Supramolecular daisy chains

Article abstract of DOI:10.1021/jo010405h

Two series of self-complementary daisy chain monomers, in which a secondary ammonium ion-containing arm is grafted onto a macrocycle with either a [24]- or [25]crown-8 constitution, have been synthesized. In the solid- and 'gas'-phases, the parent [24]crown-8-based monomer forms dimeric superstructures, as revealed by X-ray crystallography and mass spectrometry, respectively. Elucidation of the complicated solution-phase behavior of this compound was facilitated by the synthesis and study of both deuterated, and fluorinated, analogues. These investigations revealed that the cyclic dimeric superstructure also dominates in solution, except when extremes of either concentration (low), temperature (high), or solvent polarity (highly polar, e.g., dimethyl sulfoxide) are employed. Whereas, upon aggregation, the [24]crown-8-based daisy chain monomers have the capacity to form stereoisomeric superstructures further complicating the study of this series of compounds. The assembly of [25]crown-8-based monomers gives only achiral superstructures. The weaker association exhibited between secondary dialkylammonium ions and crown ethers with a [25]crown-8 constitution, however, resulted in limited oligomerization-only dimeric and trimeric superstructures were formed at experimentally attainable concentrations-of [25]crown-8-based daisy chain monomers.

Full text of DOI:10.1021/jo010405h

J . Org. Chem. 2001, 66, 6857-6872
6857
Su p r a m olecu la r Da isy Ch a in s†
Stuart J . Cantrill,‡ Gilmer J . Youn, and J . Fraser Stoddart*
Department of Chemistry and Biochemistry, University of California, Los Angeles, California 90095
David J . Williams*
Department of Chemistry, Imperial College, South Kensington, London SW7 2AY, UK
stoddart@chem.ucla.edu
Received April 19, 2001
Two series of self-complementary daisy chain monomers, in which a secondary ammonium ion-
containing arm is grafted onto a macrocycle with either a [24]- or [25]crown-8 constitution, have
been synthesized. In the solid- and ‘gas’-phases, the parent [24]crown-8-based monomer forms
dimeric superstructures, as revealed by X-ray crystallography and mass spectrometry, respectively.
Elucidation of the complicated solution-phase behavior of this compound was facilitated by the
synthesis and study of both deuterated, and fluorinated, analogues. These investigations revealed
that the cyclic dimeric superstructure also dominates in solution, except when extremes of either
concentration (low), temperature (high), or solvent polarity (highly polar, e.g., dimethyl sulfoxide)
are employed. Whereas, upon aggregation, the [24]crown-8-based daisy chain monomers have the
capacity to form stereoisomeric superstructures further complicating the study of this series of
compounds. The assembly of [25]crown-8-based monomers gives only achiral superstructures. The
weaker association exhibited between secondary dialkylammonium ions and crown ethers with a
[25]crown-8 constitution, however, resulted in limited oligomerizationsonly dimeric and trimeric
superstructures were formed at experimentally attainable concentrationssof [25]crown-8-based
daisy chain monomers.
In tr od u ction
The rise of supramolecular chemistry1, and its precise
exploitation of delicate noncovalent interactions, has
spurred investigations2 into alternative approaches for
polymer synthesis, challenging3 the covalent bond’s
monopoly of the macromolecular world. Whereas molec-
ular polymerization relies upon the formation (Figure 1,
top) of covalent bonds between monomeric building
blocks, the propagation step in a supramolecular poly-
merization proceeds via the formation (Figure 1, bottom)
of noncovalent bonds. Consequently, it is the inherent
reversibility associated with the self-assembly4 of such
a dynamic aggregate that is anticipated5 to bestow, upon
the resulting polymer, properties that are distinct from
those observed for traditional covalently linked macro-
molecules. Foremost, polymerization performed under
such a thermodynamically controlled regime allows2c,5 for
a high degree of architectural control; incorrect or
F igu r e 1. Top: A schematic representation of the covalent
polymerization of monomer M1. Bottom: A schematic repre-
sentation of the noncovalent polymerization of a different
monomer, M2.
unspecific chain extensions are simply reversed in the
constantly evolving system as it strives to find a local
energy minimum. Furthermore, once assembled, the
dynamic nature of such an assembly does not cease to
be important. In fact, this very attribute remains of
paramount importance, and renders such a species
extremely sensitive (i.e., responsive) to external environ-
mental factors5 such as mechanical stress or temperature.
Once sheared, a material linked through noncovalent
* To whom correspondence should be addressed. Tel: (310) 206 7078.
Fax: (310) 206 1843.
† Molecular Meccano, Part 64. For Part 63, see: Ashton, P. R.;
Baldoni, V.; Balzani, V.; Credi, A.; Hoffmann, H. D. A.; Mart´ınez-D´ıaz,
M.-V.; Raymo, F. M.; Stoddart, J . F.; Venturi, M. Chem. Eur. J . 2001,
7, 3482-3493.
(4) (a) Whitesides, G. M.; Mathias, J . P.; Seto, C. T. Science 1991,
154, 1312-1319. (b) Lawrence, D. S.; J iang, T.; Levett, M. Chem. Rev.
1995, 95, 2229-2260. (c) Philp, D.; Stoddart, J . F. Angew. Chem., Int.
Ed. Engl. 1996, 35, 1154-1196. (d) Stang, P. J .; Olenyuk, B. Acc. Chem.
Res. 1997, 30, 502-518. (e) Conn, M. M.; Rebek, J ., J r. Chem. Rev.
1997, 97, 1647-1668. (f) Linton, B.; Hamilton, A. D. Chem. Rev. 1997,
97, 1669-1680. (g) Fujita, M. Chem. Soc. Rev. 1998, 27, 417-425. (h)
Emrick, T.; Fre´chet, J . M. J . Curr. Opin. Colloid Interface Sci. 1999,
4, 15-23. (i) Sijbesma, R. P.; Meijer, E. W. Curr. Opin. Colloid Interface
Sci. 1999, 4, 24-32. (j) Lindoy, L. F.; Atkinson, I. M. Self-Assembly in
Supramolecular Systems; Stoddart, J . F., Ed.; RSC: Cambridge, 2000.
(5) Sijbesma, R. P.; Beijer, F. H.; Brunsveld, L.; Folmer, B. J . B.;
Ky Hirschberg, J . H. K.; Lange, R. F. M.; Lowe, J . K. L.; Meijer, E. W.
Science 1997, 278, 1601-1604.
‡ Current address: Division of Chemistry and Chemical Engineer-
ing, California Institute of Technology, Pasadena, CA 91125.
(1) (a) Lehn, J .-M. Supramolecular Chemistry; VCH: Weinheim,
1995. (b) Comprehensive Supramolecular Chemistry; Atwood, J . L.;
Davies, J . E. D.; MacNicol, D. D.; Vo¨gtle, F., Eds.; Pergamon: Oxford,
1996; 11 vols.
(2) (a) Archer, R. D. Coord. Chem. Rev. 1993, 128, 49-68. (b)
Rehahn, M. Acta Polymer. 1998, 49, 201-224. (c) Moore, J . S. Curr.
Opin. Colloid Interface Sci. 1999, 4, 108-116. (d) Supramolecular
Polymers; Ciferri, A., Ed.; Marcel Dekker: New York, 2000.
(3) Zimmerman, N.; Moore, J . S.; Zimmerman, S. C. Chem. Ind.
1998, 604-610.
10.1021/jo010405h CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/26/2001

6858 J . Org. Chem., Vol. 66, No. 21, 2001
Cantrill et al.
interactions retains, under the appropriate conditions,
the capacity to ‘heal’ itself, reforming those bonds that
were broken under the application of mechanical stress.
Conversely, no correspondingly simple repair mechanism
exists for covalent polymers. Second, changes in temper-
ature should affect6 directly (and reversibly) the degree
of polymerization (DP) exhibited by a supramolecular
polymer, and in turn, may be used to attenuate bulk
properties such as viscosity and/or rheology. In contrast,
although potentially significant,7 temperature effects
upon covalent polymers are generally less dramatic and
are rarely8 utilized in order to alter, reversibly, the degree
of polymerization.
Although both (i) metal-ligand,6,9 and (ii) π-π stack-
ing10 interactions have been utilized as the ‘glue’ with
which monomers have been strung together to form
supramolecular polymers, hydrogen bonding interactions
have been exploited5,11-15 far more widely. This prefer-
F igu r e 2. A schematic representation depicting the formation
of a threaded 1:1 complex (a pseudorotaxane) between two
complementary species wherein the cavity of a suitably sized
ring is skewered by a linear rod.
ence arises as a consequence of the easily tunable, and
therefore favorable, thermodynamic and kinetic param-
eters associated with hydrogen bonding interactions.
Whereas the reversibility of some metal-ligand interac-
tions is limited, and the strength of π-π stacking
interactions is not sufficient enough to produce polymers
with high DPs, hydrogen bonding interactions can be
fine-tuned to deliver the desired properties. Although
single,11 double,12 and triple13 hydrogen bond manifolds
have been employed in the construction of supramolecu-
lar polymers, the most successful has been that involving
a quadruple14 couple. Meijer and co-workers have ex-
ploited the strong dimerization16 of 2-ureido-4-pyrimidone
derivatives, with Ka values in excess of 106 M-1 in CHCl3,
for the reversible formation14 of hydrogen-bonded supra-
molecular polymers and networks.
Rather than the simple face-to-face association (vide
supra) of monomers, however, more complex topological
interactions may be contrived. The concept (Figure 2) of
threading a rod-shaped molecule through the macrocyclic
cavity of another ring-shaped onesto create17 an inter-
woven host-guest complex (often referred to as a
pseudorotaxane18)scan be applied to the propagation step
of a supramolecular polymerization. Covalently coupling
the two mutually recognizing components to one an-
other19 affords (Figure 3) a self-complementary mono-
(6) The degree of polymerization (DP) depends critically upon the
association constant (Ka) for two monomer units (see: Michelsen, U.;
Hunter, C. A. Angew. Chem., Int. Ed. 2000, 39, 764-767) and can be
approximated by using the following expression: DP ≈ x(Ka[mono-
mer]). Consequently, since Ka ) (T∆S - ∆H)/RT, changing the
temperature can alter dramatically the value of Ka, and hence also
that of DP.
(7) For recent examples of thermoresponsive polymers, see: (a)
Everlof, G. J .; J aycox, G. D. Polymer 2000, 41, 6527-6536. (b) Bignotti,
F.; Penco, M.; Sartore, L.; Peroni, I.; Mendichi, R.; Casolaro, M.;
D’Amore, A. Polymer 2000, 41, 8247-8256. (c) Pan, H.-Z.; Yan, Y.;
Tang, Li.; Wu, Z.-Q.; Li, F.-M. Macromol. Rapid Commun. 2000, 21,
567-573. (d) Kurisawa, M.; Yokoyama, M.; Okano, T. J . Controlled
Release 2000, 69, 127-137. (e) Kim, S. Y.; Cho, S. M.; Lee, Y. M.; Kim,
S. J . J . Appl. Polym. Sci. 2000, 78, 1381-1391.
(8) The exception to this statement relates to equilibrium polymer-
izations, in which dynamic covalent chemistry is employed for the
reversible formation of covalently bonded polymers. For a general
background on this topic, see: (a) Sawada, H. J . Macromol. Sci. Rev.
Macromol. Chem. 1972, C8, 235-288. For specific examples describing
how temperature effects can be used to control the DP in a given
polymer system, see: (b) Chikaoka, S.; Takata, T.; Endo, T. Macro-
molecules 1991, 24, 6557-6562. (c) Marsella, M. J .; Maynard, H. D.;
Grubbs, R. H. Angew. Chem., Int. Ed. Engl. 1997, 36, 1101-1103. (d)
Lowe, T. L.; Tenhu, H. Macromolecules 1998, 31, 1590-1594.
(9) (a) Constable, E. C. Chem. Commun. 1997, 1073-1080. (b) Kelch,
S.; Rehahn, M. Macromolecules 1997, 30, 6185-6193. (c) Kelch, S.;
Rehahn, M. Macromolecules 1999, 32, 5818-5828. (d) Gardner, M.;
Guerin, A. J .; Hunter, C. A.; Michelsen, U.; Rotger, C. New J . Chem.
1999, 309-316. (e) Wojaczynski, J .; Latos-Grazynski, L. Coord. Chem.
Rev. 2000, 204, 113-171. (f) Ogawa, K.; Kobuke, Y. Angew. Chem.,
Int. Ed. 2000, 39, 4070-4073. (g) Ellis, W. W.; Schmitz, M.; Arif, A.
A.; Stang, P. J . Inorg. Chem. 2000, 39, 2547-2557. In contrast with
these metallo-supramolecular polymers, which exhibit macromolecular
properties in solution, much research has been done on the crystal
engineering of solid-state coordination networks that are also referred
to as ‘polymers’. See: (h) Hirsch, K. A.; Wilson, S. R.; Moore, J . S. Inorg.
Chem. 1997, 36, 2960-2968. (i) Blake, A. J .; Champness, N. R.;
Hubberstey, P.; Li, W. S.; Withersby, M. A.; Schroder, M. Coord. Chem.
Rev. 1999, 183, 117-138. (j) Shimizu, G. K. H.; Enright, G. D.; Rego,
G. S.; Ripmeester, J . A. Can. J . Chem. 1999, 77, 313-318. (k) Carlucci,
L.; Ciani, G.; Proserpio, D. M. Chem. Commun. 1999, 449-450. (l)
Groeneman, R. H.; MacGillivray, L. R.; Atwood, J . L. Inorg. Chem.
1999, 38, 208-209. (m) Gudbjartson, H.; Biradha, K.; Poirier, K. M.;
Zaworotko, M. J . J . Am. Chem. Soc. 1999, 121, 2599-2600. (n) Kleina,
C.; Graf, E.; Hosseini, M. W.; De Cian, A.; Fischer, J . Chem. Commun.
2000, 239-240. (o) Hong, M.; Zhao, Y.; Su, W.; Cao, R.; Fujita, M.;
Zhou, Z.; Chan, A. S. C. Angew. Chem., Int. Ed. 2000, 39, 2468-2470.
(p) Robson, R. J . Chem. Soc., Dalton Trans. 2000, 3735-3744. (q)
Biradha, K.; Fujita, M. J . Chem. Soc., Dalton Trans. 2000, 3805-3810.
(r) Sharma, A. C.; Borovik, A. S. J . Am. Chem. Soc. 2000, 122, 8946-
8955.
(12) (a) Ducharme, Y.; Wuest, J . D. J . Org. Chem. 1988, 53, 5787-
5789. (b) Lillya, C. P.; Baker, R. J .; Hutte, S.; Winter, H. H.; Lin, Y.
G.; Shi, J . F.; Dickinson, L. C.; Chen, J . C. W. Macromolecules 1992,
25, 2076-2080. (c) Abed, S.; Boileau, S.; Bouteiller, L. Macromolecules
2000, 33, 8479-8487.
(13) (a) Kotera, M.; Lehn, J .-M.; Vigneron, J .-P. J . Chem. Soc., Chem.
Commun. 1994, 197-199. (b) Russell, K. C.; Lehn, J .-M.; Kyritsakas,
N.; DeCian, A.; Fischer, J . New J . Chem. 1998, 123-128. (c) Choi, I.
S.; Li, X.; Simanek, E. E.; Akaba, R.; Whitesides, G. M. Chem. Mater.
1999, 11, 684-690. (d) Klok, H.-A.; J olliffe, K. A.; Schauer, C. L.; Prins,
L. J .; Spatz, J . P.; Mo¨ller, M.; Timmerman, P.; Reinhoudt, D. N. J .
Am. Chem. Soc. 1999, 121, 7154-7155.
(14) (a) Folmer, B. J . B.; Cavini, E.; Sijbesma, R. P.; Meijer, E. W.
Chem. Commun. 1998, 1847-1848. (b) Hirschberg, J . H. K. K.; Beijer,
F. H.; van Aert, H. A.; Magusim, P. C. M. M.; Sijbesma, R. P.; Meijer,
E. W. Macromolecules 1999, 32, 2696-2705. (c) Lange, R. F. M.; Van
Gurp, M.; Meijer, E. W. J . Polym. Sci., Part A: Polym. Chem. 1999,
37, 3657-3670. (d) Folmer, B. J . B.; Sijbesma, R. P.; Versteegen, R.
M.; van der Rijt, J . A. J .; Meijer, E. W. Adv. Mater. 2000, 12, 874-
878. (e) Boileau, S.; Bouteiller, L.; Laupreˆtre, F.; Lortie, F. New J .
Chem. 2000, 845-848. (f) Folmer, B. J . B.; Sijbesma, R. P.; Meijer, E.
W. J . Am. Chem. Soc. 2001, 123, 2093-2094.
(15) (a) Castellano, R. K.; Rudkevich, D. M.; Rebek, J ., J r. Proc. Natl.
Acad. Sci. U.S.A. 1997, 94, 7132-7137. (b) Castellano, R. K.; Rebek,
J ., J r. J . Am. Chem. Soc. 1998, 120, 3657-3663. (c) Castellano, R. K.;
Rebek, J ., J r. Polym. Mater. Sci. Eng. 1999, 80, 16-17. (d) Castellano,
R. K.; Nuckolls, C.; Eichhorn, S. H.; Wood, M. R.; Lovinger, A. J .;
Rebek, J ., J r. Angew. Chem., Int. Ed. 1999, 38, 2603-2606.
(16) (a) Beijer, F. H.; Kooijman, H.; Spek, A. L.; Sijbesma, R. P.;
Meijer, E. W. Angew. Chem., Int. Ed. 1998, 37, 75-78. (b) Beijer, F.
H.; Sijbesma, R. P.; Kooijman, H.; Spek, A. L.; Meijer, E. W. J . Am.
Chem. Soc. 1998, 120, 6761-6769. (c) El-ghayoury, A.; Peeters, E.;
Schenning, A. P. H. J .; Meijer, E. W. Chem. Commun. 2000, 1969-
1970. For other recent examples of self-complementary quadruple
hydrogen-bonded dimers yet to be utilized in the construction of larger
aggregates, see: (d) Corbin, P. S.; Zimmerman, S. C. J . Am. Chem.
Soc. 1998, 120, 9710-9711. (e) Davis, A. P.; Draper, S. M.; Dunne, G.;
Ashton, P. R. Chem. Commun. 1999, 2265-2266.
(10) (a) Alcock, N. W.; Barker, P. R.; Haider, J . M.; Hannon, M. J .;
Painting, C. L.; Pikramenou, Z.; Plummer, E. A.; Rissanen, K.;
Saarenketo, P. J . Chem. Soc., Dalton Trans. 2000, 1447-1461. (b)
Hirschberg, J . H. K. K.; Brunsveld, L.; Ramzi, A.; Vekemans, J . A. J .
M.; Sijbesma, R. P.; Meijer, E. W. Nature 2000, 407, 167-170.
(11) (a) Bladon, P.; Griffin, A. C. Macromolecules 1993, 26, 6604-
6610. (b) Alexander, C.; J ariwala, C. P.; Lee, C. M.; Griffin, A. C.
Macromol. Symp. 1994, 77, 283-294. (c) Lee, C. M.; J ariwala, C. P.;
Griffin, A. C. Polymer 1994, 35, 4550-4554. (d) St Pourcain, C. B.;
Griffin, A. C. Macromolecules 1995, 28, 4116-4121. (e) Lee, C. M.;
Griffin, A. C. Macromol. Symp. 1997, 117, 281-290.

Supramolecular Daisy Chains
J . Org. Chem., Vol. 66, No. 21, 2001 6859
F igu r e 3. Conceptually, when two mutually recognizing species are merged together, a single self-complementary entity is formed.
This daisy-chain monomer can then self-assemble to form both cyclic (c) and acyclic (a) interwoven superstructures. Note: the
numerical descriptor refers to the number of monomer units that make up any particular superstructure, i.e., a cyclic dimer is a
[c2]daisy chain.
mer,20 which has the capacity to self-assemble into either
linear21 or cyclic22 daisy chain23 arrays.24 The research
reported in this paper focuses upon the study of self-
complementary molecules containing both crown ether
and dibenzylammonium ion recognition sites. The design,
synthesis, characterization, and ultimately aggregation
behavior of these daisy chain monomers, which are based
upon a well-understood25 supramolecular interaction, are
described.
Resu lts a n d Discu ssion
[24]Cr ow n -8-Ba sed System s. Conceptually, the ini-
tial target molecule (1-H‚PF6) arose as a consequence of
(20) (a) Ashton, P. R.; Baxter, I.; Cantrill, S. J .; Fyfe, M. C. T.; Glink,
P. T.; Stoddart, J . F.; White, A. J . P.; Williams, D. J . Angew. Chem.,
Int. Ed. 1998, 37, 1294-1297. (b) Ashton, P. R.; Parsons, I. W.; Raymo,
F. M.; Stoddart, J . F.; White, A. J . P.; Williams, D. J .; Wolf, R. Angew.
Chem., Int. Ed. 1998, 37, 7, 1913-1916. (c) Yamaguchi, N.; Nagvekar,
D. S.; Gibson, H. W. Angew. Chem., Int. Ed. 1998, 37, 2361-2364. (d)
Mirzoian, A.; Kaifer, A. E. Chem. Commun. 1999, 1603-1604. (e)
Bu¨lger, J .; Sommerdijk, N. A. J . M.; Visser, A. J . W. G.; van Hoek, A.;
Nolte, R. J . M.; Engbersen, J . F. J .; Reinhoudt, D. N. J . Am. Chem.
Soc. 1999, 121, 28-33. (f) Nielsen, M. B.; Hansen, J . G.; Becher, J .
Eur. J . Org. Chem. 1999, 2807, 7-2815. (g) J ime´nez, M. C.; Dietrich-
Buchecker, C.; Sauvage, J .-P.; De Cian, A. Angew. Chem., Int. Ed. 2000,
39, 1295-1298.
(21) Appropriately monosubstituted cyclo-oligosaccharides have been
shown to crystallize to give arrays of this type in the solid state, see:
(a) Hirotsu, K.; Higuchi, T.; Fujita, K.; Ueda, T.; Shinoda, A.; Imoto,
T.; Tabushi, I. J . Org. Chem. 1982, 47, 1143-1144. (b) Mentzafos, D.;
Terzis, A.; Coleman, A. W.; de Rango, C. Carbohydr. Res. 1996, 282,
125-135. (c) Liu, Y.; You, C.-C.; Zhang, M.; Weng, L.-H.; Wada, T.;
Inoue, Y. Org. Lett. 2000, 2, 2761-2763.
(22) Cyclic arrays are formed whenever chain termination of a
propagating acyclic species occurs via intramolecular ‘back-biting’. For
examples, see: ref 20a,b,d,g. In some less geometrically demanding
cases, the monomeric species is also capable of biting its own tail, to
form an intramolecular self-complexed species. See: ref 20e,f, and
Ashton, P. R.; Ballardini, R.; Balzani, V.; Boyd, S. E.; Credi, A.;
Gandolfi, M. T.; Go´mez-Lo´pez, M.; Iqbal, S.; Philp, D.; Preece, J . A.;
Prodi, L.; Ricketts, H. G.; Stoddart, J . F.; Tolley, M. S.; Venturi, M.;
White, A. J . P.; Williams, D. J . Chem. Eur. J . 1997, 3, 152-170.
(23) The term ‘daisy chain’ is used to describe an interwoven chain,
usually made of flowers (hence ‘daisy’), in which each unit acts as a
‘donor’ and an ‘acceptor’ for a threading interaction. The standard
definition (Webster’s Third New International Dictionary; Babcock-
Gove, P., Ed.; Merriam-Webster: Springfield MA, 1993) of a ‘daisy
chain’ is; “a string of daisies with stems linked to form a chain,..... such
a chain carried by chosen students at a class day or other celebration
in some women’s colleges”. With regard to chemical systems, it should
be noted that Busch and co-workers coined their own term for such a
supramolecular polymeric architecture, calling it a ‘Figure 9 pseudo-
polyrotaxane’. See: Hubin, T. J .; Busch, D. H. Coord. Chem. Rev. 2000,
200-202, 5-52.
(24) Small cyclic daisy chain structures have been trapped kineti-
cally upon postassembly covalent modification of the corresponding
complexes. For examples of dimers, see: (a) J ime´nez, M. C.; Dietrich-
Buchecker, C.; Sauvage, J .-P. Angew. Chem., Int. Ed. 2000, 39, 3284-
3287. (b) Fujimoto, T.; Sakata, Y.; Kaneda, T. Chem. Commun. 2000,
2143-2144. (c) Onagi, H.; Easton, C. J .; Lincoln, S. F. Org. Lett. 2001,
3, 1041-1044. For a captured trimer, see: (d) Hoshino, T.; Miyauchi,
M.; Kawaguchi, Y.; Yamaguchi, H.; Harada, A. J . Am. Chem. Soc. 2000,
122, 9876-9877. An alternative approach for the construction of these
novel mechanically interlocked assemblies, which relies upon covalent,
rather than noncovalent, oligo/polymerization, was communicated
before any of the above reports. See: (e) Rowan, S. J .; Cantrill, S. J .;
Stoddart, J . F.; White, A. J . P.; Williams, D. J . Org. Lett. 2000, 2, 759-
762.
(17) For recent examples, see: (a) J eon, Y. M.; Whang, D.; Kim, J .;
Kim, K. Chem. Lett. 1996, 503-504. (b) Mirzoian, A.; Kaifer, A. E.
Chem. Eur. J . 1997, 3, 1052-1058. (c) Sleiman, H.; Baxter, P. N. W.;
Lehn, J .-M.; Airola, K.; Rissanen, K. Inorg. Chem. 1997, 36, 4734-
4742. (d) Loeb, S. J .; Wisner, J . A. Angew. Chem., Int. Ed. 1998, 37,
2838-2840. (e) Smith, A. C.; Macartney, D. H. J . Org. Chem. 1998,
63, 9243-9251. (f) Asakawa, M.; Ashton, P. R.; Balzani, V.; Boyd, S.
E.; Credi, A.; Mattersteig, G.; Menzer, S.; Montalti, M.; Raymo, F. M.;
Ruffilli, C.; Stoddart, J . F.; Venturi, M.; Williams, D. J . Eur. J . Org.
Chem. 1999, 985, 5-994. (g) Cantrill, S. J .; Fyfe, M. C. T.; Heiss, A.
M.; Stoddart, J . F.; White, A. J . P.; Williams, D. J . Org. Lett. 2000, 2,
61-64. (h) Chang, T.; Heiss, A. M.; Cantrill, S. J .; Fyfe, M. C. T.; Pease,
A. R.; Rowan, S. J .; Stoddart, J . F.; Williams, D. J . Org. Lett. 2000, 2,
2943-2946. (i) Chang, T.; Heiss, A. M.; Cantrill, S. J .; Fyfe, M. C. T.;
Pease, A. R.; Rowan, S. J .; Stoddart, J . F.; White, A. J . P.; Williams,
D. J . Org. Lett. 2000, 2, 2947-2950. (j) Ryan, D.; Rao, S. N.;
Rensmo, H.; Fitzmaurice, D.; Preece, J . A.; Wenger, S.; Stoddart, J .
F.; Zaccheroni, N. J . Am. Chem. Soc. 2000, 122, 6252-6257. (k)
Chichak, K.; Walsh, M. C.; Branda, N. R. Chem. Commun. 2000, 847-
848.
(18) These complexes, in which an acyclic component is threaded
through a cyclic one, are simply rotaxanes which are lacking the bulky
end-groups, and are, therefore, termed pseudorotaxanes. In fact,
pseudorotaxanes occupy a pivotal position in the chemistry of mechani-
cally interlocked compounds, as they can serve as intermediates in
the syntheses of both catenanes and rotaxanes. For recent examples,
see: (a) Hansen, J . G.; Feeder, N.; Hamilton, D. G.; Gunter, M. J .;
Becher, J .; Sanders, J . K. M. Org. Lett. 2000, 2, 449-452. (b) Raehm,
L.; Hamann, C.; Kern, J .-M.; Sauvage, J .-P. Org. Lett. 2000, 2, 1991-
1994. (c) Safarowsky, O.; Vogel, E.; Vo¨gtle, F. Eur. J . Org. Chem. 2000,
499-505. (d) Cabezon, B.; Cao, J .; Raymo, F. M.; Stoddart, J . F.; White,
A. J . P.; Williams, D. J . Chem. Eur. J . 2000, 6, 2262-2273. (e) Reuter,
C.; Vo¨gtle, F. Org. Lett. 2000, 2, 593-595. (f) Skinner, P. J .; Blair, S.;
Kataky, R.; Parker, D. New J . Chem. 2000, 24, 265-268. (g) Kawagu-
chi, Y.; Harada, A. Org. Lett. 2000, 2, 1353-1356. (h) Loeb, S. J .;
Wisner, J . A. Chem. Commun. 2000, 845-846. (i) Buston, J . E. H.;
Young, J . R.; Anderson, H. L. Chem. Commun. 2000, 905-906. (j)
Cantrill, S. J .; Fulton, D. A.; Heiss, A. M.; Pease, A. R.; Stoddart, J .
F.; White, A. J . P.; Williams, D. J . Chem. Eur. J . 2000, 6, 2274-2287.
(k) Tachibana, Y.; Kihara, N.; Ohga, Y.; Takata, T. Chem. Lett. 2000,
806-807.
(19) An alternative approach requires the self-assembly of two
homoditopic monomers, i.e., AA and BB building blocks, rather than
a self-complementary AB one. See: (a) Yamaguchi, N.; Gibson, H. W.
Angew. Chem., Int. Ed. 1999, 38, 143-147. (b) Yamaguchi, N.; Gibson,
H. W. Chem. Commun. 1999, 789-790. Interwoven supramolecular
polymers constructed in this fashion would serve as precursors for a
novel interlocked macromolecular architecture, recently termed a
‘Figure 8 polyrotaxane’ by Busch et al. See: Hubin, T. J .; Kolchinski,
A. G.; Vance, A. L.; Busch, D. H. Adv. Supramol. Chem. 1999, 10, 237-
357.
(25) Ashton, P. R.; Chrystal, E. J . T.; Glink, P. T.; Menzer, S.;
Schiavo, C.; Spencer, N.; Stoddart, J . F.; Tasker, P. A.; White, A. J . P.;
Williams, D. J . Chem. Eur. J . 1996, 2, 709-728.

6860 J . Org. Chem., Vol. 66, No. 21, 2001
Cantrill et al.
F igu r e 5. The LSI-mass spectrum of 1-H‚PF6.
intense as the peak located at m/z ) 568, which could
correspond to the species [1-H]+. However, closer inspec-
tion of the expanded spectrum revealed that the isotope
peaks, in the area surrounding the ‘[1-H]+’ peak, are
separated by 0.5 mass units, indicating the presence of
a doubly charged species with a mass corresponding to a
dimeric entity, thus providing evidence for the existence
of the doubly charged dimer [(1-H)2]2+. It is also notable
that no higher order oligomeric species were detected,
inferring that they are either (i) not formed, (ii) do not
survive the conditions of the analysis, or (iii) are not
amenable to mass spectrometric analysis, i.e., do not
ionize as well as their smaller siblings.
F igu r e 4. Combining the features of both DB24C8 (2) and
dibenzylammonium hexafluorophosphate (3-H‚PF6) into one
and the same molecule results in the design of a self-comple-
mentary daisy chain monomer 1-H‚PF6.
splicing (Figure 4) both the dibenzo[24]crown-8 (2) and
dibenzylammonium hexafluorophosphate (3-H‚PF6) struc-
tural motifs into one and the same molecule. Upon
inspection of space-filling molecular models, it was ap-
parent that such a structure satisfied one of the most
important design criteria, i.e., that intramolecular self-
complexation is geometrically unfavorable, if not impos-
sible.
Th e P a r en t Mon om er . The synthesis of this self-
complementary daisy chain monomer is outlined in
Scheme 1. Alkylation of 3,4-dihydroxybenzaldehyde (4)
with the chloride 5 proceeded under basic conditions in
DMF to afford the diol 6 in very good yield. Tosylation of
6, under standard conditions, gave the ditosylate 7, which
was subsequently reacted with catechol (8) to give the
formyl-substituted DB24C8 derivative 9 in reasonable
yield. Condensation of 9 with benzylamine (10), followed
by borohydride reduction, protonation (HCl), and coun-
terion exchange from Cl- to PF6-, afforded the target
compound 1-H‚PF6 in a 73% yield.
Ma ss Sp ectr om etr y. The liquid secondary ion (LSI)
mass spectrum of the salt 1-H‚PF6 revealed (Figure 5)
intense peaks, encountered at m/z ) 1281 and 1135,
corresponding to the creation of dimeric supermolecules
in the ‘gas phase’. These peaks can be identified as the
supramolecular ions [(1-H)2(PF6)]+ and [(1-H)(1)]+, re-
spectively, and indicate the aggregation of two monomer
units. The peak at m/z ) 1135 was observed to be the
spectrum’s base peaksit was approximately twice as
X-r a y Cr ysta llogr a p h y. The trifluoroacetate26 salt
1-H‚O2CCF3 crystallized20a on standing from a mixture
of (EtOAc/C6H14/CD3CN, ca. 10:10:1) affording crystals
suitable for single-crystal X-ray diffraction analysis. The
X-ray structural analysis of 1-H‚O2CCF3 revealed (Figure
6) the formation of a C2 symmetric head-to-tail dimeric
structure in which the benzylammonium cationic portions
of each component thread simultaneously through the
DB24C8 portions of their adjacent counterparts. The
pairs of molecules are stabilized by a combination of π-π
stacking of the substituted catechol rings (interplanar
and centroid-centroid separations of 3.36 and 3.55 Å,
respectively) and [N+-H‚‚‚O] hydrogen bonds between
+
one of the hydrogen atoms on each NH2 center and a
polyether oxygen atom within one of the linkages of each
DB24C8 component ([N+‚‚‚O], [H‚‚‚O] distances are 2.91
and 2.02 Å, respectively, with an [N+-H‚‚‚O] angle of
170°). Interestingly, in this instance, the interaction
between the two achiral 1-H+ cations is dissymmetrizing,
rendering a chiral association and resulting in the
formation (Figure 7a,c) of a pair of enantiomeric C2-
Sch em e 1. Th e Syn th esis of th e P a r en t [24]Cr ow n -8 Da isy-Ch a in Mon om er 1-H‚P F 6

Supramolecular Daisy Chains
J . Org. Chem., Vol. 66, No. 21, 2001 6861
F igu r e 8. The 1H NMR spectra of 1-H‚PF6 dissolved in (a)
CD3SOCD3 (300 MHz, 298 K), and (b) CD3CN (400 MHz, 273
K).
F igu r e 6. (a) Ball-and-stick and (b) space-filling representa-
tions of the C2 symmetric head-to-tail dimer formed by 1-H+.
recorded in CD3SOCD3, is shown in Figure 8a. The
spectrum is relatively simple, and each of the peaks can
be assigned (as annotated in the figure) as arising from
a particular set of protons. In contrast, however, when
1
the H NMR spectrum (400 MHz, 273 K) of 1-H‚PF6 was
recorded in CD3CN, a more complicated spectrum (Figure
8b) was observed. The simpler ‘CD3SOCD3’ spectrum can
be rationalized by ascribing it to the ‘monomeric’ daisy
chain species, i.e., no complexation is occurring. As
expected,25 in the more polar CD3SOCD3, solvation of the
ammonium center occurs preferentially by the solvent
molecules rather than by the macrocyclic polyether
component, thus precluding intermolecular association
of the daisy chain monomers. Conversely, in the less polar
CD3CN solution, the preference for complexation of the
ammonium center by the solvent is less overwhelming,25
and hence ammonium ion complexation ensues. Conse-
quently the formation of numerous oligomeric species,
all in slow exchange27 with one another, could result in
the complicated spectrum observed for 1-H‚PF6 in CD3CN
solution.
The only well-resolved peaks in this spectrum are two
doublets, with coupling constants of ∼1.5 and 8.0 Hz,
respectively, that have been shifted upfield from the
remainder of the signals associated with the aromatic
protons. Upon the basis of chemical shift arguments,28
these signals can be assigned to protons attached to the
central aromatic ring of 1-H‚PF6; a meta-coupled doublet
F igu r e 7. The three cyclic stereoisomeric superstructures that
can be formed upon dimerization of 1-H‚PF6.
symmetric supramolecular stereoisomers. This kind of
interaction occurs because the enantiotopic faces of the
interacting 1-H+ cations possess the same prochirality.
However, an alternative situation can also be envis-
aged: a diastereoisomeric ‘meso’ supramolecular stereo-
isomer, endowed with Ci symmetry, would have been
created (Figure 7b) if the interacting faces of individual
1-H+ cations had maintained different prochiralities upon
crystallization. In other words, the noncovalent dimer-
ization of the 1-H+ cation proceeds diastereoselectively,
at least in the solid state, to furnish a racemic mixture
of the C2-symmetric supramolecular stereoisomers.
NMR Sp ectr oscop y. The 1H NMR spectrum (300
MHz, 298 K) of the hexafluorophosphate salt 1-H‚PF6,
(27) DB24C8-sized macrocycles and phenyl-terminated secondary
ammonium ions form threaded complexes at a rate slower than the
1H NMR time scale. Consequently, all different species present in
solution give rise to their own distinct set of NMR resonances. See:
ref 25.
(28) With regard to aromatic proton spin systems, coupling constants
of 1.5 and 8.0 Hz suggest meta (4J ) and ortho (3J ) coupling, respectively.
Therefore, such coupling patterns, as well as δ values, are unlikely to
arise from the protons of the terminal benzyl ring; the protons of such
an electron neutral (neither especially electron poor nor rich) aromatic
ring would be expected to produce signals with δ values of ∼7.1-7.6
ppm, as well as give rise to a more complicated splitting pattern.
Similar logic leads to the same conclusion for the protons of the
‘catechol’ ring (i.e., the o-C6H4O2 aromatic residue) of 1-H‚PF6.
Although in an electron rich ring, the resonances of protons attached
to 1,2-dioxy-substituted benzene rings generally give rise to second-
order multiplets (or occasionally broad singlets) centered around δ
∼6.8-7.0 ppm. Therefore, the most plausible explanation is one based
upon the protons attached to the central aromatic ring of 1-H‚PF6.
(26) Despite repeated efforts, the PF6- salt of 1+ would not crystal-
lize to give material of suitable quality for single-crystal X-ray analysis.
-
However, on changing to the CF3CO2 anion, good crystals were
obtained easily.

6862 J . Org. Chem., Vol. 66, No. 21, 2001
Cantrill et al.
F igu r e 9. 1H NMR spectra of (a) the amine 1, followed by
sequential addition of (b) CF3CO2D (10 mol equiv), and (c) Et3N
(20 mol equiv).
is expected for H1, and an ortho-coupled doublet is
anticipated for H3. The relative upfield shift of these
distinctive doublets probably results from the offset face-
to-face stacking of two of these central aromatic units,
as would be anticipated, based upon observations of the
solid-state superstructure (vide supra), for a cyclic head-
to-tail dimer. However, the ratio of the intensities of these
two doublets is ∼5:1, indicating that they must originate
from two different cyclic dimeric superstructures. As
noted previously, however, three possible stereoisomeric
head-to-tail dimers (recall Figure 7) can be formedsan
enantiomeric pair and a third diastereoisomeric ‘meso’
form. Despite the presence of only the racemate in the
solid-state superstructure, this inherently kinetic crystal-
lization process does not rule out the possibility that the
‘meso’ diastereoisomer is also present in solution. Hence,
one dimeric diastereoisomer may29 position H3 within a
shielding zone, and the other, H1.
Acid /Ba se Effects. The spectrum of the amine 1 is
shown in Figure 9a. Upon addition of an excess of
CF3CO2D (10 mol equiv) to the NMR tube, the spectrum
shown in Figure 9b was obtained. The complexity of this
spectrum is reminiscent of that exhibited (Figure 8b) by
the hexafluorophosphate salt 1-H‚PF6. Subsequently,
addition of an excess of Et3N (20 mol equiv.) to the same
NMR tube, followed by spectroscopic analysis, resulted
in the spectrum shown in Figure 9c, which is essentially
the same as that illustrated in Figure 9a, except that
there are additional peaks arising from the presence of
[Et3NH]+[CF3CO2]- and an excess of Et3N. These obser-
vations suggest that addition of CF3CO2D to the amine
1 results in deuteration to give the trifluoroacetate salt
1-D‚O2CCF3, which aggregates in a manner similar to
the hexafluorophosphate salt 1-H‚PF6. Subsequent ad-
dition of Et3N deprotonates the ammonium center, thus
reversing, and subsequently preventing, any aggregation.
Tem p er a t u r e E ffect s. The temperature-dependent
1H NMR spectra (400 MHz) of a 98:2 CD3CN/D2O solution
of 1-H‚PF6 show (Figure 10) how increasing the temper-
ature of the sample simplifies significantly the spectro-
scopic behavior. At 358 K, the aromatic region of the
spectrum is very similar to that obtained (Figure 8a)
when analyzing a CD3SOCD3 solution of 1-H‚PF6 at room
temperature. It seems reasonable to conclude, therefore,
F igu r e 10. The temperature-dependent (273-358 K) partial
1H NMR spectra of a CD3CN/D2O (98:2) solution of 1-H‚PF6.
that at 358 K in this CD3CN/D2O mixture, there is
negligible association of daisy chain monomers. However,
as the temperature decreases, other signals in the
spectrum, most notably the two upfield shifted doublets
described above, start to ‘grow’, indicative of the forma-
tion of aggregated species. The lower the temperature
becomes, the larger these peaks grow, a trend consist-
ent with such a temperature-dependent assembly/dis-
assembly. Simply put, in this particular system (with its
own enthalpic and entropic nuances) the Ka value in-
creases as the temperature decreases, resulting in a
greater degree of aggregation at these lower tempera-
tures.
Con cen tr a tion Effects. To investigate its concentra-
tion dependent behavior, a 46.6 mmol CD3CN solution
of 1-H‚PF6 was prepared30 and subjected to 1H NMR
spectroscopic analysis (400 MHz, 298 K). This solution
was diluted repeatedly with known amounts of CD3CN,
1
and H NMR spectra were recorded (partial region shown
in Figure 11) at each particular concentration. As the
solution becomes more and more dilute, eventually (at
0.026 mmol), the spectrum simplifies into one that is
reminiscent of that obtained (Figure 8a) for a 10 mM
solution of 1-H‚PF6 dissolved in CD3SOCD3. In conclu-
sion, therefore, at this very low monomer concentration,
the only species present in solution is the free (non-
aggregated) daisy chain monomer. At higher concentra-
tions, however, the region from δ ) 3.5-4.3 ppm becomes
ever more complicated, presumably as a result of the
formation of aggregated species. Furthermore, at higher
concentrations, signals clustered around δ ) 4.7 ppmsa
chemical shift value that is characteristic of DB24C8-
encircled NH2+-adjacent benzylic methylene protonss
begin to appear, thus suggesting the formation of inter-
(29) Despite studying space-filling molecular models of the different
diastereoisomeric dimeric superstructures, it is difficult to produce a
rationalization for which diastereoisomer should result in a greater
shift for the signal associated with H1, and which for H3.
(30) At concentrations greater than this one, 1-H‚PF6 begins to
precipitate only a few minutes after complete dissolution.

Supramolecular Daisy Chains
J . Org. Chem., Vol. 66, No. 21, 2001 6863
Tw o F lu or in a ted Mon om er s. We have witnessed
previously how atom-substitution was used to mask a
spectral response for this daisy chain system. Now,
however, the use of fluorine atom labeling, to provide an
additional spectroscopic probe via which a greater un-
derstanding of the solution-phase behavior of this system
could be achieved, is described. The reason for placing
fluorine atoms judiciously at strategic locations in the
daisy chain monomer was to permit 19F NMR spectros-
copy to be employed in the solution-phase analysis of
these systems. Whereas the resonances of numerous
different protons in the structure would give rise, upon
1
complexation, to many different signals in the H NMR
spectrum, one, or maybe two, fluorine atoms located in
the structure, would be expected to give a simpler, and
hence interpretable, 19F NMR spectroscopic response.
p-F System s. To test the viability of this fluorine
substitution strategy, a model system was first of all
investigated. Bis(4-fluorobenzyl)ammonium hexafluoro-
phosphate (14-H‚PF6) was prepared from commercially
available starting materials. The binding of this F-
substituted dibenzylammonium salt by DB24C8 (2) was
then investigated, in the usual manner, by recording a
1H NMR spectrum (400 MHz, 300 K) of a 1:1 mixture of
the two components dissolved in CD3CN. Subsequently,
the Ka value for the association of 14-H‚PF6 and DB24C8
at 300 K was calculated, using the single point method,33
to be 875 M-1. Gratifyingly, the 19F NMR spectrum (376
MHz, 300 K) revealed the presence of two aromatic F
peaks, separated by δ ∼ 0.9 ppm, corresponding to (i)
‘free’ 14-H‚PF6, and (ii) bound 14-H‚PF6, which, when
integrated, gave values that resulted in the calculation
of a Ka value exactly the same (875 M-1) as that already
determined using the 1H NMR spectrum. This result
prompted the synthesis of a fluoro-labeled daisy chain
monomer, namely 17-H‚PF6, in which the terminal
phenyl ring carries a fluorine atom in the para position,
in the hope that 19F NMR spectroscopic analysis would
prove much more enlightening than had previous 1H
NMR spectroscopic investigations on the unlabeled ana-
logue.
F igu r e 11. The concentration-dependent partial 1H NMR
spectra obtained from CD3CN solutions of 1-H‚PF6.
molecularly threaded aggregates, i.e., daisy chains. In
addition, the characteristic ortho- and meta-coupled
aromatic doublets (not shown) associated with dimer
formation (vide supra) disappear at low concentration,
but grow in intensity at higher concentrations. Despite
these observations, however, little more can be said other
than that 1-H‚PF6, under the appropriate conditions of
solvent polarity, concentration, and temperature, forms,
reversibly, aggregated species that are presumed to have
the threaded daisy chain architectures outlined in Figure
3.
A Deu ter a ted Mon om er . In an attempt to facilitate
the interpretation of spectra obtained from this class of
daisy chain compounds, 1-H‚PF6 was ‘resynthesized’
using tetradeuterated catechol,31 giving rise to a com-
pound (11-H‚PF6 in Figure 12) in which the most
diagnostic resonances, those arising from H1, H2, and H3,
are not obscured by other, less significant, ones.
Catechol-d4 was synthesized31 from catechol using
D2SO4. Subsequently, the synthesis of the deuterium-
masked daisy chain monomer 11-H‚PF6 was carried out
in the same manner as that which was employed in the
synthesis of the parent (all 1H) monomer.32 Subsequently,
1
the H NMR spectrum (400 MHz, 300 K) of 11-H‚PF6 (20
mM) in CD3CN revealed (Figure 12) the presence of only
two different species in solution. As expected, the two
characteristic upfield shifted doublets (*) are present.
However, in this case, the corresponding signals for the
appropriately coupled protons can be observed, for both
major (‡) and minor (†) species, in the region from δ )
6.7-7.0 ppm. This observation suggests that there are
only two significant species in solution, which, based
upon previous arguments, are likely to be the chiral and
meso [c2]daisy chains (Figure 7).
1
(31) Baban, J . A.; Goodchild, N. J .; Roberts, B. P. J . Chem. Soc.,
Perkin Trans. 2 1986, 157-161.
(32) The extent to which the catechol ring had been deuterateds
93% of all catechol 1H atoms were replaced by 2H atomsswas
calculated based upon the integrals observed in the 1H NMR spectrum
of the intermediate d4-formyl-substituted macrocycle 13.
F igu r e 12. The partial H NMR spectrum (400 MHz, 300 K)
of a CD3CN solution of the deuterium-labeled daisy chain
monomer 11-H‚PF6. Two sets (‡ and †) of signals are observed
for the resonances of the protons (H1, H2, and H3) of the central
aromatic ring.

6864 J . Org. Chem., Vol. 66, No. 21, 2001
Cantrill et al.
The synthesis of 17-H‚PF6 began with the common
CHO-substituted macrocycle precursor 9. In the standard
way, imine formation (with 4-fluorobenzylamine) was
followed by (i) reduction, (ii) protonation, and finally, (iii)
counterion exchange to afford the F-substituted daisy
chain monomer 17-H‚PF6 in very good yield. Except for
variations in the aromatic region, as a consequence of
the AA′BB′ spin system of the terminal p-F substituted
1
phenyl ring, the H NMR spectrum of a 10 mM solution
of 17-H‚PF6 dissolved in CD3CN was reminiscent of that
obtained for the unsubstituted parent daisy chain mono-
1
mer. Despite the complexity of the H NMR spectrum,
however, the 19F NMR spectrum contained (Figure 13,
bottom) only three peaks (A, B, and C) that corresponded
to the resonances of aromatic F atoms. In an effort to
assign these peaks to either complexed or uncomplexed
species, a 10 mM CD3SOCD3 solution of 17-H‚PF6 was
1
added portionwise to the original CD3CN sample, and H
and 19F NMR spectra were recorded after each addition.34
As the percentage CD3SOCD3 content of the CD3CN
solution of 17-H‚PF6 increases (Figure 13), the peaks B
and C begin to reduce in intensity as signal A grows in
intensity. Ultimately, when just over 9% of the solvent
mixture is CD3SOCD3, only one peak (A) remains. The
F igu r e 13. The partial 19F NMR spectra (376 MHz, 300 K)
of a 10 mM solution of 17-H‚PF6 dissolved in varying mixtures
of CD3CN/CD3SOCD3.
1
corresponding H NMR spectrum of this sample reveals
+
increased upon addition of CD3SOCD3, the NH2 sites
that 17-H‚PF6 is present only in the monomeric form
under these conditionssas indicated by the simple nature
of the spectrum. These experiments lead to the conclusion
that peaks B and C arise as a consequence of aggregated
species, i.e., daisy chains. In CD3CN, ammonium ion
binding is favorable and intermolecular association oc-
curs, resulting in the formation of species that give rise
to signals B and C). However, as the solvent polarity is
become solvated by these solvent molecules, resulting in
the deaggregation of daisy chains to form monomer, i.e.,
the species responsible for signal A. Furthermore, the
presence of only two different aggregated species is in
agreement with the behavior observed for the parent
daisy chain systems (vide supra). Coincidentally, the ratio
of peaks B and C is ∼5:1, the same proportion as that
observed for the two upfield shifted aromatic doublets
1
which are so distinctive in the H NMR spectra of both
(33) In a slowly equilibrating complexation/decomplexation scenario,
Ka values are determined as follows: integration of the peaks in the
1H NMR spectrum associated with (1) free host, (2) free guest, and (3)
1:1 complex are measured, and by knowing accurately the concentra-
tions of host and guest species that were dissolved initially, equilibrium
concentrations can be determined. The Ka value is then given simply
by dividing the equilibrium concentration of the 1:1 complex by the
product of the equilibrium concentrations of the free host and guest
species. To minimize errors in this study, numerous probe protons,
associated with each of the three species present in solution, were used
to calculate Ka values in any one spectrum. These Ka values were
simply averaged for each spectrum. Furthermore, at least three
independent 1H NMR spectroscopic analyses were performed for each
host:guest system, and the subsequent Ka values were once again
averaged. For leading references on this method, see: Adrian, J . C.;
Wilcox, C. S. J . Am. Chem. Soc. 1991, 113, 678-680.
(34) This experiment, therefore, allowed the change in solution-
phase behavior to be monitored as the polarity of the solvent system
was gradually increased. It is important to note, however, that in this
process the sample is not diluted: since the CD3SOCD3 solution
contains a 10 mM concentration of 17-H‚PF6, the resulting solutions
always maintain a 10 mM concentration of daisy chain monomer
despite changes in volume. This fact is important, as now changes in
the species present in solution, as monitored spectroscopically, can be
attributed solely to solvent polarity, rather than to concentration
effects.
this analogue and the parent system. It seems reason-
able, therefore, to propose that the two signals B and C,
observed in the 19F NMR spectrum under aggregation-
promoting conditions, originate from the two diastereo-
isomerically different [c2]daisy chains.35
3,5-F 2 System s. When located in the para position of
the terminal phenyl ring, the F atom occupies a position
+
that is as far away from the recognition site (the NH2
center) as possible. By placing F atoms in each of the
(35) This conclusion finds further support insofar as the presence
of significant amounts of acyclic daisy chains, which would give rise
to different patterns in the 19F NMR spectrum to those that are
observed, can be ruled out. For example, in the case of the [a2]daisy
chain, each F-atom lies in a different chemical environment, thus
potentially resulting in two equal intensity peaks appearing in the
spectrum. Upon the basis of the model two-component system (vide
supra) in which ‘free’ and bound F-containing threadlike molecules give
rise to signals ∼0.9 ppm apart, it is unlikely that the resonances of
the two different F atoms in the acyclic dimer would be isochronous.
Obviously, a linear trimer would have F atoms occupying three
different F environments, and so on.

Supramolecular Daisy Chains
J . Org. Chem., Vol. 66, No. 21, 2001 6865
meta positions, it was hoped that the influence of binding
upon the environment of a fluorine substituent would be
increased, resulting in a greater separation in the chemi-
cal shifts of the signals arising from the resonances of
the F atoms attached to the ‘free’ and bound species,
respectively. Therefore, the initial aim was to study a
model system employing this substitution patterns
namely, bis(3,5-difluorobenzyl)ammonium hexafluoro-
phosphate (18-H‚PF6). This salt was prepared starting
from 3,5-difluorobenzaldehyde and 3,5-difluorobenzyl-
amine. Imine formation, using a Dean-Stark apparatus
to trap water generated during the reaction, followed by
borohydride reduction, protonation, and counterion ex-
change, gave the desired F-labeled compound in very
good yield. The binding of this F-substituted dibenzyl-
ammonium salt by DB24C8 (2) was then investigated,
1
in the usual manner, by recording a H NMR spectrum
(400 MHz, 300 K) of a 1:1 mixture of the two components
dissolved in CDCl3/CD3CN (3:1). Initially, the spectrum
exhibited only peaks for each of the two free species, as
if it was simply an overlay of the spectra of the two
independent compounds, indicating that no complexation
had occurred. Over a matter of weeks, however, 18-H+
was observed to slowly thread through the cavity of
DB24C8 to generate the corresponding [2]pseudorotax-
ane.36
F igu r e 14. Partial 19F NMR spectra (376 MHz, 300 K)
recorded over time of a CD3SOCD3 solution of 21-H‚PF6.
analysis was performed in CD3SOCD3 in an effort to
determine whether the compound was pure, i.e., since
aggregation does not occur in this solvent, the purity of
the sample can be assessed rather easily. Therefore, the
1H NMR (400 MHz, 300 K) spectrum of 21-H‚PF6 was
greeted with some surprise. Instead of a simple spectrum,
representing only the daisy chain monomer, a spectrum
reminiscent of that obtained (recall Figure 8b) when a
self-complementary daisy chain monomer is dissolved in
aggregation-promoting CD3CN was observed. Not only
were the peaks in the aliphatic region smeared out across
a wide chemical shift range, but the two distinctive
upfield-shifted aromatic doublets (assigned up to now as
arising from [c2]daisy chains) were also present. Further-
more, the 19F NMR spectrum (vide infra) contained more
than one aromatic 19F peak, which is all that should be
expected if the monomeric species is the only one present
in solution. Not only was this initial spectrum a surprise
but it was also transient! Upon rerunning the spectrum
the next day, both the 1H and 19F NMR spectra had
changed significantly. This observation prompted a more
in-depth analysis of this compoundsone in which a
sample of 21-H‚PF6 was dissolved in CD3SOCD3 and
monitored spectroscopically at more frequent intervals.
The 19F NMR spectra recorded over a period of 6 d
revealed (Figure 14) that, although initially two peaks
(A and B) are present in the spectrum, eventually these
signals disappear and give rise to one singlet, which,
upon inspection of the corresponding simple 1H NMR
spectrum, can be assigned to a monomeric species.
Therefore, it appears that 21-H‚PF6 was isolated as a
mixture of two aggregated daisy chain species, one giving
rise to signal A, and the other, signal B, which, upon
dissolution in CD3SOCD3, dissociate extremely slowly to
give the daisy chain monomer.
The synthesis of the 3,5-difluorophenyl-substituted
[24]crown-8 based daisy chain monomer 21-H‚PF6 began
with the common formyl-substituted macrocycle precur-
sor 9. In the standard fashion, imine formation was
followed by (i) reduction, (ii) protonation, and finally, (iii)
counterion exchange. Once again, the initial spectroscopic
(36) 1H NMR spectra were recorded over subsequent weeks, reveal-
ing the formation of a complex, albeit very slowly. The diagnostic
multiplet at δ ) 4.6 increased in intensity with time while the ratio of
‘free’:bound thread can be most easily followed by monitoring the
signals arising from the resonances of the γ protons of ‘free’ and bound
DB24C8. Unfortunately, however, careful inspection of the correspond-
ing 19F NMR spectra revealed that the sample was undergoing
decomposition during the course of the analysis. Although more and
more of the cation was slowly finding its way into DB24C8 macrocycles,
-
the PF6 anion was, after ∼100 h, no longer present. Interestingly, a
-
new doublet, shifted upfield from the PF6 doublet by ∼12 ppm,
appears and then fades away with time as well. One possible explana-
tion lies in the decomposition of PF6- to give PF5, which, with a boiling
point of -75 °C, although initially dissolved in solution, is drawn into
the partial vacuum above the liquid that is formed upon sealing the
NMR tube. Therefore, even although the system appears to be
undergoing equilibration, the uncertainty surrounding the identity of
the corresponding anion ended the experiment abruptly and prema-
turely. Nonetheless, the synthesis of the corresponding 3,5-F2-
substituted daisy chain monomer was undertaken, safe in the knowl-
edge that, although slow in their passage, 3,5-F2-substituted phenyl
rings will pass through the macrocyclic cavity of DB24C8. Further-
more, the slow kinetics of threading/unthreading did not necessarily
constitute a problem since they could potentially provide an op-
portunity to follow the assembly of daisy chain species on the human,
rather than the NMR, time scale.
1
Although complicated, the initial H NMR spectrum
contains the two diagnostic upfield shifted aromatic
doublets that are indicative (vide supra) of the formation
of the two possible diastereoisomeric [c2]daisy chains.
Furthermore, the 19F NMR spectrum contains, initially,
two signals (A and B), with an intensity ratio of ∼5:1sa
feature which is reminiscent of that observed for the p-F

6866 J . Org. Chem., Vol. 66, No. 21, 2001
Cantrill et al.
species remains, i.e., there are no longer any aggregated
assemblies present.
There is, however, one feature of the spectra illustrated
in Figure 14, that remains to be explained. A small signal
(*) is evident at δ ) -108.7 ppm. Initially, it grows ever
so slightly in intensity, and then ‘shrinks back’ into the
baseline of the spectrum until it is no longer visible at
the end of the experiment. Perhaps, finally, we are
‘seeing’ a little of the [a2]daisy chain. When the [c2]daisy
chains dissociate to give the monomeric structure, unless
both 3,5-difluorophenyl terminated arms dethread si-
multaneously, some of the [a2]daisy chain will inevitably
be formed. Such a structure contains both ‘free’ and
bound ammonium ion-containing arms, each of which is
terminated by a 3,5-difluorophenyl group. The [a2]daisy
chain would be expected to give rise to two equal intensity
peaks in the 19F NMR spectrum. It is likely that the peak
at δ ) -108.7 ppm corresponds to the resonance of the
F atoms of the ‘free’ ammonium ion-containing arm of
the [a2]daisy chain, while the corresponding signal for
the bound arm remains hidden under either peak A or
B. As the intensity of this peak never ‘grows’ to any
significant proportions, such an explanation would re-
quire the rate constant for the transformation of [c2]daisy
chain into [a2]daisy chain to be smaller than that for the
conversion of [a2]daisy chain into [a1]daisy chain, i.e.,
the monomer. It is not unreasonable to propose that the
more tangled nature of the cyclic superstructure, with
less degrees of translational and rotational freedom, as
compared with the linear [a2]daisy chain superstructure,
may offer an explanation for this observation. At this
point, therefore, there is no doubt regarding the absence
of higher order daisy chain superstructures in the initial
solution. For example, the presence of a [c3]daisy chain
would, in the process of its CD3SOCD3-induced dismem-
berment, generate [a2]- and [a3]daisy chains (the latter
complex having three distinct F atom environments),
almost certainly giving rise to a more complicated set of
spectra than those observed experimentally. Obviously
the situation would be considerably more complicated for
even larger daisy chain complexes.
[25]Cr ow n -8-Ba sed System s. It is clear that gaining
an appreciation of the aggregation behavior of [24]crown-
8 based daisy chain systems is hampered by the forma-
tion of stereoisomeric superstructures. In an effort to
remove this stereochemical ambiguity, monomers based
upon a [25]crown-8 macrocycle were investigated as
candidates for the formation of aggregated daisy chains.
Although crown ethers with a [25]crown-8 constitution
have been shown37 to have much lower affinities for
secondary ammonium ions than their DB24C8-based
counterparts, the more highly symmetric nature of the
prototypical [25]crown-8 daisy chain monomer (22-H‚
PF6), as compared to the [24]crown-8 monomer (1-H‚PF6),
means that, upon assembly, only one unique [c2]daisy
chain is formed. Furthermore, only one form of each of
the higher order aggregates, i.e., [c3]daisy chain, [c4]daisy
chain, etc., is possible.
F igu r e 15. The FAB-mass spectra of 21-H‚PF6 (i) prior to
dissolution in CD3SOCD3, and (ii) after sitting in CD3SOCD3
solution for 6 d.
daisy chain system (17-H‚PF6). This observation suggests
that there are two different F atom environments arising
from two different F-containing aggregated species. Upon
the basis of all of the evidence to hand, it is reasonable
to conclude that these two aggregated species are indeed
the two diastereoisomeric [c2]daisy chains. This conclu-
sion is further supported by FAB mass spectrometric
investigations that were subsequently performed on this
compound. The FAB mass spectrum of the isolated solid,
i.e., prior to dissolution in CD3SOCD3, reveals (Figure
15i) the presence of dimeric species. The signals at m/z
values of 1207.9 and 1353.9, respectively, correspond to
unipositive dimeric structuressthe larger value arises
-
as a consequence of an associated PF6 anion. Further-
more, the cluster of isotopic peaks centered around m/z
) 604.4, are spaced by 0.5 mass units, indicating the
presence of the doubly charged dimeric superstructure.
This result suggests that no higher order daisy chain
oligomers (i.e., trimers, tetramers, etc.) are to be found
in the isolated sample, and that it consists solely of
dimeric superstructures, of which there are only three:
(i) the two diastereoisomeric [c2]daisy chains and (ii) the
linear [a2]daisy chain. The second of these possibilities
can be excluded based upon the simplicity (only two
peaks, A and B in a ratio of ca. 5:1) of the initial 19F NMR
spectrum. The reason is that, such an unsymmetric
species would give rise to two equal intensity singlets,
one each for a bound 3,5-difluorophenyl ring, and an
unbound onesthe same argument that was applied in
explaining the 19F NMR spectra observed in the case
of the p-F daisy chain system 17-H‚PF6. Subsequently,
the proposed composition of the six-day old CD3SOCD3
solution of 21-H‚PF6 was confirmed (Figure 15ii) follow-
ing FAB mass spectrometric analysis of this solution.
Only a single peak (at m/z ) 604.3) is observed in the
spectrum and, since the adjacent isotope peaks are
separated by unity, only the monomeric daisy chain
Th e P a r en t Mon om er . The synthesis of this [25]-
crown-8-based self-complementary daisy chain monomer
is outlined in Scheme 2. Tosylation of the known38 diol
23, under standard conditions, gave the ditosylate 24,
(37) Ashton, P. R.; Bartsch, R. A.; Cantrill, S. J .; Hanes, R. E., J r.;
Hickingbottom, S. K.; Lowe, J . N.; Preece, J . A.; Stoddart, J . F.;
Talanov, V. S.; Wang, Z.-H. Tetrahedron Lett. 1999, 40, 3661-3664.

Supramolecular Daisy Chains
J . Org. Chem., Vol. 66, No. 21, 2001 6867
Sch em e 2. Th e Syn th esis of th e P a r en t [25]Cr ow n -8-Ba sed Da isy-Ch a in Mon om er 22-H‚P F 6
which was subsequently reacted with 3,5-dihydroxy-
benzyl alcohol (25) to give the hydroxymethyl-substituted
BMP25C8 derivative 26 in reasonable yield. Oxidation
of the hydroxyl group of 26 with PCC gave the corre-
sponding formyl-substituted macrocycle 27 in good yield.
Subsequent condensation of 27 with benzylamine (10),
followed by borohydride reduction, protonation (HCl), and
counterion exchange from Cl- to PF6-, afforded the target
compound 22-H‚PF6 in a 15% yield.
more amenable to analysis. Not only would a slowly
exchanging system be preferable spectroscopically, but,
as observed in the [24]crown-8 series, it was felt that a
19F probe might prove helpful. One candidate immedi-
ately sprang to mind - namely, the 3,5-difluorophenyl
group, which had been shown to pass very slowly through
a DB24C8 sized macrocycle. Might it also prove useful
in the [25]crown-8-based system? Before proceeding with
the synthesis of a 3,5-difluorophenyl terminated [25]crown-
8 daisy chain monomer, the appropriate model system
was investigated. A CD3CN solution containing a 1:1
mixture of bis(3,5-difluorobenzyl)ammonium hexafluoro-
phosphate (18-H‚PF6) and BMP25C8 (28) was subjected
to both H and 19F NMR spectroscopic analyses. The H
NMR spectrum was characteristic of a slowly exchanging
crown ether/dibenzylammonium ion system. Signals were
observed for all three species present in solution, namely
(i) ‘free’ BMP25C8, (ii) ‘free’ 18-H‚PF6, and (iii) the
[2]pseudorotaxane [28‚18-H]PF6. Consequently, this spec-
trum was used to calculate, using the single point
method33, a Ka value (at 300 K in CD3CN) of 25 M-1 for
1
1
the association of these two species. Furthermore, the 19
F
NMR spectrum revealed the presence of two aromatic F
singlets, one each for the ‘free’ and bound bis(3,5-
difluorobenzyl)ammonium cation. Crucially, the Ka value,
when calculated based upon the intensities of these two
peaks, was the same as that already obtained from the
data extracted from the 1H NMR spectrum. The slow
passage, on the NMR time scale, of a 3,5-difluorophenyl
ring through the cavity of BMP25C8 prompted the
synthesis of the correspondingly substituted daisy chain
analogue, namely 29-H‚PF6.
Unlike the [24]crown-8 daisy chain analogues, 22-
H‚PF6 was found to be soluble in CD2Cl2, perhaps the
optimal solvent25 for the crown ether/ammonium ion
interaction, and, therefore, subsequent 1H NMR inves-
tigations were conducted in this solvent. Unfortunately,
as expected,18j,37 the kinetics associated with the thread-
ing/dethreading of a phenyl ring through a [25]crown-8
sized macrocycle are such that neither a slow- nor a fast-
1
exchange regime operates. Consequently, H NMR spec-
tra obtained from solutions of 22-H‚PF6 over a range
(0.05-28.5 mM) of concentrations consist of broad peaks
that are difficult to assign to any particular species. The
spectroscopic changes observed over a given concentra-
tion range indicate that aggregation, to some extent, does
occur with this [25]crown-8-based monomer. Further-
more, the FAB mass spectrum of 22-H‚PF6 contains a
peak, albeit very small (ca. 3% of the peak arising from
the monomer), corresponding to a dimeric assembly.
A F lu or in a ted Mon om er . The limited information
garnered from the study of the parent [25]crown-8 daisy
chain prompted the search for a system that would be
The synthesis of 29-H‚PF6 started from the formyl-
substituted macrocycle precursor 27. In standard fashion,
imine formation (with 3,5-difluorobenzylamine) was fol-
lowed by (i) reduction, (ii) protonation, and finally, (iii)
counterion exchange, to afford the 3,5-difluorophenyl-
substituted daisy chain monomer 29-H‚PF6 in a 17%
1
yield. The initial H and 19F NMR spectroscopic charac-
terization, which was carried out in CD3SOCD3, revealed
the daisy chain monomer to be pure: a simple, easily
assignable, 1H NMR spectrum was obtained, and the 19
F
NMR spectrum contained a single peak corresponding
to the aromatic F atoms, in addition to the doublet for
-
the PF6 anion. Mass spectrometry (FAB) revealed the
(38) Mertens, I. J . A.; Wegh, R.; J enneskens, L. W.; Vlietstra, E. J .;
van der Kerk-van Hoof, A.; Zwikker, J . W.; Cleij, T. J .; Smeets, W. J .
J .; Veldman, N.; Spek, A. L. J . Chem. Soc., Perkin Trans. 2 1998, 725-
735.
formation, in the ‘gas phase’, of dimeric superstructures,
as evidenced by the presence of signals corresponding to
both the uni- and dipositive dimers. More interesting,

6868 J . Org. Chem., Vol. 66, No. 21, 2001
Cantrill et al.
sample (71.4 mM) investigated in this study, is, ap-
propriately, the most complicated. At this concentration,
a new peak is observed to appear just downfield of that
assigned to the [c2]daisy chain, and can, therefore, be
attributed to the [c3]daisy chain. Correspondingly, peaks
that can be attributed to the acyclic trimer ([a3]daisy
chain) are also annotated on the spectrum. Two small
equal-intensity peaks (δ ) -110.12 and -110.15, respec-
tively) are observed in the same region of the spectrum
as the peak (δ ) -110.22) corresponding to the resonance
of the F atoms on the complexed arm of the [a2]daisy
chain. The [a3]daisy chain is expected (Figure 17) to give
rise to three equal intensity signals, two corresponding
to the F atom resonances of bound arms and one arising
from the resonance of the F atoms in the ‘free’ arm. It is
not unreasonable to propose that the peak associated
with this latter resonance overlaps with the peak arising
from the corresponding resonance of the ‘free’ arm of the
[a2]daisy chain. This hypothesis gains further support
when it is considered that the signal labeled ‘a2 + a3
(uc)’ has an intensity equal to the sum of the intensities
of the peak corresponding to a2 (c) and one of the a3 (c)
peaks. Although these assignments cannot be made with
absolute certainty, they are consistent with (i) the
anticipated solution phase behavior, upon increasing
concentration, of an aggregating self-complementary
system, (ii) the number and intensity ratios of peaks that
would be expected to arise from the proposed species, as
well as (iii) the relative chemical shift values that would
arise for peaks corresponding to the resonances of either
‘free’ or bound F-containing arms. Assuming the correct
interpretation of the spectrum, Ka values for the ap-
propriate assembly steps (recall Figure 3) were calcu-
lated, employing the single point method, based upon the
data contained with the 19F NMR spectrum obtained from
the 71.4 mM sample. Although the values obtained (Ka2
) 8.2 M-1; Kc2 ) 1.2; Ka3 ) 6.5 M-1; Kc3 ) 0.7) are lower
F igu r e 16. The concentration-dependent partial 19F NMR
spectra (376 MHz, 298 K, CD3CN) of 29-H‚PF6.
however, were the concentration-dependent 19F NMR
spectroscopic investigations conducted39 in CD3CN solu-
tion.
The 19F NMR spectra obtained from three samples
containing different concentrations of 29-H‚PF6 (2.0, 12.7,
and 71.4 mM, respectively) revealed (Figure 16) a dra-
matic concentration dependence. The least concentrated
sample (2.0 mM) gave rise to a 19F NMR spectrum
containing a single aromatic F signal. Inspection of the
corresponding 1H NMR spectrum, which was reminiscent
of that obtained when 29-H‚PF6 was dissolved in
CD3SOCD3, confirmed that the single F resonance could
be attributed to the monomeric species, i.e., the [a1]daisy
chain. The 19F NMR spectrum arising from the sample
of intermediate concentration (12.7 mM) contained three
additional aromatic F signals. From these extra peaks,
it can be inferred that, at this higher concentration,
intermolecular aggregation is starting to occur and
results in the formation of new species. The presence of
two small equal-intensity peaks (δ ) -110.46 and
-110.22, respectively) suggest the presence of an [a2]daisy
chain in which one of the 3,5-difluorophenyl-terminated
ammonium ion-containing arms is complexed, and the
other one is not. One peak has a chemical shift similar
to that observed for the monomeric species, whereas the
other one is shifted downfield, consistent with complex
formation, as was observed in the case of the model
BMP25C8/18-H‚PF6 system. Even further downfield
shifted is another singlet (δ ) -109.90), which, most
likely, corresponds to the [c2]daisy chain in which all of
the F atoms reside in equivalent environments. The 19F
NMR spectrum, obtained from the most concentrated
F igu r e 17. A schematic representation depicting the ‘free’ and
bound ammonium ion-containing arms of aggregated (up to,
and including, trimeric) daisy chain superstructures.
(39) The solvent of choice was CD3CN, as no meaningful 1H or 19F
NMR spectroscopic results were obtained for CD3NO2 or CD2Cl2
solutions of 29-H‚PF6.

Supramolecular Daisy Chains
J . Org. Chem., Vol. 66, No. 21, 2001 6869
(Figure 20), the enthalpic and entropic contributions to
the formation of these species can be determined. Straight
lines, each with a good fit, can be approximated to the
experimental data, revealing that (i) the formation of the
[c3]daisy chain is more enthalpically favorable than the
assembly of the [c2]daisy chain, whereas (ii) entropy
factors favor the formation of the cyclic dimeric, rather
than the trimeric, superstructure. These results are not
unexpected: assembly of the [c3]daisy chain proceeds
with the creation of more noncovalent interactions than
does the formation of the [c2]daisy chain, but it requires
the assembly of three separate components rather than
two. One important point to note is that the enthalpic
gain upon cyclic trimerization is almost twice as much
as that calculated for cyclic dimerization. Upon the basis
of the fact that the formation of a [c3]daisy chain results
from three crown ether/ammonium ion interactions, and
two such interactions are responsible for [c2]daisy chain
formation, why do the enthalpic terms not reflect this
3:2 ratio? Furthermore, in the case of the [24]crown-8
daisy chains, the stacking of the central aromatic rings
appears to account for a significant stabilization of the
[c2] superstructure, resulting in their domination of both
the solution- and solid-state properties. If extrapolated
to the [25]crown-8-based systems, this additional π-π
derived stabilization would surely result in an even
smaller ratio of the enthalpy changes observed for cyclic
trimerization versus dimerization. Perhaps the fact that
such an effect is not observed, means that, in the
[25]crown-8 series the [c2] superstructure does not
represent such an island of stability as it does in the case
of the [24]crown-8 analogues. In retrospect, this observa-
tion is also reflected in the values of Ka2 and Kc2 obtained
for 29-H‚PF6. Whereas Ka2 is over six times larger than
Kc2 in the [25]crown-8 series, the trend, if not the
magnitude, is believed to be reversed in the [24]crown-8
series.21
F igu r e 18. Expressions for the overall cyclic dimerization and
trimerization constants in a self-complementary daisy-chain
system.
F igu r e 19. The temperature-dependent partial 19F NMR
spectra (376 MHz, CD3CN) of 29-H‚PF6.
than those observed for the model system (Ka ) 25 M-1),
this outcome is not unexpected.40
For such a self-assembling system, it is possible to
define (Figure 18) overall cyclic dimerization and cyclic
trimerization constants, namely KDim and KTri, respec-
tively. By utilizing variable temperature 19F NMR spec-
troscopy (Figure 19), values for these association con-
stants were calculated at different temperatures. An
increase in the temperature of the solution resulted in a
decrease in the amount of dimeric and trimeric species
observed. Such a trend is reflected in the ever-decreasing
values of KDim and KTri as the temperature rises. By
constructing a van’t Hoff plot for each of these species
F igu r e 20. The van’t Hoff plot obtained upon plotting ln Ka
vs 1/T for the Ka values determined from the partial 19F NMR
spectra illustrated in Figure 19.
(40) The lower values of Ka than those observed in the model system
may reflect the fact that oligomerization of this self-complementary
system always requires bringing together two positively charged
species. This repulsive Coulombic effect may serve to reduce the overall
affinity for two daisy chain monomers in comparison with the model
two-component system, in which a cationic thread is bound by a neutral
macrocycle.
Con clu sion s
It has been shown that self-complementary dibenzyl-
ammonium ion substituted crown ethers, with either a
[24]- or [25]crown-8 constitution, will aggregate to give
daisy chain assemblies under the appropriate conditions

6870 J . Org. Chem., Vol. 66, No. 21, 2001
Cantrill et al.
a krypton primary atom beam, utilizing a m-nitrobenzyl
alcohol matrix. Cesium iodide or poly(ethylene glycol) was
employed as a reference compound. Electron impact mass
of concentration, temperature, and solvent polarity.
These supramolecular aggregates, however, fall far short
of the suprapolymeric systems described in the Introduc-
tion of this paper. Although intramolecular self-complex-
ation is not an option for these daisy chain molecules, a
growing chain quickly finds its own tail and cyclizes,
preventing any further chain propagation. In the case of
the [24]crown-8 systems, there is overwhelming evidence
to suggest that, over a wide concentration range, the
[c2]daisy chain is by far the most favorable assembly.
This conclusion was not reached easily or lightly. It
required the application of deuterium- as well as fluorine-
labeling, in conjunction with extensive concentration- and
temperature-dependent spectroscopic studies. Further-
more, the interpretation of the spectroscopic data col-
lected for these compounds is complicated by the forma-
tion of stereoisomeric aggregates. In an attempt to
alleviate this subtle stereochemical nuance, more sym-
metrical daisy chain monomers based upon [25]crown-8
macrocycles were prepared and studied. Not only is the
spectroscopic behavior of these modified daisy chain
monomers considerably simpler, but the [c2]daisy chain
superstructure no longer represents a dead-end for chain
growth. The weaker interaction between secondary di-
alkylammonium ions and [25]crown-8-based macrocycles,
however, undermines the effectiveness of this supra-
molecular system. Although trimeric aggregates are
observed in solution, these species are not predicted to
rise to significant levels at concentrations < 1 M.
Consequently, the concentration at which a species
considered to be supramolecular polymer would be
formed is likely to far exceed the solubility limits of these
salts. In summary, although not precursors for inter-
woven supramolecular polymers, the compounds reported
in this paper represent interesting examples of inter-
woven supermolecules that could be utilized, with ap-
propriate modification, in the formation of mechanically
interlocked polymers.
spectra (EIMS) were obtained from
a VG Prospec mass
spectrometer. Microanalyses were performed by either the
University of North London Microanalytical Service (UK) or
Quantitative Technologies, Inc..
3,4-Bis{2-[2-(2-h yd r oxyet h oxy)et h oxy]et h oxy}b en z-
a ld eh yd e (6). 3,4-Dihydroxybenzaldehyde (4) (15.1 g, 0.11
mol), K2CO3 (37.3 g, 0.27 mol), and LiBr (9.5 g, 0.11 mol) were
placed in a 2 L round-bottomed flask fitted with a condenser
and a pressure-equalized dropping funnel. The system was
flushed with N2, and anhydrous DMF (500 mL) was introduced
into the flask. 2-[2-(2-Chloroethoxy)ethoxy]ethanol (5) (40.6 g,
0.24 mol) was dissolved in DMF (100 mL) and added to the
dropping funnel, again through
a flow of nitrogen. The
suspension in the flask was heated to 100 °C while stirring,
and the solution of 5 was added dropwise over 1 h. The reaction
mixture was stirred at 100 °C for a further 3 d under a
continuous flow of N2. After this time, the reaction mixture
was filtered (in order to remove inorganic salts) and evaporated
to dryness, and the residue was partitioned between CH2Cl2
(300 mL) and 10% w/v K2CO3 solution (300 mL). The organic
layer was further washed with 10% w/v K2CO3 solution (3 ×
100 mL). The organic phase was dried (MgSO4), and the
solvents were removed in vacuo to yield 6 as a brown oil (39.1
g, 89%). This oil was used in subsequent reactions without
further purification. An analytically pure sample was obtained
as a pale yellow oil after chromatography (SiO2: gradient
elution with CH2Cl2/MeOH, 100:0 to 96:4); 1H NMR (300 MHz,
CDCl3): δ ) 3.32 (br s, 2H), 3.52-3.74 (m, 16H), 3.83-3.90
(m, 4H), 4.14-4.23 (m, 4H), 6.94 (d, J ) 8.1 Hz, 1H), 7.35-
7.42 (m, 2H), 9.77 (s, 1H); 13C NMR (75 MHz, CDCl3): δ )
61.5, 68.5, 69.2, 69.3, 70.2, 70.7, 70.8, 72.6, 111.5, 112.2, 126.7,
130.1, 148.9, 154.0, 190.8; MS (EI): m/z (%): 402 (37) [M]+;
C
19H30O9 (402.4): calcd C 56.71, H 7.51; found C 56.88, H 7.41.
3,4-Bis(2-{2-[2-(p-tolu en esu lfon yloxy)eth oxy]eth oxy}-
eth oxy)ben za ld eh yd e (7). Diol 6 (26.6 g, 66.0 mmol), Et3N
(66.8 g, 660 mmol) and a catalytic amount of 4-(dimethyl-
amino)pyridine were dissolved in CH2Cl2 (400 mL), and this
solution was stirred and cooled (0-5 °C). A solution of
p-toluenesulfonyl chloride (62.9 g, 330 mmol) in CH2Cl2 (200
mL) was then added dropwise over a period of 3 h, maintaining
the reaction temperature below 5 °C. Subsequently, the
reaction mixture was allowed to warm to ambient temperature
and left to stir for a further 8 h under a continuous flow of N2.
The reaction mixture was acidified with 5 M HCl solution (250
mL), and the organic layer was washed with 2 M HCl solution
(200 mL) and saturated brine (2 × 200 mL). The organic layer
was dried (MgSO4), and the solvents were removed in vacuo.
The residue was purified by column chromatography (SiO2:
gradient elution with EtOAc/C6H14, 50:50 to 80:20) to yield the
desired compound 7 as a pale yellow oil (26.8 g, 57%); 1H NMR
(300 MHz, CDCl3): δ ) 2.38 (s, 6H), 3.53-3.67 (m, 12H), 3.78-
3.86 (m, 4H), 4.08-4.21 (m, 8H), 6.96 (d, J ) 8.1 Hz, 1H), 7.29
(d, J ) 8.5 Hz, 4H), 7.36-7.43 (m, 2H), 7.74 (d, J ) 8.5 Hz,
4H), 9.78 (s, 1H); 13C NMR (75 MHz, CDCl3): δ ) 21.8, 68.9,
69.4, 69.6, 69.7, 70.9, 71.5, 111.5, 112.0, 112.7, 126.9, 128.1,
130.0, 130.4, 133.1, 145.0, 149.3, 154.5, 191.0; MS (FAB): m/z
(%): 711 (21) [M]+; C19H30O9 (710.8): calcd C 55.76, H 5.96;
found C 55.74, H 5.75.
(2-F or m yl)d iben zo[24]cr ow n -8 (9). Cesium carbonate
(19.3 g, 59.3 mmol) was placed in a 1 L round-bottomed flask
fitted with condenser and pressure-equalized dropping funnel.
The system was flushed with N2, and anhydrous DMF (200
mL) was added to the flask. The ditosylate 7 (8.43 g, 11.9
mmol) and catechol (8) (1.31 g, 11.9 mmol) were dissolved in
DMF (500 mL) and added to the dropping funnel, again
through a flow of N2. The suspension in the flask was heated
to 100 °C while stirring, and the ditosylate/catechol solution
was added dropwise over 24 h. This mixture was stirred at
100 °C, under an N2 atmosphere, for a further 7 d. Subse-
quently, the solvent was removed in vacuo and the residue
partitioned between PhMe (300 mL) and 10% w/v K2CO3
solution (300 mL). The aqueous layer was further extracted
Exp er im en ta l Section
Gen er a l. Chemicals were purchased from Aldrich and
Lancaster Synthesis Ltd and used as received unless indicated
otherwise. Compounds 1231 and 2338 were prepared according
to literature procedures. The synthesis and characterization
data for BMP25C8 (28) have been reported elsewhere.18j
Solvents were dried according to literature procedures.41 Thin-
layer chromatography was carried out using aluminum sheets
precoated with silica gel 60F (Merck 5554). The plates were
inspected by UV light and, if required, developed in I2 vapor.
Column chromatography was carried out using silica gel 60F
(Merck 9385, 0.040-0.063 mm). Melting points were deter-
mined on an Electrothermal 9100 apparatus and are uncor-
rected. 1H and 13C NMR spectra were recorded on either a
Bruker AC300 (300 and 75 MHz, respectively), Bruker ARX400
(400 and 100 MHz, respectively), or Bruker ARX500 (500 and
125 MHz, respectively) spectrometer, using residual solvent
as the internal standard. 1H-Decoupled 19F NMR spectra were
recorded on a Bruker ARX400 (376 MHz) spectrometer and
were referenced to C6F6 (-163.0 ppm) present as a CH2Cl2
solution in an internal capillary tube. All chemical shifts are
quoted on the δ scale, and all coupling constants are expressed
in hertz (Hz). Liquid secondary ion (LSI) mass spectra were
obtained on a VG Zabspec mass spectrometer, equipped with
a cesium ion source and utilizing a m-nitrobenzyl alcohol
matrix. Fast atom bombardment (FAB) mass spectra were
obtained using a ZAB-SE mass spectrometer, equipped with
(41) Perrin, D. D.; Armarego, W. F. L. Purification of Laboratory
Chemicals; Pergamon: Oxford, 1989.

Supramolecular Daisy Chains
J . Org. Chem., Vol. 66, No. 21, 2001 6871
with PhMe (3 × 200 mL), and the combined organic layers
were washed with 10% w/v K2CO3 solution (300 mL). The
organic phase was dried (MgSO4), and the solvents were
removed in vacuo. The residue was subjected to column
chromatography (SiO2: gradient elution with EtOAc/MeOH,
100:0 to 95:5) to yield an off-white solid which was recrystal-
lized from EtOAc/C6H14 to yield the desired compound 9 as a
white solid (2.31 g, 41%); mp 105-107 °C; 1H NMR (300 MHz,
CDCl3): δ ) 3.82-3.85 (m, 8H), 3.89-3.97 (m, 8H), 4.12-4.24
(m, 8H), 6.83-6.90 (m, 4H), 6.93 (d, J ) 8.1 Hz, 1H), 7.37 (d,
J ) 1.8 Hz, 1H), 7.42 (dd, J ) 1.8, 8.1 Hz, 2H), 9.81 (s, 1H);
13C NMR (75 MHz, CDCl3): δ ) 69.4, 69.5, 69.7, 70.0, 71.3,
71.4, 71.5, 111.1, 111.9, 114.0, 121.4, 126.8, 130.2, 148.9, 149.2,
154.3, 190.9; MS (FAB): m/z (%): 477 (100) [M + H]+, 494
(13) [M + NH4]+, 499 (24) [M + Na]+, 515 (5) [M + K]+;
Z ) 8, Fcalcd ) 1.39 g cm-3, µ (MoKR) ) 1.22 cm-1, F(000) )
3328. Crystal dimensions 0.33 × 0.33 × 0.80 mm (needles),
Siemens P4 diffractometer, graphite-monochromated MoKR
radiation, ω-scans, T ) 203 K. Of 6449 independent reflections
measured (2θ e 50°), 2715 had Io > 2σ(Io) and were considered
to be observed. The structure was solved by direct methods
and the non-hydrogen atoms refined anisotropically. The
positions of the NH2+ hydrogen atoms were located from a ∆F
map and refined isotropically subject to an N-H distance
constraint. The remaining hydrogen atoms were placed in
calculated positions, assigned isotropic thermal parameters
U(H) ) 1.2 Ueq(C), and allowed to ride on their parent atoms.
Refinement was by full-matrix least-squares, based on F2 to
give R1 ) 0.092, wR2 ) 0.222. Computations were carried out
using the SHELXTL 5.03 package.42 CCDC 100961.43
1,2-Bis(2-{2-[2-(2-p -t olylsu lfon yloxy)et h oxy]et h oxy}-
eth oxy)ben zen e (24). Diol 2338 (55.0 g, 147 mmol), Et3N (74.3
g, 734 mmol) and a catalytic amount of 4-dimethylamino-
pyridine were dissolved in CH2Cl2 (400 mL), and this solution
was stirred and cooled (0-5 °C). A solution of p-toluenesulfonyl
chloride (61.6 g, 323 mmol) in CH2Cl2 (100 mL) was then added
dropwise over a period of 2 h, maintaining the reaction
temperature below 5 °C. Subsequently, the reaction mixture
was allowed to warm to ambient temperature and left to stir
for a further 3 h under a continuous flow of N2. The reaction
mixture was acidified with 5 M HCl solution (250 mL), and
the organic layer was washed with 2 M HCl solution (2 × 200
mL). The organic layer was dried (MgSO4), and the solvents
were removed in vacuo. The residue was purified by filtering
through a pad of SiO2 (gradient elution with EtOAc/C6H14, 10:
90 to 60:40) to yield the desired compound 24 as a pale yellow
oil (82.8 g, 83%); 1H NMR (200 MHz, CDCl3): δ ) 2.39 (s, 6H),
3.53-3.81 (m, 16H), 4.08-4.14 (m, 8H), 6.88 (s, 4H), 7.29 (d,
J ) 8.2 Hz, 4H), 7.75 (d, J ) 8.2 Hz, 4H); 13C NMR (50 MHz,
CDCl3): δ ) 21.6, 68.7, 68.9, 69.4, 69.8, 70.7, 115.0, 121.7,
127.9, 129.9, 133.0, 144.8, 149.0; MS (FAB): m/z (%): 682 (26)
[M]+; C32H42O12S2 (682.8): calcd C 56.29, H 6.20; found C 56.39,
H 6.17.
Ben zo(5-h yd r oxym et h ylm et a p h en ylen e)[25]cr ow n -8
(26). Cesium carbonate (47.7 g, 146 mmol) was placed in a 2
L round-bottomed flask fitted with condenser and pressure-
equalized dropping funnel. The system was flushed with N2,
and anhydrous MeCN (700 mL) was added to the flask. The
ditosylate 24 (20.0 g, 29.3 mmol) and 3,5-dihydroxybenzyl
alcohol (25) (4.10 g, 29.3 mmol) were dissolved in dry MeCN
(600 mL) and added to the dropping funnel, again through a
flow of N2. The suspension in the flask was heated under reflux
while stirring, and the distosylate/25 solution was added
dropwise over 24 h. This mixture was stirred at reflux, under
an N2 atmosphere, for a further 3 d. Upon cooling, the reaction
mixture was filtered, the solvent removed in vacuo, and the
residue partitioned between CH2Cl2 (300 mL) and 10% w/v
K2CO3 solution (300 mL). The aqueous layer was further
extracted with CH2Cl2 (3 × 300 mL), and the combined organic
layers were washed with 10% w/v K2CO3 solution (300 mL).
The organic phase was dried (MgSO4), and the solvents were
removed in vacuo. The residue was subjected to column
chromatography (SiO2: EtOAc/C6H14, 90:10) to yield a white
solid (4.50 g, 32%); mp 68-71 °C; 1H NMR (400 MHz, CDCl3):
δ ) 3.66-3.71 (m, 8H), 3.77-3.82 (m, 8H), 4.08-4.12 (m, 8H),
4.53 (s, 2H), 6.48 (d, J ) 2.3 Hz, 2H), 6.57 (d, J ) 2.3 Hz, 1H),
6.84-6.89 (m, 4H); 13C NMR (75 MHz, CDCl3): δ ) 65.1, 68.1,
68.9, 69.8, 69.9, 71.0, 101.7, 106.2, 115.1, 121.7, 143.5, 149.0,
160.1; MS (FAB): m/z (%): 461 (75) [M - OH]+, 478 (100) [M]+,
501 (6) [M + Na]+; C25H34O9 (478.5): calcd C 62.75, H 7.16;
found C 62.73, H 7.04.
C
25H32O9 (476.5): calcd C 63.01, H 6.77; found C 63.12, H 6.54.
(2-Ben zyla m in om eth yl)d iben zo[24]cr ow n -8 (1). A solu-
tion of the formyl crown ether 9 (1.50 g, 3.15 mmol) and
benzylamine 10 (337 mg, 3.15 mmol) was heated under reflux
for 30 h in PhMe (150 mL) using a Dean-Stark apparatus.
When the reaction mixture had cooled to ambient temperature,
an aliquot (ca.. 1 mL) was removed, and the solvent was
evaporated in vacuo to give a yellow oily residue which was
shown by 1H NMR to be the imine; 1H NMR (300 MHz,
CDCl3): δ ) 3.85-3.98 (m, 16H), 4.13-4.23 (m, 8H), 4.81 (s,
2H), 6.85-6.95 (m, 5H), 7.16-7.45 (m, 7H), 8.28 (s, 1H). The
reaction mixture was diluted with dry MeOH (100 mL), and
then NaBH4 (1.19 g, 31.5 mmol) was added portionwise to the
stirring solution over a period of 1 h. Stirring was maintained
under ambient conditions for a further 24 h, after which time
5 M HCl solution (250 mL) was added to the reaction mixture.
The solvents were removed in vacuo, and the residue was
partitioned between 2 M NaOH solution (250 mL) and CH2Cl2
(250 mL). The aqueous layer was extracted with CH2Cl2 (2 ×
100 mL), and the combined organic extracts were dried
(MgSO4). The solvents were removed in vacuo, yielding a crude
product which was subjected to column chromatography (SiO2:
Me2CO) to yield the title compound 1 as a pale yellow oil (1.42
g, 79%); 1H NMR (300 MHz, CDCl3): δ ) 3.71 (s, 2H), 3.77 (s,
2H), 3.82 (s, 8H), 3.88-3.93 (m, 8H), 4.11-4.17 (m, 8H), 6.78-
6.91 (m, 5H), 7.20-7.34 (m, 7H); 13C NMR (75 MHz, CDCl3):
δ ) 52.9, 53.2, 69.5, 69.6, 70.0, 71.3, 114.0, 114.2, 121.0, 121.5,
127.0, 128.3, 128.5, 133.7, 140.4, 147.9, 149.0; MS (LSI): m/z
(%): 568 (15) [M + H]+ and [2M + 2H]2+, 1135 (100) [2M +
H]+; C32H41NO8 (567.7): calcd C 67.71, H 7.28, N 2.47; found
C 67.75, H 7.29, N 2.39.
(2-Ben zylam m on iu m m eth yl)diben zo[24]cr own -8 Hexa-
flu or op h osp h a te (1-H‚P F 6). Amine 1 (470 mg 0.83 mmol)
was dissolved in 5 M HCl solution (50 mL), and the resulting
solution was evaporated to dryness. The residue was dissolved
in hot water (ca. 15 mL), and a saturated aqueous solution of
NH4PF6 was added until no further precipitation occurred. The
white precipitate was collected, washed with copious amounts
of H2O, and dried to yield the hexafluorophosphate salt
1-H‚PF6 as a white powder (552 mg, 93%); mp >210 °C
1
(decomp); H NMR (300 MHz, CD3SOCD3): δ ) 3.68 (s, 8H),
3.74-3.83 (m, 8H), 4.04-4.16 (m, 12H), 6.86-6.99 (m, 4H),
7.02 (s, 2H), 7.13 (s, 1H), 7.47 (br s, 5H), 9.09 (br s, 2H); 13C
NMR (75.5 MHz, CD3SOCD3): δ ) 49.9, 50.0, 68.7, 69.0, 69.1,
70.4, 113.5, 114.0, 115.4, 121.1, 123.0, 124.2, 128.7, 129.0,
129.9, 132.0, 148.2, 148.4, 148.9; MS (LSI): m/z (%): 568 (51)
[M - PF6]+ and [2M - 2PF6]2+, 590 (21) [M - H-PF6 + Na]+,
700 (19) [M - H - PF6 + Cs]+, 1135 (100) [2M - H - 2PF6]+,
1281 (16) [2M - PF6]+; C32H42NO8PF6 (713.6): calcd C 53.86,
H 5.93, N 1.96; found C 53.60, H 5.84, N 2.05.
(2-Ben zyla m m on iu m m eth yl)d iben zo-24-cr ow n -8 Tr i-
flu or oa ceta te (1-H‚O2CCF 3). A CH2Cl2 solution of the amine
1 was treated with an excess of CF3CO2H. Removal of the
solvents in vacuo afforded 1-H‚O2CCF3 as a pale yellow solid.
Single crystals suitable for X-ray crystallographic analysis
Ben zo(5-for m ylm eta p h en ylen e)[25]cr ow n -8 (27). PCC
(3.68 g, 17.1 mmol) was added to a stirred solution of 26 (3.27
g, 6.83 mmol) in dry CH2Cl2 (150 mL). Although initially
were obtained when a solution of the salt in EtOAc/C6H14
/
(42) SHELXTL PC version 5.03, Siemens Analytical X-ray Instru-
ments, Inc., Madison, WI, 1994.
(43) Copies of the crystallographic data can be obtained free of
charge on application to CCDC, 12 Union Road, Cambridge CB12 1EZ,
UK (Fax: (+44) 1223-336033; E-mail: teched@ccdc.cam. ac.uk).
MeCN (10:10:1) was allowed to stand at 20 °C for ap-
proximately 1 d. Crystal data for C32H42NO8‚CF3CO2‚CF3CO2H:
Mr ) 795.7, monoclinic, space group C2/c, a ) 22.936 (4), b )
23.249 (3), c ) 15.393 (3) Å, â ) 112.14 (1)°, V ) 7603 (2) Å3,

6872 J . Org. Chem., Vol. 66, No. 21, 2001
Cantrill et al.
orange, the solution very quickly went black and was stirred
for a further 75 min. After this time, the reaction mixture was
filtered through a small layer of Celite, which was then washed
with CH2Cl2 (500 mL). The green CH2Cl2 filtrate was stirred
with 200 mL aliquots of 2 N NaOH solution until colorless
and subsequently dried (MgSO4), and the solvent was removed
in vacuo to yield a colorless oil. This oil was dissolved in EtOAc
and precipitated with C6H14 to afford the desired compound
(27) as a powdery white solid (2.55 g, 78%); mp 68-69 °C; 1H
NMR (500 MHz, CDCl3): δ ) 3.69-3.74 (m, 8H), 3.82-3.86
(m, 8H), 4.12-4.21 (m, 8H), 6.86-6.91 (m, 4H), 6.96-7.00 (m,
3H), 9.84 (s, 1H); 13C NMR (125 MHz, CDCl3): δ ) 68.6, 69.1,
70.0, 70.1, 71.2, 71.3, 108.9, 109.3, 115.4, 121.9, 138.4, 149.2,
160.7, 192.1; MS (FAB): m/z (%): 476.5 (100) [M]+; C25H32O9
(476.5): calcd C 63.01, H 6.77; found C 62.89, H 6.80.
(5-[Be n zyla m m on iu m m e t h yl]m et a p h e n yle n e )[25]-
cr ow n -8 Hexa flu or op h osp h a te (22-H‚P F 6). A solution of
the formyl-substituted crown ether 27 (685 mg, 1.44 mmol)
and benzylamine (10) (154 mg, 1.44 mmol) in PhMe (50 mL)
was heated under reflux for 24 h using a Dean-Stark
apparatus. The resulting solution was evaporated to dryness,
the residue was dissolved in dry MeOH (100 mL), and NaBH4
(272 mg, 7.19 mmol) was added portionwise over a period of 5
min. After being stirred under ambient conditions for 24 h,
the reaction mixture was evaporated to dryness, and the
residue was partitioned between NaOH solution (2 N, 200 mL)
and CH2Cl2 (200 mL). The aqueous layer was further extracted
with CH2Cl2 (2 × 250 mL), the combined organic extracts were
dried (MgSO4), and the resulting solution was evaporated to
dryness to yield an off-white solid. The solid was subsequently
dissolved in MeOH (100 mL), and 12 M HCl solution (10 mL)
was added carefully. After stirring for ca. 10 min, the solvents
were removed in vacuo, and the residue dissolved in hot H2O.
Addition of an excess of saturated aqueous NH4PF6 to this
solution resulted in the precipitation of a white solid. Upon
collection, the solid was dissolved in CH2Cl2, and precipitated
with Et2O to afford the desired compound as a powdery white
1
solid (152 mg, 15%); mp 177-179 °C (decomp); H NMR (400
MHz, CD3SOCD3): δ ) 3.56-3.62 (m, 8H), 3.69-3.85 (m, 8H),
4.03-4.09 (m, 6H), 4.13-4.17 (m, 6H), 6.65 (d, J ) 2.0 Hz,
2H), 6.71 (t, J ) 2.0 Hz, 1H), 6.84-6.89 (m, 2H), 6.92-6.98
(m, 2H), 7.36 (s, 1H), 7.39-7.50 (m, 5H), 9.12 (br s, 2H); 13C
NMR (125 MHz, CD3SOCD3): δ ) 50.1, 67.7, 68.1, 69.1, 70.1,
70.2, 102.3, 108.9, 114.3, 121.2, 128.7, 129.1, 130.0, 131.8,
133.7, 148.3, 159.8; MS (FAB): m/z (%): 568.4 (100) [M -
PF6]+, 1135.6 (3) [2M - H-2PF6]+; C32H42NO8PF6 (713.6): calcd
C 53.86, H 5.93, N 1.96; found C 53.47, H 5.85, N 1.86.
Ack n ow led gm en t. This research was supported by
the National Science Foundation. Anthony Pease is
thanked for producing the ray-traced pictures contained
within this paper.
Su p p or tin g In for m a tion Ava ila ble: Additional synthetic
schemes, supporting diagrammatic and spectroscopic figures,
supporting experimental data, and a mathematical model for
the aggregation of self-complementary daisy chain molecules.
This material is available free of charge via the Internet at
http://pubs.acs.org.
J O010405H