenes. In the KI case, one apex was occupied by the arene in an arrange-
ment essentially identical to that in the two-armed complexes. On the opposite side of the ring, Kþ was in contact with an iodide ion.
The centroid of the arene, the potassium cation, and the iodide anion were arrayed in a nearly linear fashion as shown in Fig. 19.
The complexes obtained with < 18N>CH2CH2Ph and KB(C6H4Cl)4 were similar in terms of the cation–p interaction. The
charged iodide ion was able to directly interact with the ring-bound cation. The negative charge in [B(C6H4Cl)4] is sterically inac-
cessible and the Kþ complex was stabilized by a direct interaction with a chlorine atom of the anion. 51
A different result was obtained when a complex between < 18N>CH2CH2C6H4OH and NaBPh4 was isolated. In this case, the
arene’s centroid was positioned over the ring-bound Naþ ion. However, the macroring was contracted and puckered to accommo-
date the shorter heteroatom-cation distances for Naþ compared to Kþ. The puckering made the second apex of the cation sterically
inaccessible to the bulky BPh 4 ion. Instead, the large anion was adjacent to the cation complex and did not show intimate contact.
52

1.05.5.2 Imidazole as a Sigma- or p Donor

The imidazole sidearm of histidine is relatively electron poor compared to benzene or indole. The heterocycle’s ability to form
cation–p complexes is further complicated by the potential competition with the nitrogen donor groups present in the ring. An

Figure 18 Structures of N-substituted aza-18-crown-6 lariat ethers having 2-phenylethyl and 2-(4-hydroxylphenyl)ethyl sidearms.

Figure 19 Potassium cation complexes of N-(2-phenylethyl)-aza-18-crown-6 in which the anions are iodide (left) and tetra(4-fluorophenyl)borate
(right).
 The Supramolecular Alkali Metal Cation–p Interaction 113

Figure 20 Structural drawing of N-(2-(2-imidazloyl)ethyl)aza-15-crown-5 and the solid-state structure of its complex with NaBPh4.

imidazole derivative having a 15-membered ring was prepared and complexed with sodium tetraphenylborate ( Fig. 20). In the
complex, the sodium cation was slightly above the plane of the macrocycle’s heteroatoms and the sterically hindered anion was
adjacent, but not in contact with the complex. Thus, the sigma donor atom in the ring was favored in this complex rather than
the electron-poor arene.

1.05.5.3 Control Studies for the Diaza-18-crown-6 Lariat Ether Complexes

It seemed possible that the early failure of the allyl and propargyl side-armed compounds to form complexes was the result of
experimental technique rather than geometry. Thus, two lariat ethers were prepared that differed from the successful
RCH2CH2< N18N>CH2CH2R pattern. These were RCH2CH2CH2< N18N>CH2CH2CH2R and RCH2CH2< N15N>CH2CH2R.
When an additional methylene was present in the 18-membered ring receptor, reaction with the KPF6 salt resulted in a complex
in which the ring-bound Kþ was axially solvated by three of the six fluorine atoms. In the 15-membered ring case, Kþ (from KI)
was bound in a “nesting” arrangement in the smaller macroring and formed an extended chain of  Kþ/I/Kþ/I. Thus,
one iodide anion bridged two bound Kþ ions.
Although less common in biology, a Ca2 þ-p complex is certainly possible in the larger supramolecular context. The
TyrCH2CH2< N18N>CH2CH2Tyr receptor system was treated with CaI2 and a solid-state structure was obtained. The cation, which
is closer in size to Naþ than Kþ, was bound in a somewhat contracted macrocycle. The two HOC6H4CH2CH2 sidearms were
extended from the macroring. Two molecules of water were in calcium’s ligand sphere, one of which was H bonded to iodide. 53

1.05.6 Double and Triple Bonds as p-Donor Groups

Given the fact that all three electron-rich aromatics showed cation–p interactions, the question became whether or not double or
triple bonds could function in the same way. As above, diaza-18-crown-6 molecules were prepared that had sidearms terminated by
either double or triple bonds and spaced from the macrocycle by an ethylene unit ( Fig. 21). As controls, double bonds were also
placed one carbon closer in a butylene side chain. Potassium iodide complexes of these lariat ethers were obtained as crystals with
the result that p-donor interactions were obvious in both cases.

1.05.6.1 Receptors and Other Experimental Studies

An early receptor study was reported by Ngola and Dougherty, who prepared water-soluble, toroidal host molecules differing in p
donicity. 54 Fig. 22 shows the receptors, which vary only by the identity of Y. In one case, Y ¼ four hydrogen atoms (benzene itself)

Figure 21 Double bond (left) and triple bond (right) side-armed macrocycles that formed Kþ complexes exhibiting apical cation–p interactions. The
center structure formed a Kþ complex, but the sidearms did not interact with the ring-bound cation.
114 The Supramolecular Alkali Metal Cation–p Interaction

Figure 22 A host-guest study to assess cation–p affinities with receptors having different Lewis basicities.

and in the second case, the same four positions were occupied by fluorine atoms. Since the cation–p interaction is largely electro-
static, it was anticipated that binding of electron-rich substrates would be stronger for the octafluorinated receptor and conversely,
electron-deficient guests would be better bound by the more Lewis basic host.
The guest molecules are illustrated in Fig. 22 to the right of the receptor. The compounds labeled Q1, Q2, and Ada all have
positively charged nitrogen atoms. Ada differs from Q1 and Q2 because it has no p system. Compounds Q2 and Q4 are methylated
quinolones that possess neutral nitrogen atoms and Ind is electron-rich N-methylindole. The strongest binding ( DG in
kcal mol 1) by the nonfluorinated receptor was observed for the charged arenes, Q1 (8.4), Q3 (7.2), and Ada (6.7). The lack of
any possible p–p interaction with Ada may have reduced binding, although the nearly spherical structure differs from that of
the flat quinolones. Binding was lower for the remaining compounds with the hydrogenated receptor and lowest with the fluori-
nated analog.
The compound shown in Fig. 22 as Ada is an adamantine derivative and the carbon framework is, therefore, nearly spherical.
Arnecke et al. prepared a family of calix[5]arenes, some of which were hybrid crown ether receptors (Fig. 23).55 They examined the
complexation of a variety of ammonium salts including Ada. The parent calix[5]arene was hybridized to form either a crown-5 or
and crown-6 macrocycle. The parent calix[5]arene and the hybrids were t-butylated along its aromatic periphery. The crown-6
hybrid was also de-t-butylated to form a five-benzene cycle that was more sterically accessible than when the alkyl chains were
present.
Solution complexation studies were conducted by NMR in CDCl3 or CD3COCD3 to determine which of several ammonium ions
was most strongly bound. An interesting feature of the work was an attempt to complex Ada. The five t-butyl groups prevented
complexation, but Ada was bound in the penta-aromatic collar when the sterically hindering groups were removed (center structure
in Fig. 23).
The need to understand the scope of cation–p interactions has led to efforts that involved receptor experiments, computational
studies, and spectroscopic analyses. 56,57 In part owing to the solid-state evidence shown earlier, experimental model systems were
designed to obtain structural and energetic information in solution. An early study conducted by Hunter and coworkers used pairs
of amides designed to form heterodimers by dint of H-bond formation.58 NMR analyses conducted in chloroform-detected chem-
ical shift differences in pairs of structurally modified diamides. Structural modifications to both the H-bond donors and H-bond
acceptors allowed for the determination of comparative affinities. In the cycle shown in Fig. 23, Hunter and coworkers evaluated
the energies of the various steps. From A to C, the energy was evaluated as DGA  DGC. Likewise, B to D was evaluated as DGB  DGD.
From A to B and from C to D, the energies are DGA  DGB and DGC  DGD, respectively. The cation–p interaction shown in the upper

Figure 23 Calix[5]arene parent (left) and structures having macrocyclic polyether pendants. The central and right hand figures are hybrid calixarene
crown ethers. They are identical except that the central structure is d-t-butlyated.
 The Supramolecular Alkali Metal Cation–p Interaction 115

left corner of Fig. 23 is, therefore, DDG ¼ DGA  DGB  DGC þ DGD. From these, it was concluded that when electron donating
substituents were present on a participating arene, the cation–p interaction was attractive by  8 kJ mol 1 ( 2 kcal mol 1). Elec-
tron withdrawing substituents led to a repulsive effect. The compounds involved in the cycle are shown in Fig. 24.
In work reported by Waters et al., a model compound possessing three aromatic rings in a flexible configuration was used to
probe cation–p interactions in water. 59 The cation in this case was a pyridinium salt that could orient itself to achieve the most
stable conformation. This was expected to be the one in which the cation–p interaction was strongest. The studies were conducted
in acidic D2O and monitored by NMR. The authors found that there were rotational preferences when the cation–p effect was maxi-
mized. In addition, alkylation of the nitrogen atom further enhanced the interaction energy. Fig. 25 shows the structure of the
compound studied at the left and the so-called syn and anti conformations in equilibrium at the right.
Hamilton and coworkers 60 prepared a tetrameric tetralin derivative as shown in Fig. 26. A stabilization mechanism was postu-
lated for this compound that would involve both salt bridge formation and cation–p stacking interactions. The structure shown at
the left of the lower line suggests a possible orientation for the interactions. Four peptides that are rich in acid (D) or aromatic (F, W)
side chains were prepared and are shown in the same figure as P1–P4. It was noted that P3 and P4 showed high affinity for the
tetramer suggesting that a stabilizing interaction occurred in the a-helical peptides.
A nearly simultaneous report from Waters et al. 61 reported two a-helical peptides (Fig. 27) in which the position of a phenyl-
alanine residue was varied. The other variations in the peptides are indicated by “X.” The substitutions were lysine (Lys, K), ornithine
(Orn), and diaminobutanoic acid (Dab). The three amino acids are analogs in which the side-chain amino group is extended from
the main chain by four, three, or two methylenes, respectively. The helicity of the peptide analogs was assayed by circular dichroism
and by NMR. It was concluded that a Phe–Orn interaction was most stabilizing and that  DG ¼ 0.4 kcal mol 1. The authors state
that the “results indicate that even the simplest cation–p interaction can provide significant stability to protein structure and
demonstrate the subtle factors that can influence the observed interaction energies in designed systems.”
An effort to distinguish the energies of hydrogen bonding and cation–p effects was conducted by Marshall and coworkers. 62
NOE NMR evidence suggested that in photoactivated rhodopsin, the GtR(340–350) peptide bound to it, is stabilized by

Figure 24 A thermodynamic cycle designed to evaluate cation–p effects.

Figure 25 Compounds designed to assay cation–p influence on conformation in aqueous solution.

116 The Supramolecular Alkali Metal Cation–p Interaction

Figure 26 Design of a cationic molecule that can interact with a-helical peptides using cation–p interactions.

Figure 27 a-Helical peptides studied for cation–p stabilization.

a cation–p interaction between a lysine (Lys341) and the aromatic ring of a phenylalanine (Phe350). Their work explored the
importance of this interaction by altering the substitution pattern on the Phe350 arene. The effect was assayed by changes in
receptor binding between the photoactivated rhodopsin and the peptide. It was found that the elements of a cation–p interaction
were present, but that its effect was minimal, owing to a salt bridge that shielded the ammonium group’s charge. Conversion of the
participating carboxyl to the carboxamide, enhanced the influence of the cation–p interaction. A conclusion of the work was that
estimations of cation–p interactions being stronger than salt bridge formation, may not be, considering all of the relevant structural
and electronic factors.
An early demonstration of cation–p interactions involving encapsulation by a polyarene receptor was reported by Murayama
and Aoki. 63 The vehicles for study were various calix[6]arenes (Fig. 28). Three structures are presented of Kþ complexes. In all
of them, there is extensive involvement between the cation and the phenolic hydroxyl groups. In one particular case, however,
a complex incorporating two Kþ ions shows a strong contact between the cation and an aromatic ring. The authors refer to the situ-
ation as the “ion rides on the phenyl ring.”
A metal-seamed nanocapsule that forms tight complexes with thallium was reported by Kumari et al. 64 Pyrogallol (1,2,3-
trihydroxybenzene) undergoes acid-catalyzed cyclization with aldehydes to form tetramers of the type shown in Fig. 29 as “A.”
Six tetramers self-assemble to form nanocapsules that enclose a significant volume of space. The capsules are held together by
a network of hydrogen bonds that may be replaced by appropriate metals to form metal-seamed nanocapsules. Structure B in
Fig. 29 shows a gallium-seamed nanocapsule in which each tetrameric group of arenes supports a thallium ion. The presence of
these ions, which are symmetrically disposed within the cages, was corroborated by 205Tl NMR studies.

Figure 28 The structure of calix[6]arene receptor without any cation within.

Figure 29 Cation–p complexation within metal-seamed nanocapsules.

 The Supramolecular Alkali Metal Cation–p Interaction 117

Two studies focused on ferrocene as a scaffold for the study of cation–p interactions. Wagner and coworkers 65 used stacked
ferrocenes to probe cation–p interactions with a range of alkali metal ions such as Liþ, Naþ, Kþ, Rbþ, and Csþ. In a sense, the
dicyclopentadienyl iron sandwich structure is a cation–p complex. However, its electron-rich aromatic surfaces were used in
the present case to explore its p donicity toward alkali metal cations. The authors used a combination of NMR, X-ray structural
analysis, electrochemistry, and computational methods to analyze the structures that formed. Three of the structures are shown in
Fig. 30. The two structures at the left of the figure show three types of cation stabilization, two of which (O: and N:) are
traditional.
An effort was made in the author’s lab to devise a ferrocene system 66 that would use the iron atom as an “atomic ball bearing.”67
The top line of Fig. 31 shows the anticipated interaction. It was predicted that the syn conformation would be favored by an ammo-
nium ion cation–p stabilization. The X-ray crystal structure revealed the water-linked structure shown on the lower line of Fig. 31. It
should be noted that Malezieux et al. also obtained a crystal structure of a ferrocene derivative that showed an ammonium arene
interaction in their studies of ionic hydrogenation.68
As noted earlier, Dunbar and coworkers 15 encapsulated various metal cations in a PhePhe cage. A synthetic cage was developed
in the author’s laboratory and its ability to complex cations was assayed by mass spectrometry.69 The compounds were constructed
from hydroquinone and 1,4-dichloro-2-butyne. The receptor, which was given the name pyxophane, was examined by mass spec-
trometry in the presence of either Naþ, Kþ, or both. The predominant ion observed in the presence of Naþ was a 1:1 host-guest
complex. In the presence of larger Kþ, a 2:1 host/guest complex predominated. The inference drawn from these studies was that
the complexes had the structures indicated in Fig. 32.

Figure 30 Alkali metal cation complexes in which ferrocene serves as a donor. The left and central structures show an lithium cation pi interaction
with ferrocene. In the right hand structure, M ¼ Liþ, Naþ, or Kþ.

Figure 31 Attempted observation of cation–p interactions using the rotational freedom of a ferrocene system.

Figure 32 Sodium cation fits within a single pyxophane molecule, but larger potassium cation is stabilized in the gas phase by two cation–p
interactions.
118 The Supramolecular Alkali Metal Cation–p Interaction

Figure 33 Intramolecular cation–p catalysis of a cyclization reaction.

Figure 34 Extramolecular cation–p catalysis of a cyclization reaction.

1.05.6.2 The Cation–p Interaction in Synthesis

Hydrogen bonding, orbital overlap, Lewis acid–base attraction, steric effects, and a wealth of other interactions have been used in
synthesis to affect the outcomes of chemical reactions. Applications of the cation–p interaction in synthesis have paralleled the
increasing understanding of this important supramolecular force. Yamada has offered a perspective on the value of this interaction
in a short review that appeared in 2007. 70 He calls attention to the fact that just as cation–p interactions influence protein tertiary
structure, it can influence the course of small-molecule reactions. Two recent examples illustrate the potential of cation–p interac-
tions in synthesis.
Aubé and coworkers 71 studied an azide-mediated asymmetric ring expansion reaction that benefitted from a strong cation–p
interaction. The sequence is shown in Fig. 33. When arene (benzene in the scheme) was present as a substituent, a strong interaction
was apparent with the cationic diazonium ion intermediate.
LePore and coworkers 72 reported that the cyclization reaction shown in Fig. 34 occurs more readily in the presence of an ammo-
nium ion catalyst. For example, the sodium-phenoxide-catalyzed cyclization reaction occurred when conducted in tetrahydrofuran
(THF) and had a half-life of  7500 min in the absence of Bu4NBr, compared to 33 min when one equivalent of the salt was present.
It was proposed that the ammonium ion polarized the alkyne by a cation–p interaction, making it more susceptible to attack by the
deprotonated amino group.

1.05.7 Literature Reviews

Several reviews of various aspects of cation–p interactions have appeared.73–77

1.05.8 Conclusion

The alkali metal cation–p interaction is a significant supramolecular force. The evidence suggests that it is largely an electrostatic
interaction, although other energetic contributors have been recognized. The force is greatest along a linear axis, but like hydrogen
bonds, its effects are manifested to a lesser extent off axis. The cation–p effect is strongest when solvent competition is absent or
weak. In this respect, it is like other supramolecular interactions. The strongest interaction energies were reported for gas phase
cation–p contacts studied by mass spectrometry. Numerous spectroscopic studies have been conducted in nonpolar or modestly
polar solvents and the cation–p effect is in evidence in all of these, although to varying extents. In aqueous solution, competition
with water and hydrogen bonding reduces the strength of the cation–p effect, which is thought generally to be similar in potency to
salt bridge formation.

References

1. Jeffrey, G. A.; Saenger, W. Hydrogen Bonding in Biological Structures; Springer-Verlag: Berlin, 1991, 569 pp.
2. Sutor, D. J. Nature 1962, 195, 68–69.
3. (a) Desiraju, G. R. Acc. Chem. Res. 1991, 24, 290–296; (b) Desiraju, G. Acc. Chem. Res. 1996, 29, 441–449.
 The Supramolecular Alkali Metal Cation–p Interaction 119

4. Desiraju, G. R.; Steiner, T. The Weak Hydrogen Bond in Structural Chemistry and Biology; Oxford University Press: Oxford, 2001, 526 pp.
5. Bowman-James, K. Acc. Chem. Res. 2005, 38, 671–678.
6. Pedersen, C. J. Science 1988, 241, 536–540.
7. Lehn, J. M. Prix Nobel 1988, 129–176.
8. Lehn, J.-M. Supramolecular Chemistry; VCH: Weinheim, 1995, 271 pp.
9. Pliska, V., Testa, B., Waterbeemd, H.v. d, Eds. Lipophilicity in Drug Action and Toxicology; VCH: Weinheim, 1996, 438 pp.
10. Sunner, J.; Nishizawa, K.; Kebarle, P. J. Phys. Chem. 1981, 85, 1814–1820.
11. Guo, B. C.; Purnell, J. W.; Castleman, A. W., Jr. Chem. Phys. Lett. 1990, 168, 155–160.
12. (a) Gapeev, A.; Yang, C. N.; Klippenstein, S. J.; Dunbar, R. C. J. Phys. Chem. A 2000, 104, 3246–3256; (b) Gapeev, A.; Dunbar, R. C. J. Am. Chem. Soc. 2001, 123,
8360–8365.
13. (a) Meot-Ner, M.; Deakyne, C. J. Am. Chem. Soc. 1985, 107, 469–474; (b) Meot-Ner, M.; Deakyne, C. J. Am. Chem. Soc. 1985, 107, 474–479.
14. Burley, S. K.; Petsko, G. A. FEBS Lett. 1986, 203, 139–143.
15. (a) Ryzhov, V.; Dunbar, R. C. J. Am. Chem. Soc. 1999, 121, 2259–2268; (b) Ryzhov, V.; Dunbar, R. C.; Cerda, B.; Wesdemiotis, C. J. Am. Soc. Mass Spectrom. 2000, 11,
1037–1046.
16. Dunbar, R. C.; Steill, J. D.; Oomens, J. J. Am. Chem. Soc. 2011, 133, 9376–9386.
17. Gilligan, J. J.; McCunn, L. R.; Leskiw, B. D.; Herman, Z.; Castleman, A. W., Jr. Int. J. Mass Spectrom. 2001, 204, 247–253.
18. (a) Vijay, D.; Zipse, H.; Sastry, G. N. J. Phys. Chem. B 2008, 112, 8863–8867; (b) Rao, J. S.; Zipse, H.; Sastry, G. N. J. Phys. Chem. B 2009, 113, 7225–7236.
19. Ma, J. C.; Dougherty, D. A. Chem. Rev. 1997, 97, 1303–1324.
20. (a) Mecozzi, S.; West, A. P., Jr.; Dougherty, D. A. Proc. Natl. Acad. Sci. 1996, 93, 10566–10571; (b) Mecozzi, S.; West, A. P., Jr.; Dougherty, D. A. J. Am. Chem. Soc. 1996,
118, 2307–2308.
21. Doyle, D. A.; Cabral, J. M.; Pfuetzner, R. A.; Kuo, A.; Gulbis, J. M.; Cohen, S. L.; Chait, B. T.; MacKinnon, R. Science 1998, 280, 69–77.
22. Heginbotham, L.; Lu, Z.; Abramson, T.; MacKinnon, R. Biophys. J. 1994, 66, 1061–1067.
23. MacKinnon, R. Angew. Chem. Int. Ed. Engl. 2004, 43, 4265–4277.
24. Wouters, J. J. Comput. Chem. 2000, 21, 847–855.
25. Wouters, J. Protein Sci. 1998, 7, 2472–2475.
26. (a) Tsuzuki, S.; Yoshida, M.; Uchimaru, T.; Mikami, M. J. Phys. Chem. A 2001, 105, 769–773; (b) Tsuzuki, S.; Uchimaru, T.; Mikami, M. J. Phys. Chem. A 2003, 107, 10414–
10418.
27. Biot, C.; Buisine, E.; Rooman, M. J. Am. Chem. Soc. 2003, 125, 13988–13994.
28. Mo, Y.; Subramanian, G.; Gao, J.; Ferguson, D. M. J. Am. Chem. Soc. 2002, 124, 4832–4837.
29. Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T. N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E. Nucleic Acids Res. 2000, 28, 235–242.
30. Henrick, K.; Thornton, J. M. Trends Biochem. Sci. 1998, 23, 358–361.
31. Ruan, C.; Yang, Z.; Hallowita, N.; Rodgers, M. T. J. Phys. Chem. A 2005, 109, 11539–11550.
32. Khandelia, H.; Kaznessis, Y. N. J. Phys. Chem. B 2007, 111, 242–250.
33. Wheeler, S. E.; Houk, K. N. J. Am. Chem. Soc. 2009, 131, 3126–3127.
34. (a) Lewis, M.; Bagwill, C.; Hardebeck, L. K.; Wireduaah, S. Comput. Struct. Biotechnol. J. 2012, 1. e201204004; (b) Wireduaah, S.; Parker, T. M.; Lewis, M. J. Phys. Chem. A
2013, 117, 2598–2604.
35. Marshall, M. S.; Steele, R. P.; Thanthiriwatte, K. S.; Sherrill, C. D. J. Phys. Chem. A 2009, 113, 13628–13632.
36. Gallivan, J. P.; Dougherty, D. A. J. Am. Chem. Soc. 2000, 122, 870–874.
37. Ansorg, K.; Tafipolsky, M.; Engels, B. J. Phys. Chem. B 2013, 117, 10093–10102.
38. Rapp, C.; Goldberger, E.; Tishbi, N.; Kirshenbaum, R. Proteins 2014, 82, 1494–1502.
39. Ozol, Y.; Vimba, S.; Levins, A. Krystallografiya 1962, 7, 362.
40. Hoffmann, K.; Weiss, E. J. Organomet. Chem. 1974, 67, 221–228.
41. Brooks, J. J.; Rhine, W.; Stucky, G. D. J. Am. Chem. Soc. 1972, 94, 7346–7351.
42. Üffing, C.; Köppe, R.; Schnöckel, H. Organometallics 1998, 17, 3512–3515.
43. (a) Corbelin, S.; Kopf, J.; Lorenzen, P. N.; Weiss, E. Angew. Chem. Int. Ed. Engl. 1991, 30, 825–826; (b) Corbelin, S.; Kopf, J.; Weiss, E. Chem. Ber. 1991, 2417–2422.
44. Hrncir, D. C.; Rogers, R. D.; Atwood, J. L. J. Am. Chem. Soc. 1981, 103, 4277.
45. Ting, A. Y.; Shin, I.; Lucero, C.; Schultz, P. G. J. Am. Chem. Soc. 1998, 120, 7135–7136.
46. (a) Schneider, H.-J.; Schiestel, T.; Zimmermann, P. J. J. Am. Chem. Soc. 1992, 114, 7698–7703; (b) Loewenthal, R.; Sancho, J.; Fersht, A. R. J. Mol. Biol. 1992, 224, 759–
770; (c) Kearney, P. C.; Mizoue, L. S.; Kumpf, R. A.; Forman, J. E.; McCurdy, A.; Dougherty, D. A. J. Am. Chem. Soc. 1993, 115, 9907–9919; (d) Schneider, H.-J. Chem. Soc.
Rev. 1994, 227–234.
47. Gokel, G. W.; Dishong, D. M.; Diamond, C. J. Chem. Commun. 1980, 1053–1054.
48. Arnold, K. A.; Hernandez, J. C.; Li, C.; Mallen, J. V.; Nakano, A.; Schall, O. F.; Trafton, J. E.; Tsesarskaja, M.; White, B. D.; Gokel, G. W. Supramol. Chem. 1995, 5, 45–60.
49. Arnold, K. A.; Viscariello, A. M.; Kim, M.; Gandour, R. D.; Fronczek, F. R.; Gokel, G. W. Tetrahedron Lett. 1988, 29, 3025–3028.
50. De Wall, S. L.; Meadows, E. S.; Barbour, L. J.; Gokel, G. W. J. Am. Chem. Soc. 1999, 121, 5613–5614.
51. Hu, J.; Barbour, L. J.; Gokel, G. W. Chem. Commun. 2002, 1808–1809.
52. Hu, J.; Barbour, L.; Gokel, G. W. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 5121–5126.
53. Hu, J.; Barbour, L. J.; Ferdani, R.; Gokel, G. W. Chem. Commun. 2002, 1806–1807.
54. Ngola, S. M.; Dougherty, D. A. J. Org. Chem. 1998, 63, 4566–4567.
55. Arnecke, R.; Bohmer, V.; Cacciapaglia, R.; Cort, A. D.; Mandolini, L. Tetrahedron 1997, 53, 4901–4908.
56. Lee, P. K.; Chapman, R. P.; Zhang, L.; Hu, J.; Barbour, L. J.; Elliott, E. K.; Gokel, G. W.; Bryce, D. L. J. Phys. Chem. A 2007, 111, 12859–12863.
57. Schlamadinger, D. E.; Daschbach, M. M.; Gokel, G. W.; Kim, J. E. J. Raman Spectrosc. 2011, 42, 633–638.
58. Hunter, C. A.; Low, C. M.; Rotger, C.; Vinter, J. G.; Zonta, C. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 4873–4876.
59. Rashkin, M. J.; Hughes, R. M.; Calloway, N. T.; Waters, M. L. J. Am. Chem. Soc. 2004, 126, 13320–13325.
60. Orner, B. P.; Salvatella, X.; Sanchez Quesada, J.; De Mendoza, J.; Giralt, E.; Hamilton, A. D. Angew. Chem. Int. Ed. 2002, 41, 117–119.
61. Tsou, L. K.; Tatko, C. D.; Waters, M. L. J. Am. Chem. Soc. 2002, 124, 14917–14921.
62. Anderson, M. A.; Ogbay, B.; Arimoto, R.; Sha, W.; Kisselev, O. G.; Cistola, D. P.; Marshall, G. R. J. Am. Chem. Soc. 2006, 128, 7531–7541.
63. Murayama, K.; Aoki, K. Inorg. Chim. Acta 1998, 1998, 36–42.
64. Kumari, H.; Jin, P.; Teat, S. J.; Barnes, C. L.; Dalgarno, S. J.; Atwood, J. L. J. Am. Chem. Soc. 2014, 136, 17002–17005.
65. Ilkhechi, A. H.; Mercero, J. M.; Silanes, I.; Bolte, M.; Scheibitz, M.; Lerner, H. W.; Ugalde, J. M.; Wagner, M. J. Am. Chem. Soc. 2005, 127, 10656–10666.
66. Hu, J.; Barbour, L. J.; Gokel, G. W. New J. Chem. 2004, 28, 907–911.
67. Li, C.; Medina, J. C.; Abel, E.; Maguire, G. E. M.; Atwood, J. L.; Gokel, G. W. J. Am. Chem. Soc. 1997, 119, 1609–1618.
68. Malezieux, B.; Gruselle, M.; Troitskaya, L. L.; Sokolov, V. I.; Vaissermann, J. Organometallics 1994, 13, 2979–2986.
69. Behm, R.; Gloeckner, C.; Grayson, M. A.; Gross, M. L.; Gokel, G. W. Chem. Commun. 2000, 2377–2378.
120 The Supramolecular Alkali Metal Cation–p Interaction

70. Yamada, S. Org. Biomol. Chem. 2007, 5, 2903–2912.

71. Katz, C. E.; Ribelin, T.; Withrow, D.; Basseri, Y.; Manukyan, A. K.; Bermudez, A.; Nuera, C. G.; Day, V. W.; Powell, D. R.; Poutsma, J. L.; Aube, J. J. Org. Chem. 2008, 73,
3318–3327.
72. Nagy, E.; St Germain, E.; Cosme, P.; Maity, P.; Terentis, A. C.; Lepore, S. D. Chem. Commun. (Camb.) 2016, 52, 2311–2313.
73. (a) Ma, J. C.; Dougherty, D. A. Chem. Rev. 1997, 97, 1303–1324; (b) Gallivan, J. P.; Dougherty, D. A. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 9459–9464.
74. Gokel, G. W.; Barbour, L. J.; Ferdani, R.; Hu, J. Acc. Chem. Res. 2002, 35, 878–886.
75. Schneider, H. J.; Yatsimirsky, A. K. Chem. Soc. Rev. 2008, 37, 263–277.
76. Mahadevi, A. S.; Sastry, G. N. Chem. Rev. 2013, 113, 2100–2138.
77. Schneider, H. J. Acc. Chem. Res. 2013, 46, 1010–1019.