Creating new binding sites in ligands and metal complexes using the Negishi cross-coupling reaction

Article abstract of DOI:10.1021/ic026043t

The Negishi cross-coupling reaction creates a new binding site in a ruthenium complex with high efficiency as exemplified by the synthesis of a heterodimetallic ruthenium-osmium complex.

Full text of DOI:10.1021/ic026043t

Inorg. Chem. 2003, 42, 5−7
Creating New Binding Sites in Ligands and Metal Complexes Using the
Negishi Cross-Coupling Reaction
,†
Yuan-Qing Fang, Matthew I. J. Polson,† and Garry S. Hanan*
Department of Chemistry, UniVersity of Waterloo, 200 UniVersity AVenue West,
Waterloo, Ontario N2L 3G1, Canada
Received September 18, 2002
The Negishi cross-coupling reaction creates a new binding site in
a ruthenium complex with high efficiency as exemplified by the
synthesis of a heterodimetallic ruthenium−osmium complex.
Advances in inorganic photochemistry over the past two
decades have driven the diversification of model systems
under investigation.1 As interest in energy and electron
transfer studies in mono- and polymetallic complexes grew,
so did the synthetic approaches to obtain these complexes.
Figure 1. Strategy for the synthesis of heterodimetallic complexes: (a)
Homodimetallic complexes are readily synthesized by first
preparing symmetric dinucleating ligands followed by the
direct synthesis of their dimetallic complexes. However, more
elaborate energy and electron transfer studies require the
preparation of the synthetically more challenging heterodi-
metallic complexes.
Most synthetic approaches to heterodimetallic complexes
make use of symmetric dinucleating ligands.2 The mono-
metallic complexes of dinucleating ligands are usually
obtained in moderate yield as a statistical mixture from the
reaction of the ligand with 1 equiv of metal ion, followed
by the introduction of the second metal ion. Protection-
deprotection methodology has also been developed to build
up polymetallic complexes, whereby one metal-binding site
is protected by methylation, followed by metal ion com-
plexation, deprotection, and subsequent heterometal ion
binding.3 More recently, the “chemistry-on-the-complex”
approach, in which classic organic and organometallic
reactions are performed directly on metal complexes, has
the first metal ion binds to a complete chelating site; (b) a new binding site
is created in a reaction on the metal complex; (c) the second metal ion is
then introduced giving a heterodimetallic complex.
shown significant versatility and promise.4 Typically, this
approach simplifies the purification of products, increases
the overall yields, and, in some cases, proves to be the only
method to obtain the desired products.
The “chemistry-on-the-complex” approach is even more
powerful when a new binding site is created in the complex,
which allows further metal ion complexation (Figure 1).5 The
first metal ion binds selectively to a complete chelating site
as opposed to an incomplete binding site (Figure 1a). A
catalytic reaction, in this case the Negishi reaction, creates
a new binding site in the metal complex (Figure 1b). A
second and different metal ion is introduced into the newly
created binding site, thus allowing the straightforward
synthesis of heterodimetallic complexes.
Herein, we describe the use of 2-pyridylzinc bromide to
create a new bidentate coordination site in both ligands and
* To whom correspondence should be addressed. E-mail: garry.hanan@
montreal.ca.
(4) (a) Aspley, C. J.; Gareth Williams, J. A. New J. Chem. 2001, 25, 1136.
(b) Griffiths, P. M.; Loiseau, F.; Puntoriero, F.; Serroni, S.; Campagna,
S. Chem. Commun. 2000, 2297. (c) Fanni, S.; Di Pietro, C.; Serroni,
S.; Campagna, S.; Vos, J. G. Inorg. Chem. Commun. 2000, 3, 42. (d)
Pabst, G. R.; Pfuller, O. C.; Sauer, J. Tetrahedron, 1999, 55, 8045.
(e) Dunne, S. J.; Constable, E. C. Inorg. Chem. Commun. 1998, 1,
167. (f) Patoux, C.; Launay, J.-P.; Beley, M.; Chodorowski-Kimmes,
S.; Collin, J.-P.; James, S.; Sauvage, J.-P. J. Am. Chem. Soc. 1998,
120, 3717. (g) Constable, E. C.; Cargill Thompson, A. M. W.;
Greulich, S. J. Chem. Soc., Chem. Commun. 1993, 1444. (h) Beley,
M.; Collin, J. P.; Louis, R.; Metz, B.; Sauvage, J. P. J. Am. Chem.
Soc. 1991, 113, 8521.
† Present address: De´partement de chimie, Universite´ de Montre´al,
Montre´al, QC, Canada H3T 1J4.
(1) Balzani, V.; Juris, A.; Venturi, M.; Campagna, S.; Denti, G.; Serroni,
S. Chem. ReV. 1996, 96, 759. Balzani, V.; Scandola, F. Supramolecular
Photochemistry; Ellis Horwood: Chichester, U.K., 1991.
(2) Recent work with asymmetric ligands allows heterodimetallic com-
plexes to be formed selectively: Ward, M. D.; Barigelletti, F. Coord.
Chem. ReV. 2001, 216-217, 127. Encinas, S.; Barigelletti, F.;
Barthram, A. M.; Ward, M. D.; Campagna, S. Chem. Commun. 2001,
277.
(3) Denti, G.; Serroni, S.; Campagna, S.; Juris, A.; Ciano, M.; Balzani,
V. Perspect. Coord. Chem. 1992, 153. Denti, G.; Campagna, S.;
Sabatino, L.; Serroni, S.; Ciano, M.; Balzani, V. Inorg. Chem. 1990,
29, 4750.
(5) Johansson, K. O.; Lotoski, J. A.; Tong, C. C.; Hanan, G. S. Chem.
Commun. 2000, 819. Other examples of this methodology: Tomohiro,
Y.; Satake, A.; Kobuke, Y. J. Org. Chem. 2001, 66, 8442.
10.1021/ic026043t CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/13/2002
Inorganic Chemistry, Vol. 42, No. 1, 2003 5

COMMUNICATION
Scheme 1. Synthesis of Polypyridyl Ligands 1a,ba
Chart 1. Homodimetallic Complexes 4 and 5
a The precipitation of the zinc bromide salt of 1a leads to the enhanced
yield (76%) of monobrominated product even with an excess of 2-pyridyl-
zinc bromide.
Scheme 2. Preparation of Heterodimetallic Ruthenium-Osmium
Complex 3aa
Figure 2. Luminescence data for the dimetallic complexes 3 (gray line),
4 (black line), and 5 (dashed line) on excitation at their isoabsorption
wavelengths (500, 490, and 510 nm, respectively).
2-pyridylzinc bromide (0.50 mmol) under mild Negishi cross-
coupling conditions, stirring at room temperature in DMF
with Pd(PPh3)4 as catalyst (20 mol % based on 2a), affords
2b in 97% yield. Thus, the conversion of 2,2′-dibromo-4,4′-
bipyridine to 2b occurs in 69% overall yield. Because of
the hydration of 2a, excess organozinc reagent (5 equiv) is
required for the Negishi reaction to proceed. The crude
product is purified by sequestering excess zinc halides formed
in the transmetalation step with EDTA, followed by column
chromatography on silica gel. These cross-coupling condi-
tions are considerably milder than those required for cross-
coupling stannyl or borane reagents with metal complexes.7
The newly formed ligand 1b in complex 2b is not isolated
in its metal-free state. Indeed, the reaction of the metal-free
ligand 1b with Ru(bpy)2Cl2 would require difficult purifica-
tion steps as both bpy sites could react with Ru(bpy)2Cl2.
The newly created bpy site in 2b is then allowed to react
with cis-(bpy)2OsCl2 in ethylene glycol to yield the het-
erodimetallic complex 3 in 43% yield (Scheme 2c). The
lower yield in the last step is typical for the synthesis of
osmium bpy complexes.
To interpret the emission spectrum of 3, the homodime-
tallic complexes of ruthenium (4) and osmium (5) were
required (Chart 1). Allowing excess Ru(bpy)2Cl2 and Os(bpy)2-
Cl2 to reflux in ethylene glycol with 1b afforded 48 and 5,
respectively.
Preliminary luminescence studies of heterodimetallic
complex 3 indicate that energy transfer is occurring from
the Ru(bpy)3-like site to the Os(bpy)3-like site (Figure 2).
Upon irradiation of monoruthenium complex 2b and diru-
thenium complex 4 at an isoabsorption point (490 nm), their
a Reaction conditions: (a) Ru(bpy)2Cl2, AgNO3, EtOH, 93% yield; (b)
2-pyridylzinc bromide, Pd(PPh3)4, DMF, r.t., 97% yield; (c) Os(bpy)2Cl2,
HOCH2CH2OH, 43% yield.
metal complexes by way of the Pd-catalyzed Negishi
reaction. As an example of a heterodimetallic complex, a
mixed Ru-Os dimetallic complex has been made in high
overall yield using this methodology, and its physical
properties have been examined.
The Negishi reaction affords a facile method to prepare
new binding sites in ligands. 2,2′-Dibromo-4,4′-bipyridine6
(0.75 mmol) readily reacts with 2-pyridylzinc bromide (1.5
mmol) at r.t. (room temperature) in THF with Pd(PPh3)4
(0.025 mmol, 3%) [Pd(PPh3)4 ) tetrakis(triphenylphosphine)-
palladium(0)] as catalyst to afford a mixture of terpyridine
1a and quaterpyridine 1b (Scheme 1). It is noteworthy that,
even with an excess of 2-pyridylzinc bromide, 1a forms
selectively because of its precipitation from the reaction
mixture, which disfavors the formation of dinucleating ligand
1b. The monohalogenated ligand is obtained in high yield
by avoiding reactions that generate statistical mixtures of
products.
Allowing ligand 1a to react with an equimolar amount of
Ru(bpy)2Cl2 (bpy ) 2,2′-bipyridine) in EtOH at reflux in
the presence of AgNO3 gives monometallic complex 2a in
93% yield (Scheme 2a). The Ru(bpy)22+ moiety binds much
more effectively to the bpy-like site in 1a as opposed to the
monodentate 2-bromopyridyl site because of the chelate
effect. In addition, the bromine atom hinders the pyridyl
nitrogen’s ability to bind to the metal ion.
The 2-bromopyridyl site in 2a is readily converted into a
chelating bpy-like site using the Negishi reaction (Scheme
2b). The reaction of 2a (0.10 mmol) with an excess of
(7) Stille and Suzuki reactions on metal complexes have previously been
described; however, no new binding sites were created. See refs 4a,
4d-f.
(8) Downard, J.; Honey, G. E.; Phillips, L. F.; Steel, P. J. Inorg. Chem.
1991, 30, 2259.
(6) Kelly, T. R.; Lee, Y.-J.; Mears, R. J. J. Org. Chem. 1997, 62, 2774.
6 Inorganic Chemistry, Vol. 42, No. 1, 2003

COMMUNICATION
homodimetallic complexes 4 (+1.31 V) and 5 (+0.87 V)
have a separation between the cathodic and anodic peak of
98 and 92 mV, respectively. The peak separation in 4 has
previously been reported as an indication of weak com-
munication between the metal centers.8 Likewise, we ascribe
the oxidation in 5 to two coincidental one-electron processes.
In heterodimetallic complex 3, the oxidation of the osmium
metal center (+0.87 V, (74 mV)) does not alter the oxidation
potential of the ruthenium metal center (+1.31 V, (70 mV)),
indicating weak metal-metal interaction in these complexes.
We have shown that organozinc reagents undergo the
Negishi reaction with metal complexes in high yield. The
lower toxicity of organozinc reagents compared to stannyl
reagents, the mild reaction conditions, and the commercial
availability of a wide range of heterocyclic zinc reagents
make this the method of choice for creating new binding
sites in metal complexes. The coordination of different metal
ions in the newly created binding site can lead to heterodi-
metallic complexes, one of which exhibits Ru-to-Os energy
transfer. The extension of this work to the synthesis of
polymetallic complexes is currently under investigation.
Figure 3. Oxidation and reduction potentials for homodimetallic complexes
4 (top) and 5 (middle) and heterodimetallic complex 3 (bottom). Values
reported in the text are referenced to internal ferrocene (Fc).
emission bands are observed at 667 nm and are of equal
intensity. Diosmium complex 5 exhibits a single emission
at 790 nm on irradiation at either 510 nm (1MLCT absorp-
tion) or 620 nm (3MLCT absorption). Irradiation of complex
3 at 500 nm shows a strong quenching of the ruthenium
emission with an emission band occurring at 788 nm instead.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada, the Research
Corporation, and the Universite´ de Montre´al for financial
support.
3
This implies that partial energy transfer from the MLCT
3
excited-state of ruthenium to the MLCT excited-state of
Supporting Information Available: Experimental procedures
for ligands 1a,b and complexes 2a,b, 3 and 5. This material is
available free of charge via the Internet at http://pubs.acs.org.
osmium occurs in 3.
The electrochemistry of complexes 3, 4, and 5 indicates
that limited communication exists between the metal centers
(Figure 3). The oxidation potentials of the metal centers in
IC026043T
Inorganic Chemistry, Vol. 42, No. 1, 2003 7