Palladacyclic phosphinite complexes as extremely high activity catalysts in the
Suzuki reaction
Robin B. Bedford* and Samantha L. Welch
School of Chemistry, University of Exeter, Exeter, UK EX4 4QD. E-mail: R.Bedford@ex.ac.uk
Received (in Cambridge, UK) 18th October 2000, Accepted 16th November 2000
First published as an Advance Article on the web
Phosphinite based palladacycles show extremely high activ-
ity in the Suzuki coupling of both sterically hindered and
electronically deactivated aryl bromides, especially in the
presence of one equivalent of free ligand.
The application of the complexes 4a–c as catalysts in the
Suzuki reaction was studied and the results are summarised in
Table 1. The complexes 4a and b prove to be extremely active
catalysts for both the ‘easy to couple’ substrate 4-bromoaceto-
phenone and the more challenging, electronically deactivated
substrate 4-bromoanisole. To the best of our knowledge, the
activity shown by the catalyst mixture of 4b and one equivalent
of added ligand 3b (entry 13) is the highest reported for any
Suzuki reaction—over five times higher than for the analogous
reaction catalysed by palladium complexes of the phosphine
ligands 5.6 By contrast, the previous highest reported activity
with a palladacyclic catalyst in this reaction was with complex
1 which gave a turn-over number (TON) of 1 million.2 More
importantly with the electronically deactivated substrate 4-bro-
moanisole the catalyst systems reported here show up to 45
times higher activity than the previous best catalysts—the
related bis(phosphinite) pincer complexes 2.3 Similarly the
activities seen with the sterically hindered, electronically
deactivated substrates 2-bromotoluene and 2-bromo-m-xylene
are over an order of magnitude higher than any catalyst systems
in the literature.3
There has recently been considerable interest in the synthesis of
new, high activity palladium-based catalysts that can be used in
low concentration in the Suzuki reaction (Scheme 1) since such
catalysts have the potential to be used in industrial systems. In
particular, palladacyclic catalysts in which a ligand coordinates
to the metal centre through both a donor atom and metallated
carbon have shown considerable promise. Beller and co-
workers demonstrated that palladated phosphine complexes
show good activity1 whilst we have shown that the palladated
triaryl phosphite complex 1 and the bis(phosphinite) PCP-
Complex 4b generally shows higher activity than 4a,
presumably due to the higher electron density that the
phosphinite ligand confers on the palladium centre—this
increase in electron density facilitates oxidative addition of the
aryl bromide. The observation that 4a seems to show very little
difference in activity in the coupling of phenylboronic acid with
either 4-bromoacetophenone or 4-bromoanisole indicates that
the oxidative addition of the aryl bromide may not be the rate
determining step in the catalytic cycle. Comparing entries 3 and
7 it can be seen that decreasing the steric profile of the
metallated ring has a deleterious effect on the rate of reaction,
however 4c still shows substantially higher activity than its
previously reported bis(phosphinite) PCP pincer counterpart, 2
(R = H).3 The importance of the metallation in the pre-catalysts
is demonstrated by comparing entries 3, 8 and 9. The addition of
a second phosphinite ligand to the metallated complex gives an
increase in activity, by contrast the formation of bis(phosphi-
nite) complexes in situ leads to substantially reduced reactivity.
In other words addition of a second phosphinite is only
beneficial if there is a metallated ring in the complex, otherwise
it is deleterious. This demonstrates that the orthometallation of
the pre-catalyst is extremely important; the possible role it plays
is addressed below.
By contrast with the extraordinary results obtained with the
deactivated aryl bromide 4-bromoanisole, complex 4b proved
to be a very poor catalyst when 4-chloroanisole was used as a
substrate, indeed substantial deposition of palladium metal was
observed almost immediately upon heating. Interestingly, we
found that rapid decomposition of all three complexes 4a–c
occurs even if the 4-chloroanisole is left out of the reaction
mixture. To gain evidence for the mechanism of decomposition
a reaction was performed between 4c, PhB(OH)2 (4 molar
equiv./Pd) and K2CO3 (5 molar equiv./Pd) in toluene at reflux
temperature. The reaction was performed over 24 h, although
decomposition started within seconds of the reaction reaching
about 50 °C. After this time the supernatant liquid contained a
mixture of compounds which included some unreacted 4c. A
GC-MS spectrum of the mixture revealed the presence of some
pincer complexes 2 show excellent activity.2,3 High activity is
not limited to metallated phosphorus donor systems—Milstein
and co-workers have shown that a palladated imine complex
shows excellent activity,4 whilst Zim et al. have shown that
palladated thioether complexes can also be used.5
Scheme 1 The Suzuki biaryl coupling reaction.
To the best of our knowledge, the use of P,C-bidentate
phosphinite palladacycles as catalysts in the Suzuki reaction has
not been investigated. We report here the synthesis of such
complexes, the remarkable activity they show in the Suzuki
reaction and present preliminary evidence that suggests that the
active catalysts are likely to be low coordinate palladium(0)
species.
The reactions of the ligands 3a and b with palladium chloride
in toluene at reflux temperature gives the complexes 4a and b in
good yields. Complex 4c is best prepared by the reaction of
[PdCl2(NCPh)2] with 3c in THF at reflux temperature. All the
complexes 7 are obtained as a mixture of cis and trrans
isomers.2
Scheme 2 (i) Et3N, toluene, reflux temperature, 18–20 h; (ii) For 4a and b:
PdCl2, toluene, reflux temperature 18–20 h; (iii) For 4c: [PdCl2(NCPh)2],
THF, reflux temperature, 20 h.
DOI: 10.1039/b008470k
Chem. Commun., 2001, 129–130
This journal is © The Royal Society of Chemistry 2001
129
Bedford
