Bis(allyl)-ruthenium(IV) complexes as highly efficient catalysts for the redox isomerization of allylic alcohols into carbonyl compounds in organic and aqueous media: Scope, limitations, and theoretical analysis of the mechanism

Article abstract of DOI:10.1021/ja054827a

The catalytic activity of the bis(allyl)-ruthenium(IV) dimer [{Ru(η³:η³-C₁₀H₁₆)(μ-Cl) Cl}₂] (C₁₀H₁₆ = 2,7-dimethylocta-2,6-diene-1,8- diyl) (1), and that of its mononuclear derivatives [Ru(η³: η³-C₁₀H₁₆)Cl₂(L)] (L = CO, PR₃, CNR, NCR) (2) and [Ru(η³:η³-C ₁₀H₁₆)Cl(NCMe)₂][SbF₆] (3), in the redox isomerization of allylic alcohols into carbonyl compounds, both in tetrahydrofuran and in water, is reported. In particular, a variety of allylic alcohols have been quantitatively isomerized using [{Ru(η³: η³-C₁₀H₁₆)(μ-Cl)Cl}₂] (1) as catalyst, the reactions proceeding in all cases faster in water. Remarkably, complex 1 has been found to be the most efficient catalyst reported to date for this particular transformation, leading to TOF and TON values up to 62 500 h^-1 and 1 500 000, respectively. Moreover, catalyst 1 can be recycled and is capable of performing allylic alcohol isomerizations even in the presence of conjugated dienes, which are known to be strong poisons in isomerization catalysis. On the basis of both experimental data and theoretical calculations (DFT), a complete catalytic cycle for the isomerization of 2-propen-1-ol into propenal is described. The potential energy surfaces of the cycle have been explored at the B3LYP/6-311+G(d,p)//B3LYP/6-31G(d,p) + LAN2DZ level. The proposed mechanism involves the coordination of the oxygen atom of the allylic alcohol to the metal. The DFT energy profile is consistent with the experimental observation that the reaction only proceeds under heating. Calculations predict the catalytic cycle to be strongly exergonic, in full agreement with the high yields experimentally observed.

Full text of DOI:10.1021/ja054827a

Published on Web 01/06/2006
Bis(allyl)-Ruthenium(IV) Complexes as Highly Efficient
Catalysts for the Redox Isomerization of Allylic Alcohols into
Carbonyl Compounds in Organic and Aqueous Media: Scope,
Limitations, and Theoretical Analysis of the Mechanism
Victorio Cadierno,*,† Sergio E. Garc´ıa-Garrido,† Jose´ Gimeno,*,†
Adria´n Varela-AÄ lvarez,‡ and Jose´ A. Sordo*,‡
Contribution from the Departamento de Qu´ımica Orga´nica e Inorga´nica, Instituto UniVersitario
de Qu´ımica Organometa´lica “Enrique Moles” (Unidad Asociada al CSIC), and Laboratorio de
Qu´ımica Computacional, Departamento de Qu´ımica F´ısica y Anal´ıtica, Facultad de Qu´ımica,
UniVersidad de OViedo, Julia´n ClaVer´ıa 8, E-33006 OViedo, Principado de Asturias, Spain
Received July 19, 2005; E-mail: vcm@uniovi.es; jgh@uniovi.es; jasg@uniovi.es
Abstract: The catalytic activity of the bis(allyl)-ruthenium(IV) dimer [{Ru(η3:η3-C10H16)(µ-Cl)Cl}2] (C10H16
) 2,7-dimethylocta-2,6-diene-1,8-diyl) (1), and that of its mononuclear derivatives [Ru(η3:η3-C10H16)Cl2(L)]
(L ) CO, PR3, CNR, NCR) (2) and [Ru(η3:η3-C10H16)Cl(NCMe)2][SbF6] (3), in the redox isomerization of
allylic alcohols into carbonyl compounds, both in tetrahydrofuran and in water, is reported. In particular, a
variety of allylic alcohols have been quantitatively isomerized using [{Ru(η3:η3-C10H16)(µ-Cl)Cl}2] (1) as
catalyst, the reactions proceeding in all cases faster in water. Remarkably, complex 1 has been found to
be the most efficient catalyst reported to date for this particular transformation, leading to TOF and TON
values up to 62 500 h-1 and 1 500 000, respectively. Moreover, catalyst 1 can be recycled and is capable
of performing allylic alcohol isomerizations even in the presence of conjugated dienes, which are known to
be strong poisons in isomerization catalysis. On the basis of both experimental data and theoretical
calculations (DFT), a complete catalytic cycle for the isomerization of 2-propen-1-ol into propenal is
described. The potential energy surfaces of the cycle have been explored at the B3LYP/6-311+G(d,p)//
B3LYP/6-31G(d,p) + LAN2DZ level. The proposed mechanism involves the coordination of the oxygen
atom of the allylic alcohol to the metal. The DFT energy profile is consistent with the experimental observation
that the reaction only proceeds under heating. Calculations predict the catalytic cycle to be strongly
exergonic, in full agreement with the high yields experimentally observed.
Introduction
recovery of the organic products from an aqueous medium can
be easily achieved by phase separation. Simultaneously, the
New achievements in organic synthesis have appeared very
rapidly as part of the increasing academic and industrial interest
in fulfilling the fundamental principles of “Green Chemistry”.1
Among the most important challenges of modern chemistry,
organic reactions in water are of primary interest especially when
they are performed under catalytic conditions.2 As a matter of
fact, the development of metal-mediated catalytic transforma-
tions in aqueous media represents one of the cornerstones of
modern organometallic chemistry.3,4 Although the hydrophobic
character of most organic compounds may be a drawback, the
concept of atom economy, that is, all atoms of the reactant end
up in the final product, has emerged as a desirable goal in
chemical reactions.5
To address both issues, that is, atom economical organic
catalytic reactions in water, much effort has been done during
the past few years.3,4 Isomerization reactions are typical
examples of atom economical reactions because no byproducts
are generated. Among them, the isomerization of allylic alcohols
into the corresponding carbonyl compounds is an economically
(3) See, for example: (a) Kalck, P.; Monteil, F. AdV. Organomet. Chem. 1992,
34, 219-284. (b) Herrmann, W. A.; Kohlpaintner, C. W. Angew. Chem.,
Int. Ed. Engl. 1993, 32, 1524-1544. (c) Aqueous Organometallic Chemistry
and Catalysis; Horva´th, I. T., Joo´, F., Eds.; Kluwer: Dordrecht, 1995. (d)
Aqueous-Phase Organometallic Catalysis: Concepts and Applications;
Cornils, B., Herrmann, W. A., Eds.; Wiley-VCH: Weinheim, 1998. (e)
Hanson, B. E. Coord. Chem. ReV. 1999, 185-186, 795-807. (f) Joo´, F.
Aqueous Organometallic Catalysis; Kluwer: Dordrecht, 2001. (g) Pinault,
N.; Bruce, D. W. Coord. Chem. ReV. 2003, 241, 1-25.
(4) Thematic issues of journals have been devoted to the catalytic applications
of water-soluble organometallic compounds. See, for example: (a) J. Mol.
Catal. A: Chem. 1997, 116, 1-316. (b) Catal. Today 1998, 42, 371-478.
(c) AdV. Synth. Catal. 2002, 344, 219-451.
(5) See, for example: (a) Trost, B. M. Angew. Chem., Int. Ed. Engl. 1995, 34,
259-281 and references therein. (b) Trost, B. M. Acc. Chem. Res. 2002,
35, 695-705 and references therein.
† Departamento de Qu´ımica Orga´nica e Inorga´nica.
‡ Laboratorio de Qu´ımica Computacional.
(1) See, for example: (a) Anastas, P. T.; Warner, J. C. Green Chemistry Theory
and Practice; Oxford University Press: Oxford, 1998. (b) Matlack, A. S.
Introduction to Green Chemistry; Marcel Dekker: New York, 2001. (c)
Lancaster, M. In Handbook of Green Chemistry and Technology; Clark, J.
H., Macquarrie, D. J., Eds.; Blackwell Publishing: Abingdon, 2002. (d)
Lancaster, M. Green Chemistry: An Introductory Text; RSC Editions:
Cambridge, 2002.
(2) See, for example: (a) Li, C. J. Chem. ReV. 1993, 93, 2023-2035. (b)
Lubineau, A.; Auge, J.; Queneau, Y. Synthesis 1994, 741-760. (c) Li, C.
J.; Chan, T. H. Organic Reactions in Aqueous Media; John Wiley & Sons:
New York, 1997. (d) Lindstro¨m, U. M. Chem. ReV. 2002, 102, 2751-
2772. (e) Li, C. J. Chem. ReV. 2005, 105, 3095-3165.
9
1360
J. AM. CHEM. SOC. 2006, 128, 1360-1370
10.1021/ja054827a CCC: $33.50 © 2006 American Chemical Society

Bis(allyl)−Ruthenium(IV) Complexes as Catalysts
A R T I C L E S
Scheme 1. Transformation of Allylic Alcohols into Saturated
Carbonyl Compounds
catalyst for the isomerization of allylic alcohols and ethers in
pure aqueous solution.9 In addition, the catalytic systems [Rh-
(CO)2(µ-Cl)]2,10 mer-[RuCl3(DMSO)(phen)] (DMSO ) dimeth-
yl sulfoxide; phen ) 1,10-phenanthroline),11 cis,cis-[RuCl2-
(DMSO)2(phen)],11 and [Ru(acac)3]/phen/TsOH (acac ) acetyl-
acetonate; TsOH ) p-toluenesulfonic acid)12 have been used
for allylic alcohols isomerization in monophasic water/organic
solvent mixtures.
Remarkably, we have recently reported that the mononuclear
bis(allyl)-ruthenium(IV) complex [Ru(η3:η2:η3-C12H18)Cl2]
(C12H18 ) dodeca-2,6,10-triene-1,12-diyl) is an active catalyst
for this particular transformation, showing an extremely high
performance (TOF values up to 55 000 h-1; TON values up to
106).13 As far as we know, this complex represents the first
example of a Ru(IV) catalyst for this isomerization reaction.6
With these precedents in mind, and continuing with our studies
aimed at discovering new catalytic systems, in this paper we
report that the dinuclear bis(allyl)-ruthenium(IV) complex
[{Ru(η3:η3-C10H16)(µ-Cl)Cl}2] (C10H16 ) 2,7-dimethylocta-2,6-
diene-1,8-diyl) (1), as well as its mononuclear derivatives [Ru-
(η3:η3-C10H16)Cl2(L)] (L ) CO, PR3, CNR, NCR) (2) and
[Ru(η3:η3-C10H16)Cl(NCMe)2][SbF6] (3) (see Figure 1), are also
highly efficient catalysts for the isomerization of allylic alcohols
into carbonyl compounds in both organic and aqueous media.
In particular, the parent complex 1 provides the following
remarkable features: (i) it is the most efficient catalyst described
to date,6 (ii) its catalytic activity is maintained at very low
catalyst loadings and can be recycled with only partial deactiva-
tion after three or four runs, (iii) it is a readily available air-
stable solid,14 and (iv) it remains active in the presence of
isoprene in contrast to the loss of activity observed for almost
all of the known catalysts.15
Moreover, on the basis of both experimental data and
theoretical calculations, a complete catalytic cycle is described.
The most relevant feature of the mechanism is that it involves
the formation of a π-oxoallyl intermediate, in contrast to the
classical alkyl and π-allyl intermediates commonly proposed
for this isomerization reaction.6 Although the involvement of a
π-oxoallyl intermediate has been previously postulated by B.
M. Trost and co-workers starting from half-sandwich ruthenium-
(II) catalysts,16 no theoretical confirmation was provided. Herein,
we report for the first time a theoretical study showing the key
role of the oxygen atom of the allylic alcohols, which enables
important process because it provides a simple synthetic route
to very valuable raw materials in organic synthesis. Although
pioneering transformations were performed thermally by the
treatment with acids, bases, or by using heterogeneous catalysts,
new synthetic approaches based in homogeneous catalysis (see
path A in Scheme 1) are preferred because they generally
proceed with much more efficiency, improving also the
selectivities.6 This one-pot internal redox process represents a
useful and elegant shortcut to carbonyl compounds, which
otherwise would require a two-step sequence of oxidation and
reduction reactions (paths B or C in Scheme 1).6 Several group
6, 8, 9, and 10 metal complexes have been found to be active
catalysts for this transformation.6 In particular, the best perfor-
mances in terms of both selectivity, turnover frequencies (TOF),
and turnover numbers (TON) have been reported using iron,
ruthenium, and rhodium compounds.6
Surprisingly, the catalytic isomerization of allylic alcohols
into carbonyl compounds in aqueous media has been scarcely
studied.6 The reactions are usually performed under biphasic
(water/organic solvent) conditions using as catalysts transition-
metal complexes containing water-soluble ligands (either iso-
lated or generated in situ), which allows the catalyst recycling
by phase separation. These include: [Ni(COD)2]/DPPBTS
(COD ) 1,5-cyclooctadiene; DPPBTS ) tetrasulfonated 1,4-
bis(diphenylphosphino)butane),7a [RhCl(COD)(PAr3)] (PAr3 )
carboxylated triarylphosphines),7b RhCl3/TPPTS or Rh2(SO4)3/
TPPTS (TPPTS ) P(m-C6H4SO3Na)3),7c [Rh(COD)(SUL-
PHOS)] (SULPHOS ) -O3S-C6H4-CH2C(CH2PPh2)3),7d [Rh-
(COD)(TPPMP)2][BF4] (TPPMP ) Ph2P(p-C6H4PO3Na2)),7e
and [RuCl2(η6-arene){P(CH2OH)3}] (arene ) C6H6, C6Me6,
p-cymene).7f,8 The hexaaquoruthenium(II) complex [Ru(H2O)6]-
[Tos]2 (Tos ) p-toluenesulfonate) has also proven to be an active
(9) (a) McGrath, D. V.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1991,
113, 3611-3613. (b) Karlen, T.; Ludi, A. HelV. Chim. Acta 1992, 75,
1604-1606. (c) McGrath, D. V.; Grubbs, R. H. Organometallics 1994,
13, 224-235.
(10) Alper, H.; Hachem, K. J. Org. Chem. 1980, 45, 2269-2270.
(11) van der Drift, R. C.; Sprengers, J. W.; Bouwman, E.; Mul, W. P.; Kooijman,
H.; Spek, A. L.; Drent, E. Eur. J. Inorg. Chem. 2002, 2147-2155.
(12) Stunnenberg, F.; Niele, F. G. H.; Drent, E. Inorg. Chim. Acta 1994, 222,
225-233.
(13) Cadierno, V.; Garc´ıa-Garrido, S. E.; Gimeno, J. Chem. Commun. 2004,
232-233.
(6) For reviews on transition-metal catalyzed isomerization of allylic alcohols
into carbonyls, see: (a) Uma, R.; Cre´visy, C.; Gre´e, R. Chem. ReV. 2003,
103, 27-51. (b) van der Drift, R. C.; Bouwman, E.; Drent, E. J. Organomet.
Chem. 2002, 650, 1-24. (c) The asymmetric version of this reaction has
been recently reported: Ito, M.; Kitahara, S.; Ikariya, T. J. Am. Chem.
Soc. 2005, 127, 6172-6173.
(7) (a) Bricout, H.; Monflier, E.; Carpentier, J. F.; Mortreux, A. Eur. J. Inorg.
Chem. 1998, 1739-1744. (b) Schumann, H.; Ravindar, V.; Meltser, L.;
Baidossi, W.; Sasson, Y.; Blum, J. J. Mol. Catal. A 1997, 118, 55-61. (c)
de Bellefon, C.; Caravieilhes, S.; Kuntz, E. G. C. R. Acad. Sci., Ser. IIc:
Chim. 2000, 3, 607-614. (d) Bianchini, C.; Meli, A.; Oberhauser, W. New
J. Chem. 2001, 25, 11-12. (e) Knight, D. A.; Schull, T. L. Synth. Commun.
2003, 33, 827-831. (f) Cadierno, V.; Crochet, P.; Garc´ıa-Garrido, S. E.;
Gimeno, J. Dalton Trans. 2004, 3635-3641.
(8) We also note that a micromixer reactor has been used for the screening of
a series of Rh, Ir, Ru, and Pd complexes with a library of water-soluble
phosphine ligands. The isomerization of 1-hexen-3-ol into hexan-3-one
under biphasic water/n-heptane conditions was used as a model reaction:
(a) de Bellefon, C.; Tanchoux, N.; Caravieilhes, S.; Grenouillet, P.; Hessel,
V. Angew. Chem., Int. Ed. 2000, 39, 3442-3445. (b) Abdallah, R.; Ireland,
T.; de Bellefon, C. Chem.-Ing.-Tech. 2004, 76, 633-637.
(14) Complex 1 is easily obtained from the reaction of alcoholic solutions of
RuCl3‚nH2O with isoprene: (a) Porri, L.; Gallazzi, M. C.; Colombo, A.;
Allegra, G. Tetrahedron Lett. 1965, 47, 4187-4189. (b) Salzer, A.; Bauer,
A.; Podewils, F. In Synthetic Methods of Organo-metallic and Inorganic
Chemistry; Herrmann, W. A., Ed.; Thieme Verlag: Stuttgart, 2000; Vol.
9, pp 36-38. (c) Salzer, A.; Bauer, A.; Geyser, S.; Podewils, F.; Turpin,
G. C.; Ernst, R. D. Inorg. Synth. 2004, 34, 59-65.
(15) Most of the catalysts able to perform allylic alcohol isomerization fail in
the presence of conjugated dienes. See refs 6 and 11, and: van der Drift,
R. C.; Gagliardo, M.; Kooijman, H.; Spek, A. L.; Bouwman, E.; Drent, E.
J. Organomet. Chem. 2005, 690, 1044-1055.
(16) (a) Trost, B. M.; Kulawiec, R. J. Tetrahedron Lett. 1991, 32, 3039-3042.
(b) Trost, B. M.; Kulawiec, R. J. J. Am. Chem. Soc. 1993, 115, 2027-
2036. (c) van der Drift, R. C.; Vailati, M.; Bouwman, E.; Drent, E. J. Mol.
Catal. A 2000, 159, 163-177.
9
J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006 1361

A R T I C L E S
Cadierno et al.
Figure 1. Bis(allyl)-ruthenium(IV) catalysts used in this work.
to a solution of complex [{Ru(η3:η3-C10H16)(µ-Cl)Cl}2] (1) (0.616 g;
1 mmol) in 20 mL of dichloromethane. After the mixture was stirred
for 10 min, the solvent was removed under vacuum and the resulting
yellow solid residue was washed with hexanes (3 × 10 mL) and dried
in vacuo. 2b: Yield 91% (0.699 g). Anal. Calcd for RuC13H25Cl2P:
C, 40.63; H, 6.56. Found: C, 40.57; H, 6.54. IR (KBr, cm-1): ν 501
(w), 573 (w), 678 (w), 736 (w), 783 (w), 801 (w), 821 (w), 850 (m),
948 (vs), 1025 (w), 1095 (w), 1261 (w), 1285 (m), 1304 (w), 1315
(w), 1344 (w), 1384 (m), 1418 (m), 1456 (w), 1495 (w), 2853 (m),
the chelate coordination of the substrate to the metal favoring
the catalytic activity.
Experimental Section
The manipulations were performed under an atmosphere of dry
nitrogen using vacuum-line and standard Schlenk techniques. Solvents
were dried by standard methods and distilled under nitrogen before
use. All reagents were obtained from commercial suppliers and used
without further purification with the exception of compounds [{Ru-
(η3:η3-C10H16)(µ-Cl)Cl}2] (1),14 [{Ru(η6-arene)(µ-Cl)Cl}2] (arene )
C6H6, p-cymene, C6Me6),17 [RuCl2(PPh3)3],18 [Ru(η5-C9H7)Cl(PPh3)2],19
[Ru(η5-C5H5)Cl(PPh3)2],20 [Ru(η3:η3-C10H16)Cl2(L)] (L ) CO (2a),21a
PPh3 (2e),21a PiPr3 (2g),21e P(OMe)3 (2h),21c CNtBu (2i),21c NH2Ph (2j),21d
NCMe (2k)21b), and [Ru(η3:η3-C10H16)Cl(NCMe)2][SbF6] (3),21f,g which
were prepared by following the methods reported in the literature.
Infrared spectra were recorded on a Perkin-Elmer 1720-XFT spec-
trometer. The C and H analyses were carried out with a Perkin-Elmer
2400 microanalyzer. GC measurements were made on Hewlett-Packard
HP6890 equipment using a HP-INNOWAX cross-linked poly(ethylene
glycol) (30 m, 250 µm) or a Supelco Beta-Dex 120 (30 m, 250 µm)
column. GC/MS measurements were performed on Agilent 6890N
equipment coupled to a 5973 mass detector (70-eV electron impact
ionization) using a HP-1MS column. NMR spectra were recorded on
a Bruker DPX-300 instrument at 300 MHz (1H), 121.5 MHz (31P), or
75.4 MHz (13C) using SiMe4 or 85% H3PO4 as standards. DEPT
experiments have been carried out for all of the compounds reported.
The numbering for protons and carbons of the 2,7-dimethylocta-
2,6-diene-1,8-diyl skeleton is as follows:
1
2908 (m), 2977 (m). 31P{1H} NMR (CD2Cl2): δ -4.78 (s) ppm. H
2
NMR (CD2Cl2): δ 1.58 (d, 9H, JHP ) 10.3 Hz, PCH3), 2.15 (s, 6H,
3
CH3), 2.69 (m, 2H, H4 and H6), 2.97 (d, 2H, JHP ) 4.8 Hz, H2 and
H10), 3.39 (m, 2H, H5 and H7), 4.09 (d, 2H, 3JHP ) 9.1 Hz, H1 and H9),
4.93 (m, 2H, H3 and H8) ppm. 13C{1H} NMR (CD2Cl2): δ 16.90 (d,
1JCP ) 33.9 Hz, PCH3), 20.97 (s, CH3), 36.99 (s, C4 and C5), 63.70 (d,
2
2JCP ) 5.7 Hz, C1 and C8), 108.07 (d, JCP ) 10.2 Hz, C3 and C6),
2
122.08 (d, JCP ) 1.2 Hz, C2 and C7) ppm. 2c: Yield 89% (0.794 g).
Anal. Calcd for RuC18H27Cl2P: C, 48.44; H, 6.10. Found: C, 48.39;
H, 6.25. IR (KBr, cm-1): ν 493 (s), 573 (w), 679 (s), 694 (s), 712 (s),
750 (s), 787 (m), 846 (m), 914 (vs), 941 (vs), 964 (w), 1002 (w), 1029
(m), 1109 (m), 1183 (w), 1199 (w), 1279 (m), 1294 (m), 1312 (m),
1325 (w), 1343 (w), 1383 (s), 1436 (s), 1453 (m), 1496 (w), 2855 (m),
2908 (m), 3001 (m), 3073 (m). 31P{1H} NMR (CD2Cl2): δ 1.27 (s)
1
2
ppm. H NMR (CD2Cl2): δ 1.91 and 2.00 (d, 3H each, JHP ) 10.0
Hz, PCH3), 2.14 (s, 6H, CH3), 2.70 (m, 2H, H4 and H6), 2.99 (d, 2H,
3JHP ) 4.2 Hz, H2 and H10), 3.43 (m, 2H, H5 and H7), 3.70 (d, 2H, 3JHP
) 9.7 Hz, H1 and H9), 5.02 (m, 2H, H3 and H8), 7.45-7.65 (m, 5H,
Ph) ppm. 13C{1H} NMR (CD2Cl2): δ 14.42 and 15.30 (d, 1JCP ) 35.5
2
Hz, PCH3), 21.29 (s, CH3), 37.30 (s, C4 and C5), 66.88 (d, JCP ) 5.3
2
Hz, C1 and C8), 107.80 (d, JCP ) 9.5 Hz, C3 and C6), 123.52 (s, C2
and C7), 128.65-140.41 (m, Ph) ppm. 2d: Yield 93% (0.946 g). Anal.
Calcd for RuC23H29Cl2P: C, 54.33; H, 5.75. Found: C, 54.36; H, 5.54.
IR (KBr, cm-1): ν 477 (m), 514 (s), 524 (m), 700 (s), 736 (s), 743 (s),
756 (s), 787 (w), 801 (w), 821 (w), 857 (w), 890 (vs), 969 (w), 1000
(w), 1023 (m), 1096 (w), 1167 (w), 1198 (w), 1286 (m), 1313 (w),
1343 (m), 1381 (m), 1417 (m), 1438 (s), 1486 (w), 2860 (m), 2921
(m), 2989 (m), 3003 (m), 3056 (m). 31P{1H} NMR (CD2Cl2): δ 4.42
(s) ppm. 1H NMR (CD2Cl2): δ 2.14 (s, 6H, CH3), 2.18 (s, 3H, PCH3),
2.63 (m, 2H, H4 and H6), 3.08 (d, 2H, 3JHP ) 2.9 Hz, H2 and H10), 3.43
(m, 2H, H5 and H7), 3.96 (d, 2H, 3JHP ) 10.0 Hz, H1 and H9), 5.14 (m,
2H, H3 and H8), 7.35-7.96 (m, 10H, Ph) ppm. 13C{1H} NMR (CD2-
Cl2): δ 14.11 (d, 1JCP ) 32.0 Hz, PCH3), 21.03 (s, CH3), 37.24 (s, C4
Synthesis of Complexes [Ru(η3:η3-C10H16)Cl2(PR3)] (PR3 ) PMe3
(2b), PMe2Ph (2c), PMePh2 (2d), P(p-C6H4OMe)3 (2f)). The corre-
sponding phosphine ligand (2 mmol) was added, at room temperature,
(17) (a) Bennett, M. A.; Smith, A. K. J. Chem. Soc., Dalton Trans. 1974, 233-
241. (b) Bennett, M. A.; Huang, T. N.; Matheson, T. W.; Smith, A. K.
Inorg. Synth. 1982, 21, 74-78.
2
and C5), 67.42 (s, C1 and C8), 108.22 (d, JCP ) 8.2 Hz, C3 and C6),
123.42 (s, C2 and C7), 128.10-138.00 (m, Ph) ppm. 2f: Yield 92%
(1.215 g). Anal. Calcd for RuC31H37O3Cl2P: C, 56.37; H, 5.65.
Found: C, 56.64; H, 5.86. IR (KBr, cm-1): ν 501 (w), 541 (s), 613
(w), 622 (w), 642 (w), 718 (w), 798 (s), 826 (m), 862 (w), 960 (w),
1028 (m), 1088 (m), 1112 (w), 1182 (s), 1252 (vs), 1286 (s), 1382
(w), 1409 (w), 1441 (m), 1458 (m), 1500 (s), 2835 (m), 2853 (m),
2910 (m), 2957 (m), 2999 (m), 3066 (m). 31P{1H} NMR (CD2Cl2): δ
22.39 (s) ppm. 1H NMR (CD2Cl2): δ 2.24 (s, 6H, CH3), 2.63 (m, 2H,
H4 and H6), 3.02 (s, 2H, H2 and H10), 3.43 (m, 2H, H5 and H7), 3.79 (s,
(18) Hallman, P. S.; Stephenson, T. A.; Wilkinson, G. Inorg. Synth. 1970, 12,
237-240.
(19) Oro, L. A.; Ciriano, M. A.; Campo, M.; Foces-Foces, C.; Cano, F. H. J.
Organomet. Chem. 1985, 289, 117-131.
(20) (a) Bruce, M. I.; Hameister, C.; Swincer, A. G.; Wallis, R. C. Inorg. Synth.
1982, 21, 78-84. (b) Bruce, M. I.; Hameister, C.; Swincer, A. G.; Wallis,
R. C. Inorg. Synth. 1990, 28, 270-272.
(21) The synthesis of complexes 2a, 2e, 2g-k, and 3 has been previously
reported: (a) Head, R. A.; Nixon, J. F.; Swain, J. R.; Woodard, C. M. J.
Organomet. Chem. 1974, 76, 393-400. (b) Cox, D. N.; Roulet, R. Inorg.
Chem. 1990, 29, 1360-1365. (c) Cox, D. N.; Small, R. W. H.; Roulet, R.
J. Chem. Soc., Dalton Trans. 1991, 2013-2018. (d) Steed, J. W.; Tocher,
D. A. J. Organomet. Chem. 1994, 471, 221-228. (e) Werner, H.; Stu¨er,
W.; Jung, S.; Weberndo¨rfer, B.; Wolf, J. Eur. J. Inorg. Chem. 2002, 1076-
1080. (f) Herrmann, W. A.; Schattenmann, W. C.; Nuyken, O.; Glander,
S. C. Angew. Chem., Int. Ed. Engl. 1996, 35, 1087-1088. (g) Glander, S.
C.; Nuyken, O.; Schattenmann, W. C.; Herrmann, W. A. Macromol. Symp.
1998, 127, 67-75.
3
9H, OCH3), 4.41 (d, 2H, JHP ) 9.4 Hz, H1 and H9), 5.20 (m, 2H, H3
and H8), 6.85-7.65 (m, 12H, Ph) ppm. 13C{1H} NMR (CD2Cl2): δ
20.80 (s, CH3), 37.20 (s, C4 and C5), 55.61 (s, OCH3), 67.42 (d, 2JCP
4.7 Hz, C1 and C8), 108.59 (d, JCP ) 10.4 Hz, C3 and C6), 113.10-
)
2
160.82 (m, C6H4, C2 and C7) ppm.
9
1362 J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006

Bis(allyl)−Ruthenium(IV) Complexes as Catalysts
A R T I C L E S
Table 1. Isomerization of 1-Octen-3-ol into Octan-3-one Catalyzed
by the Dinuclear Bis(allyl)-Ruthenium(IV) Complex 1a
General Procedure for the Catalytic Isomerization of Allylic
Alcohols into Carbonyl Compounds. In a Schlenk tube fitted with a
condenser, the corresponding allylic alcohol, ruthenium catalyst precur-
sor, and cocatalyst (if required) were dissolved in THF or water (20
mL), and the reaction mixture was stirred at 75 °C for the indicated
time. The course of the reaction was monitored by regular sampling
and analysis by gas chromatography. The identity of the resulting
saturated carbonyl compounds was assessed by comparison with
commercially available pure samples (Aldrich Chemical Co. or Acros
Organics) and by their fragmentation in GC/MS.
Isomerization of 1-Octen-3-ol into Octan-3-one under Prepara-
tive Conditions. Under nitrogen atmosphere, 1-octen-3-ol (6.3 mL,
40 mmol), dimer 1 (0.025 g, 0.04 mmol), Cs2CO3 (0.053 g, 0.163
mmol), and water (200 mL) were introduced in a Schlenk flask, and
the reaction mixture was stirred at 75 °C for 20 min (quantitative
conversion by GC). The resulting aqueous solution was then saturated
with NaCl and extracted with dichloromethane (2 × 50 mL). The
combined organic extracts were dried over MgSO4, concentrated, and
purified by column chromatography over silica gel, using diethyl ether
as eluent, to give 4.82 g (37.6 mmol) of analytically pure octan-3-one
(94% yield).
Theoretical Methods. The potential energy surfaces (PESs) corre-
sponding to the processes involved in the redox isomerization of allyl
alcohols into carbonyl compounds catalyzed by the Ru(IV) complexes
have been explored using density functional theory (DFT)22 with the
Becke 3-parameter Lee-Yang-Parr23,24 (B3LYP) hybrid functional.
We have carried out all of the geometry optimizations using Pople’s
6-31G(d,p) basis sets for H, C, O, Cl atoms25 and the LANL2DZ
effective core potential26 for ruthenium. All of the geometrical
parameters were fully optimized, and all of the structures located on
the PESs were characterized as minima or transition structures by
computing the corresponding Hessian matrices, examining the number
of imaginary frequencies from them. The energy predictions were
improved by carrying out single-point calculations with Pople’s valence
triple-zeta 6-311+G(d,p) basis set.27
1
c
entry
cocatalyst
yield (%)b
time
TOF (h-
)
1
2
3
4
5
6
7
8
none
96
100
100
100
100
100
100
62
24 h
3 h
20
167
545
333
857
3000
3000
17
Li2CO3 (0.2 mol %)
Na2CO3 (0.2 mol %)
K2CO3 (0.2 mol %)
Cs2CO3 (0.2 mol %)
Cs2CO3 (0.4 mol %)
Cs2CO3 (2 mol %)
AgSbF6 (0.2 mol %)
55 min
90 min
35 min
10 min
10 min
18 h
a All of the reactions were performed under N2 atmosphere at 75 °C
using 4 mmol of 1-octen-3-ol (0.2 M in THF). Substrate/Ru ratio: 500:1.
b Yield of octan-3-one determined by GC. c Turnover frequencies ((mol
product/mol Ru)/time) were calculated at the time indicated in each case.
catalytic amount of base as cocatalyst.30 Among all alkaline
carbonates M2CO3 (M ) Li, Na, K, and Cs) tested, the best
results were obtained with Cs2CO3 (entry 5 vs entries 2-4; in
all cases, a 1:1 Ru/base ratio was used), which leads to the
quantitative transformation of the unsaturated alcohol into the
saturated ketone in only 35 min (TOF ) 857 h-1). We have
also screened the optimal Ru/Cs2CO3 ratio performing experi-
ments with 1:1 to 1:10 molar ratios, and we have observed that
a notable improvement occurs by increasing the Cs2CO3/Ru mol
ratio up to 2 (entry 6 (TOF ) 3000 h-1) vs 5 (TOF ) 857
h-1)). The addition of an excess of base has no influence on
the reaction rate (entry 7 vs 6). To check whether Cs2CO3 acts
as base or chloride abstractor, the isomerization of 1-octen-3-
ol by complex 1 was studied using AgSbF6 as cocatalyst under
the same reaction conditions (substrate/Ru/AgSbF6 ratio: 500:
1:1). In this case, only 62% of conversion was attained after 18
h of reflux (entry 8; to be compared to entry 5), indicating clearly
that Cs2CO3 acts mainly as a base.
Solvation effects (water and tetrahydrofuran) have been estimated
by using a continuum solvent description: the so-called polarized
continuum model (PCM).28
In accord with these results, a substrate/Ru/Cs2CO3 ratio of
500:1:2 was considered as the optimal reaction conditions for
the rest of our catalytic studies. Remarkably, under these reaction
conditions, the dimeric complex [{Ru(η3:η3-C10H16)(µ-Cl)Cl}2]
(1) is by far much more active (TOF ) 3000 h-1) than the
classical ruthenium(II) catalysts [{Ru(η6-arene)(µ-Cl)Cl}2] (are-
ne ) C6H6,7f p-cymene,30a C6Me67f), [RuCl2(PPh3)3],30 [Ru(η5-
C9H7)Cl(PPh3)2],16b and [Ru(η5-C5H5)Cl(PPh3)2]16 (TOF )
1-500 h-1; entries 2-7 vs entry 1 in Table 2),31 its efficiency
being even higher than that shown by the mononuclear
ruthenium(IV) complex [Ru(η3:η2:η3-C12H18)Cl2] recently re-
ported by us (TOF ) 3000 vs 429 h-1; entry 1 vs entry 8).13
It is also worth mentioning that the exceptional high activity
of complex 1 is retained at lower catalyst loadings. For example,
using 10-4 or 6.67 × 10-5 mol % of Ru, 1-octen-3-ol (0.2 M
in THF; substrate/Ru/Cs2CO3 ratio: 1 000 000:1:2 or 1 500 000:
1:2) can be quantitatively transformed into octan-3-one within
16 (TOF ) 62 500 h-1; entry 9 in Table 2) or 36 h (TOF )
41 666 h-1; entry 10 in Table 2), leading to the highest turnover
The Gaussian98 and Gaussian03 packages of programs29 were used
to carry out the calculations.
Results and Discussion
Catalytic Isomerization of Allylic Alcohols into Carbonyl
Compounds in THF. First, we checked the catalytic activity
of dimer [{Ru(η3:η3-C10H16)(µ-Cl)Cl}2] (1) in the isomerization
of a simple model, 1-octen-3-ol (see Table 1). Thus, when a
0.2 M THF solution of 1-octen-3-ol was refluxed for 24 h with
a catalytic amount of 1 (0.2 mol % in Ru), octan-3-one was
formed in 96% yield (see entry 1 in Table 1). The rate of this
isomerization process can be enhanced upon addition of a
(22) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density Functional
Theory, 2nd ed.; Wiley-VCH: Weinheim, 2000.
(23) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789.
(24) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(25) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab initio Molecular
Orbital Theory; Wiley: New York, 1986.
(26) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270-283. (b) Wadt,
W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284-298. (c) Hay, P. J.; Wadt,
W. R. J. Chem. Phys. 1985, 82, 299-310.
(27) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980,
72, 650-654.
(28) (a) Cance`s, E.; Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 107, 3032-
3041. (b) Mennucci, B.; Cance´s, E.; Tomasi, J. J. Phys. Chem. B 1997,
101, 10506-10517. (c) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J.
Chem. Phys. 2001, 114, 5691-5701. (d) Cossi, M.; Scalmani, G.; Rega,
N.; Barone, V. J. Chem. Phys. 2002, 117, 43-54.
(29) (a) Frisch, M. J.; et al. Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh,
PA, 1998. (b) Frisch, M. J.; et al. Gaussian 03, revision C.01; Gaussian,
Inc.: Wallingford, CT, 2004.
(30) Improvement of catalytic activity upon addition of a base as cocatalyst
has been previously observed. See, for example: (a) Ba¨ckvall, J. E.;
Andreasson, U. Tetrahedron Lett. 1993, 34, 5459-5462. (b) Uma, R.;
Davies, M. K.; Cre´visy, C.; Gre´e, R. Eur. J. Org. Chem. 2001, 3141-
3146. See also ref 13.
(31) TOF values of 285, 33, and 8 h-1 have been previously reported for the
isomerization of 1-octen-3-ol by using [RuCl2(PPh3)3], [{Ru(η6-p-cymene)-
(µ-Cl)Cl}2], and [Ru(η5-C5H5)Cl(PPh3)2], respectively, as catalysts and 3
equiv of K2CO3 as cocatalyst. See ref 30a.
9
J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006 1363

A R T I C L E S
Cadierno et al.
Table 2. Ruthenium-Catalyzed Isomerization of 1-Octen-3-ol into
Table 3. Isomerization of 1-Octen-3-ol into Octan-3-one Catalyzed
by Mononuclear (2,7-Dimethylocta-2,6-diene-1,8-diyl)-
Ruthenium(IV) Complexesa
Octan-3-onea
1
c
entry
catalyst
yield (%)b
time
TOF (h-
)
yield
(%)b
TOF
1 c
(h-
)
1
2
3
4
5
6
7
8
[{Ru(η3:η3-C10H16)(µ-Cl)Cl}2] (1)
[{Ru(η6-C6H6)(µ-Cl)Cl}2]
[{Ru(η6-p-cymene)(µ-Cl)Cl}2]
[{Ru(η6-C6Me6)(µ-Cl)Cl}2]
[RuCl2(PPh3)3]
100
100
100
100
92
38
1
100
100
100
10 min
60 min
90 min
4 h
22 h
22 h
22 h
70 min
16 h
3000
500
333
125
21
9
<1
429
entry
catalyst
time
1
2
3
4
5
6
7
8
9
[Ru(η3:η3-C10H16)Cl2(CO)] (2a)
[Ru(η3:η3-C10H16)Cl2(PMe3)] (2b)
[Ru(η3:η3-C10H16)Cl2(PMe2Ph)] (2c)
[Ru(η3:η3-C10H16)Cl2(PMePh2)] (2d)
[Ru(η3:η3-C10H16)Cl2(PPh3)] (2e)
100
100
100
100 8.5 h
100 7.5 h
85 min
29 h
29 h
353
17
17
59
67
[Ru(η5-C9H7)Cl(PPh3)2]
[Ru(η5-C5H5)Cl(PPh3)2]
[Ru(η3:η2:η3-C12H18)Cl2]
[Ru(η3:η3-C10H16)Cl2{P(p-C6H4OMe)3}] (2f) 100 6.5 h
77
[Ru(η3:η3-C10H16)Cl2(PiPr3)] (2g)
[Ru(η3:η3-C10H16)Cl2{P(OMe)3}] (2h)
[Ru(η3:η3-C10H16)Cl2(CNtBu)] (2i)
[Ru(η3:η3-C10H16)Cl2(NH2Ph)] (2j)
[Ru(η3:η3-C10H16)Cl2(NCMe)] (2k)
[Ru(η3:η3-C10H16)Cl(NCMe)2][SbF6] (3)
100
100
100
100
100
100
40 min
90 min
4 h
750
333
125
500
9d [{Ru(η3:η3-C10H16)(µ-Cl)Cl}2] (1)
10e [{Ru(η3:η3-C10H16)(µ-Cl)Cl}2] (1)
62 500
41 666
36 h
10
11
12
60 min
a All of the reactions were performed under N2 atmosphere at 75 °C
using 4 mmol of 1-octen-3-ol (0.2 M in THF). Substrate/Ru/Cs2CO3 ratio:
500:1:2. b Yield of octan-3-one determined by GC. c Turnover frequencies
((mol product/mol Ru)/time) were calculated at the time indicated in each
case. d Reaction performed with a substrate/Ru/Cs2CO3 ratio of 1 000 000:
1:2. e Reaction performed with a substrate/Ru/Cs2CO3 ratio of 1 500 000:
1:2.
12 min 2500
12 min 2500
a All of the reactions were performed under N2 atmosphere at 75 °C
using 4 mmol of 1-octen-3-ol (0.2 M in THF). Substrate/Ru/Cs2CO3 ratio:
500:1:2. b Yield of octan-3-one determined by GC. c Turnover frequencies
((mol product/mol Ru)/time) were calculated at the time indicated in each
case.
Scheme 2. Synthesis of the Mononuclear
Bis(allyl)-Ruthenium(IV) Complexes 2 and 3
saturated ketone in quantitative yield when compared to dimer
1 (see entries 1-10; TOF ) 17-750 h-1 to be compared to
3000 h-1 using 1). Only complexes [Ru(η3:η3-C10H16)Cl2-
(NCMe)] (2k) and [Ru(η3:η3-C10H16)Cl(NCMe)2][SbF6] (3),
containing labile acetonitrile ligands, have shown a catalytic
performance comparable to that of 1 (entries 11 and 12; TOF
) 2500 h-1). This seems to indicate that dimer 1 acts as a
catalyst precursor of an active mononuclear species “[Ru(η3:
η3-C10H16)Cl]”.
Catalytic Isomerization of Allylic Alcohols into Carbonyl
Compounds in Water. Several bis(allyl)-ruthenium(IV) com-
plexes derived from 1 are prone to coordinate water, and some
of them show a significant solubility in aqueous media (e.g.,
[Ru(η3:η3-C10H16)(H2O)(κ2-O,O-CH3CO2)][BF4], [Ru(η3:η3-
C10H16)(H2O)(κ1-O-CF3SO3)2][BF4], or [Ru(η3:η3-C10H16)(κ3-
N,N,N-terpy)][BF4]2 (terpy ) 2,2’:6’,2’’-terpyridine) among
others).34 Although the hydrolitic sensitivity of metal-carbon
bonds has been a major drawback of organometallic chemistry,
the remarkable stability of these species toward water is
explained on the basis of the high electronegativity of ruthenium
in the +4 oxidation state. This fact confers a greater covalency
to the resulting metal-carbon bonds, which consequently
become quite resistant to hydrolitic conditions.35
number values (TON ) 106 and 1.5 × 106, respectively)
reported to date for this catalytic transformation.6,32
The catalytic activity of the dimeric complex 1 raised the
question whether the active catalyst is a dinuclear or mono-
nuclear species. In this respect, the catalytic activity of a series
of mononuclear derivatives [Ru(η3:η3-C10H16)Cl2(L)] (2a-k)
and [Ru(η3:η3-C10H16)Cl(NCMe)2][SbF6] (3) (see Scheme 2),
in the isomerization of 1-octen-3-ol into octan-3-one, was
studied (results are summarized in Table 3).21,33 Although all
of the complexes tested have been found to be active catalysts,
longer reaction times are in general required to obtain the
(33) The coordination chemistry of the dimeric Ru(IV) complex [{Ru(η3:η3-
C10H16)(µ-Cl)Cl}2] (1), although much less developed, is clearly related to
that of the classical arene-Ru(II) dimers [{Ru(η6-arene)(µ-Cl)Cl}2],
allowing the preparation of mononuclear adducts via chloride bridges
cleavage and abstraction. For a review on the chemistry of 1, see: (a)
Cadierno, V.; Crochet, P.; Garc´ıa-Garrido, S. E.; Gimeno, J. Curr. Org.
Chem. 2006, 10, 165-183. For reviews on the chemistry of [{Ru(η6-arene)-
(µ-Cl)Cl}2] dimers, see: (b) Le Bozec, H.; Touchard, D.; Dixneuf, P. H.
AdV. Organomet. Chem. 1989, 29, 163-247. (c) Bennett, M. A. In
ComprehensiVe Organometallic Chemistry II; Abel, E. W., Stone, F. G.
A., Wilkinson, G., Eds.; Pergamon Press: Oxford, 1995; Vol. 7, pp 549-
602. (d) Bennett, M. A. Coord. Chem. ReV. 1997, 166, 225-254. (e) Pigge,
F. C.; Coniglio, J. J. Curr. Org. Chem. 2001, 5, 757-784.
(32) (a) TON values up to 3000 (TOF up to 36 000 h-1) have been reported for
the isomerization of 2-propen-1-ol into propionaldehyde using complexes
[Ru(η5-C5H5)(PR3)(NCMe)2][PF6] (PR3 ) PPh3, PCy3) as catalysts: Slugovc,
C.; Ru¨ba, E.; Schmid, R.; Kirchner, K. Organometallics 1999, 18, 4230-
4233. (b) TON values up to 6000 have also been reported for the
isomerization of 3-buten-2-ol into butan-2-one using complex [Ru(η5-C5H5)-
Cl(PPh3)2] as catalyst and AgOTs (silver p-toluenesulfonate) as cocatalyst
(a TOF value > 200 000 h-1 has been claimed): see ref 16c. (c) We have
reported TON values up to 106 (TOF up to 55 000 h-1) for the isomerization
of 1-octen-3-ol into octan-3-one using complex [Ru(η3:η2:η3-C12H18)Cl2]
as catalyst and Cs2CO3 as cocatalyst: see ref 13.
(34) See, for example: (a) Sommerer, S. O.; Palenik, G. J. Organometallics
1991, 10, 1223-1224. (b) Steed, J. W.; Tocher, D. A. Polyhedron 1992,
11, 2729-2737. (c) Kavanagh, B.; Steed, J. W.; Tocher, D. A. J. Chem.
Soc., Dalton Trans. 1993, 327-335. (d) Belchem, G.; Schreiber, D. H. A.;
Steed, J. W.; Tocher, D. A. Inorg. Chim. Acta 1994, 218, 81-87. (e) Wache,
S. J. Organomet. Chem. 1995, 494, 235-240.
(35) The application of high oxidation state organometallic derivatives in aqueous
catalysis has been recently highlighted: Poli, R. Chem.-Eur. J. 2004, 10,
332-341 and references therein.
9
1364 J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006

Bis(allyl)−Ruthenium(IV) Complexes as Catalysts
A R T I C L E S
Table 4. Isomerization of Various Allylic Alcohols Catalyzed by the Bis(allyl)-Ruthenium(IV) Complex 1 in THF and Watera
a All of the reactions were performed under N2 atmosphere at 75 °C using 4 mmol of the corresponding allylic alcohol (0.2 M in THF or H2O). When
Cs2CO3 was used as cocatalyst, it was added in a 2:1 molar ratio with respect to Ru. b Yield of the corresponding aldehyde or ketone determined by GC.
c Turnover frequencies ((mol product/mol Ru)/time) were calculated at the time indicated in each case. d 91% yield after 31 h if the reaction is performed
at 90 °C. e 89% yield after 40 h if the reaction is performed at 90 °C.
Taking advantage of this property, we decided to study the
isomerization of 1-octen-3-ol catalyzed by complex 1 in aqueous
medium. Remarkably, the reaction proceeds at higher rates than
those observed in THF either in the presence or in the absence
of Cs2CO3. For instance, in the absence of Cs2CO3, octan-3-
one was quantitatively obtained in only 40 min versus 96% yield
after 24 h in THF (see entry 1 in Table 4). These observations
seem to suggest that the higher polarity of water versus THF
favors the dissociation of the chloride ligands in complex 1 as
well as the deprotonation of the alcohol to form the correspond-
ing oxo-allyl anion (see the theoretical analysis section).
Noteworthy, the reaction profile of this isomerization process
is characterized by an induction period of several minutes (see
Figure 2), indicating the slow formation of the real active species
in the catalytic cycle.36 Moreover, deactivation of the active
Figure 2. Isomerization of 1-octen-3-ol into octan-3-one catalyzed by
complex 1 in water and in the absence of Cs2CO3 (reaction performed under
species does not occur after total consumption of the allylic
alcohol. Thus, after a regular catalytic test, no induction period
was observed upon addition of a second charge of substrate
(Figure 2).
Catalyst 1 is also very efficient in the isomerization of a
variety of allylic alcohols, proving the wide scope of this
catalytic transformation (results are summarized in Table 4).
As a general trend, the isomerization process was found to take
place faster in water than in THF (see columns 7 and 8 vs 5
and 6, respectively), and, for most of the substrates tested, a
quantitative formation of the saturated carbonyl compound is
achieved without Cs2CO3 (see column 7 in Table 4). As far as
N2 atmosphere at 75 °C using 4 mmol of 1-octen-3-ol (0.2 M in H2O)).
Yield of octan-3-one as a function of time (b, first charge; 0, second
charge).
the allylic alcohols are concerned, there is a strong dependence
upon the substitution of the carbon-carbon double bond as
previously observed with other catalytic systems.37 Thus, the
monosubstituted alcohols 1-octen-3-ol, 1-hepten-3-ol, 1-hexen-
3-ol, 1-penten-3-ol, and 3-buten-2-ol (see entries 1-5) are
readily isomerized in aqueous medium using a ruthenium charge
of 0.2 mol %, leading to TOF values of 750-3000 h-1 without
Cs2CO3 or 1500-2000 h-1 with Cs2CO3. Efficient transforma-
(37) It is well known that the isomerization of allylic alcohols becomes more
difficult as the number of substituents increases on the CdC double bond.
See ref 6.
(36) Induction periods have been systematically observed in all catalytic reactions
regardless of the solvent (THF or water) or the presence of cocatalysts.
9
J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006 1365

A R T I C L E S
Cadierno et al.
Table 5. Isomerization of Allylic Alcohols Catalyzed by Complex
1: Catalyst Recyclinga
tion of allyl alcohol into propanal has also been observed using
1 mol % of Ru (entry 6; TOF ) 300 h-1 in water).38 The high
catalytic activity of complex 1 at a lower loading (10-4 mol %
of Ru) has also been confirmed for the isomerization of 3-buten-
2-ol into butan-2-one in water and in the absence of Cs2CO3
(100% yield in 24 h, TON ) 106, TOF ) 41 667 h-1). In
contrast, when 1,1- and 1,2-disubstituted allylic alcohols are
used, longer reaction times and higher Ru loadings (5-10 mol
%) are required to achieve total conversions (entries 8-12; TOF
) 1-20 h-1 in water). In the case of trisubstituted allylic
alcohols such as 3-methyl-2-buten-1-ol (entry 13), complex 1
was found to be almost inactive at 75 °C. Thus, after 40 h,
only 32% yield of 3-methyl-butyraldehyde could be obtained
using 1 (10 mol % of Ru) in water and in the presence of Cs2-
CO3. Nevertheless, with the reactions performed at a higher
temperature (90 °C), 3-methyl-butyraldehyde could be obtained
in 89% yield after 40 h.39
In contrast to almost all known catalysts for allylic alcohols
isomerization, which fail in the presence of conjugated dienes,15
catalyst 1 remains active in the presence of isoprene. As an
example, in the presence of 15 mmol of isoprene, 4 mmol of
1-octen-3-ol (0.2 M in water) can be quantitatively isomerized
into octan-3-one within 170 min using 0.1 mol % of 1 (0.2 mol
% of Ru; without Cs2CO3; TOF ) 176 h-1 to be compared to
entry 1 in Table 4).40 Nevertheless, it should be noted that the
isoprene poisoning observed for dimer 1 is much more severe
than that previously observed by us for the mononuclear bis-
(allyl)-ruthenium(IV) complex [Ru(η3:η2:η3-C12H18)Cl2] (C12H18
) dodeca-2,6,10-triene-1,12-diyl).13 Two major deactivation
routes have been proposed: (i) the preferential coordination of
the diene versus the substrate, and (ii) the formation of stable
π-allyl complexes after reaction of the diene with metal-hydride
intermediates.11,15 Apparently, coordination of isoprene to the
Ru(IV) unit [Ru(η3:η3-C10H16)Cl2], via chloride bridges cleavage
in dimer 1, seems to be favored versus isoprene coordination
to [Ru(η3:η3-C12H18)Cl2], via displacement of the olefinic Cd
C bond of the dodeca-2,6,10-triene-1,12-diyl chain in the
mononuclear complex [Ru(η3:η2:η3-C12H18)Cl2].
1
c
entry
substrate
cycle
yield (%)b
time
TOF (h-
)
TONd
1
n ) 0
1
2
3
4
1
2
3
4
1
2
3
4
5
100
100
100
18
100
100
100
15
100
100
100
100
7
20 min
20 min
35 min
24 h
15 min
25 min
2.5 h
1500
1500
857
4
2000
1200
200
3
2000
1500
1200
600
1
500
1000
1500
1590
500
1000
1500
1575
500
1000
1500
2000
2035
2
3
n ) 1
n ) 2
24 h
15 min
20 min
25 min
50 min
24 h
a All of the reactions were performed under N2 atmosphere at 75 °C
using 4 mmol of the corresponding allylic alcohol (0.2 M in water).
Substrate/Ru/Cs2CO3 ratio: 500:1:2. b Yield of the corresponding ketone
determined by GC. c Turnover frequencies ((mol product/mol Ru)/time) were
calculated at the time indicated in each case. d Cumulative TON values.
can be performed in a preparative scale (a representative
example is shown in the Experimental Section), the basic
research presented here is believed to be of interest to synthetic
chemists and to find wide applications in different synthetic
programs soon.
Mechanistic Aspects. From a mechanistic point of view, the
catalytic isomerization of allylic alcohols is based on the well-
known ability of transition-metal complexes to migrate carbon-
carbon double bonds.41 Thus, in the case of allylic alcohols,
such a migration generates an enol intermediate, which readily
tautomerizes into the thermodinamically more stable carbonyl
compound. Over the years, two different mechanisms have been
proposed for this isomerization reaction (A and B in Scheme
3).6
(i) The Alkyl Mechanism. In this case, the catalyst is a
metal-hydride complex that can be either isolated or generated
in situ. The first step in the catalytic cycle involves the
π-coordination of the allylic alcohol to the metal through the
CdC bond. Subsequent insertion of the alkene moiety into the
M-H bond generates metal-alkyl species, which evolve, via
â-hydrogen elimination, into the corresponding η2-enol complex.
In the final step, the decomplexation of the enol regenerates
the starting metal-hydride catalyst. Significantly, in this mech-
anism, an intermolecular hydrogen transfer takes place. R. H.
Grubbs and co-workers have demonstrated, using labeled
compounds, that this mechanism is operative when [Ru(H2O)6]-
[Tos]2 is used as precatalyst.9c
We also note that catalyst 1 can be recycled at least three or
four times after the isomerization of 3-buten-2-ol, 1-penten-3-
ol, and 1-hexen-3-ol (0.2 M in water; 0.2 mol % of Ru and 0.4
mol % of Cs2CO3) into the corresponding ketones, which can
be easily separated from the aqueous solution by fractional
distillation (see Table 5). This feature, along with the activity
observed in the presence of isoprene, is an additional proof of
the utility of these catalysts, which show an unprecedented
performance for the catalytic isomerization of allylic alcohols
in water. Moreover, because all of these isomerization reactions
(ii) The η3-Allyl Mechanism. The catalytic cycle starts again
with the π-coordination of the alkene moiety to an unsaturated
metal fragment that does not contain a M-H bond. Formal
oxidative addition of one of the C-H single bonds of the alkene
generates a π-allyl metal-hydride intermediate, which evolves,
by reductive elimination, into the corresponding η2-enol com-
plex. As in the previous mechanism, decoordination regenerates
(38) In contrast, due probably to steric reasons, the isomerization of 1-phenyl-
2-propen-1-ol requires at least 5 mol % of Ru to achieve 100% conversion
in reasonable reaction times (TOF values up to 80 h-1; see entry 7 in Table
4).
(39) A systematic study on the influence of the temperature (from 40 to 100
°C) on the catalytic activity of complex 1 has been carried out using the
isomerization of 1-octen-3-ol into octan-3-one as model reaction (0.2 M
in water; 0.2 mol % of Ru; without Cs2CO3). Selected results are as
follows: T ) 60 °C, 88% yield after 24 h (TOF ) 18 h-1); T ) 70 °C,
100% yield after 1.5 h (TOF ) 333 h-1); T ) 80 °C, 100% yield after 20
min (TOF ) 1500 h-1); T ) 90 °C, 100% yield after 10 min (TOF )
3000 h-1); T ) 100 °C, 100% yield after 7 min (TOF ) 4286 h-1). Very
low conversions (<27%) were observed after 24 h when T < 60 °C.
(40) Isoprene resistance was also observed in THF solution. Thus, in the presence
of 15 mmol of isoprene, 4 mmol of 1-octen-3-ol (0.2 M in THF) can be
quantitatively isomerized into octan-3-one within 50 min using 0.1 mol %
of 1 and 0.4 mol % of Cs2CO3 (TOF ) 600 h-1 to be compared to entry
1 in Table 4).
(41) (a) Masters, C. Homogeneous Transition-Metal Catalysis; Chapman and
Hall: London, 1981; pp 70-89. (b) Parshall, G. W.; Ittel, S. D.
Homogeneous Catalysis; John Wiley & Sons: New York, 1992; pp 9-24.
(c) Herrmann, W. A. In Applied Homogeneous Catalysis with Organome-
tallic Compounds; Cornils, B., Herrmann, W. A., Eds.; VCH: Weinheim,
1996; Vol. 2, pp 980-991.
9
1366 J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006

Bis(allyl)−Ruthenium(IV) Complexes as Catalysts
A R T I C L E S
Scheme 3. Mechanisms Commonly Proposed for the Redox Isomerization of Allylic Alcohols into Carbonyl Compounds
Scheme 4. Mechanism Proposed by B. M. Trost
the catalyst and gives the free enol, which tautomerizes into
the corresponding carbonyl. In this case, an intramolecular
hydrogen transfer is involved. Because the oxidation state of
the metal has to be raised by two, this mechanism has been
usually proposed for low-valent transition-metal catalysts.6 In
particular, Branchadell and Gre´e have recently demonstrated
by means of DFT calculations that such a mechanism is
operative in the conversion of allylic alcohols to carbonyl
compounds catalyzed by the Fe(CO)3 unit.42
The two pathways described above, which have been well
established for the metal-catalyzed isomerization of unfunc-
tionalized olefins, do not assign a specific role to the oxygen
atom of the allylic alcohols during the catalytic cycle. Moreover,
they cannot explain why some catalysts isomerize allylic
alcohols faster than unfunctionalized alkenes,16 or even exclu-
sively.11,12 To account for these facts, a third mechanism,
invoking coordination of the oxygen atom to the metal,43 has
been recently proposed by Trost and Kulawiec for the isomer-
ization of allylic alcohols catalyzed by [Ru(η5-C5H5)Cl(PPh3)2]
(the η3-oxo-allyl mechanism; see Scheme 4).16b
The first step in the catalytic cycle involves deprotonation
of the allylic alcohol and chelate coordination of the resulting
R,â-unsaturated alkoxide to the metal. Subsequent â-hydrogen
elimination yields an enone-metal-hydride species, which,
after re-addition of the hydride ligand to the substrate, generates
a π-oxo-allyl complex. Final protonation releases the enol and
regenerates the starting complex. The initial formation of the
R,â-unsaturated alkoxide is in agreement with the improvement
of the catalytic activity usually observed in the presence of
bases.30 Experimental evidence, using deuterium-labeled sub-
strates, suggests that several half-sandwich ruthenium(II) cata-
lysts operate through this mechanism.15,16,32a,44 In our case, the
rate enhancement observed in the catalytic activity of complex
1 upon addition of Cs2CO3 as cocatalyst, along with the fact
that this complex is inactive in the isomerization of simple
olefins such as 1-hexene or 1-dodecene, seems to suggest that
a π-oxo-allyl type mechanism (Scheme 4) operates. In the next
section, we will present a theoretical analysis on such a
mechanism.
Theoretical Analysis of the Isomerization of 2-Propen-1-
ol into Propenal. In the following theoretical analysis, we will
focus on the mononuclear alkoxo-π-olefin complex 4 (Figure
3), analogous to the Ru(II) species proposed in Trost’s mech-
anism (Scheme 4), that according to our experimental results
can be proposed as the catalytic active species. We have
introduced minimal modelization substituting the methyl groups
of the 2,7-dimethylocta-2,6-diene-1,8-diyl ligand in 4 by
hydrogens.
(42) Branchadell, V.; Cre´visy, C.; Gre´e, R. Chem.-Eur. J. 2003, 9, 2062-2067.
(43) In this context, Noyori and co-workers have demonstrated that isomerization
of allylic amines catalyzed by complexes [Rh(BINAP)L2][ClO4] (BINAP
) 2,2’-bis(diphenylphosphino)-1,1’-binaphthyl; L ) acetone, methanol,
amines, etc.) involves the initial coordination of nitrogen to Rh. This was
confirmed by trapping intermediates, variable-temperature NMR experi-
ments, and ab initio calculations. (a) Tani, K. Pure Appl. Chem. 1985, 57,
1845-1854. (b) Inoue, S.; Takaya, H.; Tani, K.; Otsuka, S.; Sato, T.;
Noyori, R. J. Am. Chem. Soc. 1990, 112, 4897-4905. (c) Yamakawa, M.;
Noyori, R. Organometallics 1992, 11, 3167-3169. (d) For a review on
catalytic isomerization of allylamines, see: Otsuka, S. Acta Chem. Scand.
1996, 50, 353-360.
(44) (a) van der Drift, R. C.; Bouwman, E.; Drent, E.; Kooijman, H.; Spek, A.
L.; van Oort, A. B.; Mul, W. P. Organometallics 2002, 21, 3401-3407.
(b) Standfest-Hauser, C. M.; Lummerstorfer, T.; Schmid, R.; Kirchner, K.;
Hoffmann, H.; Puchberger, M. Monatsh. Chem. 2003, 134, 1167-1175.
9
J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006 1367

A R T I C L E S
Cadierno et al.
on the reaction coordinate. A schematic and complete catalytic
cycle is shown in Scheme 5. The energy profile is collected in
Schemes 6 and 7, and the geometrical parameters for all of the
structures are provided as Supporting Information (Figure S1;
information about the relative Gibbs free energies for all of the
structures on the PESs is collected in Table S5).
Figure 3. The proposed active species in the isomerization of 2-propen-
1-ol into propanal catalyzed by complex 1.
Scheme 6 shows that the alkoxo-π-olefin complex I0 evolves
into the 16-electron alkoxo intermediate I1 through the transition
structure TS1 involving a moderate energy barrier. An alterna-
tive route proposed by Trost and Kulawiec16b for Ru(II)
complexes (see Scheme 4) in which the oxygen atom of the
hydroxyl group decoordinates from ruthenium at the same time
that the allylic hydrogen migrates to form the ruthenium hydride
bond was less favorable energetically.45 Apparently, the oxygen
To keep the computational cost of the present work within
reasonable limits, we have not considered mechanisms involving
protonations, and, therefore, the results of our theoretical analysis
should apply to the experimental observations in basic media.
On the other hand, we focused on a general mechanism valid
for the two solvents experimentally employed (water and THF)
without explicit participation of individual solvent molecules
Scheme 5. The Complete Catalytic Cycle for the Isomerization of 2-Propen-1-ol into Propanal (For a Detailed Drawing of All Structures,
See the Supporting Information)
9
1368 J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006

Bis(allyl)−Ruthenium(IV) Complexes as Catalysts
A R T I C L E S
Scheme 6. Energy Profile for the First Part of the Catalytic Cycle (For a Detailed Drawing of All Structures, See the Supporting
Information)a
a ∆G values (kcal/mol) including solvation (H2O) effects are given. A temperature of 348.15 K and a pressure of 1 atm were assumed.
Scheme 7. Energy Profile for the Second Part of the Catalytic Cycle (For a Detailed Drawing of All Structures, See the Supporting
Information)a
a ∆G values (kcal/mol) including solvation (H2O) effects are given. A temperature of 348.15 K and a pressure of 1 atm were assumed.
atom is more strongly coordinated to the electrophilic Ru(IV)
center than in the case of Trost-Kulawieck’s Ru(II) complexes,
the Ru-O bond cleavage requiring therefore more energy.
After a â-elimination process, involving the intermediate I2
and transition structures TS2 and TS3, the enal-hydride I3 is
formed. Suitable rotations (I3 f I5) of the O-coordinated enal
ligand, to allow the insertion of the olefinic moiety into the
Ru-H bond, generate the enolate intermediate I6. Finally,
transition structure TS7 leads to the η3(O,C,C)-oxo-allyl
intermediate I7. All of these transformations involve rather small
energy barriers (see Scheme 6). The experimental observation
that the catalyst is inactive in the isomerization of single olefins
strongly supports the formation of this η3-oxo-allyl complex,
in agreement with the mechanism proposed by Trost and
Kulawiec (Scheme 4).16b
Direct addition of 2-propen-1-ol to I7, leading to the
intermediate I8, takes place through a transition structure TS8
with a moderate energy barrier of 16.5 kcal/mol (see Scheme
7). The following steps in the cycle, involving intermediates
I8-I11, which are formed through relatively moderate energy
barriers (TS9, TS10, and TS11), account for the formal
exchange of the coordinated enolate ligand in I8 by the incoming
2-propen-1-ol (Scheme 7). Dissociation of the enol ligand in
I10, through the transition structure TS11, leads to the 16-
electron alkoxo complex I11. A suitable rotation involving TS12
gives rise to the intermediate I1, closing the catalytic cycle (see
Scheme 5). The formed enol spontaneously tautomerizes into
the more stable propanal, which is the desired final product of
the studied process.
Our experimental results show that the formation of carbonyl
compounds from allylic alcohols, catalyzed by the bis(allyl)-
ruthenium(IV) complexes, takes place only by increasing the
(45) An anonymous referee contributed in a decisive manner to elucidate this
important point.
9
J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006 1369

A R T I C L E S
Cadierno et al.
temperature (T > 60°) as mentioned above.39 The DFT
calculations suggest that the I7 f I11 transformation in Scheme
7, involving a global energy barrier of 27 kcal/mol, is
responsible for the required activation.46 The ∆G free energy
of the whole cycle is -15.9 kcal/mol. Therefore, it is clearly
an exergonic cycle, in full agreement with the high yields
experimentally observed for the isomerization reaction of allyl
alcohols into carbonyl compounds catalyzed by the bis(allyl)-
ruthenium(IV) complex [{Ru(η3:η3-C10H16)(µ-Cl)Cl}2] (1) (see
Table 4).
to the intermediate η3-oxo-allyl complex, and the subsequent
dissociation of the formed enol ligand, lead to a global energy
barrier (27 kcal/mol), which is consistent with the experimentally
observed activation required for the studied reactions to proceed.
Calculations also predict the whole catalytic cycle to be strongly
exergonic (-15.9 kcal/mol), in full agreement with the high
yields experimentally observed.
All of these theoretical and experimental facts together with
the catalytic activity shown by the ruthenium(IV) catalyst in
the presence of isoprene (in contrast to the loss of activity found
with other catalysts) disclose a powerful tool for the synthesis
of carbonyl compounds in an environment friendly manner.
Because the synthesis of aldehydes and ketones using readily
available dienes as raw materials, via a hydration/isomerization
sequence, is nowadays an important goal,47 the use of this
catalytic process provides a valuable synthetic approach, which
in addition can be performed in water.
Conclusions
In this work, we have shown that the air-stable bis(allyl)
ruthenium(IV) dimer [{Ru(η3:η3-C10H16)(µ-Cl)Cl}2] (1) as well
as its mononuclear derivatives [Ru(η3:η3-C10H16)Cl2(L)] (L )
CO, PR3, CNR, NCR) (2) and [Ru(η3:η3-C10H16)Cl(NCMe)2]-
[SbF6] (3) are highly efficient catalysts for the redox isomer-
ization of allylic alcohols into carbonyl compounds. In particular,
complex 1 has been found to be the most efficient catalyst
reported to date for this isomerization reaction (TOF and TON
values up to 62 500 h-1 and 1 500 000, respectively). Moreover,
the transformation can be performed in water, and the catalyst
can be recycled up to four times with only partial deactivation.
The potential energy surfaces corresponding to the catalytic
cycle were explored. Our calculations support a variant of the
mechanism proposed by Trost and Kulawiec for the catalytic
isomerization of allylic alcohols into carbonyl compounds, based
on alkoxo-π-olefin ruthenium complexes as the active species,
in which the cleavage of the Ru(IV)-olefin versus Ru(IV)-O
bond is preferred. According to the DFT calculations, the
reaction steps involving the addition of a 2-propen-1-ol molecule
Acknowledgment. We are indebted to the Ministerio de
Educacio´n y Ciencia (MEC) of Spain (Projects BQU2003-00255
and BQU2004-07405-C02-02) for financial support. J.G.,
S.E.G.-G., and V.C. also thank the Gobierno del Principado de
Asturias (Project PR-01-GE-6) for financial support. S.E.G.-G.
and V.C. thank the MCyT for the award of a Ph.D. grant and
a “Ramo´n y Cajal” contract, respectively.
Supporting Information Available: Full ref 29 and Cartesian
coordinates, energetic data (relative and absolute energies), and
some representative geometrical parameters for all of the
structures located at the PESs. This material is available free
of charge via the Internet at http://pubs.acs.org.
JA054827A
(46) A rough estimate of the half-life time based on the computed barrier height
(using 1 M standard state) leads to a value close to 5 min, which extrapolates
rather well to the experimental total reaction times reported in entry 6 in
Table 4 (20-60 min). Cramer, C. J. Essentials of Computational Chemistry.
Theory and Models; Wiley: New York, 2002.
(47) (a) Drent, E. Eur. Patent No. 457387, 1991 (to Shell). (b) van der Drift, R.
C.; Mul, W. P.; Bouwman, E.; Drent, E. Chem. Commun. 2001, 2746-
2747. See also ref 12.
9
1370 J. AM. CHEM. SOC. VOL. 128, NO. 4, 2006