Alkali metal cation - π Interactions observed by using a lariat ether model system

3092

J. Am. Chem. Soc. 2001, 123, 3092-3107

Alkali Metal Cation-π Interactions Observed by Using a Lariat Ether

Model System

Eric S. Meadows,^†Stephen L. De Wall,^†Leonard J. Barbour,^‡and George W. Gokel*^,†

Contribution from the Bioorganic Chemistry Program and Department of Molecular Biology &

Pharmacology, Washington UniVersity School of Medicine, 660 South Euclid AVenue, Campus Box 8103,

St. Louis, Missouri 63110, and Department of Chemistry, UniVersity of Missouri, 601 South College

AVenue, Columbia, Missouri 65211

ReceiVed August 16, 2000

Abstract: The Na⁺or K⁺cation-π interaction has been experimentally probed by using synthetic receptors

that comprise diaza-18-crown-6 lariat ethers having ethylene sidearms attached to aromatic π-donors. The

side chains are 2-(3-indolyl)ethyl (7), 2-(3-(1-methyl)indolyl)ethyl (8), 2-(3-(5-methoxy)indolyl)ethyl (9), 2-(4-

hydroxyphenyl)ethyl (10), 2-phenylethyl (11), 2-pentafluorophenylethyl (12), and 2-(1-naphthyl)ethyl (13).

Solid-state structures are reported for six examples of alkali metal complexes in which the cation is π-coordinated

by phenyl, phenol, or indole. Indole-containing crown, 7, adopts a similar conformation when bound by NaI,

KI, KSCN, or KPF₆. In each case, the macroring and both arenes coordinate the cation; the counteranion is

excluded from the solvation sphere. NMR measurements in acetone-d₆solution confirm the observed solid-

state conformations of unbound 7 and 7‚NaI. In 7‚Na⁺and 7‚K⁺, the pyrrolo, rather than benzo, subunit of

indole is the π-donor for the alkali metal cation. Cation-π complexes were also observed for 10‚KI and

11‚KI. In these cases, the orientation of the cation on the aromatic ring is in accord with the binding site

predicted by computational studies. In contrast to the phenyl case (11) the pentafluorophenyl group of 12

failed to coordinate K⁺. Solid-state structures are also reported for 7‚NaPF₆, 10‚NaI, 11‚NaI, 13‚KI, 13‚KPF₆,

and 9‚NaI, in which cation-π complexation is not observed. Steric and electrostatic considerations in the

π-complexation of alkali metal cations by these lariat ethers are thought to account for the observed complexation

behavior or lack thereof.

Introduction

confirmed these findings and extended them to Na⁺. Hoffmann

and Weiss had, even earlier, reported a solid-state structure of

KBPh₄in which K⁺was cradled between two benzene rings.⁶

Atwood and co-workers reported a structure in which an apical

interaction between a molecule of benzene and a crown-

complexed K⁺cation appeared to stabilize the latter.⁷Our own

efforts to develop lariat ether receptors that could be used to

assess π-interactions between alkali metal cations and double

or triple bonds appeared encouraging at first but thermodynamic

and solid-state analyses ultimately failed to confirm them.⁸

Novel solid-state studies in the cation-π complexation arena

continue to emerge.⁹

Interest in cation-π interactions was stimulated in the 1990’s

by a postulate that the aromatic side chains of amino acids might

determine K⁺transport selectivity in transmembrane protein

channels.¹⁰Site-directed mutagenesis studies by Heginbotham

and MacKinnon permitted a test of this postulate by removal

of the hydroxyl group from a critical tyrosine residue. The

phenol (Tyr) to phenyl (Phe, i.e. Y f F) mutation removed the

hydroxyl group but left the arene in tact. Cation transport

selectivity was altered in the mutant even though the arene was

Prior to the late 1960’s, the experimental study of alkali metal

cation chemistry was difficult at best. The advent of crown

ethers¹and cryptands²provided a vehicle for such studies and

the field has since burgeoned. Numerous studies have defined

the π-donor abilities of a variety of elements such as O, N, and

S toward alkali metal cations. Assessing the ability of π-donors

to interact with alkali metals has proved elusive. The importance

of alkali metal cation-π interactions is potentially great,

especially in biological systems where nearly 10% of amino

acid side chains are terminated by the aromatic residues benzene

(in phenylalanine, Phe, F), phenol (in tyrosine, Tyr, Y), and

indole (in tryptophan, Trp, W).

Experimental evidence for cation-π interactions dates from

the pioneering work of Kebarle and co-workers,³who showed

that the coordination of K⁺by a molecule of benzene or of

water in the gas phase was approximately isoenergetic. More

recent work from the laboratories of Castleman⁴and Lisy⁵has

^†Washington University School of Medicine.

^‡University of Missouri.

(1) Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 2495, 7017.

(2) (a) Dietrich, B.; Lehn, J. M.; Sauvage, J. P. Tetrahedron Lett. 1969,

2885. (b) Lehn, J. M.; Sauvage, J. P. J. Chem. Soc., Chem. Commun. 1971,

440.

(6) Hoffmann, K.; Weiss, E. J. Organomet. Chem. 1974, 67, 221-8.

(7) (a) Atwood, J. L.; Crissinger, K. D.; Rogers, R. D. J. Organomet.

Chem. 1978, 155, 1-14. (b) Hrncir, D. C.; Rogers, R. D.; Atwood, J. L. J.

Am. Chem. Soc. 1981, 103, 4277-8. (c) Atwood, J. L. J. Inclusion Phenom.

1985, 3, 13-20.

(3) Sunner, J.; Nishizawa, K.; Kebarle, P. J. Phys. Chem. 1981, 85,

1814-1820.

(4) Guo, B. C.; Purnell, J. W.; Castleman, A. W., Jr. Chem. Phys. Lett.

1990, 168, 155-160.

(8) Arnold, K. A.; Viscariello, A. M.; Kim, M.-S.; Gandour, R. A.;

Fronczek, F. R.; Gokel, G. W. Tetrahedron Lett. 1988, 3025-3028.

(9) King, B. T.; Noll, B. C.; Michl, J. Collect. Czech. Chem. Commun.

1999, 64, 1001-1012.

(5) (a) Cabarcos, O. M.; Weinheimer, C. J.; Lisy, J. M. J. Chem. Phys.

1998, 108, 5151-5154. (b) Cabarcos, O. M.; Weinheimer, C. J.; Lisy, J.

M. J. Chem. Phys. 1999, 110, 8429-8435.

(10) Kumpf, R. A.; Dougherty, D. A. Science 1993, 261, 1708-1710.

Published on Web 03/08/2001

Alkali Metal Cation-π Interactions

J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3093

still present.¹¹The recently reported solid-state structure of the

KcsA K⁺channel of Streptomyces liVidans shows that the

tyrosine hydroxyl group rather than the arene plays a key

structural role.¹²

clearly confirms the cation-π interaction of benzene, indole,

and phenol with K⁺. Fewer data are available for the Na⁺-

arene interaction but this is demonstrated as well.

The importance of cation-π interactions¹³looms large in

biology. In the nonpolar interior of a protein, such interactions

may be significant. The three amino acids phenylalanine,

tyrosine, and tryptophan have abundances in all known proteins

of 3.9%, 3.2%, and 1.3%, respectively. In any given protein,

8.4% of the amino acids may be involved in cation-π

interactions. This means that a protein having 250 amino acids

could theoretically have as many as 21 alkali metal-arene

contacts in a single molecule. The potential importance of such

interactions and the need to understand them at the molecular

level is clear.

Results

Receptors Used in This Study. Compounds 2-13 are

bibracchial (2-armed) lariat ethers.²⁰All are diaza-18-crown-6

derivatives that have two identical sidearms. The crown ether

system was chosen because it affords a well-characterized

binding site for alkali metal cations such as Na⁺and K⁺. The

receptor sidearms were designed to be sufficiently flexible to

turn either inward or outward depending on the system’s

demands. Compounds 3-6 have single carbon spacers con-

necting the macroring to an unsaturated residue. Receptors 2-6

were studied previously and found not to exhibit cation-π

complexation as judged by solution thermodynamic and X-ray

diffraction studies.²¹A subsequent study of molecular models

suggested that a 2-carbon chain would provide optimal spacing

for the formation of an intramolecular π-complex. Three

receptors in particular, 7, 10, and 11, were chosen because the

pendant residues correspond to the side chains of the aromatic

amino acids tryptophan (w 7), tyrosine (w 10), and phenyla-

lanine (w 11).

An important and related question is that of ammonium-π

interactions, discussed in an early article by Kier and Aldrich.¹⁴

They attempted to understand the effect of ammonium ion

interactions on the binding of drugs to protein receptors and in

model systems such as the methylamine‚indole complex. In a

second report,¹⁵influenced by earlier proposals by O’Brien,¹⁶

they conducted calculations to assess the binding of small

molecule models to acetylcholinesterase. The calculated binding

energy was then used to compute a theoretical catalytic rate,

which was compared to the experimentally determined value.

By studying charged and hydrophobic substrates, they found

that an anionic binding site could not account for the experi-

mentally observed rates. Instead, the best agreement with

experiment was obtained by using a model in which aromatic

residues dominated the binding of the tertiary nitrogen of

acetylcholine. The presence of the aromatic binding site was

confirmed by the X-ray structure determination of the enzyme¹⁷

and inhibitor-enzyme complexes.¹⁸The stabilization mecha-

nism was thought to involve induced dipoles rather than

cation-π interactions, but the relevance of this effort is obvious.

A similar binding mechanism has been suggested for the

nicotinic acetylcholine receptor, which is an ion channel.¹⁹In

the work described below, we report experimental evidence that

Choice of Crown Ether. A decision was made at an early

stage to use a substituted diazacrown as the receptor’s basic

scaffold. Sidearms attached to nitrogen may complex from the

same or opposite sides of the macroring due to nitrogen’s ability

to undergo rapid inversion. If the sidearms were attached at

carbon, the receptor would exhibit “sidedness”. There is also

(18) (a) Silman, I.; Harel, M.; Axelsen, P.; Raves, M.; Sussman, J. L.

Biochem. Soc. Trans. 1994, 22, 745-9. (b) Eichler, J.; Anselmet, A.;

Sussman, J. L.; Massoulie, J.; Silman, I. Mol. Pharmacol. 1994, 45, 335-

40. (c) Harel, M.; Schalk, I.; Ehret-Sabatier, L.; Bouet, F.; Goeldner, M.;

Hirth, C.; Axelsen, P. H.; Silman, I.; Sussman, J. L. Proc. Natl. Acad. Sci.

U.S.A. 1993, 90, 9031-9035. (d) Ripoll, D. R.; Faerman, C. H.; Axelsen,

P. H.; Silman, I.; Sussman, J. L. Proc. Natl. Acad. Sci. U.S.A. 1993, 90,

5128-32.

(19) Zhonge, W.; Gallivan, J. P.; Zhang, Y.; Li, L.; Lester, H. A.;

Dougherty, D. A. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 12088-12093

and references therein.

(20) (a) Gokel, G. W. Chem. Soc. ReV. 1992, 21, 39-47. (b) Gokel, G.

W.; Schall, O. F. Lariat Ethers. In ComprehensiVe Supramolecular

Chemistry; Elsevier: Oxford, 1996; pp 97-152.

(21) Arnold, K. A.; Viscariello, A. M.; Kim, M.; Gandour, R. D.;

Fronczek, F. R.; Gokel, G. W. Tetrahedron Lett. 1988, 3025-3028.

(11) Heginbotham, L.; Lu, Z.; Abramson, T.; MacKinnon, R. Biophys.

J. 1994, 66, 1061.

(12) 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.

(13) Ma, J. C.; Dougherty, D. A. Chem. ReV. 1997, 97, 1303-1324.

(14) Kier, L. B.; Aldrich, H. S. J. Theor. Biol. 1974, 46, 529-41.

(15) Hoeltje, H. D.; Kier, L. B. J. Pharm. Sci. 1975, 64, 418-20.

(16) O’Brien, R. D. Med. Chem. Ser. Monogr. 1971, 11, 161-212.

(17) Sussman, J. L.; Harel, M.; Frolow, F.; Oefner, C.; Goldman, A.;

Toker, L.; Silman, I. Science 1991, 253, 872-879.

3094 J. Am. Chem. Soc., Vol. 123, No. 13, 2001

Meadows et al.

an inherent economy in such structures since the nitrogen atom

serves as both a pivot atom and a macroring donor group.

Either 15- or 18-membered diazacrowns²²could have been

the basic subunit for the receptor. The choice of the 18-

membered ring system was made based on three issues. First,

sidearms attached at the macroring nitrogens of diaza-18-

crown-6 can interact in a symmetric fashion that is not possible

for the 15-membered-ring compounds. Second, the presence of

a sixth donor in the 18-membered ring generally enhances the

cation-binding strength for alkali metal cations. This increase

is generally an order of magnitude in anhydrous CH₃OH solution

and more in less polar solvents. Third, many more crystal

structures are known for 18-membered-ring complexes than for

the 15-membered relatives. This may be due to the fact that

synthetic access is better for the 18-membered-ring systems and

more receptors are available. In our experience the 18-

membered-ring complexes generally crystallize more readily

than do the 15-membered-ring analogues. In any event, the

existence of more 18-membered-ring examples makes for better

comparisons.

Linker Length. Early studies,⁸in which we sought to identify

interactions of alkali metal cations with double or triple bonds

or benzene, used one-carbon spacers between the arene and

macroring. Thus the sidearms attached to nitrogen in 4,13-diaza-

18-crown-6 were allyl (3), propargyl (4), and benzyl (5),

respectively. Solid-state structures of 3-6 complexing Na⁺or

K⁺failed to show sidearm involvement. Successful alkali metal

complexation was achieved when, for example, an oxygen donor

was placed 2 carbons distant from the nitrogen. In the present

studies 2-carbon, rather than 1-carbon, spacer units were favored.

The use of 3-carbon linker arms was considered but deferred

until results were obtained for the 2-carbon spacer systems, 7,

10, and 11. The 3-carbon linker would provide additional

extension if required but could potentially suffer from unfavor-

able conformational (gauche) interactions in the linker chain.

We felt it best to avoid unnecessary conformational variables

at the early stage of the project.

Molecular models (CPK) were prepared of compounds 1-13.

For the present study, particular attention was paid to macro-

cycles in which the sidearms corresponded to aromatic amino

acid side chains: 2-(3-indolyl)ethyl (7), 2-(4-hydroxyphenyl)-

ethyl (10), and 2-phenylethyl (11). The design of the system

required a cation (Na⁺or K⁺) to be bound by the macroring.

The sidearm needed to be long enough and sufficiently flexible

to permit the arene to occupy the vacant apical coordination

sites. We assumed that the coordination sphere would be a

hexagonal bipyramid in which the crown ether affords the

equatorial donors.

For compounds 7, 10, and 11, this was the case. In the 2-(3-

indolyl)ethyl (7) case, a further issue was considered. The ring-

bound cation’s apical position could be occupied by either the

benzene or pyrrole centroid of indole. Theoretical calculations

suggested that the most favorable electrostatic potential is found

in the benzene, rather than the pyrrole ring; the former is

predicted to be the preferred donor site. Molecular models

suggested that a conformation in which the benzene rings could

serve as apical donors on opposite sides of the macroring was

readily accessible.

Scheme 1

synthetic receptor prepared, 7.²⁴More recently, phenol and

phenyl, the side chains of Tyr and Phe, have been incorporated

into receptors (10 and 11, respectively).²⁵Pentafluorophenyl

derivative 12 was designed to maintain the aromaticity of 11

but to reverse the arene’s electrostatic potential surface (see

below), thereby reducing the sidearm’s π-donicity. Compounds

8 (N-methylindolylethyl sidearms), 9 (5-methoxyindolylethyl

sidearms), and 13 (1-naphthylethyl sidearms) were all designed

to be analogues of 7. N-Methyl- and 5-methoxyindole deriva-

tives 8 and 9 are obvious derivatives of 7. The 5-methoxyindole

receptor, 9, was designed to increase the cation-π donicity of

indole. The N-Me indole receptor (8) was prepared to probe

the effect of preventing the indole from participating in a

hydrogen bond with an anion or solvent. In 13, the pyrrolo ring

of 7 is replaced by a second benzene ring. In a sense, 13 is an

analogue of both 7 and 11. The centroids of the two six-

membered rings of naphthalene provide two equivalent π donor

sites on each sidearm.

Synthetic Access. Compound 1 is commercially available

and the syntheses of 2-5⁸and 6²⁶were previously reported.

Structures 7-13 were prepared either by alkylation of 1 or by

using the incipient sidearm’s amino group to form the macro-

cycle directly. Dibenzyldiaza-18-crown-6, for example, can be

prepared by reaction of benzylamine with I(CH₂CH₂O)₂CH₂-

CH₂I in the presence of K₂CO₃and KI.²⁷The approaches are

shown in Scheme 1 and detailed in the Experimental Section.

Crystal Structures. All compounds were characterized by

traditional chemical methods but X-ray crystallography was the

main focus of the present study. Crystals of alkali metal cation

complexes suitable for X-ray analysis were generally obtained

by mixing equimolar amounts of alkali metal salt with the

receptor in a polar organic solvent such as acetone. In some

cases, crystals were obtained by slow evaporation of the solvent.

In other cases, a nonpolar solvent such as diethyl ether was

introduced and crystals formed by vapor diffusion. Data were

collected by using either a 4-circle or CCD diffractometer.

Details of crystallization, data collection, and analyses are

presented in the Experimental Section.

Solid-State Conformations of Free Receptors. Solid-state

structures were obtained for the free receptor molecule whenever

possible. The results of X-ray diffraction studies of 7, 10, and

12 are shown in Figure 1. Compound 12 is the analogue of 11

in which the aromatic hydrogen atoms have been replaced by

(23) Mecozzi, S.; West, A. P., Jr.; Dougherty, D. A. J. Am. Chem. Soc.

1996, 118, 2307-2308.

(24) De Wall, S. L.; Meadows, E. S.; Barbour, L. J.; Gokel, G. W. J.

Am. Chem. Soc. 1999, 121, 5613-5614.

Sidearm Structural Features. Computational studies have

shown that among the aromatic amino acids (His, Phe, Trp,

Tyr), tryptophan (i.e. indole) has the greatest cation-π donic-

ity.²³Thus indole was used as the π-donor system in the first

(25) De Wall, S. L.; Barbour, L. J.; Gokel, G. W. J. Am. Chem. Soc.

1999, 121, 8405-8406.

(26) Gatto, V. J.; Miller, S. R.; Gokel, G. W. Org. Synth. 1989, 68, 227-

233.

(22) Gatto, V. J.; Arnold, K. A.; Viscariello, A. M.; Miller, S. R.; Morgan,

C. R.; Gokel, G. W.; J. Org. Chem. 1986, 51, 5373-5384.

(27) Gatto, V. J.; Gokel, G. W. J. Am. Chem. Soc. 1984, 106, 8240-

8244.

Alkali Metal Cation-π Interactions

J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3095

Table 1. Ethylene Dihedral Angles in the Unbound Receptors

(in deg)

compd

Crown ether ethylene units

sidearm

no.^a

7

68.3 -66.4 -175.3 -68.3 66.4 175.31 66.6 -166.6

7a^b

62.6

71.9 -150.0 -62.6 -71.9 150.01 61.5 -161.5

71.6 -58.5 -52.4 -176.2 53.3 160.4 -166.2

69.4 -63.5 -62.2 -69.4 63.5 161.5 -161.5

7b^b64.2

7c^b

10

12

62.2

79.5 -70.0 -178.0 -79.5

70.0 178.0 178.3 -178.3

68.1 163.7 168.2 -168.2

73.6 -68.1 -163.7 -73.6

^aSee Figure 1. ^bAlternative structure of 7 observed in structures

not shown (see text).

The Phenol-Sidearmed Receptor, 10. The conformations

of unbound phenol receptor 10 and that of 7 (Figure 1, panels

b and a) are essentially identical. In both cases, the macrorings

are in the “parallelogram” arrangement and all bond distances

and angles are typical of that conformation. The sidearm

ethylene groups are antiperiplanar in 10 and are oriented in

approximately opposite directions extending from the macro-

cycle. The phenolic hydroxyl groups of 10 form intermolecular

O-H‚‚‚N hydrogen bonds (D_O-N) 2.76 Å) with adjacent

molecules of 10 (not shown).

Pentafluorophenyl Receptor 12. Receptor 12 was prepared

to assess the effect of reduced electron density in the arene (see

below). It was expected to be structurally similar to 11 because

the van der Waals radii of hydrogen and fluorine are 1.2 and

1.35 Å, respectively. Otherwise, all of the atoms in receptors

11 and 12 are identical. We assume that (if available) the

structure of 11 would be similar to that obtained for 12 (Figure

1, panel c). The structure of 12 is similar to expectation and to

those observed for 7 (Figure 1a) and 10 (Figure 1b). One

interesting feature of noncomplexed 12 is that the fluorinated

phenyl groups are involved in intermolecular edge-face π-stack-

ing (apparent in the unit cell, not shown).

Alkali Metal Complexes of 7, 10, 11, and 12. Solid-state

structures were obtained for 7‚NaI, 7‚KI, 7‚KSCN, 7‚KPF₆,

10‚NaI, 10‚KI, 11‚NaI, 11‚KI, and 12‚KI. An additional

complex structure, 7‚HPF₆, was obtained that has a conformation

essentially identical to that of 7‚PF₆. The K⁺complexes of

7, 10, and 11 exhibit similar structural features. Complexes

7‚KPF₆, 7‚NaI, 7‚KSCN, 7‚KI, 10‚KI, and 11‚KI are shown

using the CPK metaphor in Figure 2, panels a and b. The

structures of unbound 7 (Figure 2, panel a, middle), 7‚NaI

(Figure 2, panel a, right), and 7‚KPF₆(Figure 2, panel a, left)

are shown as examples of the general structural relationship

between the unbound and bound states of the synthetic receptors

that form cation-π complexes.

Receptor Conformational Changes Associated with Cation

Binding. In those cases in which direct comparisons can be

made, i.e. 7 f 7 NaI, 7 f 7‚KI, 7 f 7‚KSCN, 7 f 7‚KPF6,

10 f 10‚KI, and (12) f 11‚KI, the sidearms fold back onto

the macroring to make π-contacts with Na⁺or K⁺when

complexation occurs. The ethylene sidearms adopt a gauche

conformation as they reach above and below the mean plane

of the crown. The arenes occupy the apical coordination sites

of the cation and fully envelop it, excluding the counteranion,

solvent, or adventitious water. In the four alkali metal complexes

of 7, the anions are H-bonded to the indole nitrogen (N1).

Likewise, the phenolic hydroxyl of 10 forms an H-bond to iodide

in 10‚KI.

Figure 1. Conformations observed for noncomplexed 7 (a), 10 (b),

and 12 (c).

fluorine atoms. Although 11 did not crystallize, compound 12

is presented here as a surrogate with the qualification that H

and F are electronically different albeit similar in size. Different

conformations of 7 were observed within the same crystal.

Structure determinations on crystals of either 10 or 12 revealed

nearly identical conformations as shown in panels b and c of

Figure 1. The macroring conformations observed for these

macrocycles are similar to those reported for other crowns of

similar size.^28,29Dihedral angle data for 7, 10, and 12 are

recorded in Table 1.

Receptor 7. Two different crystal forms of receptor 7 were

isolated; two independent molecules of 7 were present in the

crystallographic asymmetric unit. Thus, 7 adopts at least four

different solid-state conformations but the macrocycle confor-

mation in each case was typical of diaza-18-crown-6 derivatives.

Only one of these is shown in panel a of Figure 1. In all cases,

the macrocycle adopts a “parallelogram arrangement” in which

methylenes on opposite sides of the macrocycle rotate inward

to occupy the void space. In all cases, the sidearms turn away

from each other and no intramolecular interactions were noted

between the sidearm and the crown for unbound 7.

Cation-π Sandwich Complexation of Alkali Metal Cat-

ions. The macrocycle and linkers have been deleted in the

structures illustrated in Figure 2, panel c, to demonstrate the

π-sandwich arrangement between the arenes and cation. Both

(28) Dunitz, J. D.; Seiler, P. Acta Crystallogr. Sect. B 1974, B30, 2739-

2740.

(29) Dale, J. Isr. J. Chem. 1980, 20, 3-11.

3096 J. Am. Chem. Soc., Vol. 123, No. 13, 2001

Meadows et al.

Figure 2. (a) Representations of 7 in the CPK metaphor unbound (center) and complexed by KPF₆(left) or NaI (right). (b) Solid-state structures

of the cation-π complexes 7‚KSCN, 7‚KI, 10‚KI, and 11‚KI. (c) Cross-section in the side view of arenes, cations, and anions for the complexes

7‚NaI, 7‚KPF₆, 7‚KI, 10‚KI, and 11‚KI. (d) Top and cutaway view showing the relation of the cation position to arene for the complexes shown

in part c, above. (e) Electrostatic potential surfaces and chemical structures for arenes.

of the apical coordination sites of Na⁺or K⁺are fully occupied

by arenes. The counterion is excluded from the coordination

sphere. In all cases, the arenes on opposite sides of each cation

are parallel to each other and equidistant from the cation (see

Figure 2, panel c). The specific distances for each sandwich

complex are shown in Table 2. Molecular interactions in the

three K⁺‚indolyl receptor complexes (7‚KI, 7‚KSCN, and

7‚KPF₆) are essentially identical (see below).

The distances observed for the cation-arene contacts in the

solid state are 3.50 and 3.46 ( 0.03 Å for the 7‚Na⁺and 7‚K⁺

complexes, respectively, and are summarized in Table 3. Ionic

radii for 8-coordinate Na⁺and K⁺are reported to be 1.18 and

Alkali Metal Cation-π Interactions

J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3097

Table 2. Geometries of Cation-π Interactions

Table 4. Ethylene Dihedral Angles in Lariat Ether Complexes

distances (in Å)

angle (deg)

dihedral angles (deg) for ethylene units in the

complex

M⁺-arene arene-arene M⁺-anion arene-crown^a

macrocyclic ring^a

sidearm^b

7‚NaI

3.50

3.45

3.48

3.43

3.44

6.99

6.89

6.95

6.86

6.88

6.86

7.02

6.17

6.06

5.48^b

5.94

6.74

6.56

4.72^d

26.49

16.48

18.78

17.07

8.71

complex

1

2

3

4

5

6

1

2

7‚KI

4‚KBF₄

70.1 -65.8 69.0 65.2 -49.9 -55.3 n/a

n/a

7‚KPF₆

7‚KSCN

10‚KI

5‚KSCN^a65.8 -64.3 -50.6 -65.8 64.4 50.6 n/a

5‚KSCN^b62.2 -21.9 63.3 -62.2 21.9 -63.3 n/a

7‚NaI

7‚KI

7‚KPF₆

59.6 -61.5 -54.6 -59.6 61.5 54.6 67.7 -67.7

63.3 -60.7 -47.6 -63.3 60.7 47.6 62.0 -62.0

65.6 -61.1 -50.2 -65.6 61.1 50.2 64.6 -64.6

11‚KI

3.43

9.18

13.65

(7‚2H)²⁺(PF₆

)

23.20^c

-

^aAngle calculated between the macrocycle mean plane and the arene.

^bDistance from K⁺to the closest fluorine. ^cDistance from the hydrogen

atom on the protonated nitrogen to the pyrrole centroid. ^dDistance from

the hydrogen atom on the protonated nitrogen to the closest fluorine

7‚KSCN 63.4 -61.8 -47.3 -63.4 61.8 47.3 59.0 -59.0

(7‚2H)²⁺

-

)

2

59.2 -68.0 -48.9 -59.2 68.0 48.9 59.5 -59.5

-

(PF₆

10‚KI

11‚KI

12‚KI

63.7 -59.0 -48.8 -63.7 59.0 48.8 61.4 -61.4

62.9 -60.8 -47.9 -62.9 60.8 47.95 58.9 -58.9

63.3 -54.1 73.2 -63.3 -73.2 54.1 176.5 -176.5

Table 3. Geometry of the Cation-π Interaction Involving Indole

M⁺-R distance (Å)

^aEthylene units in the macroring are consecutively numbered.

^bEthylene units in the sidearms are arbitrarily numbered.

R

NaI

KI

KPF₆

KSCN

7‚K⁺structures is in close intramolecular contact with the C-H

of an ethylene group directed toward the benzo centroid (not

shown). Such C-H ‚‚‚π interactions have been classified as

weak “hydrogen bonds” and have received attention recently

as contributing factors in molecular recognition.³²The C-benzo

centroid distance is in the range 3.79-3.88 Å in the three

7‚K⁺structures. An intermolecular C-H ‚‚‚π contact is also

formed in 7‚KPF₆(D_C-centroid) 3.88 Å) while 7‚KSCN and

7‚KI are involved in intermolecular arene-arene contacts (not

shown).

M⁺-N(indole)

3.58

3.23

3.51

4.05

4.02

3.50

4.77

3.51

3.32

3.57

3.89

3.91

3.45

4.59

3.57

3.30

3.52

3.98

3.94

3.48

4.67

3.52

3.38

3.55

3.82

3.83

3.43

4.47

M⁺-C2(indole)

M⁺-C3(indole)

M⁺-C8(indole)

M⁺-C9(indole)

M⁺-pyrrolo centroid

M⁺-benzo centroid

1.51 Å, respectively.³⁰The van der Waals radius of an aromatic

carbon is reported to be 1.72-1.80 Å.³¹These values suggest

an arene thickness of 3.44-3.6 Å. Using an arene thickness

value of 3.5 Å, we would expect the Na⁺-indole distance to

be 1.75 Å + 1.18 Å ) 2.93 Å. The observed Na⁺-C2 distance

in 7‚NaI is 3.23 Å. Thus, the separation between the Na⁺and

the aromatic ring is approximately (3.23 - 2.93) ) 0.30 Å on

each side. For the three 7‚K⁺complexes, the corresponding

K⁺-C2 distances are 3.34 ( 0.04 Å. The sum of van der Waals

radii for an aromatic carbon and K⁺is (1.75 + 1.51) ) 3.26 Å.

The separation between indole and K⁺in the 7‚K⁺complexes

is therefore only about 0.1 Å.

The sidearms of 10 (ethylphenol) and 11 (ethylbenzene)

correspond to the side chains of tyrosine and phenylalanine and

are identical except for the presence of a hydroxyl group in the

former. In each complex, the benzene ring occupies an apical

position above and below the macroring to form a π-sandwich

complex with K⁺. The K⁺to arene centroid distances are 3.43

and 3.44 Å, respectively. The distance between the two arenes

including the complexed cation (arene-arene separation) is 6.88

Å in 10‚KI and 6.86 Å in 11‚KI. Cutaway views of the 10‚K⁺

and 11‚K⁺complexes are shown in panel c of Figure 2.

Position of the Na⁺and K⁺Cation with Respect to the

Arene’s Face in Complexes of 7, 10, and 11. The orientation

of K⁺and Na⁺with respect to the indole is shown in Figure 2,

panel d (four complexes of 7). The pyrrolo, rather than the

benzo, subunit serves as the π-donor in the four complexes of

7. Indole-M⁺distances are recorded for these complexes in

Table 3. In no case observed for complexes of 7 is the cation

aligned exactly with either the benzo or the pyrrolo centroid.

The closest contact between indole and either Na⁺or K⁺is

near C2 of the indole ring. The identity of neither the cation

(Na⁺or K⁺) nor anion (I^-, SCN^-, or PF₆^-) alters the favored

cation-π geometry with indole in these complexes of 7.

The cation-π interaction between 7 and either Na⁺or K⁺

in their respective complexes does not involve the benzene ring

but rather the pyrrole subunit. The benzo ring in all of the

The cutaway top views in Figure 2, panel d, show the

positions of the Na⁺and K⁺on the surface of the phenol and

phenyl rings, respectively (compound numbers above structure).

In 10‚KI and 11‚KI, the K⁺is positioned essentially over the

center of the aromatic ring. The position of K⁺with respect to

the aromatic ring is nearly identical in 10 and 11. In both cases,

K⁺is almost, but not exactly, in the center of the benzene ring.

The results of ab initio molecular orbital calculations are

shown in panel e of Figure 2. Electrostatic potential surfaces

(EPS’s) were mapped onto a surface of molecular electron

density and color coded for qualitative analysis. The red regions

are e-25 kcal/mol while the blue color denotes regions of

g+25 kcal/mol. Similar EPS’s have been used to predict the

calculated binding sites for Na⁺on aromatic surfaces.¹⁹

Macrocycle Conformation in Complexes of 7, 10, and 11.

The macrocycles are slightly distorted from the expected D_3d

conformation in 7‚NaI and the K⁺complexes of 7, 10, and 11.

The crown nitrogen atoms pucker slightly out of the plane and

extend toward the sidearms. The dihedral angles defined by the

ethylene bonds in the cation-π complexes of 7, 10, and 11 are

shown in Table 4. For comparison, corresponding angles for

dibenzyl complex 5‚KSCN and dipropargyl complex 4‚KBF₄

are also shown in Table 4. A “true” D_3dconformation³³was

observed for the previously reported KI complex of N,N′-bis-

(2-methoxyethyl)-4,13-diaza-18-crown-6 (14) in which both

sidearms are extensively involved in cation complexation.³⁴

Although there is some significant deviation (e.g. 7‚KSCN,

11‚KI), sidearm dihedral angles are generally near the expected

(32) Nishio, M.; Umezawa, Y.; Hirota, M.; Takeuchi, Y. Tetrahedron

1995, 51, 8665-701 and references therein.

(33) (a) Seiler, P.; Dobler, M.; Dunitz, J. D. Acta Crystallogr. 1974,

B30, 2744-5. (b) Dunitz, J. D.; Dobler, M.; Seiler, P.; Phizackerley, R. P.

Acta Crystallogr. 1974, B30, 2733-8. (c) Wipff, G.; Weiner, P.; Kollman,

P. J. Am. Chem. Soc. 1982, 104, 3249-58.

(34) (a) White, B. D.; Fronczek, F. R.; Gandour, R. D.; Gokel, G. W.

Tetrahedron Lett. 1987, 28, 1753. (b) Gandour, R. D.; Fronczek, F. R.;

Gatto, V. J.; Minganti, C.; Schultz, R. A.; White, B. D.; Arnold, K. A.;

Mazzochi, D.; Miller, S. R.; Gokel, G. W. J. Am. Chem. Soc. 1986, 108,

4078.

(30) Shannon, R. D. Acta Crystallogr., Sect. A 1976, A32, 751-767.

(31) Bondi, A. J. Phys. Chem. 1964, 68, 441-451.

3098 J. Am. Chem. Soc., Vol. 123, No. 13, 2001

Meadows et al.

Figure 3. Structures of 7‚NaI, 7‚KI, 7‚KSCN, and 7‚KPF₆shown in the tube metaphor. Dotted lines indicate cation-π and H-bond contacts.

value of 60°. The homology between the complexes of 7, 10,

and 11 is apparent from the fact that the standard deviations

observed for each of the macroring dihedral angles is <3 (data

shown in Table 4).

Na⁺compared to benzene‚K⁺(∆H ) -28.0 and -18.3 kcal/

mol, respectively, see below). Calculations have also found that

the attraction is greater between benzene and the smaller, more

charge dense, Na⁺compared to the larger K⁺.³⁶

Although the macrocycles in the cation-π complexes of 7,

10, and 11 are not in a precise D_3darrangement, the distances

between the donor atoms and the cations are typical. For

example, the K⁺-O distances for the 7‚K⁺complexes are in

the range 2.75 ( 0.1 Å. The K⁺-N distances are 3.02 ( 0.05

Å. The corresponding distances for 14‚KI (macroring in D_3d

conformation) are as follows: K⁺-O ) 2.83 ( 0.03 Å, K⁺-N

) 2.94 Å. Thus, the bond distances are typical for alkali metal

complexes of 18-membered crowns. In complexes exhibiting a

cation-π interaction with the arene, the ethylene unit in the

sidearm adopts a gauche conformation (see Table 4). In the

phenyl-terminated sidearm complexes, K⁺is also positioned

near the center of the macroring. In both 10‚KI and 11‚KI, the

macrocycle adopts a distorted D_3dconformation and the K⁺-O

(2.68-2.70 Å) and K⁺-N distances (3.04-3.06 Å) are as

expected.

In 7‚NaI, the distance between the centroids of the pyrrolo

fragments of the indole group is 6.99 Å. The iodide ion shown

in Figure 3 is excluded from the first solvation sphere of the

cation (Na⁺-I^-) 7.16 Å) but H-bonded to the indole nitrogen

(N1). In 7‚NaI, the arene is tilted 26.49° with respect to the

mean plane of the macrocycle. The tilting of the indole ring is

greater in 7‚NaI than in any of the other cation-π complexes.

The angle between the crown and the arene results in the shortest

Na⁺-C2 distance observed (see Table 3). The angle between

the crown and the indole ring is 16.48°, about 10° less tilted

than in 7‚NaI, reflecting this difference. In 7‚KPF₆, the indole

ring is canted by 18.78° with respect to the macrocycle’s mean

plane; this value is similar to that observed for 7‚KI. The angle

between the macrocycle mean plane and the indole ring in

7‚KSCN is 17.07°.

Differences in the Alkali Metal Cation Complexes. Despite

the numerous similarities in overall conformation in the

structures discussed above, there are several notable differences.

The unique features of each cation-π complex are described

below. Throughout this section, tube representations of each

structure are shown individually. Dashed lines in each figure

connect the cation with the aromatic group to show the bonding

axis between the π-system and the cation.

In all four cases, the counteranion is hydrogen bonded to the

indole nitrogen. Unlike the iodide ion, the thiocyanate ion in

7‚KSCN is disordered. This contrasts with the KSCN complex

of N,N′-bis(benzyl)-4,13-diaza-18-crown-6, 5, in which the

thiocyanate anion occupies the apical coordination sites of the

bound potassium cation.³⁷

Structures of 10‚KI and 11‚KI. The macrorings in 10‚KI

and 11‚KI are nearly planar but slightly distorted from the D_3d

conformation (see Figure 4). Heteroatom to cation distances are

as expected. The phenolic hydroxyl group is hydrogen bonded

to the iodide anion. The O-H‚‚‚I hydrogen bond angle is 170.5°.

The phenolic benzene ring is positioned nearly parallel to the

macrocycle’s mean plane; the two planes intersect at an angle

of 8.71° in 10‚KI and 9.18° in 11‚KI. The “centerline” distance

between the two aromatic centroids is 6.87 ( 0.01 Å and the

cation is positioned between them at their midpoint. Unlike

10‚KI, there is no group available to H-bond to iodide in

11‚KI. The iodide anion is positioned approximately coplanar

with the cation and macroring but excluded from the solvation

sphere.

Structures of 7‚NaI, 7‚KI, 7‚KSCN, and 7‚KPF₆. The

similarity in these four structures is apparent from Figure 3.

Several features are notable. First, the macrocyclic rings are in

the expected conformation although they are somewhat distorted

from D_3d. For the most part, the bond angles and distances are

typical. The N-N distance across the crowns is about 6.1 Å in

these four 7‚M⁺complexes. Typically, the M⁺-O distances

are shorter than the M⁺-N distances. Typical values are Na⁺-O

) 2.47 ( 0.01 Å and Na⁺-N ) 3.03 Å for 7‚NaI. The

corresponding K⁺-O and K⁺-N distances in 7‚KI are 2.70 (

0.06 and 3.06 Å, respectively. The dihedral angles (see Table

4) for the ethylene units in the crown are all near (60°, although

variations to ∼55° are observed.

The most interesting feature of 7‚NaI is the cation-π

interaction between the indole groups and Na⁺. The 7‚NaI

complex is the only example we have obtained of a cation-π

interaction involving Na⁺. This is surprising in light of the

experimental measurements^3,35in the gas phase that have

demonstrated the much stronger binding energy for benzene‚

(35) Guo, B. C.; Purnell, J. W.; Castleman, A. W., Jr. Chem. Phys. Lett.

1990, 168, 155-160.

(36) Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc. 1995, 117, 4177-

4178.

(37) Arnold, K. A.; Viscariello, A. M.; Kim, M.; Gandour, R. D.;

Fronczek, F. R.; Gokel, G. W. Tetrahedron Lett. 1988, 29, 3025-8.

Alkali Metal Cation-π Interactions

J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3099

122.2° rather than 180° as expected for a nearly planar, D_3d

macroring conformation. The two Na-N distances are 2.52 and

2.60 Å. The Na⁺-O distances are 2.35, 2.35, 2.38, and 2.47

Å. The sidearms in 10‚NaI extend upward, with the phenolic

residues contacting I^-at the apex of the structure. The two

phenolic oxygen atoms are about equidistant from iodide

(D_O‚‚‚I) 3.46 and 3.47 Å).

Structure of 11‚NaI. In 11‚NaI, the cation is bound by the

macroring but not by the sidearms. The solid-state structure of

11‚NaI (Figure 6) is similar to that of 10‚NaI and 5‚NaI. In

each case the macrocycle is puckered to more fully immerse

the cation in the macroring while the two sidearms extend

upward. There are six disordered water molecules present in

the asymmetric unit of 11‚NaI; the Na⁺cation is coordinated

by only one of them. Hydrogens could not be accurately placed

on the disordered water oxygen atoms but it appears that the I^-

counterion is close enough to form an I‚‚‚H-O hydrogen bond

to the Na⁺-coordinated water (3.67 and 3.78 Å). The Na⁺cation

in 11‚NaI is 7-coordinate and the Z-Na-Z angles are 67.9 (

0.8°. The sidearm donor groups are positioned on the same side

of the macroring (syn) rather than anti as in the π-coordinated

complexes of 7, 10, and 11.

Structure of 12‚KI. The solid-state structure of 12‚KI (Figure

7) revealed no sidearm coordination of the ring-bound K⁺ion

although the macroring conformation is similar to that of 11‚

KI. The K⁺to oxygen and to nitrogen distances are also similar

to those of 11‚K⁺. The sidearm ethylene groups are both in the

antiperiplanar (176.5°) conformation in 12‚K⁺. Because neither

arene in 12‚K⁺is involved in π-complexation, the K⁺ion is

accessible to the I^-counterion. Each iodide thus bridges two

molecules of 12‚K⁺to form an extended, linear (K⁺‚‚‚I^-)_n

network.

Structures of 13‚KI and 13‚KPF₆. Neither 13‚K⁺exhibits

cation-π complexation; the ring ethylenes are approximately

gauche. In both cases, the K⁺ion is coordinated by the six crown

ether donor groups (K⁺-O ) 2.77 Å, K⁺-N ) 3.08 Å) that

lie in a plane surrounding the cation (shown at the right of Figure

7). The apical positions are occupied by I^-, which also contacts

a neighboring K⁺to form a columnar (K⁺‚‚‚I^-)_nnetwork as

observed for 12‚KI. The naphthyl residues are oriented away

from the macrocycle (anti orientation of the sidearm ethylene

units). The naphthalenes form intermolecular edge-face contacts

with an adjacent 13‚KI complex. The disordered -CH₂CH₂-

naphthyl groups in one molecule of 13‚KI were successfully

modeled (R1 ) 7.6%) to occupy two positions, each with half

occupancy. One of the two positions is shown as an ellipsoid

plot while the other is represented by the dotted lines (see Figure

7). Crystals of 13‚KPF₆were disordered but were found to be

isostructural with 13‚KI. The structure was solved only to the

extent that the conformation of the crown and the orientation

of the sidearms were apparent.

Figure 4. Structures of 10‚KI and 11‚KI shown in the tube metaphor.

Dotted lines indicate cation-π and H-bond contacts.

Structures of (7‚2H)²⁺(PF₆^-)₂, 7‚NaPF₆, and 9‚NaI. The

three complexes shown in Figure 5 have unusual aspects.

Diprotonated 7 is similar to 7‚KPF₆although no alkali metal

cation is present. The two macroring NH bonds point toward

-

the center of the macroring. One of the PF₆counterions is

H-bonded to the indole NH, which makes contact with the

edges of the aromatic rings (not shown). The sidearms of

(7‚2H)²⁺(PF₆^-)₂are positioned above and below the positively

charged crown and the angle between the mean plane of the

macrocycle and either arene is 13.65°. The pyrrolo centroids

are separated by 7.02 Å.

-

Like the other PF₆complexes in this study, the anion in

7‚NaPF₆is H-bonded to indole. The indole rings, however, do

not π-complex Na⁺. The macroring is contracted; the transan-

nular N-N distance is 4.96 Å compared to >6 Å in other

complexes. Likewise, compound 9 does not form a Na⁺-π

complex in the solid state. Instead, the indole rings contact the

iodide ion through N-H‚‚‚I^-hydrogen bonds (d_N-I) 3.89,

3.57 Å). The macroring N-N separation in 9‚NaI is 4.42 Å. A

water molecule bridges two crown‚Na⁺units to form a head-

to-head dimer (not shown).

Solution NMR Studies: NOESY Data for 7. 2D NOESY

experiments were conducted on 7 in CD₃COCD₃in the absence

and presence of NaI. The solution data suggest a structure in

accord with that observed for the solid state. Figure 8 shows

the solid-state structure along with superimposed arrows that

indicate NOE cross-peaks. In the absence of salt (broken

arrows), strong NOE cross-peaks are observed between the

resonance due to unresolved spacer arm hydrogens and indole’s

H2 and H4 hydrogens (see Figure 16 for numbering). Other

NOEs are also observed in accord with the proposed structure,

but the cross-peaks of the indole H2 and H4 atoms with the

crown hydrogens are the most significant peaks involving

crown-sidearm interactions.

Structure of 10‚NaI. The structures of 10‚NaI and 11‚NaI

are generally similar. In both cases, the crystallographic asym-

metric unit contains two unique but similar molecules. Because

the two complexes are quite similar, only one of the 10‚NaI

complexes is shown in Figure 6. The macrocycle in 10‚NaI

adopts an overall saddle-like shape, arranging the π-donors

three-dimensionally rather than in a plane. The six macroring

heteroatoms (Z) are organized around bound Na⁺in a distorted

octahedral geometry. The Z-Na-Z angles (70.1 ( 4.4°) are

intermediate between 60°, expected for a planar structure, and

90° (octahedral). Five of the O-C-Z angles are ∼69° but the

sixth is 74.6°, causing the deviation. The N-Na-N angle is

3100 J. Am. Chem. Soc., Vol. 123, No. 13, 2001

Meadows et al.

Figure 5. Structures of 7‚H₂(PF₆)₂, 7‚NaPF₆, and 9‚NaI shown in the tube metaphor. Dotted lines indicate H-bond contact.

Figure 7. Structures of 12‚KI and 13‚KI shown in an ORTEP plot.

Figure 6. Structures of 10‚NaI and 11‚NaI (H₂O)₂shown in the tube

portion of the indole (H4-H7) exhibited modest shifts ()0.10

ppm). The latter chemical shifts are expected to be relatively

unaffected by complexation because the solid-state structures

show that they are most remote from the cation.

metaphor. Dotted lines indicate H-bond contacts.

When NaI is added, the cross-peak between the indole H4

and the crown hydrogens disappears. Cross-peaks were observed

between H2 and both Hc and the side chain methylenes in the

complex. New cross-peaks between H7 and both Hb and Hc

were observed in the complex (solid arrows in Figure 8). Thus

the structure inferred by the 2D NOESY data comports with

the solid-state result.

Titration of 7 and 8 with NaI. The side chains of 7 are

terminated by indoles attached at the 3-position. Compound 8

is identical except that the indole nitrogen atom (1-position) is

methylated. In separate experiments, each compound (5 mM

in CD₃COCD₃) was titrated with NaI (0.5 M in CD₃COCD₃)

at room temperature. Chemical shift changes were assessed by

¹H NMR and the results are graphed in Figure 9. For 7, the

largest observed chemical shift increment was for H2 (2): ∆δ_∞

) -0.30 ppm.This was expected because H2 was closer to the

bound cation than was any other hydrogen atom in every solid-

state complex of 7 obtained thus far (7‚NaI, 7‚KI, 7‚KSCN,

and 7‚KPF₆). The macroring hydrogens of 7 [H_a(∆), H_b(∇ )]

showed typical downfield shifts upon complexation. The

hydrogen on the indole nitrogen [H1 (9)] was also shifted

downfield to a comparable degree. Hydrogens on the benzo

1

The H NMR chemical shift changes observed when 8 was

titrated as above with NaI corresponded to those observed for

7 but were generally more modest. The largest chemical shift

increments were (0.2 ppm (H_a, H2) rather than nearly -0.3.

The protons most affected in 8 were the same as those in 7 and

the magnitudes of the chemical shift changes are more similar

than different for the two systems. We therefore conclude that

a similar conformation is present in both cases but this is verified

for the solid state only in the case of 7.

Discussion

Indole as a π-Donor. Tryptophan has a 2-(3-indolylethyl)

side chain and can be categorized as either hydrophobic or

hydrophilic depending on the solvent system employed in the

study,⁴¹but in either case, it possesses a π-system. The positively

(38) Meadows, E. S.; De Wall, S. L.; Barbour, L. J.; Gokel, G. W. Chem.

Commun. 1999, 16, 1555-1556.

(39) De Wall, S. L.; Meadows, E. S.; Barbour, L. J.; Gokel, G. W. Proc.

Natl. Acad. Sci. U.S.A. 2000, 97, 6271-6276.

(40) Live, D.; Chan, S. I. J. Am. Chem. Soc. 1976, 98, 3769-3778.

Alkali Metal Cation-π Interactions

J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3101

Figure 9. Graph showing ¹H NMR shifts resulting from titration of 7

(top) or 8 with NaI in CD₃COCD₃.

that either the benzo or the pyrrolo subunit can coordinate

cationic species. Aoki and co-workers observed both inter- and

intramolecular quaternary ammonium ion-π interactions in-

volving the choline ester of 3-indolylacetic acid.⁴⁶The closest

(3.36 Å) contact was between Me₃N⁺and the pyrrolo ring of

an adjacent indole. A further example, found in a search of the

Protein Data Bank (PDB),⁴⁷illustrates a bifurcated cation-π

interaction in the active site of acetylcholine esterase. One of

the carbons from decamethonium (a quaternary amine inhibitor)

is within 4.0 Å of the benzo subunit and another carbon is 4.0

Å from the pyrrolo centroid. A similar interaction was observed

in the phosphocholine complex of the antibody fragment

McPC603.⁴⁸We also note that a search of the Cambridge

Structural Database (CSD)⁴⁹for small molecule crystal structures

and a previous study by Verdonk et al.⁵⁰provide evidence of

ammonium-benzene interactions. A limited amount of solid-

state structural evidence for cation-π interactions has been

available for a number of years although many examples have

not been well recognized. An important example is the report

by Atwood and co-workers of a dibenzo-18-crown-6‚K⁺

Figure 8. The solid-state structure of unbound and sodium-complexed

7 shown with the chemical structures of 7 and 8. Dashed and solid

arrows indicate H NOE’s observed in Me₂CO solution.

1

polarized hydrogens of water may interact with the π-surface

of indole or benzene. Indole has recently been shown by gas-

phase calculations to bind water to its π-surface by formation

of “π-type hydrogen bonds” almost as well as by traditional

N_indole-H‚‚‚O_waterhydrogen bonds.⁴²A report by Atwood and

co-workers demonstrated experimentally the existence of an

O-H_waterπ hydrogen bond.⁴³Recent gas-phase measurements

have confirmed the surprisingly strong attraction between water

and benzene.⁴⁴In these cases, the water hydrogens were directed

toward the benzene centroid as expected for a cation-π

interaction.

Although calculations showed that the benzo ring is the

preferred π-donor in indole, Whipple and co-workers reported

in 1962 that indole is protonated preferentially at either C2 or

C3 (pyrrolo subunit) depending upon indole’s substitution

pattern.⁴⁵Experimental examples in the recent literature suggest

(45) Hinman, R. L.; Whipple, E. B. J. Am. Chem. Soc. 1962, 84, 2534-

2539.

(46) Aoki, K.; Maruyama, K.; Nishiyama, H. J. C. S. Chem. Commun.

1995, 2221-2222.

(47) (a) Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T.

N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E. Nucleic Acid. Res. 2000,

28, 235-242. (b) Bernstein, F. C.; Koetzle, T. F.; Williams, G. J. B.; Meyer,

E. F., Jr.; Brice, M. D.; Rodgers, J. R. K.; Olga; Shimanouchi, T.; Tasumi,

M. J. Mol. Biol. 1977, 112, 535-42.

(48) Satow, Y.; Cohen, G. H.; Padlan, E. A.; Davies, D. R. J. Mol. Biol.

1986, 190, 593-604.

(49) (a) Kennard, O. Supramol. Chem. 1993, 1, 277-95. (b) Allen, F.

H.; Kennard, O.; Taylor, R. Acc. Chem. Res. 1983, 16, 146-53.

(50) Verdonk, M. L.; Boks, G. J.; Kooijman, H.; Kanters, J. A.; Kroon,

J. J. Comput.-Aid. Mol. Des. 1993, 7, 173-82.

(41) (a) Radzicka, A.; Wolfenden, R. Biochemistry 1988, 27, 1664-70.

(b)Wolfenden, R.; Radzicka, A. Trends Biochem. Sci. 1986, 11, 69-70.

(42) (a) Mons, M.; Dimicoli, I.; Tardivel, B.; Piuzzi, F.; Brenner, V.;

Millie, P. J. Phys. Chem. A 1999, 103, 9958-9965. (b) Carney, J. R.; Zwier,

T. S. J. Phys. Chem. A 1999, 103, 9943-9957.

(43) Atwood, J. L.; Hamada, F.; Robinson, K. D.; Orr, G. W.; Vincent,

R. L. Nature 1991, 349, 683-4.

(44) (a) Rodham, D. A.; Suzuki, S.; Suenram, R. D.; Lovas, F. J.;

Dasgupta, S.; Goddard, W. A., III; Blake, G. A. Nature 1993, 362, 735-7.

(b) Suzuki, S.; Green, P. G.; Bumgarner, R. E.; Dasgupta, S.; Goddard, W.

A., III; Blake, G. A. Science 1992, 257, 942-5. (c) Zwier, T. S. Annu.

ReV. Phys. Chem. 1996, 47, 205-241.

3102 J. Am. Chem. Soc., Vol. 123, No. 13, 2001

Meadows et al.

complex in which benzene solvent occupied an apical donor

site.^7bIn the present work, we set as a goal the systematic

assessment of alkali metal cation-π interactions with benzene,

phenol, and indole, the side chains of the aromatic amino acids

Phe, Tyr, and Trp.

These interactions are expected for a conformation of 7 in which

the sidearms are extended from the macrocycle. When NaI was

added, the cross-peaks involving C4 hydrogens were no longer

observed. Instead, new peaks suggested that C2 was close to

one side of the macroring and C7 was close to the opposite

side. These interactions suggest a solution conformation con-

sistent with that observed in the solid state (Figure 3).

The effect of NaI complexation on the ¹H NMR spectra of 7

and 8 was also assessed. Addition of NaI (0 to 1 equiv) caused

a continuous alteration in the chemical shifts. It seemed possible

that the solution conformation was being controlled by H-bond

formation between the anion or solvent and the indole nitrogen.

Replacement of the 1-hydrogen with a methyl group afforded

8. The observed chemical shift differences were smaller for 8

than for 7 but the trends were similar. Moreover, C2 remained

the most shifted aromatic resonance in 8, as was the case for 7.

A Non-π-complexed Conformation for 7‚NaPF₆. The K⁺

complexes of 7 that exhibit cation-π interactions are generally

similar. They are also similar to 7‚NaI, in which cation-π

interactions are also observed. In contrast, the solid-state

structure of 7‚NaPF₆is very different from that observed for

7‚NaPF₆and 7‚KPF₆(see Figure 5) particularly with respect to

the orientation of indole-terminated sidearms. For the cation-π

complexes, the sidearm ethylene adopts a gauche arrangement

such that the indole rings of 7 form a tight cation-π sandwich

complex with K⁺. In contrast, the sidearms in 7‚NaPF₆adopt

the antiperiplanar arrangement.

Artificial Receptors as Biological Model Systems. The

receptor molecules used in the present studies are neither

peptides nor proteins but possess advantages over both. Recep-

tors 1-13 and their complexes all have molecular weights of

e850 Da. For molecules of this size, typical crystals diffract

well enough to afford resolutions better than 0.8 Å. Thus, the

atomic positions and parameters (bond lengths, bond angles,

etc.) are extremely well determined. This is a significant

advantage compared to protein structures, for which the resolu-

tion is often only 2-3 Å. In addition, the receptor systems were

designed to be closely related to each other. The arene in each

case is covalently attached to the macrocycle in an orientation

and at a distance appropriate to permit a cation-π interaction,

eliminating the need for providential contacts. As only the

receptor side chains and salts are varied in the receptors used

in this study, we can directly compare the resulting complex

structures.

A peptide containing 30 amino acids will have a molecular

weight⁵¹of 3-4000 Da. A protein of only 200 amino acids

will have a molecular weight of more than 20 kDa. Like the

crystal structure determinations, solution and computational

studies are easier using these lower molecular weight model

systems than for peptides or proteins.

The non-π-conformation adopted by 7‚NaPF₆represents an

energy minimum but its relative stability in solution compared

to other conformations is unknown. One interesting observation

is that the Na⁺-O distances in 7‚NaI and 7‚NaPF₆are nearly

the same but the Na⁺-N distances in the latter are significantly

shorter. This suggests a greater cation coordination role for the

nitrogen donors and thus the macroring. Presumably, this

increased π-donation compensates for the absence of π-interac-

tions. The more contracted ring in the non-π complex may also

force the sidearms into the “tent” arrangement apparent in Figure

5. The difference in conformation between 7‚NaI and 7‚NaPF₆

cannot be attributed solely to the difference in counteranion since

Sodium and Potassium Complexes of 7. The Na⁺and K⁺

complexes of 7 comprise the first experimental documentation

of Na⁺and K⁺cation-π sandwich interactions of indole. It is

clear from CPK molecular models that π-coordination of a ring-

bound cation in 7 could involve either the benzo or the pyrrolo

subcyclic units. Several theoretical calculations have been

reported for alkali metal cation-π complexation^4,5,13and, in

all cases, the benzo residue is predicted to coordinate the cation.

To our knowledge, no previous small molecule X-ray structure

of a complex involving a direct interaction between indole and

either Na⁺or K⁺has been reported. Wouters has recently

reported a Na⁺-tryptophan contact in the structure of lysozyme

but the distance reported for the arene-cation contact is >4

Å.⁵²An intriguing example of a Cs⁺-Trp interaction was

reported by Hol and co-workers in the crystal structure of

rhodanese.⁵³In this case, the cesium cation is clearly coordinated

by the benzo ring of the indole side chain.

-

both I^-and PF₆are excluded from the solvation sphere.

Moreover, 7‚KPF₆forms a cation-π complex.

A Dicationic Structure Having a π-Complexed Conforma-

tion. One of the most intriguing complexes studied in the effort

reported here was (7‚2H)²⁺(PF₆^-)₂. It was thought that we might

+

obtain an ammonium ion-π complex by treating 7 with NH₄

.

In contrast to theoretical predictions, it is the pyrrolo, rather

than the benzo, subcyclic unit that coordinates with the alkali

metal in 7‚NaI and 7‚KI. The preference of pyrrolo as a donor

for either Na⁺or K⁺in complexes of 7 is independent of the

counterion. Indeed, the solid-state structures of 7‚KI, 7‚KSCN,

and 7‚KPF₆are essentially superimposable, exclusive of the

anion. In all cases studied (7‚NaI, 7‚KI, 7‚KSCN, and 7‚KPF₆)

in which cation-π complexation is observed, the anion is

excluded from the cation’s coordination sphere and is at a

distance from the cation of at least 5.4 Å.

Solution Studies of 7 and 8. The interaction of sodium cation

with receptors 7 and 8 was assessed in two different ways. First,

the NOESY spectrum for 7 was obtained in CD₃COCD₃

solution. The cross-peaks observed suggested proximity between

C2 and C4 on the sidearm indoles and the adjacent macrocycle.

Evidently, proton transfer from NH₄⁺to the macroring nitrogens

of 7 occurred, affording diprotonated 7 (7‚2H)²⁺. A combination

of low-temperature X-ray analysis and high-quality crystals

allowed for the two protons to be located on the crown nitrogen

atoms. The proton positions were assigned based on the

geometry of the macrocycle and the observation of electron

density in the obvious protonation sites.

Although no metal ion is present in the macroring, the

structure parallels those found for 7 M⁺complexes in which

cation-π interactions are observed. The sidearms in (7‚2H)²⁺

-

(PF₆^-)₂are organized above and below the macroring plane as

if solvating the void in the center of the crown. The overall

compactness of this structure may be important in determining

the sidearm positions. Ammonium ion π-complexation of the

type that may occur in this system has been the subject of

theoretical studies⁵⁴and is known in protein structures.⁵⁵

(51) Calculation based on a weighted average molecular weight for an

amino acid of 119.

(52) Wouters, J. Protein Sci. 1998, 7, 2472-2477.

(53) Kooystra, P. J. U.; Kalk, K. H.; Hol, W. G. J. Eur. J. Biochem.

1988, 177, 345-349.

Receptors 8 and 9: Variations in the Indole Side Chains

of 7. The N-methylindole receptor, 8, was prepared to probe

the effect of preventing the indole group from H-bonding.

Alkali Metal Cation-π Interactions

J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3103

Unfortunately, no solid-state structure was obtained for 8‚M⁺.

The solution data described above, however, suggest that similar

interactions occur when either 7 or 8 is titrated with NaI. Thus,

the potential H-bonding role of the indolyl NH may be

unimportant in determining the conformations of the receptor

complexes.

is more charge dense, should be more strongly complexed by a

π-donor than K⁺. This has been confirmed experimentally for

benzene with Na⁺and K⁺(gas phase, difference in binding

energy is 8.8 kcal/mol),^5,35and phenol is anticipated to behave

similarly. Although essentially featureless spheres, Na⁺and K⁺

differ in size (2.36 Å vs 3.02 Å for the 8-coordinate cations).³⁰

Since no cation-π interaction is observed between Na⁺and

10 or 11, we surmise that the difference in steric interactions,

in particular the thickness of the macrocycle relative to the sizes

of Na⁺and K⁺, prevents the observation of a Na⁺-arene

interaction in these two model systems.

Molecular models suggest that the aromatic rings of the

phenol (10) and phenyl (11) receptors cannot penetrate deeply

enough into the macroring’s cavity to coordinate Na⁺. The larger

size of K⁺diminishes the role of steric interactions in this

complex. The K⁺cation (radius ) 1.51 Å) is a better fit for the

18-membered macrocycle (internal radius ) ∼1.5 Å). The size

of the K⁺cation also enables it to overcome the thickness of

the macrocycle, which is estimated to be 3.2 Å based on the

crystal structures of the K⁺complexes of receptors 10 and 11.

The Issue of Electrostatics: (a) Pentafluorophenyl-Side-

armed Complexes, 12 M⁺. Compounds 11 and 12 are identical

except that the aromatic rings of the latter are perfluorinated.

Although the van der Waals radii of aryl H (1.05 Å) and F

(1.35 Å,⁵⁶1.47 Å⁵⁷) differ slightly, 11 and 12 are essentially

isosteres. Their π-donicities differ dramatically, however (see

calculated electrostatic potential surfaces, Figure 2, panel e).

For simplicity, the surfaces were calculated for toluene and for

pentafluorotoluene rather than the entire side chain. The red

color represents negative, cation-attracting density and is

calculated for toluene to be near the arene’s center. In contrast,

pentafluorotoluene is substantially positive (blue ) positive) at

the center of the aromatic ring.

(b) Naphthyl-Sidearmed Receptor Complexes, 13‚M⁺.

There are two, nearly equivalent aromatic subunits in each

sidearm naphthyl group of 13. Either may serve as a binding

site for alkali metal cations in each side chain. Computational

studies of the Na⁺‚naphthalene complex have been reported.^58,59

The calculated gas-phase binding energy for Na⁺‚naphthalene

was 28.7 kcal/mol compared to a value for Na⁺‚benzene of 27.1

kcal/mol. In these calculations, the centroid of either of the six-

membered rings, rather than the overall molecular center, was

found to be the preferred binding site for Na⁺. Thus, Na⁺could

be aligned with either benzo ring in a π-complex of 13. In fact,

π-complexation is not observed in the K⁺‚13 complex; instead,

the sidearms are extended in a fashion similar to that of 12‚K⁺

(Figure 7). The cation in both 12‚K⁺and 13‚K⁺is complexed

by the anion rather than a π-donor. The observation of a non-π

complex (K⁺‚13) in this case may be due to steric interactions

because naphthalene is significantly larger than benzene.

Steric Effects. High-level quantum mechanical (gas phase)

calculations performed on the indole-π‚Na⁺complex predict

that the metal cation will be aligned with the benzo, rather than

the pyrrolo, centroid. In fact, the opposite is observed in the

Na⁺and K⁺complexes of 7 (Figure 2d). What is the explanation

for the discrepancy between ab initio calculations and our

experimental results? One possibility is that the numerous

theoretical calculations are in error because they ignore forces

such as polarization effects⁶⁰that are present in condensed

phases.

The binding energies of Na⁺‚5-methylindole (33.4 kcal/mol)

and Na⁺‚5-aminoindole (36.4 kcal/mol) have been calculated

to be greater than that for Na⁺indole (32.6 kcal/mol). Despite

the presence of the electron-donating methoxy group, the

electrostatic potential surfaces of 5-methoxy-3-methylindole and

3-methylindole are qualitatively similar (Figure 2, panel e). A

quantum mechanical calculation (see the Experimental Section)

of the Na⁺‚5-methoxyindole complex gave a binding energy

of 34.1 kcal/mol, about 5% greater than that of indole. The

similarity in calculated binding energies suggests that it is not

differences in π-donicity that prevent complex formation, but

the data are otherwise unrevealing. Based on π-donicity alone,

9 should form an indole sandwich complex with NaI, as was

the case for 7. Clearly, steric factors may play a role in this

case, but our inability to obtain crystals of a complex of 9 +

NaI does not constitute evidence.

Phenol- and Benzene-Sidearmed Complexes, 10‚MI and

11‚MI. The solid-state structures of 10‚KI and 11‚KI are similar

(see Results). In both cases, the macroring adopts a distorted

D_3dconformation, the apical coordination sites are filled by the

arenes, and the iodide ion is excluded from the cation’s

coordination sphere. In 10‚K⁺, the phenolic hydroxyl group is

H-bonded to iodide, but in 11‚K⁺, no hydrogen bond donor is

accessible to the iodide counterion. In both cases, the center of

the arene is aligned with the cation in accord with theoretical

predictions. The position of the cations relative to the arenes in

10‚KI and 11‚KI agrees with the electrostatic potential surface

calculated for 4-methylphenol and 4-methylbenzene shown in

panel e of Figure 2.

Similar conformations are observed for 10‚NaI and 11‚NaI,

but in neither case do the sidearms coordinate the cation. The

sidearms of both complexes are extended upward from the

approximate plane of the macroring. The phenolic hydroxyls

of 10 H-bond to iodide in the NaI complex. This is not possible

for 11‚NaI but the sidearms exhibit a similar arrangement. Iodide

anion is H-bonded in 11‚NaI but to a water molecule that is

bound to Na⁺within the macrocycle.

The difference in conformation observed between Na⁺and

K⁺complexes for receptors 10 (ethylphenol sidearm) and 11

(ethylbenzene sidearm) is noteworthy. In principle, Na⁺, which

(54) (a) Deakyne, C. A.; Meot-Ner, M. J. Am. Chem. Soc. 1985, 107,

474-479. (b) Mavri, J.; Koller, J.; Hadzˇi, D. J. Molecular Struct.

(THEOCHEM) 1993, 283, 305-312. (c) Kim, K. S.; Lee, J. Y.; Lee, S. J.;

Ha, T.-K.; Kim, D. H. J. Am. Chem. Soc. 1994, 116, 7399-7400. (d)

Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc. 1995, 117, 4177-4178.

(e) Basch, H.; Stevens, W. J. J. Molecular Struct. (THEOCHEM) 1995,

338, 303-315. (f) Lee, J. Y.; Lee, S. J.; Choi, H. S.; Cho, S. J.; Kim, K.

S.; Ha, T.-K. Chem. Phys. Lett. 1995, 232, 67-71. (g) Chipot, C.; Maigret,

B.; Pearlman, D. A.; Kollman, P. A. J. Am. Chem. Soc. 1996, 118, 2998-

3005. (h) Pullman, A.; Berthier, G.; Savinelli, R. J. Comput. Chem. 1997,

18, 2012-2022. (i) Roelens, S.; Torriti, R. J. Am. Chem. Soc. 1998, 120,

12443-12452. (j) De Sant’Anna, C. M. R.; Bicca de Alencastro, R.;

Barreiro, E. J. THEOCHEM 1998, 429, 217-227. (k) De Sant’Anna, C.

M. R.; De Alencastro, R. B.; Barreiro, E. J. THEOCHEM 1999, 167-180.

(l) Tan, X. J.; Jiang, H. L.; Zhu, W. L.; Chen, K. X.; Ji, R. Y. J. Chem.

Soc., Perkin Trans. 2 1999, 107-112. (m) Zhu, W.-L.; Jiang, H.-L.; Puah,

C. M.; Tan, X.-J.; Chen, K.-X.; Cao, Y.; Ji, R.-Y. J. Chem. Soc., Perkin

Trans. 2 1999, 2615-2622. (n) Minoux, H.; Chipot, C. J. Am. Chem. Soc.

1999, 121, 10366-10372. (o) Eriksson, M. A. L.; Morgantini, P.-Y.;

Kollman, P. A. J. Phys. Chem. B 1999, 103, 4474-4480.

(56) Emsley, J. The Elements, 2nd ed.; Clarendon Press: Oxford, 1991.

(57) Bondi, A. J. Phys. Chem. 1964, 68, 441-451.

(58) Mecozzi, S.; West, A. P., Jr.; Dougherty, D. A. J. Am. Chem. Soc.

1996, 118, 2307-2308.

(55) (a) Burley, S. K.; Petsko, G. A. FEBS 1986, 203, 139-143. (b)

Singh, J.; Thornton, J. M. J. Mol. Biol. 1990, 211, 595-615. (c) Scrutton,

N. S.; Raine, A. R. C. Biochem. J. 1996, 319, 1-8.

(59) Dunbar, R. C. J. Phys. Chem. A 1998, 102, 8946-8952.

3104 J. Am. Chem. Soc., Vol. 123, No. 13, 2001

Meadows et al.

One may view the theoretical calculations as being limited

by both the gas phase and the simplicity of the models.

Alternately, one may consider the natural systems complicated

by such interactions as H-bond formation and crystal packing

forces. Indeed, issues such as solvation energies are not directly

addressed by either gas-phase calculations or X-ray crystal

structure determination. The orientation of Na⁺or K⁺proximate

to the pyrrolo centroid is clear from the solid-state structures

of 7‚M⁺complexes. We note that in all cases, the anion forms

a hydrogen bond with the indole nitrogen, which is, of course,

located on the pyrrolo subunit. Thus the cation-anion separation

is minimized when the pyrrolo subunit serves as the π-donor.

We estimate from CPK molecular models and the results of

the ab initio calculations that the cation-anion separation would

be increased ∼0.3 Å if the benzo centroid, rather than the

pyrrolo subunit, coordinated Na⁺or K⁺.

Presumably, the arene geometry in 7‚M⁺complexes is

optimized not just for the cation-π interaction but also forces

such as dipole-dipole, van der Waals, and steric interactions.

Even in such simple receptor systems as 7, 10, and 11, the

overall conformation is controlled by countless intra- and

intermolecular contacts. Among these, steric factors may be

playing a critical role in determining the experimentally observed

arene positions. The indole rings pack tightly against the crown

ether (Figure 2, panels a and b) and are tilted, in the most

extreme case (7‚NaI) by over 26°. The observed indole‚M⁺

geometry in our lariat ether crystal structures could be deter-

mined by the limited steric accessibility of the ring bound cation.

Cation-π interactions are largely electrostatic effects. The

binding energy between a cation and an arene is therefore

dependent on the distance between the aromatic ring and the

alkali metal cation. In complexes of 7, the smaller pyrrolo unit

can more closely approach the bound alkali metal when the

cation is partially buried. Thus, the pyrrolo subunit of indole

may be more accessible to a cation when it is bound within the

crown ether.

the cation-π interaction is not only powerful but also versatile

and may be observed even when cation-π geometry is not

optimal. Such constraints may occur frequently in peptides and

proteins containing tryptophan.

Experimental Section

¹H NMR spectra were recorded at either 300 or 500 MHz in CDCl₃

unless otherwise specified. Chemical shifts are reported in ppm (δ)

downfield from internal TMS. NMR data are recorded as follows:

chemical shift, peak multiplicity (b ) broad; s ) singlet; d ) doublet;

t ) triplet; m ) multiplet, bs ) broad singlet, etc.), integration, and

assignment. Infrared spectra were recorded in KBr unless otherwise

noted and were calibrated against the 1601 cm^-1band of polystyrene.

Melting points were determined on a Thomas-Hoover apparatus in open

capillaries and are uncorrected. Thin layer chromatographic (TLC)

analyses were performed on silica gel HLO F-254 (0.25 mm thickness),

Scientific Adsorbents, Inc. Preparative chromatography columns were

packed with silica gel from Merck (230-400 mesh, 60C). Chromatotron

chromatography was performed on a Harrison Research Model 7924

Chromatotron with 1, 2, or 4 mm thick circular plates (Alltech).

All reactions were conducted under dry N₂unless otherwise noted.

All reagents were the best grade commercially available and were

distilled, recrystallized, or used without further purification, as ap-

propriate. Crystals suitable for X-ray analysis were prepared from

commercially available solvents and salts were used without further

purification. Combustion analyses were performed by Atlantic Microlab,

Inc., Atlanta, GA, and are reported as percents.

Diaza-18-crown-6, 1, was prepared according to a previously

published procedure.²⁷

N,N′-Di-n-propyl-4,13-diaza-18-crown-6, 2, was prepared as previ-

ously described.⁸

N,N′-Diallyl-4,13-diaza-18-crown-6, 3, was prepared as previously

described.⁸

N,N′-Dipropargyl-4,13-diaza-18-crown-6, 4, was prepared as pre-

viously described.⁸

N,N′-Dibenzyl-4,13-diaza-18-crown-6, 5, was prepared as previ-

ously described.²⁸

N,N′-Bis-(1-naphthylmethyl)-4,13-diaza-18-crown-6, 6, was pre-

pared as previously described.⁶¹

Steric hindrance may also explain the antiperiplanar sidearm

conformations (extended away from the macroring) that are

observed for complexes of 7‚NaPF₆, 9‚NaI, 10‚NaI, 11‚NaI, and

13‚KI. When Na⁺is bound by the receptors, the crown more

fully envelops the cation and hinders access of the arene. When

K⁺is bound, the macroring enlarges (relative to bound Na⁺)

and only very large naphthalene fails to coordinate the cation.

Thus, a combination of cavity size, cation size, macroring

thickness, and arene size all play a role in dictating the observed

complex conformations.

N,N′-Bis(2-(3-indolyl)ethyl)-4,13-diaza-18-crown-6, 7. Diaza-

18-crown-6 (0.187 g, 0.712 mmol) and 3-(2-bromoethyl)indole

(0.319 g, 1.42 mmol) were heated with Na₂CO₃(0.377 g, 3.56 mmol)

and NaI (0.011 g, 0.071 mmol) at reflux in CH₃CN (15 mL) for 24 h.

The reaction mixture was filtered and concentrated in Vacuo. The

resulting residue was dissolved in CH₂Cl₂and washed with H₂O

(3 × 15 mL). The organic phase was dried with MgSO4 (anhydrous),

filtered, and concentrated in vacuo to give a yellow solid. Compound

7 was isolated as a light yellow solid after recrystallization from ethanol

1

(0.206 g, 53% yield), mp 129-131 °C. H NMR (acetone-d₆, 0.03%

TMS): 2.83-2.89 (m, -CH₂CH₂- and -N(CH₂)₂-, 16H), 3.56 (s,

-OCH₂CH₂O-, 8H), 3.61 (t, -OCH₂CH₂N-, 8H), 6.99 (t, indole-

H5, 2H), 7.07 (t, indole-H6, 2H), 7.24 (s, indole-H2, 2H), 7.36 (d,

indole-H7, 2H), 7.58 (d, indole-H4, 2H), 9.98 (s, indole-H1, 2H). Anal.

Calcd for C₃₂H₄₄N₄O₄: C, 70.04; H, 8.08; N, 10.21. Found: C, 69.75;

H, 8.05; N, 10.05.

Conclusion

We present here the first solid-state structural evidence that

demonstrates alkali metal cation-π interactions for a family

of systematically varied, biologically relevant arenes. Cation-π

sandwich coordination with K⁺is demonstrated for benzene,

phenol, and indole and sodium-π complexation is documented

for indole. In the latter case, solution NMR studies support the

inferences drawn from the solid-state structures. In general, the

observed cation-π interactions correlate well with theoretical

predictions. The solid-state structures for indole-side-chained

7‚M⁺complexes contrast with the calculations, but the binding

preference for the pyrrolo, over the benzo, subunit of indole

may be understood to result from steric interactions. The latter

is a very important observation because it demonstrates that

Crystallization of 7 and 7‚KI. Receptor 7 was dissolved in acetone

at ambient temperature. Slow evaporation of the solvent over several

weeks gave light yellow crystals of 7. The complex was obtained by

dissolving 7 and an equivalent amount of KI in Me₂CO. Yellow crystals

of the complex formed within 24 h.

1-Methyl-3-(2-bromoethyl)indole was prepared as previously re-

ported.⁶²

N,N′-Bis(2-(3-N-methylindolyl)ethyl)-4,13-diaza-18-crown-6, 8.

Diaza-18-crown-6 (0.55 g, 2.1 mmol), 1-methyl-3-(bromoethyl)indole

(1.0 g, 4.2 mmol), Na₂CO₃(5.0 g, 45 mmol), and 10 mg of KI were

heated at reflux in CH₃CN (50 mL) for 36 h. The mixture was cooled,

(61) Meadows, E. S.; De Wall, S. L.; Barbour, L. J.; Fronczek, F. R.;

Kim, M.-S.; Gokel, G. W. J. Am. Chem. Soc. 2000, 122, 3325-3335.

(62) Murillo, O.; Abel, E.; Maguire, G. E. M.; Gokel, G. W. J. Chem.

Soc., Chem. Commun. 1996, 2147-2148.

(60) (a) Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc. 1995, 117,

4177-4178. (b) Cubero, E.; Luque, F. J.; Orozco, M. Proc. Natl. Acad.

Sci. U.S.A. 1998, 95, 5976-5980.

Alkali Metal Cation-π Interactions

J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3105

filtered, and concentrated under reduced pressure. The residue was taken

up in CH₂Cl₂(100 mL), washed with H₂O (3 × 100 mL), dried

(Na₂SO₄), and reduced in vacuo to a light yellow oil. Column

chromatography (0-5% Et₃N, Me₂CO on silica gel) afforded 8 (0.56

g, 46%) as a yellow oil. ¹H NMR (acetone-d₆): 2.85 (m, 16H,

NCH₂CH₂Ind, NCH₂CH₂Ind, OCH₂CH₂N), 3.59 (m, 16H, OCH₂CH₂N,

OCH₂CH₂O), 3.72 (s, 6H, CH₃-N), 7.03 (s, 2H, indole H2), 7.04 (t, J

) 7.8, 2H, indole H5), 7.15 (t, J ) 6.9, 2H, indole H6), 7.31 (d, J )

8.4, 2H, indole H4), 7.58 (d, J ) 8.1, 2H, indole H7). ¹³C NMR

(CDCl₃): 136.3, 127.9, 126.4, 121.4, 118.8, 118.6, 112.8, 109.1, 70.6,

69.8, 56.6, 57.5, 53.9, 32.4, 22.9. HRMS (FAB) calcd for C₃₄H₄₉N₄O₄

[M + H]⁺577.38, found 577.3775.

N,N′-Bis(2-(3-(5-methoxy)indolyl)ethyl)-4,13-diaza-18-crown-6, 9.

5-Methoxytryptamine (0.20 g, 1.05 mmol), 1,2-bis(2-iodoethoxy)-

ethane⁶³(0.39 g, 1.05 mmol), Na₂CO₃(1.0 g, 9.5 mmol), and 1 mg of

KI were heated at reflux in CH₃CN (10 mL) for 40 h. The reaction

was cooled and filtered, and the resulting solid was heated at reflux in

EtOAc (50 mL) for 20 min. The mixture was filtered, and slow cooling

(2 wk) of the filtrate afforded 9‚NaI as light yellow crystals (25 mg,

6%, mp 135-140 °C). ¹H NMR: 2.89 (m, 16H, NCH₂CH₂Ind, NCH₂-

CH₂Ind, OCH₂CH₂N), 3.59 (m, 16H, OCH₂CH₂N, OCH₂CH₂O), 3.84

(s, 6H, CH₃-O-Ind), 6.81 (d, J ) 9.0, 2H, indole H6), 7.02 (s, 2H,

indole H4), 7.09 (s, 2H, indole H2), 7.20 (d, J ) 9.0, 2H, indole H7),

8.34 (bs, 2H, indole NH). ¹³C NMR (CDCl₃): 153.8, 131.4, 127.9,

123.2, 113.6, 111.9, 111.7, 100.5, 70.6, 70.4, 69.6, 56.2, 55.9, 55.5,

54.1, 22.7. HRMS (FAB) calcd for C₃₄H₄₉N₄O₄[M + H]⁺609.37, found

609.3676.

N,N′-Bis(2-(4-Hydroxyphenyl)ethyl)-4,13-diaza-18-crown-6, 10.

A mixture of 1,2-bis(2-iodoethoxy)ethane (6.74 g, 18.2 mmol), tyramine

(2.5 g, 18.2 mmol), and Na₂CO₃(105.99 g, 91.1 mmol) was heated at

reflux in CH₃CN (50 mL) for 24 h. The mixture was cooled, filtered,

and concentrated in Vacuo to give an orange oil. The crude product

was crystallized from acetone to give 10‚NaI as colorless crystals (0.390

g, 14%, mp 213-214 °C). ¹H NMR (CD₃OD, referenced to 3.31

ppm): 2.59-2.65 (m, phenol-CH₂-, 4H), 2.74-2.80 (m, -CH₂N-

(CH₂)₂-, 12H), 3.63 (t, -NCH₂CH₂O-, 8H), 3.67 (s, -OCH₂CH₂O-,

8H), 6.69 (d, phenol, 4H), 6.96 (d, phenol, 4H). Anal. Calcd for

C₂₈H₄₂O₆N₂NaI: C, 51.54; H, 6.49; N, 4.29. Found: C, 51.55; H, 6.51;

N, 4.34. Free 10 was isolated by dissolving 10‚NaI (0.475 g) in

CH₃CN (100 mL), washing with H₂O (3 × 25 mL), and then adding

CHCl₃(20 mL). The organic phase was dried with MgSO₄and

concentrated in vacuo. Recrystallization from absolute EtOH gave 2

as a light yellow solid (0.252 g, mp 176-178 °C). X-ray quality crystals

of 10 were obtained by slow evaporation at room temperature of a

saturated solution prepared in hot absolute EtOH solution.

temperature. Slow evaporation of the solvent during several weeks

afforded the complex as colorless crystals.

N,N′-Bis[2-(2,3,4,5,6-pentafluorophenyl)ethyl]-4,13-diaza-18-crown-

6, 12. (a) 2,3,4,5,6-Pentafluorophenylacetyl Chloride. To an ice-cold

solution of 2,3,4,5,6-pentafluorophenylacetic acid (1.0 g, 4.42 mmol)

in CH₂Cl₂(50 mL) was added (dropwise) oxalyl chloride (2.0 M in

CH₂Cl₂, 2.21 mL, 4.42 mmol) and anhydrous DMF (catalytic amount).

The mixture was allowed to warm to ambient temperature during 1 h,

and the CH₂Cl₂was removed in Vacuo. The resulting solid acid chloride

was stored under N₂until further use.

(b) N,N′-Bis[2-(2,3,4,5,6-pentafluorophenyl)ethylamide]-4,13-

diaza-18-crown-6. To an ice-cold solution of 4,13-diaza-18-crown-6

(0.580 g, 2.21 mmol) and Et₃N (0.62 mL, 4.42 mmol) in CH₂Cl₂(20

mL) was added (dropwise) an ice-cold solution of 2,3,4,5,6-pentafluo-

rophenylacetyl chloride (1.08 g, 4.42 mmol) in CH₂Cl₂(20 mL). After

the addition, the mixture was stirred at room temperature (24 h), filtered,

and concentrated in Vacuo to an orange oil. Crystallization from EtOAc

gave the desired amide (1.04 g, 70%) as a white solid, mp 132 °C.

N,N′-Bis[2-(2,3,4,5,6-pentafluorophenyl)ethyl]-4,13-diaza-18-crown-

6, 12. To a 0 °C THF solution (10 mL) of the bis(amide) obtained

above (0.150 g, 0.221 mmol) was added BH₃‚THF (1M, 10 mL), and

stirring was continued for 24 h at ambient temperature. The mixture

was concentrated in vacuo, HCl (6 M, 10 mL) was added, and the

solution was heated at reflux for 30 min, cooled gradually to 0 °C, and

made basic by addition of NaOH pellets. The basic solution was

extracted with EtOAc, dried over anhydrous MgSO4, filtered, and

concentrated in vacuo. Compound 12 was isolated (0.124 g, 86%) as

a white solid, mp 61-62 °C. ¹H NMR: 2.73-2.86 (m, -CH₂-

NCH₂CH₂-perfluorophenyl, 16H), 3.56 (t, -NCH₂CH₂O-, 8H), 3.58

(s, -OCH₂CH₂O-, 8H). Anal. Calcd for C₂₈H₃₂F₁₀N₂O₄: C, 51.69;

H, 4.96; N, 4.31. Found: C, 51.50; H, 5.04; N, 4.19.

A nearly saturated ethyl acetate solution was prepared of 12. Slow

evaporation during several weeks gave crystals of 12 suitable for X-ray

analysis. To obtain the complex,1 equiv of 12 (40 mg) was combined

with KI (11 mg, 1 equiv) in EtOAc. Colorless crystals were obtained

by slow evaporation over 7 days. Anal. Calcd for C₃₆H₄₆N₂O₄KI: C,

58.69; H, 6.29; N, 3.80. Found: C, 58.95; H, 6.34; N, 3.81.

2-(1-Naphthyl)ethyl p-toluenesulfonate was prepared from 2-(1-

naphthyl)ethanol by a published procedure;⁶⁴its physical properties were

identical to those previously reported.⁶⁵

N,N′-Bis-(2-(1-naphthyl)ethyl)-4,13-diaza-18-crown-3, 13. Diaza-

18-crown-6 (0.21 g, 0.8 mmol), 1-(2-p-tosylethyl)naphthalene (0.52 g,

1.6 mmol), and Na₂CO₃(0.4 g, 6.4 mmol) were heated at reflux in

MeCN (20 mL) for 72 h. The mixture was cooled, filtered, and

concentrated in Vacuo, and the residue was taken up in CH₂Cl₂(100

mL), washed (3 × 100 mL H₂O), dried (Na₂SO₄), and concentrated in

Vacuo to afford a clear oil. Column chromatography (0-3% Et₃N in

Crystallization of 10‚KI. Equivalent amounts of 10 and KI were

dissolved in hot Me₂CO. Slow evaporation of the solvent over several

days at room temperature yielded light yellow crystals of the complex.

1

Me₂CO on silica gel) yielded 13 (0.28 g, 61%) as a colorless oil. H

NMR (acetone-d₆): 2.85 (m, 12H, OCH₂CH₂N, NCH₂CH₂naph), 3.23

(t, J ) 6.9, 4H, naphCH₂CH₂N), 3.56 (m, 16H, OCH₂CH₂O,

OCH₂CH₂O), 3.68 (m, 20H, OCH₂CH₂N, OCH₂CH₂O), 7.38 (m, 8H,

naphth), 7.73 (t, 2H, J ) 3.6, naphth), 7.87 (d, 2H, J ) 7.8, naphth),

8.12 (d, J ) 8.1, 2H, naphth). ¹³C NMR (acetone-d₆) 138.3, 135.3,

133.3, 129.9, 127.9, 127.8, 127.0, 126.8, 126.6, 125.1, 71.7, 71.3, 58.0,

55.3, 32.2. 13 (40 mg) was combined with 11 mg (1 equiv) of KI and

dissolved in EtOAc. Colorless crystals (mp 171-173 °C) were obtained

by slow evaporation over 7 days. Anal. Calcd for C₃₆H₄₆N₂O₄KI: C,

58.69; H, 6.29; N, 3.80. Found: C, 58.95; H, 6.34; N, 3.81.

X-ray Crystallography. Data were collected at 173(1) K on a Bruker

SMART CCD diffractometer (ω scan mode Mo K_Iradiation, λ )

0.7107 Å). Data were corrected for absorption using the program

SADABS.⁶⁶Structure solution and refinement proceeded similarly for

all structures (SHELX-97 software⁶⁷using the X-Seed⁶⁸interface).

Direct methods yielded all non-hydrogen atoms of the asymmetric unit.

These atoms were refined anisotropically (full-matrix least-squares

N,N′-Bis(2-phenylethyl)-4,13-diaza-18-crown-6, 11. A solution of

(2-bromoethyl)benzene (0.727 g, 3.93 mmol), 4,13-diaza-18-crown-6

(0.515 g, 1.96 mmol), and Na₂CO₃(0.519 g, 4.9 mmol) was stirred in

refluxing CH₃CN (20 mL) for 24 h. The crude product was cooled to

room temperature, filtered, concentrated in vacuo, redissolved in

CH₂Cl₂(100 mL), and washed with H₂O (3 × 15 mL); the organic

phase was dried over MgSO₄, filtered, and concentrated again in vacuo

to afford a yellow oil. The oil was purified by flash column

chromatography (silica, 5% Et₃N in acetone (v/v)) to give the desired

product as a colorless oil. After 24 h under high vacuum the oil

solidified (0.554 g, 60% yield, mp 48-50 °C). ¹H NMR: 2.77 (s,

crown-CH₂CH₂-phenyl, 8H), 2.86 (t, -NCH₂CH₂O-, 8H), 3.62 (m,

-CH₂OCH₂-, 16H), 7.15-7.30 (m, phenyl, 10H). Anal. Calcd for

C₂₈H₄₂N₂O₄: C, 71.46; H, 8.99; N, 5.95. Found: C, 71.20; H, 9.00; N,

5.90.

The complexes were obtained as follows. Equivalent amounts of 11

and NaI were mixed in EtOAc at ambient temperature. The solution

was placed in a freezer at -20 °C. Slow evaporation of the solvent

during several weeks afforded colorless crystals of 11‚NaI. Likewise,

equivalent amounts of 11 and KI were mixed in EtOAc at ambient

(64) Saari, W. S.; Schwering, J. E.; Lyle, P. A.; Smith, S. J.; Engelhardt,

E. L. J. Med. Chem. 1990, 33, 2590-5.

(65) (a) Bentley, M. D.; Dewar, M. J. S. J. Am. Chem. Soc. 1970, 92,

3996-4002. (b) Cram, D. J.; Dalton, C. K. J. Am. Chem. Soc. 1963, 85,

1268-1273.

(63) Kulstad, S.; Malmsten, L. A. Acta Chem. Scand. 1979, B33, 469-

74.

(66) Blessing, R. H. Acta Crystallogr. 1995, A51, 33-38.

3106 J. Am. Chem. Soc., Vol. 123, No. 13, 2001

Meadows et al.

method on F²). Hydrogen atoms were placed in calculated positions

with their isotropic thermal parameters riding on those of their parent

atoms. When structures required modeling of disorder, the disordered

hydrogen atoms were not placed. Mean planes of the macrocycles were

calculated with PARST.⁶⁹All X-ray structure figures were prepared

with X-Seed and POV-Ray.⁷⁰

Quantum Mechanical Calculations. The electrostatic potential

surfaces were calculated by performing a full ab initio geometry

optimization of the relevant aromatic molecule. The 6-31G** basis set

was used as implemented by SPARTAN 5.0 or PC Spartan Pro 1.0.3

(Wavefunction, Inc.). An electron density surface was calculated for

0.002 electron/Å and the electrostatic potential was mapped onto the

surface. All potential energy mapping was completed in the range of

+25.0 kcal/mol to -25.0 kcal/mol. The blue color signifies a positive

potential energy equal to or greater than +25.0 kcal/mol. A red color

denotes an equivalent negative (-25 kcal/mol) potential energy. The

binding energy of Na⁺complexes of the aromatic molecules was

performed at the 6-31G** level of theory for both Na⁺and the aromatic

group.

Titration Studies. A 0.5 M solution of NaI in acetone-d₆(100%)

was prepared in a 1 mL volumetric flask. The NaI was dried for 24 h

in a vacuum oven at 100 °C. In a small, predried glass vial the diindole

receptor, 7 (5 mg), was dissolved in 1 mL of acetone-d₆(100%). The

solution was transferred to a 5 mm NMR tube and then diluted to 2

mL with acetone-d₆. The final concentration of the solution of diindole

receptor, 7, was 5 mM.

Data were collected using Gemini 2300 (Varian, Inc.) at a frequency

of 300 MHz. The NMR probe temperature was set to 25 °C. The sample

tube was inserted into the spectrometer and allowed to equilibrate with

the probe temperature. Each spectrum was obtained by collecting 16

transients. The signal from residual acetone in the solvent was used as

the reference (2.05 ppm). Each spectrum was collected after adding a

1 µL aliquot of the 0.5 M NaI solution. Mixing of the titrant in the

sample tube was accomplished by inverting the tube several times. A

total of 25 data points were collected by this method. Identical results

were obtained for the titration of a 10 mM sample of 7 with 1.0 M NaI

solution in acetone-d₆. Assignment of the resonances in the spectra

was possible for all protons. However, the resonances from the protons

in the ethylene sidearm and those next to the nitrogens in the crown

ether overlapped over the course of the titration making unequivocal

assignment of these protons impossible.

2D-NMR Experiments. Experiments were done using either a

Varian or Bruker 500 MHz NMR spectrometer. Both COSY and

NOESY experiments were performed on a 10 mM solution of the indole

receptor, 7, in acetone-d₆. The total solution volume was 750 µL. The

experiments were also carried out using the same concentration of 7

([7] ) 10 mM) and 1 equiv of NaI ([NaI] ) 10 mM). Prior to doing

the NOESY experiments the sample solutions were degassed by

sonication. Experiments were carried out at 25 °C.

Crystal data for 7 (first crystal form): M ) 548.71; light yellow

rhombohedroid, 0.50 × 0. 50 × 0.45 mm³; monoclinic, P21/c; a )

18.0798(9) Å, b ) 12.1004(6) Å, c ) 14.5384(7) Å, â ) 110.2130(10)°;

Z ) 4; V ) 2984.7(3) Å³; D_c) 1.221 Mg/m³; 2θ_max) 54.32°, 17035

reflections collected, 6508 unique [R(int) ) 0.0246]; final GoF ) 1.033,

R1 ) 0.0458, wR2 ) 0.1676, R indices based on 5039 reflections with

I > 2σ(I), µ ) 0.081 mm^-1, minimum transmission factor ) 0.763.

Polymorph of 7: M ) 548.17; light brown parallelepiped, 0.30 ×

0.25 × 0.20 mm³; monoclinic, P21/n; a ) 11.2857(8) Å, b ) 31.408(2)

Å, c ) 12.7040(8) Å, â ) 102.3590(10)°; Z ) 6; V ) 4398.7(5) Å³;

D_c) 1.243 Mg/m³; 2θ_max) 54.26°, 27176 reflections collected, 9655

unique [R(int) ) 0.0554]; final GoF ) 0.932, R1 ) 0.0736, wR2 )

0.0625, R indices based on 4725 reflections with I > 2σ(I), µ ) 0.082

mm^-1, minimum transmission factor ) 0.9757.

12.1395(6) Å, c ) 15.4777(8) Å; â ) 91.8540(10)°; Z ) 4; V )

3264.6(3) Å³; D_c) 1.421 Mg/m³; 2θ_max) 54.32°, 9866 reflections

collected, 3609 unique [R(int) ) 0.0176]; final GoF ) 1.041, R1 )

0.0277, wR2 ) 0.0649, R indices based on 3172 reflections with I >

2σ(I), µ ) 1.035 mm^-1, minimum transmission factor ) 0.759.

Crystal data for 7‚KI: M ) 714.71; light yellow rhombohedroid,

0.40 × 0.30 × 0.20 mm³; monoclinic, C2/c; a ) 17.2370(10) Å, b )

12.0426(7) Å, c ) 15.9510(9) Å, â ) 90.2610(10)°; Z ) 4; V )

3311.0(3) Å³; D_c) 1.434 g/cm³; 2θ_max) 54.28°, 9687 reflections

collected, 3620 unique [R(int) ) 0.0323]; final GoF ) 1.034, R1 )

0.0291, wR2 ) 0.0682, R indices based on 3060 reflections with I >

2σ(I), µ ) 1.134 mm^-1, minimum transmission factor ) 0.747.

Crystal data for 7‚KPF6: M ) 732.78; pale pink rhombohedroids,

0.40 × 0.35 × 0.30 mm³; monoclinic, P21/n; a ) 10.9698(8) Å, b )

13.0812(9) Å, c ) 11.9828(9) Å; â ) 95.0620(10)°; Z ) 2; V )

1712.8(2) Å³; D_c) 1.421 Mg/m³; 2θ_max) 54.18°, 10365 reflections

collected, 3754 unique [R(int) ) 0.0210]; final GoF ) 1.049, R1 )

0.0325, wR2 ) 0.0842, R indices based on 1698 reflections with I >

2σ(I), µ ) 0.277 mm^-1, minimum transmission factor ) 0.8972.

Crystal data for 7‚(HPF₆)₂: M ) 838.65; pale orange rhombohe-

droids, 0.40 × 0.30 × 0.25 mm³; monoclinic, P21/c; a ) 10.0769(16)

Å, b ) 17.045(3) Å, c ) 11.2044(18) Å; â ) 105.547(3)° ; Z ) 2; V

) 1854.0(5) Å³; D_c) 1.502 Mg/m³; 2θ_max) 52.84°, 15372 reflections

collected, 3787 unique [R(int) ) 0.0433]; final GoF ) 1.055, R1 )

0.0652, wR2 ) 0.1844, R indices based on 1698 reflections with I >

2σ(I), µ ) 0.220 mm^-1, minimum transmission factor ) 0.9171.

Crystal data for 10: M ) 502.64; colorless rhombohedroid, 0.35

× 0.20 × 0.10 mm³; monoclinic, P21/c; a ) 8.1816(10) Å, b )

17.039(2) Å, c ) 9.7101(12) Å, â ) 101.142(2)°; Z ) 2; V ) 1328.1(3)

Å³; D_c)1.257 g/cm³; 2θ_max) 54.24°, 8248 reflections collected, 2926

unique [R(int) ) 0.0679]; final GoF ) 1.000, R1 ) 0.0561, wR2 )

0.1062, R indices based on 1438 reflections with I > 2σ(I), µ ) 0.088

mm^-1, minimum transmission factor ) 0.9700.

Crystal data for 10‚NaI: M ) 652.53; colorless rhombohedroid,

0.35 × 0.35 × 0.25 mm³; tetragonal, P43; a ) 10.0920(4) Å, c )

60.278(3) Å; Z ) 8; V ) 6139.2(5) Å³; D_c) 1.412 g/cm³; 2θ_max

)

54.38°, 37186 reflections collected, 13363 unique [R(int) ) 0.0509];

final GoF ) 1.161, R1 ) 0.0499, wR2 ) 0.0954, R indices based on

11815 reflections with I > 2σ(I), µ ) 1.098 mm^-1, minimum

transmission factor ) 0.6998.

Crystal data for 10‚KI: M ) 668.64; colorless rhombohedroid,

0.35 × 0.30 × 0.20 mm³; monoclinic, C2/c; a ) 19.8231(15) Å, b )

9.5298(7) Å, c ) 17.4222(13) Å, â ) 113.1490(10)°; Z ) 4; V )

3026.2(4) Å³; D_c) 1.468 g/cm³; 2θ_max) 54.38°, 9095 reflections

collected, 3342 unique [R(int) ) 0.0313]; final GoF ) 0.904, R1 )

0.0335, wR2 ) 0.0755, R indices based on 2468 reflections with I >

2σ(I), µ ) 1.237 mm^-1, minimum transmission factor ) 0.6713.

Crystal data for 11‚NaI: M ) 668.53; colorless rhombohedroid,

0.30 × 0.30 × 0.25 mm³; triclinic, P1h; a ) 10.0425(19) Å, b )

10.0425(19) Å, c ) 31.775(6) Å, R ) 90.191(3)°, â ) 98.094(4)°, γ

) 90.485(3)°; Z ) 4; V ) 3236.7(10) Å³; D_c) 1.372 g/cm³; 2θ_max

)

54.54°, 20010 reflections collected, 13857 unique [R(int) ) 0.0581];

final GoF ) 1.144, R1 ) 0.1055, wR2 ) 0.2470, R indices based on

11404 reflections with I > 2σ(I), µ ) 1.046 mm^-1, minimum

transmission factor ) 0.7444.

Crystal data for 11‚KI: M ) 636.64; colorless rhombohedroids,

0.40 × 0.30 × 0.15 mm³; monoclinic, C2/c; a ) 19.9098(9) Å, b )

9.6181(4) Å, c ) 17.2441(8) Å, â ) 116.4390(10)°; Z ) 4; V )

2956.8(2) Å³; D_c) 1.430 g/cm³; 2θ_max) 54.24°, 8740 reflections

collected, 3248 unique [R(int) ) 0.0185]; final GoF ) 1.046, R1 )

0.0231, wR2 ) 0.0594, R indices based on 2982 reflections with I >

2σ(I), µ ) 1.258 mm^-1, minimum transmission factor ) 0.6331.

Crystal data for 12: M ) 650.56; colorless rhombohedroids, 0.25

× 0.25 × 0.10 mm³; monoclinic, P21/c; a ) 15.0714(17) Å, b )

9.5904(11) Å, c ) 9.9806(11) Å, â ) 95.185(2)°; Z ) 2; V ) 1436.7(3)

Å³; D_c) 1.504 g/cm³; 2 θ_max) 54.22°, 8742 reflections collected,

3162 unique [R(int) ) 0.0392]; final GoF ) 0.887, R1 ) 0.0406, wR2

) 0.0871, R indices based on 1821 reflections with I > 2σ(I), µ )

0.143 mm^-1, minimum transmission factor ) 0.9652.

Crystal data for 7‚NaI: M ) 698.60; colorless rhombohedroids,

0.45 × 0.40 × 0.40 mm³; monoclinic, P21/n; a ) 17.3841(9) Å, b )

(67) Sheldrick, G. M., 1997, University of Go¨ttingen.

(68) Barbour, L. J., 1999, University of Missouri-Columbia. (http://

www.lbarbour.com/xseed/).

(69) (a) Nardelli, M. J. Appl. Crystallogr. 1995, 28, 659. (b) Nardelli,

M. Comput. Chem. 1983, 7, 95-8.

(70) http://www.povray.org.

Crystal data for 12‚KI: M ) 816.56; colorless rhombohedroids,

0.25 × 0.25 × 0.20 mm³; triclinic, P1; a ) 8.0839(6) Å, b ) 8.2103(6)

Alkali Metal Cation-π Interactions

J. Am. Chem. Soc., Vol. 123, No. 13, 2001 3107

Å, c ) 13.9553(11) Å, R ) 97.9020(10)°, â ) 115.0700(10)°; Z ) 1;

V ) 813.38(11) Å³; D_c) 1.667 g/cm³; 2θ_max) 54.22°, 5052 reflections

collected, 3482 unique [R(int) ) 0.0207]; final GoF ) 1.041, R1 )

0.0526, wR2 ) 0.1311, R indices based on 3042 reflections with I >

2σ(I), µ ) 1.206 mm^-1, minimum transmission factor ) 0.7526.

based on 1698 reflections with I > 2σ(I), µ ) 1.612 mm^-1, minimum

transmission factor ) 0.6433.

Acknowledgment. We thank the NIH (GM-36262) and the

NSF (CHE-9805840) for grants that supported this work. We

are also grateful for an ACS Division of Organic Chemistry

fellowship, funded by Procter and Gamble, to E.S.M. The

2-dimensional NMR experiments were accomplished with the

assistance of Drs. D. Andre´ d’Avignon and Janet Braddock-

Wilking whom we thank.

Crystal data for 13 KI: M ) 736.75; colorless rhombohedroids,

0.30 × 0.20 × 0.15 mm³; orthorhombic, Cmca; a ) 14.852(2) Å, b )

7.6873(11) Å, c ) 30.609(4) Å; Z ) 6; V ) 3494.7(9) Å³; D_c) 2.100

Mg/m³; 2θ_max) 54.22°, 10428 reflections collected, 2003 unique [R(int)

) 0.0316]; final GoF ) 1.176, R1 ) 0.0762, wR2 ) 0.1866, R indices

JA003059E

Article Doi

Alkali metal cation - π Interactions observed by using a lariat ether model system

DOI: 10.1021/ja003059e

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja003059e

Full text of DOI:10.1021/ja003059e

Products guided by the article

R&D Labs maybe for 289477-27-6

Relevant to this article

Hot Product