Efficient synthesis of calix[6]tmpa: A new calix[6]azacryptand with unique conformational and host-guest properties

Article abstract of DOI:10.1002/chem.200600278

A new C_3v-symmetrical calix[6]azacryptand, that is, calix[6] tmpa (11), was synthesized by efficient [1+1] macrocyclization reactions. Remarkably, both linear and convergent synthetic strategies that were applied lead to equally good overall yields. Calix[6]tmpa behaves as a single proton sponge and appeared reluctant to undergo polyprotonation, unlike classical tris(2-pyridylmethyl)amine (tmpa) derivatives. It also acts as a good host for ammonium ions. Interestingly, it strongly binds a sodium ion and a neutral guest molecule, such as a urea, an amide, or an alcohol, in a cooperative way. A ¹H NMR study indicated that the ligand, as well as its complexes, adopt a major flattened cone conformation that is the opposite of that observed with the previously reported calix[6]cryptands. Characterization of the monoprotonated derivative 11·H⁺ by X-ray diffraction also revealed the presence of a 1,3-alternate conformation, which is the first example of its kind in the calix[6]arene family. This conformer is probably also present in solution as a minor species. The important covalent constraint induced by the polyaromatic tmpa cap on the calixarene skeleton, and conversely from the calix core onto the tmpa moiety, is the likely basis for the unique conformational and chemical properties of this host.

Full text of DOI:10.1002/chem.200600278

FULL PAPER
DOI: 10.1002/chem.200600278
Efficient Synthesis of Calix[6]tmpa: A New Calix[6]azacryptand with Unique
Conformational and Host–Guest Properties
Xianshun Zeng,[a] David Coqui›re,[b] AurØlie Alenda,[b] Eva Garrier,[a] Thierry PrangØ,[c]
Yun Li,[b] Olivia Reinaud,*[b] and Ivan Jabin*[a]
Abstract: A new C3v-symmetrical cal-
ix[6]azacryptand, that is, calix[6]tmpa
(11), was synthesized by efficient [1+1]
macrocyclization reactions. Remarka-
bly, both linear and convergent syn-
thetic strategies that were applied lead
to equally good overall yields. Cal-
ix[6]tmpa behaves as a single proton
sponge and appeared reluctant to un-
dergo polyprotonation, unlike classical
tris(2-pyridylmethyl)amine (tmpa) de-
rivatives. It also acts as a good host for
ammonium ions. Interestingly, it
strongly binds a sodium ion and a neu-
tral guest molecule, such as a urea, an
amide, or an alcohol, in a cooperative
way. A 1HNMR study indicated that
the ligand, as well as its complexes,
adopt a major flattened cone confor-
mation that is the opposite of that ob-
served with the previously reported
calix[6]cryptands. Characterization of
the monoprotonated derivative 11·H+
by X-ray diffraction also revealed the
presence of a 1,3-alternate conforma-
tion, which is the first example of its
kind in the calix[6]arene family. This
conformer is probably also present in
solution as a minor species. The impor-
tant covalent constraint induced by the
polyaromatic tmpa cap on the calixar-
ene skeleton, and conversely from the
calix core onto the tmpa moiety, is the
likely basis for the unique conforma-
tional and chemical properties of this
host.
Keywords: calixarenes · cryptands ·
host–guest systems · supramolecular
chemistry
Introduction
Efficient molecular receptors toward either charged or neu-
tral species have been designed from calixarenes.[1] Among
the different cyclic oligomers, calix[6]arenes possess a hy-
drophobic cavity that is well adapted for the inclusion of or-
ganic guests. However, they are highly flexible and it is first
necessary to constrain them in a cone conformation.[2] For
this, a possible strategy consists of grafting a tripodal cap on
the narrow rim.[3] In this regard, we have developed an orig-
inal class of molecular receptors, which consist of a calixar-
ene core rigidified by a tripodal aza cap at the narrow rim.[4]
These C3v-symmetrical calix[6]azacryptands (Figure 1) dis-
play remarkable host properties toward either ammonium
ions, neutral polar molecules, or metal ions, depending on
the way they are polarized.[5] These versatile receptors have
also been successfully used in chiral recognition processes[6]
and in the modeling of active sites of enzymes.[7] In the
course of our research, we were interested in introducing a
tris(2-pyridylmethyl)amine (tmpa) cap on the calixarene
core. Indeed, tmpa has been widely used as an N4 donor
ligand, which mimicks the polyhistidine binding sites in met-
alloenzymes. Hence, tmpa-based Cu[8] and Fe[9] complexes
have been shown to reproduce some aspects of biocatalytic
[a] Dr. X. Zeng, Dr. E. Garrier, Dr. I. Jabin
UnitØ de Recherche en Chimie Organique et MacromolØculaire
(URCOM)
UniversitØ du Havre, 25 rue Philippe Lebon, BP 540, 76058 Le Havre
cedex (France)
Fax : (+33)232-744-391
E-mail: ivan.jabin@univ-lehavre.fr
[b] D. Coqui›re, A. Alenda, Dr. Y. Li, Prof. O. Reinaud
Laboratoire de Chimie et Biochimie Pharmacologiques et Toxicologi-
ques
UMR CNRS 8601, UniversitØ RenØ Descartes, 45 rue des Saints
P›res
75270 Paris cedex 06 (France)
Fax : (+33)142-862-183
E-mail: Olivia.Reinaud@univ-paris5.fr
[c] Dr. T. PrangØ
Laboratoire de Cristallographie et RMN Biologiques, UMR CNRS
8015
UniversitØ RenØ Descartes, 4 avenue de l’Observatoire
75270 Paris Cedex 06 (France)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2006, 12, 6393 – 6402
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
6393

tion of ammonia in THF con-
taining 4 produced the tmpa de-
rivative 5 in 82% yield. It is re-
markable that, under these con-
ditions, the [3+1] product
5
formed instead of the [1+1]
adduct or mixtures of com-
pounds. Finally, a classical two-
step sequence led to the desired
tris[6-(chloromethyl)-2-pyridyl-
Figure 1. The calix[6]azacryptands[4–7] and tmpa ligand.
methyl]amine (7) in 44% over-
all yield from 2 (Scheme 1).[18]
cycles, giving rise to interesting reactive species. The present
work describes the synthesis, conformational behaviour, and
preliminary host–guest properties of this novel calix[6]aza-
cryptand, namely calix[6]tmpa.
The [1+1] macrocyclization reaction between X6H3Me3
(1) and the tmpa derivative 7 was performed in the presence
of a Cs2CO3/K2CO3 mixture (0.5 and 3 equiv, respective-
ly).[19] Under these basic conditions, calix[6]tmpa (11) was
isolated in 47% yield. The use of the stronger base NaHled
to a lower yield (36%) (Scheme 2). It should be noted that,
Results and Discussion
Synthesis of the calix[6]tmpa
(11): Different synthetic strat-
egies have been described in
the literature for the covalent
linkage of a tripodal cap onto
the narrow rim of calix[6]ar-
enes.[10] The convergent strategy
consists of using a [1+1] macro-
cyclization reaction between
Scheme 1. i) NaBH4, MeOH/CH2Cl2, RT, 87%; ii) PBr3, CH C3l, RT, 99%; iii) NH3, K2CO3, THF, 508C, 82%;
iv) NaBH4, EtOH, reflux, 82%; v) SOCl2, RT, 76%.
two tripodal partners, that is,
the 1,3,5-tris-O-methylated cal-
ix[6]arene 1 (X6H3Me3)[11] or an
appropriately 2,4,6-tris-O-functionalized derivative and a tri-
podal cap bearing either nucleophilic or electrophilic group-
s.[4a,c,12] The linear strategy has been successfully applied
with a [3+1] macrocyclization reaction between a calix[6]-
trisamine and formaldehyde[4b] or through a proton-cata-
lyzed cyclotrimerization of veratryl units.[13] For the synthe-
sis of calix[6]tmpa (11), we have investigated both strategies.
On the one hand, a tmpa derivative that can be grafted onto
1 through a [1+1] macrocyclization reaction was prepared.
On the other hand, we introduced, at the narrow rim of 1,
6-substituted picolyl arms that can lead to the formation of
the tmpa unit through a macrocyclization reaction with am-
monia. To our knowledge, only one example of such a direct
bi(macrocycle) formation with ammonia has been report-
ed.[14]
For the first strategy, it was necessary to synthesize a
tmpa derivative bearing electrophilic groups on the 6-posi-
tion of the pyridyl residues. The tris-6-hydroxymethyl pre-
cursor 6 was described in the literature,[15] but its synthesis
was reinvestigated in the course of this work.[16] Our synthet-
ic strategy was based on the recently reported regioselective
reduction of the dimethyl pyridine-2,6-dicarboxylate (2) into
3 by NaBH4 (87% yield).[17] Treatment of 3 with PBr3 in
chloroform afforded the corresponding 6-bromomethyl de-
rivative 4 in 99% yield. Heating to 508C in a saturated solu-
Scheme 2. i) 7, K2CO3, Cs2CO3, KI, DMF, 908C, 47%; ii) 7, NaH, NaI,
THF/DMF, reflux, 36%; iii) 4, NaH, THF, reflux, 99%; iv) LiAlH4, Et2O,
reflux, 88%; v) PBr3, CH Cl, RT, 97%; vi) NH3, Na2CO3, THF, 608C,
3
62%.
6394
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6393 – 6402

Calix[6]arene Azacryptand
FULL PAPER
in both cases, no other macro-
cyclization products could be
clearly detected by 1HNMR
spectroscopy.
For the second synthetic
pathway, X6H3Me3 1 was first
alkylated in quantitative yield
with an excess of methyl 6-(bro-
momethyl)picolinate
4
(5 equiv) in the presence of a
strong base (NaH). Reduction
of the ester groups of 8 by lithi-
um aluminum hydride (LAH)
afforded the corresponding tris-
hydroxy derivative 9 in 88%
yield. Subsequent treatment
with PBr3 led to the tris-bromo
derivative 10 in high yield
1
(97%). The HNMR spectra of
the calix[6]arene derivatives 8,
9, and 10 show the presence of
a major flattened C3v-symmetri-
cal cone conformation with the
anisole tBu substituents in the
out
position
(dtBu =1.36–
1.38 ppm), their corresponding
methoxy group being oriented
towards the center of the cavity
(dOMe =2.34–2.35 ppm) (see the
Supporting Information: Fig-
ures S15, S17, S19). It is note-
worthy however that minor dis-
Figure 2. 1HNMR spectra (300 MHz, CDCl 3): a) 11 (293 K); b) endo complex 11·Na+ꢀEtOH (293 K); c)
11.H+,PicÀ (263 K); d) endo complex 11ꢀPrNH3+,PicÀ (263 K); : free PrNH2. Residual solvents, water and
!
the minor conformers (see the text) have been labeled “S”, “W” and “Minor conf.”, respectively.
symmetrical alternate conformations are detectable, as pre-
viously observed for similar calix[6]arenes bearing picolyl
residues.[20] Similarly to the preparation of compound 5, the
reaction of calixarene 10 with a saturated solution of ammo-
nia in THF at 608C led, in the presence of Na2CO3, to 11 in
62% yield (Scheme 2). This high yield does not result from
a templating effect of the sodium cation since 11 was isolat-
ed in a comparable 55% yield when the reaction was con-
ducted in absence of Na2CO3.
If we compare the two different pathways for the synthe-
sis of 11, they appear equally efficient as they lead to similar
overall yields from 1 (47% and 52%, respectively), and are
both easy to carry out on a large scale.
units linked to the tmpa cap present the more high-field
shifted resonances (dArH =6.80 ppm, dtBu =0.86 ppm), where-
as those belonging to the tBu-anisole units are relatively
down-field shifted (dArH =7.26 ppm, dtBu =1.37 ppm) (see the
Supporting Information: Figure S23). Hence, the former
protons experience the anisotropic ring current from the
tBu-anisole aromatic cores. This shows that the tBuArO-
AHCTRE(UNG tmpa) aromatic units are in an in-position, whereas the ani-
sole walls stand away from the C3v axis, thus forming the ex-
ternal shell of the calixarene cavity. Despite the presence of
the tmpa cap that precludes the inclusion of all three O-
methyl substituents in the calixarene cone conformation due
to overcrowding, the corresponding CH3 protons present a
high-field shifted resonance (dOMe =2.53 ppm). This can be
explained by their relative proximity to the C3v axis and, as
a consequence, to the aromatic units of the tmpa and calix-
arene cores.
Such a conformation differs from those adopted by all
other calixazacryptands which we have described so far.
Indeed, in the case of calix[6]tac (Figure 1), the capping of
the narrow rim produced a calixarene with an inverted flat-
tened cone conformation compared with 11, that is, the ani-
sole tBu substituents (dtBu =0.73 ppm) being in an in-posi-
1H NMR conformational study of calix[6]tmpa (11): The
conformational properties of 11 have been studied by 1H
NMR spectroscopy. The spectrum recorded at 293 K (Fig-
ure 2a) is characteristic of a major C3v-symmetrical flattened
cone conformation with sharp doublets at d=3.43 and
4.45 ppm for the ArCH2 methylene protons. The NMR pro-
file barely changed in the 263–330 K temperature range, be-
sides some broadening of the OMe signal at low tempera-
ture (see the Supporting Information: Figure S1). An heter-
onuclear multiple bond correlation (HMBC) experiment al-
lowed the attribution of the tBuAr resonances: the aromatic
tion with their OCH3 groups far from the C3v axis (dOMe
=
3.75 ppm). For calix[6]tren and PN3, a straight and regular
Chem. Eur. J. 2006, 12, 6393 – 6402
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
6395

I. Jabin, O. Reinaud et al.
cone conformation was observed (DdtBu < 0.09 ppm and
dOMe ꢁ3 ppm). This original conformational behavior of 11
may be related to the more rigid tmpa cap that is construct-
ed from aromatic amines instead of aliphatic amines. Inter-
estingly, such a conformation produces a cavity with a differ-
ent environment at the narrow rim because the oxygen
atoms bearing the tmpa cap are moved away from the C3v
axis, directing their lone pair electrons toward the outside of
the cavity (Scheme 2 and Figure 2a).
ing of the signals belonging to the included EtOHmolecule
was observed at 330 K. This observation emphasizes the re-
markable affinity of the Na+ complex for this neutral guest.
Lastly, the co-existence of a minor species was evidenced by
the presence of a small triplet at d=À1.45 ppm (Figure 2b,
insert) which was associated to different calixarene resonan-
ces, some of which were clearly observable in the 0–2 ppm
region. This probably corresponds to a minor alternate con-
formation, as discussed below.[23] Interestingly, the binding
of Na+ and EtOHappeared cooperative. Indeed, when the
endo complex 11·Na+ꢀEtOH was strongly dried in the solid
state under high vacuum, part of the bound EtOHwas re-
moved. Redissolving the resulting solid in CDCl3 showed a
mixture of 11·Na+ꢀEtOH and free ligand 11. This shows, in
solution and in the absence of EtOH, that the Na+ complex
is not stable and the tmpa cap spontaneously releases the
cation (probably as NaCl). However, adding approximately
three equivelants of ethanol to the solution quantitatively
restored the initial complex. This demonstrates that without
the EtOHguest, 11 is inefficient at binding Na+. Converse-
ly, without Na+, 11 is inefficient at binding EtOH.
A sodium “funnel complex” based on calix[6]tmpa (11): As
in the case of all other calix[6]azacryptands, no inclusion of
small neutral molecules by 11 could be evidenced by 1H
NMR spectroscopy in CDCl3. Interestingly however, when a
chloroform/EtOHsolution of 11 was treated with solid
NaCl, a new species was produced. After filtration and re-
1
moval of the solvents, the resulting solid was analyzed by H
NMR spectroscopy in CDCl3 at 293 K (Figure 2b). It
showed new resonances in the high-field region that were
unambiguously (COSY and NOESY, see the Supporting In-
formation: Figures S25 and S27) attributed to an ethanol
guest molecule. The latter appeared strongly bound to the
calix cavity as no free EtOHwas observable. The main spe-
cies is C3v-symmetrical with one equivalent of EtOHin the
In a chloroform solution, the displacement of the bound
EtOHwas performed upon the addition of a few molar
equivalents of other neutral guests such as (Æ)-propane-1,2-
diol (PPD), pyrrolidin-2-one (PYD), imidazolidin-2-one
(IMI), EtNH2 or DMF to the endo-complex 11·Na+ꢀEtOH
(Scheme 3). The complexation-induced upfield shifts (CIS)
and relative affinities obtained with these guests are report-
ed in Table 1. The relative affinities decrease according to
the sequence order IMI @ PYD > EtOH > EtNH2. Thus,
in contrast to the previously reported Zn2+ funnel complex-
es,[5b,24] an alkali metal ion such as Na+ allows the stronger
binding of cyclic ureas, amides, and primary alcohols com-
pared with primary amines. This is obviously related to the
strong oxophilic character of
calixarene
cavity
(dCH =À1.56,
dCH =1.44,
dOH =
3
2
2.16 ppm),[21] possesses a flattened cone conformation simi-
lar to that in 11 (dtBu =1.43 ppm (for the anisole unit as con-
firmed by HMBC) and DdtBu =0.62 ppm) and up-field shift-
À
ed CH2N and Py Hresonances. Therefore, the identity of
this new species has been assigned as 11·Na+ꢀEtOH, a
funnel complex in which the sodium ion allows the efficient
hosting of one equivalent of ethanol, as depicted in
Scheme 3.[22] The whole NMR spectrum barely changed in
the 261–330 K temperature range, and only a slight broaden-
the harder Na+ ion.
Characterization by X-ray dif-
fraction and hosting properties
of the mono-protonated deriva-
tive 11·H+: The host properties
of 11 with a protonated aza cap
were investigated as well. At
263 K, the addition of approxi-
mately 0.5 equivalents of picric
acid (PicH) to a solution of 11
in CDCl3 led to the formation
of an approximately 1:1 mixture
of a new C3v-symmetrical spe-
cies in slow exchange with 11
when compared with the NMR
time scale (see the Supporting
Information: Figure S2).[25]
When one equivalent of PicH
was added, 11 was fully con-
verted into this new species
(Figure 2c), which therefore
Scheme 3. i) NaCl, CHCl3, EtOH; ii) L (> 1 equiv), CDCl3; iii) PicH(1 equiv), CDCl 3; iv) RNH2 (> 1 equiv),
CDCl3.
6396
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6393 – 6402

Calix[6]arene Azacryptand
FULL PAPER
Table 1. Relative affinities of the neutral guests L towards host 11·Na+
and NMR complexation-induced upfield shifts (CIS) observed upon their
endo complexation (solvent: CDCl3).
tonated calix[6] azacryptands (i.e. calix[6]tren·H+ and cal-
ix[6]PN3·H+).[4c,5b] The third calix structure observed in the
solid state possesses an unexpected 1,3-alternate conforma-
tion (Figure 3, right). One anisole unit presents its tBu sub-
stituent in an out position with its OMe group pointing to-
wards the center of the calix cavity. The two others are
upside down, with their tBuAr moieties folded back at the
level of the small rim of the calixarene. Interestingly, this
1,3-alternate conformation appears to be stabilized by a
series of CH–p interactions: all three methoxy groups are
oriented toward the inside of the cavity at relatively short
distances from two p bonds belonging to the calix aromatic
core, one on each side; the upside down tBu groups are also
found next to the pyridine units at distances denoting CH–p
interactions (Figure 4). To the best of our knowledge, this is
Entry
L
Rel. aff.[a]
CIS [ppm][b]
a
b
g
1
2
3
4
5
6
EtOH1
(Æ)-PPD
EtNH2
DMF
PYD
IMI
À2.27
n.d.[c]
À2.17
À1.74
–
À2.79
n.d.[c]
À2.72
–
-
0.015
0.16
0.33
1.9
À2.53[d]
–
À2.61[e]
À2.48, À2.86[f]
À2.74
À2.63
12.3
–
–
[a] Relative affinity calculated at 293 K and defined as [Lin]/
[EtOHT]/[LT], where indexes in included and T total amount.
Errors estimated Æ10%. [b] CIS calculated at 293 K and defined as Dd=
(free L). a, b, and g refer to the relative position of
ACHTRE[UNG EtOHin]”
=
=
d(complexed L) À d
ACHTREUNG
the protons to the oxygen atom or to the nitrogen atom in the case of
PrNH2. [c] Not determined due to overlapping of the host and guest sig-
nals. [d] g refers to the relative position of the protons to the oxygen
atom of the primary alcohol. [e] Average value of the two inequivalent
methyl groups. [f] Values determined for CH2CH2CH2 and CH2N, respec-
tively.
corresponds to the monoprotonated derivative 11·H+
(Scheme 3).[26] It is noteworthy that the addition of a large
excess of PicH(ca. 100 equiv) or a stronger acid such as
perchloric acid did not affect the NMR spectrum much,
beside some broadening of the resonances. This shows that
further protonation of the tmpa cap is highly disfavored, in
contrast to a simple tmpa unit[27] and to the other calix[6]-
ACHTREUNG
Figure 4. Top view of the 1,3-alternate conformer of 11·H+ displaying the
CH–p interactions (dashed lines) between the OMe and the aryl rings on
one hand, and between the tBu substituents and the pyridine groups on
the other. The C···p distances are between 3.1–3.5 and 3.7–4.1 Š, respec-
tively.
tion of 11 in CHCl3 that had been treated by a few molar
equivalents of HClO4 diluted in EtOH. The crystal unit con-
tains three calixarene cryptands that are monoprotonated as
attested by the presence of three perchlorate counterions.
Two calixarene cores adopt a flattened cone conformation
(see Figure 3, left, for one of these two cone conformations)
the first example in the solid state of a 1,3-alternate confor-
mation of a calix[6]arene derivative.
The NMR spectrum of 11·H+ (Figure 2c) suggests that in
solution too, the protonated calix cryptand adopts two dif-
ferent conformations.[26] The major conformation clearly cor-
responds to the cone conformation observed in the solid-
state structure because it displays a C3v-symmetrical pattern
with several data attesting to the monoprotonation of the
tmpa cap: 1) a downfield shift of the CH2N and pyridine
À
(Py H) signals; 2) the presence of a broad singlet at d=
10.01 ppm, integrating for one proton and corresponding to
the NH+ resonance (D2O-exchanged). The presence of a
minor dissymmetrical conformation is shown by extra reso-
nances clearly observable in the d=0–2 ppm region[28] and
may well correspond to the 1,3-alternate conformation ob-
served in the solid state. Such a conformation may also be
relevant to the minor species detected in the spectrum of
complex 11·Na+ꢀEtOH (Figure 2b).
Figure 3. Crystal structure of 11·H+,ClO4À. Left: one of the two cone
conformations (see the text); Right: 1,3-alternate conformation. Hydro-
gen atoms and some crystallization solvent molecules have been omitted
for clarity.
Upon the addition of a large excess of small polar neutral
molecules such as DMSO, EtOH, MeOH, and IMI into a
solution of 11·H+ in CDCl3, no inclusion was detected even
at low temperature (223 K). Indeed, we have previously
shown that the recognition of polar neutral molecules
that fully corresponds to that described for 11 above. It is
noteworthy that these flattened cone conformations of
11·H+ present aromatic units that are in alternate in and out
positions compared with the previously reported monopro-
Chem. Eur. J. 2006, 12, 6393 – 6402
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
6397

I. Jabin, O. Reinaud et al.
through charge–dipole interactions and hydrogen bonding
requires a polycationic cap with multiple NH+ sites.[5b,29]
Yet, the addition of a few equivalents of a primary amine
RNH2 (with R=Et, nPr, or nBu) to 11·H+ led to the forma-
tion of C3v-symmetrical endo complexes with characteristic
high-field-shifted signals for the included alkyl chain (see
Figure 2d for R=Pr, and the CIS values reported in
Table 2). It is noteworthy that, upon complexation of the
of the cavity (as indicated by similar CIS values) and an
identical guest preference for R=Pr.[5b]
Conclusion
The synthesis of a new C3v-symmetrical calix[6]azacryptand,
that is, calix[6]tmpa (11), was achieved according to two effi-
cient strategies in remarkably similar overall yields from cal-
ixarene 1 (47% and 52%, respectively). Calix[6]tmpa (11)
and its sodium and protonated derivatives (11·Na+ and
11·H+) display conformational properties that differ from
the properties previously observed for all the other calix[6]-
azacryptands, and are probably due to the high sterical con-
strains inferred from the tmpa cap. Also, interestingly, the
NMR study in CDCl3 showed that whereas 11 readily under-
goes monoprotonation in the presence of an acid, it appears
reluctant to undergo polyprotonated, unlike the more flexi-
ble calix[6]azacryptands presenting alkylamines instead of
aromatic amines. The monoprotonated derivative 11·H+ be-
haves as a good receptor for amines, leading to inclusion
complexes 11ꢀRNH3+, but it is not polarized enough to
bind small polar neutral molecules in chloroform. Finally,
the cooperative complexation of Na+ in the tmpa cap and
neutral guests in the cavity was evidenced, providing the
first example of a funnel complex binding an alkali-metal
cation. This Na+ complex possesses unique host properties
as compared with the related Zn2+ funnel complexes[24]
since it preferentially includes cyclic ureas, amides, or alco-
hols rather than primary amines. Hence, 11 provides a new
example of synergistic combination of a polyaza site and a
calix[6]arene structure. Indeed, on the one hand, a simple
tmpa molecule undergoes trisprotonation in the presence of
perchloric acid and leads to a bis-tmpa Na+ complex with
NaClO4.[31] Such different behavior shows the importance of
geometrical control owing to the calixarene macrocycle. On
the other hand, whereas calix[4]arenes functionalized at the
small rim with O-donors have been shown to efficiently
complex sodium ions,[32] cooperative guest coordination is
only possible with a calix[6]arene that provides a wider
small rim leaving access for guest binding to Na+. In conclu-
sion, calix[6]tmpa (11) constitutes a new member of the cal-
ix[6]azacryptand family with different conformational be-
havior and promising binding properties toward neutral or
charged species.
+
Table 2. Relative affinities of the ammonium ion RNH3 towards host
11 and NMR complexation-induced upfield shifts (CIS) observed upon
their endo-complexation in CDCl3.
Entry Ammonium Ion
Rel.
CIS [ppm][b]
+
RNH3
aff.[a]
a
b
g
d
+
1
2
3
EtNH3
0.20
1
0.48
À2.41 À2.38
–
–
–
+
nPrNH3
nBuNH3
n.d.[c] À1.87 À2.81
+
n.d.[c] À1.28 À2.76 À2.63
+
[a] Relative affinity calculated at 293 K and defined as [RNH3
+
in]/
and T=total amount. Errors estimated Æ10%. [b] CIS calculated at
263 K and defined as Dd=d(complexed ammon
ACHTREUNG ACHTREUGN
[PrNH3 in]”[PrNH2(H+)T]/[RNH2(H+)T] where indexes in=included
A
ium ion). a, b, g, and d refer to the relative position of the protons to the
charged nitrogen atom of the ammonium ion (i.e. Cd-Cg-Cb-Ca-NH3+).
ACHTREUNG[c] Not determined due to overlapping of the host and guest signals.
amine, the CH2N signal of the tmpa cap of 11·H+ experi-
enced an impressive high-field shift (DdCH N =1.23 ppm rela-
2
+
À
tive to 11·H ꢀ1), whereas the Py Hresonances were only
slightly affected. This evidences a proton transfer from the
tmpa cap to the amine, hence showing that the endo com-
+
plex indeed corresponds to 11ꢀRNH3 rather than to 11·H+
ꢀRNH2 (as schematized in Scheme 3). Although we were
unsuccessful in assigning the tBu peaks by HMBC, we can
1
gather from the only moderate changes observed in the H
NMR spectrum (Figure 2) that the calixarene core of
+
11ꢀRNH3 likely remains in the flattened cone conforma-
tion characteristic of 11. When the spectrum was recorded
at room temperature, the in and out exchange process of
+
RNH3 remained slower than the NMR time scale, high-
lighting the strong affinity of the receptor for these ammoni-
um ions. Very interestingly, in the case of the endo complex
11ꢀPrNH3+, a splitting of some resonances of the host (i.e.
CH2O and CH2N of the cap) and of the guest (i.e. CH3CH2)
was observed at 220 K (see the Supporting Information:
Figure S3). This is clearly due to the well-known helical
twisting of the three nitrogenous arms of the cap,[20,30] the
host providing a chiral C3 environment sensed by the guest.
Finally, competitive binding experiments monitored by 1H
NMR spectroscopy at 263 K (see the Experimental Section
Experimental Section
General: CH2Cl2 was distilled over CaH2 under argon. CHCl3 was distil-
led over P2O5 under argon. DMF was distilled over silica gel/MgSO4
under argon. Diethyl ether and THF were distilled over sodium/benzo-
phenone under argon. Ethanol was distilled over sodium/diethylphthalate
under argon. All reactions were performed under an inert atmosphere. In
all cases, NaHwas not washed prior to use. Silica gel (230–400 mesh) was
used for flash chromatography separations. 1Hand 13C NMR spectra
were recorded either with a Bruker AC 200, ARX 250, Avance 300 or
Avance 500 apparatus. Chemical shifts are expressed in ppm. In most
+
for the detailed procedure) showed that PrNH3 is the best
À
guest, thus corresponding to optimal cavity filling and CH
p interactions (Table 2). Hence, in spite of the conforma-
tional differences with calix[6]tren, calix[6]tmpa (11) exhib-
its analogous complexation properties toward ammonium
ions, with a comparable positioning of the ions in the heart
6398
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6393 – 6402

Calix[6]arene Azacryptand
FULL PAPER
À
cases, they were assigned through HMQC and/or HMBC, COSY,
NOESY experiments. Traces of residual solvent were used as internal
standard. IR spectra were recorded on a Perkin-Elmer Spectrum One
FT-IR spectrometer using either KBr pellets or the Attenuated Total Re-
flectance (ATR) method. EIMS analyses were performed with a Finnigan
LCQ Advantage apparatus. Elemental analyses were performed at the
Service de Microanalyse (I.C.S.N., Gif sur Yvette, France) and at the
Laboratoire de Microanalyse Organique (IRCOF, France).
6H; CH2N), 4.93 (s, 6H; CH2O), 7.04 (d, J=8 Hz, 3H; Py H), 7.06 (d,
J=8 Hz, 3H; Py H), 7.55 ppm (t, J=8 Hz, 3H; Py H); 13C NMR
(75 MHz, CDCl3): d=59.8, 64.1, 120.5, 121.8, 137.5, 157.9, 160.6 ppm;
ESI-MS (CH3CN): m/z (%): 381.4 ([M+H]+, 100).
À
À
Tris[6-(chloromethyl)-2-pyridylmethyl]amine (7): SOCl2 (22.0 mL,
302 mmol) was added dropwise to compound 6 (3.18 g, 8.36 mmol) at
08C. The resulting suspension was stirred for 5.5 h at room temperature
and the excess of SOCl2 was removed under reduced pressure. The resi-
due was dissolved in CH2Cl2 (200 mL) and an aqueous solution of NaOH
(1m) was added slowly at 08C until the mixture became basic (pH > 12).
The organic layer was separated and the aqueous layer was extracted
with CH2Cl2 (2”200 mL). The combined organic layers were washed
with water (3”100 mL), dried over Na2SO4, and evaporated to dryness.
Et2O (2”150 mL) was added to the resulting brown solid and the solu-
tion was stirred for 1 h and filtered. After removal of Et2O under re-
duced pressure, compound 7 was obtained as a pure colorless solid
(2.77 g, 76%). This product is not stable at room temperature and de-
composes within a few days. However, it is stable at lower temperature
(À188C) for approximately one month. M.p. 818C; IR (KBr): n˜ =1589,
Methyl 6-(hydroxymethyl)picolinate (3): NaBH4 (390 mg, 10.3 mmol) was
slowly added, at 08C, to a solution of dimethyl pyridine-2,6-dicarboxylate
(2; 2.00 g, 10.2 mmol) in a dry 7:3 mixture of MeOH/CH2Cl2 (100 mL).
The reaction mixture was stirred for 3 h at room temperature and then
neutralized with an aqueous saturated NH4Cl solution. After extraction
with CH2Cl2 (3”50 mL), the combined organic layers were dried with
Na2SO4 and the solvent was removed under reduced pressure. The result-
ing crude residue was purified by column chromatography (hexane/
EtOAc; 1:1 then 1:2) giving the mono-alcohol 3 (1.49 g, 87%) as a white
solid. The melting point was identical to the one described in the litera-
ture:[33] m.p. 888C; IR (KBr): n˜ =1742 cmÀ1; 1HNMR (300 MHz, CDCl 3):
d=3.43 (brs, 1H; OH), 3.98 (s, 3H; OCH 3), 4.86 (s, 2H; CH2O), 7.54 (d,
1574 cmÀ1 1HNMR (300 MHz, CDCl 3): d=3.88 (s, 6H; CH2N), 4.62 (s,
;
À
À
À
À
J=8 Hz, 1H; Py H), 7.85 (t, J=8 Hz, 1H; Py H), 8.02 ppm (d, J=8 H z,
6H; CH2Cl), 7.30 (d, J=8 Hz, 3H; Py -H), 7.51 (d, J=8 Hz, 3H; Py
1H; Py H); 13C NMR (75 MHz, CDCl3): d=53.0, 64.7, 123.9, 124.2,
H), 7.66 ppm (t, J=8 H z; 3H , PyH); 13C NMR (75 MHz, CDCl3) d=
À
À
137.8, 147.1; 160.4, 165.7 ppm; EIMS: m/z (%): 167 ([M]+, 9), 109 (100),
46.8, 60.0, 121.1, 122.3, 137.5, 155.9, 159.1 ppm; EIMS (MeOH): m/z (%):
91 (81).
435 ([M
H]+, 100).
Methyl 6-(bromomethyl)picolinate (4): PBr3 (1.04 mL, 11.1 mmol) was
added, at 08C, to mono-alcohol 3 (1.72 g, 10.3 mmol) in anhydrous
CHCl3 (150 mL). The reaction mixture was stirred for 4 h at room tem-
perature and then neutralized at 08C with an aqueous saturated K2CO3
solution. After extraction with CH2Cl2 (2”50 mL), the combined organic
layers were dried with Na2SO4 and the solvent was removed under re-
duced pressure, leading to pure compound 4 as a white solid (2.34 g,
99%). No further purification of the compound 4 was necessary. The syn-
thesis of compound 4 was previously reported with a different proce-
dure[34] and, to our knowledge, no characterization data has been report-
5,11,17,23,29,35-Hexa-tert-butyl-37,39,41-trimethoxy-38,40,42-tris-[(6-me-
thoxycarbonyl-2-pyridyl)methoxy]calix[6]arene (8): NaH(60 wt% in oil,
1.48 g, 37.0 mmol) was added in small portions into a solution of methyl
6-(bromomethyl)picolinate (4; 4.40 g, 19.1 mmol) and X6H3Me3 (1; 3.89 g,
3.83 mmol) in anhydrous THF (160 mL) at 08C. After refluxing for 23 h,
the reaction mixture was cooled to 08C and water was slowly added until
the gas release ceased. The solvent was removed under reduced pressure,
the resulting residue was dissolved in CH2Cl2 (200 mL) and washed with
water (2”25 mL). After solvent evaporation, the residue was triturated
with EtOH(10 mL) and the resulting solid was isolated by filtration,
ed. m.p. 988C; IR (ATR): n˜ =1739 cmÀ1
;
1HNMR (300 MHz, CDCl 3):
À
d=4.00 (s, 3H; OCH3), 4.65 (s, 2H; CH2), 7.69 (d, J=8 Hz, 1H; Py H),
yielding compound
8
(5.55 g, 99%) as
a
white solid. m.p. 2248C
1HNMR
7.87 (t, J=8 Hz, 1H; Py H), 8.05 ppm (d, J=8 Hz, 1H; Py H); 13C
NMR (75 MHz, CDCl3): d=33.0, 53.2, 124.6, 127.4, 138.6, 147.5, 157.5,
165.2 ppm; ESI-MS (CH3CN): m/z (%): 230.0 ([M+H]+, 100), 232.0
([M+H]+, 92); elemental analysis calcd (%) for C8H8BrNO2 (230.06): C
41.77, H3.50, N 6.09; found: C 41.79, H3.27, N 6.12.
(decomp); IR (KBr): n˜ =2957, 1727, 1592, 1483 cmÀ1
;
À
À
(300 MHz, CDCl3): d=0.82 (s, 27H; tBu), 1.36 (s, 27H; tBu), 2.34 (s, 9H;
OCH3calix), 3.42 (d, J=15 Hz, 6H; ArCH2), 3.98 (s, 9H; OCH3pyr), 4.57
(d, J=15 Hz, 6H; ArCH2), 5.20 (s, 6H; OCH2), 6.70 (s, 6H; ArHcalix),
À
7.26 (s, 6H; ArHcalix), 7.96 (t, J=7 Hz, 3H; Py H), 8.07 (d, J=7 Hz, 3H;
Py H), 8.19 ppm (d, J=7 Hz, 3H; Py H); 13C NMR (75 MHz, CDCl3):
d=29.7, 31.2, 31.7, 34.1, 34.3, 53.0, 60.4, 74.7, 123.9, 124.1, 124.7, 128.1,
133.0, 133.6, 138.1, 146.1, 146.4, 147.2, 151.4, 154.4, 158.9, 165.8 ppm; ele-
mental analysis calcd (%) for C93H111N3O12·2.5H2O (1507.94): C 74.07, H
7.75, N 2.79; found: C 73.87, H7.66, N 2.59.
À
À
Tris[6-(methoxycarbonyl)-2-pyridylmethyl]amine (5): NH3 was bubbled
into a solution of 4 (2.00 g, 8.69 mmol) in THF (150 mL) at room temper-
ature for 30 min. K2CO3 (2.70 g, 19.5 mmol) was then added, the mixture
1
was stirred at 508C and the reaction was monitored by HNMR spectro-
scopy in CDCl3. Once the reaction appeared complete, the solvent was
removed under reduced pressure, the resulting residue was dissolved in
CH2Cl2 (30 mL) and washed with water (20 mL). The aqueous layer was
extracted twice with CH2Cl2 (2”15 mL) and the combined organic layers
were dried with Na2SO4. After removal of the solvent under reduced
pressure, recrystallization from CH2Cl2/Et2O (1:1) afforded pure product
5,11,17,23,29,35-Hexa-tert-butyl-37,39,41-trimethoxy-38,40,42-tris-[(6-hy-
droxymethyl-2-pyridyl)methoxy]calix[6]arene (9): LiAlH4 (50 mg,
1.3 mmol) was added into
a solution of calix[6]arene 8 (200 mg,
0.14 mmol) in anhydrous Et2O (14 mL). After refluxing for 16 h, the re-
action mixture was cooled to 08C and HCl (4n) was slowly added until
pH<3. The solvent was removed under reduced pressure and the result-
ing residue was dissolved in CH2Cl2 (50 mL), washed with an aqueous
solution of NaOH(1 m, 15 mL) and then twice with water (2”10 mL).
After solvent evaporation, EtOH(2 mL) was added to the residue and
the resulting solid was isolated by filtration, yielding compound
(167 mg, 88%) as a white solid. m.p. 2258C (decomp); IR (KBr): n=
2961, 1597, 1482 cmÀ1 1HNMR (300 MHz, CDCl 3): d=0.84 (s, 27H;
5 (1.10 g, 82%) as a white solid. m.p. 948C; IR (ATR): n˜ =1738 cmÀ1; H
NMR (250 MHz, CDCl3): d=4.02 (s, 9H; OCH3), 4.05 (s, 6H; CH2),
1
7.80–7.90 (m, 6H; Py H), 8.01 ppm (d, J=7 Hz, 3H; Py H); 13C NMR
(75 MHz, CDCl3): d=53.3, 60.3, 124.1, 126.7, 137.8, 147.9, 160.1,
166.2 ppm; ESI-MS (CH3CN): m/z (%): 487.3 ([M+H]+, 100); elemental
analysis calcd (%) for C24H24N4O6·0.33H2O (470.42): C 61.27, H5.28, N
11.91; found: C 61.29, H5.05, N 11.86.
À
À
9
;
tBu), 1.38 (s, 27H; tBu), 2.35 (s, 9H; OCH3), 3.43 (d, J=15 Hz, 6H;
ArCH2), 4.61 (d, J=15 Hz, 6H; ArCH2), 4.76 (s, 6H; CH2OH), 5.11 (s,
Tris[6-(hydroxymethyl)-2-pyridylmethyl]amine (6): NaBH4 (360 mg,
9.52 mmol) was added to
a suspension of tris-ester 5 (230 mg,
À
6H; PyrCH2O), 6.72 (s, 6H; ArHcalix), 7.18 (t, J=4 Hz, 3H; Py H), 7.28
0.495 mmol) in anhydrous EtOH(8 mL) at 0 8C. The reaction mixture
was refluxed for 2 h and the solvent was removed under reduced pres-
sure. The resulting crude residue was dissolved in CH2Cl2 (10 mL),
washed with brine (2”10 mL), and with an aqueous saturated Na2CO3
solution (2 mL). The organic layer was dried with Na2SO4 and concen-
trated. Recrystallization from CH2Cl2/Et2O (1:1) yielded pure 6 (154 mg,
82%) as a white solid. The characterization data were identical to those
described in the literature.[15] 1HNMR (200 MHz, CDCl 3): d=3.70 (s,
(s, 6H; ArHcalix), 7.75–7.86 ppm (m, 6H; Py H); 13C NMR (75 MHz,
À
CDCl3): d=29.8, 31.3, 31.8, 34.2, 34.4, 60.4, 64.0, 75.0, 119.4, 120.2, 123.9,
128.1, 133.2, 133.8, 137.8, 146.0, 146.3, 151.6, 154.5, 157.0, 158.2 ppm; ele-
mental analysis calcd (%) for C90H111N3O9·2H2O (1414.90): C 76.40, H
8.19, N 2.97; found: C 76.66, H8.03, N 2.65.
5,11,17,23,29,35-Hexa-tert-butyl-37,39,41-trimethoxy-38,40,42-tris-[(6-bro-
momethyl-2-pyridyl)methoxy]calix[6]arene
(10):
PBr3
(0.300 mL,
Chem. Eur. J. 2006, 12, 6393 – 6402
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
6399

I. Jabin, O. Reinaud et al.
3.19 mmol) was added to
a
solution of calix[6]arene
9
(600 mg,
27H; tBu), 1.40 (s, 27H; tBu), 2.85 (s, 9H; OCH3), 3.40 (d, J=15 Hz,
0.435 mmol) in CHCl3 (50 mL) and the reaction mixture was stirred for
2 h at room temperature. The reaction mixture was poured into an aque-
ous K2CO3 (sat.) solution (150 mL) at 08C. The aqueous layer was ex-
tracted with CH2Cl2 (2”30 mL) and the combined organic layers were
dried with Na2SO4. After filtration and removal of the solvent under
vacuum, pure compound 10 (661 mg, 97%) was obtained as a white
6H; ArCH2), 4.29 (d, J=15 Hz, 6H; ArCH2), 4.95 (sb, 6H ; CHN), 5.56
2
(s, 6H ; CHO), 6.85 (s, 6H; ArHcalix), 7.32 (s, 6H; ArHcalix), 7.46 (d, J=
2
À
À
8 Hz, 3H; Py H), 7.92 (t, J=8 Hz, 3H; Py H), 8.04 (d, J=8 Hz, 3H;
À
À
Py H), 8.97 (s, 2H; Pic ), 10.09 ppm (brs, 1H; NH +).
RNH2 (> 1 equiv) was then added to the NMR tube leading to the quan-
titative formation of complex 11ꢀRNH3+,PicÀ as shown by 1HNMR
spectroscopy.
solid. m.p. 1518C (decomp); IR (KBr): n˜ =2961, 1593, 1482 cmÀ1 1H
;
NMR (300 MHz, CDCl3): d=0.83 (s, 27H; tBu), 1.38 (s, 27H; tBu), 2.34
(s, 9H; OCH3), 3.42 (d, J=15 Hz, 6H; ArCH2), 4.57 (s, 6H; CH2Br),
4.58 (d, J=15 Hz, 6H; ArCH2), 5.11 (s, 6H; CH2O), 6.71 (s, 6H;
For R =Et: 1HNMR (CDCl 3, 500 MHz, 263 K) d= À1.28 (t, J=7 H z,
3H; CH3guest), 0.33 (m, 2H; CH3CH2guest), 0.79 (s, 27H; tBu), 1.40 (s,
27H; tBu), 3.32 (s, 9H; OCH3), 3.48 (d, J=15 Hz, 6H; ArCH2), 3.63 (sb,
6H; CH2N), 4.25 (d, J=15 Hz, 6H; ArCH2), 5.31 (s, 6H; CH2O), 6.69 (s,
À
ArHcalix), 7.27 (s, 6H; ArHcalix), 7.41 (d, J=7 Hz, 3H; Py H), 7.80–
7.89 ppm (m, 6H; Py H); 13C NMR (75 MHz, CDCl3): d=29.8, 31.3,
À
À
6H; ArHcalix), 7.31–7.34 (m, 9H; 6 ArHcalix + 3Py H), 7.75 (d, J=7 H z,
31.8, 34.0, 34.2, 34.4, 60.4, 75.0, 120.9, 122.4, 123.9, 128.1, 133.2, 133.7,
138.1, 146.0, 146.3, 151.5, 154.5, 156.1, 158.2 ppm; elemental analysis
calcd (%) for C90H108Br3N3O6·2H2O (1603.59): C 67.41, H7.04, N 2.62;
found: C 67.18, H6.69, N 2.55.
À
À
À
3H; Py H), 7.81 (t, J=7 Hz, 3H; Py H), 8.85 ppm (s, 2H; Pic ).
For R =Pr: 1HNMR (CDCl 3, 300 MHz, 263 K): d=À1.91 (t, J=7 H z,
3H; CH3guest), À0.33 (sb, 2H ; CHCH2guest), 0.78 (s, 27H; tBu), 1.41 (s,
3
27H; tBu), 3.10 (s, 9H; OCH3), 3.44 (d, J=15 Hz, 6H; ArCH2), 3.72 (sb,
6H; CH2N), 4.30 (d, J=15 Hz, 6H; ArCH2), 5.38 (s, 6H; CH2O), 6.75 (s,
Calix[6]tmpa (11):
Method A: X6Me3H3 (1; 508 mg, 0.500 mmol) was added to a mixture of
NaH(60 wt% in oil, 90 mg, 2.3 mmol) and NaI (50 mg, 0.33 mmol) in an-
hydrous THF (80 mL) and anhydrous DMF (80 mL). The solution was
stirred for 30 min and compound 7 (327 mg, 0.750 mmol) was added in
one portion. The reaction mixture was refluxed for 48 h and then the sol-
vent was removed under reduced pressure. The resulting residue was dis-
solved in dichloromethane (100 mL) and washed with water (3”50 mL).
The organic layer was separated and concentrated to dryness. The result-
ing residue was triturated with ethanol (10 mL) and pure calix[6]tmpa
(11) was isolated by centrifugation as a white solid (240 mg, 36%).
À
6H; ArHcalix), 7.30–7.35 (m, 9H; 6 ArHcalix + 3Py H), 7.66 (d, J=7 H z,
+
À
À
3H; Py H), 7.87 (t, J=7 Hz, 3H; Py H), 7.89 (brs, 3H; NH3 guest),
8.87 ppm (s, 2H; PicÀ).
endo complex 11·Na+,ClÀꢀEtOH: A mixture of 11 (5 mg, 3.7 mmol) and
NaCl (9.5 mg, 0.16 mmol) in CHCl3 (0.8 mL) and EtOH(0.3 mL) was
stirred at room temperature for 1 h and then filtered. After evaporation
of the solvents, CDCl3 (0.6 mL) was added and the quantitative forma-
1
tion of complex 11·Na+,ClÀꢀEtOH was shown by HNMR spectroscopy.
1HNMR (CDCl 3, 300 MHz, 293 K): d=À1.56 (t, J=7 Hz, 3H; CH3guest),
0.81 (s, 27H; tBu), 1.35–1.50 (m, 29H; tBu + CH2guest; determined by the
COSY experiment), 2,16 (t, J=6 Hz, 1H; OHguest), 3.13 (s, 9H; OCH3),
3.45 (d, J=15 Hz, 6H; ArCH2), 3.96 (brs, 6H; CH2N), 4.25 (d, J=15 Hz,
6H; ArCH2), 5.38 (s, 6H; CH2O), 6.71 (s, 6H; ArHcalix), 7.34 (s, 6H;
Method B: A suspension of X6Me3H3 (1; 300 mg, 0.295 mmol), compound
7 (142 mg, 0.326 mmol), K2CO3 (124 mg, 0.897 mmol), Cs2CO3 (48 mg,
0.15 mmol), and KI (25 mg, 0.15 mmol) in anhydrous DMF (20 mL) was
stirred vigorously at room temperature for 2 h and then at 908C for 64 h.
After the mixture was allowed to cool back to room temperature, DMF
was removed under vacuum to dryness. CH2Cl2 (100 mL) and HCl (2m,
1.5 mL) were added to the residue and the solution was stirred for 1 h at
room temperature. The organic layer was separated, washed with H2O
(2”50 mL), then stirred with an aqueous NaOHsolution (5 m, 4.5 mL)
for 1 h at room temperature. The organic layer was again separated and
washed three times with H2O (first with 2”50 mL for a few minutes,
then with 100 mL for 1 h) to remove any trace of sodium cation. After
evaporation of CH2Cl2 under reduced pressure, the aqueous suspension
was filtered over celite. The solid phase was washed successively with
H2O (3”50 mL), EtOH(3”50 mL), CH 2Cl2 (3”20 mL), and finally ex-
tracted with CHCl3 (4”20 mL). Evaporation of the CHCl3 fractions
yielded the desired pure product 11 (187 mg, 47%).
À
À
ArHcalix), 7.44 (d, J=6 Hz, 3H; Py H), 7.82–7.96 ppm (m, 6H; Py H);
13C NMR (CDCl3, 75 MHz): d=15.7 (CH3CH2OHin), 29.5, 31.3, 31.8,
34.3, 34.5, 57.6 (CH3CH2OHin), 58.7, 61.4, 75.6, 118.4, 123.7, 124.1, 129.6,
131.8, 132.7, 139.7, 146.6, 147.0, 150.8, 153.6, 158.2, 159.3 ppm.
+
Determination of the relative affinities of the ammonium ions RNH3
(with R=Et, Pr or nBu) toward host 11 through 1H NMR competitive
binding studies: For a typical procedure: EtNH2 (ca. 8 equiv) and PrNH2
(ca. 5 equiv) were successively added into a CDCl3 solution (0.60 mL)
containing 11·H+,PicÀ (prepared as above) at room temperature. A 1H
NMR spectrum recorded at 293 K showed the guest resonances of both
endo complexes 11ꢀEtNH3+,PicÀ and 11ꢀPrNH3+,PicÀ besides the sig-
nals corresponding to the free amines. The integrations of the methyl
group of the free amines and of the included ammonium guests were
+
+
used to calculate the relative affinity defined as [EtNH3 in]/
ACHTRE[UNG PrNH3 in]”
[PrNH2(H+)T]/[EtNH2(H+)T], where indexes “in” and “T” stand for “in-
AHCTREUNG
Method C: NH3 was bubbled for 1 h into a solution of calix[6]arene 10
(170 mg, 0.108 mmol) in THF (100 mL) at room temperature. Na2CO3
(20 mg, 0.19 mmol) was added and the reaction mixture was heated at
608C for 16 h. The solvent was removed under reduced pressure and the
crude solid was dissolved in CH2Cl2 (20 mL), washed with an aqueous
solution of NaOH(1 m, 5 mL) and with water (2”100 mL). After evapo-
ration of CH2Cl2, the suspension was filtered over celite. The solid phase
was washed successively with H2O (2”10 mL), CH2Cl2 (2”5 mL), and fi-
nally extracted with CHCl3 (4”10 mL) to yield pure 11 (90 mg, 62%).
cluded” and “total amount”, respectively (errors estimated Æ10%).
Given the large excess of free amines versus 11·H+, the relative affinities
were calculated considering that the slight difference of pKa between the
different free amines was negligible.
Determination of the relative affinities of the neutral guests L (with L=
(Æ)-PPD, PYD, EtOH, DMF, IMI, EtNH2) toward host 11·Na+ through
1H NMR competitive binding studies: For a typical procedure: EtNH2
(ca. 7 equiv) was added into a CDCl3 solution (0.60 mL) containing
11·Na+ꢀEtOH (prepared as above) at room temperature. A 1HNMR
spectrum recorded at 293 K showed the guest resonances of both endo
complexes 11·Na+ꢀEtNH2 and 11·Na+ꢀEtOH besides the signals corre-
sponding to the free EtOHand EtNH 2. The integrations of the free and
m.p.
>
2608C; IR (KBr): n˜ =2951, 1591 cmÀ1
;
1HNMR (CDCl
,
3
300 MHz): d=0.86 (s, 27H; tBu), 1.37 (s, 27H; tBu), 2.53 (s, 9H; OCH3),
3.43 (d, J=15 Hz, 6H; ArCH2), 3.63 (s, 6H; CH2N), 4.45 (d, J=15 Hz,
6H; ArCH2), 5.65 (s, 6H; CH2O), 6.80 (s, 6H; ArHcalix), 7.08 (d, J=7 H z,
À
À
3H; Py H), 7.26 (s, 6H; ArHcalix), 7.61–7.66 ppm (m, 6H; Py H);
13C NMR (CDCl3, 75 MHz): d=31.0, 31.2, 31.9, 34.2, 34.4, 60.1, 61.4,
74.9, 117.8, 121.9, 124.3, 128.6, 133.5, 134.0, 136.6, 145.8, 146.0, 151.7,
155.9, 157.5, 160.0 ppm; elemental analysis calcd (%) for
included EtOHand EtNH allowed us to calculate the relative affinity
defined as [EtNH2in]/ACHTRE[UNG EtOHin]”[EtOHT]/CAHTRE[UGN EtNH2T], where indexes “in”
and “T” stand for “included” and “total amount”, respectively (errors es-
2
timated Æ10%).
.
XRD structure determination of 11·H+,ClO4À: A prismatic colorless crys-
tal was rapidly fished out of the mother liquor with a cryoloop and
frozen under a cold nitrogen stream (100 K). Diffraction data were re-
corded at the BM-14 beamline (ESRF Synchrotron, Grenoble) and con-
sist of 100 three-degree rotation frames. Distance was set as to get the
highest possible resolution (1.01 Š) compatible with the wavelength in
C90H108N4O6 CH2Cl2·3H2O (1480.82): C 73.81, H7.90, N 3.78; found: C
74.12, H7.69, N 3.59.
11·H+,PicÀ and endo complexes 11ꢀRNH3+,PicÀ: PicH(1 equiv) was
added to a solution of 11 (3 mg, 2 mmol) in CDCl3 (0.6 mL). The corre-
1
sponding HNMR spectrum showed the formation of the monoprotonat-
ed species 11·H+,PicÀ: 1HNMR (CDCl 3, 300 MHz, 263 K): d=0.87 (s,
6400
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6393 – 6402

Calix[6]arene Azacryptand
FULL PAPER
use (0.964 Š). Data reductions and averages were done with the HKL
package of programs.[35] The structure was solved with the SHELXD pro-
gram, part of the SHELX package.[36] Refinements with anisotropic ther-
mal factors were conducted with the SHELXH program. Final statistics
[16] We first prepared 6 according to this previously reported synthesis.
However, the obtained low overall yield (< 20%, 22% in the litera-
ture) as well as long time and hazardous procedures led us to devel-
op a new synthetic strategy.
[17] S.-S. Jew, B.-S. Park, D.-Y. Lim, M. G. Kim, I. K. Chung, J. H. Kim,
C. I. Hong, J.-K. Kim, H.-J. Park, J.-H. Lee, H.-G. Park, Bioorg.
Med. Chem. Lett. 2003, 13, 609–612.
¯
are as following: space group P1 (triclinic) with three independent mole-
cules (Z=6); parameters: a=17.806(1), b=28.698(1), c=31.229(1) Š;
a=67.175(5), b=79.256(4), g=83.793(5)8; final R factor=0.0975 for
21031 observed Fo (criterion 4s(Fo)) and 0.1003 for all data (25549 Fo).
[18] It is noteworthy that 7 decomposed within a few days at room tem-
perature and thus should be kept at low temperature.
À
Associated to the three ligands are eight CHCl3 and three ClO4 ions.
Two additional individual peaks were refined as water molecules. Mw=
5318.01. Total chemical formula: [C90H109N4O6
[19] We have recently reported that the best yields for similar [1+1]
+
,
ClO4À]3, 8·CHCl3,
macrocyclization reactions were obtained with
a 1:6:2 Cs2CO3/
2·H2O. CCDC-298120 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data re-
quest/cif.
K2CO3/calix ratio: S. Le Gac, X. Zeng, O. Reinaud, I. Jabin, J. Org.
Chem. 2005, 70, 1204–1210.
[20] S. Blanchard, M.-N. Rager, A. F. Duprat, O. Reinaud, New J. Chem.
1998, 22, 1143–1146.
[21] The carbon resonances of the included EtOHmolecule were also
determined and attributed thanks to an HMBC spectrum (see the
Experimental Section and the Supporting Information: Figure S26).
[22] A similar NMR pattern was obtained when a solution of EtONa
(21%) in EtOHwas added to a CDCl solution of 11 (see the Sup-
porting Information: Figure S4). Similarly, the addition of a solution
3
Acknowledgements
of EtOK in EtOHto a CDCl solution of 11 led to the formation of
3
Part of this work was supported by the French Ministry of Research and
New Technologies. We thank Martin Walsh (ESRF, Grenoble, France)
for access and advice on the BM14 beam line.
the endo complex 11·K+ꢀEtOH. However, a total conversion was
not obtained even in presence of an excess of EtOK, showing a
weaker binding of K+.
[23] The relative proportion of this minor dissymmetrical species did not
change upon the addition of a large excess of EtOH.
[24] O. SØn›que, M.-N. Rager, M. Giorgi, O. Reinaud, J. Am. Chem. Soc.
2000, 122, 6183–6189.
[1] a) C. D. Gutsche in Calixarenes Revisited, Monographs in Supramo-
lecular Chemistry (Ed.: J. F. Stoddart), The Royal Society of Chem-
istry, Cambridge, 1998; b) Calixarenes in Action (Eds.: L. Mandolini,
R. Ungaro), Imperial College Press, London, 2000; c) C. Redshaw,
Coord. Chem. Rev. 2003, 244, 45–70.
[2] A. Ikeda, S. Shinkai, Chem. Rev. 1997, 97, 1713–1734.
[3] Y. Chen, S. Gong, J. Inclusion Phenom. Macrocyclic Chem. 2003, 45,
165–184.
[4] a) I. Jabin, O. Reinaud, J. Org. Chem. 2003, 68, 3416–3419; b) U.
Darbost, M. Giorgi, O. Reinaud I. Jabin, J. Org. Chem. 2004, 69,
4879–4884; c) X. Zeng, N. Hucher, O. Reinaud, I. Jabin, J. Org.
Chem. 2004, 69, 6886–6889.
[5] a) U. Darbost, X. Zeng, M.-N. Rager, M. Giorgi, I. Jabin, O. Rein-
aud, Eur. J. Inorg. Chem. 2004, 4371–4374; b) U. Darbost, M.-N.
Rager, S. Petit, I. Jabin, O. Reinaud, J. Am. Chem. Soc. 2005, 127,
8517–8525; c) See also: K. Kim, H. J. Lee, J.-I. Choe, Bull. Korean
Chem. Soc. 2005, 26, 645–651.
[25] When the spectrum was recorded at 293 K, the exchange rate be-
tween 11 and 11·H+ was still slower than the NMR time scale, with
only a broadening of the signals from the new species (i.e. 11·H+)
being observed.
[26] Similar NMR spectra of 11·H+ were obtained upon the addition of
either EtCOOHor trifluoroacetic acid or perchloric acid to
a
CDCl3 solution of 11 (see the Supporting Information for 11·H+
,ClO4À: Figure S22).
[27] For a simple tmpa ligand, three successive protonation constants
have been reported: K=2.7, 4.5 and 6.3. Hence, if the calix-tmpa
ligand were to present the same basic properties, it should easily un-
dergo polyprotonation in a strong acidic medium. See: J. C. Mare-
que-Rivas, R. Prabaharan, R. Torres Martin de Rosales, Chem.
Commun. 2004, 76–77.
[28] The proportion of this minor dissymmetrical species did not increase
upon the addition of a large excess of PicH.
[6] E. Garrier, S. Le Gac, I. Jabin, Tetrahedron: Asymmetry 2005, 16,
3767–3771.
[29] a) U. Darbost, M. Giorgi, N. Hucher, I. Jabin, O. Reinaud, Supra-
mol. Chem. 2005, 17, 243–250; b) U. Darbost, X. Zeng, M. Giorgi, I.
Jabin, J. Org. Chem. 2005, 70, 10552–10560.
[30] a) O. SØn›que, Y. Rondelez, L. Le Clainche, C. Inisan, M.-N. Rager,
M. Giorgi, O. Reinaud, Eur. J. Inorg. Chem. 2001, 2597–2604; b) O.
SØn›que, M. Giorgi, O. Reinaud, Supramol. Chem. 2003, 15, 573–
580.
[31] H. Toftlund, S. Ishiguro, Inorg. Chem. 1989, 28, 2236–2238.
[32] See for example: a) S. G. Bott, A. W. Coleman, J. L. Atwood, J. Am.
Chem. Soc. 1986, 108, 1709–1710; b) A. Arduini, E. Ghidini, A. Po-
chini, R. Ungaro, G. D. Andreetti, L. Calestani, F. Ugozzoli, J. Incl.
Phenom. Mol. Recogn. Chem. 1988, 6, 119–134; c) M. Baaden, G.
Wipff, M. R. Yaftian, M. Burgard, D. Matt, J. Chem. Soc. Perkin
Trans. 2 2000, 1315–1321; d) J. Wickens, R. A. W. Dryfe, F. S. Mair,
R. G. Pritchard, R. Hayes, D. W. M. Arrigan, New J. Chem. 2000, 24,
149–154; e) F. Arnaud-Neu, S. Barboso, S. Fanni, M.-J. Schwing-
Weil, V. McKee, M. A. McKerveyl, Ind. Eng. Chem. Res. 2000, 39,
3489–3492; f) L. Baklouti, R. Abidi, P. Thuery, M. Nierlich, Z.
Asfari, J. Vicens, J. Inclusion Phenom. Macrocyclic Chem. 2001, 40,
323–325.
[7] G. Izzet, B. Douziech, T. PrangØ, A. Tomas, I. Jabin, Y. Le Mest, O.
Reinaud, Proc. Natl. Acad. Sci. USA 2005, 102, 6831–6836.
[8] K. D. Karlin, S. Kaderli, A. D. Zuberbuhler, Acc. Chem. Res. 1997,
30, 139–147.
[9] M. Costas, M. P. Mehn, M. P. Jensen, L. Que, Jr., Chem. Rev. 2004,
104, 939–986.
[10] U. Lüning, F. Lçffler, J. Eggert in Calixarenes (Eds.: Z. Asfari,
et al.), Kluwer, 2001, pp. 71–88.
[11] For the preparation of 1, see: R. G. Janssen, W. Verboom, D. N.
Reinhoudt, A. Casnati, M. Freriks, A. Pochini, F. Uggozoli, R.
Ungaro, P. M. Nieto, M. Carramolino, F. Cuevas, P. Prados, J. de
Mendoza, Synthesis 1993, 380–385.
[12] a) K. Araki, K. Akao, H. Otsuka, K. Nakashima, F. Inokuchi, S.
Shinkai, Chem. Lett. 1994, 1251–1254; b) H. Otsuka, K. Araki, H.
Matsumoto, T. Harada, S. Shinkai J. Org. Chem. 1995, 60, 4862–
4867; c) Y.-Y. Chen, J.-S. Li, Z.-L. Zhong, X.-R. Lu, Tetrahedron
1998, 54, 15183–15188; d) Y. Zhang, H. Yuan, Z. Huang, J. Zhou,
Y. Kawanishi, J. Schatz, G. Maas, Tetrahedron 2001, 57, 4161–4165.
[13] R. G. Janssen, W. Verboom, J. P. M. van Duynhoven, E. J. J. van Vel-
zen, D. N. Reinhoudt, Tetrahedron Lett. 1994, 35, 6555–6558.
[14] J.-C. Rodriguez-Ubis, B. Alpha, D. Plancherel, J.-M. Lehn, Helv.
Chim. Acta 1984, 67, 2264–2269.
[33] For the melting point and the elemental analysis of 3, see: W.
Mathes, W. Sauermilch, T. Klein, Chem. Ber. 1953, 86, 584–587.
[34] D. I. C. Scopes, N. F. Hayes, D. E. Bays, D. Belton, J. Brain, D. S.
Brown, D. B. Judd, A. B. McElroy, C. A. Meerholz, A. Naylor, A. G.
[15] Z. He, P. J. Chaimungkalanont, D. C. Craig, S. B. Colbran, J. Chem.
Soc. Dalton Trans. 2000, 1419–1429.
Chem. Eur. J. 2006, 12, 6393 – 6402
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
6401

I. Jabin, O. Reinaud et al.
Hayes, M. J. Sheehan, P. J. Birch, M. B. Tyers, J. Med. Chem. 1992,
35, 490–501.
[36] G. Sheldrick, SHEXL97, a program for the refinement of crystal
structures, University of Gçttingen, Germany, 1997.
[35] Z. Otwinowski, W. Minor, Processing of X-Ray Diffraction Data
Collected in Oscillation Mode, Methods in Enzymology 1997, 276,
307–325.
Received: February 28, 2006
Published online: July 5, 2006
6402
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6393 – 6402