Tripodal oxazoline-based homochiral coordination cages with internal binding sites

Article abstract of DOI:10.1039/b514697f

Homochiral coordination cages, which have two well-defined internal binding sites for ammonium and organoammonium ions, have been constructed by Pd(ii)-mediated self-assembly of preorganized tripodal oxazolines containing pyridine pendant groups. The Roya

Full text of DOI:10.1039/b514697f

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Tripodal oxazoline-based homochiral coordination cages with internal
binding sites{
Jeongryul Kim,a Dowook Ryu,a Yoshihisa Sei,b Kentaro Yamaguchib and Kyo Han Ahn*a
Received (in Cambridge, UK) 18th October 2005, Accepted 6th January 2006
First published as an Advance Article on the web 26th January 2006
DOI: 10.1039/b514697f
Homochiral coordination cages, which have two well-defined
internal binding sites for ammonium and organoammonium
ions, have been constructed by Pd(II)-mediated self-assembly of
preorganized tripodal oxazolines containing pyridine pendant
groups.
Metal-directed self-assembly has been widely used to construct
supramolecular container molecules because of its operational
simplicity compared to syntheses based on non-coordinative
covalent bond formation. Thus, a variety of so-called coordination
cages or capsules have been constructed for potential applications
such as molecular receptors, catalysts, delivery systems, and so on.1
In spite of such efforts in the synthesis of coordination cages in
recent years,2 the construction of coordination cages that have
homochirality and/or internal binding sites specific to certain
guests has been rarely addressed and remains as a challenging
task.3 This may be due to the difficulty of incorporating chirality
or guest binding sites in the cavity of coordination cages. Herein,
we wish to report for the first time the construction of homochiral
supramolecular cages by metal mediated self-assembly of chiral
tripodal oxazolines containing pyridine pendant groups. The cages
contain two well-defined internal binding sites suitable for
ammonium and organoammonium ions, provided by the homo-
chiral oxazoline ligands.
Scheme 1 Reagents and conditions: (a) (COCl)2, Et3N, CH2Cl2; (b)
MsCl, Et3N, DMAP, CH2Cl2; (c) 1.0 M aq. NaOH, MeOH, rt, 6 h; (d)
NaH, DMF; 4-bromomethylpyridine, CH2Cl2, rt, 24 h.
Based on our experience of molecular recognition using tripodal
oxazoline receptors,4 we envisioned that they might be used as new
platforms for the construction of homochiral coordination cages
because they readily provide screw-sense chiral pockets through
preorganization. A molecular modelling study suggested that
tripodal oxazolines based on (3-substituted-phenyl)glycinols, such
as 1 in Scheme 1, would form dimeric supramolecular cages. Upon
binding an ammonium ion through tripodal hydrogen bonds to
three oxazoline nitrogens, the 3-substituted metal binding ligands
point upwards, ready for metal binding to form dimeric cages.
Thus, a protected (3-hydroxyphenyl)glycinol was chosen as the
precursor for the tripodal oxazolines, because introducing the
metal binding ligands via ether formation was readily conceivable.
The tripodal oxazoline 5a was thus synthesized from amino
alcohol 35 and tricarboxylic acid 4a in a one-pot reaction
according to the established procedure.4 Then, deprotection of
the silyl group of 5a, followed by generation of its phenoxide ion
and subsequent treatment with 4-bromomethylpyridine, afforded
pyridyl-oxazoline 1a as a white powder in overall 62% yield.
Similarly, a more preorganized analogue 1b was synthesized
starting from 5b in 53% yield (ESI{). Both 1a and 1b exhibited
rather simple 1H and 13C NMR spectra at room temperature,
consistent with their C3 symmetric nature.
Next, the formation of metal-mediated supramolecular cages
from homochiral tripodal oxazolines 1 was investigated using
trans-palladium(II) bis(triflate) complexes, following literature
conditions.2 Judging from their NMR spectra, the more
preorganized ligand 1b produced a more discrete cage structure
compared to 1a. Thus, mixing 1b and trans-Pd(OTf)2(PEt3)2 in a
2 : 3 molar ratio at room temperature in a solvent such as CH2Cl2,
CHCl3, or CH3COCH3 produced dimeric cage 2b (Fig. 1), whose
structure was characterized by the following experiments.
aDepartment of Chemistry and Center for Integrated Molecular
Systems, Division of Molecular and Life Science, POSTECH, San 31,
Hyoja-dong, Pohang, 790-784, Republic of Korea.
E-mail: ahn@postech.ac.kr; Fax: (+82) 54 279 3399;
Tel: (+82) 54 279 2105
bDepartment of Pharmaceutical Technology, Faculty of Pharmaceutical
Sciences at Kagawa Campus, Tokushima Bunri University, 1314-1
Shido, Sanuki-city, Kagawa, 769-2193, Japan.
E-mail: yamaguchi@kph.bunri-u.ac.jp; Fax: (+81) 87 894 0181;
Tel: (+81) 87 894 5111
{ Electronic Supplementary Information (ESI) available: Experimental
section, NMR and MS data of the cages, and binding studies by NMR.
See DOI: 10.1039/b514697f
Formation of cage 2b{ could be monitored by following the 1H
NMR spectrum upon changing the molar ratio of the ligand and
Pd source (1b/[Pd(II)]).§ When the molar ratio was less than 2 : 3,
1136 | Chem. Commun., 2006, 1136–1138
This journal is ß The Royal Society of Chemistry 2006

View Article Online
Fig. 1 Self-assembly of 1b and trans-[Pd(OTf)2(PEt3)2], monitored by 1H NMR in CD2Cl2 at 25 uC: (a) Free 1b (the peaks assigned are for the ligand
itself), (b) a mixture of 1b and trans-Pd(OTf)2(PEt3)2 in a ratio of 2 : 1, (c) 1 : 1 and (d) 2 : 3.
two sets of signals were observed, including the ligand’s peaks. The
peak’s relative intensities were dependent upon their molar ratio,
which was clearly indicated by following the a-pyridyl protons
(indicated with dotted lines in Fig. 1). These two sets of signals
merged into one set at the molar ratio 2 : 3; a point equivalent of
dimeric cage formation. The 31P NMR spectrum of 2b exhibited a
sharp singlet at 26.2 ppm, which indicates that the six Et3P groups
are equivalent and the cage is highly symmetrical. Its 19F NMR
spectrum also exhibited a sharp singlet at 277.7 ppm, which
indicates no localization of the triflate ions. The molecular entity of
the dimeric cage was also supported by vapor pressure osmometry
in CH2Cl2 at 25 uC, which gave a molecular weight (3930) close to
the calculated value for the dimeric cage (3841). Further evidence
for cage formation was obtained by cold-spray ionization mass
(CSI MS) analysis of the 2 : 3 mixture, which showed rather weak
but discernable peaks at m/z = 3691 and 3841.9, corresponding to
[M 2 CF3SO3]+ and [M + H]+ fragments. It is worthwhile to note
that cage formation was either incomplete or inhibited by polar
solvents such as CH3CN, Me2SO and DMF. The less preorga-
nized ligand 1a also produced the corresponding dimeric cage
under similar conditions; however, in this case, a broad and less
was observed below 10 uC, suggesting that open structures exist in
only slight amounts, if at all.{
We studied the disassembly of 2b using Et3N as a competing
ligand. Upon adding Et3N, the cage in CH2Cl2 at 25 uC began to
disassemble into ligand 1b, and a complete disassembly was
observed in the presence of 10 equivalents of Et3N.
Based on our successful cage formation, we investigated the
molecular encapsulation ability of cage 2b towards ammonium
ions. As we have demonstrated previously, ligands 1 can efficiently
bind ammonium and organoammonium ions through tripodal
hydrogen bonding and cation p-interactions.4 When cage 2b was
+
2
titrated with NH4 PF6 in CD2Cl2 at room temperature,
significant complexation-induced chemical shifts (CICS) were
observed when two equivalents of the ammonium ion were added,
readily observable for one of the oxazoline ring protons (d 3.97 A
4.23; Dd = 0.26) and the benzylic (d 2.84 A 2.56; Dd = 0.28)
protons.{ This result indicates, as expected, that 2b binds two
equivalents of the ammonium ion at both of the internal oxazoline
binding sites. This binding experiment also confirms that the cage
has a dimeric structure. At this point, we confirmed that a 1 : 1
+
inclusion complex between 1b and NH4 PF62 could also form the
1
well-resolved H NMR spectrum was observed compared to the
corresponding guest-included cage complex, upon addition of the
palladium precursor through a self-assembly process. Thus, we can
generate the ammonium ion-included cage either by cage
formation followed by guest inclusion or vice versa.
case of 1b.
Cage 2b seems to provide a cylindrical cavity, the volume of
which was calculated to be about 2100 s3 by fitting an ellipsoid of
rotation whose principal axes were a = 28 s, b = 10 s and c =
10 s." This cage may have ‘‘twist chirality’’6 owing to the screw-
sense chirality of the oxazoline phenyl substituents. However, we
could not confirm such helicity either for the cage itself or for its
(R)-a-phenylethylammonium complex by circular dichroism
spectroscopy, possibly due to their flexible nature. VT-NMR
experiments from 235 to 40 uC for cage 2b resulted in further
splitting of the peaks; for example, two major peaks for the
a-pyridyl protons, which might have been due to the twist
chirality. A slight peak at 8.6 ppm due to free a-pyridyl protons
As the tripodal oxazoline ligands derived from phenylglycinol
itself bind ammonium and organoammonium ions efficiently,4
cages 2, which contain similar binding pockets, are also expected to
selectively recognize other chiral and/or achiral organoammonium
ions that fit the cavity. A preliminary binding study towards the
a-phenylethylammonium ion (as its perchlorate salt) indicated that
2b forms distinct (R,R)- and (S,S)-diastereomeric inclusion
complexes. Upon addition of the (R)- or (S)-guest to 2b in a 2 :
1 molar ratio, the a-methyl group of each of the enantiomeric
guests showed a significant upfield shift (0.17 and 20.04 ppm for
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 1136–1138 | 1137

View Article Online
dimeric cage is preferred over other complexes due to the geometry of 1b.
The result also excludes the possibility of cis–trans equilibration of the
trans-Pd source under the given conditions.
" Molecular mechanics computations were performed using Spartan
Windows ’04, purchased from Wavefunction, Inc.
the (R,R)- and (S,S)-inclusion complexes, respectively; 1.63 ppm in
the absence of 2b). A similar upfield shift was observed previously
in the case of ‘‘open’’ tripodal oxazoline receptors similar to 5,
explained by the diamagnetic shielding of the guest surrounded by
the oxazoline phenyl rings.4 When four molar equivalents of a
racemic mixture of the a-phenylethylammonium ion were added to
2b, interestingly, the a-methyl group of the guest was split into
three peaks of ratio 10 : 7 : 2, in which the two major peaks could
be assigned to the (R,R)- and (S,S)-inclusion complexes,
respectively; the new minor peak appearing at 20.21 ppm was
tentatively assigned to the (R,S)-inclusion complex.{ A further
study of the enantiodiscriminatory behavior of 2b towards other
chiral organoammonium ions is necessary to address the scope and
limitation of our cage system as a chiral selector.
1 M. Fujita, M. K. Umemoto, M. Yoshizawa, N. Fujita, T. Kusukawa
and K. Biradha, Chem. Commun., 2001, 509–518; D. L. Caulder and
K. N. Raymond, Chem. Commun., 2001, 509–518; S. R. Seidel and
P. J. Stang, Acc. Chem. Res., 2002, 35, 972–983; F. Hof, S. L. Craig,
C. Nuckolls and J. Rebek, Jr., Angew. Chem., Int. Ed., 2002, 41,
1488–1508.
2 For selected recent examples of coordination cages, see: O. D. Fox, M. G.
B. Drew and P. D. Beer, Angew. Chem., Int. Ed., 2000, 39, 136–140;
S. J. Park and J.-I. Hong, Chem. Commun., 2001, 1554–1555; C. J. Kuehl,
Y. K. Kryschenko, U. Radhakrishnan, S. R. Seidel, S. D. Huang and
P. J. Stang, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 4932–4936;
L. Pirondini, F. Bertolini, B. Cantadori, F. Ugozzoli, C. Massera and
E. Dalcanale, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 4911–4915;
R. Pinalli, V. Cristini, V. Sottili, S. Geremia, M. Campagnolo,
A. Caneschi and E. Dalcanale, J. Am. Chem. Soc., 2004, 126,
6516–6517; K. Kobayashi, Y. Yamada, M. Yamanak, Y. Sei and
K. Yamaguchi, J. Am. Chem. Soc., 2004, 126, 13896–13897.
3 For examples of coordination cages that include guests, see: A. Ikeda,
M. Yoshimura, H. Udzu, C. Fukuhara and S. Shinkai, J. Am. Chem.
Soc., 1999, 121, 4296–4297; F. Fochi, P. Jacopozzi, E. Wegelius,
K. Rissanen, P. Cozzini, E. Marastoni, E. Fisicaro, P. Manini,
R. Fokkens and E. Dalcanale, J. Am. Chem. Soc., 2001, 123,
7539–7552; K. Kobayashi, K. Ishi, S. Sakamoto, T. Shirasaka and
K. Yamaguchi, J. Am. Chem. Soc., 2003, 125, 10615–10624;
A. H. Mahmoudkhani, A. P. Coˆte´ and G. K. H. Shimizu, Chem.
Commun., 2004, 2678–2679; S. Tashiro, M. Tominaga, M. Kawano,
B. Therrien, T. Ozeki and M. Fujita, J. Am. Chem. Soc., 2005, 127,
4546–4547; T. Haino, M. Kobayashi, M. Chikaraishi and Y. Fukawawa,
Chem. Commun., 2005, 2321–2323.
In summary, we have constructed homochiral coordination
cages for the first time that provide two well-defined internal
binding sites for ammonium and organoammonium ions. The
cages are constructed by Pd(II)-mediated dimerization of C3
symmetric tripodal oxazoline units bearing pyridine pendant
groups. A preliminary binding study indicates that the cages can
recognize organoammonium ions, including chiral examples.{
This work was supported by the Center for Integrated
Molecular Systems, POSTECH.
Notes and references
{ Cage 2b was assembled by mixing 1b with Pd(OTf)2(PEt3)2 in a molar
ratio 2 : 3 in CH2Cl2 at rt for 5 min. Removal of the solvent in vacuo
afforded the desired cage in quantitative yield: dH (500 MHz, CD2Cl2,
Me4Si, 323 K) 9.1 (12 H, s), 7.5 (12 H, br s), 7.2 (6 H, br s), 6.8 (18 H, br s),
5.0 (18 H, br s), 4.5 (6 H, br s), 4.0 (6 H, br s), 3.8 (12 H, br s), 2.8 (12 H, br
s), 1.9–1.8 (36 H, m), 1.7 (18, br s) and 1.3–1.25 (54 H, m); dP (81 MHz,
CD2Cl2) 26.2; dF (188.3 MHz, CD2Cl2) 277.7; CSI-MS: m/z 3200.9 [M 2
CF3SO32]+ and 3841.9 [M + H]+.
4 S.-G. Kim and K. H. Ahn, Chem.–Eur. J., 2000, 6, 3399–3403; S.-G. Kim,
K.-H. Kim, J. Jung, S. K. Shin and K. H. Ahn, J. Am. Chem. Soc., 2002,
124, 591–596; K. H. Ahn, S.-G. Kim, K.-H. Kim, Y. K. Kim and
S. K. Shin, J. Am. Chem. Soc., 2003, 125, 13819–13824.
5 K. H. Ahn, unpublished results.
6 A. Ikeda, H. Udzu, Z. Zhong, S. Shinkai, S. Sakamoto and
K. Yamaguchi, J. Am. Chem. Soc., 2001, 123, 3872–3877.
§ With the corresponding cis Pd(II) source, no cage formation was
observed. This result corresponds to the modelling study that indicates the
1138 | Chem. Commun., 2006, 1136–1138
This journal is ß The Royal Society of Chemistry 2006