Simple rhodium-chlorophosphine pre-catalysts for the ortho-arylation of phenols

Article abstract of DOI:10.1039/b718128k

Simple chlorodiisopropylphosphine adducts of rhodium, either pre-formed or formed in situ, prove to be highly effective catalysts for the ortho-arylation of phenols. The Royal Society of Chemistry.

Full text of DOI:10.1039/b718128k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Simple rhodium–chlorophosphine pre-catalysts for the ortho-arylation of
phenolsw
Robin B. Bedford,*a Michael Betham,a Andrew J. M. Caffyn,*b
Jonathan P. H. Charmant,a Lesley C. Lewis-Alleyne,ab Philip D. Long,a
ac
Dorian Polo-Ceron and Sanjiv Prashar*c
´
Received (in Cambridge, UK) 23rd November 2007, Accepted 14th December 2007
First published as an Advance Article on the web 10th January 2008
DOI: 10.1039/b718128k
Simple chlorodiisopropylphosphine adducts of rhodium, either
pre-formed or formed in situ, prove to be highly effective
catalysts for the ortho-arylation of phenols.
produced in situ from commercially available materials, are
indeed at least as active as phosphinite systems, making them
the catalysts of choice for these reactions.
In the first instance, we examined the use of commercially
i
available Pr2PCl as the ligand since we previously found the
Aromatic functionalisation by catalytic C–H activation is
rapidly establishing itself as a viable alternative to classical
cross-coupling chemistry, not least because it obviates the need
to introduce an organometallic leaving group onto either of
the coupling partners.1 In the absence of an organometallic
fragment, two general C–H activation protocols can be en-
visaged; the first is to use oxidative coupling (Scheme 1, path
a) and the second relies on the oxidative addition of an organic
halide (path b). The first process is obviously more atom-
economical than the second,2,3 although this is dependent on
the complexity of the oxidising agent, but can lack the
selectivity inherent in the second reaction type.
corresponding phosphinites iPr2P(OAr) to be particularly
useful for the ortho-arylation of a range of phenols. Table 1
summarises the results obtained in the coupling of 2,4-di-tert-
butylphenol with 4-bromoanisole. As can be seen, good
activity is observed at 5 mol% Rh with iPr2PCl, but changing
to Ph2PCl is deleterious. Similarly, the other dialkylchloropho-
sphines employed both perform poorly. It is obvious that
nucleophilic attack of the phenol at a coordinated ClPtBu2 is
severely impeded by the bulk of the alkyl groups, however
comparison of simple models of ‘ClPiPr2–Rh’ and
‘ClPCy2–Rh’ fragments also show the approach of the nucleo-
phile to the latter to be sterically disfavoured.w It seems that
the iPr group holds a privileged position with sterically
encumbered phenol substrates—it is sufficiently bulky to
accelerate orthometallation by forcing the phenolic residue
over the metal centre,6 and yet provides an access route for the
incoming phenolate during transesterification.
An example of the second methodology is provided by the
catalytic ortho-arylation of phenols (Scheme 2).4,5 In this
instance the phosphinite ligand is required as a co-catalyst;
this is the species that undergoes C–H activation via ortho-
metallation (Scheme 3), in effect acting as an intramolecular
‘tether’. The phosphinite ligand then undergoes ‘transesterifi-
cation’ with the starting phenol, in the presence of base, which
liberates the 2-arylated phenol product and regenerates the
starting phosphinite.
Changing the base to NaOtBu or K3PO4 leads to reduced
activity. Changing the rhodium precursor to Wilkinson’s
catalyst has little effect on performance, whereas
[{RhCl(ethene)2}2] behaves highly capriciously with poor
Unfortunately, the use of phosphinite co-catalysts presents
some obvious problems. Firstly, these ligands are not generally
commercially available and thus need to be pre-synthesised;
secondly, the phosphinite used should ideally contain the same
phenoxide residue as the substrate to avoid contamination of
the product with coupled by-products. These issues can limit
the appeal of the methodology in synthetic applications. One
way around these problems would be to use rhodium systems
with commercially available chlorophosphine complexes as
pre-catalysts—this should lead to the formation of the desired
phosphinite ligands in situ. We now report that simple
rhodium chlorophosphine complexes, either pre-formed or
Scheme 1 Aromatic C–H functionalisation.
a School of Chemistry, University of Bristol, Cantock’s Close, Bristol,
UK BS8 1TS. E-mail: r.bedford@bristol.ac.uk
Scheme 2 The catalytic ortho-arylation of phenols.
b Department of Chemistry, The University of the West Indies, St
Augustine, Trinidad and Tobago
c
´
´
´
Departamento de Quımica Inorganica y Analıtica, Universidad Rey
Juan Carlos, Tulipan; s/n, 28993Mostoles, Madrid, Spain
´
´
w Electronic supplementary information (ESI) available: Experimental
details, models of Rh–PClR2 fragments. See DOI: 10.1039/b718128k
Scheme 3 Orthometallation of the phosphinite ligand.
ꢀc
This journal is The Royal Society of Chemistry 2008
990 | Chem. Commun., 2008, 990–992

View Article Online
Table 1 Catalyst optimisationa
Rh loading
(mol%)
Yield
(%)b
Entry
Rh precursor
ClPR2
1
2
3
4
5
6
7
8
[{RhCl(COD)}2]
5
5
5
5
5
5
5
5
ClPiPr2
ClPPh2
ClPCy2
ClPtBu2
ClPiPr2
ClPiPr2
ClPiPr2
ClPiPr2
ClPiPr2
ClPiPr2
ClPiPr2
92
27
19
0
Scheme 4 Conditions: (i) ClPiPr2, CH2Cl2, rt, 2 h; (ii) HOC6H3-
2,4-tBu2, NaOtBu, toluene, 80 1C, 1 h; (iii) NaOtBu, toluene, 80 1C,
1 h; (iv) PiPr2(OC6H3-2,4-tBu2), CH2Cl2, rt, 1 h.
76c
66d
85
55e
90
80
23
[RhCl(PPh3)3]
[{RhCl(C2H4)2}2]
[{RhCl(COD)}2]
9
2.5
1
0.5
chlorophosphine, implying that transesterification is achiev-
able but that the rate of C–H activation is not enhanced. In all
three cases, the only product observed is the di-arylated
phenol.7 A synthetically useful alternative, which yields the
mono-arylated 6-H 2-arylphenols, is to exploit the 2-tert-
butylphenol substrates which, as shown above, undergo highly
selective reactions in good yields and then remove the tert-
butyl group, a process which can easily be achieved using mild
conditions.8
10
11
a
Conditions: 2,4-tBu2C6H3OH (0.5 mmol), 4-BrC6H4OMe (0.6
mmol), catalyst loading 5 mol%, ClPR2 : Rh ¼ 2 : 1, base (0.85
b
mmol), toluene (5 ml), 18 h, reflux. Spectroscopic yield determined
by 1H NMR spectroscopy (1,3,5-(MeO)3C6H3 internal standard),
c
d
average of 2 runs. NaOtBu used as base. K3PO4 used as base.
e
Average of 9 runs, ꢁ38%.
With regards to the mechanism, it seems likely that the
chlorophosphine ligand, either free or coordinated to the
rhodium centre, reacts with the phenolic substrates in the
presence of base to generate phosphinite complexes in situ.
In order to test this we prepared the rhodium chlorophosphine
complex 2 (Scheme 4).w The 31P{1H} NMR spectrum of 2
shows a doublet at 173.2 ppm with a 1JPRh of 175 Hz, similar
to the data reported for a related rhodium complex of a bulky
dialkylchlorophosphine.9 Complex 2 is air-stable in the solid
state, showing no sign of decomposition after several days,
while a CDCl3 solution stored under air shows less than 5%
decomposition within three days.10 The crystal structure of 2
was determined and the molecule is shown in Fig. 1.z To the
best of our knowledge this is the first structure of a chloro-
diisopropylphosphine transition metal complex to be deter-
mined.
reproducibility across several runs. [{RhCl(COD)}2] was se-
lected for the remainder of the studies as it is commercially
available, easy to handle and, with fewer, less massive ligands,
offers lower potential for product contamination than Wilk-
inson’s catalyst. We were pleased to find that the catalyst
loading could be reduced to 1 mol% Rh without too much loss
in activity.
Having established the optimum catalyst and conditions, we
next briefly examined the application of these to the coupling
of a range of substrates and the results are summarised in
Table 2.w In most cases the catalyst loading used was not
optimised but held at 5 mol% Rh, although lowering the
loading to 2.5 mol% and repeating the reactions shown in
entries 1 and 3 led to only a modest decrease in yield (entries 2
and 4, respectively). In all cases the performances obtained
were comparable within a few percent to those reported
previously for the pre-formed phosphinite ligands,4b demon-
strating that chlorodiisopropylphosphine is indeed an ideal
and convenient substitute. The aryl bromide employed in entry
7, 2-bromo-meta-xylene, has not been used previously in this
reaction and demonstrates that the process is tolerant of bulky
aryl bromides.
Complex
2
is catalytically competent, giving the
coupled product 1 in 89% yield at 2.5 mol% loading, indicat-
ing that it is a viable model compound for mechanistic
studies.11 The reaction of complex 2 with 2,4-di-tert-butyl-
phenol in the presence of excess NaOtBu generates
a
new complex assigned as the orthometallated complex 3 on
the basis of the NMR spectroscopic and MS data.w Complex 3
can also be prepared by orthometallation of the k1-phosphi-
nite complex 4. When 2 is reacted with 2,4-di-tert-butylphenol
in the presence of one equivalent of NaOtBu then a mixture of
products is obtained; 31P NMR spectroscopy reveals the major
components to be 4 and 3. The conversion of 4 to 3 reaction is
dependent on the presence of base; no orthometallation is
observed even at reflux temperature in toluene for 24 h in its
As can be seen from entries 8–10, unsubstituted phenol
represents more of a challenge. This is because the rate of
orthometallation falls rapidly with decreasing bulk of the
i
t
phenolic substrate. Replacing the Pr2PCl with Bu2PCl leads
to some di-arylated product, however the conversion obtained
is far from useful (entry 9). We reasoned that whilst ClPCy2 is
too large for bulky phenols to undergo rapid transesterifica-
tion (see above) it may allow nucleophilic attack by smaller
phenols; the increase in steric bulk may be sufficient to
promote a good rate of C–H activation. Unfortunately, there
appears to be little difference in activity on changing to this
absence. This suggests
a
base-assisted deprotonation
mechanism for the C–H activation rather than either
electrophilic displacement or oxidative addition of the
C–H bond.12
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 990–992 | 991

View Article Online
Table 2 Coupling of selected substratesa
istic studies indicate that the rhodium chlorophosphine com-
plexes react with phenolic substrates in the presence of base to
give phosphinite complexes that undergo orthometallation by
base-assisted C–H activation. The good activity shown by
these simple catalysts, comparable with the best reported
previously, coupled with the commercial availability of both
the rhodium precursors and the ClPiPr2 ligand, considerably
increases the attractiveness of the catalytic ortho-arylation of
phenols in synthesis. We are currently exploring the applica-
tion of this class of catalysts to a range of substrates; in
particular we are exploring methodologies that allow the
simple mono-arylation of ortho-unsubstituted phenols and
probing the mechanism of the reaction further.
Entry Phenol Aryl halide
1
Product
Yield (%)b
89
2
3
87c
77
4
5
70c
74
We thank the EPSRC (Advanced Research Fellowship to
RBB, PDRA to MB, DTA studentship to PDL), the Royal
Society of Chemistry (Sir Edward Frankland Fellowship to
RBB), University of the West Indies, St. Augustine (Graduate
Research Grant to LCL-A), the Spanish Ministerio de Educa-
cion y Ciencia (CTQ2005-07918-C02-02/BQU) (SP and DP-C)
and the Universidad Rey Juan Carlos (graduate fellowship
and travel grant to DP-C) for funding, Johnson-Matthey for
the loan of rhodium precursors and the EPSRC national
crystallography service (University of Southampton) for
X-ray data collection.
6
94
7
8
91
(83)d
26e
9
10
10e,f
24e,g
Notes and references
a
Conditions: ArOH (0.5 mmol), ArBr (0.6 mmol), [{RhCl(COD)}2]
(5 mol% Rh), ClPiPr2 (10 mol%), Cs2CO3 (0.85 mmol), toluene
z Crystal data for 2: C14H26Cl2PRh, M ¼ 399.13, monoclinic, a ¼
7.4343(4), b ¼ 20.7246(14), c ¼ 10.6522(7) A, b ¼ 92.041(2)1, V ¼
1640.17(18) A, T ¼ 120(2) K, space group P21/n, Z ¼ 4, m ¼ 1.447
mmꢂ1, Rint ¼ 4.9% (for 16 093 measured reflections), R1 ¼ 4.4% [for
3175 unique reflections with 42s(I)], wR2 ¼ 9.6% (for all 3776 unique
reflections). CCDC 669413. For crystallographic data in CIF format
see DOI: 10.1039/b718128k
b
(5 ml), reflux, 18 h. Spectroscopic yield determined by 1H NMR
spectroscopy (1,3,5-(MeO)3C6H3 internal standard), average of 2
c
runs. 2.5 mol% Rh. Isolated yield.
d
e
2
Equiv. of aryl
f
g
bromide. ClPtBu2 used as ligand. ClPCy2 used as ligand.
1 For an overview see: Handbook of C–H Transformations, ed. G.
Dyker, Wiley, VCH, 2005.
2 For recent reviews see: (a) S. S. Stahl, Angew. Chem., Int. Ed., 2004,
43, 3400; (b) A. R. Dick and M. S. Sanford, Tetrahedron, 2006, 62,
2439.
3 For recent rare examples of biaryl formation using this methodol-
ogy see: (a) D. R. Stuart and K. Fagnou, Science, 2007, 316, 1172;
(b) T. A. Dwight, N. R. Rue, D. Charyk, R. Josselyn and B.
DeBoef, Org. Lett., 2007, 9, 3137; (c) K. L. Hull and M. S.
Sanford, J. Am. Chem. Soc., 2007, 129, 11904.
In summary, simple rhodium chlorodiisopropylphosphine
complexes serve as excellent catalysts for the ortho-arylation
of phenols. These complexes can be either pre-formed or
produced in situ; the former has the benefit of simplicity whilst
the latter yields an air-stable, easily handled species and
requires a lower loading of chlorophosphine. Initial mechan-
4 (a) R. B. Bedford, S. J. Coles, M. B. Hursthouse and M. E.
Limmert, Angew. Chem., Int. Ed., 2003, 42, 112; (b) R. B. Bedford
and M. E. Limmert, J. Org. Chem., 2003, 68, 8669.
5 S. Oi, S. Watanabe, S. Fukita and Y. Inoue, Tetrahedron Lett.,
2003, 44, 8665.
6 Previous studies suggest that orthometallation is the rate-determin-
ing step with aryl bromide substrates and PiPr2(OAr) acting as
co-catalyst; see ref. 4b.
7 Diarylated product is obtained because the rate of the second C–H
activation is much faster than the first since the mono-arylated
phenol is larger.
8 See, for example: Z. Hua, V. C. Vassar and I. Ojima, Org. Lett.,
2003, 5, 3831.
9 B. D. Murray, H. Hope, J. Hvoslef and P. P. Power, Organo-
metallics, 1984, 3, 657.
10 31P NMR spectroscopy reveals the presence of a single new Rh–P
species as a doublet at 154.7 ppm (1JRhP ¼ 140 Hz).
11 Conditions as in Table 1, using Cs2CO3 as base.
Fig. 1 X-Ray crystal structure of complex 2. Selected bond lengths
(A) and angles (1): Rh1–Cl1, 2.3609(11); Rh1–P1, 2.2634(11);
Rh1–C1, 2.232(4); Rh1–C2, 2.251(4); Rh1–C5, 2.120(4); Rh1–C6,
2.143(4); P1–Cl2, 2.0849(15); Cl1–Rh–P1, 88.04(4); Rh1–P1–Cl2,
114.71(6).
12 (a) D. Garcı
´
a-Cuadrado, A. A. C. Braga, F. Maseras and A. M.
a-
Echavarren, J. Am. Chem. Soc., 2006, 128, 1066; (b) D. Garcı
´
Cuadrado, P. de Mendoza, A. A. C. Braga, F. Maseras and A. M.
Echavarren, J. Am. Chem. Soc., 2007, 129, 6880.
ꢀc
This journal is The Royal Society of Chemistry 2008
992 | Chem. Commun., 2008, 990–992