Halide-free ethylation of phenol by multifunctional catalysis using phosphinite ligands

Article abstract of DOI:10.1039/b610038d

The ortho-alkylation of phenols or aniline by catalytic C-H activation and multifunctional catalysis is described. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b610038d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Halide-free ethylation of phenol by multifunctional catalysis using
phosphinite ligands
M. Carmen Carri o´ n and David J. Cole-Hamilton*
Received (in Cambridge, UK) 13th July 2006, Accepted 15th August 2006
First published as an Advance Article on the web 18th September 2006
DOI: 10.1039/b610038d
The ortho-alkylation of phenols or aniline by catalytic C–H
obtained in all cases, with Ph P(OPh) giving slightly higher yields
2
activation and multifunctional catalysis is described.
than Et P(OPh). We think that this may be because of the very
2
high oxygen sensitivity of Et P(OPh). GC analysis of the spent
2
The formation of carbon–carbon bonds is a key component of
many transformations in the pharmaceutical, agrochemical and
organic chemical industries. Generally speaking, these transforma-
tions involve the reaction of alkyl or aryl halides with an alkyl or
arylating agent, or with an alkene (Heck, Suzuki, Stille etc.
couplings). In all these cases, the halogen is first introduced into
one of the substrates and is then lost, necessitating extra synthetic
steps and generating waste. Replacing such reactions by the direct
reaction solutions showed that all of the phosphinite was oxidised
to Et or 2,6-Et
2
P(O)(OR) (R = Ph, 2-EtC
6
H
4
2 6 3
C H ).
2 2
Using the [RhCl(cyclooctene) ] as the catalyst precursor, the
yields are generally lower than with Wilkinson’s catalyst under all
of the studied conditions. This may be because triphenylphosphine
helps to ensure that stable rhodium complexes are present at all
times, even if some of the phosphinite is degraded.
A possible reaction mechanism for the alkylation reaction is
shown in Scheme 1. After coordination of the phosphinite ligand
and ortho-metallation, ethene coordinates and hydride migration
gives a coordinated ethyl group. Reductive elimination of the
ethylated product occurs and transesterification at P with phenol
leads to 2-ethylphenol, which can itself enter the catalytic cycle to
give diethylphenol. Diethylphenol may also be formed by a second
alkylation prior to transesterification.
1
replacement of C–H bonds would be a major step towards cleaner
more efficient syntheses. The ruthenium-catalysed Murai reac-
2,3
tion allows the insertion of ethene into an ortho C–H bond of
acetophenone, but is restricted to ketones because ortho-metalla-
tion is only favoured if a 5-membered ring is formed. We now
report that the Murai reaction can be extended to include
substrates such as phenols or anilines by attaching the substrate to
phosphorus. After alkylation, the product is released and the new
substrate introduced onto P by a transesterification reaction. We
have previously used such transesterifications for the hydrogena-
Given the high efficiency of the rhodium catalysts, we studied
the behaviour of other metals (Table 2), although the results were
not as good as with rhodium.
4
tion of acrylic acids via mixed anhydrides with phosphinic acids,
With the Pd(II) catalyst, there was no reaction at any
temperature studied, although a small amount of activity to give
whilst Bedford and Limmert have shown that a similar type of
reaction can be used for the ortho-arylation of phenols by aryl
2-ethylphenol was found when using Pd(0) at 120–150 uC. Since
5
6
bromides. There has been a report of the stoichiometric
alkylation of triphenylphosphite with ethylene, catalysed by
ruthenium complexes; weak catalytic activity (15 total turnovers)
was observed when the reaction was carried out in the presence of
excess phenol and base.
oxidative addition (ortho-metallation) is important in the catalytic
cycle, a low-valent catalyst precursor is essential. The ruthenium
catalyst was more active than the palladium complexes, although
the maximum conversion reached was 31.5% at 150 uC. In this
case, diethylphenol was produced, but 2-ethylphenol was the main
product at all temperatures.
In order to study the versatility of the reaction, we have
investigated other substrates for ethylation, such as aniline and
cresols. We have tried the reaction using aniline as the substrate,
ð1Þ
Reactions between phenol and ethene (30 bar) were carried out
at different temperatures (80 to 150 uC) using Wilkinson’s catalyst
or [RhCl(cyclooctene) ] (30 mol%) in the presence of R P(OPh)
using Wilkinson’s catalyst in the presence of Ph
2
P(NHPh). The
results are collected in the Table 3.
2
2
2
With aniline, the result was much less efficient than with phenol.
Even with 30% catalyst, the maximum conversion was 27.3% at
(R = Et or Ph). These studies revealed that 120 uC was the best
temperature to conduct the reaction. No reaction was observed in
1
50 uC. It is also interesting that in this case, the yields of both
the absence of the phosphinite. The coordination of Et P(OPh) to
2
products increased with temperature, and that the result at 150 uC
was better than at 120 uC. The fact that the yields under these
conditions were barely more than stoichiometric may suggest that
the transamination step is not occurring. Using lower catalyst
loadings (5%), the conversion after 15 h (9.4% 2-ethylaniline, 8.3%
these metallic precursors was confirmed by NMR experiments,
2 3
where [RhCl(Et P(OPh)) ] was observed in both cases.{
Results obtained with smaller catalyst loadings are collected in
Table 1. High conversions, mainly to 2,6-diethylphenol, are
2,6-diethylaniline) could also be accounted-for by reactions not
EaStCHEM, School of Chemistry, University of St Andrews, The
Purdie Building, North Haugh, St Andrews, Fife, UK KY16 9ST.
E-mail: djc@st-and.ac.uk; Fax: +44 (0)1334 463808;
Tel: +44 (0)1334 463805
involving transesterification. However, the reaction did not stop,
and continued at an increased rate (to give 15.1% 2-ethylaniline
and 33.2% diethylaniline), which must have involved some
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 4527–4529 | 4527

View Article Online
a
Table 1 Rhodium-catalyzed ethylation of phenol
b
b
Entry
Catalyst
Mol% catalyst
Ligand
Et P(OPh)
Yield ethylphenol (%)
Yield diethylphenol (%)
1
2
3
4
5
6
7
8
a
[RhCl(PPh
[RhCl(PPh
[RhCl(PPh
[RhCl(PPh
3
)
3
)
3
)
3
)
3
]
]
]
]
5
5
2.5
1
5
5
2.5
1
2
10.7
0.0
0.9
7.0
5.7
4.3
5.4
3.7
76.3
87.5
93.1
19.6
30.6
65.3
52.4
16.2
3
3
3
Ph
Ph
Ph
2
P(OPh)
P(OPh)
P(OPh)
2
2
{RhCl(cyclooct)
{RhCl(cyclooct)
{RhCl(cyclooct)
{RhCl(cyclooct)
2
2
2
2
}
2
}
2
}
2
}
2
Et
Ph
Ph
Ph
2
P(OPh)
2
P(OPh)
P(OPh)
P(OPh)
2
2
Reaction conditions: 6 mmol phenol; Pethylene = 30 bar; T = 120 uC; reaction time = 15 h; solvent = 10 ml toluene; catalyst/ligand ratio =
: 3.3. Yields were determined from GC, based on the total amount of phenoxide present at the start (including ligand).
b
1
transamination. It seems that there is probably an induction period
for forming the active catalyst, but that the overall catalytic
reaction is very slow.
In the reaction with aniline as the substrate, small amounts of
N-ethylaniline and 2-methylquinoline were formed; however, they
are sub-stoichiometric. We have also observed 2-ethylphenyldi-
3
phenylphosphine from the ethylation of PPh in reactions
involving Wilkinson’s catalyst.
To study the possible effect of the methyl group in different
positions on the phenol ring, the reaction has also been
investigated using Wilkinson’s catalyst and different cresols . The
ligands were synthesized by the same procedure as the phenol-
derived ligand, using the corresponding cresol in each case.{
When the reaction was run with the Ph P(OPh) ligand, the
2
result was similar to that obtained with a ligand derived from
the relevant cresol, but a mixture of ethyl and diethylphenol, and
ethyl and diethylcresol (when possible) was observed. Because of
this, each cresol-derived ligand was synthesized to facilitate
analysis of the results by GC. The results are summarized in
Table 4.
Scheme 1 Possible mechanism of the ethylation of phenol by multi-
functional catalysis.
a
Table 2 Catalytic ethylation of phenol
b
b
Entry
Catalyst
T/uC
Ligand
Yield ethylphenol (%)
Yield diethylphenol (%)
1
2
3
4
5
6
7
8
9
[PdCl
[PdCl
[PdCl
2
2
2
(CH
(CH
(CH
3
3
3
CN)
CN)
CN)
2
2
2
]
]
]
80
120
150
80
120
150
80
Et
Et
Et
Ph
Ph
Ph
Ph
Ph
Ph
2
2
2
P(OPh)
P(OPh)
P(OPh)
—
—
—
—
4.5
5.6
11.3
14.0
17.2
—
—
—
—
—
—
7.5
8.9
14.3
[Pd
[Pd
[Pd
2
2
2
(dba)
3
3
3
]
]
]
2
P(OPh)
P(OPh)
P(OPh)
P(OPh)
P(OPh)
P(OPh)
(dba)
(dba)
2
2
2
2
2
[RuH
[RuH
[RuH
2
2
2
(CO)(PPh
(CO)(PPh
(CO)(PPh
3
3
3
)
)
)
3
3
3
]
]
]
120
150
a
Reaction conditions: 1 mmol phenol; 0.3 mmol catalyst; Pethylene = 30 bar; reaction time = 15 h; solvent = 10 ml toluene; catalyst/ligand ratio
1 : 3.3. Yields were determined from GC.
b
=
a
Table 3 Rhodium catalyzed ethylation of aniline
b
b
Entry
Aniline/mmol
Mol% catalyst
T/uC
Ligand
Yield ethylaniline (%)
Yield diethylaniline (%)
1
2
3
4
5
1
1
1
6
6
6
30
30
30
5
5
5
80
Ph
Ph
Ph
Ph
Ph
Ph
2
P(NHPh)
P(NHPh)
P(NHPh)
P(NHPh)
P(NHPh)
P(NHPh)
4.9
6.8
17.1
3.2
9.4
15.1
3.6
5.9
10.2
0.7
8.3
33.2
120
150
120
150
150
2
2
2
2
2
c
6
a
3 3
Reaction conditions: 3 mmol RhCl(PPh ) catalyst; Pethylene = 30 bar; reaction time = 15 h; solvent = 10 ml toluene; catalyst/ligand ratio =
c
: 3.3. Yields were determined from GC. Reaction time = 25 h.
b
1
4
528 | Chem. Commun., 2006, 4527–4529
This journal is ß The Royal Society of Chemistry 2006

View Article Online
a
Table 4 Rhodium-catalyzed ethylation of different cresols
with 1% catalyst (Table 4, entry 8), the first ethylation was almost
complete before the second ethylation had started.
Yield
Yield
For ortho-cresol, no C–H activation of the methyl group
occurred, and only ethylation in the other ortho position was
observed. Thus, this ligand gives the worst activity results, and the
result was only slightly better when the diphenyl ligand was used
(Table 4, entry 9) than when the ligand with ortho-cresol was used
(Table 4, entry 10). Therefore, it seems than the ortho-cresol may
impart some steric hindrance around the rhodium centre, which
could inhibit the ortho-metallation step.
Mol%
Entry catalyst
ethylcresol diethylcresol
b b
Ligand
Ph P(OPh)
(%)
(%)
1
2
3
4
5
6
7
8
9
5
5
2.5
1
5
5
2.5
1
2
0.7
1.6
5.0
5.8
44.6
44.1
69.5
84.4
98.9
88.9
69.1
22.3
99.1
98.1
89.0
51.6
55.4
55.9
30.5
1.3
—
—
—
—
Ph
Ph
Ph
2
2
2
P(O(p-MeC
P(O(p-MeC
P(O(p-MeC
6
6
6
H
H
H
4
4
4
))
))
))
Ph
2
P(OPh)
Ph
Ph
Ph
2
P(O(m-MeC
P(O(m-MeC
P(O(m-MeC
6
6
6
H
H
H
4
))
))
))
2
2
4
4
In summary, we have developed a system for the ethylation of
different aromatic compounds by multifunctional catalysis based
5
Ph
2
P(OPh)
1
1
1
0
1
2
5
2.5
1
Ph
2
Ph
2
Ph
2
P(O(o-MeC
P(O(o-MeC
P(O(o-MeC
6
6
6
H
4
H
4
H
4
))
))
))
2
on rhodium complexes of R PXPh ligands (R = Ph or Et, X = O
or NH) that involves C–H activation and 100% atom economy.
Financial support from the Ministerio de Educaci o´ n y Ciencia
of Spain (EX-2005-0237) is gratefully acknowledged.
a
Reaction conditions: 6 mmol cresol; RhCl(PPh
3
)
3
catalyst; Pethylene
=
30 bar; T = 120 uC; reaction time = 15 h; solvent = 10 ml toluene;
b
catalyst/ligand ratio = 1 : 3.3. Yields were determined from GC,
based on the ratio of the integration of each product in the
corresponding FID chromatogram.
Notes and references
{
and only 0.5 h when the dimer was used. [RhCl(Et
The reaction needed 2 h for completion when using Wilkinson’s catalyst,
31
2
P(OPh))
3
], P NMR
For para-cresol, it seems that the methyl group does not impart
any steric hindrance, and the results are comparable to those
obtained with phenol. Under all conditions, diethylcresol was the
main product of the reaction. With 5% catalyst, the conversion
was almost total to the diethylated product, and the yield was only
slightly lower with 2.5% catalyst. However, with 1% catalyst, the
yield was, as with phenol, significantly reduced.
1
2
(C D ): d = 155.6 (dd, 2 P, JP–Rh = 158.3 Hz, JP–P = 37.8 Hz) and 161.7
6 6
1
(dt, 1 P, J
2
P–Rh
P–P
= 206.1 Hz, J = 37.7 Hz).
3
1
{
P NMR (C
P(O(m-MeC
6
D
6
):
d
=
112.7 (Ph
))) and 111.5 (Ph P(O(o-MeC H
2
P(O(p-MeC
))).
6 4
H ))), 111.7
(
Ph
2
6
H
4
2
6 4
1
For reviews on catalytic C–H activation, see: (a) R. H. Crabtree, J. Chem.
Soc., Dalton Trans., 2001, 2437; (b) V. Ritleng, C. Sirlin and M. Pfeffer,
Chem. Rev., 2002, 102, 1731; (c) F. Kakiuchi and S. Murai, Acc. Chem.
Res., 2002, 35, 826.
For meta-cresol, the second ethylation is much more difficult
because the 2-position of the phenyl ring is less accessible due to
the nearby methyl group. This means that with 5% catalyst, the
selectivity for the diethylated product was only 55.9%, while the
rest was only monoethylated. As the amount of catalyst was
decreased, the ratio of ethylated/diethylated product increased, and
2
S. Murai, F. Kakiuchi, S. Sekine, Y. Tanaka, A. Kamatani, M. Sonoda
and N. Chatani, Nature, 1993, 366, 529.
3 S. Murai, N. Chatani and F. Kakiuchi, Pure Appl. Chem., 1997, 68, 589.
4
A. Iraqi, N. R. Fairfax, S. A. Preston, D. C. Cupertino, D. J. Irvine and
D. J. Cole-Hamilton, J. Chem. Soc., Dalton Trans., 1991, 1929.
R. B. Bedford and M. E. Limmert, J. Org. Chem., 2003, 68, 8669.
5
6 L. N. Lewis and J. F. Smith, J. Am. Chem. Soc., 1986, 108, 2728.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 4527–4529 | 4529