Extended structures built on a triphenoxymethane platform - C3- symmetric, conformational mimics of calix[n]arenes

Article abstract of DOI:10.1002/1099-0690(200007)2000:13<2467::AID-EJOC2467>3.0.CO;2-N

A series of C₃-symmetric tris(3,5-dialkyl-2-hydroxyphenyl)methanes (alkyl = tert-butyl, methyl, tert-pentyl) have been synthesized in high yield from their respective phenols and fully characterized, including single crystal X-ray structures for two examples. The di-tert-butyl-substituted compound, 1a, has been derivatized to a tris-acid chloride, 4, which was treated with a variety of amines (dimethylamine, benzylamine, glycine, and alanine) from which the corresponding trisamides 5-8 were formed. The absolute geometry and conformation of the dimethylamine, glycine, and alanine derived systems were determined by X-ray analyses, and in all cases, the three phenolate arms point up with respect to the central methine. Alkali metal binding studies (Li, Na, K, Rb, and Cs) were carried out for the dimethylamine, benzylamine, and glycine compounds 5-7, and these compounds were found to have some selectivity for potassium cations. NMR studies demonstrate that C₃ symmetry is retained in all the compounds and the stoichiometry for lithium ion complexation is 1:1, whereas two ligands are needed to complex sodium cations. Crystal structures of the dimethylamine derivative with lithium picrate and the benzylamine and glycine derivatives with sodium tetraphenytborate were also determined.

Full text of DOI:10.1002/1099-0690(200007)2000:13<2467::AID-EJOC2467>3.0.CO;2-N

FULL PAPER
Extended Structures Built on a Triphenoxymethane Platform ؊ C3-Symmetric,
Conformational Mimics of Calix[n]arenes
Maarten B. Dinger[a] and Michael J. Scott*[a]
Keywords: Phenoxides / Calixarenes / Molecular recognition / Extraction
A series of C3-symmetric tris(3,5-dialkyl-2-hydroxyphenyl)- the three phenolate arms point up with respect to the central
methanes (alkyl = tert-butyl, methyl, tert-pentyl) have been
synthesized in high yield from their respective phenols and
fully characterized, including single crystal X-ray structures
for two examples. The di-tert-butyl-substituted compound,
1a, has been derivatized to a tris-acid chloride, 4, which was
methine. Alkali metal binding studies (Li, Na, K, Rb, and Cs)
were carried out for the dimethylamine, benzylamine, and
glycine compounds 5−7, and these compounds were found
to have some selectivity for potassium cations. NMR studies
demonstrate that C3 symmetry is retained in all the com-
treated with a variety of amines (dimethylamine, benzylam- pounds and the stoichiometry for lithium ion complexation is
ine, glycine, and alanine) from which the corresponding tris-
amides 5−8 were formed. The absolute geometry and con-
formation of the dimethylamine, glycine, and alanine derived
systems were determined by X-ray analyses, and in all cases,
1:1, whereas two ligands are needed to complex sodium cat-
ions. Crystal structures of the dimethylamine derivative with
lithium picrate and the benzylamine and glycine derivatives
with sodium tetraphenylborate were also determined.
Introduction
many complex synthetic manipulations.[19] Hence, the devel-
opment of simple, rigid, compact molecules with only three,
oriented oxygen donor arms has been problematic.
By virtue of their predisposition to serve as host molec-
ules, binding various neutral and ionic guest species, much
attention has focused on the development of calix[n]arene-
based macrocyclic ligands.[1Ϫ4] These macrocycles maintain
reactive hydroxyl groups about their rims, and with the ap-
propriate pendant arms, macrocycles with diverse proper-
ties can be produced. Chemical modifications with esters,[5]
glycols,[3] sugars,[6] Schiff bases,[7] and peptides[8] produce
compounds capable of binding a range of substrates from
alkali[9] and lanthanide metals[9] to proteins.[10] Unfortu-
nately, calixarene-like macrocycles with only three oxygen
donors are often difficult to prepare. Compounds such as
hexahomooxacalix[3]arenes and trihydroxy[3.3.3]metacy-
clophanes, require longer linkers between the phenoxide
ligands.[11Ϫ13] With the increased flexibility associated with
the larger cyclic structure, these macrocycles adopt multiple
configurations in solution (Scheme 1), an impediment to
molecular recognition. Moreover, the systematic substitu-
tion of only three of the phenolic oxygens is quite trouble-
some for most of the calix[n]arenes. Trisubstituted macro-
cycles can be isolated through the selective functionali-
zation of calix[6]arene,[14Ϫ17] but, as before, the larger back-
bone of this compound engenders the macrocycle with
immense flexibility,[18] complicating the solution structure
of the larger constructs. Conformationally rigid molecules
with three oxygen donor arms have been developed from
scyllo-inositol and cis,cis-cyclododecane-1,5,9-triol, but
these molecules are quite difficult to prepare, requiring
Scheme 1
Molecules containing three appendages offer many ad-
vantages for selective recognition, especially for metal ions.
Each of the three arms can be designed to provide two
donor groups to a metal center allowing the metal to adopt
either octahedral or trigonal prismatic stereochemistry de-
pending on its preference. Nature has found this to be a
particularly adept binding motif for metal ions. For ex-
ample, the three catecholate arms of the protein Enterobac-
tin are perfectly poised to seize an iron atom.[20]
Intent on the isolation of small, uncomplicated, con-
formational mimics of macrocycles including calix[4]arene,
we examined the literature and came across an interesting
report of the preparation of a series of compounds in which
three phenoxide groups are joined to a common carbon
atom forming tris-phenoxidemethanes.[21] If sufficient rigid-
ity could be built into these unique molecules, we theorized
that they would serve as an ideal platform for the prepara-
tion of extended structures containing only three append-
ages. Herein, we report our investigations into the prepara-
tion of materials utilizing the tris(3,5-dialkyl-2-hydroxy-
phenyl)methane framework and an initial examination into
the alkali metal binding properties of these extended struc-
tures.
[a]
Department of Chemistry, University of Florida
P. O. Box 117200 Gainesville, FL 32611-7200, USA
Fax: (internat.) ϩ1-352/392-3255
Email: mjscott@chem.ufl.edu
Eur. J. Org. Chem. 2000, 2467Ϫ2478
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1434Ϫ193X/00/0713Ϫ2467 $ 17.50ϩ.50/0
2467

M. B. Dinger, M. J. Scott
FULL PAPER
Results and Discussion
action scheme is easier to scale up for the preparation of
large quantities of 1a, 1b, and 1d since it avoids the use of
dry hydrogen chloride, and we routinely isolate the mat-
erials in 25 gram quantities using this procedure. Unfortu-
nately, this synthesis is less general than the phenol/salical-
dehyde condensation, and is unsatisfactory for the synthesis
of 1c, or any tris-phenoxymethane featuring a methyl sub-
stituent in the ortho position.[21]
As outlined in Scheme 2, ligands with divergent steric
properties can be prepared through two different synthetic
routes. Following addition of dry hydrogen chloride to a
solution of a salicylaldehyde and phenol, the tris(3,5-dial-
kyl-2-hydroxyphenyl)methanes 1 are isolated in high yield
directly from the methanol solution. Offering an added
advantage, both three fold symmetric and mirror-symmetric
compounds can be formed using this methodology through
the judicious choice of the two starting materials. Facilitat-
ing isolation and purification of the products, the pre-
cursors are all readily soluble in methanol, yet the tris(3,5-
dialkyl-2-hydroxyphenyl)methanes show only negligible
solubility. Product isolation and purification is quite simple;
the desired products are simply filtered off and any re-
maining starting materials or by-products are removed with
washing of the solid. Typically, condensation reactions of
this type are carried out in acetic acid,[25Ϫ27] but this solvent
is unsatisfactory for the formation of 1aϪ1d, with only very
low yields collected from the acid solution, despite the in-
solubility of the compounds in the solvent. There is also
precedence for the use of concentrated sulfuric acid as the
catalyst; indeed the first triarylmethanes formed by the con-
densation of benzaldehyde and phenols were obtained by
this route.[28] Nonetheless, when concentrated sulfuric acid
was added to a methanolic solution of the aldehyde and
phenol, no color change occurred and tris(3,5-dialkyl-2-hy-
droxyphenyl)methanes could not be detected, with NMR
showing only starting materials.
As opposed to the calix[n]arene macrocycles where intra-
molecular hydrogen-bonding induces a cone-like conforma-
tion,[2] the central carbon atoms in 1a؊d prevent the ligand
arms from aligning properly for this type of interaction.
Instead, intermolecular communication between the rigid,
C3 symmetric ligands is favored. In the solid-state structure
of 1a depicted in Figure 1 (crystallographic data presented
in Table 2), the two rotamers of the ligand form a dimeric
structure producing six equivalent hydrogen bonds. The
shortest oxygenϪoxygen distance between the two conver-
˚
gent ligands is more than ഠ2.95 A, suggesting only a minor
degree of orbital communication through this hydrogen
bond. Decreasing the steric constraints at the 2- and 4-posi-
tions of the phenol engenders the ligand with a higher de-
gree of rotational flexibility. In the solid state structure of
1c (Figure 2, Table 2), the intermolecular oxygenϪoxygen
˚
separation is reduced to ഠ2.73 A, and the ligands adopt a
tetrahedral arrangement with tris-phenoxide ligands resid-
ing on the four corners with each of the three oxygen
donors orientated towards a different corner. Concomitant
Scheme 2
Although in somewhat lower yields, the ligands 1a, 1b,
and 1d can also be prepared directly from the phenolic pre-
cursors following a synthetic scheme originally developed
by Casnati and co-workers.[21] Deprotonation of the phenol
with a Grignard reagent and addition of ethyl orthoformate
affords the tris-phenoxides in only moderate yield, but given
the difficulties often associated with the synthesis of the sal-
icaldehydes from the phenol, including the use of hazardous
tin reagents, this procedure is normally much more conven-
ient for the production of the threefold symmetric phenox-
ides. Moreover, in contrast to the acid condensation, the
metal directed synthesis allows for the preparation of tris(3-
alkyl-2-hydroxyphenyl)methanes lacking alkyl substitution
Figure 1. Diagram of 1a · C6H6 hydrogen-bonded dimer, showing
30% probability ellipsoids for the oxygen atoms, and atom labeling
scheme. Only one molecule of the two unique molecules in the
asymmetric unit is shown. The numbering scheme for the re-
maining molecule is analogous. tert-Butyl and methyl groups and
the benzene solvate are omitted, for clarity. Selected bond lengths
˚
[A] and angles [°]: C(1)ϪC(2), 1.530(3); O(1)···O(2Ј), 3.114(5);
O(2)···O(3Ј), 2.972(5); O(3)···O(1Ј), 3.009(5); O(1)H···O(2Ј), 2.529;
O(2)H···O(3Ј), 2.368; O(3)H···O(1Ј); 2.507; O(1)ϪHϪO(2Ј),
at the para-position of the phenoxide arms. Finally, this re- 132.4(2); O(2)ϪHϪO(3Ј), 127.8(2); O(3)ϪHϪO(1Ј), 121.7(2)
2468
Eur. J. Org. Chem. 2000, 2467Ϫ2478

C3-Symmetric, Conformational Mimics of Calix[n]arenes
FULL PAPER
Figure 2. Crystal structure of 1c showing half of the hydrogen-bonded tetramer, with 30% probability ellipsoids for the oxygen atoms,
and atom-labeling scheme. C(26) resides on a C3 crystallographic axis, and the unlabelled atoms of the molecule are generated by this
˚
symmetry operation. Selected bond lengths [A] and angles [°]: C(1)ϪC(2), 1.518(5); O(1)O(3Ј), 2.732(5); O(2)O(2Ј), 2.746(5); O(3)O(4),
2.816(5); O(4")O(1), 2.730(5); O(1)HO(3Ј), 1.825; O(2)HO(2Ј), 1.872; O(3)HO(4); 1.948; O(4")HO(1), 2.092; O(1)ϪHϪO(3Ј), 151.6(4);
O(2)ϪHϪO(2Ј), 161.3(4); O(3)ϪHϪO(4), 134.2(4); O(4")ϪHϪO(1), 139.0(4)
with the strengthening of the hydrogen-bonding interaction, literature provides an overabundance of methodology for
the frequency of the hydroxide infrared stretch in the solid the derivatization of calix[n]arenes,[1Ϫ3,29,30] and many of
state decreases from 3503 cmϪ1 in 1a to 3460 cmϪ1 in 1c.
the procedures are directly applicable to the tris-phenoxide
As is the case for calix[n]arenes,[1Ϫ3] the rigidity of the system. Following the simple procedures outlined in
tris(3,5-dialkyl-2-hydroxyphenyl)methane platforms, no- Scheme 3, reactive groups can be placed on the periphery
tably 1a, permits the construction of larger ligand systems of the ligand, thus allowing a wide scope of organic frag-
with well-defined structures. A careful examination of the ments to be incorporated onto the platform. In the develop-
Scheme 3
Eur. J. Org. Chem. 2000, 2467Ϫ2478
2469

M. B. Dinger, M. J. Scott
FULL PAPER
ment of these procedures, the reactivity of 1a was found to of the ligand. An O diagram of 7, highlighting several
be quite comparable to calix[4]arene, although depro- of these hydrogen bonds, is contained in Figure 3 and crys-
tonation of all three phenoxide groups requires the use of a tallographic information can be found in Table 2. In the
stronger base, cesium carbonate, instead of the more typical crystal, the molecule is rigorously three fold symmetric with
reagent, potassium carbonate. The yields for these reactions the amide nitrogen interacting with an intra-arm phenolic
˚
are all excellent, and the insolubility of the products in po- group [N···O 2.658(9) A], in addition to an adjacent car-
˚
lar solvents greatly facilitates the purification and eliminates bonyl oxygen atom [N···O 2.905(9) A].
any need for chromatography. The acid chloride 4 can be
isolated in multigram quantities and it is stable indefinitely
in the absence of moisture, facilitating the preparation of
extended structures.
Perhaps one of the most interesting features of the com-
plex structures derived from the calix[n]arene platform is
their ability to selectively bind metal ions, and as depicted
in Scheme 3, addition of six equivalents of an amine to the
phenoxyacetyl chloride group in 4 allows for incorporation
of many different groups, from simple amines to extended
peptides, onto the tris-phenoxide backbone. Three different
extended structures were prepared to test the ability of these
compounds to extract alkali metal ions: tris(3,5-di-tert-bu-
tyl-2-[(dimethylamido)methoxy]phenyl)methane (5) which
contains alkyl amido groups, normally one of the most pro-
ficient alkali metal binders in calix[4]arenes,[31] tris(3,5-di-
tert-butyl-2-[(benzylaminocarbonyl)methoxy]phenyl)-
methane (6), and tris{3,5-di-tert-butyl-2-[N-(methyl-
glycyl)carbonylmethoxy]phenyl}methane (7) with longer
amino-acid arms. The ligand tris{3,5-di-tert-butyl-2-[N-
(methylalanyl)carbonylmethoxy]phenyl]methane (8) was
also prepared to allow for a comparison of the hydrogen
bonding interactions within arms with different amino
acid groups.
Upon irradiation of the hydrogen on the central carbon
Figure 3. Diagram of 7 · 1/2 C5H12, showing 30% probability ellips-
oids and atom-labeling scheme. Primed, double primed, and un-
primed atoms are all related by a 3-fold symmetry operation. Only
one of two unique arms in the asymmetric unit is shown. The re-
maining molecule follows an analogous numbering scheme. The
disordered pentane solvate has been omitted for clarity. Selected
atom in 1a, NOE enhancements are clearly evident in the
1H NMR transitions for one set of aryl rings as well as the
oxo protons, intimating that the aryl rings freely rotate on
the NMR timescale. Incorporation of groups onto the
framework, however, appears to preclude this process as no
NOE enhancements are evident between the ligand arms
and the aryl rings in compounds 2؊8 in solution at room
temperature. Supporting this assignment, the solid-state
structures of compounds 2b, 3, 4, 5, 7, and 8 were all deter-
˚
bond lengths [A] and angles [°]: N(1)ϪO(3Ј), 2.905(9); N(1)ϪO(1),
2.658(9);
N(1)HϪO(3Ј),
2.304;
N(1)HϪO(1),
2.263;
N(1)ϪH(1)ϪO(3Ј), 127.1(8); N(1)ϪH(1)ϪO(1), 108.0(8)
Reasoning that exchange of the glycine arms with other
mined (Tables 2 and 3), and the arms in all of these com- amino acids would disrupt the hydrogen bonding interac-
pounds are oriented in the same direction. From a compar- tions evident in the solid-state structure of 7, compound 4
ison of the crystal structures of 2a[32] and 2b, a few addi- was treated with six equivalents of the methyl ester of
tional conclusions can be drawn. For the preparation of -alanine to afford tris{3,5-di-tert-butyl-2-[N-(methyl-
compounds with extended arms, it is essential to include alanyl)carbonylmethoxy]phenyl}methane (8) in high yield.
sterically encumbering groups at the para-position of the The solid-state structure is depicted in Figure 4 while im-
aryl rings. With small substituents at this location, the aryl portant metric data can be found in Table 3. As expected,
rings can rotate to reduce unfavorable steric interaction and placement of a methyl group on the arm disrupts the hydro-
the arms splay out away from each other. To maintain fa- gen-bonding between adjacent arms, and the amide nitro-
vorable binding properties, the platform must include gen instead interacts with both the ester carbonyl [N···O
˚
groups such as tert-butyl or tert-pentyl at this para-position, 2.680(9) A] and a phenolic oxygen of the same residue
˚
which restrict the rotation of the phenoxide rings relative to [N···O 2.635(9) A]. While both 7 and 8 exhibit extensive
each other.
hydrogen-bonding in the solid lattice, these interactions are
Often, the arms in these extended structures exhibit only weak in solution, highlighted by the temperature de-
extensive intramolecular hydrogen-bonding interactions, pendent shifts ([D6]DMSO) of the amide nitrogens from
particularly between amide hydrogens and the phenolic 293 to 353 K of 5.4·10Ϫ3 ppm KϪ1 and 4.7·10Ϫ3 ppm
oxygens, and these contacts may help to rigidify the arms
K
Ϫ1, respectively.[33,34]
2470
Eur. J. Org. Chem. 2000, 2467Ϫ2478

C3-Symmetric, Conformational Mimics of Calix[n]arenes
FULL PAPER
Extraction of Alkali Metal Ions
Following the methodology of Cram and co-workers,[22]
the association constants (Ka) for the complexation of 5, 6,
and 7 to the alkali metals were determined in H2O/CDCl3,
and the results are presented in Table 1. Although 5 does
Table 1. Comparison of R, Ka, and Ϫ∆G° for tripodal phenoxide
hosts complexing picrate salt guests in CHCl3 at 25°C[22]; figures
within Ϯ10% in multiple experiments; stoichiometry, guest/host,
for extractions was assumed as 1:1; NMR data may suggest a dif-
ferent stoichiometry for the larger cations
Compound Picrate R[a]
Ka ϫ 10Ϫ3 Ϫ∆G°
[Ϫ1 [b]
[kcal/mol] extracted
%
salt
]
5
6
7
Li
0.1988 1210
8.3
8.3
8.5
8.3
8.2
7.4
7.4
7.8
7.6
7.4
7.2
7.1
7.7
7.3
7.2
22
25
36
30
32
9
Na
K
0.2249 1233
0.3231 1815
0.2394 1191
0.2359 977
0.0655 251
0.0852 284
0.1641 490
0.1152 364
0.1018 260
0.0521 191
0.0513 153
0.1538 443
0.0824 234
0.0848 205
Rb
Cs
Li
Na
K
Rb
Cs
Li
Na
K
Rb
Cs
12
17
17
16
6
6
18
13
10
Figure 4. Diagram of 8 · CHCl3, showing 30% probability ellips-
oids for the oxygen and nitrogen atoms, and atom-labeling scheme.
Primed, double primed, and unprimed atoms are all related by a
3-fold symmetry operation. The remaining molecule follows an
analogous numbering scheme. The chloroform solvate has been
[a]
[b]
[Guest]/[host] in CHCl3 layer at equilibrium. Ϫ
Association
˚
omitted for clarity. Selected bond lengths [A] and angles [°]:
constant (see Equation 1).
N(1)ϪO(3), 2.680(9); N(1)ϪO(1), 2.635(9); N(1)HϪO(3), 2.352;
N(1)HϪO(1),
2.248;
N(1)ϪH(1)ϪO(3),
103.1(8);
N(1)ϪH(1)ϪO(1), 107.3(8)
Figure 5. 1H NMR spectra (300 MHz) of a) free 5, b) 5 ϩ excess lithium picrate and c) 5 ϩ excess sodium picrate, in CDCl3 after filtration.
The peaks labeled * are from residual CHCl3
Eur. J. Org. Chem. 2000, 2467Ϫ2478
2471

M. B. Dinger, M. J. Scott
FULL PAPER
not exhibit a particularly high degree of selectivity for the of 5. The bridging molecules are further stabilized through
alkali metal ions, the association values do indicate that the the formation of two hydrogen bonding interactions, one to
molecule is an effective binder, on the order normally asso- the carbonyl group on the remaining dimethylamido arm
ciated with calix[n]arenes.[5,31,35] As depicted in Figure 5, and one to a phenoxide oxygen atom. The asymmetry wit-
the solution structure of the lithium and sodium complexes nessed in these species is inconsistent with the NMR data
of 5 are threefold symmetric on the time scale of the experi- presented in Figure 5, suggesting an alternative, symmetric
ment, with one ligand ligating each lithium ion, while two species exists in solution on the time scale of the NMR ex-
ligands are required to bind to a sodium atom. Removing periment.
one of the alkyl substituents from the amide nitrogen de-
creases the electron density of the carbonyl oxygen, and 6 complex of 5 were unsuccessful, but crystals of a complex
is a much less effective binder for all the alkali metal ions. of 6 and sodium tetraphenylborate were isolated. Although
Repeated attempts to grow single crystals of the sodium
In order to further probe the binding properties of these the tetraphenylborate anion was severely disordered in the
ligands, the solid-state structure of the lithium picrate com- crystal lattice, which prohibits a detailed discussion of the
plex with 5 was determined, and it is depicted in Figure 6. structure, the coordination environment of the metal center
In this complex, two water molecules bridge two metal cen- was unambiguously identified. As portrayed in Figure 7,
ters, forming a dinuclear species. The water bridge is some- the oxygen atoms from two separate threefold symmetric
what asymmetric with LiϪOH2 distances of 2.009(5) and ligands bind to the metal in an octahedral arrangement
˚
˚
2.082(5) A, and the coordination sphere of the metal con- with NaϪO distances ranging from 2.366(4) to 2.375(4) A.
tains three additional oxygen atom ligands, two carbonyl While the local environment about the sodium ion closely
groups and one phenoxide oxygen from two different arms resembles the structure of potassium bound to the carbonyl
groups in the cyclic dodecadepsipeptide valinomycin,[36]
a
Figure 7. Crystal structure of 62NaBPh4 · 4.5 THF, showing 30%
probability ellipsoids for the oxygen, nitrogen and sodium atoms,
and atom labeling scheme. The complex resides on a crystallo-
graphic inversion center and only the symmetry unique atoms are
labeled. The disordered benzyl and tert-butyl groups have been
omitted, together with the BPh4Ϫ, THF solvates, and hydrogen
atoms. Hydrogen bonds are evident between amide nitrogen and
Figure 6. Crystal structure of 52Li2[Li2picrate4] · 2 C6H6 · 2 H2O,
showing 30% probability ellipsoids for the oxygen, nitrogen and
lithium atoms, and atom labeling scheme. The complex is centro-
symmetric and only the unique atoms are labeled. The methyl car-
bons of tert-butyl groups and the [Li2picrate4]2Ϫ counter-ion have
˚
˚
been omitted for clarity. Selected bond lengths [A] and angles [°]:
phenolic oxygen atoms. Selected distances [A] and angles [°]:
LiϪO(3), 2.136(5); LiϪO(5), 2.004(5); LiϪO(6), 1.951(5); LiϪO(7), NaϪO(5), 2.366(4); NaϪO(6), 2.372(4); NaϪO(4), 2.375(4);
2.009(5); LiϪO(7Ј), 2.082(5); O(7)HϪO(2), 2.082; O(7)HϪO(4),
N(1)HϪO(1), 2.290; N(2)HϪO(2), 2.220; N(3)HϪO(3), 2.214;
83.6(2); O(5)ϪNaϪO(4), 86.6(2);
O(6)ϪNaϪO(4), 87.4(1)
1.982; O(3)ϪLiϪO(7), 146.4(2); O(3)ϪLiϪO(7Ј), 120.1(2); O(5)ϪNaϪO(6),
O(5)ϪLiϪO(6), 164.3(3); LiϪO(7)ϪLiЈ, 86.5(2)
2472
Eur. J. Org. Chem. 2000, 2467Ϫ2478

C3-Symmetric, Conformational Mimics of Calix[n]arenes
FULL PAPER
to present the metal ion with two different sets of donor
groups, the three carbonyl oxygens from the amide linkage
and the three ester carbonyl groups. Even though the gly-
cine arms present the metals with an unfavorable seven-
membered chelate ring, an enhanced preference for potas-
sium over all the other alkali metals is evident in Table 1.
1
As was the case for both 5 and 6, the solution H NMR
data exhibits threefold symmetry for the metal complexes,
but in contrast to 5 and 6, integration of the NMR data
indicates only one ligand is required to extract the larger
metal ions. On the time scale of the NMR experiment, each
of the three arms provides, on average, two donor atoms to
these metal centers.
Entirely inconsistent with the C3 symmetry witnessed in
the solution NMR data, the complex of 7 and sodium tetra-
phenylborate adopts an asymmetric orientation in the solid-
state (Figure 8, Table 3), but this structure does confirm the
one to one ligand to metal stoichiometry. The difference
between the solid-state and solution structure may be
brought about by the unfavorable chelate rings presented to
the metal by the oxygen donors of the glycine arms. The
complex crystallizes as a dinuclear species with one ester
carbonyl group bridging to a sodium center of an adjacent
ligand. The three amide carbonyl linkages in 7 complete the
˚
coordination sphere of the metal. At 2.344(2) A, one of the
bonds to an amide carbonyl is longer than the other three
˚
NaϪO distances [2.264(2) to 2.279(3) A] presumably to al-
low for an extremely long interaction of 2.825(2) A with a
phenoxide oxygen atom on the same glycine residue. All
attempts to crystallize 7 with potassium ions were unsuc-
cessful.
˚
Figure 8. Crystal structure of 72Na2(BPh4)2 · 1/8 C6H14 · 2/3 THF,
showing 30% probability ellipsoids for the oxygen, nitrogen and
sodium atoms, and atom labeling scheme. The complex is centro-
symmetric and prime and unprimed atoms are related by symmetry.
Ϫ
The methyl carbons of the tert-butyl groups, BPh4 counter-ion
and solvent molecules have been omitted for clarity. Selected bond
˚
lengths [A] and angles [°]: NaϪO(7), 2.264(2); NaϪO(10), 2.271(3);
Conclusion
NaϪO(8Ј), 2.279(3); N(1)HϪO(1), 2.290; NaϪO(1), 2.825(2);
O(7)ϪNaϪO(10),
115.3(1);
O(7)ϪNaϪO(8Ј),
87.9(1);
O(7)ϪNaϪO(4), 123.98(9); O(4)ϪNaϪO(10), 120.5(1); O(4)
ϪNaϪO(8Ј), 94.96(9)
With only three arms locked in alignment, the ligands
constructed on the framework of 1a are clearly distinct
from the well-studied calix[n]arenes, and yet, this scaffold
retains the simplicity of the synthetic manipulation of the
calix[n]arene system. Having only three appendages, a vari-
ety of extended arms can be secured to the tris-phenoxide
platform and this ligand has the ability to present a metal
with an octahedral array of binding sites. The development
of extended macromolecular complexes with these plat-
forms may afford compounds capable of selectively binding
metal ions, clusters, salts, and organic substrates. Investi-
gations into the binding properties of this unique ligand
class with different appendages including amino acid chains
and carbamoylmethylphosphane oxides (CMPO),[38] de-
signed to recognize actinide ions, are currently underway.
direct comparison to the structure of the sodium complex
of valinomycin[37] is uninformative. In the presence of pic-
rate, the sodium ion will only coordinate to four of the car-
bonyl groups of valinomycin with a water molecule and the
picrate ion filling the remaining coordination sphere of the
metal ion. In metal complexes prepared with both 5 and 6,
there was no direct evidence of any interaction between the
metal ions and the picrate anion.
Given the tendency of 6 to form this octahedral structure
with two different ligands binding to the one sodium ion,
the reaction of 5 and sodium picrate may form an analo-
gous species, and this formulation would be consistent with
1
the H NMR data presented in Figure 5. Interestingly, in
the complex presented in Figure 7, the position of each li-
gand arm and the orientation of the carbonyl group is sta-
bilized for metal binding by a strong intramolecular hydro-
gen bond between the phenoxide oxygen and amide [avg.
N···O 2.625(7) A], but this type of interaction cannot occur
in complexes of 5.
Experimental Section
General Remarks: 1H- and 13C NMR spectra were recorded in
CDCl3 solvent, unless otherwise stated, with a Varian VXR-300
spectrometer at 299.95 and 75.43 MHz for the proton and carbon
channels, respectively. IR spectra were recorded as KBr discs, un-
less otherwise stated, with a Bruker Vector-22 instrument with a 4
˚
In an attempt to improve the binding selectivity of the
extended structures for the alkali metal ions, 7 was designed cmϪ1 resolution.
Eur. J. Org. Chem. 2000, 2467Ϫ2478
2473

M. B. Dinger, M. J. Scott
FULL PAPER
Tris(3,5-dialkyl-2-hydroxyphenyl)methanes (1). ؊ General Proced-
ure: Powdered 2,4-dialkylphenol (2 equiv.) and 3,5-dialkyl-4-
hydroxybenzaldehyde[39] (1 equiv.) were dissolved in methanol. The
resulting solutions were immersed in ice baths and charged with
dry hydrogen chloride. The reaction mixtures rapidly turned red
concomitant with precipitate formation. After two hours the HCl
source was replaced with a drying tube, and the reactions allowed
to stir at room temperature for two days. CAUTION: The flask
should be vented to prevent pressure buildup. The solids were fil-
tered off and thoroughly washed with methanol to give pure 1 as
white powders.
150.6 (CAr). Ϫ C49H76O3 (713.1): calcd. C 82.52, H 10.74; found C
82.42, H 11.34.
Tris(3,5-dialkyl-2-ethoxycarbonylmethoxyphenyl)methanes (2).
؊
General Procedure: To an acetone solution of tris(3,5-dialkyl-2-hy-
droxyphenyl)methane (1 equiv.) was added cesium carbonate (3.5
equiv.) and ethyl bromoacetate (3.5 equiv.). The resulting mixture
was refluxed under an atmosphere of dry nitrogen for about 12 h.
After cooling, the acetone was removed under reduced pressure,
and diethyl ether was added to the residue. The insoluble materials
were filtered off, and the ether was removed under reduced pressure
to leave colorless oils that slowly crystallized. The crude products
2 were essentially pure by NMR, but additional purification was
readily facilitated by washing the solids with methanol.
Alternatively, compounds 1a, 1b, and 1d could be prepared by us-
ing a modified procedure reported by Casnati et al.[21] To a solution
of phenol in anhydrous diethyl ether was added, dropwise, an
equivalent of ethylmagnesium bromide. When the evolution of eth-
ane had ceased, a third of an equivalent of triethyl orthoformate
was added. The resulting solution was heated, allowing the ether
to evaporate, and occasionally a small amount of toluene was ad-
ded to aid mixing. The mixture was heated to 100°C for 12 hours,
during which time the solution became deep blue, then gradually
becoming yellow often with precipitate formation. After allowing
the mixture to cool, hydrochloric acid (2 ) was added, and the
product extracted into diethyl ether. Evaporation of the solvents
left a viscous oil which, upon addition of methanol, precipitated
out a white solid. Filtration and subsequent washing with methanol
gave the corresponding tris(3,5-dialkyl-2-hydroxyphenyl)methanes
as white powders in ca. 65% yield.
2a: Yield 96%. Ϫ M.p. 154Ϫ155°C. Ϫ IR: ν˜ [cmϪ1] 1763, 1738
(CO). Ϫ 1H NMR: δ ϭ 1.21 (s, 27 H, tBu), 1.27 (t, 9 H, J ϭ
7.2 Hz, OCH2CH3), 1.33 (s, 27 H, tBu), 4.21Ϫ4.28 (m, 12 H,CH2),
6.33 (s, 1 H, CH), 7.19 (m, 6 H, Ar-H). Ϫ 13C NMR: δ ϭ 13.9
(OCH2CH3), 30.8, 31.3 [C(CH3)3], 34.4, 35.2 [C(CH3)3], 37.6 (CH),
60.5 (OCH2CH3), 69.4 (ArOCH2), 122.4, 126.5, 136.4, 142.3, 145.1,
152.3 (CAr), 168.8 (CϭO). Ϫ MS (FAB) m/z (%) ϭ 926 [M ϩ K]ϩ
(35), 910 [M ϩ Na]ϩ (100), 887 [M]ϩ (18). Ϫ C55H82O9 (887.2):
calcd. C 74.46, H, 9.32; found C 74.54, H 9.44.
2b: Yield 89%. Ϫ M.p. 145Ϫ146°C. Ϫ IR: ν˜ [cmϪ1] 1766, 1732
(CO). Ϫ 1H NMR: δ ϭ 1.24 (t, J ϭ 7.2 Hz, 9 H, OCH2CH3), 1.30
(s, 27 H, tBu), 2.24 (s, 9 H, ArCH3), 4.06 (s, 6 H, ArOCH2), 4.19
(q, 6 H, J ϭ 7.2 Hz, OCH2CH3), 6.31 (s, 1 H, CH), 6.94 (s, 3 H,
Ar-H), 7.01 (s, 3 H, Ar-H). Ϫ 13C NMR: δ ϭ 14.0 (OCH2CH3),
21.4 (ArCH3), 30.8 [C(CH3)3], 35.0 [C(CH3)3], 37.1 (CH), 60.6
(OCH2CH3), 69.6 (ArOCH2), 126.9, 129.9, 132.5, 137.3, 143.2,
152.6 (CAr), 168.8 (CϭO). Ϫ MS (FAB) m/z (%) ϭ 799 [M ϩ K]ϩ
(100), 783 [M ϩ Na]ϩ (85), 761 [M]ϩ (20). Ϫ C46H64O9 (761.3):
calcd. C 72.59, H 8.48; found C 72.46, H 8.83.
1a: Yield 78%. Ϫ M.p. 260°C (decomposed without melting). Ϫ
IR: ν˜ [cmϪ1] 3503 (OH). Ϫ 1H NMR: δ ϭ 1.14 (s, 27 H, tBu), 1.34
(s, 27 H, tBu), 4.83 (s, 3 H, OH), 5.64 (s, 1 H, CH), 6.64 (d, J ϭ
2.4 Hz, 3 H, Ar-H), 7.28 (d, J ϭ 2.4 Hz, 3 H, Ar-H). Ϫ 13C NMR:
δ ϭ 29.8, 31.4 [C(CH3)3], 34.3, 35.0 [C(CH3)3], 42.5 (CH), 123.5,
125.3, 137.1, 143.2, 150.8 (CAr). Ϫ MS (FAB) m/z (%) ϭ 629 [M]ϩ
(40), 423 [M Ϫ tBu2C6H3OH]ϩ (100), 365 [M Ϫ tBu2C6H3OH Ϫ
tBu]ϩ (76). Ϫ C43H64O3 (628.97): calcd. C 82.11, H 10.26; found
C 82.17, H 10.32.
Tris(3,5-di-tert-butyl-2-carboxymethoxyphenyl)methane (3):
Tris(3,5-di-tert-butyl-2-ethoxycarbonylmethoxyphenyl)methane
(3.50 g, 3.95 mmol) was dissolved in a methanol/water/dichloro-
methane (50:10:4 mL) solvent mixture and potassium carbonate
(5.50 g, 39.8 mmol) added. The mixture was refluxed for 4 hours.
The resulting clear solution was poured into water (400 mL), acidi-
fied to pH 4 with dilute hydrochloric acid, and extracted with di-
ethyl ether. The solvent was removed under reduced pressure and
chloroform was added to the residue. The insoluble material was
filtered and washed with additional chloroform to give 3 as white
powder (2.92 g, 92%). The product can be recrystallized from di-
ethyl ether/hexanes, although the crude workup yields very pure
material, suitable for further reaction. M.p. 254Ϫ256°C. Ϫ IR: ν˜
[cmϪ1] 1740, 1713 (CO). Ϫ 1H NMR (CD3OD): δ ϭ 1.21 (s, 27 H,
tBu), 1.36 (s, 27 H, tBu), 4.04 (s, 6 H, ArOCH2), 6.46 (s, 1 H, CH),
7.11 (d, J ϭ 2.4 Hz, 3 H, Ar-H), 7.32 (d, J ϭ 2.4 Hz, 3 H, Ar-
H). Ϫ 13C NMR (CD3OD): δ ϭ 31.7, 31.9 [C(CH3)3], 35.5, 36.5
[C(CH3)3], 40.0 (CH), 71.2 (ArOCH2), 124.3, 128.0, 138.4, 144.0,
147.1, 154.1 (CAr), 172.3 (CϭO). Ϫ MS (FAB) m/z (%) ϭ 842 [M
ϩ K]ϩ (65), 826 [M ϩ Na]ϩ (100), 803 [M]ϩ (10). Ϫ C49H70O9
(803.8): calcd. C 73.27, H 8.79; found C 73.16, H 9.17%.
1b: Yield 75%. Ϫ M.p. 240°C (decomposed without melting). Ϫ
IR: ν˜ [cmϪ1] 3507 (OH). Ϫ 1H NMR: δ ϭ 1.37 (s, 27 H, tBu), 2.19
(s, 9 H, ArϪCH3), 4.79 (s, 3 H, OH), 5.58 (s, 1 H, CH), 6.55 (s, 3
H, Ar-H), 7.08 (s, 3 H, Ar-H). Ϫ 13C NMR: δ ϭ 21.0 (CH3) 29.7
[C(CH3)3], 34.7 [C(CH3)3], 42.4 (CH), 125.9, 127.2, 127.7, 130.1,
137.9, 151.1 (CAr). Ϫ MS (FAB) m/z (%) ϭ 502 [M]ϩ (42), 339 [M
Ϫ tBu Ϫ Me Ϫ C6H3OH]ϩ (100), 365 [M Ϫ tBu Ϫ Me Ϫ C6H3OH
Ϫ tBu]ϩ (91). Ϫ C34H46O3 (502.7): calcd. C 81.23, H 9.22; found
C 81.15, H 9.34.
1c: Yield 60%. Ϫ M.p. 268Ϫ269°C (melts with decomposition). Ϫ
1
IR: ν˜ [cmϪ1] 3460 (OH). Ϫ H NMR: δ ϭ 2.18 (s, 3 H, CH3), 2.20
(s, 9 H, CH3), 4.75 (s, 3 H, OH), 5.84 (s, 1 H, CH), 6.58 (s, 3 H,
Ar-H), 6.89 (s, 3 H, Ar-H),. Ϫ 13C NMR: δ ϭ 16.0, 20.7 (CH3),
40.1 (CH), 124.6, 126.4, 127.5, 129.8, 130.8, 149.8 (CAr). Ϫ MS
(FAB) m/z (%) ϭ 376 [M]ϩ (100), 255 [M Ϫ Me2C6H3OH]ϩ (97).
Ϫ C25H28O3 (376.5): calcd. C 79.74, H 7.50; found C 79.28, H 7.80.
1d: Yield 65%. Ϫ M.p. 233Ϫ235°C. Ϫ IR: ν˜ [cmϪ1] 3500 (OH). Ϫ Tris(3,5-di-tert-butyl-2-chlorocarbonylmethoxyphenyl)methane (4):
1H NMR: δ ϭ 0.54 (t, J ϭ 7.5 Hz, 6 H, CH2CH3), 0.60 (t, J ϭ
Tris(3,5-di-tert-butyl-2-carboxymethoxyphenyl)methane was dis-
7.5 Hz, 6 H, CH2CH3), 1.09 (s, 21 H, CMe2), 1.32 (s, 21 H, CMe2), solved in dichloromethane and an excess of thionyl chloride added.
1.44 (q, J ϭ 7.5 Hz, 6 H, CH2CH3), 1.77 (q, J ϭ 7.5 Hz, 6 H, The solution was refluxed for five hours under a nitrogen atmo-
CH2CH3), 4.76 (s, 3 H, OH), 5.67 (s, 1 H, CH), 6.57 (d, J ϭ 2.4 Hz, sphere. The solvent and excess thionyl chloride were both removed
3 H, Ar-H), 7.14 (d, J ϭ 2.4 Hz, 3 H, Ar-H). Ϫ 13C NMR: δ ϭ
under vacuum to leave 4 as a white residue. The product was pure
9.0, 9.7 (CH2CH3), 27.9, 28.4 [C(CH3)3], 33.2, 37.0 [C(CH3)3], 37.4, by NMR, and additional purification was not carried out for its
38.6 (CH2CH3), 41.9 (CH), 124.3, 125.5, 123.8, 125.3, 135.3, 141.0, use in subsequent reactions. 1H NMR: δ ϭ 1.15 (s, 27 H, tBu),
2474
Eur. J. Org. Chem. 2000, 2467Ϫ2478

C3-Symmetric, Conformational Mimics of Calix[n]arenes
1.27 (s, 27 H, tBu), 4.38 (s, 6 H, ArOCH2), 6.19 (s, 1 H, CH), 7.13 Alkali Metal Complexes. ؊ General Procedure: To a small quantity
FULL PAPER
(d, J ϭ 2.4 Hz, 3 H, Ar-H), 7.22 (d, J ϭ 2.4 Hz, 3 H, Ar-H). Ϫ
of amide (15 mg) dissolved in CDCl3 (1 mL) was added excess (10
13C NMR: δ ϭ 31.1, 31.2 [C(CH3)3], 34.5, 35.4 [C(CH3)3], 37.2 equiv.) alkali metal salt. The resulting slurries were vigorously
(CH), 76.1 (ArOCH2), 126.4 127.3, 140.9, 142.7, 146.0, 146.6 (CAr), stirred for 24 hours. The solutions were filtered and, for the pic-
169.1 (CϭO).
rates, 1H-, 13C-NMR, and IR spectra were directly acquired. When
sodium tetraphenylborate was used, the products were recrystal-
lized by diffusion of pentane into a THF solution.
Compounds 5؊8. ؊ General Procedure: The crude tris(3,5-di-tert-
butyl-2-chlorocarbonylmethoxyphenyl)methane prepared above
was redissolved in dichloromethane and the corresponding amine
(6 equiv.) added. Immediate precipitates of the amine hydro-
chlorides formed. The solutions were refluxed under nitrogen for 1
hour to ensure completion. Dilute hydrochloric acid was added and
the organic layers collected and dried with magnesium sulfate. Re-
moval of the solvent gave the corresponding amides 5؊8 as white
solids. Although all the crude products were essentially pure by
NMR, purification by recrystallization from THF/pentane was
readily effected.
1
5 ؉ Lithium Picrate: IR (CDCl3): ν˜ [cmϪ1] 1652, 1619 (CO). Ϫ H
NMR: δ ϭ 1.24 (s, 27 H, tBu), 1.29 (s, 27 H, tBu), 2.59 (s, 9 H,
NMe), 2.97 (s, 9 H, NMe), ca. 4 (s, v br, 6 H, ArOCH2), 6.77 (s, 1
H, CH), 7.28 (d, J ϭ 2.4 Hz, 3 H, Ar-H tripod), 7.35 (d, J ϭ
2.4 Hz, 3 H, Ar-H tripod), 9.02 (s, 2H, Ar-H picrate). Ϫ 13C NMR:
δ ϭ 31.3, 31.8 [C(CH3)3], 34.6 [C(CH3)3], 35.5 (CH), 35.6, 35.7
(NMe2), 36.0 [C(CH3)3], 70.2 (ArOCH2), 124.3, 126.3, 128.0, 128.7,
136.0, 140.7, 142.0, 147.0, 151.3, 163.2 (CAr), 168.8 (CϭO).
5 ؉ Sodium Picrate: IR (CDCl3): ν˜ [cmϪ1] 1664, 1635, 1616 (CO).
Tris(3,5-di-tert-butyl-2-dimethylamidomethoxyphenyl)methane (5):
Yield 89%. Ϫ M.p. 254Ϫ257°C. Ϫ IR: ν˜ [cmϪ1] 1685, 1666, 1653
1
Ϫ H NMR: δ ϭ 1.21 (s, 54 H, tBu), 1.29 (s, 54 H, tBu), 2.52 (s,
18 H, NMe), 2.91 (s, 18 H, NMe), ഠ3.8 (s, v br, 12 H, ArOCH2),
6.60 (s, 2H, CH), 7.17 (d, J ϭ 2.4 Hz, 6 H, Ar-H tripod), 7.27 (d,
J ϭ 2.4 Hz, 6 H, Ar-H tripod), 8.91 (4 H, s, Ar-H picrate). Ϫ 13C
NMR: δ ϭ 31.3, 31.8 [C(CH3)3], 34.5 [C(CH3)3], 35.5, 35.6 (NMe2),
35.8 [C(CH3)3], 37.5 (CH), 70.7 (ArOCH2), 123.9, 126.5, 127.0,
136.6, 141.8, 142.3, 146.5, 152.1 (CAr), 168.2 (CϭO).
1
(CO). Ϫ IR (CHCl3) 1673, 1639 (CO). Ϫ H NMR: δ ϭ 1.09 (s,
27 H, tBu), 1.23 (s, 27 H, tBu), 2.68 (s, 9 H, NMe), 2.81 (s, 9 H,
NMe), 4.02 (s, 6 H, ArOCH2), 6.27 (s, 1 H, CH), 6.94 (d, J ϭ
2.4 Hz, 3 H, Ar-H tripod), 7.15 (d, J ϭ 2.4 Hz, 3 H, Ar-H tripod).
Ϫ
13C NMR: δ ϭ 30.8, 31.1 [C(CH3)3], 34.2 [C(CH3)3], 35.1
(NMe2), 36.2 [C(CH3)3], 39.6 (CH), 72.0 (ArOCH2), 122.5, 126.6,
136.7, 142.2, 145.3, 153.1 (CAr), 167.5 (CϭO). Ϫ C55H85O6N3
(884.3): calcd. C 74.69, H 9.69, N 4.75; found C 74.38, H 9.93,
N 4.66.
6 ؉ NaBPh4: IR: ν˜ [cmϪ1] 3427 br, 3366 br (NH), 1671 (CO). Ϫ
1H NMR: δ ϭ 1.08 (s, 54 H, tBu), 1.35 (s, 54 H, tBu), 4.34 (br, 12
H, CH2), 5.86 (s, 2H, CH tripod), 6.18 (s, 6 H, NH), 6.87 (t, J ϭ
7.1 Hz, 4 H, BPh4), 6.96Ϫ7.01 (m, 14 H, Ar-H), 7.04 (t, J ϭ 7.4 Hz,
12 H, Ar-H), 7.18Ϫ7.22 (m, 18 H, Ar-H), 7.31 (d, J ϭ 2.4 Hz, 6
H, Ar-H tripod), 7.44 (br, 8 H, BPh4). Ϫ 13C NMR: δ ϭ 31.2, 31.5
[C(CH3)3], 34.4 [C(CH3)3], 35.5, 40.0 (CH), 42.3 (NHCH2), 72.6
(ArOCH2), 121.5, 123.5, 125.5, 126.5, 126.6, 127.2, 128.4, 136.1,
Tris(3,5-di-tert-butyl-2-benzylaminocarbonylmethoxyphenyl)-
methane (6):Yield 93%. Ϫ M.p. 247Ϫ248°C. Ϫ IR: ν˜ [cmϪ1] 3426,
1
3294 v br (NH), 1679, 1656 (CO). Ϫ H NMR: δ ϭ 1.19 (s, 27 H,
tBu), 1.28 (s, 27 H, tBu), 4.16 (s, 6 H, ArOCH2), 4.34 (d, J ϭ
5.4 Hz, 6 H, NHCH2), 6.48 (s, 1 H, CH), 7.07 (s, 3 H, Ar-H tripod),
8.35 (m, 21 H, Ar-H and NH). Ϫ 13C NMR: δ ϭ 31.29, 31.31
[C(CH3)3], 34.5, 35.3 [C(CH3)3], 37.3 (CH), 42.6 (NHCH2), 71.8
(ArOCH2), 123.2, 126.5, 127.2, 127.4, 128.5, 136.4, 138.1, 141.8,
146.1, 151.3 (CAr), 168.5 (CϭO). Ϫ C70H91O6N3 (1070.5 ): calcd.
C 78.53, H 8.57, N 3.93; found C 78.24, H 8.81, N 3.90.
136.3, 137.9, 141.7, 147.0, 151.3 (CAr), 168.4 (CϭO).
Ϫ
C140H182O12N6NaB(C6H5)4 · 4.5 THF (2483.2): calcd. C 77.84, H
8.55, N 2.99; found C 78.03, H 8.92, N 2.94.
1
7 ؉ NaBPh4: IR: ν˜ [cmϪ1] 3369 br (NH), 1750, 1681 (CO). Ϫ H
NMR (CD2Cl2): δ ϭ 1.15 (s, 27 H, tBu), 1.21 (s, 27 H, tBu), 3.67
(s, 9 H, OMe), ഠ3.8 (s, v br, 6 H, CH2), 6.24 (s, 1 H, CH tripod),
6.42 (br, 3 H, NH), 6.72 (t, 4 H, J ϭ 7.2 Hz, 4 H, BPh4), 7.08 (t,
8 H, J ϭ 7.5 Hz, 4 H, BPh4), 7.20 (br, 8 H, BPh4), 7.27 (d, J ϭ
2.4 Hz, 3 H, Ar-H tripod). Ϫ C58H85O12N3NaB(C6H5)4 (1358.5):
calcd. C 72.47, H 7.79, N 3.09; found C 71.91, H 7.84, N 3.01.
Tris{3,5-di-tert-butyl-2-[N-(methyl glycyl)carbonylmethoxy]-
phenyl}methane (7):Yield 88%. Ϫ M.p. 191Ϫ193°C. Ϫ IR: ν˜ [cmϪ1
]
3413 br, 3307 br (NH), 1759, 1689, 1667 (CO). Ϫ 1H NMR
(CD2Cl2): δ ϭ 1.16 (s, 27 H, tBu), 1.21 (s, 27 H, tBu), 3.61 (s, 9 H,
OMe), 3.69 (br, 6 H, NHCH2), 4.04 (s, 6 H, ArOCH2), 6.57 (s, 1
H, CH tripod), 7.19 (d, J ϭ 2.4 Hz, 3 H, Ar-H), 7.22 (d, J ϭ
2.4 Hz, 3 H, Ar-H), 7.64 (t, 3 H, J ϭ 5.9 Hz, NH). Ϫ 13C NMR
(CD2Cl2): δ ϭ 31.2, 31.4 [C(CH3)3], 34.6, 35.4 [C(CH3)3], 36.9 (CH
tripod), 40.8 (NHCH2), 52.3 (OCH3), 72.2 (ArOCH2), 123.7, 127.1,
137.1, 142.6, 146.6, 152.0 (CAr), 169.3 (NHCϭO), 171.3 (MeOCϭ
O). Ϫ C58H85O12N3 (1016.3): calcd. C 68.53, H 8.43, N 4.14; found
C 68.81, H 8.76, N, 4.02.
Determination of Binding Affinities: All extraction studies were car-
ried out at 23 °C, with freshly distilled analytical grade solvents,
using the methodology and equations described by Cram and co-
workers.[22] Absorbance values were determined using a Varian
Cary 50, at 380 nm. The picrate salts were prepared according to
literature procedures,[23] and were thoroughly dried under vacuum
prior to use. Aqueous solutions of the picrates were prepared that
were 0.015  for the Liϩ, Naϩ, and Kϩ salts, and 0.010  for the
less soluble Ruϩ and Csϩ salts. Aliquots of the each of the picrate
Tris{3,5-di-tert-butyl-2-[N-(methyl alanyl)carbonylmethoxy]-
phenyl}methane (8):Yield 92%. Ϫ M.p. 92Ϫ94°C. Ϫ IR: ν˜ [cmϪ1
]
3411 br (NH), 1748, 1683 (CO). Ϫ 1H NMR: δ ϭ1.17 (s, 27 H, solution were transferred into five glass vials using an Eppendorf
tBu), 1.30 (s, 27 H, tBu), 1.45 (d, J ϭ 7.2 Hz, 9 H, CH3CH), 3.91 micropipetter, 0.50 mL for the Liϩ, Naϩ, and Kϩ, and 0.75 mL for
(s, 6 H, ArOCH2), 4.67 (quintet, 3 H, J ϭ 7.2 Hz, NHCH), 6.41
the Ruϩ and Csϩ solutions. To these were added 0.5 mL of 0.015
(s, 1 H, CH tripod), 7.01 (d, J ϭ 5.4 Hz, 3 H, Ar-H), 7.21 (d, J ϭ  solutions of the host in CHCl3, and capped. The vials were thor-
5.4 Hz, 3 H, Ar-H), 7.49 (s, 3 H, br, NH). Ϫ 13C NMR: δ ϭ 17.9
(CH3CH), 13.1, 31.3 [C(CH3)3], 34.4, 35.2 [C(CH3)3], 38.3 (CH tri-
oughly shaken for 2 min, and allowed to separate. Aliquots of
0.100 mL and 0.200 mL were carefully syringed from the organic
pod), 42.6 (NHCH2), 47.5 (NHCH), 52.3 (OCH3), 71.8 (ArOCH2), and aqueous phases, respectively, using Hamilton Gas-tight syr-
123.1, 126.7, 136.6, 141.9, 146.0, 152.0 (CAr), 168.2 (NHCϭO), inges, and transferred to 10 mL volumetric flasks, which were sub-
173.0 (MeOCϭO ). Ϫ C61H91O12N3 · 1/2 CHCl3 (1058.4): calcd. C sequently bought to the mark with CH3CN. The absorbance of
66.09, H 8.26, N 3.76; found C 66.00, H 8.47, N 3.65.
each sample was determined, and the R, Ka, ∆G° and percentage
Eur. J. Org. Chem. 2000, 2467Ϫ2478
2475

M. B. Dinger, M. J. Scott
FULL PAPER
extracted values calculated. The values used for the extinction coef-
ficients (ε) of the picrate salts in CH3CN and the distribution con-
stants (Kd) of the picrate salts between water and CHCl3, which
are required for the calculations, were those previously determined
by Cram and co-workers.[22]
X-ray Crystallographic Study: Unit cell dimensions and intensity
data for all the structures were obtained on a Siemens CCD
SMART diffractometer at Ϫ100°C, with the exception of
62NaBPh4 · 4.5 THF which was collected at Ϫ125°C, with mono-
˚
chromatic Mo-Kα X-rays (λ ϭ 0.71073 A). The data collections
Table 2. X-ray data for the crystal structures of 1a, 1d, 2b, 4, and 5
1a · 1/2 C6H6
1d
2b · 11/3 CH3OH
4
5
Total reflections
Unique reflections
Collection range
Tmax,min
27220
24668
12140
29403
55165
19934
5094
2520
10101
9238
1.74 Ͻ θ Ͻ 29.37°
1.00, 0.84
C46H67O3
668
1.72 Ͻ θ Ͻ 26.36°
1.00, 0.86
C25H28O3
376.47
trigonal
R3c
21.9188(2)
Ϫ
30.2450(2)
Ϫ
1.31 Ͻ θ Ͻ 23.26°
1.00, 0.88
C47.33H68.33O10.33
812.02
1.46 Ͻ θ Ͻ 26.41°
1.00, 0.47
C49H67O6Cl3
858.38
monoclinic
P21/n
12.2997(8)
22.743(2)
18.203(1)
Ϫ
1.88 Ͻ θ Ͻ 24.71°
1.00, 0.78
C55H85N3O6
884.26
orthorhombic
Pbca
18.928(3)
21.703(3)
26.376(5)
Ϫ
Empirical formula
Mr
Crystal system
Space group
triclinic
trigonal
¯
¯
P1
P3
˚
a [A]
14.0330(2)
14.5409(1)
24.159(1)
74.853(1)
78.696(1)
65.297(1)
4301.72(7)
1.031
17.9708(3)
˚
b [A]
Ϫ
˚
c [A]
9.3776(2)
Ϫ
α [°]
β [°]
γ [°]
Ϫ
120
Ϫ
120
104.460(1)
Ϫ
Ϫ
Ϫ
3
˚
Vc [A ]
12584.0(2)
1.192
2622.75(8)
1.028
4930.7(6)
1.156
10835(3)
Dc [g cmϪ3
Z
]
1.084
4
24
2
4
8
F(000)
1468
4848
1028
1840
3872
µ(Mo-Kα) [mmϪ1
R1 [I Ն 2σ(I)]
]
0.062
0.077
0.08
0.23
0.069
0.0705 [15421]
0.0603 [3754]
0.0805 [1923]
0.0776 [5820]
0.0597 [5266]
wR2 (all data); X, Y[a] 0.1680; 0.0367, 4.04 0.1489; 0.0633, 12.46 0.2651; 0.1675, 1.98 0.2346; 0.1057, 2.60 0.1726; 0.0867, 2.51
GoF
1.117
ϩ0.75, Ϫ0.37
1.102
ϩ0.87, Ϫ0.21
1.099
ϩ0.84, Ϫ0.38
1.063
ϩ0.42, Ϫ0.35
1.01
ϩ0.64, Ϫ0.33
Largest peak,
Ϫ3
˚
deepest trough [e A
]
[a]
2
2
2
w ϭ 1/[σ2(Fo ) ϩ (XP)2 ϩ YP] where P ϭ (Fo ϩ 2Fc )/3.
Table 3. X-ray data for the crystal structures of 7 · 1/2 C5H12, 8 · CHCl3, 52Li4picrate4 · 2 C6H6 · 2 H2O, 62NaBPh4 · 41/2 THF, and
72Na2(BPh4)2 · 1/8 C6H14 · 2/3 THF
7 · 1/2 C5H12
8 · CHCl3
52Li2[Li2picrate4]
· 2 C6H6 · 2 H2O
62NaBPh4
· 41/2 THF
72Na2(BPh4)2
· 1/8 C6H14 · 2/3 THF
Total reflections
Unique reflections
Collection range
Tmax,min
30003
5825
7205
3085
23725
15772
19013
11347
24616
16868
1.80 Ͻ θ Ͻ 22.50°1.94 Ͻ θ Ͻ 23.26° 1.48 Ͻ θ Ͻ 26.37° 1.02 Ͻ θ Ͻ 22.50°
1.03 Ͻ θ Ͻ 26.38°
1.00, 0.80
1.00, 0.79
C60.5H90.5N3O12 C62H92N3O12Cl3
1051.86
hexagonal
P63
12.4486(5)
Ϫ
1.00, 0.55
1.00, 0.92
1.00, 0.81
Empirical formula
Mr
C140H188Li4N18O42 C182H238BN6O16.5Na C167.42H217.08B2N6O24.66Na2
1177.74
hexagonal
R32
12.226(4)
Ϫ
77.63(4)
Ϫ
Ϫ
2822.84
triclinic
2807.58
triclinic
2775.92
triclinic
Crystal system
Space group
¯
¯
¯
P1
P1
P1
˚
a [A]
14.2094(6)
17.4119(8)
17.5694(8)
71.068(1)
83.519(1)
72.079(1)
3911.8(3)
1.198
14.0674(6)
15.8351(7)
21.6376(9)
108.384(1)
100.678(1)
98.822(1)
4377.3(3)
1.065
15.4947(7)
15.6292(7)
20.1703(9)
78.594(1)
88.268(1)
61.132(1)
4181.0(3)
1.102
˚
b [A]
˚
c [A]
50.871(3)
Ϫ
α [°]
β [°]
γ [°]
Ϫ
120
120
3
˚
Vc (A )
6827.1(5)
1.023
10048(7)
1.102
Dc [g cmϪ3
Z
]
4
6
1
1
1
F(000)
2282
3792
1502
1520
1493
µ(Mo-Kα) [mmϪ1
R1 [I Ն 2σ(I)]
]
0.07
0.194
0.088
0.069
0.077
0.0926 [4996]
0.0647 [1629]
0.1693; 0.0750,
0.00
0.0749 [10635]
0.2438; 0.1365,
2.45
0.1384 [8529]
0.3916; 0.2000,
16.00
0.0725 [8889]
0.2503; 0.1640,
0.00
wR2 (all data); X, Y[a] 0.2599; 0.1641,
4.05
1.167
GoF
0.967
ϩ0.40, Ϫ0.39
1.048
ϩ0.96, Ϫ0.56
1.039
ϩ0.74, Ϫ0.67
0.934
ϩ0.78, Ϫ0.80
Largest peak, deepest ϩ0.85, Ϫ0.56
Ϫ3
˚
trough (e A
)
[a]
2
2
2
w ϭ 1/[σ2(Fo ) ϩ (XP)2 ϩ YP] where P ϭ (Fo ϩ 2Fc )/3.
2476
Eur. J. Org. Chem. 2000, 2467Ϫ2478

C3-Symmetric, Conformational Mimics of Calix[n]arenes
FULL PAPER
[8]
[9]
B. C. Gibb, A. R. Mezo, A. S. Causton, J. R. Fraser, F. C. S.
Tsai, J. C. Sherman, Tetrahedron 1995, 51, 8719Ϫ8732.
F. Arnaud-Neu, S. Cremin, S. Harris, M. A. McKervey, M.-J.
nominally covered over a hemisphere of reciprocal space, by a com-
bination of three sets of exposures; each set had a different ϕ angle
for the crystal and each exposure covered 0.3° in ω. The crystal
to detector distance was ca. 5.0 cm. The data sets were corrected
empirically for absorption using SADABS.[24]
´
Schwing-Weill, P. Schwinte, A. Walker, J. Chem. Soc., Dalton
Trans. 1997, 329Ϫ334.
[10]
Y. Hamuro, M. C. Calama, H. S. Park, A. D. Hamilton, Angew.
Chem. 1997, 109, 2797Ϫ2799; Angew. Chem. Int. Ed. Engl.
1997, 36, 2680Ϫ2683.
All the structures were solved by using the Bruker SHELXTL soft-
ware package for the PC and the direct methods option of
SHELXS. Space groups for all of the structures were determined
from an examination of the systematic absences in the data, and
the successful solution and refinement of the structure confirmed
these assignments. When compounds crystallized in polar space
groups, both enantiomers were refined and the final structure
yielded the lower residuals. Moreover, the Flack parameter refined
to the lowest value for the chosen enantiomer. Except for the hydro-
gens involved in hydrogen-bonding interactions, all hydrogen atoms
were assigned idealized locations and were given a thermal para-
meter equivalent to 1.2 or 1.5 times the thermal parameter of the
carbon atom to which it was attached. For the methyl groups, where
the location of the hydrogen atoms is uncertain, the AFIX 137 card
was used to allow the hydrogen atoms to rotate to the maximum
area of residual density, while fixing their geometry. In cases of
extreme disorder, the non-hydrogen atoms were refined only iso-
tropically, and hydrogen atoms were not included in the model.
Structural and refinement data for all the compounds are presented
in Table 2 and Table 3 while complete experimental details are con-
tained in the supplementary information.
[11]
[12]
[13]
[14]
K. Tsubaki, T. Otsubo, K. Tanaka, K. Fuji, T. Kinoshita, J.
Org. Chem. 1998, 63, 3260Ϫ3265.
T. Yamato, M. Haraguchi, J. Nishikawa, S. Ide, J. Chem. Soc.,
Perkin Trans. 1 1998, 609Ϫ614.
T. Yamato, L. K. Doamekpor, H. Tsuzuki, Liebigs Ann./Re-
cueil 1997, 1537Ϫ1544.
R. G. Janssen, W. Verboom, D. N. Reinhoudt, A. Casnati, M.
Freriks, A. Pochini, F. Ugozzoli, R. Ungaro, P. M. Nieto, M.
Carramolino, F. Cuevas, P. Prados, J. de Mendoza, Synthesis
1993, 380Ϫ386.
C. Dinse, N. Baglan, C. Cossonnet, J. F. Le Du, Z. Asfari, J.
Vicens, J. Alloy Compd 1998, 271, 778Ϫ781.
H. Otsuka, Y. Suzuki, A. Ikeda, K. Araki, S. Shinkai, Tetrahe-
dron 1998, 54, 423Ϫ446.
S. Blanchard, L. Le Clainche, M.-N. Rager, B. Chansou, J.-P.
Tuchagues, A. F. Duprat, Y. Le Mest, O. Reinaud, Angew.
Chem. 1998, 110, 2861Ϫ2864; Angew. Chem. Int. Ed. 1998,
37, 2732Ϫ2735.
[15]
[16]
[17]
[18]
W. P. van Hoorn, F. C. J. M. van Veggel, D. N. Reinhoudt, J.
Phys. Chem. A 1998, 102, 6676Ϫ6681 and references tin.
B. Tse, Y. Kishi, J. Org. Chem. 1994, 59, 7807Ϫ7814.
J. R. Telford, K. N. Raymond, Comprehensive Supramolecular
Chemistry (Ed.: J. M. Lehn; Elsevier Science Ltd.: Oxford,
1996; Vol. 1, pp 245 and references therein.
[19]
[20]
[21] [21a]
G. Casiraghi, G. Casnati, M. Cornia, V. D’Azeglio, Tetra-
[21b]
hedron Lett. 1973, 679Ϫ682. Ϫ
G. Casiraghi, G. Casnati,
Crystallographic data (excluding structure factors) for the structures
reported in this paper have been deposited with the Cam-
bridge Crystallographic Data Centre as supplementary publication
nos. CCDC-106701 (1a), CCDC-106702 (1d), CCDC-131462 (2b),
M. Cornia, A. Pochini, Gazz. Chim. Ital. 1978, 108, 79Ϫ84.
K. E. Koenig, G. M. Lein, P. Stuckler, T. Kaneda, D. J. Cram,
J. Am. Chem. Soc. 1979, 101, 3553Ϫ3567.
[22]
[23]
S. S. Moore, T. L. Tarnowski, M. Newcomb, D. J. Cram, J.
Am. Chem. Soc. 1977, 99, 6405Ϫ6410.
CCDC-131463
(4),
CCDC-131464
(5),
CCDC-131465
[24]
[25]
R. H. Blessing, Acta Cryst. 1995, A51, 33.
{(5OH2)2[Li4picrate4]}, CCDC-131466 {(62Na)BPh4}, CCDC-
131467 (7), CCDC-131468 {(7Na)2(BPh4)2}, CCDC-140533 (8).
Copies of the data can be obtained free of charge on application to
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK [Fax: (internat.)
ϩ 44Ϫ1223/336-033; E-mail: deposit@ccdc.cam.ac.uk].
M. E. McGreal, V. Niederl, J. B. Niederl, J. Am. Chem. Soc.
1939, 61, 345Ϫ348.
[26]
[27]
J. C. Lockhart, M. B. McDonnell, W. Clegg, M. N. S. Hill, J.
Chem. Soc., Perkin Trans. 2 1987, 639Ϫ649.
C. Grüttner, V. Böhmer, R. Assmus, S. Scherf, J. Chem. Soc.,
Perkin Trans. 1 1995, 93Ϫ95.
[28]
[29]
Th. Zincke, Justus Liebigs Ann. Chem. 1939, 363, 268Ϫ274.
F. Arnaud-Neu, E. M. Collins, M. Deasy, G. Ferguson, S. J.
Harris, B. Kaitner, A. J. Lough, M. A. McKervey, E. Marques,
B. L. Ruhl, M. J. Schwing-Weill, E. M. Seward, J. Am. Chem.
Soc. 1989, 111, 8681Ϫ8691.
Acknowledgments
We thank the University of Florida, the National Science Founda-
tion ( Award 9874966) and the donors of the American
Chemical Society Petroleum Research Fund for funding this re-
search. Support from the Research Corporation is also gratefully
acknowledged. We also thank Dr. Khalil Abboud at the University
of Florida for granting access to X-ray instrumentation.
[30]
[31]
[32]
F. Sansone, S. Barboso, A. Casnati, M. Fabbi, A. Pochini, F.
Ugozzoli, R. Ungaro, Eur. J. Org. Chem. 1998, 897Ϫ905.
F. Arnaud-Neu, M.-J. Schwing-Weill, K. Ziat, S. Cremin, S. J.
Harris, M. A. McKervey, New J. Chem. 1991, 15, 33Ϫ37.
In all attempts to crystallize 2a, the molecule resided on a C3
symmetry axis and the ethoxy groups were severely disordered
in the lattice, precluding an accurate determination of the
structure. Nevertheless, the overall orientation of the arms with
respect to the platform could be unambiguously determined,
allowing for a cursory comparison to 2b. When crystallized
from methanol, the unit cell parameters for 2a were determined
[1]
C. D. Gutsche, Calixarenes Revisited; Royal Society of Chem-
˚
¯
istry: Cambridge, England, 1998.
to be hexagonal, space group R3, a ϭ 12.178(1) A, c ϭ 68.28(1)
3
[2]
˚
˚
A, V ϭ 8770(3) A , Z ϭ 6.
J. Vicens, V. Böhmer, Calixarenes: a Versatile Class of Macro-
[33]
[34]
[35]
cyclic Compounds; Kluwer Academic: Boston, 1991.
H. Kessler, Angew. Chem. 1982, 94, 509Ϫ520; Angew. Chem.
[3]
V. Böhmer, Angew. Chem. 1995, 107, 785Ϫ817; Angew. Chem.
Int. Ed. Engl. 1982, 21, 512Ϫ523.
Int. Ed. Engl. 1995, 34, 713Ϫ745.
Y. A. Ovchinnikov, V. T. Ivanov, Tetrahedron 1975, 31,
2177Ϫ2209.
[4]
C. Wieser, C. B. Dieleman, D. Matt, Coord. Chem. Rev. 1997,
165, 93Ϫ161.
F. Arnaud-Neu, G. Barret, S. Cremin, M. Deasy, G. Ferguson,
S. J. Harris, A. J. Lough, L. Guerra, M. A. McKervey, M.
J. Schwing-Weill, P. Schwinte, J. Chem. Soc., Perkin Trans 2
1992, 1119Ϫ1125.
[5]
G. Barrett, M. A. McKervey, J. F. Malone, A. Walker, F. Ar-
naud-Neu, L. Guerra, M.-J. Schwing-Weill, C. D. Gutsche, D.
R. Stewart, J. Chem. Soc., Perkin Trans. 2 1993, 1475Ϫ1479.
[6]
[36]
[37]
A. Dondoni, A. Marra, M.-C. Scherrmann, A. Casnati, F. San-
J. A. Hamilton, M. N. Sabesan, L. K. Steinrauf, J. Am. Chem.
sone, R. Ungaro, Chem. Eur. J. 1997, 3, 1774Ϫ1782.
Soc. 1981, 103, 5880Ϫ5885.
[7]
R. Seangprasertkij, Z. Asfari, F. Arnaud-Neu, J. Vicens, J. Org.
L. K. Steinrauf, J. A. Hamilton, M. N. Sabesan, J. Am. Chem.
Soc. 1982, 104, 4085Ϫ4091.
Chem. 1994, 59, 1741Ϫ1744.
Eur. J. Org. Chem. 2000, 2467Ϫ2478
2477

M. B. Dinger, M. J. Scott
FULL PAPER
[38]
[39]
F. Arnaud-Neu, V. Böhmer, J.-F. Dozol, C. Grüttner, R. A.
G. Casiraghi, G. Casnati, G. Puglia, G. Sartori, G. Terenghi, J.
Chem. Soc., Perkin Trans. 1 1979, 1862Ϫ1865.
Received February 11, 2000
Jakobi, D. Kraft, O. Mauprivez, H. Rouquette, M.-J. Schwing-
Weill, N. Simon, W. Vogt, J. Chem. Soc., Perkin Trans. 2
1996, 1175Ϫ1182.
[O00076]
2478
Eur. J. Org. Chem. 2000, 2467Ϫ2478