DOI: 10.1002/anie.201006470
Ligand Design
Azatriquinane as a Platform for Tripodal Metal Complexes and
Calixiform Scaffolds**
Martyn Jevric, Tao Zheng, Nabin K. Meher, James C. Fettinger, and Mark Mascal*
Azatriquinane 1 is a bowl-shaped tricycle featuring an apical,
acutely pyramidalized nitrogen.[1,2] The trefoil fusion of three
five-membered rings in 1 results in an averaged C3v-symmetric
platform, which presents the opportunity to build out from
the sites a to the nitrogen (C-1, 4, and 7) to position
substituents in an array with a conic focus. Herein, we report a
synthetic approach to substitution at these sites, leading both
to a novel, conformationally fixed tripodal ligand system and
calix-like derivatives with a basic nitrogen at the bottom of
the cavity.
Scheme 1. Reagents and conditions: a) H2CCHCH2SiMe3, AlCl3,
CH2Cl2, 16 h, 90%; b) Me3SiCCSiMe3, AlCl3, CH2Cl2, 10 h, 68%.
The centerpiece of this chemistry is perchloroazatriqui-
nacene 3, a synthetic precursor to the parent azatriquinacene
2. It can be produced in a single step and high yield by
photochemical chlorination of 1 with SO2Cl2, and is available
in multigram quantities.[1] The presence of doubly allylic a
chloroamine functions in 3 suggested the potential to
generate cationic intermediates which could alkylate elec-
tron-rich p systems. Thus, the Lewis acid catalyzed reaction of
3 with either allyltrimethylsilane or bis(trimethylsilyl)acetyl-
ene led to the corresponding triallyl 4 or triyne 5 derivatives in
good yields (Scheme 1). Although a number of opportunities
for further elaboration present themselves in 4 and 5, we were
first attracted to the prospect of applying the azatriquinane
framework as a scaffold for tripodal ligands. To this end, we
envisaged a threefold “click” reaction[3,4] with the alkyne
functions in 5 to give a tris(1,2,3-triazole) similar to the
Sharpless TBTA ligand,[5] but with the nitrogen sites preor-
ganized in a trigonal-pyramidal arrangement. This was
accomplished as shown in Scheme 2. Desilylation of 5 with
fluoride gave 6, and subsequent dehalogenation with lithium
in tert-butanol gave triethynylazatriquinacene 7. Reaction of 7
with phenyl azide provided the tris(1,2,3-triazole) derivative
8, which could be hydrogenated to the ligand 9.
Simple mixing of 9 with either ZnII acetate or CoII acetate
in methanol gave rise to metal complexes, and X-ray quality
crystals of both could be obtained by capping the metals with
a terminal chloride ligand and exchanging the counterion for
PF6À. In each case, the coordination geometry is pseudo-
À
trigonal-bipyramidal, with the axial N M bond longer than
II
À
the equatorial N M bonds. For example, in the Co complex
(Figure 1), the former is 2.382(2) ꢀ and the latter are
2.012(2), 2.020(2), and 2.030(2) ꢀ. As noted above, an
attractive characteristic of this host is its degree of preorga-
nization, compared to flexible tris(2-aminoethyl)amine
(tren)-based ligands.[6,7]
The reaction of 3 with aromatic rings gave rise to triaryl-
substituted hexachloroazatriquinacenes 10 (Scheme 3). X-ray
[*] Dr. M. Jevric, Dr. T. Zheng, Dr. N. K. Meher, Dr. J. C. Fettinger,
Prof. M. Mascal
Department of Chemistry, University of California Davis
1 Shields Avenue, Davis, CA 95616 (USA)
E-mail: mascal@chem.ucdavis.edu
[**] This work was financially supported by National Science Foundation
grant CHE-0957798.
Figure 1. ORTEP[8] views of the X-ray crystal structure of the [9-
CoCl]PF6 complex (counterion omitted for clarity). Thermal ellipsoids
are drawn at the 60% probability level.
Supporting information for this article is available on the WWW
Angew. Chem. Int. Ed. 2011, 50, 717 –719
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
717