Metallotectons: Using enantiopure tris(dipyrrinato)cobalt(III) complexes to build chiral molecular materials

Article abstract of DOI:10.1039/b705212j

Resolution of a tricarboxylic acid derived from a chiral tris(dipyrrinato)cobalt(III) complex provides a series of enantiopure metallotectons that crystallise as porous hydrogen-bonded networks. The Royal Society of Chemistry.

Full text of DOI:10.1039/b705212j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Metallotectons: using enantiopure tris(dipyrrinato)cobalt(III)
complexes to build chiral molecular materials{
Shane G. Telfer*ab and James D. Wuesta
Received (in Austin, TX, USA) 5th April 2007, Accepted 30th April 2007
First published as an Advance Article on the web 29th May 2007
DOI: 10.1039/b705212j
To circumvent these problems, we sought to develop an
Resolution of a tricarboxylic acid derived from a chiral
tris(dipyrrinato)cobalt(III) complex provides a series of enan-
tiopure metallotectons that crystallise as porous hydrogen-
bonded networks.
alternative strategy based on robust preformed transition metal
complexes decorated with substituents designed to induce the
formation of hydrogen-bonded networks. Goldberg has demon-
strated that metalloporphyrin-based networks can be constructed
using this approach,6 and Braga et al. have used substituted
metallocenes in a similar way.7
Reliable supramolecular synthons,1 such as cyclic hydrogen-
bonded dimers of carboxylic acids, can be used to control the
spacing and relative orientation of individual molecules in
crystalline solids. This strategy has become one of the guiding
principles of crystal engineering. Our group has sought to develop
this method, combine it with the power of organic synthesis, and
thereby create new crystalline molecular materials with predictable
structures and properties.2 In this strategy for purposeful
construction, the individual molecular subunits have been called
tectons.3 Tectons have tended to be purely organic compounds
that can engage in strong directional intermolecular interactions
such as hydrogen bonds. Recently, however, we and others have
begun to explore a complementary approach in which the tectons
are coordination complexes of transition metals.4–7 The unique
geometries of such complexes, as well as their special catalytic,
magnetic, and optoelectronic properties, expand the scope for
developing useful new materials.
The present report focuses on the use of tris(dipyrrinato)cobalt-
(III) complexes (1) as metallotectons (Scheme 1).8 The octahedral
cobalt(III) core of these complexes is diamagnetic and kinetically
inert, and the three monoanionic chelating dipyrrinato ligands are
oriented in a chiral, propeller-like arrangement with D3 symmetry.
Diverse functional groups can be introduced on the periphery of
these complexes, either during synthesis of the dipyrrin ligands or
subsequently by reactions occurring after complexation has taken
place. Such dipyrrinato complexes are attractive sources of
metallotectons because they are robust enough to withstand
conditions typically used in many functional group transforma-
tions, and their neutrality facilitates purification by conventional
chromatography. Previously reported tris(dipyrrinato)cobalt(III)
complexes include compounds 1a, 1b, and 1f–h, which were
reported by Dolphin et al.,9 and complexes 1d and 1i–j, which were
elegantly used by Cohen et al. as ligands for coordination
polymers.10,11
Although coordinative interactions have already been used in
concert with hydrogen bonding to create crystalline hybrid
inorganic–organic materials,4–7 much of this work has involved
grafting hydrogen-bonding groups onto simple ligands such as
pyridines, nitriles, and carboxylic acids. Moreover, the coordina-
tive interactions and the hydrogen bonds have frequently been
formed in a single synthetic step, and discrete metallotectons have
not been isolated. This approach has several drawbacks. Most
notably, there is a lack of control over the coordination sphere of
the metal and consequently over the ultimate structure and
properties of the resulting material. Further disadvantages include
(i) difficulties in identifying which factors determine the observed
crystalline architecture, (ii) the need to crystallise under the same
experimental conditions used in the synthesis, and (iii) the
possibility that functional groups introduced to favour hydrogen
bonding may coordinate to the metal instead.
The published synthesis of complex 1a involves generating
dipyrrin 2, followed by reaction with [Co(py)4Cl2]Cl.9 We modified
this procedure by replacing [Co(py)4Cl2]Cl with commercially
available Na3[Co(NO2)6], which has been used to prepare related
complexes.11,12 As reported, triester 1a is readily hydrolysed to give
tricarboxylic acid 1b.9
The presence of free carboxyl groups in chiral complex 1b
allowed us to resolve it via classical procedures involving the
formation of diastereomeric salts with enantiomerically pure bases.
aDe´partement de Chimie, Universite´ de Montre´al, Montre´al, Que´bec,
Canada H3C 3J7
bMacDiarmid Institute for Advanced Materials and Nanotechnology,
Institute of Fundamental Sciences, Massey University, Private Bag 11
222, Palmerston North, New Zealand H3C 3J7.
E-mail: s.telfer@massey.ac.nz
{ Electronic supplementary information (ESI) available: Detailed descrip-
tions of experimental procedures and X-ray crystallographic studies. See
DOI: 10.1039/b705212j
Scheme 1 Structures of selected tris(dipyrrinato)cobalt(III) complexes
and synthesis of complexes 1b–e. Reaction conditions: (a) KOH (aq); (b)
(i) oxalyl chloride, (ii) NH3; (c) Tf2O, pyridine; (d) dicyandiamide, KOH.
3166 | Chem. Commun., 2007, 3166–3168
This journal is ß The Royal Society of Chemistry 2007

View Article Online
In testing various bases, we found that cinchonidine precipitated
complex 1b from THF as a crystalline salt. Optimisation of the
resolution afforded the two enantiomers D- and L-1b in high yields
and with ees of 100% (as determined by chiral HPLC). Despite the
long tradition of resolving chiral organic compounds by forming
diastereomeric salts, examples in which neutral transition metal
complexes have been resolved by this method are rare.13
related tris(dipyrrinato)cobalt(III) complex to be made on the basis
of CD spectroscopy alone, without the requirement for analysis by
chiral HPLC.
Crystals of metallotecton D-1b could be grown by slowly
evaporating a solution of the complex in ethyl acetate–toluene, and
their structure was determined by X-ray diffraction (Fig. 2).{ The
complex was found to crystallise in the enantiomorphic space
group P1, with two independent molecules in the unit cell. Both
molecules proved to have the same absolute configuration (D), as
anticipated by analysis of the CD spectra (vide supra).
Resolved complexes D- and L-1b could be easily converted into
enantiomerically pure derivatives D- and L-1a and 1c–e by
straightforward transformations (Scheme 1). Analysis by HPLC
and circular dichroism (CD) spectroscopy (vide infra) indicated
that no detectable racemisation occurred during these reactions.
The CD spectra of enantiopure complexes 1b, 1d, and 1e in
solution all displayed bisignate curves with peaks near 515 nm and
468 nm (Fig. 1). These signals presumably originate from an intra-
ligand S0 A S1 electronic transition that is polarised along the long
axis of the fully conjugated dipyrrinato chromophore.9,14 Due to
the spatial proximity of the three ligands in these complexes,
exciton coupling of intra-ligand transitions is possible, resulting in
the appearance of a characteristic split Cotton effect (‘exciton
couplet’).15 The sign of this exciton couplet is directly related to the
relative disposition of the dipyrrin ligands and thus to the absolute
stereochemical configuration of the metal centre. This correlation
allowed us to assign the L configuration to complexes displaying a
positive peak at higher wavelengths, and the D configuration to
those displaying a negative peak in this region. Although exciton
coupling theory reliably correlates the sign of the observed couplet
and the relative stereochemistry of the interacting chromophores,
our stereochemical assignments are also supported by X-ray
crystallography (vide infra),{ and they agree with a large body of
literature describing [M(bipy)3] and [M(phen)3] complexes (bipy =
The D3-symmetric metallotecton 1b is designed so that its three
carboxyl groups extend outwards from the periphery and can
freely hydrogen bond with neighbouring tectons, thereby generat-
ing a multidimensional network. In the observed structure,
however, only two carboxyl groups engage directly in intertectonic
hydrogen bonding, creating 1D zig-zag chains (Fig. 2b). This
primary network motif is also found in the crystal structure of rac-
1b.17 The chains are then linked into a complex open 3D network
by additional hydrogen bonds involving included molecules of
H2O. These additional hydrogen bonds link the third carboxyl
group of each metallotecton (the one not engaged in forming the
primary chains) to the oxygen atoms of two dimer-forming
carboxyl groups belonging to two different chains (see the ESI for
details{). Approximately 32% of the volume of the crystal is
accessible to guests, as estimated by standard methods.18 No
significant continuous channels are present; instead, the available
volume consists of discrete cavities occupied predominantly by
toluene (three molecules per unit cell).
Supramolecular synthons involving diaminotriazine (DAT)
groups are useful in crystal engineering because multiple hydrogen
bonds are present. Networks held together by interactions of DAT
groups have proven to be highly porous and robust,19 thereby
encouraging us to synthesise metallotecton L-1e (Scheme 1) and to
determine its crystal structure. Single crystals of complex L-1e were
grown by allowing MeOH to diffuse into a solution of the tecton
in DMSO. The compound was found to crystallise in the space
group P1. Two independent complexes, 3.5 molecules of H2O, and
3.5 molecules of DMSO comprise the unit cell.
2,29-bipyridine; phen
= 1,10-phenanthroline) in which the
polarisation of intra-ligand transitions is analogous to that of the
dipyrrin ligands in complexes 1.16 As expected, the CD spectrum
of complex D-1b mirrors that of enantiomer L-1b.
The amplitudes of the CD spectra of complexes L-1b, 1d, and 1e
(defined as De1 2 De2, where De1 and De2 are the heights of the
high- and low-energy peaks, respectively) lie in the narrow range
1025–1100 M21 cm21. The spectral amplitudes are presumably
similar because the transitions responsible for the CD signal
originate in the dipyrrinato groups, and there is little electronic
communication with substituents located on the periphery. This is
a useful observation, because it allows an estimate of the ee of any
The resulting structure can be considered to be constructed from
sheets in which each metallotecton is hydrogen bonded to three
neighbours (Fig. 3). Each DAT group participates in four
hydrogen bonds according to established motifs. A noteworthy
Fig. 2 (a) ORTEP view of the structure of one of the two crystal-
lographically-independent molecules of metallotecton D-1b. Thermal
ellipsoids are set at the 30% level. Atoms of Co appear in orange, N in
blue, O in red, and C in grey. (b) Representation of 1D zig-zag chains of
metallotecton D-1b linked by hydrogen bonds involving two of the three
carboxyl groups of each tecton. Guest molecules and most hydrogen
atoms have been omitted for clarity.
Fig. 1 CD spectra of solutions of metallotectons D-1b (grey), L-1b (red),
D-1d (green), and D-1e (blue) in THF (1025 M).
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 3166–3168 | 3167

View Article Online
Notes and references
{ CCDC 643751–643752. For crystallographic data in CIF format see
DOI: 10.1039/b705212j
1 G. R. Desiraju, Angew. Chem., Int. Ed. Engl., 1995, 34, 2311–2327.
2 J. D. Wuest, Chem. Commun., 2005, 5830–5837.
3 M. Simard, D. Su and J. D. Wuest, J. Am. Chem. Soc., 1991, 113,
4696–4697.
4 For reviews, see: D. Braga, L. Brammer and N. R. Champness,
CrystEngComm, 2005, 7, 1–19; L. Brammer, Chem. Soc. Rev., 2004, 33,
476–489; C. B. Aakero¨y and A. M. Beatty, in Comprehensive Coordina-
tion Chemistry II, ed. J. A. McCleverty, T. J. Meyer and A. B. P. Lever,
Elsevier, Oxford, 1st edn, 2004, vol. 1, ch. 37, pp. 679–688; A. M. Beatty,
Coord. Chem. Rev., 2003, 246, 131–143.
5 For additional references, see: A. Bacchi, E. Bosetti and M. Carcelli,
CrystEngComm, 2007, 9, 313–318; M. B. Smith, S. H. Dale, S. J. Coles,
T. Gelbrich, M. B. Hursthouse, M. E. Light and P. N. Horton,
CrystEngComm, 2007, 9, 165–175; P. Dechambenoit, S. Ferlay,
M. W. Hosseini, J.-M. Planeix and N. Kyritsakas, New J. Chem.,
2006, 30, 1403–1410; L. M. Gruia, F. D. Rochon and A. L. Beauchamp,
Can. J. Chem., 2006, 84, 949–959; R. Sekiya, S.-i. Nishikiori and
K. Ogura, Inorg. Chem., 2006, 45, 9233–9244; S. U. Son, J. A. Reingold,
G. B. Carpenter, P. T. Czech and D. A. Sweigart, Organometallics,
2006, 25, 5276–5285; S. Kitagawa and K. Uemura, Chem. Soc. Rev.,
2005, 34, 109–119; T. J. Burchell, D. J. Eisler and R. J. Puddephatt,
Chem. Commun., 2004, 944–945; C. B. Aakero¨y, J. Desper and J. Valde´s-
Mart´ınez, CrystEngComm, 2004, 6, 413–418; M. Tadokoro, H. Kanno,
T. Kitajima, H. Shimada-Umemoto, N. Nakanishi, K. Isobe and
K. Nakasuji, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 4950–4955;
C. B. Aakero¨y, A. M. Beatty and D. S. Leinen, Angew. Chem., Int. Ed.,
1999, 38, 1815–1819; C. L. Schauer, E. Matwey, F. W. Fowler and
J. W. Lauher, J. Am. Chem. Soc., 1997, 119, 10245–10246;
M. Munakata, L. P. Wu, M. Yamamoto, T. Kuroda-Sowa and
M. Maekawa, J. Am. Chem. Soc., 1996, 118, 3117–3124; A. D. Burrows,
C.-W. Chan, M. M. Chowdhry, J. E. McGrady and D. M. P. Mingos,
Chem. Soc. Rev., 1995, 24, 329–339.
Fig. 3 (a) ORTEP view of the structure of metallotecton L-1e. Thermal
ellipsoids are set at the 30% level. Atoms of Co appear in orange, N in
blue, and C in grey. (b) The hydrogen-bonded network formed by
compound L-1e, with the hexameric rosette of DAT groups highlighted in
pink. Most H atoms and all guest molecules have been omitted for clarity.
feature of the structure is the formation of hexameric rosettes of
DAT groups (Fig. 3b). Although each sheet consists of an open
hydrogen-bonded network, adjacent layers are offset, and no
significant channels penetrate the structure. Nevertheless, 32% of
the volume is accessible to guests.18
6 I. Goldberg, Chem. Commun., 2005, 1243–1254.
7 D. Braga, S. L. Giaffreda, F. Grepioni, L. Maini and M. Polito, Coord.
Chem. Rev., 2006, 250, 1267–1285.
8 For a review of dipyrrins and their complexes, see: T. E. Wood and
A. Thompson, Chem. Rev., 2007, 107, 1831–1861.
Various supramolecular synthons involving DAT groups have
been identified,19 but we believe that metallotecton L-1e forms the
first hexameric DAT rosette characterised by X-ray crystallo-
graphy. This motif has been postulated to explain the aggregation
of certain DAT-substituted compounds in liquid crystals,20 in 2D
assemblies on graphite surfaces,21 and in reverse micelles.22 The
internal diameter of the rosettes formed by L-1e (average
9 C. Bru¨ckner, Y. Zhang, S. J. Rettig and D. Dolphin, Inorg. Chim. Acta,
1997, 263, 279–286.
10 S. R. Halper, L. Do, J. R. Stork and S. M. Cohen, J. Am. Chem. Soc.,
2006, 128, 15255–15268; D. L. Murphy, M. R. Malachowski, C. F.
Campana and S. M. Cohen, Chem. Commun., 2005, 5506–5508.
11 S. R. Halper and S. M. Cohen, Inorg. Chem., 2005, 44, 486–488.
12 S. R. Halper, J. R. Stork and S. M. Cohen, Dalton Trans., 2007,
1067–1074.
13 For an early example, see: L. Westman and K. L. Rinehart, Jr., Acta
Chem. Scand., 1962, 16, 1199–1205.
…
˚
˚
transannular N N distance = 7.7 A) is close to the value (7 A)
determined by analysis of STM images.21 The crystal structure of
compound L-1e gives firm support to these earlier propositions
and underscores the importance of hexameric rosettes in the
supramolecular chemistry of DAT groups.
14 L. Yu, K. Muthukumaran, I. V. Sazanovich, C. Kirmaier, E. Hindin,
J. R. Diers, P. D. Boyle, D. F. Bocian, D. Holten and J. S. Lindsey,
Inorg. Chem., 2003, 42, 6629–6647; J. Karolin, L. B.-A. Johansson,
L. Strandberg and T. Ny, J. Am. Chem. Soc., 1994, 116, 7801–7806.
15 S. G. Telfer, N. Tajima and R. Kuroda, J. Am. Chem. Soc., 2004, 126,
1408–1418.
16 B. Bosnich, Acc. Chem. Res., 1969, 2, 266–273.
17 S. G. Telfer and J. D. Wuest, unpublished results.
18 A. L. Spek, PLATON, A Multipurpose Crystallographic Tool, Utrecht
University, Utrecht, The Netherlands, 2001.
19 K. E. Maly, E. Gagnon, T. Maris and J. D. Wuest, J. Am. Chem. Soc.,
2007, 129, 4306–4322; J.-H. Fournier, T. Maris and J. D. Wuest, J. Org.
Chem., 2004, 69, 1762–1775; P. Brunet, M. Simard and J. D. Wuest,
J. Am. Chem. Soc., 1997, 119, 2737–2738.
20 K. E. Maly, C. Dauphin and J. D. Wuest, J. Mater. Chem., 2006, 16,
4695–4700.
21 A. Miura, P. Jonkheijm, S. De Feyter, A. P. H. J. Schenning,
E. W. Meijer and F. C. De Schryver, Small, 2005, 1, 131–137;
P. Jonkheijm, A. Miura, M. Zdanowska, F. J. M. Hoeben, S. De
Feyter, A. P. H. J. Schenning, F. C. De Schryver and E. W. Meijer,
Angew. Chem., Int. Ed., 2004, 43, 74–78.
We are now investigating the enantioselective inclusion of chiral
guests in networks built from metallotecton L-1e and related
dipyrrinato complexes, and we are exploring the utility of these
compounds as ligands for the construction of chiral metal–organic
frameworks.
We gratefully acknowledge the expert assistance of Dr Thierry
Maris in X-ray crystallography, and we thank Prof. Andreea
Schmitzer for the use of a CD spectrometer. We are grateful to the
Marsden Fund of the Royal Society of New Zealand, the Natural
Sciences and Engineering Research Council of Canada, the
´
Ministe`re de l’Education du Que´bec, the Canada Foundation for
Innovation, and the Canada Research Chairs Program for
financial support.
22 C. Thalacker and F. Wu¨rthner, Adv. Funct. Mater., 2002, 12, 209–218.
3168 | Chem. Commun., 2007, 3166–3168
This journal is ß The Royal Society of Chemistry 2007