High-yield macrocyclization via glaser coupling of temporary covalent templated bisacetylenes

Full text of DOI:10.1021/jo970350c

4556
J . Org. Chem. 1997, 62, 4556-4557
Sch em e 1a
High -Yield Ma cr ocycliza tion via Gla ser
Cou p lin g of Tem p or a r y Cova len t
Tem p la ted Bisa cetylen es
Sigurd Ho¨ger,* Anne-De´sire´e Meckenstock, and
Heike Pellen
Max-Planck-Institut fu¨r Polymerforschung,
Ackermannweg 10, D-55128 Mainz, Germany
Received February 25, 1997 (Revised Manuscript Received April
8, 1997 )
The synthesis of well-defined, shape-persistent nanom-
eter-scale objects has obtained increasing attention dur-
ing the last several years. Besides one-dimensional
structures, those of higher dimensionality are especially
interesting1 because they can be used as host molecules
if they contain appropriate binding sites.2 Some of the
most established cyclic structures are based on the
phenylacetylene backbone. Two extreme synthetic strat-
egies can be distinguished. Staab and co-workers pre-
pared cyclic hexa-m-phenylacetylene in a one-pot reaction
by a 6-fold Stephens-Castro coupling of 3-iodophenyl-
acetylene in 4.6% yield.3 Despite the low yield for the
cyclization step, this method is impressive because the
starting materials are readily accessible. A completely
different approach was investigated by the group of
Moore.4 They prepared similar and other phenylacet-
ylene macrocycles in good to excellent yields by an
intramolecular coupling of the corresponding R-iodo-ω-
ethynyl precursor; however, the synthesis of the precur-
sors is relatively time consuming. One way to overcome
this dilemma, which also occurs when the Glaser coupling
is used as the bond-forming step,5 has been investigated
by the group of Sanders, using the template directed
synthesis of cyclic porphyrin-acetylene structures; how-
ever, this method is restricted to metal-containing com-
pounds.6 We are interested in shape-persistent macro-
cycles containing both nonpolar and polar side groups in
a switchable arrangement, which have been prepared by
the intermolecular Glaser coupling of two rather large
and rigid bisacetylenes. In order to reduce the number
of synthetic steps, we became interested in using small
and easily preparable starting materials like 6 for the
cyclization step.
a
Key: (a) 3-bromo-1-propanol, K2CO3, DMF, 70 °C (92%); (b)
t-BuMe2SiCl, imidazole, DMF, rt (87%); (c) 3, PdCl2(PPh3)2, CuI,
piperidine, 60 °C (91%); (d) Bu4NF, THF, rt (92%); (e) p-toluic acid
chloride, pyridine, THF, rt (80%); (f) CuCl/CuCl2, pyridine, rt; GPC
diagram of 6 (I) and of the crude product of the cyclization of 6
(II).
Scheme 1 illustrates the synthesis of the protected
amphiphilic bisacetylene 6 and its cyclization. Protection
of the free OH group of 5 as an aryl ester and cyclization
of 6 by a modified Eglington-Glaser coupling using high
dilution conditions gave a mixture of different oligomers,
as determined by gel permeation chromatography (GPC).7
The fact that we were not able to detect residual
1
absorptions for acetylenic protons in the H-NMR spectra
supports the assumption that a mixture of cyclic oligo-
mers and cyclic or noncyclic polymers was formed. The
smallest cyclic product of the reaction, the cyclic trimer,
was only formed in about 20-25% yield and, moreover,
due to their similar physical properties we were not able
to separate the reaction products by recrystallization or
column chromatography.8
* To whom correspondence should be addressed. Fax: Int. + 6131/
379 100. E-mail: hoeger@mpip-mainz.mpg.de.
(1) Selected reviews and some new examples: (a) Tour, J . M. Chem.
Rev. 1996, 96, 537. (b) Bunz, U. H. F. Angew. Chem., Int. Ed. Engl.
1994, 33, 1073. (c) Timmermann, P.; Verboom, W.; van Veggel, F. C.
J . M.; van Hoorn, W. P.; Reinhoudt, D. N. Angew. Chem., Int. Ed. Engl.
1994, 33, 1292. (d) Ho¨ger, S.; Enkelmann, V. Angew. Chem., Int. Ed.
Engl. 1995, 34, 2713.
Our approach now is to circumvent the low yield of the
statistical intermolecular reaction described above and
still retain the advantage of small and easily preparable
bisacetylenes. Therefore, we present here a template-
directed synthesis of shape-persistent macrocycles based
on the phenylacetylene backbone by using the Glaser
coupling as the bond-forming step and a covalent linkage
of the monomers to the template.9 For this purpose, the
(2) Host molecules based on phenylacetylene structures: (a) Mor-
rison, D. L.; Ho¨ger, S. J . Chem. Soc., Chem. Commun. 1996, 2313. (b)
Anderson, H. L.; Sanders, J . K. M. J . Chem. Soc., Chem. Commun.
1989, 1714. (c) Mackeay, L. G.; Anderson, H. L.; Sanders, J . K. M. J .
Chem. Soc., Chem. Commun. 1992, 43. (d) Anderson, S.; Neidlein, U.;
Gramlich, V.; Diederich, F. Angew. Chem., Int. Ed. Engl. 1995, 34,
1596.
(3) Staab, H. A.; Neunhoeffer, K. Synthesis 1974, 424.
(4) Zhang, J .; Pesak, D. J .; Ludwig, J . L.; Moore, J . S. J . Am. Chem.
Soc. 1994, 116, 4227.
(5) (a) de Meijere, A.; Kozhushkov, S.; Haumann, T.; Boese, R.; Puls,
C.; Cooney, M. J .; Scott, L. T. Chem. Eur. J . 1995, 1, 124. (b) Boldi, A.
M.; Diederich, F. Angew. Chem., Int. Ed. Engl. 1994, 33, 486.
(6) For recent articles dealing with template reactions see: (a)
Anderson, S.; Anderson, H. L.; Sanders, J . K. M. Acc. Chem. Res. 1993,
26, 469. (b) Hoss, R.; Vo¨gtle, F. Angew. Chem., Int. Ed. Engl. 1994,
33, 375. (c) Dietrich, B.; Viout, P.; Lehn, J .-M. Macrocyclic Chemistry;
VCH: Weinheim, 1993; Chapter 3.3.
(7) The GPC diagrams were measured in THF, and a UV detector
operating at λ ) 254 nm was used. The molecular weight was obtained
from polystyrene calibration of the GPC columns.
(8) To confirm that it is actually the cyclic trimer, we esterified
compound 9 and compared the GPC diagram with the one obtained
for the cyclization of 6.
S0022-3263(97)00350-2 CCC: $14.00 © 1997 American Chemical Society

Communications
J . Org. Chem., Vol. 62, No. 14, 1997 4557
Sch em e 2a
Sch em e 3a
a
Key: (a) 1,3,5-benzenetricarboxylic acid, DEAD, PPh3, THF,
rt (79%); (b) CuCl/CuCl2, pyridine, rt; (c) NaOH, LiOH, MeOH,
H2O, THF, 80 °C (95%); GPC diagram of 7 (I) and of the crude
product of the cyclization of 7 (II).
a
Key: (a) CuCl/CuCl2, pyridine, rt.
bisacetylene 5 was protected with 1,3,5-benzenetricar-
boxylic acid under Mitsunobu conditions to give the
triester 7 in good yields (Scheme 2).
Both macrocycles are probably a mixture of two isomers.
One of these corresponds to the case where the polar side
groups of a ring point in the same direction, while in the
other case one of the polar groups points in an opposite
direction. We were not able to separate these mixtures
by simple column chromatography.
The results shown above motivated us to extend our
strategy and prepare the tetraacetylene 10 (Scheme 3).10
Under conditions identical to those used above, the
crude product (92%) contained 11 in ∼86-88% yield. The
reason for the lower yield compared to the cyclization of
7 is at the present not clear, and future variations of the
spacer geometry will show if steric restrictions after the
first diyne formation prevent a fast intermolecular second
reaction.
In summary, we can say that macrocyclic geometries
of well-defined size and shape can be easily prepared by
cyclization of relatively small and readily accessible
monomers if they are temporarily covalently bound to a
template. The length of the spacer betwen the template
and the monomer is of only minor importance, and even
attachement of the template at the outside of the final
ring works well.
Cyclization was carried out under the same conditions
used for the coupling of 6, except that the crude product
of the reaction (94% yield) now contained more than 95%
of the template-bound cyclic trimer (according to the GPC
data). If instead of the 3-hydroxypropyl ether 5 the
corresponding 11-hydroxyundecyl ether was used for the
cyclization under the same conditions, the crude product
(88%) again contained more than 95% of the desired
templated macrocycle. This result is quite important
since it shows that for a high-yield intramolecular
coupling of the monomers it is not necessary to arrange
them in a special preorganized geometry. Rather, the
high yield of the cyclization process can be attributed to
an overall low triester concentration in the reaction
mixture (as a result of the high dilution conditions)
together with a high local concentration of terminal
acetylenes (as a result of the intramolecular nature of
the coupling reaction).
Both templated products show, probably due to their
restricted flexibility, complex NMR spectra. Contrary to
our expectations, we could not observe, even in the case
of the undecyl derivative, any simplification of the spectra
upon heating, even up to 130 °C in tetrachloroethane.
Base-catalyzed hydrolysis of 8 gave the macrocycle 9 in
nearly quantitative yield. 9, as well as the undecyl
compound, now show the expected simple NMR spectra.
Su p p or tin g In for m a tion Ava ila ble: Preparation of all
the starting materials including spectral data (1H NMR and
13C NMR). Experimental procedures for the preparation of 6
and 9, including copies of the 1H NMR spectra (10 pages).
J O970350C
(9) The preparation of C3 macrocyclic receptors by a templated triple
macrolactamization (Hong, J .-I.; Namgoong, S. K.; Bernardi, A.; Still,
W. C. J . Am. Chem. Soc. 1991, 113, 5111) and the formation of
macrocyclic tetraesters by reaction on a porphyrin template (Mackey,
L. G.; Bonar-Law, R. P.; Sanders, J . K. M. J . Chem. Soc., Perkin Trans.
1 1993, 1377) are also reported. However, in both cases the yields are
relatively low, and the template was not removed after cyclization.
(10) 10 was prepared using our convergent strategy for the prepara-
tion of arylacetylene oligomers with ethynyl end groups (Ho¨ger, S.;
Mu¨ller, S.; Karcher, L.; Polym. Prepr. (Am. Chem. Soc. Div. Polym.
Chem.) 1997, 38(1), 72). A detailed experimental procedure for the
preparation of 10 as well as the functionalization of the detemplated
macrocycle will be published elsewhere.