Macrocyclic hexaureas: Synthesis, conformation, and anion binding

Article abstract of DOI:10.1002/chem.200802573

Five macrocylic compounds XXXXXX, XXDXXD, XDXDXD,XDDXDD, and DDDDDD with 48-membered rings, in which six xanthene and/or diphenyl ether fragments are linked through six urea (-NH-C(O)-NH-) groups, have been synthesized. In the cyclization step, a linear d

Full text of DOI:10.1002/chem.200802573

FULL PAPER
DOI: 10.1002/chem.200802573
Macrocyclic Hexaureas: Synthesis, Conformation, and Anion Binding
Denys Meshcheryakov,[a] Volker Bçhmer,*[a] Michael Bolte,[b]
Vꢀronique Hubscher-Bruder,[c] and FranÅoise Arnaud-Neu[c]
Abstract: Five macrocylic compounds
XXXXXX, XXDXXD, XDXDXD,
XDDXDD, and DDDDDD with 48-
membered rings, in which six xanthene
and/or diphenyl ether fragments are
linked through six urea (-NH-C(O)-
NH-) groups, have been synthesized. In
the cyclization step, a linear diamine
was allowed to react with the appropri-
ate diisocyanate by using a [5+1] (i.e.,
“XDXDX+D” for XDXDXD), [4+2]
(DDDDDD), or [3+3] (XDDXDD)
procedure. Compounds XXXXXX and
XXDXXD were prepared from two
molecules of the dimeric amine XX
and two molecules of the respective
monomeric diisocyanate (X or D) in a
[2+1+2+1] (or 2ꢀ[2+1]) reaction.
The (nonoptimized) yields in the cycli-
zation step ranged from 45 to 80%.
The linear precursor diamines or diiso-
cyanates were obtained by analogous
condensation reactions by using partial
protection with a tert-butoxycarbonyl
group. All the macrocyclic compounds
and synthetic intermediates were char-
acterized by 1H NMR and mass spec-
tra. Three different crystal structures
were obtained for XDDXDD, which
show the molecule in a more or less
strongly folded conformation deter-
mined by intramolecular hydrogen
bonding. The interaction of the hexaur-
eas with selected anions was studied by
1H NMR spectroscopy and UV absorp-
tion spectrophotometry.
Keywords: anion binding · host–
guest systems
·
macrocycles
·
template synthesis · urea
Introduction
Initially guided by the idea that the cyclic arrangement of
three urea groups as hydrogen-bond donors separated by a
suitably sized spacer should lead to a selective ligand for the
nitrate anion,[6] we started with the synthesis of macrocyclic
triurea compounds. According to molecular modeling stud-
ies, the 4,5-substituted xanthene skeleton[7] should be an ap-
propriate rigid spacer (“X”). Furthermore, the spacer could
be combined with the respective diphenyl ether fragment[8]
(“D”) to adjust the appropriate balance between rigidity
(i.e., preorganization of the hydrogen-bond donors) and
flexibility of the ligand. We thus synthesized and studied all
possible cyclic tri- and tetraurea[9,10] derivatives of building
blocks X and D. During these syntheses, we did not only ob-
serve the formation of a “monomeric”[11] and a dimeric
cyclic urea,[12] but we also isolated cyclic hexamers. In the
case of XXDXXD, the formation of the hexamer was obvi-
ously favored by a template effect exerted by two chloride
ions.[13] To obtain a deeper understanding of the factors re-
sponsible for the easy formation of such large cyclic hexam-
ers, we decided to prepare and study further examples. In
addition to macrocycles XXXXXX and DDDDDD, eleven
cyclic combinations of the X and D units are possible. We
selected the most symmetric combinations, namely
XDDXDD, XDXDXD, and XXDXXD, for synthesis.
Anion-receptor chemistry has continued to attract great in-
terest over recent years.[1] Numerous synthetic hosts contain-
ing amide,[2] pyrrole,[3] or urea[4] groups incorporated in in-
creasingly complicated supramolecular skeletons target the
efficiency of natural receptors.[5]
[a] Dr. D. Meshcheryakov, Dr. V. Bçhmer
Fachbereich Chemie, Pharmazie und Geowissenschaften
Abteilung Lehramt Chemie
Johannes Gutenberg-Universitꢁt Mainz
Duesbergweg 10–14, 55099 Mainz (Germany)
Fax : (+49)6131-3925419
E-mail: vboehmer@mail.uni-mainz.de
[b] Dr. M. Bolte
Fachbereich Chemie und Pharmazeutische Wissenschaften
Institut fꢂr Anorganische Chemie
Johann Wolfgang Goethe-Universitꢁt Frankfurt
Max-von-Laue-Strasse 7, 60438 Frankfurt/Main (Germany)
[c] Dr. V. Hubscher-Bruder, Dr. F. Arnaud-Neu
IPHC-DSA, ULP, CNRS
25 rue Becquerel, 67087 Strasbourg (France)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802573.
Chem. Eur. J. 2009, 15, 4811 – 4821
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4811

Results and Discussion
quantity of triphosgene in the presence of trialkylamines.
However, this reaction, which is only suitable for hexamers
that consist of the same unit, was not effective even in this
case.
On the other hand, the [2+1] reaction between the di-
meric diamine 5 (XX) connected through a urea group and
diisocyanates 1 or 2, initially planned for the synthesis of
cyclic trimers, produced hexamers in surprisingly good yield.
The conditions were finally found under which even the
cyclic hexamer was obtained as the main product of the re-
action.
Syntheses: There are numerous ways to prepare a cyclic
hexamer, which neither can be exhaustively discussed nor
experimentally checked or compared. Scheme 1 summarizes
the last, ring-closing condensation step for all the five exam-
ples that were realized (i.e., 12–16) and the synthesis of es-
sential intermediates.
Considering the number and size of the molecules in the
final reaction step, we found [3+3], [4+2], and [5+1] reac-
tions for 15, 16, and 14, respectively. These strategies could
also be used for 12 and 13, but a 2ꢀ[2+1] reaction (e.g.,
XX+D+XX+D) appeared to be more convenient for
both. However, a systematic study was neither done nor in-
tended, and the synthetic pathway realized for a certain
compound was (partly) also determined by economic rea-
sons (e.g., the use of the same fragment for the preparation
of different products).
Hexamer 12 (XXXXXX) was the first example of such an
unexpected product obtained by the reaction of diamine 5
(XX) and diisocyanate 1 (X). With the intention of prepar-
ing the cyclic trimer XXX, we added a solution of the dia-
mine in dichloromethane dropwise to a solution of the diiso-
cyanate in dichloromethane. The crude product was triturat-
ed with hexane and a white solid was collected by filtration.
The isolated product was practically insoluble in deuterated
Traces of six-membered macrocycles are usually detected
by mass spectrometry of the reaction mixture when a “mon-
omeric” diamine is allowed to react with an equimolar
1
dimethyl sulfoxide ([D6]DMSO) and the obtained H NMR
spectra in CDCl3 were broad and unclear. ESI mass spec-
Scheme 1. Syntheses of cyclic hexaureas and their precursors.
4812
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4811 – 4821

Anions as Templates in the Formation of Macrocycles
FULL PAPER
trometry revealed the nature of the compound: the solid
was the cyclic product with six xanthene units. The pure
compound 12 was isolated in a yield of as much as 49%
under these conditions, while a significant amount remained
in solution. The formation of the cyclic trimer, however, was
not observed at all.
When diamine 5 (XX) was treated with diisocyanate 2
(D) under the same conditions, the corresponding hexamer
13 (XXDXXD) was formed along with the trimer XXD, but
the highest isolated yield of the hexaurea was 20%. In this
case, the cyclic trimer was the main product of the reaction.
When the reaction was carried with the more-polar acetoni-
trile as the solvent instead of dichloromethane, the hexamer
was not formed at all. This outcome indicates the impor-
tance of molecular association through hydrogen bonds for
the formation of cyclic hexamers from four independent
molecules, that is, “2ꢀ[2+1]”
tion, the influence of tetrabutylammonium (TBA) chloride
on the reaction between diamine 5 (XX) and D-isocyanate 2
revealed a remarkable example in which two (!) chloride
ions as a template control the formation of the hexamer 13
(XXDXXD). It was found that the molar ratio of trimer/
hexamer changes drastically from 5:1 to 1:5 in the presence
of TBA chloride. A significant but less pronounced increase
in the ratio of trimer/hexamer to 1:3 was observed in the
presence of the bromide ion.[13]
It has been already mentioned that the “ACTHUNRTGNEUNG[XX+X]” reac-
tion does not yield the expected trimer but hexamer 12
(XXXXXX) as the only cyclic product. Further studies re-
vealed that anions have a strong influence on the product of
the reaction as well. Particularly, the presence of TBA ace-
tate in the reaction mixture changes the outcome complete-
ly, because the trimer XXX is the only product formed in a
significant quantity.[9]
The formation of hexamers 15 (XDDXDD) and 16
(DDDDDD) was also observed in two other examples of in-
tended [2+1] reactions, that is, DD+X and DD+D, in
which the cyclic hexamers could be obtained in up to 29 and
20% yield, respectively. However, the isolation and purifica-
tion of these compounds was difficult because of the close
similarity of their physical properties (i.e., solubility and
RF values) to other reaction products, such as the corre-
sponding cyclic trimers.
These two striking examples of the templated formation
of cyclic oligourea compounds gives rise to questions about
the roles that anion-induced and structure-dependent preor-
ganization factors play in the outcome of the reaction. With
this idea in mind, a series of analogous [2+1] cyclizations
(i.e., XX+X, XX+D, DD +X, and DD+D) were per-
formed with and without the addition of acetate or chloride
ions to the reaction mixture. The general conditions were:
dichloromethane as a solvent with stirring for 18 h at room
temperature. The cyclic products were isolated by column
chromatography after the reaction, and the composition of
the product mixtures were analyzed by NMR spectroscopy.
The compositions of the trimer and hexamer are compared
in Table 1.
Hexamer 15 (XDDXDD) was finally prepared by using
the “ACHTUNGTRENNUNG[3+3]” method, which produces a smaller number of
intermediates and consequently of by-products on its way to
the target compound. The trimeric diamine 8 (DXD) was al-
lowed to react with diisocyanate 9 (DXD, obtained from the
same diamine). After trituration of the crude product with
acetonitrile, the hexamer was isolated in 45% yield.
Table 1. Influence of selected salts on the [2+1] reaction.
For the preparation of hexamer 14 (XDXDXD), the [3+
3] and [5+1] strategies were checked, in which the latter
gave better results in the cyclization step. A yield of 89%
for the macrocycle was obtained by reaction of the linear di-
amino pentamer 11 (XDXDX) with D-diisocyanate 2. The
necessary pentameric diamine was prepared by a stepwise
synthesis, in which trimeric diamine 8 (DXD) was allowed
to react with two molecules of the isocyanate 3 (X-based)
followed by subsequent deprotection of the resulting linear
tetraurea XDXDX.
Salt
Compound
Yield [%]
5+1
5+2
6+1
6+2
(XX+X) (XX+D) (DD+X) (DD+D)
no anion
trimer
hexamer
–
69
–
75
15
8
46
29
60
20
63
37
51
4
TBA chloride trimer
hexamer
TBA acetate trimer
hexamer
86
[a]
55
82
40
–
ꢀ100
71
–
[b]
–
–
[a] The pentameric diamine DDXDD was detected. [b] The formation of
up to 27% of the pentameric oligourea XXDXX was observed.
The isolation of the hexamer 16 (DDDDDD) was espe-
cially difficult. Preorganization seemed to play a less signifi-
cant role in the assembly of the hexamers because of the
flexibility of the diphenyl ether skeleton. Although the
hexamer was formed in the reaction of diamine 6 (DD) with
D-diisocyanate 2, the chromatographic separation of the
hexamer and trimer was nearly impossible. The hexaurea
was finally isolated as the product of the [4+2] reaction of
diamine 10 (DDDD) with diisocyanate 7 (DD).
In each case, the anions have a rather strong influence on
the result, but the effect was most pronounced for the reac-
tions of “XX+X+acetate” and “XX+D+chloride”. Not
only the total yield of the cyclic products varied, but also
the ratio of both oligomers changed drastically. A practically
quantitative conversion of the starting compounds into
cyclic products was observed in some cases, whereas a mac-
rocycle was obtained only in 60–70% yield in others and the
rest remained uncyclized. This outcome may be explained
by different individual reaction rates in the different mix-
tures. The isocyanates are relatively stable as solids, but in
solution they are destroyed within a few hours by moisture
Anions as templates: The oligoureas interact in solution
through hydrogen bonds with other ureas and anions. In
previous studies, we have demonstrated the complexation of
anions by cyclic tri- and tetraurea compounds.[9,10] In addi-
Chem. Eur. J. 2009, 15, 4811 – 4821
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4813

V. Bçhmer et al.
introduced most likely by the extremely hygroscopic TBA
salts. To achieve higher yields of the cyclic compounds, the
corresponding [2+1] and/or 2ꢀ[2+1] reactions must be
able to compete with the rate of isocyanate hydrolysis.
In general, the formation of more rigid skeletons was
more strongly influenced by the presence of the anions.
bonyl oxygen atom of the neighboring urea groups that con-
nect X and D units as acceptors. The ethanol molecules re-
place the acetone of the a structure as a hydrogen-bond ac-
ceptor, while simultaneously donating a hydrogen bond to a
carbonyl oxygen atom of a xanthene-attached urea. The
ethyl acetate molecules are not involved in hydrogen bond-
ing.
Crystal-structure analysis: Three different single-crystal X-
ray structures were obtained for the hexamer XDDXDD.
All the samples were prepared by the slow evaporation of a
solution of the hexamer in a multicomponent mixture of sol-
vents (see the Experimental Section). The conformations of
the single molecules are shown and compared in Figure 1
In the third case (g), the molecular conformation is more
complicated as it no longer possesses an inversion center or
any other symmetry element. Intramolecular bifurcated hy-
drogen bonds exist between one urea group that connects
À
D X units (acceptor) and the adjacent urea group that con-
À
nects D D units (donor) and also between two opposite
À
urea compounds that connect D X units. One ethanol mole-
cule is bound through bifurcated hydrogen bonds by a urea
group that connects X and D units, while an ethyl acetate
molecule is bound to two adjacent urea groups that connect
À
X D units and additionally forms hydrogen bonds with a
À
urea group that connects D D units as an acceptor. Two
cyclic hexaurea molecules form centrosymmetric dimers
through a hydrogen-bonding system that also involves two
acetonitrile, two water, and two ethanol molecules. The
complicated system is illustrated in Figure 2, and the hydro-
gen-bonded dimer is shown in Figure 3.
Further information about the conformation of the three
structures 15a, 15b, and 15g is collected in Table 2, in
which the interplanar angles for the aromatic rings are
listed. At first glance (Figure 1) the two centrosymmetric
structures/molecules 15a and 15b seem to have a similar
conformation and although they even have the same hydro-
gen-bonding pattern (Figure 2), there are rather strong dif-
ferences between these interplanar angles, mainly for the
angles around the urea groups.
Figure 1. The conformation of 15 found in three different single crystals
(a, b, g). Solvent molecules, non-urea hydrogen atoms, and alkyl groups
have been omitted for clarity. Xanthene: red; diphenyl ether: blue.
In the first case (a), the crystal contains eight molecules
of acetone per hexamer (Figure 1). Two of these acetone
molecules are strongly bound by two urea groups that con-
nect X and D units (a typical six-membered cyclic binding
motif[14]), whereas two others are weakly coordinated
through a single hydrogen bond by a xanthene NH group.
The remaining four acetone molecules fill gaps in the crystal
lattice. The folded conforma-
As in cyclic tri- and tetraurea molecules, no (direct) inter-
molecular hydrogen bonds between the hexaurea molecules
were observed for 15a and 15b, although the flexibility of
tion of the hexamer molecule
with an inversion centre is de-
termined by a bifurcated hydro-
gen bond between the urea
group that connects two
D
units and the carbonyl oxygen
atom of the neighboring urea
group (i.e., connecting X and D
units). The system of hydrogen
bonds is illustrated in Figure 2.
The second structure (b) con-
tains two molecules of ethyl
acetate and two molecules of
ethanol per hexamer. The mo-
lecular conformation of b is
very similar to that of a, again
determined by the bifurcated
hydrogen bonds between urea
groups that connect the D units
Figure 2. Illustration of the system of hydrogen bonds present in the crystals of cyclic hexaurea 15 and diamine
as hydrogen donors and the car- 17.
4814
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4811 – 4821

Anions as Templates in the Formation of Macrocycles
FULL PAPER
finding is especially surprising because single crystals were
grown from a rather polar solvent mixture (acetone, di-
chloromethane, water). Some of the NH groups form hydro-
gen bonds with acetone, dichloromethane, and eventually
water molecules, but no direct intermolecular interaction
was found between separate molecules of the diamine. Two
molecules of 17 form a hydrogen-bonded “cluster” that in-
cludes eight water molecules. Double layers of 17, which
also contain acetone molecules that extend in the ab plane,
are separated by layers of dichloromethane (Figure 5b).
NMR spectroscopic studies: Similar to the cyclic tetraurea
compounds, there were intrinsic restrictions in the NMR
spectroscopic studies of the cyclic hexamers. The most rigid
hexamers XXXXXX and XXDXXD are insoluble in
DMSO, which is most frequent-
Figure 3. The network of hydrogen bonds formed with the participation
of solvent molecules (ethanol, water, and acetonitrile) that connect two
molecules of hexamer 15 in g. Non-urea hydrogen atoms and xanthene
alkyl groups have been removed for clarity.
ly used as a polar solvent in
Table 2. Interplanar angles [8] found in the X-ray structures of 15 (XDDXDD) for aromatic rings within xan-
thene and diphenyl ether units (first part of the table) and the angles [8] between the averaged planes of the
urea groups (U) and the adjacent phenyl rings of units X and D (second part).[a]
NMR spectroscopy. In other
cases, it was impossible to
obtain a sharp spectrum of a
hexamer alone or in the pres-
ence of salt, and other solvents
had to be tried. Thus, a general
comparison of 12–16 by using a
single solvent was impossible.
Compound 12 (XXXXXX)
Angle
15a
15b
15g
17
Ph1/Ph2
À3.3 (X)
À27.3 (X)
+70.0 (D)
+83.6 (D)
+27.3 (X)
À70.0 (D)
À83.6 (D)
À11.1 (X)
+59.1 (D)
À58.8 (D)
+22.4 (X)
À83.7 (D)
+72.8 (D)
À19.2
+7.9
À6.1
+6.7
+5.7
–
Ph3–Ph4
Ph5–Ph6
Ph7–Ph8
Ph9–Ph10
Ph11–Ph12
+71.6 (D)
+80.3 (D)
+3.3 (X)
À71.6 (D)
+80.3 (D)
shows
a
very well-resolved
Ph2-U-Ph3
Ph4-U-Ph5
Ph6-U-Ph7
Ph8-U-Ph9
Ph10-U-Ph11
Ph12-U-Ph1
+30.7/+9.1 (X-U-D)
+38.1/+28.1 (D-U-D)
À6.3/À51.0 (D-U-X)
À30.7/À9.1 (X-U-D)
À38.1/À28.1 (D-U-D)
+6.3/+51.0 (D-U-X)
+26.2/À14.1 (X-U-D)
+30.0/À27.8 (D-U-D)
+25.6/+83.6 (D-U-X)
À26.2/+14.1 (X-U-D)
À30.0/+27.8 (D-U-D)
À25.6/À83.6 (D-U-X)
À12.8/À14.7 (X-U-D)
À141.3/À47.2 (D-U-D)
+43.2/À29.0 (D-U-X)
+41.7/À65.1 (X-U-D)
À9.4/À7.2 (D-U-D)
+16.9/+29.5
À133.3/+162.8
À29.6/+40.9
À79.9/À139.8
–
1H NMR spectrum in [D8]THF
and [D5]pyridine, which is in
full agreement with the pro-
posed structure. However, the
clear signal pattern was entirely
unexpected: 12 m-coupled dou-
blets for the aromatic protons
(2 protons each), six urea NH
À35.9/À15.3 (D-U-X)
–
[a] The corresponding values for the linear diamine 17 (XXXXX) are also included.
the macrocyclic skeleton would surely allow this. And even
the hydrogen-bonded, solvent-bridged dimeric structural el-
ement found for 15g (compare with Figure 3) shows no fur-
ther interaction through hydrogen bonds in the crystal lat-
tice. Intramolecular hydrogen bonding is obviously more fa-
vorable for these macrocyclic oligourea compounds; as an
example, the packing of 15a is shown in Figure 4.
Unfortunately, single crystals suitable for X-ray diffraction
studies were not obtained for the “rigid” hexamer 12
(XXXXXX). However, we succeeded in growing single crys-
tals for its immediate precursor, that is, the linear diamine
that consists of five xanthene units. This compound was iso-
lated as the product of an incomplete cyclization during the
“2ꢀ[2+1]” synthesis of cyclic hexamer 12.
The molecule assumes a densely folded conformation
with strongly distorted urea groups determined again by
strong intramolecular hydrogen bonds (Figure 5a). The
linear chain is contorted and forms a compact “ball” in
which most of the urea and amine hydrogen atoms are in-
wardly oriented, thus participating in the network of hydro-
gen bonds, whereas the alkyl groups (i.e., tert-butyl, methyl)
of the xanthene units form the outer surface (Figure 6). This
peaks (2 protons each), six tert-butyl singlets (18 protons
each), and six CH3 singlets (6 protons each) could be clearly
distinguished (Figure 7). All the peaks are very sharp and
the signals of the same intensity have almost the same
height, thus indicating a very stable conformation with a
twofold symmetry of the molecule.
Whereas the signals for all the aromatic protons and five
signals for the NH group appear in the range d=6.6–
8.9 ppm, one NH signal is strongly shifted downfield (d=
11.7 ppm). Tentatively, this signal may be interpreted as a
strong, bifurcated hydrogen bond between two opposite
urea functions, thus “fixing” the molecule in an “eight” or
propeller-shaped conformation. The spectrum remained
practically unchanged when the sample was heated to 558C.
A similar NMR spectrum with [D5]pyridine as the solvent
showed no essential changes up to 1058C.
The addition of TBA chloride, acetate, or dihydrogen
phosphate to a solution of 12 in THF led to very fast and
complete precipitation of the hexamer (the same was true
for pyridine as the solvent). The addition of other salts (bro-
mide, iodide, nitrate, hydrogen sulfate) had no effect on the
spectrum. Thus, there must be a system of intramolecular
Chem. Eur. J. 2009, 15, 4811 – 4821
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4815

V. Bçhmer et al.
Figure 4. Packing of hexamer 15 in a seen along the bisection of the
angle between the c and b axes.
hydrogen bonds that is stable in THF but destroyed upon
the addition of salts or in the more polar DMSO.
The conformationally more flexible 13 (XXDXXD)
showed a different behavior. The pure compound is either
insoluble ([D6]DMSO) or has a broad and unclear spectrum
([D8]THF, [D5]pyridine, CDCl3). However, upon the addi-
tion of chloride ions a very characteristic spectrum was
found (Figure 8). The strong complexation of chloride and
bromide ions with 13 has already been reported.[13] Stable
species were also formed with dihydrogen phosphate and
hydrogen sulfate ions.
Figure 5. a) Molecular conformation of 17 (XXXXX diamine); the tert-
butyl and methyl groups from three xanthene units have been omitted
for clarity. b) Packing of 17 seen along the a axis.
In contrast with chloride ions, a sharp spectrum was ob-
2À
tained with H2PO4 ions in a 1:1 ratio, thus suggesting the
1
formation of a 1:1 complex. The H NMR spectrum has five
signals for urea protons at d=11.05, 10.55, 9.95, 9.15 (two
protons), and 8.20 ppm. Four meta-coupled doublets at d=
8.35, 8.20. 7.00, and 6.85 for the xanthene units are clearly
seen, whereas the remaining signals appear close to one sol-
vent peak (near d=7.2 ppm). From four doublets and four
triplets with ortho coupling for the diphenyl ether groups,
three doublets at d=8.40. 8.00, and 5.80 ppm and a triplet at
d=6.5 ppm are separated, while the remaining signals over-
lap at around d=6.8 and 6.9 ppm (Figure 8). Four singlets
for the methyl and four singlets for the tert-butyl groups
Figure 6. Pair of molecules of 17 (XXXXX diamine) in the crystal. Eight
water molecules fill the cavity formed by two pentamer molecules, and
the lypophilic alkyl groups form the outer surface.
4816
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4811 – 4821

Anions as Templates in the Formation of Macrocycles
FULL PAPER
Figure 7. 1H NMR spectrum of the xanthene-based cyclic hexamer 12 ([D8]THF, 258C). Signals for the aromatic protons are marked with X, the urea
protons with U, and the methyl groups with M. The signals for the six tBu groups (highfield) are not marked.
not achieved at a 1:5 ratio (hex-
amers/salt), in which both sets
are observed in a much higher
intensity for the second set.
Two sets of signals are ob-
À
served in pyridine for HSO4
ions. In contrast to the H2PO4
2À
ions, the more simple “four-
fold” spectrum strongly prevails
over the more complicated
“twofold” spectrum, even if
only 0.5 equivalents of the salt
are added. The “fourfold” spec-
trum seems to correspond to a
1:1 complex, but it is difficult to
judge the composition of anoth-
er complex that corresponds to
the ”twofold“ species because
of the weakness of the signals.
Upon further addition of
À
HSO4 ions, the ”twofold“ set
of signals weakens further, but
does not completely disappear
even when ten equivalents of
the salt are added. Because the
Figure 8. Top: a section of the 1H NMR spectrum of the complex formed by hexamer 13 (XXDXXD) with
two molecules of TBA chloride in [D5]pyridine; the signals correspond to dynamic D2 symmetry in solution.
Bottom: a section of the 1H NMR spectrum of the 1:1 complex of TBA dihydrogen phosphate/hexamer 13 in
[D5]pyridine; the spectrum corresponds to dynamic C2 (or CS) symmetry in solution (one possibility of a C2
axis or s-plane has been indicated).
”twofold“ set weakens upon the
addition of the salt, it could be
either a 2:1 or 1:1 (ligand/ion)
À
complex with HSO4 ions.
The D3d symmetrical hexa-
appear in the aliphatic region. This pattern agrees with a
conformation with a single C2 axis or a plane of symmetry
that crosses the urea groups between X units (or the oxygen
atoms of the D units).
mer 14 (XDXDXD) shows a far less-expressed response to
the addition of various salts, although it is soluble in DMSO
without limitations. The hexamer itself has a broad and un-
clear NMR spectrum. The interaction with anions could be
usually detected in the spectrum only by general changes in
the baseline shape. Acetate and dihydrogen phosphate ions
seem to have the strongest influence on the spectra. In both
cases, weak signals of a species with twofold symmetry
2À
When the addition of the H2PO4 ions is continued, an-
other set of signals starts to arise. This new set contains
fewer signals and obviously is similar to that with two chlor-
ides, although the signals are not so sharp. Saturation is still
Chem. Eur. J. 2009, 15, 4811 – 4821
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4817

V. Bçhmer et al.
appear, but their quantitative evaluation was not possible.
TBA chloride and hydrogen sulfate cause minor spectral
changes and nitrate and bromide have no influence at all.
The NMR spectroscopic studies brought only limited re-
sults in the case of 15 (XDDXDD) as well. The conforma-
tions of the molecule are neither flexible nor stable enough
in solution to result in a sharp NMR spectrum, which was
also found in the presence of salts. Thus, only cautious con-
clusions can be made. In contrast to XDXDXD, hexamer 15
than those observed with the tetrameric and trimeric ana-
logues,[9,10] especially in the presence of the spherical ClÀ
and BrÀ ions. With these anions, the changes were indeed
not significant enough to allow any interpretation, with the
exception of ligand 14 (XDXDXD), for which the 2:1 com-
plex could be found. The complexation is stronger with the
ClÀ than with BrÀ ion (Dlogb=1.8). In contrast, different
stoichiometries, that is, 1:1 and 1:2, were found for the more
rigid ligand 13.
2À
seemed to be most responsive to the addition of H2PO4
Remarkably, no spectral variation was observed at all
with the most flexible hexamer 16 and all the anions stud-
ions (small peaks appeared beside the broad “hills”), fol-
lowed by chloride and acetate ions (general changes in the
baseline shape).
À
ied, even with the oxoanions AcOÀ and H2PO4 , which led
to important spectral changes with the other ligands. In
these cases, complexes of different stoichiometries could be
found, such as bisanion (1:2) complexes of similar stability
with the more rigid ligand 12. More stable complexes of the
same stoichiometry were found with ligand 15 and an addi-
tional 1:1 acetate complex. The formation of bis-ligand com-
plexes was shown with ligand 14 and the anions studied.
Again, the acetate ion formed the most stable complex.
The most flexible hexamer 16 (DDDDDD) has a clear
spectrum in [D6]DMSO, with four signals for aromatic pro-
tons and a singlet for the urea protons in accordance with
the expected D6h symmetry of this compound. The addition
of TBA chloride, TBA acetate, and TBA dihydrogen phos-
phate affected the spectrum, whereas other salts had either
marginal or no influence. In the case of the chloride or
2À
H2PO4 ions, the spectrum is absolutely broad. Only the in-
teraction of acetate ions led to an interpretable spectrum, in
which the urea singlet, in addition to its broadening, was
shifted downfield by approximately Dd=0.8 ppm, whereas
the signals for the aromatic protons were shifted slightly up-
field (Dd=0.1 ppm).
Conclusions and Outlook
Inspired by the results obtained with the first example
XXDXXD, we extended our studies of cyclic oligoureas
composed of xanthene and diphenyl ether units from trimers
and tetramers to a series of five hexamers. Among 13 possi-
ble combinations of X and D units, we chose the most sym-
metrical structures with a Cn axis “perpendicular” to an ide-
alized, planar, macrocyclic ring. The reaction between a dia-
mine and diisocyanide group according either to a [5+1],
[4+2], [3+3], or 2ꢀ[2+1] strategy was used in the cycliza-
tion step. In all cases, an influence of the anions on the
trimer/hexamer ratio of the products was found, but the ace-
tate-templated formation of the trimer XXX and the chlo-
ride-templated fomation of the hexamer XXDXXD was es-
pecially effective.
The interaction with various anions through hydrogen
bonds has been established for all the hexaurea compounds.
In general, the most stable complexes were formed by the
most rigid hexamers XXXXXX and XXDXXD. Results
similar to the complexation of two chloride anions by
XXDXXD were not found for the other hexamers. Three
different crystal structures for XDDXDD suggest that at
least this compound prefers folded conformations with intra-
molecular hydrogen bonding over hydrogen-bonded net-
works formed by intermolecular hydrogen bonds.
UV spectrophotometric studies: UV absorption spectropho-
tometric analysis was also used to study the interactions be-
tween the four hexamers 12 and 14–16 and chloride, bro-
mide, acetate, and dihydrogen phosphate anions. The experi-
ments were carried out under the same conditions as for 13,
that is, in
Table 3).
a
mixture of THF/acetonitrile[13] (1:3, v/v;
In general, the spectral variations induced by the addition
of the anions to solutions of the ligand were much weaker
[a]
Table 3. Overall stability constants (logbÆsnÀ1
)
of the complexes
formed between the hexaurea ligands and anions in THF/acetonitrile
(1:3, v/v).
Anion
ClÀ
Ratio
ligand/
anion
12
13[13]
14
15
16
[b]
[b]
[b]
1:2
1:1
2:1
–
–
11.4
6.16
–
–
–
11.2Æ0.2
9.4Æ0.3
BrÀ
1:2
2:1
7.7
–
–
[b]
[b]
[b]
[b]
Studies are in progress to synthesize even larger cyclic
oligomers based on xanthene and diphenyl ether units.
AcOÀ
1:2
1:1
2:1
8.8Æ0.3
–
10.4Æ0.9
6.0Æ0.7
[c]
11.5Æ0.1
10.2Æ0.3
À
[c]
[b]
H2PO4
1:2
2:1
8.7Æ0.4
–
9.9Æ0.2
–
Experimental Section
[a] Mean values of at least three independent experiments; the precision
corresponds to the standard deviation on these mean values. [b] Spectral
changes were too small. [c] Not determined.
All the solvents were of analytical quality (p.a.) and were used without
additional purification. Deuterated solvents for the NMR spectroscopic
analysis were purchased from Deutero GmbH. The 1H NMR spectra
4818
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4811 – 4821

Anions as Templates in the Formation of Macrocycles
FULL PAPER
were recorded on a Bruker Avance DRX 400 spectrometer at 400 MHz
and 258C using the solvent signals as internal references. Mass spectra
were recorded on the Finnigan MAT 8230 instrument. The melting
points were not corrected.
Diamine 10 (DDDD): A solution of diamine 6 (158 mg, 0.37 mmol) in di-
chloromethane (10 mL) was added to a solution of isocyanate 4 (242 mg,
0.75 mmol) in dichloromethane (10 mL) with stirring under nitrogen.
Acetonitrile (50 mL) was added and the solvent was removed under re-
duced pressure after 18 h of stirring. The crude product was dissolved in
dichloromethane (10 mL), trifluoroacetic acid (10 mL) was added, and
the mixture was stirred for 3 h. The reaction mixture was poured slowly
to a solution of sodium carbonate (5n, 200 mL) to quench the reaction.
The pH level was adjusted to 9–10, the organic layer was separated, and
the aqueous layer was washed with dichloromethane (2ꢀ25 mL). The
combined dichloromethane extracts were dried over MgSO4. The solvent
was removed under reduced pressure and the crude product was dis-
solved in methanol (20 mL). After 12 h the unprotected tetrameric di-
Xanthene-based/derived isocyanate 3: Monoprotected diamine 18
(200 mg, 0.44 mmol) and N-diisopropylethylamine (57 mg, 0.22 mmol)
dissolved in dichloromethane (20 mL) were added dropwise over 1 h to a
stirred solution of triphosgene (66 mg, 0.44 mmol) in dichloromethane
(20 mL) under nitrogen. The reaction mixture was filtered after 3 h
through silica gel (50 g), which was subsequently washed with dichloro-
methane (2ꢀ30 mL). The solvent was removed under reduced pressure,
and the oily crude product was heated on a rotatory evaporator for 1 h at
808C to remove excess triphosgene and cooled. Isocyanate 3 (190 mg,
89%) was obtained as a beige solid. M.p. 1728C; 1H NMR (400 MHz,
[D6]DMSO): d=7.90 (s, NH, 1H), 7.56 (s, ArH, 1H), 7.37 (d, ArH, 1H,
AHCTUNGTRENNUNG
4J
1H, 4J
9H), 1.26 ppm (s, tBu, 9H).
G
E
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
Diphenyl ether-based isocyanate 4: Monoprotected diamine 19 (1.11 g,
3.69 mmol) and N-diisopropylethylamine (477 mg, 3.69 mmol) dissolved
in dichloromethane (50 mL) were added dropwise over 1 h to a stirred
solution of triphosgene (548 mg, 1.85 mmol) in dichloromethane (30 mL)
under nitrogen. The reaction mixture was filtered after 3 h through silica
gel (50 g), which was subsequently washed with dichloromethane (3ꢀ
50 mL). The solvent was removed under reduced pressure and the oily
crude product was heated on a rotatory evaporator for 1 h at 808C to
remove excess triphosgene and cooled. Isocyanate 4 (1.07 mg, 89%) was
obtained as a light-brown oil. The product is liquid at room temperature,
cannot be stored, and should be used immediately after preparation.
AHCTUNGTRENNUNG
1
0.25 mmol) dissolved in dichloromethane (25 mL) was added dropwise
with stirring over 20 min under nitrogen to this solution. The solvent was
removed under reduced pressure after 18 h of stirring. The crude product
was triturated with hexane (25 mL). A white solid was filtered off to
yield hexamer 12 (129 mg, 49%). M.p. >3708C (decomp); 1H NMR
(400 MHz, [D8]THF): d=11.66 (s, NH, 2H), 8.89 (s, NH, 2H), 8.82 (d,
ArH, 2H, 4J
(H,H)=2.0 Hz), 8.36 (s, NH, 2H), 7.85 (d, ArH, 2H, 4J
ACHTUNGTRENNUNG
7.80 (s, NH, 2H), 7.61 (d, ArH, 2H, J
4J
ACHTUNGTRENNUNG
1H NMR (400 MHz, CDCl3): d=8.20 (d, ArH, 1H, 3J
7.14 (m, ArH, NH, 4H), 6.98 (m, ArH, 2H), 6.88 (dd, ArH, 1H, 3J-
(H,H)=8 Hz, 4J(H,H)=2 Hz), 6.79 (dd, ArH, 1H, 3J(H,H)=8 Hz, 4J-
(H,H)=2 Hz), 1.53 ppm (s, tBu, 9H).
ACHTUNGTREN(NUNG H,H)=7.8 Hz),
U
4
ACHTUNGTRENNUNG
A
R
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
N
ACHTUNGTRENNUNG
Diisocyanate 9 (DXD): A solution of the diamine 8 (136 mg, 0.17 mmol)
G
ACHTUNGTRENNUNG
with diisopropylethylamine (44 mg, 0.34 mmol) in dichloromethane
G
ACHTUNGTRENNUNG
(25 mL) was added dropwise to
a solution of triphosgene (50 mg,
6.68 (s, NH, 2H), 1.62 (s, CH3, 6H), 1.49 (s, CH3, 6H), 1.44 (s, CH3, 6H),
1.36 (s, tBu, 18H), 1.33 (s, tBu, 18H), 1.29 (s, CH3, 6H), 1.22 (s, tBu,
18H), 1.03 (s, CH3, 6H), 0.99 (s, tBu, 18H), 0.96 (s, tBu, 18H), 0.95 (s,
0.17 mmol) in acetonitrile (25 mL) with stirring under nitrogen. The reac-
tion mixture was filtered after 3 h through silica gel, which subsequently
was washed with dichloromethane (3ꢀ25 mL). The solvent was removed
under reduced pressure and the crude product was heated on a rotatory
evaporator for 1 h at 808C to remove excess triphosgene and cooled. Dii-
tBu, 18H), 0.71 ppm (s, CH3, 6H); MS
[M++Na].
ACHTUNGTNER(NUNG ESI): m/z (%): 2293.3 (100%)
Cyclic hexaurea 14 (XDXDXD): Diamine 11 (XDXDX, 18 mg,
0.015 mmol) and diisocyanate 2 (2.9 mg, 0.015 mmol) were dissolved in
dichloromethane (3 mL) and stirred for 18 h. The solvent was removed
under reduced pressure and the residue was triturated with acetonitrile.
The white solid was filtered off to yield hexamer 14 (16 mg, 80%).
socyanate
9 was isolated as a glass-like solid (95%, 138 mg). The
1H NMR spectrum in CDCl3 was broadened and unclear. M.p. >1858C
(decomp).
Diamine 11 (XDXDX): A solution of diamine 8 (196 mg, 0.243 mmol) in
M.p. >2908C (decomp); MS
Cyclic hexaurea 15 (XDDXDD): Diamine 8 (DXD, 130 mg, 0.16 mmol)
dissolved in dichloromethane (25 mL) was added to diisocyanate
(ESI): m/z (%): 1837.2 (100%) [M++Na].
ACHTUNGTRENNUNG
THF (20 mL) was added to
a solution of isocyanate 3 (234 mg,
0.487 mmol) in THF (20 mL) with stirring under nitrogen. The solvent
was removed after 18 h under reduced pressure and the residue was dis-
solved in acetonitrile. A white precipitate appeared after 30 min to yield
the tert-butoxycarbonyl (Boc)-protected pentameric diamine as a white
powder after filtration. The compound was dissolved in dichloromethane
(10 mL), trifluoroacetic acid (10 mL) was added, and the mixture was
stirred for 3 h. The reaction mixture was poured slowly to a solution of
sodium carbonate (5n, 200 mL) to quench the reaction. The pH level
was adjusted to 9–10. The organic layer was separated, and the aqueous
layer was washed with dichloromethane (2ꢀ25 mL). The combined di-
chloromethane extracts were dried over MgSO4 and the solvent was re-
moved under reduced pressure, thus yielding the dimeric diamine 11
(0.22 g, 58%) as a brownish glass-like solid. M.p. > 1908C (decomp);
1H NMR (400 MHz, [D6]DMSO): d=9.11 (s, NH, 2H), 9.00 (s, NH, 2H),
9
(DXD, 138 mg, 0.16 mmol) dissolved in acetonitrile (75 mL) with stirring
and under nitrogen. The solvent was removed after 18 h under reduced
pressure, and the oily residue was triturated with acetonitrile. Hexamer
15 was isolated as a white powder (45%, 120 mg). M.p. 2438C; MS
ACHTUNGTRENNUNG(ESI):
m/z (%): 1684.8 (100%) [M++Na].
Cyclic hexaurea 16 (DDDDDD): Diamine 10 (DDDD, 111 mg,
0.13 mmol) dissolved in ethyl acetate/acetonitrile (1:1 (v/v), 25 mL) was
added to diisocyanate 7 (DD, 60 mg, 0.13 mmol) dissolved in the same
solvent (30 mL) with stirring and under nitrogen. The solvent was re-
moved under reduced pressure after 18 h, and the residue was separated
by column chromatography on silica gel (chloroform/acetone, 50:1 (v/v),
200 mL). The isolated hexamer was slightly contaminated with other
cyclic by-products. Therefore, the solid was triturated with methanol to
remove the insoluble by-products. The filtrate was evaporated under re-
duced pressure and hexamer 16 was isolated as a white powder (53%,
91 mg). M.p. 254–2578C; 1H NMR (400 MHz, [D6]DMSO): d=9.07 (brs,
8.80 (s, NH, 2H), 8.68 (s, NH, 2H), 8.33 (dd, ArHdiph, 2H, 3J
4J 2H, 4J
(H,H)=1.2 Hz), 8.15 (d, ArHxan (H,H)=2.4 Hz), 8.02 (d,
ArHxan, 2H, 4J(H,H)=2.4 Hz), 7.88 (dd, ArHdiph, 2H, 3J(H,H)=7.0. 4J-
(H,H)=2.5 Hz), 7.07 (m, ArH, 12H), 6.74 (d, ArHdiph, J(H,H)=7.0 Hz),
6.61 (m, ArH, 6H), 6.46 (d, ArHxan, 2H, 4J
(H,H)=2 Hz), 4.80 (s, NH2,
4H), 1.55 (s, CH3xan, 6H), 1.51 (s, CH3xan, 12H), 1.27 (s, tBuxan, 18H), 1.18
ACHTUNGTRENUN(NG H,H)=8.2,
N
,
ACHTUNGTRENNUNG
N
ACHTUNGTRENNUNG
3
T
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
NH, 12H), 8.17 (d, ArH, 12H, 3J
(H,H)=7.6 Hz), 6.90 (t, ArH, 12H, J
12H, J(H,H)=7.6 Hz); MS
(ESI): m/z (%): 1380.1 (100%) [M++Na].
ACHTUNGTRENNUNG
(s, tBuxan, 18H), 1.13 ppm (s, tBuxan, 18H); MS
(100%) [M++Na].
A
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
Chem. Eur. J. 2009, 15, 4811 – 4821
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4819

V. Bçhmer et al.
Stability-constant determination: UV
absorption spectrophotometric titra-
tions were carried out to obtain the
stability constants (logb) of the com-
plexes formed between the hexameric
ligands and the chloride, bromide, ace-
tate, and dihydrogen phosphate anions
provided as their tetrabutylammonium
salts. All the experiments were carried
out in a mixture of THF/acetonitrile
(1:3, v/v). The procedure has been pre-
viously reported in detail.[9,10] The con-
centration of the ligand was in the
range 2ꢀ10À6–1.2ꢀ10À5 m, and the
final anion/ligand ratios reached at the
end of the titrations varied from 2:1 to
44:1, according to the strength of the
complexes. The measurements were
treated with the numerical program
SPECFIT[15] to give the stoichiome-
tries and the stability constants of the
complexes.
Table 4. Crystallographic data.
17
15a
15b
15g
chemical formula
C
119H152N10O9·
C100H100N12O12·
8C3H6O
C100H100N12O12·
2C4H8O2·
2C2H6O
705737
C100H100N12O12·
3C2H6O·C4H8O2·
C2H3N·H2O
705738
C3H6O·2CH2Cl2·
4H2O
CCDC ref. numbers
706372
705736
2126.54
triclinic
Mr
2166.50
1930.26
1947.30
crystal system
space group
a [ꢄ]
b [ꢄ]
c [ꢄ]
triclinic
P1
triclinic
triclinic
¯
¯
¯
¯
P1
P1
P1
14.9057(8)
20.4172(9)
21.4080(11)
88.307(4)
80.779(4)
76.888(4)
6263.1(5)
2
15.2902(12)
15.3882(13)
15.7353(12)
117.903(6)
101.301(6)
103.215(6)
2981.3(4)
1
173
31249
11153
0.0567
12.3418(11)
15.1520(13)
15.2483(11)
99.182(6)
109.839(7)
97.578(7)
2594.2(4)
1
173
24886
9647
0.0892
15.5508(6)
18.1537(7)
21.1074(8)
78.610(3)
81.262(3)
78.204(3)
5679.2(4)
2
173
144878
21582
0.1483
0.3163
a [8]
b [8]
g [8]
V [ꢄ3]
Z
T [K]
173
reflections
unique reflections
Rint
83963
24000
0.0522
0.3233
wR(F2) all data
0.1989
0.2050
Single-crystal X-ray diffraction studies
Preparation of crystals: In general,
mixtures containing acetone, acetoni-
trile, chloroform, dichloromethane,
Acknowledgements
ethanol, ethyl acetate, methanol, and THF were checked for crystalliza-
tion to increase the probability of finding a suitable combination of sol-
vents. The compound was dissolved in a solvent in which it was highly
soluble, and the solution was mixed with another solvent in which the
compound was poorly soluble. The resulting solution was allowed to
evaporate slowly through a very small hole made by a needle in the plas-
tic cap of the bottle. When a precipitate was formed in the bottle but no
crystallization occurred, the compound was dissolved again in another
pair of solvents, and so on until single crystals were obtained. Thus, in
each single case every solvent listed above may be present in the mixture,
but two of them were always the main components. The main pair of sol-
vents were dichloromethane/acetone, ethyl acetate/ethanol, and ethyl
acetate/acetonitrile for a, b, and g, respectively.
Financial support from the Deutsche Forschungsgemeinschaft (Bç523/17–
1-3) and CNRS is gratefully acknowledged.
[1] J. L. Sessler, P. A. Gale, W. Cho in Anion Receptor Chemistry, Mon-
ographs in Supramolecular Chemistry (Ed.: J. F. Stoddart), RSC,
Cambridge, 2006; P. A. Gale, R. Quesada, Coord. Chem. Rev. 2006,
250, 3219–3244; P. A. Gale, S. E. Garcꢅa-Garrido, J. Garric, Chem.
Soc. Rev. 2008, 37, 151–190; G. W. Bates, P. A. Gale, Struct. Bonding
(Berlin) 2008, 129, 1–44; P. D. Beer, P. A. Gale, Angew. Chem. 2001,
113, 502–532; Angew. Chem. Int. Ed. 2001, 40, 486–516; P. Prados,
R. Quesada, Supramol. Chem. 2008, 20, 201–216.
[2] C. R. Bondy, S. J. Loeb, Coord. Chem. Rev. 2003, 240, 77–99; M. J.
Chmielewski, J. Jurczak, Chem. Eur. J. 2005, 11, 6080–6094; S. O.
Kang, D. Powell, V. W. Day, K. Bowman-James, Angew. Chem. 2006,
118, 1955–1959; Angew. Chem. Int. Ed. 2006, 45, 1921–1925; S. O.
Kang, R. A. Begum, K. Bowman-James, Angew. Chem. 2006, 118,
8048–8061; Angew. Chem. Int. Ed. 2006, 45, 7882–7894; J. L. Sess-
ler, S. K. Kim, D. E. Gross, C.-H. Lee, J. S. Kim, V. M. Lynch, J. Am.
Chem. Soc. 2008, 130, 13162–13166.
[3] For a recent review, see: J. L. Sessler, S. Camiolo, P. A. Gale, Coord.
Chem. Rev. 2003, 240, 17–55; E. A. Katayev, E. N. Myshkovskaya,
N. V. Boev, V. N. Khrustalev, Supramol. Chem. 2008, 20, 619–624.
[4] B. H. M. Snellink-Ruꢆl, M. M. G. Antonisse, J. F. J. Engbersen, P.
Timmermann, D. N. Reinhoudt, Eur. J. Org. Chem. 2000, 165–170;
D. Ranganathan, C. Lakshmi, Chem. Commun. 2001, 1250–1251; R.
Herges, A. Dikmans, U. Jana, F. Kçhler, P. G. Jones, I. Dix, T.
Fricke, B. Kçnig, Eur. J. Org. Chem. 2002, 3004–3014; V. Amendola,
D. Esteban-Gꢇmez, L. Fabbrizzi, M. Licchelli, Acc. Chem. Res.
2006, 39, 343–353; S. J. Brooks, S. E. Garcꢅa-Garrido, M. E. Light,
P. A. Cole, P. A. Gale, Chem. Eur. J. 2007, 13, 3320–3329; S. E.
Garcꢅa-Garrido, C. Caltagirone, M. E. Light, P. A. Gale, Chem.
Commun. 2007, 1450–1452; D. R. Turner, M. J. Paterson, J. W.
Steed, Chem. Commun. 2008, 1395–1397.
Analysis: Data were collected on a STOE-IPDS-II two-circle diffracto-
ACHTUNGTRENNUNGmeter with graphite-monochromated MoKa radiation (0.71073 ꢄ). Data
reduction was performed with the X-AREA software.[16] Structures were
solved by direct methods with SHELXS-90 and refined by using the full-
matrix least-squares techniques with SHELXL-97.[17]
All non-hydrogen atoms, excluding the disordered t-butyl groups in 15a
and 15g, the water molecules, and disordered dichloromethane molecules
in 17, were refined with anisotropic displacement parameters. Hydrogen
atoms bonded to carbon atoms were included at calculated positions and
allowed to ride on their parent atoms. The hydrogen atoms bonded to ni-
trogen atoms in 15a were freely refined, the hydrogen atoms bonded to
À
oxygen atoms in 15g were refined with the O H lengths restrained to
0.84 ꢄ and with the H···H distances restrained to 1.4 ꢄ. The hydrogen
atoms of the water molecules in 17 could not be found in a difference
map and were omitted from refinement. The two t-butyl groups in 15a
are disordered over two positions with a ratio of the site-occupation fac-
tors of 0.532(6)/0.468(6) and 0.588(6)/0.412(6), respectively. One of the t-
butyl groups in 56 is disordered over two positions with a ratio of the site
occupation factors of 0.767(8)/0.233(8). Bond lengths and angles of the
disordered atoms were restrained to have the same geometric parameters
as those of the non-disordered tert-butyl group. One of the tert-butyl
groups in 17 is disordered over two positions with a ratio of the site-occu-
pation factors of 0.50(2)/0.50(2) and the dichloromethane with site-occu-
[5] For example, see: H. Luecke, F. A. Quiocho, Nature 1990, 347, 402–
406; B. L. Jacobson, F. A. Quiocho, J. Mol. Biol. 1988, 204, 783–787;
R. Dutzler, E. B. Campbell, M. Cadene, B. T. Chait, R. MacKinnon,
Nature 2002, 415, 287–294.
[6] B. P. Hay, M. Gutowski, D. A. Dixon, J. Garza, R. Vargas, B. A.
Moyer, J. Am. Chem. Soc. 2004, 126, 7925–7934; B. P. Hay, T. K.
Firman, B. A. Moyer, J. Am. Chem. Soc. 2005, 127, 1810–1819; for
À
pation factors of 0.575(5)/0.425(5). The C Cl bond lengths in this di-
chloromethane molecule were restrained to 1.7 ꢄ and the Cl···Cl distan-
ces to 2.8 ꢄ. The contribution of the ethyl acetate molecules in 15b was
suppressed using the SQUEEZE[18] option in PLATON.[19] The crystallo-
graphic data are summarized in Table 4.
4820
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4811 – 4821

Anions as Templates in the Formation of Macrocycles
FULL PAPER
general review on nitrate receptors, see O. A. Okunola, P. V. Santa-
croce, J. T. Davis, Supramol. Chem. 2008, 20, 169–190.
[7] Preparation: B. C. Hamann, N. R. Branda, J. Rebek, Jr., Tetrahedron
Lett. 1993, 34, 6837–6840; Application: P. Bꢂhlmann, S. Nishizawa,
K. P. Xiao, Y. Umezawa, Tetrahedron 1997, 53, 1647–1654; V.
Alcꢈzar, M. Segura, P. Prados, J. de Mendoza, Tetrahedron Lett.
1997, 38, 1033–1036.
[12] Unpublished results.
[13] D. Meshcheryakov, V. Bçhmer, M. Bolte, V. Hubscher-Bruder, F.
Arnaud-Neu, H. Herschbach, A. Van Dorsselaer, I. Thondorf, W.
Mçgelin, Angew. Chem. 2006, 118, 1679–1682; Angew. Chem. Int.
Ed. 2006, 45, 1648–1652.
[14] M. C. Etter, Z. Urbanczyk-Lipkowska, M. Zia-Ebrahimi, T. W. Pan-
untom, J. Am. Chem. Soc. 1990, 112, 8415–8426.
[8] P. R. Ashton, B. Hçrner, O. Kocian, S. Menzer, A. J. P. White, J. F.
Stoddart, D. J. Williams, Synthesis 1996, 930–940.
[15] H. Gampp, M. Maeder, C. J. Meyer, A. D. Zuberbꢂhler, Talanta
1985, 32, 257–264.
[9] D. Meshcheryakov, F. Arnaud-Neu, V. Bçhmer, M. Bolte, V.
Hubscher-Bruder, E. Jobin, I. Thondorf, S. Werner, Org. Biomol.
Chem. 2008, 6, 1004–1014.
[10] D. Meshcheryakov, F. Arnaud-Neu, V. Bçhmer, M. Bolte, J. Cava-
leri, V. Hubscher-Bruder, I. Thondorf, S. Werner, Org. Biomol.
Chem. 2008, 6, 3244–3255.
[16] X-AREA, 2001. Stoe & Cie, Darmstadt, Germany.
[17] G. M. Sheldrick, Acta Crystallogr. Sect. A 2007, 64, 112–122.
[18] P. van der Sluis, A. L. Spek, Acta Crystallogr. Sect. A 1990, 46, 194–
201.
[19] A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7–13.
[11] V. Bçhmer, D. Meshcheryakov, I. Thondorf, M. Bolte, Acta Crystal-
logr. Sect. A 2004, 60, o136–o139.
Received: December 8, 2008
Published online: March 18, 2009
Chem. Eur. J. 2009, 15, 4811 – 4821
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4821