Shape-persistent chiral alleno-acetylenic macrocycles and cyclophanes by acetylenic scaffolding with 1,3-diethynylallenes

Article abstract of DOI:10.1002/anie.200501621

Rings on her fingers: The chiral allenoacetylenic macrocycle 1 and the related cyclophane 2 with intriguing three-dimensional shapes were prepared and isolated in diastereoisomerically pure form. The symmetries and structures of these novel unsaturated hy

Full text of DOI:10.1002/anie.200501621

Communications
Macrocycles
DOI: 10.1002/anie.200501621
Shape-Persistent Chiral Alleno-Acetylenic
Macrocycles and Cyclophanes by Acetylenic
Scaffolding with 1,3-Diethynylallenes**
Severin Odermatt, J. Lorenzo Alonso-Gómez,
Paul Seiler, M. Magdalena Cid,* and
François Diederich*
The chemistry of chiral allenes has attracted increasing
interest in recent years due to the development of improved
syntheses and their emerging use in pharmaceuticals.[1] While
strained small-ring allenes have been investigated in greater
detail for their theoretical properties and their limits of
[*] J. L. Alonso-Gómez, Prof. Dr. M. M. Cid
Facultade de Química, Universidade de Vigo
Lagoas-Marcosende, 36310 Vigo (Spain)
Fax: (+34)986812262
E-mail: mcid@uvigo.es
S. Odermatt, P. Seiler, Prof. Dr. F. Diederich
Laboratorium für Organische Chemie
ETH Hönggerberg
HCI, 8093 Zürich (Switzerland)
Fax: (+41)1-632-1109
E-mail: diederich@org.chem.ethz.ch
[**] This work was supported by a grant from the ETH Research Council,
the Fonds derChemischen Industrie, MCYT, Spain (SAF2001-3288),
and the Ministerio de Educación y Ciencia (PhD fellowship to J.L.A.-
G.). We thank Dr. C. Thilgen for his help with the stereochemical
assignments.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
5074
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 5074 –5078

Angewandte
Chemie
stability and isolability,[2] macrocyclic allenes, in particular
shape-persistent ones,[3] are nearly unknown.[4] The only such
macrocycle known is a [34]cyclophane, reported by Krause
and co-workers,[5] in which four para-phenylene moieties are
bridged by four 1,3-dimethylallene-1,3-diyl linkers and which
was isolated as a mixture of several stereoisomers. This
absence of allenic macrocycles is rather surprising in viewof
the opportunities for creating newnonplanar, chiral top-
ologies and for developing newchiral host molecules.
The recently described synthesis of the first stable 1,3-
diethynylallenes[6] nowopens up the way for acetylenic
scaffolding, and here we report the preparation of the new
chiral, unsaturated macrocycles 1 and 2 (Figure 1). Com-
Figure 1. Two pairs of enantiomers and three achiral diastereoisomers
are expected for macrocycle 1; two pairs of enantiomers and two achi-
ral diastereoisomer are expected for cyclophane 2.
Scheme 1. Synthesis of the allene–alkyne macrocycle 1. Reagents and
ꢀ
conditions: a) R-C CH, [Pd(PPh3)4], CuI, iPr2NEt, (CH2Cl)2, 508C; 74%
((Æ)-4), 53% ((Æ)-5), 69% ((Æ)-6); b) K2CO3, MeOH/THF, 208C;
90%; c) nBu4NF, THF, 08C; 67%; d) (Æ)-3, [Pd(PPh3)4], CuI, (CH2Cl)2,
Cy2NMe, 608C; 71%; e) KOH, C6H6, 808C; 65%; f) CuCl, TMEDA,
(CH2Cl)2, 508C; 91%; g) 1. nBu4NF, ortho-nitrophenol, THF, 208C;
2. CuCl, CuCl2, pyridine, [11]ꢁ10À4 m; 80% total yield of seven stereo-
isomers. Cy=cyclohexyl, TMEDA=N,N,N’,N’-tetramethylethylenedi-
amine.
pound 1 is the first alleno-acetylenic macrocycle without
aromatic rings in the backbone. It exists as seven stereo-
isomers, two pairs of enantiomers and three achiral diaster-
eoisomers, which could be isolated as pure compounds. For
one of the enantiomeric pairs, the relative configuration could
be assigned unambiguously. Cyclophane 2 was prepared as a
mixture of six stereoisomers, two pairs of enantiomers and
two achiral diastereoisomers; remarkably, both racemates
and the two achiral isomers could all be isolated and their
relative configurations assigned.
The synthesis of allenes (Æ )-4 through (Æ )-6 (Scheme 1)
by Pd-catalyzed cross-coupling of ester (Æ )-3 with the
corresponding alkynes followed the previously described
protocol;[6] however, a considerable improvement in reaction
yields was obtained by using pentafluorobenzoate as the
leaving group.[7] Removal of the silyl groups from (Æ )-4 and
(Æ )-6 afforded the free terminal alkynes (Æ )-7 and (Æ )-8,
respectively. A second cross-coupling of (Æ )-3 with (Æ )-8
provided diallene 9 in 71% yield as a mixture of two
diastereoisomeric pairs of enantiomers as evident from the
13C NMR spectrum (125 MHz, C6D6). All attempts to sepa-
rate the two diastereoisomeric pairs by HPLC failed.[8] The
acetonide protecting group in 9 was removed with KOH in
hot benzene to give the free alkyne 10. Subsequent oxidative
homocoupling under Hay conditions[17] yielded tetraallene 11
in high yield (91%). The formation of different stereoisomers
was expected, but the spectroscopic data did not give precise
information on the composition of the mixture and HPLC
analyses showed mainly one peak. Deprotection of 11 with
nBu4NF in THF in the presence of ortho-nitrophenol
provided the crude free terminal bis(alkyne) which, after a
filtration through SiO2, was subjected directly to the final ring
closure under high dilution (10À4 m). The Eglinton–Galbraith
conditions[9] proved to be the most effective, providing a total
yield of 80% of the macrocyclic tetraallene 1: higher
molecular weight products were not detected by analytical
gel permeation chromatography (GPC) after filtration
through SiO2 (for experimental details see the Supporting
Information).
The individual detection of the expected stereoisomers
(two racemates and three achiral diastereoisomers) was
possible by analytical HPLC on SiO2 (see the Supporting
Information).[10] The increased rigidity of the formed macro-
cycles 1, compared to the flexible acyclic precursors 11, most
likely had beneficial effects on the separation. Preparative
separation was hampered by the highly apolar nature and the
Angew. Chem. Int. Ed. 2005, 44, 5074 –5078
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5075

Communications
reduced solubility (despite the tert-butyl groups) of the
equivalent cumulenic C atoms of the four allene moieties. The
relative configuration of the other isolated stereoisomers
could not be assigned based on symmetry considerations. For
the second racemate, D2-symmetric (M,M,M,M)/(P,P,P,P)-1,
and the three achiral isomers, C2h-symmetric (M,M,P,P)-1 and
(M,P,P,M)-1, and C2v-symmetric (M,P,M,P)-1, two magneti-
cally different tert-butyl groups and one signal for the central
C atom of the four allene moieties was expected and observed
in the 13C NMR spectra (see the Supporting Information).
The synthesis of anthracenophane 2 started with the
Sonogashira cross-coupling[13] of (Æ )-7 and (Æ )-8 with 9,10-
dibromoanthracene to give (Æ )-12 and (Æ )-13, respectively
(Scheme 2). A series of four Sonogashira cross-couplings
alternating with three deprotection steps afforded, after
acetonide cleavage, the linear tetraallene , the precursor to
cyclophane 2, as a mixture of stereoisomers (15% overall
yield starting from (Æ )-13), which were not separated.
Macrocyclization with one equivalent of [Pd(PPh3)4] and
CuI provided 2 in 60% yield as a mixture of two racemates
and two achiral diastereoisomers, which were separated by
preparative HPLC (“Buckyclutcher 1”; for experimental
details see the Supporting Information).[11]
compounds. The five stereoisomeric products were finally
isolated with the aid of two different preparative HPLC
separations on “Buckyclutcher 1” and Kromasil columns.[11]
The detected ratio of the stereoisomers roughly corresponded
to the values of 1:1:4:1:1 expected for an unselective ring
closure. The formation of racemic, C1-symmetric (M,P,P,P)/
(P,M,M,M)-1 (Figure 2)[12] is statistically favored over the
formation of the other isomers.
In the case of cyclophane 2, the relative configuration of
all the stereoisomers could be assigned. Thanks to its
symmetry the major product was identified by 1H NMR
spectroscopy (300 MHz, CDCl3) as C2-symmetric (M,P,P,P)/
(P,M,M,M)-2 with four magnetically different tert-butyl
groups (Figure 3a). On the other hand, the spectrum of the
achiral C2h-symmetric (M,M,P,P)-2 features only two differ-
ent tert-butyl resonances (Figure 3b). The 1H NMR spectra of
the other two products display only one tert-butyl signal as
expected as a result of their higher symmetry (D4 and D2d,
respectively). Fortunately, the structure of D4-symmetric
(P,P,P,P)/(M,M,M,M)-2 could be assigned unambiguously by
an X-ray crystal structure analysis of moderate resolution
(Figure 3c; for further details of the crystal structure see the
Figure 2. Tube model and 1H NMR spectrum (300 MHz, C6D5CD3) of
twisted, C1-symmetric (M,P,P,P)/(P,M,M,M)-1. The conformation of
the (M,P,P,P) isomerwas optimized by AM1 calculations using
Spartan Pro.[12]
The assigned relative configuration of (M,P,P,P)/
(P,M,M,M)-1 was clearly proven spectroscopically. In the
1H NMR spectrum in deuterated benzene or toluene, all eight
expected resonances of the eight nonequivalent tert-butyl
groups were clearly separated (Figure 2). Furthermore, the
13C NMR spectrum showed four signals for the four non-
Scheme 2. Synthesis of anthracenophane 2. Reagents and conditions: a) 9,10-dibromoanthracene, [PdCl2(PPh3)2], CuI, TMEDA, toluene, 1108C;
56% ((Æ)-12), 73% ((Æ)-13); b) (Æ)-7, [PdCl2(PPh3)2], CuI, TMEDA, toluene, 1108C; 83%; c) nBu4NF, THF, 208C; 95% (15); 96% (17); 86%
(19); d) (Æ)-12, [PdCl2(PPh3)2], CuI, TMEDA, toluene, 1108C; 63% (16); 65% (18); e) 9,10-dibromoanthracene, [PdCl2(PPh3)2], CuI, TMEDA,
toluene, 1108C; 64%; f) NaOH, C6H6, 908C, 85%; g) [Pd(PPh3)4], CuI, iPr2NEt, toluene, 1108C, 2 h, then air, CH2Cl2, 208C, 24 h; 60%.
5076 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 5074 –5078

Angewandte
Chemie
Figure 3. Tube models and 1H NMR spectra of a) C2-symmetric (M,P,P,P)/(P,M,M,M)-2, ((P,M,M,M) isomershown), b) C2h-symmetric (M,M,P,P)-
1
2, c) (P,P,P,P)/(M,M,M,M)-2, and d) D2d-symmetric (M,P,M,P)-2. H NMR spectra in CDCl3. The models in (a), (b), and (d) show conformations
optimized by AM1 calculations.[12] The X-ray crystal structure of (M,M,M,M)-2 is shown in (c). (Note that in the crystal structure the actual sym-
metry is C1 in contrast to the observation from NMR spectroscopy; for further details see the Supporting Information.)
Supporting Information and ref. [14]). Finally, the structure of
the last isomer, D2d-symmetric (M,P,M,P)-2, was assigned by
exclusion (Figure 3d).
D2d). This ratio differs slightly from the statistically expected
ratio of 4:2:1:2. Photoisomerization of 2 but not of 1 is
probably a result of the anthracene moieties, which can act as
intramolecular sensitizers.[16]
Macrocycle 1 and cyclophane 2 differ substantially in their
properties. While the isomers of 1 are colorless solids, those of
2 are brown-colored. All compounds are stable for weeks
upon storage in the air at ambient conditions, when protected
from light. The characteristic anthracene bands in the visible
region of the electronic absorption spectra of the isomers of 2
are retained but red-shifted by nearly 100 nm (see the
Supporting Information). Similar to anthracene, the macro-
cycles showfluorescence ( F = 0.2; lmax = 478–481 and 510–
515 nm at lexc = 315 nm). The isomers of 2 undergo two
reversible 2eÀ oxidation and one reversible 4eÀ reduction
steps under cyclovoltametric conditions.[15] Whereas the
isomers of 1 are stable upon irradiation, photoisomerization
of the allene moieties in 2 occurs. The different stereoisomers
interconvert slowly upon irradiation (within days under
sunlight or within hours under xenon-lamp irradiation,
450 W; lexc = 300 nm or 463 nm) to finally reach the photo-
stationary state with an isomeric ratio of 10:5:1:5 (C2/C2h/D4/
In summary, we have prepared, separated, and charac-
terized the first members of a newclass of unsaturated alleno-
acetylenic macrocycles as well as four new cyclophanes which
undergo photoisomerization. The three-dimensional shapes
and symmetries of these novel hydrocarbons are intriguing,
and the preparation of optically pure derivatives with
potentially promising chiroptical properties is being pursued.
Furthermore, the host–guest complexation properties of 2
and related macrocycles, which are obtained by Diels–Alder
reaction of the anthracene moieties, are nowunder inves-
tigation.
Experimental Section
(M,P,P,P)/(P,M,M,M)-1: A solution of 11 (216 mg, 0.203 mmol) and
ortho-nitrophenol (43 mg, 0.31 mmol) in THF (100 mL) was treated
with nBu4NF (0.30 mL of a 1m solution in THF, 0.30 mmol). After the
Angew. Chem. Int. Ed. 2005, 44, 5074 –5078
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5077

Communications
reaction mixture had been stirred for 1 h at 208C, the solvent was
[7] a) M. I. Calaza, E. Hupe, P. Knochel, Org. Lett. 2003, 5, 1059 –
1061; b) N. Harrington-Frost, H. Leuser, M. I. Calaza, F. Kneisel,
P. Knochel, Org. Lett. 2003, 5, 2111 – 2114.
[8] R. W. Saalfrank, M. Haubner, C. Deutscher, W. Bauer, Eur. J.
Org. Chem. 1999, 2367 – 2372.
removed under reduced pressure. The orange oil was taken up in
cyclohexane/CH2Cl2 and filtered through a plug of SiO2 (eluent:
cyclohexane/EtOAc 20:1). After concentration, the residue was
dissolved in pyridine (0.30 L) and added to a degassed solution of
CuCl (1.50 g, 15.1 mmol) and CuCl2 (0.20 g, 1.51 mmol) in pyridine
(1.20 L). The green mixture was degassed with Ar and stirred for 3 d
at 208C. Pyridine was removed under vacuum until a volume of about
100 mL remained, and cyclohexane and toluene were added. The Cu
salts were extracted with aq NH4Cl solution (3 ” ) and aq EDTA
(0.2m) solution. The organic layer was dried over MgSO4 and
concentrated. The residual brown solid was filtered twice through
SiO2 and eluted with toluene and cyclohexane successively to give a
light brown solid (160 mg). Analytical HPLC analysis (Kromasil,
hexane, 1 mLminÀ1) showed roughly a 1:1:4:1:1 mixture of stereo-
isomers. Isolation of the most abundant C1-symmetric isomer
( ꢁ 60 mg) on preparative scale was successful on a “Buckyclutcher 1”
column (eluent: hexane, 10 mLminÀ1). White solid. M.p. > 2908C
(decomp); 1H NMR (300 MHz, C6D5CD3): d = 1.007 (s, 9H), 1.040 (s,
9H), 1.075 (s, 9H), 1.093 (s, 9H), 1.106 (s, 9H), 1.116 (s, 9H), 1.127 (s,
9H), 1.142 ppm (s, 9H); 13C NMR (75 MHz, C6D5CD3): d = 29.51,
29.54, 29.56, 29.62, 29.73, 29.75, 29.80, 29.84, 36.01 (2 ” ), 36.06 (2 ” ),
36.18, 36.21, 36.34, 36.43, 76.43, 76.49 (3 ” ), 78.86, 78.93, 79.11, 79.21,
87.52, 87.64, 88.06, 88.16, 104.04, 104.13, 104.24, 104.44, 105.43, 105.47,
105.60 (2 ” ), 214.56, 215.07, 215.20, 215.46 ppm; UV/Vis (hexane):
lmax (e) = 237 (126700), 266 (sh, 48800), 281 (sh, 32500), 295 (sh,
21300), 321 nm (12800); MALDI-MS (%): m/z = 783.53 (19,
[9] G. Eglinton, A. R. Galbraith, Chem. Ind. 1956, 737 – 738.
[10] For good separations, the HPLC columns should be very well
conditioned. Best analytical results were obtained with
a
Kromasil 100 Si column (5 mm) and pure hexane as the eluent.
[11] Rexchrom “Buckyclutcher 1” Prep, 10 mm 100 Š Trident-Tri-
DNP, 50 cm ” 21.1 mm ID and Spherical Silica Kromasil Prep,
5 mm, 100 Š, 30 cm ” 21.1 mm ID.
[12] Spartan 02, Wavefunction, Inc. Irvine, CA 2002.
[13] J. A. Marsden, M. M. Haley in Metal-Catalyzed Cross-Coupling
Reactions, Vol. 1 (Eds.: A. de Meijere, F. Diederich), Wiley-
VCH, Weinheim, 2004, pp. 317 – 394.
[14] X-ray crystal structure of (P,P,P,P)/(M,M,M,M)-2: Crystal data
at 220(2) K for C116H104 ” 4CH3OH ” 4H2O. Mr = 1698.22. Tri-
clinic, space group P1 (no. 2), 1calcd = 0.997 gcmÀ3, Z = 2, a =
¯
11.2927(2), b = 13.4550(3), c = 37.9436(8) Š, a = 83.672(1), b =
89.873(1), g = 80.841(1)8, V= 5656.4(2) Š3. Bruker-Nonius
Kappa-CCD diffractometer, MoKa radiation, l = 0.7107 Š, m =
0.061 mmÀ1. A yellowcrystal of ( P,P,P,P)/(M,M,M,M)-2 (linear
dimensions ca. 0.25 ” 0.15 ” 0.13 mm) was obtained by slow
evaporation of a CH2Cl2/CH3OH solution. It was mounted at
lowtemperature to prevent evaporation of enclosed solvent. The
numbers of measured and unique reflections are 15697 and
11184, respectively (Rint = 0.030). The structure was solved by
direct methods (SHELXS-97; G. M. Sheldrick, SHELXS-97
Program for the Solution of Crystal Structures, University of
Göttingen, Germany, 1997) and refined by full-matrix least-
squares analysis (SHELXL-97; G. M. Sheldrick, SHELXL-97
Program for the Refinement of Crystal Structures, University of
Göttingen, Germany, 1997), using an isotropic extinction
correction. The subunit C(88)–C(102) is disordered over two
orientations (for arbitrary atom numbering, see the Supporting
Information). For C(91)–C(98), C(100), and C(101), two sets of
atomic parameters with population parameters of 0.5 were
refined. The derived solvents exhibit static and dynamic
disorder. All heavy atoms were refined anisotropically, H
atoms of the ordered part isotropically, whereby H-atom
positions are based on stereochemical considerations. Final
R(F) = 0.138, wR(F2) = 0.325 for 1277 parameters and 7418
reflections with I > 2s(I) and 2.93 < q < 21.018 (corresponding
R-values based on all 11184 reflections are 0.185 and 0.347
respectively). CCDC-271068 contains the supplementary crys-
tallographic data for this paper. These data can be obtained free
of charge from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[M+K]+), 767.55 (34, [M+Na]+), 745.57 (100, [M+H]+), 689.50 (22,
[M+HÀtBu]+); HR-MALDI-MS: m/z = 745.5694 ([M+H]+, C56H73
;
+
calcd 745.5707).
Received: May 11, 2005
Published online: July 20, 2005
Keywords: allenes · chirality · cross-coupling · macrocycles ·
.
photoisomerization
[1] a) Modern Allene Chemistry, Vols. 1 and 2 (Eds.: N. Krause,
A. S. K. Hashmi), Wiley-VCH, Weinheim, 2004; b) A. Hoff-
mann-Röder, N. Krause, Angew. Chem. 2002, 114, 3057 – 3059;
Angew. Chem. Int. Ed. 2002, 41, 2933 – 2935; c) A. Hoffmann-
Röder, N. Krause, Angew. Chem. 2004, 116, 1216 – 1236; Angew.
Chem. Int. Ed. 2004, 43, 1196 – 1216.
[2] H. Hopf, Classics in Hydrocarbon Chemistry, Wiley-VCH,
Weinheim, 2000, chap. 9, pp. 171 – 196.
[3] a) C. S. Jones, M. J. OꢀConnor, M. M. Haley in Acetylene
Chemistry: Chemistry, Biology, and Material Science (Eds.: F.
Diederich, P. J. Stang, R. R. Tykwinski), Wiley-VCH, Weinheim,
2005, pp. 303 – 385; b) S. Höger in Acetylene Chemistry: Chemis-
try, Biology, and Material Science (Eds.: F. Diederich, P. J. Stang,
R. R. Tykwinski), Wiley-VCH, Weinheim, 2005, pp. 427 – 452;
c) for a recent report: W. Zhang, J. S. Moore, J. Am. Chem. Soc.
2004, 126, 12796.
[15] M. Gross, J.-P. Gisselbrecht, C. Boudon, F. Diederich, J. L.
Alonso-Gꢁmez, S. Odermatt, unpublished results.
[16] O. Rodriguez, H. Morrison, J. Chem. Soc. Chem. Commun. 1971,
679.
[17] A. S. Hay, J. Org. Chem. 1962, 27, 3320 – 3321.
[4] In contrast to allenes, [3]- and [5]cumulenes are found more
frequently in macrocyclic structures, such as planar dehydroan-
nulenes; see: a) ref. [2]; b) M. Nakagawa, Angew. Chem. 1979,
91, 215 – 226; Angew. Chem. Int. Ed. Engl 1979, 18, 202 – 214.
[5] a) S. Thorant, F. Vögtle, N. Krause, Angew. Chem. 1999, 111,
3929 – 3931; Angew. Chem. Int. Ed. 1999, 38, 3721 – 3723; b) C.
Kind, M. Reiher, J. Röder, B. A. Hess, Phys. Chem. Chem. Phys.
2000, 2, 2205 – 2210.
[6] a) R. C. Livingston, L. R. Cox, V. Gramlich, F. Diederich,
Angew. Chem. 2001, 113, 2396 – 2399; Angew. Chem. Int. Ed.
2001, 40, 2334 – 2337; b) R. Livingston, L. R. Cox, S. Odermatt,
F. Diederich, Helv. Chim. Acta 2002, 85, 3052 – 3065.
5078
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 5074 –5078