COMMUNICATION
www.rsc.org/chemcomm | ChemComm
A tris-pyrazolylborate ligand with hemilabile O-donor groups.
Examples of g , g , g and bridging modes of bonding to Li , Na , K ,
3
5
6
+
+
+
+
2+
Tl and Ca ions{
Malcolm H. Chisholm,* Judith C. Gallucci and G u¨ l s¸ ah Yaman
Received (in Berkeley, CA, USA) 14th December 2005, Accepted 21st February 2006
First published as an Advance Article on the web 21st March 2006
DOI: 10.1039/b517640a
A tris-pyrazolylborate ligand bearing ether appendages is
remarkably flexible responsiveness of this new ligand towards
shown to be a potential hemilabile ligand based on NMR
3
the demands of the metal. We shall subsequently describe the use
of this ligand in metal mediated catalysis.
5
6
studies and structural characterization of its g , g , g , and
+
+
+
+
The pyrazolyl group was first synthesized by the stepwise series
of reactions shown in Scheme 1. Then, as in the synthesis of the
tris-3-tert-butylpyrazolylborate ligand, the pyrazolyl was allowed
m-binding modes in coordination with Li , Na , K , Tl , and
2+
Ca ions.
The development of single-site catalysts is one of the major success
stories in the field of contemporary inorganic coordination
chemistry. For any given reaction, the development process rests
upon the delicate interplay between the metal ions’ coordination
appetite and the ligands’ ability to satisfy this hunger. If the metal
ion is completely satisfied, then it will show little affinity for
substrate binding and activation. On the other hand, if the metal
ions’ requirements are too poorly satisfied, the kinetic lability of
that specific ligand–metal combination will lead to deleterious side
reactions such as ligand scrambling and thereby loss of single-site
to react with the group 1 metal borohydride at y180 uC and H
evolution was monitored with time with the intent of satisfying the
2
stoichiometric reaction given in eqn.
1 (see Supporting
Information).{
0
180 C
ꢀ
ꢀ
MBH4z3pz H DCCA MTp z3H2ðgÞ
(
1)
Although the reaction depicted by eqn. 1 can reasonably be
understood to proceed in a stepwise manner, the kinetics of the
reaction profile can lead to a mixture of products. Thus, while the
complexes MTp* where Tp* = tris[3-(2-methoxy-1,1-dimethyl-
ethyl)pyrazolyl]borohydride and M = Li, Na and K, have been
isolated from 1, so have some ‘‘over reaction’’ products. For
2+
2+
activity. Metal ions such as Mg and Ca are particularly labile,
2+
2
as evidenced by their rates of H O exchange in their M aqua ions
5
9
21
1
which are y10 and 10 s , respectively. In order to suppress
ligand scrambling, chelating ligands are desirable, but even the
i
-2,6- Pr
example, in the reaction involving LiBH , when the temperature
4
bulky b-diiminate ligand CH(CMeN-C
6
H
3
2
+
)
2
which has
3
1
2
exceeds 180 uC, the four-coordinate g -N , g -N complex 1 shown
3
enjoyed so much success in combination with Zn ions for ring-
2
opening polymerization of lactides and related cyclic esters. and
3
in Fig. 1 was isolated. As can be seen, the B–H bond has
4
in the copolymerization of 1,2-oxiranes and carbon dioxide, failed
somewhat mysteriously been replaced by a B–O bond and a
3
methyl group has been lost. The g -Tp* ligand is further
2+
2+
to meet the requirements of Mg and Ca . These form dimeric
6
1
+
2
5
2
2
products [(g -L)Mg(m-OR)] or by ligand scrambling (g -L) Ca.
supported by a g -bound pz* ligand. The electrophilic Li ion is
clearly demonstrating its preference for nitrogen ligation, over the
available O-donor of the ether. However, in the LiTp* complex
whose structure has not yet been determined, we can reasonably
assume that one or more ether oxygens are involved in ligation.
The structure of the complex NaTp* 2 is shown in Fig. 2. The
2
3
The more bulky g tris-pyrazolylborate, HB(3-Bu pz)
t
3
coordinates
2+
2+
3
well to Mg and Zn to form discrete g -LMOR complexes, but
2+
for Zn , the coordination of the ligand is too strong, leading to
2+
7
loss of substrate binding. For the Ca ion, however, even this
3
bulky g -ligand is insufficient to secure kinetic persistence, and
+
6
6
after initial rapid reaction, catalytic activity dies. We reasoned
3 3
Na ion is completely encapsulated by the g -N ,O -Tp* ligand
2+
2+
that, for the larger ion Ca and the extremely oxophilic Mg , it
could be useful to employ a tris-pyrazolylborate ligand bearing
O-donor appendages which might stabilize the complex toward
ligand scrambling, yet leave the metal center kinetically accessible
to substrate binding. We describe here the synthesis of one such
+
ligand and its coordination to the M ions of Li, Na, K and Tl,
2+
together with Ca . The results described herein reveal the
Newman and Wolfrom Laboratories, The Ohio State University, 100 W.
18th Avenue, Columbus, OH 43210, USA.
E-mail: chisholm@chemistry.ohio-state.edu; Fax: +1-614-292 0368;
Tel: +1-614-292 7216
{
parative details, H and C{ H}-NMR data and crystallographic data.
Electronic supplementary information (ESI) available: Selected pre-
1 13 1
See DOI: 10.1039/b517640a
Scheme 1 Synthesis of 3-(2-methoxy-1,1-dimethylethyl)pyrazole, pz*H.
This journal is ß The Royal Society of Chemistry 2006
1
872 | Chem. Commun., 2006, 1872–1874