Palladium bis(phosphinite) 'PCP'-pincer complexes and their application as catalysts in the Suzuki reaction

Article abstract of DOI:10.1039/b004793g

Palladium complexes of inexpensive, easily synthesised his(phosphinite) 'PCP'-pincer ligands show good activity in the Suzuki coupling of deactivated and sterically hindered aryl bromides.

Full text of DOI:10.1039/b004793g

View Article Online / Journal Homepage / Table of Contents for this issue
Palladium bis(phosphinite) “PCPÏ-pincer complexes and their
application as catalysts in the Suzuki reaction
Robin B. Bedford,*a Sylvia M. Draper,b P. Noelle Scullyb and Samantha L. Welcha
a School of Chemistry, University of Exeter, Exeter, UK EX4 4QD. E-mail: R.Bedford=ex.ac.uk
b Department of Chemistry, T rinity College Dublin, Dublin 2, Ireland
L e t t e r
Received (in Montpellier, France) 14th June 2000, Accepted 4th August 2000
First published as an Advance Article on the web 15th September 2000
Palladium complexes of inexpensive, easily synthesised
bis(phosphinite) “PCPÏ-pincer ligands show good activity in the
Suzuki coupling of deactivated and sterically hindered aryl bro-
Scheme 1 The Suzuki biaryl coupling reaction
mides.
Palladium catalysed Suzuki coupling reactions (Scheme 1)
generally require fairly high catalyst concentrations, limiting
their attractiveness to large-scale commercial applications.
This is due to technical problems and costs associated with
removal of the spent palladium from the products. Conse-
quently, there has recently been an interest in developing
high-activity catalysts that can be used in very low concentra-
tions. Beller and co-workers reported the use of the pallada-
cyclic complex 1,1 formed by the metallation of tri(2-tolyl)
phosphine, as an efficient catalyst for the Suzuki reaction.2 We
recently investigated the possibility of using related palladated
triarylphosphite complexes and found that the complex 2
shows extremely high activity with both electronically activat-
ed and deactivated aryl bromidesÈconsiderably higher than
the activity reported for 1.3
Since then Fu and coworkers,4 as well as Guram et al.,5
have reported good, and Buchwald et al.6 excellent, activity in
the Suzuki reaction using palladium catalysts with the phos-
phine ligands PBut , 3 and 4, respectively. However, the major
3
drawbacks of these catalyst systems is that the phosphine
ligands are comparatively difficult to make or are rather
expensiveÈconsiderably more expensive than the palladium
precursors.7 We were therefore interested in continuing our
studies on high activity catalysts derived from inexpensive,
easily synthesised ligand sets. Inspired by the observations of
Milstein et al.8 and Shibasaki et al.9 that palladated
bis(phosphine) “PCPÏ-complexes of type 5 and bis(phosphite)
“PCPÏ-complexes of type 6 show very high activity in the
Heck reaction, we wished to see whether related
bis(phosphinite) “PCPÏ-pincer complexes would show high
activity in the Suzuki reaction.
The ligands 7a and 7b are easily synthesised in good yields
by the reaction of the appropriate aromatic diol with chloro-
diphenylphosphine in toluene in the presence of triethylamine
(Scheme 2). The reactions of these ligands with palladium tri-
Ñuoroacetate in THF at room temperature gives the “PCPÏ-
pincer complexes 8a and 8b in good yields. The complexes
show excellent air and moisture stability; for instance 8a
shows no sign of decomposition in aerobic solution in the
presence of water after 10 days.
The crystal structure of one of the catalysts (8a) has been
determined10 and the molecular structure and relevant data
are given in Fig. 1. The geometry of the palladium centre is
distorted square planar and is grossly similar to that of the
complex 6.9 The X-ray analysis conÐrms that palladation of
the pincer ligand has occurred and that the two P-donor
groups are in a distorted trans conÐguration. The tri-
Ñuoroacetate ligand coordinates in
a unidentate, non-
symmetrical fashion, giving greater steric repulsion between
the non-coordinated oxygen and the phenyl residues on P1
than those on P2. This is reÑected in the considerably larger
angle between O3ÈPd1ÈP1 than between O3ÈPd1ÈP2.
To the best of our knowledge no one has reported the
application of palladium “PCPÏ-pincer complexes as catalysts
in the Suzuki reaction. Initially, the complexes were tested for
activity in the coupling of phenylboronic acid with the “easy-
DOI: 10.1039/b004793g
New J. Chem., 2000, 24, 745È747
745
This journal is ( The Royal Society of Chemistry and the Centre National de la Recherche ScientiÐque 2000

View Article Online
electronically deactivated substrate 4-bromoanisole. In both
cases good conversions were observed at one ten-thousandth
catalyst concentration and very high turnover numbers
(TONs) were observed at one hundred-thousandth catalyst
concentration (Table 1). In contrast to the results obtained
with 4-bromoacetophenone, catalyst 8a shows [6 times
higher activity with the more challenging substrate 4-
bromoanisole than the previously reported catalyst 2 under
identical conditions.3 We believe this activity to be the highest
yet reported for this substrate. Particularly interesting is the
fact that catalysts 8a and 8b couple 4-bromoanisole with
nearly the same activity as they show in the coupling of 4-
bromoacetophenone. This may imply that the oxidative addi-
tion of the aryl bromide is not the rate-limiting step in the
catalytic cycle. The catalysts show moderate to good activity
at low concentrations in the coupling of the sterically hin-
dered, electronically deactivated substrates 2-bromotoluene, 2-
bromo-p-xylene and even 2-bromo-m-xylene. By comparison,
Scheme 2 (i) 2 ClPPh , 2.2 Et N, toluene, reÑux. (ii) Pd(TFA) , THF.
2
3
2
to-coupleÏ substrate 4-bromoacetophenone. The results were
encouraging with the catalysts showing good activity at
130 ¡C (Table 1), although not as high as obtained previously
with 2.3
palladium complexes of the ligands 4 and PBut show TONs
3
in the range 82È94 for these substrates, albeit run at lower
temperatures.4b,6
The catalysts also show moderate activity in the coupling of
activated aryl chlorides (Table 1); however, little or no activity
was observed with deactivated chlorides. Comparing entries
14 and 17 to 20 it can be seen that the nature of the base plays
a major role in the efficiency of coupling of the chlorides; this
is an area we will investigate further. Particularly interesting
was the fact that KF proved to be a very poor choice of base,
even when the toluene was replaced with THF. This is in
direct contrast with the Ðnding by Buchwald et al. that this
baseÈsolvent mixture can give very high activity with palla-
dium phosphine systems.6
Next we examined the activity of the complexes with the
Whilst it is difficult to establish any clear trends in catalyst
activity on these preliminary data, on average 8b shows slight-
ly higher activity than 8a in reactions at 0.01 to 0.001 mol%
catalyst concentration, possibly due to the fact that 8b has the
more electron-rich Pd centre. At 0.0001 mol% catalyst con-
centration, 8a seems more e†ective; this may be due to slightly
higher stability and thus longevity of the catalyst.
In summary, we have demonstrated for the Ðrst time that
palladium “PCPÏ-pincer complexes can be used as high activ-
ity catalysts in the Suzuki reaction and that bis(phosphinite)
“PCPÏ-pincer complexes are ideal catalysts for the coupling of
deactivated and sterically hindered aryl bromides as they are
inexpensive and very easily synthesised. We are currently
Fig. 1 The molecular structure of 8a. All H atoms are omitted for
clarity. Selected distances (A) and angles (¡): Pd1ÈC1 1.99(2), Pd1ÈP1
2.287(6), Pd1ÈP2 2.281(6), Pd1ÈO3 2.093(18); C1ÈPd1ÈP1 81.5(6), C1È
Pd1ÈP2 79.3(6), O3ÈPd1ÈP1 104.8(5), O3ÈPd1ÈP2 94.5(5), P1ÈPd1È
P2 160.2(2), C1ÈPd1ÈO3 173.7(8).
Table 1 Suzuki coupling of aryl halides with phenylboronic acid catalysed by palladium “PCPÏ-pincer complexesa
Entry
Aryl halide
Catalyst (mol%)
Base
Conv. (%)b
TONc
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
4-bromoacetophenone
4-bromoacetophenone
4-bromoacetophenone
4-bromoacetophenone
4-bromoanisole
4-bromoanisole
4-bromoanisole
8a(0.001)
8b(0.001)
8a(0.0001)
8b(0.0001)
8a(0.01)
8b(0.01)
8a(0.0001)
8b(0.0001)
8a(0.01)
8a(0.01)
8b(0.01)
8a(0.01)
8b(0.01)
8a(0.01)
8b(0.01)
8b(0.1)
K CO
59
92
19
18
61
72
19
15
87
67
63
29
30
43
40
67
25
0
59 000
92 000
190 000
180 000
6100
7200
190 000
150 000
8700
6700
6300
2900
3000
4300
4000
670
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
K CO
2
K CO
2
K CO
2
K CO
2
K CO
2
K CO
2
4-bromoanisole
2-bromotoluene
K CO
2
K CO
2
2-bromo-p-xylene
2-bromo-p-xylene
2-bromo-m-xylene
2-bromo-m-xylene
4-chloronitrobenzene
4-chloronitrobenzene
4-chloronitrobenzene
4-chloronitrobenzene
4-chloronitrobenzene
4-chloronitrobenzene
4-chloronitrobenzene
K CO
2
K CO
2
K CO
2
K CO
2
K CO
2
K CO
2
K CO
2
8a(0.01)
8a(0.01)
8a(0.01)
8a(0.01)
K PO
2500
0
500
3
4
NaOBut
KF
5
3
KFd
300
a Reaction conditions: 1.0 mmol aryl halide, 1.5 mmol PhB(OH) , 2.0 mmol base, 5 ml toluene, catalyst added in THF solution (0.1 ml), 130 ¡C,
2
18 h. b Determined by GC, based on aryl halide. c DeÐned as mol product per mol catalyst. d Solvent: 5 ml THF, 75 ¡C.
746
New J. Chem., 2000, 24, 745È747

View Article Online
investigating the use of a range of palladium “PCPÏ-pincer
complexes with phosphinite, phosphite and phosphine resi-
dues in order to elucidate which will make the optimum cata-
lyst for a given substrate in the Suzuki and related reactions.
hexane. Anal. calc. for C
H
F O P Pd: C 55.15; H 3.3.
32 23 3 4 2
Found: C 54.85; H 3.55%. NMR (CDCl ): 31P: d 146.2; 1H:
3
d 6.74 (d, 2H, 2J \ 8.0 Hz, 3,5-Hs of resorcinol ring); 7.12 (t,
HH
1H, 3J \ 8.0 Hz, 4-H of resorcinol ring); 7.57 (m, 12H,
HH
PPh ); 7.90 (m, 8H, PPh ).
Complex 8b was obtained as a colourless solid (0.166 g,
78%). Anal. calc. for C F O P Pd: C 55.75; H 3.5.
2
2
Experimental
H
33 25 3 4 2
Found: C 55.1; H 3.3% NMR (CDCl ): 31P: d 145.6; 1H: d
3
2.28 (s, 3H, CH ); 6.57 (s, 2H, 4,6-Hs of orcinol ring); 7.49 (m,
Syntheses
3
12H, PPh ); 7.86 (m, 8H, PPh ).
2
2
Resorcinolbis(diphenyl)phosphinite, 7a. To a solution of res-
orcinol (2.5 g, 23.0 mmol) and chlorodiphenylphosphine (8.4
ml, 46.0 mmol) in toluene (40 ml) was added triethylamine (7.0
ml, 50.0 mmol) dropwise. The resultant mixture was then
heated at reÑux for 18 h. After cooling, the volatiles were
removed in vacuo, the residue was extracted with THF (20 ml)
and the resultant solution was Ðltered through celite. The
celite was washed with THF (2 ] 20 ml), the extracts were
combined and the solvent was removed in vacuo to yield the
title product as a yellowÈorange solid that was not puriÐed
Acknowledgements
This work was supported by Forbairt (now Enterprise,
Ireland; studentship for P. N. S), EPSRC and Johnson
Matthey (studentship for S. L. W) and the Trinity Trust Foun-
dation. We thank Johnson Matthey for the loan of precious
metals and for access to their rapid catalyst-screening
programme.
References
further (8.2 g, 75%). NMR (CDCl ): 31P: d 114.0; 1H: d 6.84
3
1
For a review on the catalytic activity of 1 in Heck-type reactions
see: W. A. Herrmann, V. P. W. Bohm and C.-P. Reisinger, J.
Organomet. Chem., 1999, 576, 23.
(d, 2H, 3J \ 8.0 Hz, 3,5-Hs of resorcinol ring); 7.02 (s, 1H,
HH
1-H of resorcinol ring); 7.14 (t, 1H, 3J \ 8.0 Hz, 4-H of res-
orcinol ring); 7.44 (m, 12H, PPh ); 7.63 (m, 8H, PPh ).
HH
2
3
4
M. Beller, H. Fischer, W. A. Herrmann, K. Ofele and C. Bross-
mer, Angew. Chem., Int. Ed. Engl., 1995, 34, 1848.
D. A. Albisson, R. B. Bedford, S. E. Lawrence and P. N. Scully,
Chem. Commun., 1998, 2095.
(a) A. F. Littke and G. C. Fu, Angew Chem., Int. Ed., 1998, 37,
3387; (b) A. F. Littke, C. Dai and G. C. Fu, J. Am. Chem. Soc.,
2000, 122, 4020.
X. Bei, H. W. Turner, W. H. Weinberg and A. S. Guram, J. Org.
Chem., 1999, 64, 6797.
J. P. Wolfe, R. A. Singer, B. H. Yang and S. L. Buchwald, J. Am.
Chem. Soc., 1999, 121, 9550.
Strem Chemicals Catalog, 1999È2000, no. 18.
M. Oh†, A. Oh†, M. E. van der Boom and D. Milstein, J. Am.
Chem. Soc., 1997, 119, 11687.
F. Miyazaki, K. Yamaguchi and M. Shibasaki, T etrahedron L ett.,
1999, 40, 7379.
2
2
Orcinolbis(diphenyl)phosphinite, 7b. The ligand was
obtained as a yellow oil using a method analogous to that for
7a with 3,5-dihydroxytoluene (orcinol; 2.18 g, 17.6 mmol),
chlorodiphenylphosphine (7.2 ml, 39.4 mmol) and tri-
ethylamine (7.0 ml, 50 mmol) in toluene (40 ml). Yield 8.5 g,
98%. NMR (CDCl ): 31P: d 111.3; 1H: d 2.29 (s, 3H, CH );
5
6
3
3
6.71 (s, 2H, 4,6-H of orcinol ring); 6.84 (s, 1H, 2-H of orcinol
ring); 7.41 (m, 12H, PPh ); 7.60 (m, 8H, PPh ).
2
2
7
8
Complexes 8a and 8b.
A
solution of palladium tri-
Ñuoroacetate (0.100 g, 0.302 mmol) and the appropriate
bis(diphenyl)phosphinite (0.302 mmol) in THF (5.0 ml) was
stirred under nitrogen at room temperature for 2 h. The
solvent was then removed in vacuo. The residue was washed
with ethanol and then recrystallised from dichloromethaneÈ
ethanol. Complex 8a was obtained as a colourless solid (0.175
g, 85%). Crystals suitable for single crystal X-ray analysis were
obtained by layering a solution of 8a in dichloromethane with
9
10 Crystal data for 8a:
C H F O P Pd, M \ 696.84, ortho-
32 23 3 4 2
rhombic, space group P2 2 2 , Z \ 4, a \ 10.469(2),
1 1 1
b \ 15.8692(9), c \ 17.2811(14) A, U \ 2871.0(6) AŽ 3, T \ 293(2)
K, k(Mo-Ka) \ 0.815 mm~1, 2729 independent reÑections mea-
sured, the Ðnal R factor was 0.0863 for 2189 reÑections with I [ 2
p(I). CCDC reference number 440/207. See http://www.rsc.org/
suppdata/nj/b0/b004793g/ for crystallographic Ðles in .cif format.
New J. Chem., 2000, 24, 745È747
747