Extremely high activity catalysts for the Suzuki coupling of aryl chlorides: The importance of catalyst longevity

Article abstract of DOI:10.1039/b209490h

The incorporation of a new π-acidic, orthometallated phosphite ligand into a palladium pre-catalyst gives considerably enhanced catalyst longevity and thus extremely high activity in the Suzuki coupling of deactivated, activated and sterically hindered ar

Full text of DOI:10.1039/b209490h

Extremely high activity catalysts for the Suzuki coupling of aryl
chlorides: the importance of catalyst longevity
Robin B. Bedford,* Samantha L. Hazelwood (née Welch) and Michael E. Limmert
School of Chemistry, University of Exeter, Exeter, UK EX4 4QD. E-mail: r.bedford@ex.ac.uk
Received (in Cambridge, UK) 26th September 2002, Accepted 1st October 2002
First published as an Advance Article on the web 15th October 2002
The incorporation of a new p-acidic, orthometallated
phosphite ligand into a palladium pre-catalyst gives con-
siderably enhanced catalyst longevity and thus extremely
high activity in the Suzuki coupling of deactivated, activated
and sterically hindered aryl chloride substrates.
The coupling of aryl halides with aryl boronic acids, the Suzuki
reaction (Scheme 1), is one of the most powerful and versatile
methods for the synthesis of biaryls.1 There has recently been a
considerable interest in the development of new catalysts that
can couple aryl chlorides due to the lower cost and greater
availability of these substrates compared with their bromide or
Scheme 2 Reagents and conditions: (i) NEt3, toluene, D, 18 h; (ii)
[PdCl2(NCMe)2], toluene, D, 18 h.
iodide counterparts.2 Unfortunately the comparatively high
CUCl bond strength makes aryl chlorides difficult to activate.3
Consequently most catalysts that are able to catalyse aryl
Despite the fact that complex 6 could, in principle, exist in
both cis and trans forms, in practice essentially only one isomer
exists in solution, at least in chloroform, as evidenced by 31P
NMR spectroscopy which shows a very large peak at d 124.4
with only a very small peak at d 123.5 corresponding to the
second isomer. This observation is also borne out by the 1H and
13C NMR spectra which again show essentially only one isomer
present. This is in contrast with complex 3 which shows two
peaks at d 119.2 and 118.7 in the 31P NMR spectrum in CDCl3
in an approximately 2+1 ratio.
We have previously shown that the best conditions for the
coupling of the challenging, electronically deactivated (electron
rich) substrate 4-chloroanisole with aryl boronic acids are when
Cs2CO3 is used as base and 1,4-dioxane is employed as the
solvent and so these conditions were used for the initial studies.
Fig. 1 shows a plot of conversion against time in the Suzuki
coupling of 4-chloroanisole with phenylboronic acid catalysed
by a range of catalysts formed in situ from tricyclohex-
ylphosphine (one equiv. per Pd) and four differing pallada-
cycles.
chloride coupling reactions still need to be used in relatively
high loadings. Therefore the advantages associated with the use
of aryl chlorides may be negated by the high cost of the catalyst
systems.
Scheme 1 The Suzuki biaryl coupling reaction.
We have recently found that the PCy3-containing complexes
1 and 2, the latter of which is typically formed in situ from 3 and
PCy3, show good and excellent activity respectively in the
Suzuki coupling of aryl chlorides.2a,d It was argued that the high
activity obtained with the phosphite-containing catalyst is not
necessarily due to any spectacular increase in rate, but rather to
greater catalyst longevity which is conferred by the p-acidic
nature of the phosphite ligand. Thus the catalyst was still active
after one day, while the catalyst 1 shows no further activity after
2 h. If this explanation is correct, then it is reasonable to expect
that increasing the p-acidity of the supporting ligand should
lead to even greater longevity and therefore even higher total
turn-over numbers (TONs, mol product/mol catalyst) should be
achieved at low catalyst loadings. This indeed proves to be the
case and the results of this study are reported below.
In order to further increase the p-acidity of the orthopalla-
dated ligand we decided to prepare a pre-catalyst related to 3,
but in which the two non-orthometallated aryloxide substituents
were replaced by a single salicylate residue. In order to do this
we first synthesised the new ligand 4 by reaction of commer-
cially available 2-chloro-4H-1,3,2-benzodioxaphosphorin-
4-one, 5, with 2,4-di-tert-butylphenol in the presence of
triethylamine in toluene at reflux temperature for 18 h (Scheme
2). Reaction of the ligand 4 with [PdCl2(NCMe)2] in toluene at
reflux temperature for 18 h leads to the formation of the new
dimeric complex 6 in good yield.
Fig. 1 Suzuki coupling of 4-chloroanisole with phenylboronic acid.
Reagents and conditions: MeOC6H4-4-Cl (10.0 mmol), PhB(OH)2 (15.0
mmol), Cs2CO3 (20.0 mmol), catalyst (0.001 mol% Pd), 1,4-dioxane (30
mL), hexadecane (internal standard, 0.204 mmol). Conversion to 4-methox-
ybiphenyl determined by GC.
2610
CHEM. COMMUN., 2002, 2610–2611
This journal is © The Royal Society of Chemistry 2002

Table 1 Suzuki coupling of aryl chloride substrates catalysed by complex
6/PCy3 2+1 mixtures.a
As can be seen the catalyst formed in situ from 6 and PCy3
shows considerably enhanced longevity compared with the
analogous catalyst formed from 3/PCy3. Consequently, by 30 h
6/PCy3 gives higher conversion and the catalyst remains active
even after 54 h. In order to verify that the increase in longevity
is indeed a function of the p-acidity of the orthometallated co-
ligand, the reactions catalysed by species formed in situ from
PCy3 and the palladacyles 7 and 8 were also studied. These
complexes have been shown previously to give excellent
activity in the coupling of aryl bromide substrates.4 It is
apparent that the catalyst longevity and hence overall perform-
ance falls in the order 6 > 3 > 7 > 8. This trend is in line with
decreasing p-acidity of the orthopalladated ligands. By compar-
ison when no p-acidic co-ligand is used and the complex 1 is
employed as a catalyst, then it has been found previously that no
activity is observed after the first 2 h.2a These observations lend
substantial weight to our previous supposition that the increase
in longevity is due to a stabilisation of a palladium(0) ‘resting
state’,2a since the more p-acidic the ligand, the better it would
be expected to coordinate to such zerovalent species. Given that
the rate-determining step is almost certainly oxidative addition
of the aryl chloride to a Pd(0) active catalyst, then it is
reasonable to suppose that the active catalyst spends most of its
time ‘resting’ in the zerovalent state. Without the presence of
additional stabilising ligands, such resting state species would
be expected to be highly susceptible to catalyst decomposition
by aggregation and subsequent precipitation of bulk metal.
Pd
loading
TON (mol
Time/ Conv.b product/
Entry Aryl chloride
(mol%) Base
h
(%)
mol Pd)
1
2
3
4
0.001
0.001
0.001
0.001
K2CO3
K3PO4
KF
KF/K3PO4 24
(1:1)
24
24
24
4
19.5
20
7
4,000
19,500
20,000
7,000
5
6
7
0.0005 Cs2CO3
0.0005 Cs2CO3
24
48
48
43
64
90
86,000
128,000
90,000
0.001
Cs2CO3
8
9
0.00005 Cs2CO3
0.00005 Cs2CO3
24
48
65
84
1,300,000
1,860,000
10
11
0.00005 Cs2CO3
0.00005 Cs2CO3
24
48
98
100
1,960,000
2,000,000
12
13
0.00005 Cs2CO3
0.00005 Cs2CO3
24
48
85
97
1,700,000
1,940,000
14
15
0.0005 Cs2CO3
0.0005 Cs2CO3
24
48
62
84
124,000
168,000
16
17
0.0005 Cs2CO3
0.0005 Cs2CO3
24
48
71
87
142,500
174,000
18
19
0.0005 Cs2CO3
0.0005 Cs2CO3
24
48
41
58
82,000
116,000
a Reaction conditions: ArCl (1.00 mmol), PhB(OH)2 (1.50 mmol), base
(2.00 mmol), 1,4-dioxane (5 mL), 100 °C, N2. b Conversion to coupled
product, determined by GC (hexadecane standard).
We next studied the performance of the catalyst formed in
situ from complex 6 and PCy3 (1 equiv. per Pd) in a range of
aryl chloride Suzuki couplings and these results are summarised
in Table 1.
In summary a simple catalyst precursor made by the
orthopalladation of a phosphite ligand containing a salicylate
residue shows by far the best activity yet reported in the Suzuki
coupling of deactivated, activated and sterically hindered aryl
chlorides. The basis of this performance is the high stability and
longevity of the active catalyst, which in turn is a function of the
p-acidic nature of the co-ligand.
Good results are obtained when the cheaper bases K3PO4 or
KF are employed, however the use of Cs2CO3 leads to
astonishingly high turn-over numbers. Thus TONs of up to
128,000 are seen for the electronically deactivated substrate
4-chloroanisole. By contrast, to the best of our knowledge, the
highest TON reported previously for this reaction was ca.
48,000 using 3/PCy3 mixtures for the same period of time and
under the same conditions.2a High conversions are seen with
this substrate at 0.001 mol% Pd catalyst loading. Despite the
fact that 4-chlorotoluene is considered to be electronically
deactivated, it was still coupled with TONs of up to 1.86
million. When the activated substrates 4-chloroacetophenone or
4-chloronitrobenzene are employed, then essentially quantita-
tive conversions are seen at one two-millionth catalyst loadings.
Very high TONs are also seen for the sterically hindered
substrates 2-chloroanisole, 2-chorotoluene and 2-chloro-m-
xylene. By contrast, to the best of our knowledge, the previous
highest TON observed in the coupling of 2-chlorotoluene with
phenylboronic acid was 100,000, again with a complex 3/PCy3
mixture acting as catalyst.2a
We thank the EPSRC for funding (studentship for S. L. H.
and PDRF for M. E. L.) and Johnson Matthey for funding and
the loan of palladium salts.
Notes and references
1 For recent reviews, see: (a) N. Miyaura and A. Suzuki, Chem. Rev., 1995,
95, 2457; (b) S. P. Stanforth, Tetrahedron, 1998, 54, 263; (c) A. Suzuki,
J. Organomet. Chem., 1999, 576, 147.
2 Recent examples of Suzuki coupling reactions with aryl chlorides: (a) R.
B. Bedford, C. S. J. Cazin and S. L. Hazelwood, Angew. Chem. Int. Ed.,
2002, in press (b) L. Botella and C. Nájera, Angew. Chem., Int. Ed., 2002,
41, 179; (c) S.-Y. Liu, M. J. Choi and G. C. Fu, Chem. Commun., 2001,
2408; (d) R. B. Bedford and C. S. J. Cazin, Chem. Commun., 2001, 1540;
(e) M. R. Netherton and G. C. Fu, Org. Lett., 2001, 3, 4295; (f) A. Zapf,
A. Ehrentraut and M. Beller, Angew. Chem., Int. Ed., 2000, 39, 4153; (g)
M. G. Andreu, A. Zapf and M. Beller, Chem. Commun., 2000, 2475; (h)
D. W. Old, J. P. Wolfe and S. L. Buchwald, J. Am. Chem. Soc., 1998, 120,
9722; (i) J. P. Wolfe, R. A. Singer, B. H. Yang and S. L. Buchwald, J. Am.
Chem. Soc., 1999, 121, 9550; (j) X. Bei, H. W. Turner, W. H. Weinberg
and A. S. Guram, J. Org. Chem., 1999, 64, 6797; (k) C. Zhang, J. Huang,
M. L. Trudell and S. P. Nolan, J. Org. Chem., 1999, 64, 3804; (l) A. F.
Littke and G. C. Fu, Angew. Chem., Int. Ed., 1998, 37, 3387.
3 For a discussion, see: V. V. Grushin and H. Alper, Chem. Rev., 1994, 94,
1047.
It can be seen that for all the reactions excellent TONs are
obtained after 24 h, however it is also apparent that the catalyst
is still active after this time and in all cases reactions run for 48
h lead to greater conversions. This highlights the importance of
catalyst longevity and has major implications for industrial
processes where catalyst stability is often of paramount
importance.
4 R. B. Bedford and S. L. Welch, Chem. Commun., 2001, 129.
CHEM. COMMUN., 2002, 2610–2611
2611