A new method for the conversion of allyl alcohol into π-allyl species promoted by nucleophilic interaction with a CO ligand

Article abstract of DOI:10.1039/b601143h

Upon treatment with an iridium carbonyl complex, [(PN)Ir(CO)₂]⁺, allyl alcohol can be smoothly converted into π-allyliridium species at ambient temperature via nucleophilic interaction of the alcohol with a CO ligand followed by C(allyl)-O bond cleavage in the resultant protonated allyloxycarbonyl intermediate. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b601143h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
A new method for the conversion of allyl alcohol into p-allyl species
promoted by nucleophilic interaction with a CO ligand{
Christian Dubs, Toshiki Yamamoto, Akiko Inagaki and Munetaka Akita*
Received (in Cambridge, UK) 24th January 2006, Accepted 13th March 2006
First published as an Advance Article on the web 23rd March 2006
DOI: 10.1039/b601143h
c) or a protic tether ligand (path d) have been reported.2 During the
Upon treatment with an iridium carbonyl complex,
[(PN)Ir(CO)2]+, allyl alcohol can be smoothly converted into
course of our study on heterodinuclear complexes with P/N-based
multidentate ligands,3 we have found an unprecedented method
p-allyliridium species at ambient temperature via nucleophilic
for activation of allyl alcohol 1 on the iridium-carbonyl complex,
[(PN)Ir(CO)2]+ 2+ (PN = 2-diphenylphosphinomethylpyridine),4,5
interaction of the alcohol with a CO ligand followed by
C(allyl)–O bond cleavage in the resultant protonated allyloxy-
through interaction with a CO ligand (path e), which is applicable
carbonyl intermediate.
to a catalytic conversion.
Interaction of allyl alcohol 1 with the cationic iridium-carbonyl
complex 2?BF45 at ambient temperature afforded different types of
products depending on the reaction conditions (Scheme 2).
A 1 : 1 reaction gave an unstable intermediate 3+, which was
soon converted to a mixture (see below). Although 3+ could not be
characterized spectroscopically due to its instability, 3+ turned out
to be the protonated allyloxycarbonyl species as revealed by (1)
deprotonation with NEt3 (immediately after addition of 1) leading
to the neutral product 4a (R = H; 61% isolated yield; consisting of
two isomers){ and (2) comparison of 4a with 4b (R = Ph; 61%
isolated yield){ derived from cinnamyl alcohol and characterized
by X-ray crystallography (Fig. 1a).§6,7 Complex 4b contains the
five coordinate, trigonal-bipyramidal Ir center, and the remarkable
structural features are the chelating g1-C(LO)–CH2(g2-CHLCHR)
linkage resulting from (1) nucleophilic addition of the OH moiety
in 1 to the CO ligand and (2) the g2-coordination of the olefinic
part. Appearance of a n(CLO) vibration (1644 (4a), 1637 cm21
(4b)) and upfield shift of the olefinic 1H and 13C NMR signals are
consistent with this structure.
A wide variety of stoichiometric and catalytic transformations of
allyl compounds mediated by transition metal species have been
developed and utilized as indispensable tools in organic synthesis.
p(g3)-Allyl species are regarded as the key intermediates of the
transformations,1 and various allyl sources such as halides and
ester derivatives have been employed to extend the synthetic utility
of the transformations. Activation of allyl alcohol has attracted
increasing attention due to economical and environmental
advantages but only a limited number of conversions of allyl
alcohol into p-allyl species have been reported so far.2 Although,
to date, a couple of methods for activation of allyl alcohol have
been established as summarized in Scheme 1, the concerted
oxidative addition (path a) and the SN2-type mechanisms (path b)
are not always applicable to allyl alcohol because of the OH group
being a bad leaving group. Another feasible approach is
dehydration under acidic conditions. The dehydration, however,
usually requires action of an excess amount of Lewis acid or severe
reaction conditions, while rare examples of successful catalytic
activation of allyl alcohol mediated by acidic hydrido species (path
On the other hand, when a mixture of 1 and 2?BF4 was left for a
longer time (1 h) (in particular, in the presence of an excess amount
of 1), a mixture of products was obtained, from which two major
species, 5?BF4 (72% isolated yield when 1/2?BF4 = 20) and 6?BF4,"
were isolated and characterized by X-ray crystallography
Scheme 1
Chemical Resources Laboratory, Tokyo Institute of Technology, R1-27
4259 Nagatsuta, Midori-ku, Yokohama, 226-8503, Japan.
E-mail: makita@res.titech.ac.jp; Fax: +81-45-924-5230
{ Electronic supplementary information (ESI) available: X-ray structural
data for 4b, 5?BF4 and 6?BF4. See DOI: 10.1039/b601143h
Scheme 2
1962 | Chem. Commun., 2006, 1962–1964
This journal is ß The Royal Society of Chemistry 2006

View Article Online
Fig. 1 ORTEP plots of 4b, 5+ and 6+ drawn with thermal ellipsoids at the 30% probability level.
1
(Fig. 1b,c)§ and H and 31P NMR.{ It is notable that (1) 5?BF4
produce a mixture of mono- (8) and diallylaniline (9) (eqn. 1).6 The
reaction at 80 uC in the presence of a catalytic amount of 2 (2
mol%) was completed within 3 h. The catalytic activity of the
decarbonylated species 6?BF4 was slightly better than those of
2?BF4 and 5?BF4.
and 6?BF4 contain two allyl units, one is the g3-allyl ligand and the
other is the non- (5?BF4) or g2-coordinated allyloxycarbonyl
ligand (6?BF4), and (2) the g3-allyl ligands result from oxidative
addition of 1. It was confirmed that thermolysis of 5?BF4 caused
decarbonylation leading to 6?BF4.
Comparison of the structures of 3+ (4) and 5+ strongly suggests
that the g3-allyl ligand in 5+ is formed via C(allyl)–O bond
cleavage of the g2-coordinated allyloxy moiety (3+ A A) followed
by dehydrative condensation with a second molecule of 1
(Scheme 3). The unstable intermediate 3+ should be formed by
g2-coordination of 1 (1 + 2+ A B) followed by nucleophilic
interaction between the OH group in 1 and the Lewis acidic CO
moiety. Cationic five-coordinate (g2-olefin)Ir complexes like B
have precedents8 and the rather strong g2-coordination of the
olefinic part in 1 (B) should assist the subsequent intramolecular
interaction between the OH and CO groups leading to 3+. Thus it
is remarkable that the present study reveals a new method for
conversion of allyl alcohol 1 into g3-allyl species (5+ and 6+) under
mild reaction conditions (at ambient temperature), which involves
oxidative addition via heterolytic C–O bond cleavage of the
protonated allyloxycarbonyl intermediate 3+ (path e in Scheme 1).
These results, in particular, the C–O bond cleavage under mild
conditions, suggest that the iridium complexes appearing in this
communication may show activity for catalytic conversion of allyl
alcohol 1. As expected, the iridium complexes (2?BF4, 5?BF4 and
6?BF4) were active for catalytic allylation of aniline 7 with 12a,b to
(1)
In conclusion, we have demonstrated a novel method for
conversion of allyl alcohol into an g3-allyl species under mild
reaction conditions (path e in Scheme 1), which can be extended to
a catalytic allylation reaction with allyl alcohol. While carbonyl
species have seldom been used as a catalyst precursor for
conversion of allyl alcohol because of their apparently low ability
with respect to oxidative addition (cf. nucleophilic phosphine
complexes), the present result implies that (1) C–O bond oxidative
addition of 1 is promoted by nucleophilic interaction of the OH
group in 1 with the CO ligand and (2) the [Ir(CO)]+ part in 2+
serves as an effective Lewis acid to promote the OC–O bond
formation as well as the subsequent C(allyl)–O bond cleavage, and
thus cationic iridium-carbonyl complexes like 2+ have a potential
to serve as an effective catalyst for allylic transformations with
allylic alcohol 1. Extension of the catalytic reaction is now under
study and the results will be reported in due course.
We thank Prof. Ryo Takeuchi (Aoyama Gakuin University) for
fruitful discussions and Mr Kim Dong Hwan (Sunchon
University, Korea) for his assistance with the catalytic study. A
Monbukagakusho scholarship for C.D. from the Japanese
Government is gratefully acknowledged. The present research is
financially supported by the Japanese Government and the Japan
Society for the Promotion of Science, which are gratefully
acknowledged.
Notes and references
{ Selected spectroscopic data: 2:5 dP (in CDCl3) 27.1, dH (in CDCl3) 4.57
(2H, d, J = 12.0 Hz, PCH2), 7.5–7.8 (11H, m), 7.95 (1H, t, J = 7.6 Hz, py),
Scheme 3
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 1962–1964 | 1963

View Article Online
8.93 (1H, d, py), dC (in CDCl3) 40.9 (d, J = 35 Hz, PCH2), 125–142 (py and
Ph), 162.4 (d, J = 5 Hz, CO), FD-MS m/z = 526 (2), IR (KBr) 2085, 2018
(nCO), 1061 cm21 (nBF); for 4a (1 : 1 mixture of two isomers): dP (in CDCl3)
22.05, 23.65, dH (in CD2Cl2; too complicated to be assigned) 0.82–0.86
(0.5H, m), 1.82 (0.5H, dt, J = 5.4 and 3.2 Hz), 1.71–1.76 (0.5H, m), 2.12–
2.15 (1H, m), 3.08 (0.5H, dd, J = 9.6 and 2.7 Hz), 3.45–3.51 (0.5H, d, J =
9.7 Hz), 3.90 (0.5H, dd, J = 17.6 and 9.8 Hz), 4.2–4.4 (2.5 H, m), dC (in
CD2Cl2) 20.9 (d, J = 8 Hz, CH2LCH), 21.3 (d, J = 23 Hz, CH2LCH), 41.1
(d, J = 7 Hz, LCH), 45.0 (d, J = 28 Hz, PCH2), 45.3 (d, J = 31 Hz, PCH2),
46.4 (d, J = 27 Hz, =CH), 69.5 (d, J = 3 Hz, OCH2), 71.4 (d, J = 5 Hz,
OCH2), FD-MS m/z = 583 (4a), IR (KBr) 1961 (nCO), 1644 cm21 (nCLO);
for 4b: dP (in CD2Cl2) 23.10, dH (in CD2Cl2) 2.56 (1H, dd, J = 16.9 and
9.3 Hz, PhCH=), 3.53 (t, J = 8.0 Hz, CH2O), 3.78 (1H, dd, J = 16.9 and
11.5 Hz, LCHCH2), 3.9 (1H, m, CH2O), dC (in CD2Cl2) 37.0 (d, J = 6 Hz,
Ph–CH), 41.5 (d, J = 29 Hz, –CHCH2), 44.48 (d, J = 28 Hz, CH2P), FD-
MS m/z = 659 (4b), IR (KBr) 1954 (nCO), 1637 cm21 (nCLO); for 5?BF4: dP
(in CDCl3) 0.97, dH (in CDCl3) 3.40 (1H, d, J = 12.3 Hz, g3-allyl), 3.63
(1H, dd, J = 13.7 and 6.1 Hz, g3-allyl), 4.15–4.95 (6H, m), 5.00–5.17 (3H,
m), 5.62 (1H, m, g3-allyl), ESI-MS m/z = 624 (5+), 556 (6+-allyl), IR (KBr)
2057 (nCO), 1655 (nCLO), 1084 cm21 (nBF); for 6?BF4 (4 : 1 mixture of two
isomers): dP (in CDCl3) 26.0, 27.3 (major), dH (in CDCl3) 3.05–3.90 (6H,
m), 4.30–4.82 (5H, m), 5.62 (1H, m, g3-allyl), IR (KBr) 1657 (nCLO),
1081 cm21 (nBF), ESI-MS m/z = 296 (6+). Satisfactory 13C NMR data for
5?BF4 and 6?BF4 could not be obtained due to their low solubility in
organic solvents. The presence of the isomers hampered detailed spectro-
scopic characterization of 4a and 6+. The isomer 69+ could be detected as
the minor component of the disordered structure (6+ : 69+ = 0.8 : 0.2) and
the structure of 4a9 was proposed taking into account the structure of 4b.
1 J. Tsuji, Palladium Reagents and Catalysts: New Perspectives for the
21st Century, Wiley, New York, 2004; J. Tsuji, Transition Metal
Reagents and Catalysts, Wiley, New York, 2000; J. A. Davis, in
Comprehensive Organometallic Chemistry II, ed. E. W. Abel, F. G. A.
Stone and G. Wilkinson, Pergamon Press, Oxford, 1995, Vol. 9,
p. 291; S. A. Godleski, in Comprehensive Organic Synthesis, ed.
B. M. Trost and I. Fleming, Pergamon, New York, 1991,
Vol. 3, p. 585; B. M. Trost and M. L. Crawley, Chem. Rev., 2003, 103,
2921.
2 (a) F. Ozawa, H. Okamoto, S. Kawagishi, S. Yamamoto, T. Minami and
M. Yoshifuji, J. Am. Chem. Soc., 2002, 124, 10968; (b) F. Ozawa,
T. Ishiyama, S. Yamamoto, S. Kawaguchi, H. Murakami and
M. Yoshifuji, Organometallics, 2004, 23, 1698; (c) H. Saburi, S. Tanaka
and M. Kitamura, Angew. Chem., Int. Ed., 2005, 44, 1730; (d) C. Garc´ıa-
Yebra, J. P. Janssen, F. Rominger and G. Helmchen, Organometallics,
2004, 23, 545; (e) R. Takeuchi and M. Kashio, J. Am. Chem. Soc., 1998,
120, 8647. See also references cited therein.
3 (a) S. Tanaka and M. Akita, Angew. Chem., Int. Ed., 2001, 40, 2865; (b)
S. Tanaka, C. Dubs, A. Inagaki and M. Akita, Organometallics, 2004, 23,
317; (c) C. Dubs, A. Inagaki and M. Akita, Chem. Commun., 2004, 2760;
(d) S. Tanaka, C. Dubs, A. Inagaki and M. Akita, Organometallics, 2005,
24, 163; (e) C. Dubs, T. Yamamoto, A. Inagaki and M. Akita,
Organometallics, 2006, 25, 1344; (f) C. Dubs, T. Yamamoto, A. Inagaki
and M. Akita, Organometallics, 2006, 25, 1359.
4 PN complexes: P. Braunstein, B. T. Heaton, C. Jacob, L. Manzi and
˚
X. Moise, Dalton Trans., 2003, 1396; B. Akermark, B. Krakenberger,
S. Hansson and A. Vitagliano, Organometallics, 1987, 6, 620.
5 Complex 2?BF4 was prepared by repeated carbonylation (1 atm)3f of
[(PN)Ir(g4-cod)]BF4, which was obtained by treatment of (m-Cl)2{Ir(g4-
cod)}2 with the PN ligand4 in the presence of AgBF4.
6 In a previous paper,3f we reported (1) formation of a dinuclear
allyloxycarbonyl species analogous to 4a (but incorrectly assigned as
the g1-species because of lack of crystallographic structural information)
and (2) allylation reaction of 7 with 1 catalyzed by heterodinuclear
complexes such as [(OC)2Ir(m-PNNP)Pd(g3-allyl)]+ (PNNP = 3,5-
bis(diphenylphosphinomethyl)pyrazolato). But the mechanism of
the activation of 1 could not be clarified because of (1) the
dinuclear system being too complicated to be studied in detail and
(2) the presence of stereoisomers (e.g. four isomers for (allyl–
OOC)(OC)Ir(m-PNNP)Pd(g3-allyl) corresponding to 4a). Meanwhile
the present result apparently supports the mononuclear mechanism at
Ir or at Pd discussed in ref. 3f but the dinuclear mechanism cannot
be always ruled out on the basis of different reaction features (e.g. effect
of CO). Further study is needed to clarify the mechanism of the dinuclear
system.
§ X-ray data collections were carried out with a Rigaku RAXIS-IV
imaging plate area detector at 260 uC. 4b: C29H25IrNO3P, fw = 658.72,
˚
˚
monoclinic, space group P21/n, a = 10.4334(8) A, b = 14.1368(9) A, c =
˚
3
16.992(1) A, b = 94.627(4)u, V = 2498.0(3) A , Z = 4, dcalcd = 1.751 g?cm23
,
˚
R1 = 0.0346 (refined on F2) for 4960 data (I > 2s(I)) and 332 parameters.
5?BF4: C26H26IrNO3BF4P, fw = 710.49, monoclinic, space group C2/c, a =
˚
3
˚
˚
33.60(1) A, b = 10.959(2) A, c = 15.568(5) A, b = 112.406(8)u, V =
5298(2) A , Z = 8, dcalcd = 1.781 g?cm23, R1 = 0.0633 (refined on F2) for
˚
4549 data (I > 2s(I)) and 334 parameters. 6?BF4?CH2Cl2:
C26H28NO2BF4Cl2PIr, fw = 767.42, triclinic, space group P-1, a =
˚
˚
˚
9.9562(9) A, b = 11.5147(7) A, c = 12.548(1) A, a = 91.706(5)u, b =
103.789(3)u, c = 91.982(5)u, V = 1395.2(2) A , Z = 2, dcalcd = 1.827 g?cm23
,
3
˚
7 An example of an g1:g2-allyloxy carbonyl complex: R. J. Madhushaw,
S. R. Cheruku, K. Narkunan, G.-H. Lee, S.-M. Peng and R.-S. Liu,
Organometallics, 1999, 18, 748.
R1 = 0.0494 (refined on F2) for 5644 data (I > 2s(I)) and 368 parameters.
CCDC 296203–296205. For crystallographic data in CIF or other
electronic format see DOI: 10.1039/b601143h
8 A related five coordinate (g2-olefin)iridium complex with the PN
ligand, [(m-g2:g2-cod){Ir(PN)(CO)2}2](BF4)2, was characterized by X-ray
crystallography (see supporting information of ref. 3e (complex 6)).
" The isolated yield of 6?BF4 was variable and dependent on the solvent
and crystallization condition. A better yield was obtained by the reaction in
and slow crystallization from THF.
1964 | Chem. Commun., 2006, 1962–1964
This journal is ß The Royal Society of Chemistry 2006