m-Diethynylbenzene macrocycles: Syntheses and self-association behavior in solution

Article abstract of DOI:10.1021/ja012458m

m-Diethynylbenzene macrocycles (DBMs), buta-1,3-diyne-bridged [4_n]metacyclophanes, have been synthesized and their self-association behaviors in solution were investigated, Cyclic tetramers, hexamers, and octamers of DBMs having exo-annular octyl, hexadecyl, and 3,6,9-trioxadecyl ester groups were prepared by intermolecular oxidative coupling of dimer units or intramolecular cyclization of the corresponding open-chain oligomers. The aggregation properties were investigated by two methods, the ¹H NMR spectra and the vapor pressure osmometry (VPO). Although some discrepancies were observed between the association constants obtained from the two methods, the qualitative view was consistent with each other. The analysis of self-aggregation by VPO revealed unique aggregation behavior of DBMs in acetone and toluene, which was not elucidated by the NMR method. Namely, the association constants for infinite association are several times larger than the dimerization constant, suggesting that the aggregation is enhanced by the formation of dimers (a nucleation mechanism). In polar solvents, DBMs aggregate more strongly than in chloroform due to the solvophobic interactions between the macrocyclic framework and the solvents. Moreover, DBMs self-associate in aromatic solvents such as toluene and o-xylene more readily than in chloroform. In particular, the hexameric DBM having a large macrocyclic cavity exhibits extremely large association constants in aromatic solvents. By comparing the aggregation properties of DBMs with the corresponding acyclic oligomers, the effect of the macrocyclic structure on the aggregation propensity was clarified. Finally, it turned out that DBMs tend to aggregate more readily than the corresponding phenylacetylene macrocycles, acetylene-bridged [2_n]metacyclophanes, owing to the withdrawal of the electron density from the aromatic rings by the butadiyne linkages which facilitates π-π stacking interactions.

Full text of DOI:10.1021/ja012458m

Published on Web 04/18/2002
m-Diethynylbenzene Macrocycles: Syntheses and
Self-Association Behavior in Solution
Yoshito Tobe,* Naoto Utsumi, Kazuya Kawabata, Atsushi Nagano, Kiyomi Adachi,
Shunji Araki, Motohiro Sonoda, Keiji Hirose, and Koichiro Naemura
Contribution from the Department of Chemistry, Faculty of Engineering Science,
Osaka UniVersity, and CREST, Japan Science and Technology Coporation (JST),
Toyonaka, Osaka 560-8531, Japan
Received October 30, 2001
Abstract: m-Diethynylbenzene macrocycles (DBMs), buta-1,3-diyne-bridged [4n]metacyclophanes, have
been synthesized and their self-association behaviors in solution were investigated. Cyclic tetramers,
hexamers, and octamers of DBMs having exo-annular octyl, hexadecyl, and 3,6,9-trioxadecyl ester groups
were prepared by intermolecular oxidative coupling of dimer units or intramolecular cyclization of the
corresponding open-chain oligomers. The aggregation properties were investigated by two methods, the
1H NMR spectra and the vapor pressure osmometry (VPO). Although some discrepancies were observed
between the association constants obtained from the two methods, the qualitative view was consistent
with each other. The analysis of self-aggregation by VPO revealed unique aggregation behavior of DBMs
in acetone and toluene, which was not elucidated by the NMR method. Namely, the association constants
for infinite association are several times larger than the dimerization constant, suggesting that the aggregation
is enhanced by the formation of dimers (a nucleation mechanism). In polar solvents, DBMs aggregate
more strongly than in chloroform due to the solvophobic interactions between the macrocyclic framework
and the solvents. Moreover, DBMs self-associate in aromatic solvents such as toluene and o-xylene more
readily than in chloroform. In particular, the hexameric DBM having a large macrocyclic cavity exhibits
extremely large association constants in aromatic solvents. By comparing the aggregation properties of
DBMs with the corresponding acyclic oligomers, the effect of the macrocyclic structure on the aggregation
propensity was clarified. Finally, it turned out that DBMs tend to aggregate more readily than the
corresponding phenylacetylene macrocycles, acetylene-bridged [2n]metacyclophanes, owing to the
withdrawal of the electron density from the aromatic rings by the butadiyne linkages which facilitates π-π
stacking interactions.
Introduction
suffer from fluctuation arising from cis-trans isomerization,
small steric demand, and the facility in connecting an sp carbon
During the last two decades, conformationally rigid and
shape-persistent molecules of nanometer scale, such as large
macrocycles,1 molecular wires,2 and dendrimers,1a,3 have at-
tracted a great deal of interest because of their potential as
functional materials. Rational design of building blocks plays
a crucial role in construction of such molecular architectures.
The carbon-carbon triple bond is a useful connecting unit in
this respect, because of the structural linearity that does not
to an sp2 or sp carbon center owing to the numerous methods
developed over many years.4 Moreover, because of the isotropic
distribution of the π electrons along the C-C axis, acetylene
linkages are capable of transmitting electronic perturbation
efficiently from one end of conjugated π-systems to the other.
These characteristics have been successfully exploited in
molecular wires2 and conjugated dendrimers.1a,3
Moore and co-workers reported the synthesis of the acetylene-
bridged [2n]metacyclophanes, called phenylacetylene macro-
cycles (PAMs),5,6 and their interesting properties based on their
association by weak noncovalent interactions. These include self-
association in solution7 and organization to porous molecular
crystals,8 liquid crystals,9 and monolayer surfaces.10 Moreover,
* Address correspondence to this author. E-mail: tobe@chem.es.
osaka-u.ac.jp.
(1) (a) Moore, J. S. Acc. Chem. Res. 1997, 30, 402-413. (b) Bong, D. T.;
Clark, T. D.; Granja, J. R.; Ghadiri, M. R. Angew. Chem., Int. Ed. 2001,
40, 988-1011. (c) Ho¨ger, S. J. Polym. Sci. Part A: Polym. Chem. 1999,
37, 2685-2698. (d) Haley, M. M. Top. Curr. Chem. 1999, 201, 81-130.
(2) (a) Tour, J. M. Chem. ReV. 1996, 96, 537-553. (b) Tour, J. M. Acc. Chem.
Res. 2000, 33, 791-804. (c) Swager, T. M. Acc. Chem. Res. 1997, 311,
201-207. (d) McQuade, D. T.; Pullen, A. E,; Swager, T. M. Chem. ReV.
2000, 100, 2537-2574. (e) Martin, R. E.; Diederich, F. Angew. Chem.,
Int. Ed. 1999, 38, 1350-1377. (f) Bunz, U. H. F. Chem. ReV. 2000, 100,
1605-1644.
(3) (a) Smith, D. K.; Diederich, F. Chem. Eur. J. 1998, 4, 1353-1361. (b)
Adronov, A.; Fre´chet, J. M. J. Chem. Commun. 2000, 1701-1710. (c)
Hecht, S.; Fre´chet, J. M. J. Angew. Chem., Int. Ed. 2001, 40, 74-91. (d)
Watson, M. D.; Fechtenko¨tter, A.; Mu¨llen, K. Chem. ReV. 2001, 101, 1267-
1300.
(4) (a) Brandsma, L.; Verkruijsse, H. D. Synthesis of Acetylenes, Allenes and
Cumulenes: A Laboratory Manual; Elsevier: New York, 1981. (b)
Brandsma, L. PreparatiVe Acetylene Chemistry; Elsevier: Amsterdam,
1988. (c) Sonogashira, K. In ComprehensiVe Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 3, pp 521-
550. (d) Sonogashira, K. In ComprehensiVe Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 3, pp 551-
561. (e) Siemsen, P.; Livingston, R. C.; Diederich, F. Angew. Chem., Int.
Ed. 2000, 39, 2632-2657.
9
5350
J. AM. CHEM. SOC. 2002, 124, 5350-5364
10.1021/ja012458m CCC: $22.00 © 2002 American Chemical Society

m-Diethynylbenzene Macrocycles
A R T I C L E S
Ho¨ger et al. reported the synthesis of macrocyclic metapara-
cyclophanes having a large cavity constructed by acetylene and
buta-1,3-diyne bridges11 and their organization in solution,12 in
condensed phase,13 and on a solid surface,12 as well as the guest
binding ability toward an amine derivative.14 Recently, Yamagu-
chi et al. reported the synthesis of PAM analogues incorporating
chiral benzo[c]phenanthrene units and the diastereoselectivity
in their self-aggregation in solution.15 Cation binding properties
of PAMs bearing endo-annular methoxy groups were reported
by the group of Kawase and Oda.16 Thus these properties based
on noncovalent interactions can, in principle, be fine-tuned by
modifying the size and shape of the macrocycles and the nature
of the functional groups attached to the periphery and interior
of the macrocyclic framework.
DBMs due to the π-π stacking interaction, because it has been
well documented, both theoretically21 and experimentally,22 that
such interaction is susceptible to the electronic effect of the
substituents. (iii) There is ample experimental evidence which
indicates that buta-1,3-diyne units stack in the crystalline state
to parallel alignment with intermolecular distances of 3.5 to 5.0
Å because of their rodlike shape.23 Moreover, buta-1,3-diyne
derivatives are known to undergo topochemically controlled
polymerization in the solid state23,24 or in the liquid crystalline
state25 to give polydiacetylenes, which have interesting electronic
and optical properties. It is therefore expected that DBMs would
yield upon polymerization highly ordered three-dimensional
polydiacetylene structures provided that proper intermoleculer
arrangement were attained.26
We embarked on a research program aimed at the develop-
ment of the chemistry of new macrocyclic structures 1 based
on the butadiyne-bridged metacyclophanes, diethynylbenzene
macrocycles (DBMs) or “big brothers” of PAMs,17,18,19 because
we anticipated that the buta-1,3-diyne units would make a
substantial difference in the properties of DBMs from those of
PAMs in the following respects. (i) Since the ring sizes of the
diyne-bridged macrocycles are larger than those of the corre-
sponding PAMs having the same number of aromatic rings, it
is possible to introduce binding functionalities along the interior
of the macrocycles for a large gust molecule.20 (ii) The electron
density of the aromatic rings of DBMs would suffer from
stronger perturbation owing to a stronger electron-withdrawing
effect of the buta-1,3-diyne unit than that of the ethyne unit as
described later. This will alter the association properties of
(5) (a) Zhang, J.; Pesak, D. J.; Ludwick, J. L.; Moore, J. S. J. Am. Chem. Soc.
1994, 116, 4227-4239. (b) Bedard, T. C.; Moore, J. S. J. Am. Chem. Soc.
1995, 117, 10662-10671. (c) Young, J. K.; Moore, J. S. In Modern
Acetylene Chemistry; Stang, P. J., Diederich, F., Eds.; VCH: Weinheim,
1995; pp 415-442.
In this connection, we planned to prepare DBMs, i.e.,
tetramers 2a-c, hexamers 3a-c, and octamers 4a,c, having exo-
annular ester groups with straight alkyl chains (a and b) or
methyl ether of triethylene glycol (c) and to study their properties
due to self-aggragation both in solution and in the condensed
phase.27 The exo-annular ester groups would promote self-
association owing to the electron-withdrawing effect.7 The long
(6) For unsubstituted PAMs see: (a) Staab, H. A.; Neunhoeffer, K. Synthesis
1974, 424. (b) Kawase, T.; Ueda, N.; Darabi, H. R.; Oda, M. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 1556-1558. (c) Kawase, T.; Ueda, N.; Oda, M.
Tetrahedron Lett. 1997, 38, 6681-6684.
(7) (a) Zhang, J.; Moore, J. S. J. Am. Chem. Soc. 1992, 114, 9701-9702. (b)
Shetty, A. S.; Zhang, J.; Moore, J. S. J. Am. Chem. Soc. 1996, 118, 1019-
1027. (c) Lahiri, S. L.; Thompson, J. L.; Moore, J. S. J. Am. Chem. Soc.
2000, 122, 11315-11319.
(8) Venkataraman, D.; Lee, S.; Zhang, J.; Moore, J. S. Nature 1994, 371, 591-
593.
(21) For theoretical investigations on π-π interactions: (a) Brawn, N. M. D.;
Swinton, F. L. J. Chem. Soc., Chem. Commun. 1974, 770-771. (b)
Ravishanker, G.; Beveridge, D. L. J. Am. Chem. Soc. 1985, 107, 2565-
2566. (c) Jorgensen, W. L.; Sererance, D. L. J. Am. Chem. Soc. 1990, 112,
4768-4774. (d) Hunter, C. A.; Sanders, J. K. J. Am. Chem. Soc. 1990,
112, 5525-5534. (e) Linse, P. J. Am. Chem. Soc. 1992, 114, 4366-4373.
(f) Herna´ndez-Trujillo, J.; Costas, M. J. Chem. Soc., Faraday Trans. 1993,
89, 2441-2443. (g) Hobza, P.; Selzle, L.; Schlag, E. W. J. Am. Chem.
Soc. 1994, 116, 3500-3506. (h) Chipot, C.; Jaffe, R.; Maigret, B.; Pearlman,
D. A. Kollman, P. A. J. Am. Chem. Soc. 1996, 118, 11217-11224.
(22) For experimental investigations on π-π interactions: (a) Tucker, E. E.;
Christian, S. D. J. Phys. Chem. 1979, 83, 426-427. (b) Desiraju, G.;
Gavezzotti, A. Acta Crystallogr. 1989, B45, 473-482. (c) Cozzi, F.;
Cinquini, M.; Annunziata, R.; Dwyer, T.; Siegel, J. S. J. Am. Chem. Soc.
1992, 114, 5729-5733. (d) Williams, J. H. Acc. Chem. Res. 1993, 26, 593-
598. (e) Laatikainen, R.; Ratilainen, J.; Sebastian, R.; Santa, H, J. Am.
Chem. Soc. 1995, 117, 11006-11010.
(9) Zhang, J.; Moore, J. S. J. Am. Chem. Soc. 1994, 116, 2655-2656.
(10) Shetty, A. S.; Fischer, P. R.; Stork, K. F.; Bohn, P. W.; Moore, J. S. J.
Am. Chem. Soc. 1996, 118, 9409-9414.
(11) (a) Ho¨ger, S.; Enkelman, V. Angew. Chem., Int. Ed. Engl. 1995, 34, 2713-
2716. (b) Ho¨ger, S.; Meckenstock, A.-D.; Mu¨ller, S. Chem. Eur. J. 1998,
4, 2423-2434. (c) Ho¨ger, S.; Meckenstock, A.-D. Chem. Eur. J. 1999, 5,
1686-1691.
(12) Ho¨ger, S.; Bonrad, K.; Mourran, A.; Beginn, U.; Mo¨ller, M. J. Am. Chem.
Soc. 2001, 123, 5651-5659.
(13) Ho¨ger, S.; Enkelman, V.; Bonras, K.; Tschierske, C. Angew. Chem., Int.
Ed. 2000, 39, 2268-2270.
(14) Morrison, D. L.; Ho¨ger, S. Chem. Commun. 1996, 2313-2314.
(15) Nakamura, K.; Okubo, H.; Yamaguchi, M. Org. Lett. 2001, 3, 1097-1099.
(16) (a) Kawase, T.; Hosokawa, Y.; Kurata, H.; Oda, M. Chem. Lett. 1999,
845-846. (b) Hosokawa, Y.; Kawase, T.; Oda, M. Chem. Commun. 2001,
1948-1949.
(17) For related butadiyne-bridged orthocyclophanes: (a) Zhou, Q.; Carroll, P.
J.; Swager, T. M. J. Org. Chem. 1994, 59, 1294-1301. (b) Guo, L.;
Bradshaw, J. B.; Tessiew, C. A.; Youngs, W. J. J. Chem. Soc., Chem.
Commun. 1994, 243-244. (c) Haley, M. M.; Brand, S. C.; Pak, J. J. Angew.
Chem., Int. Ed. Engl. 1997, 36, 836-838. (d) Pak, J. J.; Weakley, T. J. R.;
Haley, M. M. J. Am. Chem. Soc. 1999, 121, 8182-8192. (e) Wan, W. B.;
Brand, S. C.; Pak, J. J.; Haley, M. M. Chem. Eur. J. 2000, 6, 2044-2052.
(18) For a preliminary account of this study, see: (a) Tobe, Y.; Utsumi, N.;
Kawabata, K.; Naemura, K. Tetrahedron Lett. 1996, 37, 9325-9328. The
synthesis and properties of the nonaggregating DBMs with exo-annular
tert-butyl groups have also been reported: (b) Tobe, Y.; Utsumi, N.;
Nagano, A.; Sonoda, M.; Naemura, K. Tetrahedron 2001, 57, 8075-8083.
(19) Synthesis and aggregation of water-soluble DBMs have been reported by
Araki et al.: Ayabe, M.; Araki, K. Unpublished results.
(23) For representative examples of X-ray crystallographic structure analyses:
Huntsman, W. D. In The Chemistry of triple-bonded functional groups (The
Chemistry of functional groups. Supplement C); Patai, S., Rapporport, Z.,
Eds.; John Wiley & Sons: Chichester, 1983; pp 918-980 and references
therein.
(24) (a) Baughman, R. H. J. Appl. Phys. 1972, 43, 4362-4370. (b) Wegner, G.
Pure. Appl. Chem. 1977, 49, 443-454.
(25) (a) Izuoka, A.; Okuno, T.; Ito, I.; Sugawara, T.; Saito, N.; Kamei, S.;
Tohyama, K. Mol. Cryst. Liq. Cryst. 1993, 226, 201-205. (b) Chang, Y.;
Baik, J. H.; Lee, C. B.; Han, M. J. J. Am. Chem. Soc. 1997, 119, 3917-
3918.
(26) For an attempt at polymerization of cyclic diacetylenes leading to highly
ordered three-dimensional structures: Yee, K. C. J. Polym. Sci.: Polym.
Chem. 1979, 17, 3637.
(27) All DBMs polymerized thermally, some in the solid phase and the others
in the liquid crystalline phase. These results will be published elsewhere.
(20) Tobe, Y.; Utsumi, N.; Nagano, A.; Naemura, K., Angew. Chem., Int. Ed.
Engl. 1998, 37, 1285-1287.
9
J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002 5351

A R T I C L E S
Tobe et al.
alkyl groups are necessary to ensure enough solubility to
examine the aggregation behavior in apolar organic solvents
such as chloroform and toluene and the polyether chains would
make 2c and 3c slightly soluble in polar solvents such as
acetone, acetonitrile, and methanol. This paper reports the
general, efficient synthetic method for constructing the macro-
cyclic frameworks of DBMs and their self-aggregation behavior
in apolar and polar solvents. The aggregation properties were
investigated by two methods, one based on the 1H NMR
chemical shifts and the other based on the molal osmotic
coefficient determined by the vapor pressure osmometry, and
the results obtained by both methods were compared.
employed to prepare tetrameric and hexameric DBMs 2a,c and
3a-c by intramolecular coupling of the corresponding open-
chain oligomers 11a,c and 13a-c.
Synthesis of even-numbered linear precursors, dimers (2mers),
linear tetramers (4mers), and linear hexamers (6mers) was
carried out as outlined in Scheme 1.29 Oxidative coupling with
the Eglinton method30 of 5a-c gave doubly protected 2mers
6a-c. Exhaustive deprotection of 6a,b yielded fully deprotected
2mers 7a,b, in 91% and 96% yields, respectively, while partial
deprotection of 6a-c under carefully controlled conditions gave
the corresponding singly protected 2mers 8a-c as the major
products, together with 7a-c and the starting materials 6a-c.
2mers 8a and 8c were coupled under the usual Hay coupling
conditions31 to afford the corresponding linear 4mers 10a and
10c, which were deprotected to give linear 4mers 11a and 11c,
respectively. Bromoacetylenes 9a-c were prepared by the
bromination of singly protected 2mers 8a-c. Hetero coupling
of 7a-c with 2 equiv of the corresponding bromide 9a-c under
the modified literature conditions32 gave linear 6mers 12a-c,
respectively. Removal of the triisopropylsilyl (TIPS) groups gave
linear 6mers 13a-c.
Cyclization reactions were carried out under Eglinton’s
conditions. Thus, reaction of 7b under pseudo-high-dilution
conditions (1.3 × 10-3 M) afforded cyclic 4mer 2b (dimer of
7b) and cyclic 8mer 4b (tetramer of 7b) in 10% and 2% yields,
respectively, after purification by preparative HPLC. However,
we were not able to isolate the corresponding 6mer 3b (trimer
of 7b) under these conditions. Although we have no explanation
for the absence (or very low yield, if any) of the hexameric
DBM, such anomalous behavior has been frequently observed
in the copper-mediated oxidative coupling reactions.17a,18,30b The
cyclization of open-chain 4mers 11a and 11c under high dilution
conditions (1.3 × 10-3 M) yielded the respective cyclic 4mers
2a and 2c and 8mers 4a and 4c (dimers of 11a and 11c),
respectively. Eglinton coupling of linear 6mers 13a-c under
high dilution conditions yielded cyclic 6mers 3a-c in 27%,
15%, and 65% yields, respectively. We also prepared 4mer 2e
having peripheral butyl ester groups by transesterification of
2a with 1-butanol.33 However, since this compound was
sparingly soluble in organic solvents, we did not investigate its
association behavior.
Self-Aggregation Behavior. The self-aggregation behavior
of DBMs and the linear oligomers was investigated quan-
titatively on the basis of the concentration dependence of the
1H NMR chemical shifts and the molal osmotic coefficient (φ)
determined by vapor pressure osmometry (VPO). The ag-
gregation behavior of linear tetramers 11a,c and hexamers 13a,c
was also investigated. In addition, concentration dependence
of the UV spectra of DBMs and the linear oligomers was
studied.
1H NMR spectroscopy is the most frequently used technique
for investigation of aggregation due to stacking interaction of
nucleic acids and aromatic compounds because of its ease,
precision, and the fact that the chemical shift data provide
Results
Synthesis. To minimize the possible number of products, we
first prepared even numbered tetrameric, hexameric, or oc-
tameric DBMs 2b, 3b, and 4b by intermolecular oxidative
coupling of the dimer unit 7b.28 However, since this method
was not satisfactory in view of the low yields of 2b and 4b and
the absence of cyclic hexamer 3b, the stepwise method was
(29) The synthesis of the monomer units 5a-c is described in the Supporting
Information.
(30) (a) Eglinton, G.; Galbraith, A. R. Proc. Chem. Soc. 1957, 350-351. (b)
Behr, O. M.; Eglinton, G.; Galbraith, A. R.; Raphael, R. A. J. Chem. Soc.
1960, 3614-3625.
(28) Although it has been reported that oxidative coupling of 1,3-diethynyl-
benzene gave the unsubstituted hexamer, we were unable to reproduce the
reaction: Ghose, B. N. J. Prakt. Chem. 1982, 324, 1052-1054.
(31) Hay, A. S. J. Org. Chem. 1962, 27, 3320-3321.
(32) Cai, C.; Vasella, A. HelV. Chim. Acta 1995, 78, 2053-2064.
(33) Synthesis of 2e is described in the Supporting Information.
9
5352 J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002

m-Diethynylbenzene Macrocycles
A R T I C L E S
Scheme 1 a
a Key: a, R ) CO2C8H17; b, R ) CO2C16H33; c, R ) CO2(CH2CH2O)3CH3. (a) Cu(OAc)2, pyridine, room temperature; (b) TBAF, THF, H2O, room
temperature; (c) NBS, AgNO3, acetone, room temperature; (d) CuCl, TMEDA, acetone, O2, room temperature; (e) TBAF, THF, room temperature; (f)
Pd2(dba)3‚CHCl3, CuI, i-Pr2NH, benzene, room temperature; (g) Cu(OAc)2, pyridine, benzene, room temperature.
structural information for the aggregates.34 Indeed, the 1H NMR
chemical shifts of the aromatic protons of DBMs 2a-c, 3a-c,
and 4a in CDCl3 showed remarkable concentration depen-
dence.35 For example, the chemical shifts of the two anisoch-
ronous aromatic protons (Ha and Hb) of cyclic 6mer 3a moved
upfield from δ 8.12 to 7.50 (exo-annular proton, Ha) and from
δ 7.84 to 7.30 (endo-annular proton, Hb) as the concentration
increased from 1.53 × 10-4 to 5.36 × 10-3 M at 30 °C. The
upfield shift indicates that the aggregates adopt a structure in
which the macrocyclic frameworks stack in a face-to-face
geometry, as observed frequently in self-association due to π-π
stacking interactions.7,12,15,34 The association constants (K2) were
determined by assuming that monomer-dimer equilibrium is
the predominant process of the self-association, since, in most
cases, the analysis based on the osmometric method agreed with
this assumption as described later. The least-squares curve fitting
to eq 136 was carried out to determine K2, where δm and δd are
of the DBMs were elucidated on the basis of the van’t Hoff
plots (Figure 2), using K2 determined at 4-5 different temper-
atures. The dimerization constants calculated for exo-annular
proton Ha and the corresponding thermodynamic parameters for
the dimerization of DBMs 2a-c, 3a,b,and 4a in CDCl3 are listed
in Table 1, together with those of PAM 14a reported by Moore7b
for comparison.
1 - 8K C + 1
x
2
t
δ ) δm + (δd - δm) 1 +
(1)
(
)
4K2Ct
chemical shifts of the monomer and the dimer, respectively,
and K2 and Ct are dimerization constant and total concentration
of the substrate, respectively. The dilution curves for 3a and
3b are shown in Figure 1. In most cases, the association
constants calculated from exo-annular proton Ha gave slightly
larger K2 than those calculated from the data for endo-annular
proton Hb.37 The thermodynamic parameters for the dimerization
It has been demonstrated that solvophobic interactions of
organic solvents exert a crucial effect on the dimerization of
rigid cyclophanes38 and inclusion of an aromatic guest in
diphenylmethane-based cyclophanes.39 Self-association behavior
of PAMs and related macrocycles is also sensitive to the polarity
of the solvent.7c,12 Moreover, a remarkable solvophobic effect
on the intramolecular association of linear oligomers of m-
(34) Martin, R. B. Chem. ReV. 1996, 96, 3043-3064.
(35) However, the 1H NMR chemical shifts of 8mer 3c having the polyether
chains did not exhibit concentration dependence in CDCl3.
(36) Horman, I.; Drewx, B. HelV. Chim. Acta 1984, 67, 754-764.
(37) All NMR and osmometric data and some UV spectra are listed in the
Supporting Information.
(38) Cram, D. J.; Choi, H.-J.; Bryant, J. A.; Knobler, C. B. J. Am. Chem. Soc.
1992, 114, 7748-7765.
(39) Smithrud, D. B.; Diederich, F. J. Am. Chem. Soc. 1990, 112, 339-343.
9
J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002 5353

A R T I C L E S
Tobe et al.
Table 2. Association Constants and the Corresponding Free
Energies for Self-Aggregation of DBMs 2c and 3c in Polar
Solvents at 303 Ka
DBM
solvent
KE (M-1
)
∆G (kJ‚mol-1
)
2c
CD3COCD3
19100 ( 1900
184 ( 14
1150 ( 70
1990 ( 150
3170 ( 260
8070 ( 1190
27000
111 ( 3
307 ( 6
2260 ( 140
13000 ( 1000
150000
102 ( 4
316 ( 16
1050 ( 80
2610 ( 550
580000
-24.8 ( 0.3
CD3CN/CDCl3 ) 3/7
CD3CN/CDCl3 ) 5/5
CD3CN/CDCl3 ) 6/4
CD3CN/CDCl3 ) 7/3
CD3CN/CDCl3 ) 8/2
CD3CNb
CD3OD/CDCl3 ) 2/8
CD3OD/CDCl3 ) 3/7
CD3OD/CDCl3 ) 5/5
CD3OD/CDCl3 ) 7/3
CD3ODb
CD3OD/CDCl3 ) 1/9
CD3OD/CDCl3 ) 2/8
CD3OD/CDCl3 ) 3/7
CD3OD/CDCl3 ) 4/6
CD3ODb
-26
-30
-33
3c
a Determined on the basis of the infinite association model by the NMR
method. b Extrapolated to pure solvent from the plot of ∆G vs solvent
composition shown in Figure 3.
Figure 1. Concentration dependence of 1H NMR chemical shifts for exo-
annular aromatic protons (Ha) and endo-annular aromatic protons (Hb) of
3a and 3b in CDCl3 at 30 °C.
nitrile, and methanol on the association of DBMs 2c and 3c
bearing ester groups of triglyme monomethyl ether. Unfortu-
nately, because of significant peak broadening, the dilution
experiment was not carried out for 3c in CD3COCD3 and
CD3CN. Since 2c and 3c were not soluble in neat CD3CN and
CD3OD at the NMR concentration level, the NMR experiments
were carried out in mixed solvent systems, CD3CN/CDCl3 and
CD3OD/CDCl3, of several different compositions. Although the
chemical shift data fit nicely to the theoretical curves derived
from eq 1 for dimerization, the data were analyzed by the infinite
(isodesmic) association model, assuming the formation of higher
aggregates shown in Scheme 1, since the VPO analysis clearly
indicated the formation of higher aggregates as described later.
The least-squares curve fitting was thus carried out with eq 2 ,
1 - 4K C + 1
x
E
t
δ ) δm + (δa - δm) 1 +
(2)
(
)
2KECt
assuming K2 ) K3 ) K4 ) ... ) Kn ) KE, in which δa is the
average chemical shift of the aggregates.41 As expected, 2c and
3c exhibited an increasing tendency to self-associate with an
increasing portion of the polar solvent relative to CDCl3 (Table
2), indicating that the aggregate formation was enhanced by
the solvophobic interactions. The linear relationship between
the free energy of association versus the composition of the
polar solvent in CDCl3 was observed as shown in Figure 3,
from which KE in neat CD3CN and CD3OD was estimated by
extrapolation. Similar effect of the solvent composition on the
conformation of linear phenylacetylene oligomers was reported.40b
Comparison of the association constants summarized in Table
2 with those in Table 1 clearly indicates that the self-aggregation
Figure 2. van’t Hoff plots for self-association of DBMs 2a, 3a, and 4a in
CDCl3.
Table 1. Dimerization Constants and Thermodynamic Parameters
for Self-Aggregation of DBMs 2a-c, 3a-c, and 4a and PAM 14a
in CDCl3 at 303 Ka
compd
K (M-1
)
∆G (kJ‚mol-1
)
∆H (kJ‚mol-1
)
∆S (J‚mol-1‚K-1
)
2
2a
2b
2c
3a
3b
3c
27.9 ( 0.7
28.7 ( 2.4
19.9 ( 1.3
173 ( 17
150 ( 4
42.7 ( 1.4
22.8 ( 0.9
39
-8.38 ( 0.07
-8.46 ( 0.22
-7.53 ( 0.17
-13.0 ( 0.3
-12.6 ( 0.1
-9.45 ( 0.08
-7.88 ( 0.10
-9.3
-21 ( 1
-24 ( 1
-15 ( 1
-52 ( 2
-39 ( 2
-31 ( 1
-34 ( 1
-21 ( 0.8
-42 ( 2
-50 ( 3
-26 ( 2
-128 ( 6
-88 ( 6
-71 ( 3
-86 ( 4
-39 ( 3
4a
14ab
(40) (a) Nelson, J. C.; Saven, J. G.; Moore, J. S.; Wolynes, P. G. Science 1997,
277, 1793-1796. (b) Prince, R. B.; Saven, J. G.; Wolynes, P. G.; Moore,
J. S. J. Am. Chem. Soc. 1999, 121, 3114-3121.
(41) Since all the NMR dilution data fit nicely to the assumptions of both eqs
1 and 2, we did not examine the model in which the association constant
for dimerization (K2) is different from those for the formation of higher
aggregates (K2 * KE).34 A case has recently been found in the aggregation
of an imine-containing PAM, in which NMR data did not fit the K2 ) KE
model but fit the K2 * KE model; personal communication from Prof.
Moore.
a Determined on the basis of the dimerization model by the NMR method.
b Calculated from the data in ref 7b.
phenylacetylene, forming helical structures in polar solvents,
has been reported by Moore and co-workers.7c,40 We examined
the effect of polar solvents such as deuterated acetone, aceto-
9
5354 J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002

m-Diethynylbenzene Macrocycles
A R T I C L E S
Table 3. Association Constants and Thermodynamic Parameters for Self-Aggregation of DBMs 2a,c, 3a,c, and 4a and PAM 14b in
Aromatic Solvents at 303 Ka
compd
solvent
KE (M-1
)
∆G (kJ‚mol-1
)
∆H (kJ‚mol-1
)
∆S (J‚mol-1‚K-1
)
2a
toluene-d8/CDCl3 ) 1/9
toluene-d8/CDCl3 ) 2/8
toluene-d8/CDCl3 ) 3/7
toluene-d8/CDCl3 ) 4/6
toluene-d8/CDCl3 ) 5/5
63.6 ( 4.1
75.3 ( 6.2
83.9 ( 4.4
98.9 ( 12.2
115 ( 26
240
b
toluene-d8
-14
c
2c
3a
toluene-d8
toluene-d8
163 ( 11
-12.8 ( 0.2
-31
-28 ( 1
-120 ( 10
-100 ( 10
-90 ( 3
-51 ( 2
-300 ( 30
-260 ( 30
-210 ( 10
-330 ( 50
c
210000
19000
30000
c
o-xylene-d10
toluene-d8
-25
-26
-14.7 ( 0.6
-18
c
3c
4a
14b
toluene-d8
benzene-d6
340 ( 70
1200
-120 ( 20
d
a Determined on the basis of the infinite association model by the NMR method. b Extrapolated to pure solvent from the plot of ∆G vs solvent composition.
c Extrapolated from the association constants determined at higher temperatures. d Data from ref 7c at room temperature (not specified).
in toluene (vide infra).37 In the case of 3a and 3c, it was not
possible to measure the precise chemical shifts at 30 °C owing
to the significant peak broadening. Therefore, the measurements
were carried out at several different temperatures to elucidate
thermodynamic parameters in order to make direct comparison
of the association propensities possible (see Supporting Infor-
mation for the van’t Hoff plots). Since DBMs 2a was much
less soluble in toluene-d8 than in CDCl3, its association con-
stant at 30 °C was estimated by extrapolating from the data in
toluene-d8/CDCl3 mixtures of several different compositions
(see Supporting Information for the plot of ∆G vs solvent
composition), as in the case of the dilution experiments in polar
solvents. The results are summarized in Table 3. As shown,
the association was remarkably enhanced in the aromatic
solvents compared to that in CDCl3, which we had not expected
at the outset.
Besides the NMR method, self-aggregation of solutes is
frequently analyzed on the basis of the molal osmotic coeffi-
cient (φ) determined by vapor pressure osmometry (VPO).34
Since the mosmotic measurements are colligative and count
Figure 3. Relationship between ∆G of aggregation of DBMs 2c and 3c
moles in solution, this method may sense aggregates which do
and the composition of the polar solvents in CH3CN/CHCl3 and CH3OH/
not exhibit upfield shift in the NMR spectra. Moreover, there
CHCl3 systems.
exist problems in the use of NMR for the investigation of higher
aggregation behavior due to the oversimplification of the
of DBMs is remarkably enhanced by the solvophobic interac-
tions.42
analysis models, significant line broadening, and the concentra-
tion limits arising from solubility and sensitivity. To check the
reliability of the results obtained by the NMR experiments, we
examined the association behavior of 2a (CHCl3, 40 °C), 2c
(CHCl3, 40 °C; toluene, 60 °C; acetone 40 °C), 3a (CHCl3, 40
°C), and 3c (CHCl3, 40 °C; toluene, 60 °C; acetone 40 °C) by
VPO.
In the VPO measurements, osmotic coefficient (φ), activity
coefficient (γ), and monomer concentration (m1) were calculated
for each sample solution with use of the coefficients of the sixth
order polynominal, which were obtained by curve fitting, using
the experimental data, stoichiometric molal concentration (ms),
and colligative molal concentration (mc), to the polynominal
equation (see the Experimental Section).37 For analysis of the
data, we employ two models, the equal K model,44 which
assumes K2 ) K3 ) K4 ) ... ) Kn ) KE and Kn+1 ) 0, and the
K2 * K model,45 which assumes K2 * K together with a few
different hypotheses regarding the relationship between the
association constants for the formation of higher aggregates. In
As for the effect of aromatic solvents on the association
behavior of PAMs, Moore reported that PAM 14a did not show
concentration-dependent chemical shifts in benzene-d6.7b How-
ever, in the subsequent paper, its derivative 14b bearing
polyether chains was reported to aggregate more strongly in
benzene-d6 than in CDCl3.7c Enhancement of self-aggregation
in an aromatic solvent was also noted by Yamaguchi and co-
workers for chiral macrocycles.15 We had also been aware of
the unexpected aggregation behavior of DBMs in aromatic
solvents.43 Because of the low solubility in benzene, the NMR
dilution experiments were undertaken in toluene-d8 for 2a,c,
3a,c and 4a. For 3a, self-aggregation was also examined in
p-xylene-d10. The data were treated with eq 2, assuming the
infinite association, since the VPO analysis for 2c and 3c
strongly indicated the formation of aggregates larger than dimers
(42) Although the definitions of K2 and KE are different, comparison of eqs 1
and 2 indicates KE can formally be regarded to be twice as large as K2.
(43) Araki, S.; Adachi, K.; Nagano, A.; Kawabata, K.; Utsumi, N.; Sonoda,
M.; Hirose, K.; Tobe, Y. XI International Symposium on Supramolecular
Chemistry, July 30 to August 4, 2000, Fukuoka, Japan: Abstract of papers,
p 411.
(44) Ts’o, P. O.; Chan, S. I. J. Am. Chem. Soc. 1964, 86, 4176-4181.
(45) Rossotti, F. J. C.; Rossotti, H. J. Phys. Chem. 1961, 65, 930-934.
9
J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002 5355

A R T I C L E S
Tobe et al.
and acetone did not fit the theoretical curves based on the equal
K model. For these case, extremely large KE (>104) was
determined by the NMR method. It turned out that the data fit
to the K2 * K model developed by Rossotti.45 In this model,
four different hypotheses I-IV are considered, which relate the
association constants for the formation of higher aggregates as
shown in eqs 4, 6, 8, and 10, respectively. For each model, T,
defined in eq 12, is related to m1 in eqs 5, 7, 9, and 11,
respectively.
Hypothesis (I):
K3 ) K4 ) K5 ) ... ) Kn ) KE
(4)
∞
Km1(2 - Km1)
T ) n(Km )n-1
)
(5)
∑
1
(1 - Km1)2
2
Hypothesis (II):
1
2
2
3
3
4
n - 2
n - 1
K3 ) K, K4 ) K, K5 ) K, ..., Kn )
K (6)
Figure 4. Master curves of [1 - (Km1)n]/(1 - Km1) vs log Km1 and data
plots for determination of association constants of DBM 2c in chloroform
at 40 °C, toluene at 60 °C, and acetone at 40 °C according to the equal K
model.
n-1
∞
(Km1)
Km1
1 - Km1
T )
)
- ln(1 - Km1)
(7)
∑
n - 1
2
Table 4. Association Constants for Self-Aggregation of DBMs
2a,c and 3a,c Determined by the VPO Method on the Basis of the
Equal K Model
Hypothesis (III):
1
1
3
1
4
1
K3 ) K, K4 ) K, K5 ) K, ..., Kn )
K (8)
2
n - 1
-1
DBM
solvent
temp (°C)
K or KE (n) (M
)
2
2a
2c
CHCl3
CHCl3
toluene
CHCl3
CHCl3
40
40
60
40
40
K2: 146
K2: 45
KE: 139 (n ) ∞)
KE: 350 (n ) 6)
K2: 428
n-1
∞
n(Km1)
) eKm (1 + Km1) - 1
(9)
1
T )
∑
3a
3c
(n - 1)!
2
Hypothesis (IV):
the first model, on the basis of eq 3, which relates φ/γ to K and
m1, the master curves are plotted for [1 - (KEm1)n]/(1 - KEm1)
versus log KEm1 for different n (n ) 1, 2, 3, 4, 5, 6, and ∞) of
the monomer units in the largest aggregate present in a solution.
3
2
4
3
n - 1
n - 2
K3 ) 2K, K4 ) K, K5 ) K, ..., Kn )
K (10)
∞
2Km1
(1 - Km1)3
T ) n(n - 1)(Km )n-1
)
(11)
∑
1
1 - (KEm1)n
1 - KEm1
2
φ
)
(3)
γ
K(ms - m1)
T )
(12)
Fitting a plot of the experimentally obtained φ/γ versus log
m1 to the plots yielded log K as the translation along the
abscissa.44 As examples, Figure 4 shows plots of the data for
2c in chloroform, toluene, and acetone. As can be seen, the
data in CHCl3 fit the master curve of n ) 2, yielding K2 )
44.8 M-1, while the data in toluene fit the curve of n ) ∞,
furnishing KE ) 139 M-1. Importantly, this analysis allowed
the elucidation of the mode of association, i.e., dimerization or
further aggregation. The results are summarized in Table 4. The
aggregation of 2a, 2c, and 3c in chloroform was confirmed to
form dimers mainly, while 2a forms higher aggregates (n ) ∞)
in toluene. The aggregation of 3a in chloroform is an intermedi-
ate case, which forms up to hexamer (n ) 6) with a comparable
association constant to that obtained by the NMR method.42
However, taking into account the difference between the
temperature of the measurements, the association constants
determined by VPO method are, in general, considerably larger
than the corresponding data determined by NMR.46
K2m1
The master curves are plotted for log T versus log Km1 by
using eqs 5, 7, 9, and 11. Fitting a plot of experimentally
obtained log [(ms - m1)/m1] versus log m1 to the master plots
gave log K as the translation along the abscissa and log(K/K2)
as the translation along the ordinate. As an example, Figure 5
shows a plot for 2c in acetone. The data fit best to model IV
among the hypotheses I-IV, furnishing K2 ) 552 M-1 and K
) 3370 M-1. Since, in general, the data fit best to hypothesis
IV, K2 and K shown in Table 5 were calculated with this
model.47 Table 5 also lists the higher aggregation constants, K3,
K4, and K∞ calculated from eq 10, to show clearly the respective
binding constants. At first glance, it is rather surprising that,
compared to the large KE determined by NMR, the association
constants (K2 and K) determined by the K2 * K model are
(46) The problems which would reduce the reliability of the analysis based on
NMR would not apply in the monomer-dimer models. Accordingly, we
do not have explanations at this moment for the relatively large discrepancy
between the association constants determined by NMR and VPO.
As shown in Figure 4, the plot for 2c in acetone did not fit
any of the master curves. Similarly, the data for 3c in toluene
9
5356 J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002

m-Diethynylbenzene Macrocycles
A R T I C L E S
To elucidate the effect of the macrocyclic structure on the
aggregation property of DBMs 2a,c and 3a,c, the aggregation
behavior of the corresponding linear tetramers 11a,c and
hexamers 13a,c was also investigated by NMR and VPO. In
CDCl3 and toluene-d8, none of the linear oligomers exhibit
concentration dependence of the chemical shifts, indicating the
critical importance of the macrocyclic structure of DBMs for
the self-aggregation in apolar and aromatic solvents. On the
other hand, 11c and 13c aggregate in polar solvents although
much less strongly than DBMs.49 The association constants K2
and KE were determined by NMR and VPO in CD3COCD3 (or
acetone), CD3CN, and CD3OD/CDCl3 ) 7/3. Owing to the
broadening of the NMR peaks, the data for 13c in CD3CN
were not obtained. Since the VPO data of 4mer 11c fit well
the dimerization curve (n ) 2) of the equal K model, the
NMR data were analyzed assuming the dimer formation (eq
1). On the other hand, the VPO data of 6mer 13c fit the
master plot of infinite aggregation (n ) ∞) for the equal K
model. This was confirmed by the K2 * K model, which gave
comparable K2 and K. The NMR data for 13c were, therefore,
treated with the infinite association model (eq 2). The results
are listed in Table 6. Comparison of the data in Table 6 with
those in Tables 2, 4, and 5 indicates that the aggregation pro-
pensity of the linear oligomers 11c and 13c is much less than
that of the corresponding DBMs 2c and 3c. It is worth noting,
however, that the aggregation of liner oligomers is observed
only in polar solvents, indicating that the aggregation is
solvophobically driven.
Figure 5. Master curves of log T vs log Km1 and a data plot for
determination of the association constant of DBM 2c in acetone at 40 °C
according to the K2 * K model.
Table 5. Association Constants for Self-Aggregation of DBMs 2c
and 3a,c Determined by the VPO Method on the Basis of the K2
*
K Modela
temp
(°C)
K
K
K
K
4
K∞
(M-1
2
3
DBM
solvent
(M-1
)
(M-1
)
(M-1
)
(M-1
)
)
2c
toluene
acetone
CHCl3
toluene
acetone
60
40
40
60
40
111
552
566
350
1180
104
3370
342
2580
3170
208
6740
682
5160
6340
156
5060
513
3870
4760
104
3370
342
2580
3170
The electronic spectrum is another proof for the investigation
of π-π stacking interactions, since electronic interactions
between the chromophores cause spectral changes, in most cases
a hypochromic shift.50 The absorption spectra of DBMs 2c and
3c and liner oligomers 11c and 13c were recorded in CHCl3,
CH3CN/CHCl3 ) 9/1 and CH3OH/CHCl3 ) 9/1 in a concentra-
tion ranging from ca. 1 × 10-6 to 5 × 10-5 M. Under these
conditions, little concentration dependence was observed in the
spectra of 2c and 3c in CHCl3 and those of 11c and 13c in all
solvents. On the other hand, the spectra of 2c and 3c in polar
solvents exhibit remarkable hypochromic shift with slight
hypsochromic shift of the abosorption bands at 280-340 nm,
as a result of strong aggregation. Figure 6 shows the spectra of
2c in CH3OH/CHCl3 ) 9/1 as an example.37
3a
3c
a Analyzed on the basis of the hypothesis (IV).
considerably smaller. However, when taking the difference
between the temperature into account, it becomes apparent that
the discrepancy is moderate. Namely, with the difference of 30
°C between the temperature of the NMR and VPO measure-
ments,48 one can roughly estimate K3 ) 16000 M-1 and K4 )
12000 M-1 and so on for the aggregation of 2c in acetone, which
are still smaller than but in the same order as that determined
by the NMR method (KE ) 28000 M-1). The data for 2c in
toluene and those for 3a in CHCl3 were also analyzed by this
model to give the results consistent with those obtained by the
equal K model, i.e., K2 ≈ K. It should be pointed out that the
analysis of self-aggregation by VPO thus revealed the unique
aggregation behavior of DBMs in acetone and toluene, which
was not elucidated by the NMR method.
Discussion
General Features of Self-Association of DBMs. Owing to
the limitations due to solubility, peak broadening in NMR, and
the sensitivity limit of the VPO reading particularly when
extremely large aggregates were formed, it was not possible to
obtain the data which would allow direct comparison of all
association constants determined by different methods under
(47) The difference between hypotheses I and IV are the relationship between
the higher association constants, either constant (I), gradually increasing
(II) to K∞ ) K, rapidly dropping to K∞ ) 0 (III), or gradually decreasing
K∞ ) K (IV). The difference must be due to the steric and electrostatic
interactions in the step-by-step formation of higher aggregates as discussed
in ref 47a. In all the cases we examined, hypothesis II gave the worst fit
to the experimental data, excluding the possibility of this model. Although
the model IV gave the best fit in most cases, a similarly good fit was
obtained with hypotheses I and III in some cases. With the available data,
it is not possible to explain why hypothesis IV gave better fits. However,
hypothesis IV also turns out to be satisfactory in the analysis of aggreagation
reported in the literature using Rossotti’s model. For example, see: (a)
Ghosh, A. K.; Mukerjee, P. J. Am. Chem. Soc. 1970, 92, 6408-6412. (b)
Klofutar, C.; Paljk, S.; Abram, V. J. Chem. Soc., Faraday Trans. 1993,
89, 3065-3069.
(49) The monomer unit 15 did not show concentration dependence of the NMR
chemical shift in the same polar solvents, suggesting that at least a few
diethynylbenzene units are necessary to effect aggregation.
The preparation of 15 is described in the Supporting Information.
(50) For example: Diederich, F. Cyclophanes; Stoddart, J. F., Ed.; The Royal
Society of Chemistry: Cambridge, 1991; pp 19-21.
(48) By lowering the temperature by 10 °C, we assume a 1.5 times increase of
association constants.
9
J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002 5357

A R T I C L E S
Tobe et al.
Table 6. Association Constants for Self-Aggregation of Linear Oligomers 11c and 13c in Polar Solvents at 303 K
compd
solvent
method (model for data analysis)
K , KE (n), or K (M-1
)
2
11c
CD3COCD3
acetone
NMR (dimer model)
VPO (equal K model)
K2: 9.12 ( 0.38
K2: 38.2
CD3CN
NMR (dimer model)
NMR (dimer model)
NMR (infinite association model)
VPO (equal K model)
VPO (K2 * K model))
NMR (infinite association model)
K2: 22.0 ( 6.9
K2: 28.0 ( 1.9
KE: 86.1 ( 5.6
KE: 107 (n ) ∞)
K2: 75.6, K2: 105
KE: 841 ( 43
CD3OD/CDCl3 ) 7/3.
CD3COCD3
acetone
acetone
CD3OD/CDCl3 ) 7/3
13c
reasons. (1) There is no functional group capable of forming a
hydrogen bond. (2) Polar solvents remarkably enhance the
propensity toward aggregation. (3) In the NMR spectra, the
upfield shift of the aromatic protons was observed due to the
influence of the ring current of the aromatic rings of the
neighboring molecules. Only the aromatic protons show remark-
able concentration-dependent chemical shifts. The chemical
shifts of the aliphatic protons did not move in the same
concentration range. (4) Hypochomic shift of the UV spectra
was observed in polar solvents, when the degree of aggregation
was high. (5) Linear oligomers 11a,c and 13a,c did not show a
concentration-dependent chemical shift in apolar solvents.
Polyether esters 11c and 13c aggregate only weakly in polar
solvents. Accordingly, the cyclic array of aromatic rings is
indispensable for the self-association.
Self-Association in Chloroform. As mentioned above,
DBMs and PAMs showed qualitatively similar self-association
behavior in CDCl3. The common features are as follows: (1)
As shown in Table 1, the dimerization of DBMs in CDCl3 is
enthalpy driven (∆H < 0) and entropy opposed (∆S < 0). (2)
In the series of DBMs having the same functional group and a
different ring size, cyclic 6mers 3a-c show the most negative
∆H values, indicating that cyclic 6mers self-associate most
strongly due to the π-π stacking interaction. The smaller
association force of 4mers 2a-c than that of 3a-c is ascribed
to the smaller number of possible π-π stacking interaction sites.
The less negative ∆H value of 8mer 4a than that of 6mer 3a is
ascribed to the nonplanar geometry and the conformational
flexibility of 4a. Molecular modeling51 for model hydrocarbon
4d with the AM1 method indicates the presence of a number
of stable nonplanar conformers within a small energy difference
(<0.2 kJ‚mol-1).18b On the contrary, the calculated geometry
of the parent 6mer 3d adopts a planar conformation.18b (3) The
effect of the length of the alkyl chain is negligible. On the other
hand, the aggregation propensity of the polyether esters 2c and
3c is lower than that of the alkyl esters, presumably because
the polyether moieties are better solvated by chloroform than
the hydrocarbon chain. In other words, the nonpolar hydrocarbon
chain may facilitate, to some extent, the aggregation of DBMs
due to solvophobic interactions between the alkyl chains and
the solvent.
Figure 6. UV spectra of DBM 2c in CH3OH/CHCl3 ) 9/1 at 30 °C.
different conditions. Qualitatively the data derived from the
NMR method agree with those from VPO. However, there exist
discrepancies between the data derived by the two methods;
the VPO method tends to give larger association constants than
those from the NMR method when the extent of aggregation is
small.46 On the contrary, VPO tends to give smaller association
constants than those from NMR when large aggregates are
formed with large association constants. The latter may be due
to the problem in the use of NMR for analysis of higher
aggregation as described before.
The analysis of self-aggregation by VPO revealed unique
aggregation behavior of DBMs 2c in acetone and 3c in toluene
and acetone, which was not obtained from the NMR method as
long as the infinite association model was used. Namely, the
osmotic data can only be analyzed by the K2 * K model, and
it appeared the association constants for infinite association (K3,
K4, and K∞) are several times larger than the dimerization con-
stant (K2), indicating that the formation of the higher aggregates
is more favorable than that of the dimers due to the solvophobic
interactions. In other words, the aggregation is enhanced by the
formation under these conditions, suggesting a nucleation
mechanism. These results demonstrate that the methods and
models for analysis are critical for investigation of aggregations
particularly when extensive aggregation takes place.
Comparison of the association constants and thermodynamic
parameters in CDCl3 of DBM 3a and those of PAM 14a, having
the same number of aromatic rings, reveals several remarkable
differences between their association properties. (1) First, the
association of 3a is enthalpically more favored than that of 14a.
For example, the association constant of 3a is more than four
times larger than that of 14a at 30 °C. This implies that the
butadiyne units are a more effective activator for π-π stacking
Like self-association of PAMs7 and the related macro-
cycles,12,15 the driving forces of self-association of DBMs are
attributed to π-π stacking and solvophobic interactions between
the planar macrocyclic framework of DBMs for the following
(51) SPARTAN, verson 5.0 program package; Wavefunctin, Inc.: Irvine, CA.
9
5358 J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002

m-Diethynylbenzene Macrocycles
A R T I C L E S
because such interactions between the solute and solvent would
prevail over those between solutes.53 On the contrary, like PAM
14b7c and the related chiral macrocycles,15 DBMs 2a,b and 3a,c
turned out to self-associate in aromatic solvents, toluene and
o-xylene. Moreover, whereas the association constants of 4mers
2a and 2c in toluene are in the same order of magnitude as
those in chloroform, 6mers 3a and 3c exhibit much larger
association constants in aromatic solvents than in chloroform.
The ∆H term in toluene-d8 determined by the NMR method is
about twice as favorable as that in CDCl3. Although the role of
the aromatic solvents in the present system is not clear, the large
cavities of 6mers, through which the solvent molecules can
penetrate freely, may be responsible for this unusual behavior.
Conclusion
Figure 7. Calculated electrostatic potentials on the van der Waals molecular
surfaces of model compounds 3d and 14c, based on HF/3-21G single point
calculations for the geometries optimized by semiempirical AM1 calcula-
tions. The potentials are drawn in the same color scale, with red indicating
more negative electrostatic potentials and blue more positive potentials.
We synthesized the novel rigid diethynylbenzene macrocycles
(DBMs), which showed self-association properties in solution
to form dimers and higher aggregates owing to π-π stacking
interactions. The size of the DBMs is critical to this behavior;
cyclic hexamers having planar and rigid macrocyclic framework
exhibit the strongest tendency to self-associate. By comparing
the association behaviors of the DBMs and PAMs, it is revealed
that the buta-1,3-diyne units promote π-π stacking interactions
effectively owing to their electron-withdrawing effect. In polar
solvents, the association of DBMs is much enhanced due to
the solvophonic effect compared to that in an apolar solvent
(chloroform), and in aromatic solvents, DBMs aggregate more
strongly than in chloroform for a reason that is not clarified. In
these solvents, the degree of aggregation is large enough to form
nanotubular aggregates. The analysis of self-aggregation by VPO
revealed unique aggregation behavior of DBMs in acetone and
toluene, suggesting a nucleation mechanism.
interactions than the acetylene units. Since it has been well-
documented that an electron-withdrawing group activates an
aromatic ring toward π-π stacking interactions,7,22c the activat-
ing effect of the butadiyne units is attributed to the strong
electron-withdrawing effect of this group. Indeed, as shown in
Figure 7, the electrostatic potentials of the model compound
3d and 14c on the basis of the molecular orbital calculations
(AM1//3-21G) indicate that the aromatic rings of 3d are more
electron-deficient than those of 14c. (2) DBM 6mers 3a-c and
8mer 4a show large negative ∆S values compared to PAM 14a.
In general, the reported entropy of association of two compo-
nents in an apolar solvent ranges from about -46 to -63
J‚mol-1‚K-1 52
Unusually large negative ∆S values observed
.
for 3a,b and 4a may be indicative of some ordering of the alkyl
chains in solution due to solvophobic interactions. (3) Moore
reported that the values of ∆H and ∆S could not be obtained
for flexible PAM having seven phenylacetylene units (hep-
tameric PAM) by the van’t Hoff plot because of the change in
proportion of conformers at different temperatures.7b On the
contrary, DBM 8mer 4a exhibits the straight van’t Hoff plot as
shown in Figure 2. Probably the population of the conformers
of 4a does not change significantly over the temperature range
of the measurements.
Experimental Section
General. 1H NMR spectra were recorded at 270, 300, or 400 MHz
and 13C NMR spectra at 67.5, 75, or 100 MHz on a JEOL JNM-AL-
400, a Varian Mercury 300, or a JEOL JNM-GSX-270 in CDCl3 and
with Me4Si or residual solvent as an internal standard at 30 °C unless
otherwise stated. IR spectra were recorded as a KBr disk or a neat
film on a JASCO FTIR-410 and Raman spectra on a Bio-Rad Ft-Raman
II spectrometer. Electronic spectra were recorded on a HITACHI
U-3310 spectrometer. Mass spectral analyses were performed on a
JEOL JMS-DX303HF spectrometer or a JEOL JMS-700 spectrometer
for EI, FAB, and FD ionization. For most of the MALDI-TOF mass
measurements, a Shimadzu/Kratos AXIMA-CFR spectrometer was
employed. Elemental analyses were performed on a Perkin-Elmer
2400II analyzer. Vapor pressure osmometry was conducted with a
HITACHI 117 vapor pressure osmometer. Melting points were
measured with a hot-stage apparatus and are uncorrected. Column
chromatography and TLC were performed with Merck silica gel 60
(70-230 mesh ASTM) and Merck silica gel 60 F254, respectively. Flash
chromatography was performed with Fuji Silysia silica gel (BW-300).
Preparative HPLC separation was undertaken with a JAI LC-908
chromatograph using 600-mm × 20-mm JAIGEL-1H and 2H GPC
columns with CHCl3 as an eluent. All reagents were obtained from
commercial suppliers and used as received. Solvents were dried (drying
agent in parentheses) and distilled prior to use: THF (LiAlH4 followed
by sodium benzophenone ketyl), benzene (CaH2), CDCl3 (P4O10), and
Self-Association in Polar Solvents and in Aromatic Sol-
vents. As expected, DBMs self-associate in polar solvents much
more strongly than in an apolar solvent, chloroform. An estimate
of the association constant (KE) of 2c and 3c for infinite
association in methanol extends to the order of 106 M-1
according to the analysis based on NMR. Since the degree of
aggregation is estimated to be considerably large, nanotubular
aggregates of DBMs must be formed.19 Although precise
comparison is not possible, the association constant of DBM
3c in polar solvents seems to be larger than that of PAM 14b,
due to the larger solvophobic area of the former. It should be
pointed out that linear oligomers 11c and 13c aggregate only
in polar solvents. This suggests that, owing to solvophobic
interactions, the backbone of these compounds may coil to adopt
a pseudocyclic conformer that induces aggregation by stacking
interactions like those of DBMs.
(52) Bryant, J. A.; Ericson, J. L.; Cram, D. J. J. Am. Chem. Soc. 1990, 112,
1255-1256.
(53) Sanders, G. M.; van Dijk, M.; van Veldhuizen, A.; van der Plas, H. C.;
Hofstra, U.; Schaafsma, T. J. J. Org. Chem. 1988, 53, 5272-5281.
It is intuitively reasonable to assume that aromatic solvents
suppress π-π stacking interactions between aromatic solutes
9
J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002 5359

A R T I C L E S
Tobe et al.
pyridine (KOH). All reactions which required anhydrous conditions
were conducted under a dry nitrogen atmosphere.
with water and brine and dried over MgSO4. Removal of the solvent
in vacuo afforded crude products. Purification by flash chromatography
afforded 7a (763 mg, 39%), 8a (1.25 g, 50%), and the starting material
6a (207 mg).
7a: white solid; mp 89-90 °C; 1H NMR (400 MHz, CDCl3) δ 8.15
(t, J ) 1.5 Hz, 2H), 8.13 (t, J ) 1.5 Hz, 2H), 7.79 (t, J ) 1.5 Hz, 2H),
4.33 (t, J ) 6.8 Hz, 4H), 3.16 (s, 2H), 1.77 (quintet, J ) 6.8 Hz, 4H),
1.46-1.30 (m, 20H), 0.89 (t, J ) 6.8 Hz, 6H); 13C NMR (100 MHz,
CDCl3) δ 164.71 (s), 139.33 (d), 133.64 (d), 133.47 (d), 131.40 (s),
123.19 (s), 122.33 (s), 81.46 (d), 80.18 (s), 79.16 (s), 74.93 (s), 65.86
(t), 31.84 (t), 29.28 (t), 29.21 (t), 28.69 (t), 26.05 (t), 22.70 (t), 14.15
TIPS-Protected 2mer 6a. To a solution of Cu(OAc)2 (8.28 g, 45.6
mmol) in pyridine (800 mL) was added a solution of 5a29 (9.75 g,
22.2 mmol) in pyridine (100 mL). After the solution was stirred for 4
h, additional Cu(OAc)2 (4.14 g, 22.8 mmol) was added. After an
additional 19 h of stirring, the solvent was removed and the green
residue was passed through a short plug of alumina. After removed of
the solvent under reduced pressure, purification by flash chromatog-
raphy afforded 6a as a pale yellow solid (8.32 g, 89%): mp 52-54
1
°C; H NMR (400 MHz, CDCl3) δ 8.09 (d, J ) 1.5 Hz, 4H), 7.77 (t,
(q); IR (KBr) 3257, 2114, 1720, 1307, 1223, 897, 771, 709, 669 cm-1
;
J ) 1.5 Hz, 2H), 4.33 (t, J ) 6.8 Hz, 4H), 1.78 (quintet, J ) 6.8 Hz,
MS (FAB) m/z 562 (M+), 433 (M+ - OC8H17), 321(M+ + H - OC8H17
- C8H17), 248 (M+ - (CO2C8H17)2). Anal. Calcd for C38H42O4: C,
81.11; H, 7.52. Found: C, 81.01; H, 7.42.
4H), 1.45-1.29 (m, 20H), 1.14 (s, 42H), 0.89 (t, J ) 6.7 Hz, 6H); 13
C
NMR (100 MHz, CDCl3) δ 164.92 (s), 139.39 (d), 133.44 (d), 132.83
(d), 131.24 (s), 124.52 (s), 122.17 (s), 104.59 (s), 93.39 (s), 80.33 (s),
74.81 (s), 65.80 (t), 31.83 (t), 29.28 (t), 29.22 (t), 28.71 (t), 26.05 (t),
22.69 (t), 18.70 (d), 14.14 (q), 11.35 (q); IR (KBr) 2154, 1726, 1221,
8a: pale yellow oil; 1H NMR (400 MHz, CDCl3) δ 8.14 (t, J ) 1.5
Hz, 1H), 8.12 (t, J ) 1.5 Hz, 1H), 8.10 (m, 2H), 7.78 (m, 2H), 4.33 (t,
J ) 6.6 Hz, 2H), 4.32 (t, J ) 6.6 Hz, 2H), 3.17 (s, 1H), 1.77 (m, 4H),
1.45-1.29 (m, 20H), 1.14 (s, 21H), 0.89 (t, J ) 6.8 Hz, 6H); 13C NMR
(100 MHz, CDCl3) δ 164.74 (s), 164.54 (s), 139.29 (d), 139.21 (d),
133.52 (d), 133.39 (d), 133.37 (d), 132.76 (d), 131.32 (s), 131.21 (s),
124.48 (s), 123.15 (s), 122.31 (s), 122.08 (s), 104.54 (s), 93.32 (s),
81.41 (d), 80.37 (s), 80.01 (s), 79.15 (s), 74.98 (s), 74.76 (s), 65.76 (t),
65.71 (t), 31.81 (t), 31.79 (t), 29.25 (t), 29.19 (t), 28.68 (t), 28.67 (t),
26.02 (t), 22.66 (t), 18.65 (d), 14.10 (q), 11.31 (q); IR (neat) 3306,
2158, 1727, 1219, 768, 678 cm-1; MS (FAB) m/z 719 (M+ + H), 675
(M+ - i-Pr), 563 (M+ + H - i-Pr - C8H17).
Partial Deprotection of 6b: 2mer 7b and Singly-Protected 2mer
8b. To a solution of TIPS-protected acetylene 6b (50 mg, 0.046 mmol)
in THF (1.5 mL) and water (0.2 mL) was added dropwise 0.046 mL
of a solution of TBAF (1.0 M) in THF and the solution was stirred at
room temperature. The reaction was monitored by HPLC and another
TBAF solution (each 0.04 mL portion) was added seven times at 30-
min intervals. After 4 h, the reaction mixture was worked up as
described for the reaction of 6a. Purification by flash chromatography
afforded 7b (8 mg, 22%), 8b (17 mg, 40%), and the starting material
6b (11 mg).
968, 882, 768, 677 cm-1; MS (FAB) m/z 875 (M+ + H), 831 (M+
+
H - i-Pr), 719 (M+ + H - CO2C8H17). Anal. Calcd for C56H82O4Si2:
C, 76.83; H, 9.44. Found: C, 77.11; H, 9.56.
TIPS-Protected 2mer 6b. To a solution of Cu(OAc)2 (3.30 g, 18.0
mmol) in pyridine (300 mL) was added a solution of 5b29 (5.00 g,
9.08 mol) in pyridine (70 mL). After being stirred for 4 h, the mixture
was worked up as described for the reaction of 6a. Purification by flash
chromatography afforded 6b as a pale yellow solid (4.77 g, 96%): mp
1
48-49 °C; H NMR (270 MHz, CDCl3) δ 8.09 (d, J ) 1.7 Hz, 4H),
7.78 (m, 2H), 4.33 (t, J ) 6.8 Hz, 4H), 1.78 (quintet, J ) 6.8 Hz, 4H),
1.46-1.25 (br s, 52H), 1.13 (s, 42H), 0.88 (t, J ) 6.8 Hz, 6H); 13C
NMR (67.5 MHz, CDCl3) δ 164.87 (s), 139.37 (d), 133.42 (d), 132.81
(d), 131.24 (s), 124.52 (s), 122.17 (s), 104.59 (s), 93.36 (s), 80.32 (s),
74.82 (s), 65.76 (t), 31.97 (t), 29.75 (t), 29.71 (t), 29.70 (t), 29.64 (t),
29.57 (t), 29.42 (t), 29.33 (t), 28.72 (t), 26.04 (t), 22.75 (t), 18.69 (d),
14.16 (q), 11.35 (q); IR (KBr) 2150, 1720, 1220, 890, 670 cm-1; UV
(CHCl3) λmax (log ꢀ) 337 (4.39), 313 (4.49), 294 (4.33), 296 (4.78),
259 (4.80) nm; MS (FAB) m/z 1099 (M+ + H), 1054 (M+ + H -
i-Pr). Anal. Calcd for C72H114O4Si2: C, 78.63; H, 10.45. Found: C,
78.70; H, 10.81.
7b: white solid; mp 87-88 °C; 1H NMR (400 MHz, CDCl3) δ 8.14
(t, J ) 1.5 Hz, 2H), 8.13 (t, J ) 1.5 Hz, 2H), 7.78 (t, J ) 1.5 Hz, 2H),
4.33 (t, J ) 6.8 Hz, 4H), 3.20 (s, 2H), 1.77 (quintet, J ) 6.8 Hz, 4H),
1.46-1.25 (m, 52H), 0.88 (t, J ) 6.8 Hz, 6H); 13C NMR (100 MHz,
CDCl3) δ 164.66 (s), 139.31 (d), 133.62 (d), 133.44 (d), 131.37 (s),
123.18 (s), 122.32 (s), 81.45 (d), 80.17 (s), 79.16 (s), 74.95 (s), 65.83
(t), 31.97 (t), 29.74 (t), 29.71 (t), 29.64 (t), 29.55 (t), 29.41 (t), 29.32
(t), 28.69 (t), 26.04 (t), 22.74 (t), 14.16 (q); IR (KBr) 3270, 1720, 1208,
895 cm-1; UV (CHCl3) λmax (log ꢀ) 335 (4.28), 311 (4.31), 292 (4.15),
246 (4.62) nm; MS (FAB) m/z 786 (M+), 393 (M+ - C2C6H3-
(CO2C16H33)(C2H)). Anal. Calcd for C54H74O4: C, 82.40; H, 9.47.
Found: C, 82.52; H, 9.75.
8b: white solid; mp 56-57 °C; 1H NMR (400 MHz, CDCl3) δ 8.14
(t, J ) 1.5 Hz, 1H), 8.13 (t, J ) 1.5 Hz, 1H), 8.10 (m, 2H), 7.78 (t, J
) 1.5 Hz, 2H), 4.330 (t, J ) 6.6 Hz, 2H), 4.326 (t, J ) 6.6 Hz, 2H),
3.16 (s, 1H), 1.77 (m, 4H), 1.45-1.25 (m, 52H), 1.14 (s, 21H), 0.88
(t, J ) 6.8 Hz, 6H); 13C NMR (100 MHz, CDCl3) δ 164.85 (s), 164.66
(s), 139.37 (d), 139.29 (d), 133.58 (d), 133.44 (d), 133.44 (d), 132.82
(d), 131.36 (s), 131.25 (s), 124.53 (s), 123.18 (s), 122.38 (s), 122.12
(s), 104.57 (s), 93.38 (s), 81.46 (d), 80.42 (s), 80.04 (s), 79.14 (s),
75.00 (s), 74.77 (s), 65.81 (t), 65.77 (t), 31.97 (t), 29.75 (t), 29.72 (t),
29.70 (t), 29.64 (t), 29.57 (t), 29.41 (t), 29.32 (t), 28.72 (t), 28.69 (t),
26.04 (t), 22.74 (t), 18.69 (d), 14.16 (q), 11.34 (q); IR (KBr) 3299,
2164, 1730, 1221, 899, 767, 677, 631 cm-1; MS (FAB) m/z 943 (M+),
899 (M+ - i-Pr - H), 675 (M+ - i-Pr - C16H33), 393 (M+ - C2C6H3-
(CO2C16H33)(C2TIPS)). Anal. Calcd for C63H94O4Si: C, 80.20; H, 10.04.
Found: C, 80.42; H, 9.84.
TIPS-Protected 2mer 6c. A suspension of CuCl (4.95 g, 50.0 mmol)
and tetramethylethylenediamine (TMEDA) (1.98 g, 17.0 mmol) in
acetone (80 mL) was stirred at room temperature for 30 min. The blue
skim of this mixture was used as a catalyst. To an oxygen-saturated
solution of 5c29 (501 mg, 1.06 mmol) in acetone (8 mL) was added a
solution of the above catalyst (0.4 mL). The mixture was stirred at
room temperature for 24 h while oxygen was bubbled through a glass
tube immersed into the solution. Additional catalyst (0.2 mL) was added,
and the solution was stirred for an additional 21 h. After addition of 1
N HCl (1 mL) and water (20 mL), most of the acetone was evaporated
in vacuo and the residue was extracted with ether. The extract was
washed with brine and dried over MgSO4. The solvent was removed
under reduced pressure to give a crude product. Purification by silica
gel chromatography afforded 6c as a pale yellow oil (458 mg, 92%):
1H NMR (300 MHz, CDCl3) δ 8.12-8.10 (m, 4H), 7.78 (t, J ) 1.5
Hz, 2H), 4.51-4.48 (m, 4H), 3.86-3.83 (m, 4H), 3.74-3.64 (m, 12H),
3.55-3.52 (m, 4H), 3.37 (s, 6H), 1.14-1.13 (m, 42H); 13C NMR (75
MHz, CDCl3) δ 164.82 (s), 139.56 (d), 133.52 (d), 132.94 (d), 130.88
(s), 124.55 (s), 122.19 (s), 104.47 (s), 93.43 (s), 80.22 (s), 74.82 (s),
71.90 (t), 70.67 (t), 70.61 (t), 70.58 (t), 69.02 (t), 64.57 (t), 58.94 (q),
18.60 (d), 11.22 (q); IR (neat) 2156, 1729, 1220, 890, 670 cm-1; UV
(CHCl3) λmax (log ꢀ) 337 (4.39), 313 (4.49), 294 (4.33), 296 (4.78),
259 (4.80) nm; MS (FAB) m/z 965 (M+ + Na), 943 (M+ + H).
Partial Deprotection of 6a: 2mer 7a and Singly-Protected 2mer
8a. To a solution of TIPS-protected acetylene 6a (3.04 g, 3.48 mmol)
in THF (107 mL) and water (4.2 mL) was added dropwise 3.6 mL of
a solution of tetrabuthylammonium fluoride (TBAF) in THF (1.0 M).
After being stirred at room temperature for 70 min, the mixture was
poured to ice water and extracted with CHCl3. The extract was washed
Partial Deprotection of 6c: 2mer 7c and Singly-Protected 2mer
8c. To a solution of TIPS-protected acetylene 6c (320 mg, 0.339 mmol)
9
5360 J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002

m-Diethynylbenzene Macrocycles
A R T I C L E S
in THF (12 mL) and water (0.6 mL) was added dropwise 0.34 mL of
a solution of TBAF (1.0 M) in THF. After being stirred at room
temperature for 1.5 h, the reaction mixture was worked up as described
for the reaction of 6a except that ether was employed instead of CHCl3
for extraction. Purification by flash chromatography afforded the
products 7c (51 mg, 24%) and 8c (122 mg, 46%), and the starting
material 6c (84 mg) was recovered.
7c: pale yellow solid; mp 60-61 °C; 1H NMR (300 MHz, CDCl3)
δ 8.17-8.15 (m, 4H), 7.78 (t, J ) 1.6 Hz, 2H), 4.51-4.48 (m, 4H),
3.86-3.82 (m, 4H), 3.73-3.64 (m, 12H), 3.56-3.53 (m, 4H), 3.37 (s,
6H), 3.17 (s, 2H); 13C NMR (75 MHz, CDCl3) δ 164.61 (s), 139.48
(d), 133.73 (d), 133.55 (d), 131.02 (s), 123.22 (s), 122.33 (s), 81.32
(d), 80.08 (s), 79.28 (s), 74.93 (s), 71.88 (t), 70.64 (t), 70.57 (t), 70.55
(t), 68.97 (t), 64.60 (t), 58.93 (q); IR (KBr) 3257, 2222, 2104, 1725,
1307, 1223, 897, 771, 709, 669 cm-1; MS (FAB) m/z 631 (M+ + H),
511 (M+ - O(CH2CH2O)2CH3). Anal. Calcd for C36H38O10: C, 68.56;
H, 6.07. Found: C, 68.21; H, 6.03.
(m, 40H), 0.94-0.85 (m, 12H); 13C NMR (100 MHz, CDCl3) δ 164.66
(s), 164.46 (s), 139.47 (d), 139.32 (d), 133.98 (d), 133.67 (d), 133.47
(d), 131.60 (s), 131.40 (s), 123.20 (s), 122.63 (s), 122.56 (s), 122.26
(s), 100.55 (s), 81.43 (d), 80.40 (s), 79.90 (s), 79.13 (s), 75.28 (s),
75.20 (s), 74.88 (s), 65.95 (t), 65.85 (t), 31.85 (t), 29.28 (t), 29.21 (t),
28.69 (t), 26.05 (t), 22.70 (t), 14.15 (q); IR (KBr) 3260, 2150, 2110,
1730, 1260, 894, 768, 672 cm-1; MS (MALDI-TOF) m/z 1185.5 (M+
+ Cu), 1145.3 (M+ + Na).
TIPS-Protected Linear 4mer 10c. The catalyst was prepared by
srirring a suspension of CuCl (1.00 g, 10.1 mmol) and TMEDA (391
mg, 3.37 mmol) in acetone (20 mL) at room temperature for 30 min.
To an oxygen-saturated solution of 8c (3.85 g, 4.89 mmol) in acetone
(50 mL) was added a solution of the above catalyst (2.9 mL). The
mixture was stirred at room temperature for 6 h while oxygen was
bubbled through a glass tube immersed into the solution. Additional
catalyst (1.5 mL) was added, and the solution was stirred for an
additional 26 h. The reaction mixture was worked up as described for
the synthesis of 6c. Purification by silica gel chromatography afforded
10c as a pale yellow oil (3.61 g, 94%): 1H NMR (300 MHz, CDCl3)
δ 8.21-8.19 (m 4H), 8.14-8.11 (m, 4H), 7.83 (t, J ) 1.5 Hz, 2H),
7.80 (t, J ) 1.5 Hz, 2H), 4.52-4.48 (m, 8H), 3.86-3.83 (m, 8H), 3.74-
3.64 (m, 24H), 3.57-3.52 (m, 8H), 3.38 (s, 6H), 3.37 (s, 6H), 1.14 (br
s, 42H); 13C NMR (75 MHz, CDCl3) δ 164.85 (s), 164.47 (s), 139.69
(d), 139.62 (d), 134.12 (d), 134.05 (d), 133.64 (d), 132.99 (d), 131.28
(s), 130.92 (s), 124.59 (s), 122.74 (s), 122.60 (s), 122.10 (s), 104.46
(s), 93.52 (s), 80.58 (s), 80.05 (s), 79.70 (s), 75.35 (s), 75.20 (s), 74.71
(s), 71.93 (t), 71.90 (t), 70.68 (t), 70.61 (t), 69.03 (t), 68.97 (t), 64.72
(t), 64.59 (t), 58.91 (q), 18.60 (d), 11.22 (q); IR (neat) 2156, 1730,
8c: pale yellow oil; 1H NMR (300 MHz, CDCl3) δ 8.17-8.14 (m,
2H), 8.13-8.11 (m, 2H), 7.80-7.78 (m, 2H), 4.51-4.47 (m, 4H), 3.86-
3.82 (m, 4H), 3.74-3.64 (m, 12H), 3.56-3.53 (m, 4H), 3.372 (s, 3H),
3.370 (s, 3H), 3.16 (s, 1H), 1.14 (br s, 21H); 13C NMR (75 MHz, CDCl3)
δ 164.86 (s), 164.67 (s), 139.61 (d), 139.52 (d), 133.74 (d), 133.59
(d), 132.99 (d), 131.05 (d), 130.91 (s), 124.59 (s), 123.24 (s), 122.43
(s), 122.16 (s), 104.48 (s), 93.50 (s), 81.36 (d), 80.37 (s), 79.98 (s),
79.24 (s), 75.02 (s), 74.78 (s), 71.92 (t), 70.68 (t), 70.59 (t), 69.04 (t),
69.00 (t), 64.63 (t), 64.59 (t), 58.97 (q), 18.60 (d), 11.31 (q); IR (neat)
3306, 2158, 1727, 1219, 768, 678 cm-1; MS (FAB) m/z 787 (M+
H), 667 (M+ - O(CH2CH2O)2CH3).
+
1585, 1260, 766, 677 cm-1; MS (MALDI-TOF) m/z 1593.7 (M+
+
2mer 8c was also prepared by transesterification of 8a. Thus, to a
stirred solution of 8a (2.46 g, 2.78 mmol) and triethylene glycol
monomethyl ether (56.0 g, 341 mmol) were added K2CO3 (3.84 g, 27.8
mmol) and 18-crown-6 ether (0.367 g, 1.39 mmol). After the mixture
was stirred at room temperature for 9 h, the solvent was removed with
an evaporator to give a mixture of 8c, two isomers of monoexchanged
2mers, and 8a (recovered). Purification by flash chromatography
afforded 8c as a pale yellow oil (1.50 g, 56%).
Na).
Linear 4mer 11c. To a solution of TIPS-protected acetylene 10c
(3.61 g, 2.30 mmol) in THF (115 mL) and water (5 mL) was added
dropwise 11.5 mL of a solution of TBAF (1.0 M) in THF. After being
stirred at room temperature for 2 h, the reaction mixture was worked
up as described for the synthesis of 7a and 8a. Purification by flash
chromatography afforded 11c (2.49 mg, 86%) as a pale yellow solid:
1
mp 81-82 °C; H NMR (300 MHz, CDCl3) δ 8.20 (d, J ) 1.5 Hz,
TIPS-Protected Linear 4mer 10a. The catalyst was prepared by
stirring a suspension of CuCl (4.95 g, 50.0 mmol) and TMEDA (1.98
g, 17.0 mmol) in acetone (80 mL) at room temperature for 30 min. To
an oxygen-saturated solution of 8a (7.58 g, 10.5 mmol) in acetone (100
mL) was added a solution of the above catalyst (80 mL). The mixture
was stirred at room temperature for 48 h while oxygen was bubbled
through a glass tube immersed into the solution. The reaction mixture
was worked up as described for the synthesis of 6c. Purification by
flash chromatography afforded 10a as a pale yellow oil (6.65 g, 88%):
1H NMR (400 MHz, CDCl3) δ 8.17 (d, J ) 1.5 Hz, 4H), 8.10 (t, J )
1.5 Hz, 4H), 7.81 (t, J ) 1.5 Hz, 2H), 7.79 (t, J ) 1.5 Hz, 2H), 4.34
(t, J ) 6.8 Hz, 4H), 4.33 (t, J ) 6.8 Hz, 4H), 1.78 (quintet, J ) 6.8
4H), 8.17 (t, J ) 1.5 Hz, 2H), 8.16 (t, J ) 1.5 Hz, 2H), 7.84 (t, J )
1.5 Hz, 2H), 7.81 (t, J ) 1.5 Hz, 2H), 4.52-4.48 (m, 8H), 3.86-3.83
(m, 8H), 3.75-3.65 (m, 24H), 3.57-3.53 (m, 8H), 3.38 (s, 6H), 3.37
(s, 6H), 3.17 (s, 2H); 13C NMR (100 MHz, CDCl3) δ 164.66 (s), 164.47
(s), 139.71 (d), 139.55 (d), 134.13 (d), 133.83 (d), 133.62 (d), 131.29
(s), 131.07 (s), 123.27 (s), 122.69 (s), 122.62 (s), 122.31 (s), 81.35
(d), 80.33 (s), 80.06 (s), 79.84 (s), 79.30 (s), 75.30 (s), 75.23 (s), 74.90
(s), 71.93 (t), 70.70 (t), 70.61 (t), 69.00 (t), 64.74 (t), 64.65 (t), 58.99
(q); IR (KBr) 3290, 3260, 3080, 2875, 2224, 2111, 1727, 1585, 1455,
1435, 1263, 1219, 1114, 897, 765, 673 cm-1; MS (MALDI-TOF) m/z
1281.9 (M+ + Na). Anal. Calcd for C72H74O20: C, 68.67; H, 5.92.
Found: C, 68.30; H, 6.10.
Hz, 8H), 1.46-1.14 (m, 40H), 1.14 (s, 42H), 0.92-0.88 (m, 12H); 13
C
NMR (100 MHz, CDCl3) δ 164.78 (s), 164.41 (s), 139.41 (d), 139.34
(d), 133.90 (d), 133.47 (d), 132.80 (d), 131.57 (s), 131.24 (s), 124.51
(s), 122.66 (s), 122.52 (s), 122.02 (s), 104.53 (s), 93.41 (s), 80.65 (s),
80.13 (s), 79.78 (s), 75.34 (s), 75.20 (s), 74.70 (s), 65.93 (t), 65.78 (t),
31.87 (t), 31.86 (t), 29.30 (t), 28.23 (t), 28.73 (t), 26.08 (t), 22.72 (t),
18.72 (d), 14.17 (q), 11.37 (q); IR (neat) 2160, 1730, 1590, 1260, 766,
Monobrominated 2mer 9a. In a flask covered with aluminum foil,
8a (12.3 g, 17.1 mmol) was dissolved in acetone (450 mL), and then
AgNO3 (0.87 g, 5.1 mmol) and N-bromosuccinimide (NBS) (4.27 g,
24.0 mmol) were added. After being stirred at room temperature for
50 min, the mixture was diluted with water and extracted with CHCl3.
The extract was washed with brine and dried over MgSO4. After
removal of the solvent in vacuo, purification by flash chromatography
afforded 9a as a pale yellow oil (13.2 g, 97%): 1H NMR (400 MHz,
CDCl3) δ 8.13 (m, 1H), 8.09 (m, 3H), 7.78 (m, 1H), 7.73 (m, 1H),
4.33 (t, J ) 6.6 Hz, 2H), 4.32 (t, J ) 6.6 Hz, 2H), 1.77 (m, 4H), 1.44-
1.29 (m, 20H), 1.14 (s, 21H), 0.89 (m, 6H); 13C NMR (100 MHz,
CDCl3) δ 164.89 (s), 164.66 (s), 139.38 (d), 139.14 (d), 133.48 (d),
133.33 (d), 132.83 (d), 131.41 (s), 131.26 (s), 124.53 (s), 123.73 (s),
122.43 (s), 122.10 (s), 104.56 (s), 93.42 (s), 80.48 (s), 80.01 (s), 78.11
(s), 75.04 (s), 74.75 (s), 65.86 (t), 65.80 (t), 52.60 (s), 31.85 (t), 29.28
(t), 29.21 (t), 28.71 (t), 28.69 (t), 26.05 (t), 22.70 (t), 18.70 (d), 14.15
677 cm-1; MS (MALDI-TOF) m/z 1497.9 (M+ + Cu), 1458.0 (M+
+
Na).
Linear 4mer 11a. To a solution of TIPS-protected acetylene 10a
(261 mg, 0.173 mmol) in THF (4 mL) and water (1 mL) was added
dropwise 2 mL of a solution of TBAF (1.0 M) in THF. After being
stirred at room temperature for 8 h, the reaction mixture was worked
up as described for the synthesis of 7a and 8a. Purification by flash
chromatography afforded 11a (190 mg, 98%) as a white solid: mp
116-118 °C; 1H NMR (400 MHz, CDCl3) δ 8.17-8.13 (m, 8H), 7.81
(t, J ) 1.5 Hz, 2H), 7.79 (t, J ) 1.5 Hz, 2H), 4.34 (t, J ) 6.9 Hz, 4H),
4.33 (t, J ) 6.6 Hz, 4H), 3.17 (s, 2H), 1.81-1.74 (m, 8H), 1.46-1.30
(q), 11.35 (q); IR (neat) 2204, 2160, 1728, 1221, 896, 768, 678 cm-1
;
9
J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002 5361

A R T I C L E S
Tobe et al.
MS (FAB) m/z 799, 797 (M+ + H), 755, 753 (M+ - i-Pr), 669, 667
(M+ - OC8H17).
(dba)3‚CHCl3 (6 mg, 0.006 mmol), CuI (0.95 mg, 0.0050 mmol), and
benzene (1.4 mL). After being stirred at room temperature for 5 min,
i-Pr2NH (0.041 mL) was added dropwise and the mixture was stirred
at room temperature for 45 min. After completion of the reaction, the
mixture was concentrated in vacuo and the residue was subjected to
flash chromatography on silica gel to afford 12b as a brown oil (34
mg, 44%): 1H NMR (400 MHz, CDCl3) δ 8.18 (m, 8H), 8.10 (m,
4H), 7.81 (m, 4H), 7.78 (t, J ) 1.7 Hz, 2H), 4.34 (m, 12H), 1.78 (m,
12H), 1.46-1.25 (m, 156H), 1.14 (s, 42H), 0.87 (t, J ) 6.8 Hz, 18H);
13C NMR (100 MHz, CDCl3) δ 164.85 (s), 164.48 (s), 164.45 (s),
139.46 (d), 139.39 (d), 134.00 (d), 133.91 (d), 133.51 (d), 132.83 (d),
131.60 (s), 131.59 (s), 131.26 (s), 124.54 (s), 122.70 (s), 122.60 (s),
122.55 (s), 122.05 (s), 104.55 (s), 93.43 (s), 80.66 (s), 80.15 (s), 80.11
(s), 80.09 (s), 79.77 (s), 75.35 (s), 75.24 (s), 74.71 (s), 65.95 (t), 65.79
(t), 31.97 (t), 29.75 (t), 29.71 (t), 29.65 (t), 29.56 (t), 29.42 (t), 29.32
(t), 28.72 (t), 28.69 (t), 26.04 (t), 22.75 (t), 22.74 (t), 18.70 (d), 14.17
Monobrominated 2mer 9b. In a flask covered with aluminum foil,
8b (366 mg, 0.39 mmol) was dissolved in acetone (10 mL) with
warming at 45 °C. To the solution were added AgNO3 (26 mg, 0.16
mmol) followed by NBS (76 mg, 0.43 mmol) and the mixture was
stirred at 45 °C. After 1 h, additional NBS (33 mg, 0.18 mmol) was
added and the solution was stirred for an additional 1 h. The reaction
mixture was worked up as described for the reaction of 8a. Purification
by flash chromatography afforded 9b as a pale yellow solid (356 mg,
90%): mp 49-50 °C; 1H NMR (400 MHz, CDCl3) δ 8.13 (t, J ) 1.5
Hz, 1H), 8.10 (m, 3H), 7.78 (t, J ) 1.5 Hz, 1H), 7.73 (t, J ) 1.5 Hz,
1H), 4.33 (t, J ) 6.8 Hz, 2H), 4.32 (t, J ) 6.6 Hz, 2H), 3.19 (m, 4H),
1.43-1.25 (m, 52H), 1.14 (s, 21 H), 0.88 (t, J ) 6.8 Hz, 6H); 13C
NMR (100 MHz, CDCl3) δ 164.88 (s), 164.65 (s), 139.38 (d), 139.14
(d), 133.48 (d), 133.32 (d), 132.82 (d), 131.41 (s), 131.26 (s), 124.53
(s), 123.73 (s), 122.43 (s), 122.11 (s), 104.57 (s), 93.41 (s), 80.48 (s),
80.01 (s), 78.11 (s), 75.05 (s), 74.76 (s), 65.85 (t), 65.79 (t), 52.59 (s),
31.98 (t), 29.76 (t), 29.73 (t), 29.71 (t), 29.65 (t), 29.57 (t), 29.56 (t),
29.42 (t), 29.33 (t), 28.72 (t), 28.70 (t), 26.04 (t), 22.75 (t), 18.71 (d),
14.17 (q), 11.35 (q); IR (KBr) 2206, 2160, 1727, 1223, 897, 768, 677
cm-1; MS (FAB) m/z 1024, 1022 (M+ + H), 981, 979 (M+ + H -
i-Pr), 755, 753 (M+ + H - CO2C16H33). Anal. Calcd for C63H93O4-
BrSi: C, 74.01; H, 9.17. Found: C, 74.10; H, 9.40.
(q), 11.35 (q); IR (neat) 2228, 2160, 1730, 1256, 896, 768, 677 cm-1
;
MS (MALDI-TOF) m/z 2691.3 (M+ + Na).
TIPS-Protected Linear 6mer 12c. A flask was charged with linear
2mer 7c (1.78 g, 2.83 mmol), bromide 9c (5.14 g, 5.94 mmol),
Pd2(dba)3‚CHCl3 (141 mg, 0.141 mmol), CuI (26 mg, 0.141 mmol),
and benzene (150 mL). After being stirred at room temperature for 5
min, i-Pr2NH (0.95 mL) was added dropwise and the mixture was stirred
at room temperature for 45 min. After completion of the reaction, the
mixture was concentrated in vacuo and the residue was subjected to
flash chromatography on silica gel to afford 12c as a yellow oil (1.91
Monobrominated 2mer 9c. In a flask covered with aluminum foil,
8c (122 mg, 0.155 mmol) was dissolved in acetone (3 mL), and then
AgNO3 (7.9 mg, 0.046 mmol) and NBS (38.7 mg, 0.217 mmol) were
added. After being stirred at room temperature for 1 h, the reaction
mixture was worked up as described for the synthesis of 9a except
that ether was employed instead of CHCl3 for extraction. Purification
by flash chromatography afforded 9c as a pale yellow oil (101 mg,
75%): 1H NMR (270 MHz, CDCl3, 30 °C) δ 8.15 (t, J ) 1.6 Hz, 1H),
8.12-8.10 (m, 3H), 7.78 (t, J ) 1.6 Hz, 1H), 7.74 (t, J ) 1.6 Hz, 1H),
4.51-4.47 (m, 4H), 3.85-3.81 (m, 4H), 3.73-3.63 (m, 12H), 3.55-
3.52 (m, 4H), 3.37 (s, 3H), 3.36 (s, 3H) 1.138-1.133 (m, 21H);13C
(67.5 MHz, CDCl3, 30 °C) δ 164.72 (s), 164.47 (s), 139.50 (d), 139.23
(d), 133.54 (d), 133.50 (d), 133.35 (d), 132.89 (d), 130.98 (s), 130.82
(s), 124.49 (s), 123.69 (s), 122.38 (s), 122.06 (s), 104.41 (s), 93.44 (s),
80.36 (s), 79.90 (s), 77.98 (s), 75.04 (s), 74.76 (s), 71.91 (t), 71.89 (t),
70.67 (t), 70.60 (t), 70.59 (t), 70.58 (t), 70.57 (t), 69.02 (t), 68.98 (t),
64.62 (t), 64.57 (t), 58.97 (q), 52.68 (s), 18.65 (q), 11.28 (t); IR (neat)
3070, 2942, 2865, 2202, 2155, 1727, 1586, 1460, 1434, 1256, 1112,
895, 767, 677 cm-1; MS (FAB) m/z 867, 865 (M+ + H), 747, 745
(M+ - O(CH2CH2O)2CH3).
TIPS-Protected Linear 6mer 12a. A flask was charged with linear
2mer 7a (50 mg, 0.090 mmol), bromide 9a (150 mg, 0.190 mmol),
Pd2(dba)3‚CHCl3 (4.6 mg, 0.0045 mmol), CuI (0.85 mg, 0.0045 mmol),
and benzene (3 mL). After being stiired at room temperature for 5 min,
i-Pr2NH (0.030 mL) was added dropwise and the mixture was stirred
at room temperature for 45 min. After completion of the reaction, the
mixture was concentrated in vacuo and the residue was subjected to
flash chromatography on silica gel to afford 12a as a yellow oil (67
mg, 38%): 1H NMR (400 MHz, CDCl3) δ 8.17 (m, 8H), 8.10 (m,
4H), 7.81 (m, 4H), 7.79 (t, J ) 1.7 Hz, 2H), 4.34 (t, J ) 6.6 Hz, 8H),
4.33 (t, J ) 6.6 Hz, 4H), 1.78 (m, 12H), 1.44-1.30 (m, 60H), 1.14 (s,
42H), 0.90 (m, 18H); 13C NMR (100 MHz, CDCl3) δ 164.85 (s), 164.47
(s), 164.44 (s), 139.45 (d), 139.38 (d), 133.51 (d), 133.91 (d), 133.51
(d), 132.83 (d), 131.60 (s), 131.59 (s), 131.26 (s), 124.54 (s), 122.69
(s), 122.58 (s), 122.54 (s), 122.04 (s), 104.54 (s), 93.43 (s), 80.66 (s),
80.16 (s), 80.12 (s), 80.09 (s), 79.78 (s), 75.34 (s), 75.24 (s), 75.19 (s),
74.70 (s), 65.95 (t), 65.79 (t), 31.85 (t), 31.83 (t), 29.29 (t), 29.21 (t),
28.71 (t), 28.69 (t), 26.06 (t), 26.04 (t), 22.70 (t), 18.69 (t), 14.16 (d),
14.14 (q), 11.34 (q); IR (neat) 2235, 2160, 1730, 1256, 895, 767, 676
cm-1; MS (MALDI-TOF) m/z 2057.7 (M+ + Cu), 2017.8 (M+ + Na).
TIPS-Protected Linear 6mer 12b. A flask was charged with linear
2mer 7b (23 mg, 0.029 mmol), bromide 9b (90 mg, 0.088 mmol), Pd2-
1
g, 31%): H NMR (270 MHz, CDCl3, 30 °C) δ 8.20-8.19 (m, 8H),
8.12 (t, J ) 1.6 Hz, 2H), 8.11 (t, J ) 1.6 Hz, 2H), 7.84 (t, J ) 1.6 Hz,
2H), 7.83 (t, J ) 1.6 Hz, 2H), 7.79 (t, J ) 1.6 Hz, 2H), 4.52-4.48 (m,
12H), 3.86-3.83 (m, 12H), 3.73-3.63 (m, 36H), 3.56-3.53 (m, 12H),
3.377 (s, 12H), 3.370 (s, 6H), 1.140-1.135 (m, 42H); 13C NMR (67.5
MHz, CDCl3, 30 °C) δ 164.72 (s), 164.38 (s), 139.63 (d), 139.54 (d),
134.09 (d), 133.51 (d), 132.93 (d), 131.23 (s), 131.21 (s), 130.85 (s),
127.70 (s), 124.54 (s), 122.69 (s), 122.58 (s), 122.54 (s), 122.04 (s),
104.42 (s), 97.19 (s), 93.50 (s), 80.58 (s), 80.07 (s), 80.02 (s), 79.70
(s), 75.35 (s), 75.24 (s), 75.19 (s), 74.71 (s), 71.94 (t), 71.92 (t), 70.70
(t), 70.68 (t), 70.62 (t), 70.61 (t), 70.59 (t), 69.05 (t), 68.99 (t), 64.75
(t), 64.61 (t), 59.01 (q), 59.00 (q), 18.68 (q), 11.30 (d); IR (neat) 3069,
2943, 2865, 2223, 2155, 1728 1584, 1434, 1258, 1217, 1113, 896, 765,
675 cm-1; MS (MALDI-TOF) m/z 2222.1 (M+ + Na).
Linear 6mer 13a. To a solution of TIPS-protected acetylene 12a
(2.11 g, 1.06 mmol) in THF (50 mL) and water (0.25 mL) was
added dropwise 53 mL of a solution of TBAF/acetic acid (0.2
M/0.3 M) in THF during 30 min. To complete the reaction, an addi-
tional 21.2 mL of a solution of TBAF/acetic acid (0.5 M/0.75 M) in
THF was added over 10 min. After the mixture was stirred at room
temperature for further 30 min, 5.0 g of silica gel was added at 0 °C
and the solvent was removed in vacuo. The residue was loaded on the
top of a silica gel chromatography column and subsequent flash
chromatography afforded 13a as a pale yellow solid (1.74 g,
1
97%): mp 89-91 °C; H NMR (400 MHz, CDCl3) δ 8.17 (m, 8H),
8.15 (m, 2H), 8.13 (m, 2H), 7.82 (m, 4H), 7.79 (m, 2H), 4.34 (m, 12H),
3.17 (s, 2H), 1.78 (m, 12H), 1.44-1.30 (m, 60H), 0.90 (m, 18H); 13C
NMR (100 MHz, CDCl3) δ 164.68 (s), 164.48 (s), 139.48 (d), 139.33
(d), 134.00 (d), 133.68 (d), 133.48 (d), 131.60 (s), 131.40 (s), 123.21
(s), 122.64 (s), 122.59 (s), 122.56 (s), 122.27 (s), 81.44 (d), 80.40 (s),
80.13 (s), 79.91 (s), 79.19 (s), 75.28 (s), 75.23 (s), 75.20 (s), 74.89 (s),
65.97 (t), 65.86 (t), 31.86 (t), 29.29 (t), 29.22 (t), 28.69 (t), 26.06 (t),
22.71 (t), 14.16 (q); IR (KBr) 3255, 2217, 1728, 1258, 892, 766, 672
cm-1; MS (MALDI-TOF) m/z 1745.7 (M+ + Cu), 1705.8 (M+ + Na).
Anal. Calcd for C114H122O12: C, 81.30; H, 7.30. Found: C, 81.31; H,
7.47.
Linear 6mer 13b. To a solution of TIPS-protected acetylene 12b
(1.57 g, 0.59 mmol) in THF (57 mL) and water (0.15 mL) was added
dropwise 12.9 mL of a solution of TBAF/acetic acid (0.1 M/0.15 M)
in THF over 10 min. To complete the reaction, an additional 12.9 mL
9
5362 J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002

m-Diethynylbenzene Macrocycles
A R T I C L E S
of a solution of TBAF/acetic acid (0.1 M/0.15 M) in THF was added
duoverring 10 min. After being stirred at room temperature for further
2 h, the reaction mixture was worked up according to the above
procedure. Purification by flash chromatography afforded 13b as a pale
yellow solid (1.34 g, 97%): mp 44-47 °C; 1H NMR (400 MHz, CDCl3)
δ 8.17 (m, 8H), 8.15 (m, 2H), 8.13 (m, 2H), 7.81 (m, 4H), 7.78 (m,
2H), 4.33 (m, 12H), 3.16 (s, 2H), 1.78 (m, 12H), 1.44-1.25 (m, 156H),
0.87 (t, J ) 6.8 Hz, 18H); 13C NMR (100 MHz, CDCl3) δ 164.64 (s),
166.44 (s), 139.46 (d), 139.32 (d), 133.98 (d), 133.66 (d), 133.45 (d),
131.59 (s), 131.38 (s), 123.20 (s), 122.64 (s), 122.58 (s), 122.27 (s),
81.43 (d), 80.39 (s), 80.11 (s), 79.90 (s), 79.19 (s), 75.30 (s), 75.25
(s), 74.91 (s), 65.95 (t), 65.84 (t), 31.97 (t), 29.75 (t), 29.71 (t), 29.65
(t), 29.56 (t), 29.41 (t), 29.32 (t), 28.69 (t), 26.04 (t), 22.75 (t), 14.17
(q); IR (KBr) 3307, 2217, 1729, 1257, 894, 766, 673 cm-1; MS
(MALDI-TOF) m/z 2377.9 (M+ + Na). Anal. Calcd for C162H218O12:
C, 82.54; H, 9.32. Found: C, 82.49; H, 9.54.
mL of the same solvent over 72 h. After additional stirring at room
temperature for 3 h, the reaction mixture was worked up as described
for the synthesis of 2a and 4a. Purification by flash chromatography
followed by preparative HPLC afforded 2b (40 mg, 10%) and 4b (8
mg, 2%).
2b: mp 133-134 °C; 1H NMR (270 MHz, CDCl3, 30 °C, 8.54 mM)
δ 7.94 (t, J ) 1.6 Hz, 4H), 7.86 (d, J ) 1.6 Hz, 8H), 4.29 (t, J ) 6.8
Hz, 8H), 1.77 (qunitet, J ) 6.8 Hz, 8H), 1.53-1.26 (m, 104H), 0.88
(t, J ) 6.6 Hz, 12H);13C NMR (67.5 MHz, CDCl3, 30 °C) δ 164.50
(s), 144.66 (d), 131.60 (s), 131.22 (d), 122.78 (s), 81.94 (s), 76.24 (s),
65.89 (t), 31.93 (t), 29.72 (t), 29.67 (t), 29.63 (t), 29.54 (t), 29.37 (t),
29.31 (t), 28.63 (t), 26.01 (t), 22.69 (t), 14.10 (q); IR (KBr) 2217, 1729,
1250, 893, 766, 720, 672 cm-1; UV (CHCl3, 25 °C) λmax (log ꢀ) 339
(5.15), 315 (5.31), 296 (5.02), 281 (4.73), 242 (5.00) nm; MS (FD)
m/z 1570 (M+ + H).
1
4b: mp 156-157 °C; H NMR (270 Hz, CDCl3, 30 °C) δ 8.16 (d,
Linear 6mer 13c. To a solution of TIPS-protected acetylene 12c
(332 mg, 0.150 mmol) in THF (7.5 mL) and water (0.3 mL) was added
dropwise 0.75 mL of a solution of TBAF (1 M) in THF. After being
stirred at room temperature for 4.5 h, the reaction mixture was worked
up as described for the synthesis of 7a and 8a. Purification by flash
chromatography followed by preparative HPLC afforded 13c as a pale
yellow solid (195 mg, 67%): mp 72-73 °C;1H NMR (270 MHz,
CDCl3) δ_8.20 (t, J ) 1.5 Hz, 8H), 8.17 (t, J ) 1.6 Hz, 2H), 8.15 (t,
J ) 1.6 Hz, 2H), 7.84 (t, J ) 1.6 Hz, 2H), 7.83 (t, J ) 1.6 Hz, 2H),
7.80 (t, J ) 1.6 Hz, 2H), 4.52-4.47 (m, 12H), 3.86-3.82 (m, 12H),
3.71-3.64 (m, 36H), 3.56-3.52 (m, 12H), 3.38 (s, 12H), 3.37 (s, 6H),
3.16 (s, 2H); 13C NMR (67.5 MHz, CDCl3, 30 °C) δ 164.56 (s), 164.38
(s), 139.64 (d), 139.48 (d), 134.10 (d), 133.73 (d), 133.56 (d), 131.24
(s), 131.01 (s), 123.22 (s), 122.65 (s), 122.59 (s), 122.57 (s), 122.27
(s), 81.35 (s), 80.33 (s), 80.07 (s), 80.05 (s), 80.04 (s), 79.83 (s), 79.28
(s), 75.31 (s), 75.26 (s), 75.23 (s), 74.91 (s), 71.95 (t), 71.94 (t), 70.71
(t), 70.65 (t), 70.64 (t), 69.01 (t), 64.76 (t), 64.67 (t), 59.03 (q); IR
(KBr) 3252, 3072, 2875, 2223, 2109, 1726 1585, 1433, 1261, 1112,
891, 764, 672 cm-1; MS (MALDI-TOF) m/z 1910.0 (M+ + Na). Anal.
Calcd for C108H110O30: C, 68.70; H, 5.87. Found: C, 68.44; H, 5.75.
Cyclization of Linear 4mer 11a: DBMs 2a and 4a. To a solution
of Cu(OAc)2 (8.88 g, 48.9 mmol) in 2 L of pyridine/benzene (3:2, v/v)
was added dropwise a solution of 11a (3.67 g, 3.20 mmol) in 500 mL
of the same solvent with a Hershberg dropping funnel over 7 h. After
additional stirring at room temperature for 6 d, the solvent was removed
in vacuo and the green residue was passed through a short plug of
SiO2 and the solvent was removed to afford crude products. Purification
by flash chromatography followed by preparative HPLC afforded 2a
(1.53 g, 43%) and 4a (310 mg, 9%).
2a: mp 218-220 °C; 1H NMR (400 MHz, CDCl3, 30 °C, 1.1 mM)
δ 8.02 (t, J ) 1.6 Hz, 4H), 7.96 (d, J ) 1.6 Hz, 8H), 4.32 (t, J ) 6.8
Hz, 8H), 1.78 (quintet, J ) 6.8 Hz, 8H), 1.50-1.26 (m, 40H), 0.91 (t,
J ) 7.0 Hz, 12H); 13C NMR (100 MHz, CDCl3, 30 °C) δ 164.34 (s),
144.51 (d), 131.46 (s), 131.08 (d), 122.69 (s), 81.92 (s), 76.26 (s), 65.89
(t), 31.90 (t), 29.53 (t), 29.28 (t), 28.69 (t), 26.10 (t), 22.74 (t), 14.19
(q); IR (KBr) 2216, 1727, 1270, 892, 767, 721, 670 cm-1; Raman 2222
cm-1; UV (CHCl3, 25 °C) λmax (log ꢀ) 338 (5.08), 315 (5.23), 296 (4.96),
281 (4.73), 265 (4.61) nm; MS (FAB) m/z 1121.4 (M+ + H).
4a: mp 208-209 °C; 1H NMR (400 MHz, CDCl3, 30 °C, 3.4 mM)
δ 8.12 (d, J ) 1.6 Hz, 16H), 7.75 (t, J ) 1.6 Hz, 8H), 4.31 (t, J ) 6.7
Hz, 16H), 1.77 (quintet, J ) 6.7 Hz, 16H), 1.43-1.26 (m, 80H), 0.90
(t, J ) 6.7 Hz, 24H); 13C NMR (100 MHz, CDCl3, 30 °C) δ 163.78
(s), 138.20 (d), 133.69 (d), 130.91 (s), 122.34 (s), 80.18 (s), 75.70 (s),
65.72 (t), 31.90 (t), 29.36 (t), 29.27 (t), 28.61 (t), 26.08 (t), 22.73 (t),
14.18 (q); IR (KBr) 2223, 1728, 1256, 893, 766, 674 cm-1; UV (CHCl3,
25 °C) λmax (log ꢀ) 337 (5.31), 315 (5.41), 296 (5.29), 276 (5.37), 263
(5.37) nm; MS (MALDI-TOF) m/z 2263.9 (M+ + Na).
J ) 1.0 Hz, 16H), 7.81 (t, J ) 1.0 Hz, 8H), 4.33 (t, J ) 6.7 Hz, 16H),
1.75 (quintet, J ) 6.7 Hz, 16H), 1.53-1.25 (m, 208H), 0.87 (t, J )
6.6 Hz, 24H); IR (KBr) 2224, 1728, 1257, 893, 766, 672 cm-1; UV
(CHCl3, 25 °C) λmax (log ꢀ) 338 (5.16), 315 (5.26), 296 (5.13), 276
(5.19) nm; MS (MALDI-TOF) m/z 3139.45 (M+ + H).
Cyclization of Linear 4mer 11c: DBMs 2c and 4c. To a solution
of Cu(OAc)2 (5.10 g, 28.1 mmol) in 765 mL of pyridine/benzene (3:2,
v/v) was added dropwise a solution of 11c (2.36 g, 1.87 mmol) in 765
mL of the same solvent over 74 h. After additional stirring at room
temperature for 3 h, the reaction mixture was worked up as described
for the synthesis of 2a and 4a. Purification by preparative HPLC
afforded 2c (1.00 g, 43%) and 4c (331 mg, 14%) both as pale yellow
solid.
1
2c: dec at 145 °C; H NMR (270 MHz, CDCl3, 30 °C, 2.3 mM) δ
8.06 (t, J ) 1.5 Hz, 8H), 8.04 (d, J ) 1.5 Hz, 4H), 4.51-4.48 (m,
8H), 3.86-3.83 (m, 8H), 3.74-3.65 (m, 24H), 3.57-3.54 (m, 8H),
3.39 (s, 12H); 13C NMR (75 MHz, CDCl3, 30 °C) δ 164.04 (s), 144.47
(d), 130.92 (s), 131.88 (d), 122.59 (s), 81.76 (s), 76.35 (s), 71.94 (t),
70.64 (t), 70.58 (t), 68.86 (t), 64.58 (t), 58.99 (q); IR (KBr) 2216, 1727,
1261, 892, 767, 721, 670 cm-1; MS (MALDI-TOF) m/z 1279.4 (M+
+ Na). Anal. Calcd for C72H72O20: C, 68.78; H, 5.77. Found: C, 68.60;
H, 5.68.
1
4c: dec at 109 °C; H NMR (270 MHz, CDCl3, 30 °C, 2.7 mM) δ
8.19 (d, J ) 1.6 Hz, 16H), 7.83 (t, J ) 1.5 Hz, 8H), 4.52-4.48 (m,
16H), 3.86-3.83 (m, 16H), 3.74-3.65 (m, 48H), 3.59-3.53 (m, 16H),
3.37 (s, 24H); 13C NMR (67.5 MHz, CDCl3, 30 °C) δ 164.34 (s), 139.75
(d), 134.03 (d), 131.25 (s), 122.58 (s), 80.05 (s), 75.26 (s), 71.98 (t),
70.74 (t), 70.66 (t), 69.03 (t), 64.79 (t), 59.06 (q); IR (KBr) 2360, 1726,
1261, 1111, 893, 766, 674 cm-1; MS (MALDI-TOF) m/z 2535.7 (M+
+ Na).
Transesterification of 2a: DBM 2e. To a stirred solution of octyl
ester 2a (18 mg, 1.6 µmol) and 1-butanol (1.5 mL, 16 mmol) in 3 mL
of toluene were added K2CO3 (11 mg, 8.1 µmol) and 18-crown-6 ether
(2 mg, 8.1 µmol). After the mixture was stirred at 60 °C for 16 h, the
solvent was removed with an evaporator. Purification by flash chro-
matography followed by preparative HPLC afforded butyl ester 2e as
a white solid (10 mg, 71%): dec at 247 °C; 1H NMR (400 MHz, CDCl3,
40 °C, 0.55 mM) δ 8.07 (t, J ) 1.6 Hz, 4H), 8.03 (d, J ) 1.6 Hz, 8H),
4.35 (t, J ) 6.9 Hz, 8H), 1.77 (quintet, J ) 6.9 Hz, 8H), 1.48 (m, 8H),
1.00 (t, J ) 7.4 Hz, 12H); IR (KBr) 2219, 1726, 1249, 892, 766, 671
cm-1; MS (MALDI-TOF) m/z 896.1 (M+).
Cyclization of Linear 6mer 13a: DBM 3a. To a solution of
Cu(OAc)2 (5.39 g, 29.7 mmol) in 1.6 L of pyridine/benzene (3:2, v/v)
was added dropwise a solution of 13a (1.0 g, 0.59 mmol) in 400 mL
of the same solvent over 53.5 h. After additional stirring at room
temperature for 2.5 h, the reaction mixture was worked up as described
for the synthesis of 2a and 4a. Purification by flash chromatography
followed by preparative HPLC afforded 3a as a pale yellow solid (270
Cyclization of 2mer 7b: DBMs 2b and 4b. To a solution of
Cu(OAc)2 (900 mg, 4.9 mmol) in 500 mL of pyridine/benzene (3:2,
v/v) was added dropwise a solution of 7b (390 mg, 0.5 mmol) in 50
1
mg, 27%): dec at 255 °C; H NMR (400 MHz, CDCl3, 30 °C, 14.9
mM) δ 7.23 (br s, 12H), 7.07 (br s, 6H), 4.02 (br s, 12H), 1.69 (m,
9
J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002 5363

A R T I C L E S
Tobe et al.
was also performed by the IGOR program.55 The obtained K2 (or KE),
δm, and δd (or δa) values are consistent with each other. The standard
deviations for the calculated values were estimated by using the IGOR
or SOLVSTAT56 macro of Excel.
Scheme 2. Equilibrium between Monomer (M), Dimer (M2), Trimer
(M3), and Higher Aggregates (Mn) and the Corresponding
Association Constants (K2, K3, ..., Kn)
Analysis of Self-Association by Vapor Pressure Osmometry.37
In the VPO measurements, the osmotic coefficient (φ) is related to
activity coefficient (γ) in the following eqs 13 and 14 according to the
Gibbs-Duhem equation:
mc
φ )
) 1 + ams + bms2 + cms3 + dms4 + ems5 + fms6 (13)
ms
3
2
4
3
5
4
6
5
7
6
ln γ ) 2ams + bms2 + cms3 + dms4 + ems5 + fms6 (14)
12H), 1.40 (m, 60H), 0.97 (t, J ) 6.7 Hz, 18H); 13C NMR (100 MHz,
CDCl3, 30 °C, 14.9 mM) δ 163.21 (s), 138.34 (d), 132.40 (d), 130.29
(s), 122.02 (s), 79.84 (s), 75.72 (s), 65.48 (t), 32.00 (t), 29.48 (t), 29.34
(t), 28.49 (t), 26.14 (t), 22.82 (t), 14.26(q); IR (KBr) 2223, 1728, 1249,
893, 766, 673 cm-1; Raman 2223 cm-1; UV (CHCl3) λmax (log ꢀ) 338
(5.21), 315 (5.30), 296 (5.08), 278 (4.93) nm; MS (MALDI-TOF) m/z
1680.54 (M+). Anal. Calcd for C114H120O12: C, 81.40; H, 7.19. Found:
C, 81.22; H, 7.26.
where ms is stoichiometric molal concentration and mc is colligative
molal concentration. Using eq 13, the experimental data obtained for
samples of several (typically 15-16) different concentrations ranging
from ca. 10-2 to 10-4 M are fitted to the sixth order polynominal, from
which the monomer concentrations (m1 ) γms) for each sample solution
were calculated. In the equal K model,44 eq 3 which relates φ/γ to K
and m1 is derived from the following eqs 15 and 16.
Cyclization of Linear 6mer 13b: DBM 3b. To a solution of
Cu(OAc)2 (5.16 g, 28.4 mmol) in 1.6 L of pyridine/benzene (3:2, v/v)
was added dropwise a solution of 13b (1.4 g, 0.58 mmol) in 400 mL
of the same solvent over 48 h. After additional stirring at room
temperature for 24 h, the reaction mixture was worked up as described
for the synthesis of 2a and 4a. Purification by flash chromatography
followed by preparative HPLC afforded 3b as a pale yellow solid (202
mg, 15%): mp 178-180 °C; 1H NMR (270 MHz, CDCl3, 30 °C, 12.1
mM) δ 7.69 (br s, 12H), 7.46 (br s, 6H), 4.20 (t, J ) 6.4 Hz, 12H),
1.77 (m, 12H), 1.42-1.27 (m, 156H), 0.88 (t, J ) 6.6 Hz, 18H); 13C
NMR (67.5 MHz, CDCl3, 30 °C, 12.1 mM) δ 163.87 (s), 139.16 (d),
133.11 (d), 130.98 (s), 122.41 (s), 80.06 (s), 75.71 (s), 65.73 (t), 31.94
(t), 29.76 (t), 29.70 (t), 29.65 (t), 29.42 (t), 29.39 (t), 28.60 (t), 26.05
(t), 22.70 (t), 14.10 (q); IR (KBr) 2217, 1730, 1255, 891, 765, 720,
671 cm-1; UV (CHCl3, 25 °C) λmax (log ꢀ) 338 (5.24), 315 (5.32), 296
(5.10), 278 (4.95), 243 (5.11) nm; MS (MALDI-TOF) m/z 2376.73
(M+ + Na), 2353.71 (M+). Anal. Calcd for C162H216O12: C, 82.61; H,
9.24. Found: C, 82.50; H, 9.39.
Cyclization of Linear 6mer 13c: DBM 3c. To a solution of
Cu(OAc)2 (150 mg, 0.826 mmol) in 25 mL of pyridine/benzene (3:2,
v/v) was added dropwise a solution of 13c (595 mg, 0.529 mmol) in
25 mL of the same solvent over 48 h. After additional stirring at room
temperature for 24 h, the reaction mixture was worked up as described
for the synthesis of 2a and 4a. Purification by flash chromatography
followed by preparative HPLC afforded 3c as a pale yellow solid (67
mg, 65%): dec at 242 °C;1H NMR (300 MHz, CDCl3) δ 8.06 (d, J )
1.6 Hz, 12H), 7.77 (t, J ) 1.6 Hz, 6H), 4.49-4.46 (m, 12H), 3.86-
3.83 (m, 12H), 3.75-3.67 (m, 36H), 3.66-3.54 (m, 12H), 3.39 (s, 18H);
13C NMR (67.5 MHz, CDCl3, 30 °C) δ 163.62 (s), 139.12 (d), 132.98
(d), 130.52 (s), 122.32 (s), 79.93 (s), 75.74 (s), 71.97 (t), 70.65 (t),
70.62 (t), 70.60 (t), 68.80 (t), 64.51 (t), 59.01 (q); IR (KBr) 3067, 2874,
2222, 1727, 1584, 1444, 1259, 1112, 890, 764, 673 cm-1; MS (MALDI-
TOF) m/z 1907.7 (M+ + Na). Anal. Calcd for C108H108O30: C, 68.78;
H, 5.77. Found: C, 68.44; H, 5.75.
mc ) m1 + m2 + m3 + ... + mn )
m1 + K2m12 + K2K3m13 + ... + K2K3...Knm1 (15)
n
m
c ) 1 + Km1 + K2m12 + ... + Kn-1m1
(16)
n-1
m1
Fitting a plot of the experimentally obtained φ/γ versus log m1 to
the master curves, which were plotted for [1 - (Km1)n]/(1 - Km1)
versus log Km1 for different n (n ) 1, 2, 3, 4, 5, 6, and ∞), yielded log
K as the translation along the abscissa.
In the second model (K2 * K),45 for each hypothesis I-IV, T defined
in eq 12 is related to log m1 as eqs 5, 7, 9, and 11. Fitting a plot of
experimentally obtained log[(ms - m1)/m1] versus log m1 to the master
curves of log T versus log Km1 for eqs 5, 7, 9, and 11 gave log K as
the translation along the abscissa and log(K/K2) as the translation along
the ordinate. Since, in general, the data fit best to the model IV among
the models I-IV, K2 and K shown in Table 5 were calculated by using
this model.
Acknowledgment. This work was partly supported by the
Ministry of Education, Science, Sports and Culture of Japan
through Grants-in-Aid. The authors are grateful to Professor T.
Kitayama of the same department for the use of vapor pressure
osmometer, Professor J. Moore of the University of Illinois for
sharing with us his unpublished results and discussion, and to
Shin-Etsu Chemical Co. for a generous gift of an organosilicon
reagent.
Supporting Information Available: Synthesis of 2e, 5a,b,
and 15, determination of association constants of DBMs 2a-c,
3a-c, and 4a, the linear oligomers 11c and 13c by H NMR
spectra and VPO, and UV spectra of 2c and 3c in polar solvents
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
1
Analysis of Self-Association by 1H NMR Spectra.37 For treatment
of the experimental NMR data, the sum of deviations between the
chemical shifts (δcalcd) calculated according to eq 1 for the dimer model
or eq 2 for the infinite association model and those of experimental
chemical shifts (δobs) were calculated, and K2 (or KE), δm, and δd (or
δa) were optimized to reduce the sum of deviations by using the “solver”
function implemented in the Excel program.54 The same curve fitting
JA012458M
(54) Microsoft Excel 98; Microsoft Corporation: Tokyo, Japan.
(55) IGOR pro, version 3.03; Wave Metrics, Inc.: Lake Oswego, OR.
(56) Billo, E. J. Excel for Chemists; Wiley-VCH: New York, 1997; pp 295-
300.
9
5364 J. AM. CHEM. SOC. VOL. 124, NO. 19, 2002