Addition of the ortho-C-H bonds of phenol across an olefin catalysed by a chiral iridium(I) diphosphine complex

Article abstract of DOI:10.1039/b211672c

A chiral iridium(1) diphosphine complex efficiently catalyses the unprecedented ortho-alkylation of phenol with norbornene in the absence of solvent leading to the formation of one and two C-C bonds.

Full text of DOI:10.1039/b211672c

Addition of the ortho-C–H bonds of phenol across an olefin catalysed by
a chiral iridium( ) diphosphine complex
I
Romano Dorta*a and Antonio Tognib
a Departamento de Química, Universidad Simón Bolívar, Caracas 1080A, Venezuela.
E-mail: rdorta@usb.ve; Fax: +58 212 9063961
b Department of Chemistry, Swiss Federal Institute of Technology, ETH Hönggerberg, CH-8093 Zürich,
Switzerland
Received (in Cambridge, UK) 27th November 2002, Accepted 6th February 2003
First published as an Advance Article on the web 19th February 2003
A chiral iridium(
I
) diphosphine complex efficiently catalyses
from the reaction of the monoalkylated phenol 2 with a second
equiv. of norbornene and shows that the same reactivity is
obtained with ortho-substituted phenols such as 2. The third
product was the dimer of norbornene, exo-norbornyl-norbor-
nene 4 (see eqn. 2). Although we do not yet know whether the
addition of the sp2 C–H bonds across the double bond of
norbornene in 2 and 3 occurs in an endo or exo fashion, the
selectivity for either one of the two possibilities is complete.
Based on numerous precedents and preliminary NMR data we
clearly favor exo-addition.7
the unprecedented ortho-alkylation of phenol with norbor-
nene in the absence of solvent leading to the formation of one
and two C–C bonds.
Arenes such as benzenes, phenols, and anilines are important
commodity chemicals. Catalytically efficient activation of
aromatic C–H bonds leading to new C–C bond formation is thus
a highly desirable goal and would provide a clean and economic
method for adding value to such compounds. Although there are
numerous examples of stoichiometric reactions of aromatic C–
H bonds with transition metal compounds, catalytic systems for
addition to C–C multiple bonds are rare.1 Brunet et al.
demonstrated the feasability of rhodium catalysed ortho-
alkylations of anilines by norbornene.2 Ruthenium3a,b and
rhodium3c complexes were shown to catalyse ortho-alkylations
of aromatic ketones by olefins. A protocol for the reaction of
phenols with alkynoates catalysed by electrophilic palladium
complexes in acidic media appeared recently.4 One of us has
reported the first example of a direct, asymmetric hydro-
arylation of norbornene with benzamide catalysed by electron-
(1)
(2)
The optical rotations of 2 (22.5, c 1.55, CHCl3) and 3 (2 6.5,
c 1.34, CHCl3) indicated some degree of enantioselectivity. The
enantiomeric excess (ee) of 2 was 4.5% (determined by chiral
HPLC, see Experimental footnote†). Unfortunately, the three
possible exo-(or endo-)isomers of 3 (R,R, S,S, and meso) could
not be completely separated on a chiral column (best results
obtained with Chiracel OD-H, 1% i-PrOH–hexane, 0.5 mL
min21, l = 214 nm).
A plausible reaction mechanism is outlined in Scheme 1 and
involves initial coordination of phenol cleaving the chloro-
bridges in dimer 1 leading to the monomeric square-planar
species A. This electron-rich species then undergoes selective
ortho-metallation of phenol, which is probably the rate-
determining step.8 This is followed by insertion of the double
bond of norbornene giving species C. Reductive elimination of
2 leads to the regeneration of 1. We have shown that complexes
rich Ir(
) complexes.5
I
We present here first results on the Ir-catalysed, enantiose-
lective (although marginally) ortho-norbornylation of phenol.
To the best of our knowledge, this reaction represents the first
example of a coupling of a non-functionalised olefin with
phenol catalysed by a transition-metal complex. Furthermore, it
proceeds under mild conditions, does not require any solvent
and can be conducted in nearly stoichiometric amounts of
reactants. We also present an efficient Ir-catalysed dimerisation
of norbornene.
The dinuclear Ir complex [IrCl{(R)-(S)-PPFPPh2}]2 (1, Fig.
1, (R)-(S)-PPFPPh2 = (R)-1-{(S)-2-(diphenylphosphino)ferro-
cenyl}ethyldiphenylphosphine) is isolable in high yield as a ca.
60 + 40 cis/trans mixture by reaction of a benzene solution of
(R)-(S)-PPFPPh2 with a benzene slurry of [IrCl(COE)2]2 (COE
= cyclooctene).6
Complex 1 was used as catalyst precursor in the model
reaction of norbornene with phenol. All the reactions were
performed without solvent under anaerobic and anhydrous
conditions. Heating a solution of 2 equiv. of norbornene, 1
equiv. of phenol, and 1 mol% of complex 1 at 373 K for 72 h
resulted in the formation of a viscous oil (eqn. 1).† A TLC
analysis of the reaction mixture revealed three reaction products
and no starting material. Upon separation by flash chromatog-
raphy 2 was isolated in 69% yield as an oil and 3 in 13% yield
as fine yellowish needles. Both products were analytically pure
and characterised by GC-MS and NMR.† Compound 3 arises
Fig. 1
Scheme 1
760
CHEM. COMMUN., 2003, 760–761
This journal is © The Royal Society of Chemistry 2003

View Article Online
mg, 0.0585 mmol) resulting in a yellow slurry which was heated at 373 K
of the type [IrCl(PP)*]2 undergo smooth and reversible C–H
bond activation,9 and catalyse the addition of benzamide to
norbornene via C–H activation.5 Furthermore, an example of an
iridium complex utilising the halogen of haloarenes as a
directing group for the selective ortho-C–H activation of such
compounds will appear soon.10
Different control experiments underline the validity of such a
catalytic cycle. Experiments at lower temperature (348 K)
resulted in poorer yields. The presence of water, which was
shown to oxidatively add to complex 1 at room temperature,
significantly lowered the yields.11 Using the hydroxo-bridged
for 72 h. During the course of the reaction the mixture turned limpid orange
and small amounts of a black deposit were observed. Purification through a
silica plug and drying in vacuo afforded a colorless oil (4, 834 mg, 76%).
Characterisation of 2-norbornyl-phenol (2). Anal. calc. for C13H16O:
C, 82.94; H, 8.57. Found: C, 82.94; H, 8.80%. 1H NMR (CDCl3, 250.13
MHz): d 1.25–1.45 (m, 3H), 1.45–1.70 (m, 4H), 1.75–1.90 (m, 1H), 2.37 (m,
1H), 2.45 (m, 1H), 2.86 (m, 1H), 4.74 (s, 1H), 6.75–6.85 (m, 1H), 6.85–6.95
(m, 1H), 7.00–7.15 (m, 1H), 7.15–7.25 (m, 1H). Rf = 0.42 (hexane +
20
ethylacetate
=
10). [a]D
= 22.5 (c = 1.5525 in CHCl3). The
enantiomers were separated by HPLC using Daicel’s Chiracel OD-H
column (eluent: i-PrOH–hexane 4+96 v/v, 0.5 mL min21, 298 K) with
retention times of 13.8 and 15.4 min.
Ir(III) hydride complex [{(R)-(S)-PPFPPh2}(H)IrCl(m-OH-
Characterisation of bis-2,5-norbornyl-phenol (3). Anal. calc. for
C20H26O: C, 85.06; H, 9.28. Found: C, 85.27; H, 9.31%. 1H NMR (CDCl3,
250.13 MHz): d 1.15–1.45 (m, 6H), 1.45–1.70 (m, 8H), 1.70–1.90 (m, 2H),
2.37 (m, 2H), 2.44 (m, 2H), 2.83 (m, 2H), 4.78 (s, 1H), 6.86 (t, 1H), 7.06 (m,
)2(H)IrCl{(R)-(S)-PPFPPh2}], which is the reaction product of 1
with water, as catalyst precursor resulted in ca. 35% lower
yields. Addition of co-catalysts such as ‘naked’ fluoride
(phosphazenium fluoride-P2), TlF, or ZnCl2 had no or detri-
mental effects on the activity. Furthermore, the more challeng-
ing sp3 C–H activation was not observed. Indeed, stirring a
mixture of 2 equiv. of norbornene, 1 equiv. of t-BuOH, and 1
mol% of 1 at 373 K for 72 h afforded exo-norbornyl-norbornene
4 in 76% isolated yield as the only reaction product along with
unreacted t-BuOH (eqn. 2).12† Moreover, this surprising
catalytic dimerisation of norbornene does not proceed in
absence of t-BuOH, indicating that initial coordination of t-
BuOH to complex 1 is crucial insofar as it enhances the
nucleophilicity of the iridium centre and allows the following
olefin sp2 C–H activation step. Finally, control experiments
without the use of catalyst 1 did not show any reaction between
norbornene and phenol or t-BuOH.
20
2H). [a]D
ethylacetate = 10).
Characterisation of norbornyl-norbornene (4). Anal. calc. for C14H20
= 26.5 (c = 1.3405 in CHCl3). Rf = 0.61 (hexane +
:
C, 89.30; H, 10.70. Found: C, 89.36; H, 10.68. 1H NMR (C6D6, 250.13
MHz): d 1.10–1.30 (m, 6H), 1.35–1.80 (m, 8H), 2.20 (m, 1H), 2.29 (m, 2H),
2.71 (m, 1H), 2.86 (m, 1H), 5.60 (m, 1H). Rf = 0.88 (hexane + ethylacetate
= 10).
1 (a) E. Shilov and G. B. Shulpin, Chem. Rev., 1997, 97, 2879; (b)
Activation of Unreactive Bonds and Organic Synthesis, Topics in
Organometallic Chemistry, vol. 3, ed. S. Murai, Springer Verlag, New
York, 1999; (c) F. Kakiuchi and S. Murai, Acc. Chem. Res., 2002, 35,
826.
2 (a) J.-J. Brunet, D. Neibecker and K. Philippot, J. Chem. Soc., Chem.
Commun., 1992, 1215; (b) J.-J. Brunet, G. Commenges, D. Neibecker
and K. Philippot, J. Organomet. Chem., 1994, 469, 221.
3 (a) S. Murai, F. Kakiuchi, S. Sekine, Y. Tanaka, A. Kamatani, M.
Sonoda and N. Chatani, Nature, 1993, 366, 529; (b) S. Murai, N.
Chatani and F. Kakiuchi, Pure Appl. Chem., 1997, 69, 589; (c) C. P.
Lenges and M. Brookhart, J. Am. Chem. Soc., 1999, 121, 6616.
4 (a) B. M. Trost and F. D. Toste, J. Am. Chem. Soc., 1996, 118, 6305; (b)
C. Jia, W. Lu, J. Oyamada, T. Kitamura, K. Matsuda, M. Irie and Y.
Fujiwara, J. Am. Chem. Soc., 2000, 122, 7252; (c) C. Jia, D. Piao, J.
Oyamada, W. Lu, T. Kitamura and Y. Fujiwara, Science, 2000, 287,
1992; (d) C. Jia, T. Kitamura and Y. Fujiwara, Acc. Chem. Res., 2001,
34, 633.
5 R. Aufdenblatten, S. Diezi and A. Togni, Monatsh. Chem., 2000, 131,
1345.
6 R. Dorta, P. Egli, F. Zürcher and A. Togni, J. Am. Chem. Soc., 1997,
119, 10857.
7 For N–H and C–H bonds adding across the double bond of norbornene
exclusively in an exo-fashion, see references 3, 5, 6, and also: H. L.
Casalnuovo, J. C. Calabrese and D. Milstein, J. Am. Chem. Soc., 1988,
110, 6738.
8 In fact, prolonged heating at 373 K of 1 in a benzene or toluene solution
does neither alter the cis/trans ratio of 1, nor is any C–H activation of the
solvents detected.
In summary we have demonstrated the feasibility of an
enantioselective, atom-economical ortho-alkylation of phenols
without the need of a solvent. We propose a catalytic cycle that
features an sp2 C–H bond activation with subsequent coupling
to the double bond of norbornene. These first results show that
not only phenol, but also ortho-alkylated phenols (in this case
compound 2) undergo such a reaction. Alkylation of t-BuOH by
norbornene is not observed, but instead we observe selective
dimerisation of norbornene via olefin sp2 C–H activation and
subsequent C–C coupling with a second olefin.
We are currently working on the full stereochemical
characterisation and separation of the isomers of compounds
2–4. Future studies on the reaction will include (a) exploring the
effect of different donor/acceptor groups on the phenol ring, (b)
expanding the reaction to other olefin substrates,13 and (c)
investigating the steric and/or electronic effects of different
chiral (PP)* ligands on the performance of the catalyst
precursor (of general formula [IrCl(PP)*]2) described
herein.14
We thank Reto Dorta for valuable discussions and for his help
in the preparation of the manuscript. R. D. thanks Novartis AG
(Switzerland) and FONACIT (Venezuela) for financial sup-
port.
9 R. Dorta and A. Togni, Organometallics, 1998, 17, 3423. Note that also
in this case, C–H activation is achieved after initial coordination of an
amine group to the metal centre. Future efforts will be directed at
isolating the proposed intermediates A and B.
10 E. Ben-Ari, M. Gandelmann, H. Rozenberg, L. J. W. Shimon and D.
Milstein, submitted for publication.
Notes and references
†
Catalytic condensation of phenol with norbornene. Norbornene
11 Separate experiments showed that 1 does not mediate any reaction
between water and norbornene at 373 K.
(1136 mg, 12.06 mmol) was gently warmed in the presence of phenol (568
mg, 6.04 mmol) affording a solution which was added to [IrCl{(R)-(S)-
PPFPPh2}]2·0.2C5H12 (1, 98.5 mg, 0.0603 mmol). The resulting limpid
orange solution was stirred at 373 K for 72 h. The viscous reaction mixture
was quenched by adding CH2Cl2 in air. Drying in vacuo afforded 1555 mg
of a brownish oil. FLASH chromatography (hexane + ethylacetate = 10)
allowed us to separate the reaction products: 2-norbornylphenol (2, 782 mg,
yellowish oil), 2,5-bis-norbornylphenol (3, 217 mg, yellowish needles), and
norbornyl-norbornene (4, 330 mg, colourless oil).
12 It appears that diastereomeric control in the C–C bond forming step is
complete as determined by GC-MS and 13C-NMR. Theoretically, two
diastereomeric pairs of enantiomers of the exo-product are possible.
13 Norbornene belongs to a rather special class of strained cyclic olefins
and the question of whether C–H activation or olefin insertion are rate-
limiting may depend in part on the olefin used.
14 In the closely related olefin hydroamination protocol (cf. ref. 6),
employing complex 1 gave low yields (27%, compared to a maximum
of 81% with [IrCl{(R)-(S)-Josiphos}]2) and low ee values (9%,
compared to a maximum value of 95% with [IrCl{(S)-BINAP}]2).
Catalytic dimerisation of norbornene. A mixture of norbornene (1101
mg, 11.69 mmol) and t-BuOH (433 mg, 5.84 mmol) was added to 1 (95.5
CHEM. COMMUN., 2003, 760–761
761