A highly selective arene hydrogenation catalyst that operates in ionic liquid

Article abstract of DOI:10.1021/ja026361r

The synthesis and structural characterization of [Ru(η⁶-p-cymene)(η²-TRIPHOS)Cl][PF₆] is described. The complex is a highly active, homogeneous arene hydrogenation catalyst that is selective toward the hydrogenation of aromatic rings in preference to alkenes, as demonstrated by the hydrogenation of allylbenzene to allylcyclohexane. The catalyst operates in both dichloromethane and ionic liquids and undergoes no decomposition in the latter solvent. Copyright

Full text of DOI:10.1021/ja026361r

Published on Web 07/17/2002
A Highly Selective Arene Hydrogenation Catalyst that Operates in Ionic
Liquid
Clive J. Boxwell,† Paul J. Dyson,*,‡ David J. Ellis,† and Thomas Welton§
Department of Chemistry, The UniVersity of York, Heslington, York, YO10 5DD, U.K.,
Institut de chimie mole´culaire et biologique, Ecole polytechnique fe´de´rale de Lausanne,
EPFL - BCH, CH-1015 Lausanne, Switzerland, and Department of Chemistry,
Imperial College of Science, Technology, and Medicine, South Kensington, London, SW7 2AY, U.K.
Received March 29, 2002
Aromatic compounds are notoriously difficult substrates to
hydrogenate, although a wide range of reagents have been used
for this purpose.1 It has been proposed that because chemistry of
aromatic compounds is so vast, the most obvious route to
stereodefined unsaturated cyclic compounds is to construct the
appropriate aromatic compound and then reduce it in a stereose-
lective fashion. The majority of the reagents used to hydrogenate
aromatic compounds are used in stoichiometric quantities, such as
the Birch method that employs group 1 or 2 metals dissolved in
liquid ammonia.2 In contrast, effective homogeneous catalysts are
relatively rare, although several ingenious systems have been
reported. For example, Rothwell has developed a series of
homogeneous niobium- and tantalum-based catalysts,3 Su¨ss-Fink
has described a number of robust cluster catalysts that operate under
aqueous-organic biphasic conditions,4 and Angelici has supported
some rhodium complexes onto metal-impregnated porous supports
that results in highly active catalysts.5 We have also shown that
arenes can be hydrogenated using an ionic liquid-organic biphasic
protocol.6 In this paper, we describe a new arene hydrogenation
catalyst that is active in organic solvents, but considerably more
active in ionic liquids, and, remarkably, will hydrogenate the arene
ring of allylbenzene without hydrogenating the alkene bond.
The reaction of the triply bridged chloro-dimer, [(η6-p-cymene)2-
Ru2(µ-Cl)3][PF6], with 2 equiv of 1,1,1-tris(diphenylphosphinom-
ethyl)ethane (TRIPHOS) in the presence of 1 equiv of ammonium
hexafluorophosphate in methanol under reflux affords [Ru(η6-p-
cymene)(η2-TRIPHOS)Cl][PF6] 1‚[PF6] in high yield.7 The 1H
NMR spectrum of 1 is quite complicated and shows that the six-
membered ring formed by the ruthenium and TRIPHOS ligand
exists in chair and boat conformations in a similar way to other
compounds previously reported.8 The structure of 1 has been
established by single-crystal X-ray diffraction using a crystal grown
from a dichloromethane-methanol solution by slow diffusion.9 The
structure of 1 is shown in the Supporting Information. The geometry
around the Ru atom is essentially octahedral, made up by the arene
[mean Ru-C 2.28 Å], the chlorine ligand [Ru-Cl 2.3900(10) Å],
and two of the three phosphine donors [mean Ru-P 2.338 Å] which
form a six-membered ring with a chair conformation. The third
phosphine unit resides at quite a distance from the Ru atom. The
structure around the ruthenium center is similar to other cationic
Ru(arene)(diphosphine)chloride complexes previously described.10
The bonding mode for the TRIPHOS ligand has also been observed
previously, although those in which the free phosphine remains
unoxidized are very rare.11
The extensive hydrophobic region provided by the TRIPHOS
ligand has been proposed to facilitate catalytic processes by
promoting the interaction of the coordinated metal center with small
organic molecules.12 Compound 1 has been evaluated in dichlo-
romethane under homogeneous conditions and in 1-butyl-3-meth-
ylimidazolium tetrafluoroborate ionic liquid, [bmim][BF4], under
biphasic conditions and has proven to be a highly active arene
hydrogenation catalyst. The results from some of these reactions
are summarized in Table 1, and a more extensive list is given in
the Supporting Information. Overall, the results show the expected
trend for the hydrogenation of arene substrates,4 in that the highest
turnover is observed for the hydrogenation of benzene, with
turnovers decreasing as the size or number of alkyl substituents
increases on the arene substrate. What is surprising is that 1 is
essentially inactive toward arenes with R-alkene substituents such
as styrene and 1,3-divinylbenzene, whereas the turnover for the
hydrogenation of allylbenzene to allylcyclohexane is considerably
higher than expected, ca. 329 mol mol-1 h-1 as compared to 205
and 127 mol mol-1 h-1 for toluene and ethylbenzene, respectively.
Furthermore, the alkene bond remains unhydrogenated. Hydrogena-
tion of alkene substrates such as 1-hexene and 1-octene was
attempted with 1 as the catalyst, which gave very low turnover
numbers, in part because of catalyst decomposition, but suggesting
that this catalyst has considerable potential in organic synthesis.
Arene hydrogenation using 1 is believed to take place via a
slippage mechanism, the preferred mechanism for arene hydrogena-
tion in organic synthesis. Although we do not have direct evidence
that such a mechanism is in operation, there is convincing
circumstantial evidence. First, the presence of mercury, employed
as a selective poison toward colloids, did not effect activity. Second,
the catalyst is recovered, albeit with the substrate arene in place of
the p-cymene in the starting material, intact. Third, hydrogenation
of C6D6 affords only a single isomer of C6D6H6, demonstrating
that the hydrogen atoms have been introduced to the arene ring
from one side only, indicative of the addition occurring to a
coordinated arene.14 It is not immediately obvious why styrene and
divinylbenzene are resistant to hydrogenation. It is possible that
they form an η4 interaction via the vinyl double bond and the arene
which inhibits hydrogenation. It would appear that the third, free
phosphine is essential in providing selectivity toward the hydro-
genation of arenes. When this phosphine is blocked by coordination
to a nonactive metal, the resulting material does not hydrogenate
arenes, but becomes active toward alkene substrates. We propose
that the free phosphine assists the hydrogenation of arenes by
* To whom correspondence should be addressed. E-mail: paul.dyson@epfl.ch.
† The University of York.
‡ Ecole polytechnique fe´de´rale de Lausanne.
§ Imperial College of Science, Technology, and Medicine.
9
9334
J. AM. CHEM. SOC. 2002, 124, 9334-9335
10.1021/ja026361r CCC: $22.00 © 2002 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Hydrogenation of Some Arenes Using 1‚[PF6] in
coordinates, full bond lengths and bond angles (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
[bmim][BF4]13
References
(1) Donohoe, T. J.; Garg, R.; Stevenson, C. A. Tetrahedron: Asymmetry 1996,
7, 317.
(2) For example, see: Rabideau, P. W. Tetrahedron 1989, 45, 1579.
(3) Rothwell, I. P. Chem. Commun. 1997, 1331.
(4) (a) Meister, G.; Rheinwald, G.; Stoecki-Evans, H.; Su¨ss-Fink, G. J. Chem.
Soc., Dalton Trans. 1994, 3215. (b) Plasseraud, L.; Su¨ss-Fink, G. J.
Organomet. Chem. 1997, 539, 163. (c) Fidalgo, E. G.; Plasseraud, L.;
Su¨ss-Fink, G. J. Mol. Catal. A 1998, 132, 5. (d) Su¨ss-Fink, G.; Faure,
M.; Ward, T. R. Angew. Chem., Int. Ed. 2002, 41, 99.
(5) Yang, H.; Gao, H. R.; Angelici, R. J. Organometallics 2000, 19, 622.
(6) Dyson, P. J.; Ellis, D. J.; Parker, D. G.; Welton, T. Chem. Commun. 1999,
25.
(7) Synthesis and characterisation of 1: A solution of [(η6-p-cymene)2Ru2-
(µ-Cl)3][PF6] (500 mg, 0.693 mmol), 1,1,1-tris(diphenylphosphinomethyl)-
ethane, TRIPHOS (866 mg, 1.386 mmol), and NH4PF6 (113 mg, 0.693
mmol) in methanol (200 mL) was refluxed for 4 h. After being cooled to
room temperature, filtration of the solution, followed by the removal of
the solvent under reduced pressure and washing with cold ethanol and
diethyl ether, afforded a yellow microcrystalline solid (1424 mg, 1.368
mmol, 98.7% yield/Ru). Single crystals for X-ray structural determination
were grown using liquid diffusion of hexane into a dichloromethane
solution of 1. Positive ion electrospray mass spectrum m/z 895 [Ru(η6-
p-cymene)(η2-TRIPHOS)Cl]+ m/z 761 [Ru(TRIPHOS)Cl]+; 31P-{1H}
,
NMR (CDCl3) 26.51 (s), 24.72 (s), -28.40 (s), -29.66 (s), -142.95
(septet, J ) 708.95 Hz) ppm; 1H NMR* (CDCl3) 7.6-6.8 (m, boat/chair),
5.72-5.46 (m, boat/chair), 3.37 (d, J ) 14.5 Hz, boat 1H), 3.13 (dt, J )
15.0 Hz, J ) 4.0 Hz, chair 1H), 2.32 (m, chair 1H), 2.25 (s, chair 2H),
2.13 (m, boat 1H), 2.07 (septet, J ) 6.86 Hz, boat 1H), 1.75 (s, boat 3H),
1.69 (s, boat 2H), 1.56 (septet, J ) 6.86 Hz, chair 1H), 1.20 (s, chair
3H), 1.03 (s, boat 3H), 0.74 (m, boat/chair), 0.14 (s, chair 3H) ppm.
(8) (a) Landgrafe, C.; Sheldrick, W. S.; Su¨dfeld, M. Eur. J. Inorg. Chem.
1998, 407. (b) Gibson, D. H.; He, H. Y.; Mashuta, M. S. Organometallics
2001, 20, 1456.
(9) Crystal data: (120 K) C51H53ClF6P4Ru, MW ) 1040.33, monoclinic P21/
n, a ) 12.431(3) Å, b3 ) 21.714(4) Å, c1 ) 17.416(4) Å, â ) 98.23(3)°,
Z ) 4, V ) 4652.7(16) Å3, µ ) 0.592 mm-1, 10 461 unique reflections,
R1 ) 0.0579 (0.0919 for all data), wR2 ) 0.1384 (0.1575 for all data).
Crystallographic data (excluding structure factors) for the structure in this
paper have been deposited with the Cambridge Crystallographic Data
Centre as supplementary publication no. CCDC-189682 (1‚PF6). Copies
of the data can be obtained, free of charge, on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK (fax: +44 1223 336033 or e-mail
deposit@ccdc.cam.ac.uk).
(10) For example, see: (a) Polan, J. R.; Porter, L. C. Inorg. Chim. Acta 1993,
205, 119. (b) Jensen, S. B.; Rodger, S. J.; Spicer, M. D. J. Organomet.
Chem. 1998, 556, 151.
(11) (a) Lin, S. C.; Cheng, C. P.; Lee, T.-Y.; Lee, T.-J.; Peng, S.-M. Acta
Crystallogr., Sect. C 1986, 42, 1733. (b) Blake, A. J.; Gould, R. O.;
Halcrow, M. A.; Schroder, M. J. Chem. Soc., Dalton Trans. 1993, 2909.
(12) Bianchini, C.; Farnetti, E.; Graziani, M.; Kaspar, J.; Vizza, F. J. Am. Chem.
Soc. 1993, 115, 1753.
(13) In a typical hydrogenation, the catalyst (40-50 mg) was dissolved in
solvent (10 mL), and the arene substrate (2 mL) was added. The autoclave
was pressurized with H2 to 60 atm, sealed, and heated to 90° for 1 h. The
products were identified using a combination of GC versus known
standards and NMR spectroscopy. Turnover frequenciess are quoted in
number of moles of substrate converted per mole of catalyst per hour.
(14) Eisen, M. S.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 10358.
(15) (a) Ott, J.; Venanzi, L. M.; Ghilardi, C. A.; Midollini, S.; Orlandini, A. J.
Organomet. Chem. 1985, 291, 89. (b) Thaler, E. G.; Folting, K.; Caulton,
K. G. J. Am. Chem. Soc. 1990, 112, 2664.
(16) Stuhl, L. S.; DuBois, M. R.; Hirsekorn, F. J.; Bleeke, J. R.; Stevens, A.
E.; Muetterties, E. L. J. Am. Chem. Soc. 1978, 100, 2405.
(17) Elian, M.; Hoffmann, R. Inorg. Chem. 1975, 14, 1058.
(18) (a) Dyson, P. J.; Ellis, D. J.; Parker, D. G.; Welton, T. J. Mol. Catal. A
1999, 150, 71. (b) Dyson, P. J.; Russel, K.; Welton, T. Inorg. Chem.
Commun. 2001, 4, 571.
reversible coordination to the ruthenium involving an “arm-on/off”
mechanism proposed previously.15 This could have the effect of
causing the ring to slip from the η6 to η4 mode, thereby promoting
hydrogenation of the uncoordinated arene CdC double bond. We
intend to report more fully on this in the future. A cobalt complex
has previously been shown to catalyze the reduction of allylbenzene
to allylcyclohexane in 2% yield along with other hydrogenation
products,16 and it has been postulated that such a conversion is not
unreasonable when catalyzed by a d6 metal center,17 although we
are not aware of any other examples.
As mentioned above, 1 was examined in both dichloromethane
and ionic liquid solutions for comparison purposes. Although the
latter system is biphasic, it is considerably superior, although we
have previously shown that ionic liquids are not always superior
solvents for conducting hydrogenation reactions.18 Not only are the
turnovers significantly higher, but, over a number of runs, the
catalyst decomposes in dichloromethane, whereas in [bmim][BF4]
no depreciation in activity is observed after five runs.
Acknowledgment. We also thank the EPSRC X-ray crystal-
lography service (Southampton) for collecting data on 1‚[PF6] and
Johnson-Matthey for the loan of ruthenium salts. We are grateful
to the EPSRC and the Swiss National Science Foundation for
financial support and the Royal Society for a University Research
Fellowship (P.J.D.).
Supporting Information Available: Tables of hydrogenation data
for 1‚[PF6], crystal data, figure of the molecular structure of the cation
and selected bond parameters, structure solution and refinement, atomic
JA026361R
9
J. AM. CHEM. SOC. VOL. 124, NO. 32, 2002 9335