Mechanistic investigations on the hydrogenation of alkenes using ruthenium(II)-arene diphosphine complexes

Article abstract of DOI:10.1021/om049665q

The ruthenium(II) complexes [Ru(η²-P-P)(η⁶-p- cymene)Cl]Cl, P-P = diphenylphosphinomethane (dppm), diphenylphosphinoethane (dppe), or diphenylphosphinopropane (dppp), and the highly water-soluble analogues, [Ru(η²-P-P)(η⁶-arene)Cl]Na₃, P-P = 1,2-bis(di-4-sulfonatophenylphosphino)benzene (dppbts), arene = p-cymene, benzene, or [2.2]paracyclophane, are efficient catalyst precursors for the hydrogenation of styrene in an aqueous biphase. By the use of high gas pressure NMR techniques and electrospay ionization mass spectrometry, the active species in the hydrogenation have been indirectly identified to be a dihydrogen complex, which also catalyzes H/D isotope exchange. Using the ruthenium(II) dppbts derivatives as precatalysts, evidence is provided for an arene exchange process that takes place during the catalytic hydrogenation of styrene. Together, these results lead to the proposition of a catalytic cycle for the hydrogenation of the C=C double bond of styrene using ruthenium(II)-arene diphosphine complexes.

Full text of DOI:10.1021/om049665q

Organometallics 2004, 23, 4849-4857
4849
Mech a n istic In vestiga tion s on th e Hyd r ogen a tion of
Alk en es Usin g Ru th en iu m (II)-a r en e Dip h osp h in e
Com p lexes
Corinne Daguenet, Rosario Scopelliti, and Paul J . Dyson*
Institut des Sciences et Inge´nierie Chimiques, Ecole Polytechnique Fe´de´rale de Lausanne,
EPFL-BCH, CH-1015 Lausanne, Switzerland
Received May 12, 2004
The ruthenium(II) complexes [Ru(η2-P-P)(η6-p-cymene)Cl]Cl, P-P ) diphenylphosphi-
nomethane (dppm), diphenylphosphinoethane (dppe), or diphenylphosphinopropane (dppp),
and the highly water-soluble analogues, [Ru(η2-P-P)(η6-arene)Cl]Na3, P-P ) 1,2-bis(di-4-
sulfonatophenylphosphino)benzene (dppbts), arene ) p-cymene, benzene, or [2.2]paracyclo-
phane, are efficient catalyst precursors for the hydrogenation of styrene in an aqueous
biphase. By the use of high gas pressure NMR techniques and electrospay ionization mass
spectrometry, the active species in the hydrogenation have been indirectly identified to be
a dihydrogen complex, which also catalyzes H/D isotope exchange. Using the ruthenium(II)
dppbts derivatives as precatalysts, evidence is provided for an arene exchange process that
takes place during the catalytic hydrogenation of styrene. Together, these results lead to
the proposition of a catalytic cycle for the hydrogenation of the CdC double bond of styrene
using ruthenium(II)-arene diphosphine complexes.
In tr od u ction
employed to induce water solubility of the catalysts. The
mechanism of hydrogenation reactions involving ruthe-
nium(II) complexes is thought, in most cases, to operate
via the so-called “hydride” route, as do, for example, the
well-known complex [RuCl2(PPh3)3] and the arene dimer
[RuCl2(arene)]2, which react under hydrogen to form
active hydride species by loss of HCl.9,17 Other ruthe-
nium(II) complexes such as [Ru(O2CCH3)2(arene)] and
[Ru(O2CCH3)2(diphosphine)] also heterolytically cleave
H2 to give active hydride species.17-19 In addition to the
hydride route a few groups have postulated that hydro-
gen transfer could directly occur from a dihydrogen
complex, as in the case with the dimeric species [Ru-
(diphosphine)Cl2]2, which binds H2 to form the triply
bridged chloro complex [Ru(diphosphine)(η2-H2)(µ-Cl)3-
RuCl(diphosphine)].20 Here the hydrogen can transfer
to a π-bonded alkene. With the arene ruthenium(II)
complex [Ru(CH3CN)3(benzene)][BF4]2, which operates
in an aqueous biphase, both the hydride and dihydrogen
routes have been proposed.21
To improve product separation and catalyst recycling
using molecular (homogeneous) species, the use of
biphasic conditions has emerged as a judicious solution,
especially with the environmentally friendly aqueous
biphase,1-3 although supercritical fluids,4 ionic liquids,5
and fluorous processes6 are also being evaluated. The
use of water in biphasic catalysis not only derives from
its benign nature but is also an easily tuneable medium,
which provides many unique opportunities in cataly-
sis.7,8 A large number of catalysts have been investi-
gated in water, including many hydrogenation cata-
lysts.1,8
The catalytic hydrogenation of the CdC bonds by
ruthenium(II) complexes has been widely reported, and
while the majority of catalysts operate under homoge-
neous conditions,9-13 aqueous biphasic conditions have
been applied,14-16 in which hydrophilic ligands are
(1) J oo´, F. Aqueous Organometallic Catalysis; Kluwer: Dordrecht,
2001.
While these mechanistic studies were conducted in
organic solvent, with the exception of the latter example,
it is also possible to study catalyst reaction mechanisms
in aqueous-organic biphasic processes.21-23
(2) Cornils, B., Herrmann, W. A., Eds. Aqueous-Phase Organome-
tallic Catalysis. Concepts and Applications; Wiley-VCH: Weinheim,
1998.
(3) J oo´, F.; Papp, EÄ .; Katho´, AÄ . Top. Catal. 1998, 5, 113.
(4) J essop, P. G.; Leitner, W. Chemical Synthesis using Supercritical
Fluids; Wiley-VCH, Weinheim, 1999.
(5) Dupont, J .; D. Souza, R. F.; Suarez, P. A. Z. Chem. Rev. 2002,
102, 3667.
(6) Horva´th, I. T. Acc. Chem. Res. 1998, 31, 641.
(7) J oo´, F.; Kova´cs, J .; Be´nyei, A. Cs.; Katho´, AÄ . Angew. Chem., Int.
Ed. 1998, 37, 969.
(14) Andriollo, A.; Bol´ıvar, A.; Lo´pez, F. A.; Pa´ez, D. E. Inorg. Chim.
Acta 1995, 238, 187.
(15) Ellis, D. J .; Dyson, P. J .; Parker, D. G.; Welton, T. J . Mol. Catal.
A 1999, 150, 71.
(8) J oo´, F. Acc. Chem. Res. 2002, 35, 738.
(9) J ames, B. R. Homogeneous Hydrogenation; Wiley: New York,
1973.
(10) Halpern, J .; Harrod, J . F.; J ames, B. R. J . Am. Chem. Soc. 1966,
88, 5150.
(11) Hallman, P. S.; Evans, D.; Osborn, J . A.; Wilkinson, G. Chem.
Comm. 1967, 7, 305.
(12) Suarez, T.; Fontal, B. J . Mol. Catal. 1988, 45, 335.
(13) Corma, A.; Iglesias, M.; Del Pino, C.; Sanchez, F. J . Organomet.
Chem. 1992, 431, 233.
(16) )Rojas, I.; Linares, F. L.; Valencia, N.; Bianchini, C. J . Mol.
Catal. A 1999, 144, 1.
(17) Brothers, P. J . Prog. Inorg. Chem. 1981, 28, 1.
(18) Ashby, M. T.; Halpern, J . J . Am. Chem. Soc. 1991, 113, 589.
(19) Bennett, M. A.; Ennett, J . P. Inorg. Chim. Acta 1992, 198-
200, 583.
(20) J oshi, A. M.; MacFarlane, K. S.; J ames, B. R. J . Organomet.
Chem. 1995, 488, 161.
(21) Chan, W.-C.; Lau, C.-P.; Cheng, L.; Leung, Y.-S. J . Organomet.
Chem. 1994, 464, 103.
10.1021/om049665q CCC: $27.50 © 2004 American Chemical Society
Publication on Web 09/09/2004

4850 Organometallics, Vol. 23, No. 21, 2004
Daguenet et al.
Ch a r t 1
In this work, we report the use of ruthenium(II)-arene
diphosphine complexes as precatalysts for the hydro-
genation of styrene, including water-soluble complexes
that operate in an aqueous biphase. Remarkably, the
biphasic conditions provide a way to study the evolution
of the various species present after catalysis employing
in particular electrospray ionization mass spectrometry
(ESI-MS), which has been previously shown to be an
efficient technique to observe transition metal com-
plexes.24 In addition, high gas pressure NMR spectros-
copy25 gives further clues to the mechanism.
p-cymene (4), benzene (5), or [2.2]paracyclophane (6)).
Direct reaction of the dimer with dppbts in a mixture
of water-ethanol (1:1) for a much shorter period of time,
followed by precipitation of the complex after addition
of ethanol, affords 4-6 in moderate to good yield.
The electrospray ionization (ESI) mass spectra of 4,
5, and 6 in water, recorded in negative ion mode, show
strong peak envelopes corresponding to the triply
charged anion [Ru(dppbts)(η6-arene)Cl]3- ([M]3-), cen-
tered at m/z ) 344.3, 325.7, and 369.0, respectively. Less
intense peaks, which correspond to species with a
countercation, [M + H]2- and [M + Na]2-, are also
observed at m/z ) 516.9 and 527.8 for 4, 489.0 and 499.9
for 5, and 554.0 and 565.0 for 6; these data are in
keeping with those observed for other sulfonated phos-
phine complexes.29 The 31P NMR spectra of these
complexes in methanol-d4 exhibit singlets at δ 67.1 for
4, δ 65.1 for 5, and δ 72.9 for 6, which is consistent with
the ring contribution of a five-membered ring formed
by chelation.30 Confirmation of the identity of 4-6 was
provided by two-dimensional 13C-1H correlation NMR
spectroscopy in methanol-d4 (see Supporting Informa-
tion for full details). Cross-peaks, characteristic of the
η6-coordinated arenes, appear at δ 6.27 and δ 98.09,
93.64 for 4, δ 6.22 and δ 97.80 for 5, and δ 5.21 and δ
Resu lts a n d Discu ssion
A series of ruthenium(II)-arene diphosphine com-
plexes have been synthesized from the appropriate
dimer, [{Ru(η6-arene)Cl2}2] (Chart 1). The complexes
[Ru(diphosphine)(η6-p-cymene)Cl]Cl, diphosphine )
diphenylphosphinomethane (dppm) (1), diphenylphos-
phinoethane (dppe) (2), or diphenylphosphinopropane
(dppp) (3), were prepared according to a method de-
scribed by Faraone et al. with some modifications.26 In
our experience, the literature procedure results in the
formation of byproducts including phosphine-bridged
bimetallic complexes and bis-diphosphine species re-
sulting from arene displacement, as noted elsewhere.27
To avoid bridged species, a large excess of diphosphine
was used and lower temperatures were applied to
prevent loss of the arene. In this way, the cationic
diphosphine complexes were formed as the main species
with chloride as the counteranion. The chloride is easily
replaced with tetrafluoroborate in aqueous solution.
NMR data for these complexes were in accordance with
the literature data.28 To preclude side-reactions, another
route has also been developed, but it involves a supple-
mentary step, that is, the formation of [Ru(η6-arene)-
(CH3CN)2Cl][PF6].27,28
1
94.68 for 6, in the H and 13C dimensions, respectively.
Compounds 4-6 contain approximately 10% dppbts
dioxide with a characteristic 31P NMR signal at ca. δ
33.2 in methanol-d4. A small amount of the hydride
[Ru(dppbts)(η6-arene)H]3- is detected by negative ion
ESI-MS in 5 and 6 [m/z ) 314.3 and 357.9, respectively],
as well as the hydrolysis product, [Ru(dppbts)(η6-arene)-
(H2O)]2- [m/z ) 480.0 for 5 and 545.0 for 6].
The structures of 1‚BF4 and 2‚BF4 were established
by single-crystal X-ray diffraction in order to ascertain
the P-Ru-P angle and correlate this bite angle with
the kinetics of hydrogenation (see below, Table 2). X-ray
suitable crystals of the two tetrafluoroborate salts were
obtained by slow evaporation from acetone, and the
resulting structures of 1‚BF4 and 2‚BF4 are depicted in
Figures 1 and 2, respectively, with key bond parameters
listed in the captions. The two structures are similar to
related ruthenium-arene diphosphine structures that
have been published previously.31,32
Using an adaptation of the above method, the water-
soluble diphosphine 1,2-bis(di-p-sulfonatophenylphos-
phino)benzene (dppbts) was reacted with three different
dimers to afford [Ru(η6-arene)(dppbts)Cl]Na3 (arene )
(22) Ogo, S.; Abura, T.; Watanabe, Y. Organometallics 2002, 21,
2964.
(23) J ayasree, S.; Seayad, A.; Sarkar, B. R.; Chaudhari, R. V. J . Mol.
Catal. A 2002, 181, 221.
(24) Dyson, P. J .; McIndoe, J . S. Inorg. Chim. Acta 2003, 354, 68.
(25) Elsevier: C. J . J . Mol. Catal. 1994, 92, 285.
(26) Faraone, F.; Loprete, G. A.; Tresoldi, G. Inorg. Chim. Acta 1979,
34, L251.
(29) Bryce, D. J . F.; Dyson, P. J .; Nicholson, B. K.; Parker, D. G.
Polyhedron 1998, 17, 2899
(27) Fogg, D.; J ames, B. R. J . Organomet. Chem. 1993, 462, C21.
(28) J ensen, S. B.; Rodger, S. J .; Spicer, M. D. J . Organomet. Chem.
1998, 556, 151.
(30) Garrou, P. E. Chem. Rev. 1981, 81, 229.
(31) De los Rios, I.; J ime´nez Tenorio, M.; J ime´nez Tenorio, M. A.;
Puerta, M. C.; Valerga, P. J . Organomet. Chem. 1996, 525, 57.

Hydrogenation of Alkenes
Organometallics, Vol. 23, No. 21, 2004 4851
Ta ble 1. Cr ysta l Da ta a n d Deta ils of th e Str u ctu r e
Deter m in a tion for 1‚BF 4 a n d 2‚BF 4
1‚BF4
2‚BF4
chemical formula
fw
cryst syst
space group
a (Å)
C
35H36BClF4P2Ru
C36H38BClF4P2Ru
755.93
741.91
orthorhombic
P212121
11.861(5)
15.411(5)
18.360(3)
3356.1(18)
4
orthorhombic
P212121
12.4176(6)
15.2102(7)
17.3187(10)
3271.0(3)
4
b (Å)
c (Å)
volume (Å3)
Z
D
calc (g cm-3
)
1.468
1.535
F(000)
1512
0.688
293(2)
1544
0.708
140(2)
µ (mm-1
)
temp (K)
wavelength (Å)
measd reflns
unique reflns
unique reflns
[I > 2σ(I)]
no. of data/params
Ra [I > 2σ(I)]
wR2a (all data)
GoFb
0.71070
20 159
5883
0.71073
20 135
5750
4216
5351
5883/430
0.0405
0.0978
0.956
5750/406
0.0275
0.0601
1.032
F igu r e 1. Ball-and-stick representation of the crystal
structure of 1‚BF4. Selected bond distances (Å) and angles
(deg): Ru1-Cl1, 2.397(2); Ru1-P1, 2.316(2); Ru1-P2,
2.309(2); P1...P2, 2.695(3); P1-Ru1-P2, 71.29(6); Cl1-
Ru1-P1, 83.47(5); Cl1-Ru1-P2, 84.25(6); P1-C23-P2,
95.3(3).
R ) ∑|Fo| - |Fc|/∑|Fo|, wR2 ) {∑[w(Fo2 - Fc2)2]/∑[w(Fo2)2]}1/2
.
a
GoF ) {∑[w(Fo - Fc2)2]/(n - p)}1/2 where n is the number of
data and p is the number of parameters refined.
b
2
Ta ble 2. Effect of th e Ster ic Con str a in ts of
[Ru (p-cym en e)(P -P )Cl]Cl on th e In itia l Ra te of
Styr en e Hyd r ogen a tion a
P-P
bite angle/deg
v0/M min-1
dppm
dppe
dppp
71.3b
83.0b
91.1c
0.125 ( 0.013
0.377 ( 0.017
4.2 ( 0.9
a
Reaction conditions: 80 °C, p(H2) ) 45 bar, substrate/catalyst
) 1500, in toluene-d8. This work, c Estimation from ref 40.
b
Ca ta lytic In vestiga tion s. Compounds 1-6 were
tested for catalytic hydrogenation of styrene with mo-
lecular hydrogen in order to assess their activity. Using
toluene as a cosolvent with complexes 1-3, in an
autoclave at 100 °C and at a partial pressure of
hydrogen of 45 bar with a ratio substrate/catalyst of
about 2000, 100% conversion was observed when the
reaction is stopped after 2 h, which provides a high
turnover frequency (TOF) compared to other related
ruthenium complexes.33-36 Similar activities are ob-
served in an aqueous biphase under the same condi-
tions. Addition of mercury, as a selective poison for
colloidal/nanoparticle catalysts,37 does not affect the
conversion, neither in a pure organic phase nor under
biphasic conditions, indicating that the active catalyst
is homogeneous.
F igu r e 2. Ball-and-stick representation of the crystal
structure of 2‚BF4. Selected bond distances (Å) and angles
(deg): Ru1-Cl1, 2.430(1); Ru1-P1, 2.329(1); Ru1-P2,
2.336(1); P1...P2, 3.092(2); P1-Ru1-P2, 83.03(3); Cl1-
Ru1-P1, 89.43(5); Cl1-Ru1-P2, 80.87(3).
monitored by 1H NMR spectroscopy (Figure 3, A). Initial
rates were determined for each diphosphine complex
(Table 2). A trend is observed showing an enhancement
of the initial rate with the increase of the alkyl chain
length of the diphosphine. Progressing from dppm to
dppp, the P-Ru-P bite angle becomes larger, increas-
ing the steric constraints around the coordination
sphere of the metal. The bite angles of diphosphines
have previously been shown to effect the catalytic
activity and selectivity of several important reac-
tions.39,40 In the system described herein, the correlation
of the bite angle with the activity is consistent with a
dissociative process occurring during a rate-determining
step. Dissociation of the coordinated chloride to give a
dihydrogen complex (or dihydride) is then postulated.
Hydrogenations using 1-3 were also performed in
deuterated toluene in a sapphire NMR tube38 and
(32) Mashima, K.; Kusano, K.; Ohta, T.; Noyori, R.; Takaya, H. J .
Chem. Soc., Chem. Commun. 1989, 1208.
(33) Moldes, I.; de la Encarnacio´n, E.; Ros, J .; Alvarez-Larena, AÄ .;
Piniella, J . F. J . Organomet. Chem. 1998, 566, 165.
(34) Parmar, D. U.; Bhatt, S. D.; Bajaj, H. C.; J asra, R. V. J . Mol.
Catal. A 2003, 202, 9.
(35) Baricelli, P. J .; Rodr´ıguez, G.; Rodr´ıguez, M.; Lujano, E.; Lo´pez-
Linares, F. Appl. Catal. A 2003, 239, 25.
(36) Baricelli, P. J .; Izaguirre, L.; Lo´pez, J .; Lujano, E.; Lo´pez-
Linares, F. J . Mol. Catal. A 2004, 208, 67.
(37) Whitesides, G. M.; Hackett, M.; Brainard, R. L.; Lavalleye, J .-
P. P. M.; Sowinski, A. F.; Izumi, A. N.; Moore, S. S.; Brown, D. W.;
Staudt, E. M. Organometallics 1985, 4, 1819.
(38) Cusanelli, A.; Frey, U.; Richens, D. T.; Merbach, A. E. J . Am.
Chem. Soc. 1996, 118, 5265.
(39) Subongkoj, S.; Lange, S.; Chen, W.; Xiao, J . J . Mol. Catal. A
2003, 196, 125.
(40) Dierkes, P.; van Leeuwen, P. W. N. M. J . Chem. Soc., Dalton
Trans. 1999, 1519.

4852 Organometallics, Vol. 23, No. 21, 2004
Daguenet et al.
F igu r e 3. Influence of the diphosphine on the rate of styrene hydrogenation using 1-3 as catalyst precursors: 1 (solid
squares), 2 (empty triangles), 3 (empty circles). (A) Homogeneous organic phase. Concentrations were determined by in
1
situ high-pressure H NMR spectroscopy by integration of the signals at long relaxation delay. Data was fitted using a
monoexponential (straight lines). Reaction conditions: 80 °C, p(H2) ) 45 bar, substrate/catalyst ) 1500, in toluene-d8. (B)
Aqueous biphase. Conversions were determined by GC analysis after reaction in an autoclave, addition of mercury with
1 (empty diamonds). Conditions: 100 °C, p(H2) ) 45 bar, substrate/catalyst ) 2000, in water.
The kinetics of hydrogenation with 1-3 under bipha-
sic conditions does not follow the same behavior as those
observed under homogeneous conditions (Figure 3, B).
In fact, an induction period is observed showing that
additional factors influence the reaction. The bite angle
effect is presumably counterbalanced by other phenom-
ena including differences in solubility of the complexes
in water as well as phase transfer of the complexes into
the organic layer. Indeed, after biphasic hydrogenation,
the organic phase is colored, indicating contamination
by the complexes and limiting their utility in biphasic
catalysis. A mechanistic study is not feasible since the
31P NMR spectrum of the organic phase recorded after
catalysis is complicated and the presence of free p-
1
cymene is detected in the H NMR spectrum.
F igu r e 4. Kinetics of styrene hydrogenation using 4 as
the catalyst precursor. Conversions determined by GC
analysis. Conditions: 100 °C, p(H2) ) 45 bar, substrate/
catalyst ) 2000, styrene (1 mL), H2O (1.5 mL).
To delineate the differences between the aqueous-
biphase and homogeneous organic systems, dppbts
derivatives were used to immobilize the ruthenium(II)-
arene complexes in water. Thus, complexes 4-6 were
evaluated as catalysts for hydrogenation of styrene in
that sulfonate groups participate in catalytic activity,
water. Under the same conditions as those mentioned
possibly by increasing the solubility of styrene in water.
above, 100% conversion to ethylbenzene as the only
The behavior of the kinetics for the hydrogenation of
styrene with 4 under biphasic conditions is presented
in Figure 4. An induction period of ca. 20 min is
observed, which might correspond to the formation of
product was observed for each compound after 2 h.
Poisoning experiments with mercury were performed,
and the activity of complexes 4-6 was completely
suppressed. However, it is known that mercury can
an active catalytic species. The kinetics are then defined
sometimes also react with a single-metal complex and
by an essentially linear trend, probably because of the
deactivate the catalyst, leading then to an ambiguous
low solubility of styrene in water, which provides a
conclusion.41,42 In the present case, the sulfonate groups
constant concentration until almost all the substrate is
might be responsible for this interaction since they have
converted to ethylbenzene. Considering the concentra-
been reported to adsorb on mercury.43 Furthermore,
tion of styrene as a limiting factor, giving rise to a zero-
below pH 3, catalysis is inhibited; in this pH range, the
order dependence, it seems likely that the reaction is
first protonation of a sulfonate group could partially
taking place in the water bulk.
occur and deactivate the complex. Indeed, at pH < 2,
The catalytic hydrogenation performed with 6 results
in 100% conversion of the substrate even after 1 h of
reaction, demonstrating a higher activity for the [2.2]-
the protonated complex [M + H]2- is the only ruthenium
species observed on the ESI mass spectrum of 4, and
although the proton affinity in the gas phase does not
paracyclophane derivative, which will be discussed later
exactly reflect the one in solution, it is reasonable to
on.
assume that protonation is still taking place to some
ESI-MS was used to analyze the aqueous phase after
extent in the aqueous phase. These elements indicate
catalysis; previously this technique has been used to
provide crucial information toward understanding reac-
tion mechanisms in water.44 The ESI mass spectra of
(41) Widegren, J . A.; Bennett, M. A.; Finke, R. G. J . Am. Chem. Soc.
2003, 125, 10301.
(42) Widegren, J . A.; Finke, R. G. J . Mol. Catal. A 2003, 191, 187.
(43) Hubbard, H. M.; Reynolds, C. A. J . Am. Chem. Soc. 1954, 76,
4300.
(44) Hayashi, H.; Ogo, S.; Abura, T.; Fukuzumi, S. J . Am. Chem.
Soc. 2003, 125, 14266.

Hydrogenation of Alkenes
Organometallics, Vol. 23, No. 21, 2004 4853
Sch em e 1. Heter olytic Activa tion of Dih yd r ogen
by Com p lex 4
Ta ble 3. Hyd r ogen a tion of Styr en e in Ba sic
Aqu eou s Solu tion (p H 8)a
compound
conversion/%
selectivity/%b
2
100
<2
100
100
10
8
4
5c
6
62
97
100
6d
a
Conditions: 100 °C, p(H2) ) 45 bar, substrate/catalyst ) 2000,
b
2 h. Selectivity toward ethylbenzene, byproduct ) ethylcyclo-
hexane. c Decomposition of the complex. Reaction time ) 1 h.
d
dihydrogen complex intermediate, as illustrated in
Scheme 1 in the case of the p-cymene derivative. Due
to the poor electron-donating properties of the ancillary
ligands and the lower stability of the +IV oxidation state
of ruthenium, the dihydrogen complex is preferred
compared to the dihydride, although a dynamic equi-
librium between these two species cannot be completely
ruled out.
The second feature to note from the ESI mass spectra
is that arene exchange has taken place during catalysis.
This point provides clues concerning the mechanism of
coordination of the substrate and will be discussed
below.
To establish the correlation between the activation
of dihydrogen and the hydrogenation of styrene, the
catalytic reactions were carried out in a basic buffer
solution at pH 8. Conversion and selectivity for the
different precursors are collected in Table 3. As previ-
ously reported for complex 1, the reaction performed
under basic conditions using 2 results in the hydrogena-
tion of aromatics due to nanoparticle formation.45 Thus,
the selectivity for the reaction with 2 is very low since
further hydrogenation to ethylcyclohexane takes place
in basic solution. If nanoparticles were involved in the
hydrogenation using the dppbts derivatives, the same
behavior should be observed. This is the case for the
benzene complex 5, which decomposes to give black
particles under these conditions and provides 38% of the
fully hydrogenated product. In contrast, with the p-
cymene analogue 4 the reaction is inhibited, with less
than 2% conversion observed. ESI-MS analysis of the
aqueous phase showed that 4 has been completely
converted into the hydride 8, and since no hydrogena-
tion of the aromatic ring has taken place, the formation
of nanoparticles seems unlikely. Furthermore, enhance-
ment of the heterolytic cleavage of dihydrogen by a base
F igu r e 5. Species observed in the aqueous phase by ESI-
MS after 2 h catalysis starting from precursors 4 (A),
5 (B), and 6 (C, top: 1 h, bottom: 2 h), P-P ) dppbts.
Conditions: 100 °C, p(H2) ) 45 bar, substrate/catalyst )
2000.
4-6 are shown in Figure 5. The first feature to note is
the formation of hydride species, which is accompanied
by a decrease of the pH that was measured after the
reaction was quenched. Combined, these observations
demonstrate the heterolytic cleavage of dihydrogen.
Activation of H2 presumably takes place via a η2-
(45) Daguenet, C.; Dyson, P. J . Catal. Commun. 2003, 4, 153.

4854 Organometallics, Vol. 23, No. 21, 2004
Daguenet et al.
dramatically decreases the activity toward hydrogena-
tion of the vinyl group. The hydride 8, which displays a
triplet at δ -11.4 (2J HP ) 35 Hz) in the 1H NMR
spectrum, has been isolated and redissolved in pure
water (pH 7) for catalysis, but still no hydrogenation
was observed. Kalck et al. evaluated a related hydride
ruthenium(II)-arene complex in diethyl ether, and al-
though the conditions were different, they also noted
that it was inactive toward alkene hydrogenation.46 This
is a rather unexpected result since activation of dihy-
drogen by ruthenium complexes is known to afford
hydride compounds, which are believed to be active
species for CdC bond hydrogenation.18,19,47-53 In the
present study, the active catalyst is thought to be the
dihydrogen complex formed after the thermally induced
dissociation of the chloride. Such active species have
been previously proposed for alkene hydrogenation
involving hydrogen transfer to the substrate directly
from a molecular hydrogen complex.20,54 Assuming that
the same catalytic pathway occurs in a homogeneous
organic phase, such as toluene, using precursors 1-3,
a dihydrogen complex as the active species is more likely
than a monohydride complex since the heterolytic
cleavage of H2, in nonpolar organic solvents and in the
absence of a base, is not favored. Attempts have been
made to detect the dihydrogen complex 7 under a high
pressure of H2 using 1H NMR spectroscopy in methanol-
d4, but unfortunately, due to the high lability of the
dihydrogen ligand in this complex, only the hydride 8
was detected even at low temperatures.
Concerning the [2.2]paracyclophane derivative, there
is also a decrease of the activity under basic conditions,
and after 1 h of reaction, analysis of the solution by ESI-
MS reveals a large proportion of the η6-[2.2]paracyclo-
phane derivatives. Nevertheless after 2 h of reaction the
same conversion as in pure water is measured and the
arene exchange product [Ru(dppbts)(η6-ethylbenzene)H]3-
dominates the ESI-MS spectrum. The rate of reaction
is reduced in a basic medium, but catalytic hydrogena-
tion can still take place without much decomposition,
as indicated by the high selectivity. In relation to that,
the [2.2]paracyclophane dihydrogen complex must be
more stable than the p-cymene analogue, i.e., has a
higher pKa. A possible explanation would be that the
lower electron density at the metal, in the case of the
dihydrogen complex relative to the hydride, could be
more easily stabilized by the [2.2]paracyclophane ligand
compared to p-cymene due to an additional transition
from orbitals of unbound deck π-character via the bound
deck to the metal center.55
although the complexes are not lost to the organic phase
and are well retained in the aqueous phase, catalytic
activity is considerably reduced even after the second
run. This reduction of activity is consistent with the
previous results since most of the starting complex,
[Ru(η6-arene)(dppbts)Cl]Na3, is converted into hydride
species, which have been shown to be inactive. The pH
of the solution is decreased after the first run, presum-
ably due to the heterolytic cleavage of dihydrogen,
depicted in Scheme 1. The protonation/deprotonation
equilibrium that takes place under high pressure of
dihydrogen, in addition to ligand exchange equilibrium
(see section on H/D isotope exchange), must be com-
pletely displaced toward the formation of the hydride
after the gas is released. These acidic conditions, as
discussed above, are not conducive to the catalytic
hydrogenation of styrene.
Ar en e Exch a n ge. As mentioned above, arene ex-
change takes place when catalyses are carried out with
compounds 4-6. In fact [Ru(dppbts)(η6-ethylbenzene)H]3-
is observed to varying extents in the ESI mass spectra
of each reaction (Figure 5). In addition, the 1H NMR
spectrum of the aqueous phase after catalysis with 5,
for example, showed a triplet for the hydride signal at
δ -11.20 with J HP ) 34 Hz, and the chemical shifts
corresponding to the coordinated ethylbenzene appeared
at δ 0.62 (t, 3H), 1.99 (q, 2H), 5.80 (d, 2H), and 5.87-
5.93 (m, 3H). Investigations on the exchange reaction
have been conducted on 4 using slightly different
conditions than those applied to the catalytic experi-
ments. To see if the exchange could occur without
catalytic hydrogenation of the vinyl group, ethylbenzene
was used instead of styrene while the partial pressure
of dihydrogen and the temperature were kept the same
as for catalysis. The ESI mass spectrum of the reaction
mixture showed the presence of 4 as well as the
corresponding hydride, but no hydride resulting from
arene exchange was observed. It must be noted that
chloride complexes with an exchanged arene have not
been detected after catalysis. In the same way, reaction
of 4 with a mixture of styrene/ethylbenzene without
pressurization with dihydrogen did not lead to any
arene exchange. According to these results, both the
substrate and the dihydrogen are required to induce
arene exchange. In addition, the exchange must occur
after the loss of chloride, namely, on the dihydrogen
complex. Due to the strong trans-influence of hydride
ligands, the labilization of the arene, generating an
exchange reaction, could also occur on the complex
[Ru(dppbts)(η6-arene)H]3-. Therefore the same reactions
have been carried out starting from compound 8, but
only traces of [Ru(dppbts)(η6-ethylbenzene)H]3- were
detected. It can therefore be concluded that the ex-
change on complex 4 is facilitated during the catalytic
hydrogenation of styrene and is taking place on the
dihydrogen complex 7 through a labilization of the
p-cymene ligand by the substrate. Moreover, the use of
an aromatic cosolvent such as toluene in catalysis leads
to the formation of [Ru(dppbts)(η6-toluene)H]3- in ad-
dition to the other species. Bennett and co-workers have
previously described the labilization of coordinated
naphthalene (a relatively labile arene ligand), in which
two-electron donor ligands such as acetonitrile or phos-
phines can bind to the metal inducing a ring-slippage
Recycling of the aqueous phase containing the ruthe-
nium(II)-arene complexes has been carried out, and
(46) Hernandez, M.; Kalck, P. J . Mol. Catal. A 1997, 116, 131.
(47) Evans, D.; Osborn, J . A.; J ardine, F. H.; Wilkinson, G. Nature
1965, 208, 1203.
(48) Rose, D.; Gilbert, J . D.; Richardson, R. P.; Wilkinson, G. J .
Chem. Soc. A 1969, 2610.
(49) Ogata, I.; Iwata, R.; Ikeda, Y. Tetrahedon Lett. 1970, 34, 3011.
(50) Fahey, D. R. J . Org. Chem. 1973, 38, 3343.
(51) Iwata, R.; Ogata, I. Tetrahedron 1973, 29, 2753.
(52) Yi, C. S.; Lee, D. W.; He, Z.; Rheingold, A. L.; Lam, K.-C.;
Concolino, T. E. Organometallics 2000, 19, 2909.
(53) Wiles, J . A.; Bergens, S. H.; Vanhessche, K. P. M.; Dobbs, D.
A.; Rautenstrauch, V. Angew. Chem., Int. Ed. 2001, 40 (5), 914.
(54) Kirss, R. U.; Eisenschmid, T. C.; Eisenberg, R. J . Am. Chem.
Soc. 1988, 110, 8564.
(55) Swann, R. T.; Hanson, A. W.; Boelkelheide, V. J . Am. Chem.
Soc. 1986, 108, 3324.

Hydrogenation of Alkenes
Organometallics, Vol. 23, No. 21, 2004 4855
Sch em e 2. P r op osed Mech a n ism of Styr en e
spectra shown in Figure 5. The lability of the arene
follows the series benzene > [2.2]paracyclophane >
p-cymene. The greater lability of the benzene ligand,
which can be displaced without addition of substrate,
can be related to the lower stability of complex 5
observed under more drastic conditions (basic medium).
When the catalytic hydrogenation is performed with
vinylcyclohexane instead of styrene, no conversion is
observed for the reactions with 4 and 6, whereas
complex 5 is active, but decomposes during catalysis
because no aromatics are present that may coordinate
to the metal. Since vinylcyclohexane is more crowded
than styrene, its coordination to the metal center is
difficult unless the coordinated arene is (partially)
displaced, which appears to be the case with 5. This test
indicates that, for 4 and 6 at least, the aromatic ring
(primary or exchanged) remains coordinated to ruthe-
nium during the hydrogenation of styrene.
Hyd r ogen a tion Usin g 4, P -P ) d p p bts
H/D Isotop e Exch a n ge. Catalytic isotope exchange
between H2 and D2O or D2 and H2O by transition metal
complexes is believed to take place via a η2-coordinated
dihydrogen.21,60-63 Such dihydrogen complexes must
have both a high kinetic acidity, to incorporate the other
isotope by H/D exchange, and a great lability toward
ligand dissociation, to allow the evolution of the gas.
Recording the evolution of HD in a H2-D2O mixture
by 1H NMR spectroscopy can indirectly demonstrate the
existence of a η2-H2 species. Following the procedure
applied by Lau and co-workers,21,62 a toluene-d8 column
was added over a solution of 5 in D2O and pressurized
with 50 bar of H2 in a sapphire NMR tube. After 20 h
of heating at 80 °C, the 1H NMR spectrum of the organic
phase was recorded. The spectrum displays a triplet at
δ 4.47 with J HD ) 42.7 Hz, corresponding to HD (Figure
S1 in the Supporting Information), indicating the pres-
ence of a η2-H2 intermediate.
from η6 to η4 coordination mode, which is then followed
by the substitution of the naphthalene by another
arene.56,57 Accordingly, in the present case, the alkene
is the two-electron donor ligand inducing the arene
exchange. Hence, a η4 arene complex with a coordinated
substrate can be postulated as an intermediate in the
catalytic cycle proposed in Scheme 2. Species with a η6
coordinated styrene have not been observed after hy-
drogenation even if the reaction is stopped before
completion. Consequently, it is not unreasonable to
assume that the binding of free styrene to the catalytic
intermediate goes preferentially via the CdC bond,
which undergoes hydrogenation, than via the aromatic
ring. Mechanisms involving a ring-slippage to allow the
coordination of an entering ligand (substrate or reac-
tant) have already been suggested for transfer hydro-
genation of ketones.22,58 Assuming a η6 to η4 arene
coordination shift as a step in the catalytic cycle, the
higher activity of complex 6 compared to 4 can be
rationalized by the geometry of the [2.2]paracyclophane
ligand, with a nonplanar aromatic ring, being better
suited to η4 bonding.55 The electronic nature of the [2.2]-
paracyclophane ligand might also play a part in in-
creasing catalytic activity. Indeed, the ease of reduction,
related to the feasible η4 coordination, and the ease of
oxidation, due to the additional electronic donation from
the unbound ring, might facilitate both oxidative addi-
tion and reductive elimination steps in catalytic cy-
cles.55,59
The H/D isotope exchange process has also been
investigated under conditions of catalytic hydrogenation
of styrene. To study this process, 5, which converts
completely into the arene-exchanged hydride complex
during catalysis, was used as precursor. The aqueous
phase of each catalytic experiment, i.e., in D2-D2O, D2-
H2O, H2-D2O, and H2-H2O, was analyzed by ESI-MS
after the reaction (Figure 6). The gray band in Figure 6
represents one unit on the m/z scale, which corresponds
to the difference of three units of molecular weight
between the trianionic species [Ru(dppbts)(η6-C6H5-
CH2CH3)H]3- and [Ru(dppbts)(η6-C6H5CHDCH2D)D]3-
.
Under the D2-H2O and H2-D2O conditions, the spectra
obtained correspond to a mixture of partially deuterated
products (Figure 6, B and C). Considering M as the mass
of the fully hydrogenated compound, the relative inten-
sities of the peaks for the experimental spectra (B and
C) were fitted using a linear combination of the relative
intensities of the simulated isotopic patterns for the
species M, M+1, M+2, and M+3 in order to get an
approximation of their proportions in the mixture. In
The efficiency of arene exchange varies according to
the precursor used, as is evident from the ESI mass
(59) Bond, A. M.; Dyson, P. J .; Humphrey, D. G.; Lazarev, G.;
Suman, P. J . Chem. Soc., Dalton Trans. 1999, 443.
(60) Collman, J . P.; Wagenknecht, P. S.; Hembre, R. T.; Lewis, N.
S. J . Am. Chem. Soc. 1990, 112, 1294.
(61) Kubas, G. J .; Burns, C. J .; Khalsa, G. R. K.; Van Der Sluys, L.
S.; Kiss, G.; Hoff, C. D. Organometallics 1992, 11, 3390.
(62) Lau, C.-P.; Cheng, L. J . Mol. Catal. 1993, 84, 39.
(63) Kova´cs, G.; Na´dasdi, L.; Laurenczy, G.; J oo´, F. Green Chem.
2003, 5, 213.
(56) Bennett, M. A.; Neumann, H.; Thomas, M.; Wang, X. Organo-
metallics 1991, 10, 3237.
(57) Bennett, M. A.; Lu, Z.; Wang, X.; Bown, M.; Hockless, D. C. R.
J . Am. Chem. Soc. 1998, 120, 10409.
(58) Ghebreyessus, K. Y.; Nelson, J . H. J . Organomet. Chem. 2003,
669, 48.

4856 Organometallics, Vol. 23, No. 21, 2004
Daguenet et al.
dihydrogen complex in equilibrium with the hydride.
The dihydrogen species then enters the catalytic cycle
by coordination of the substrate accompanied by a η6 to
η4 ring-slippage, labilizing the arene and facilitating
arene exchange. The dihydrogen is then cleaved and one
H is added to the CdC bond to form a ruthenium(IV)
alkyl hydride intermediate in which the arene coordi-
nates in the usual η6 mode. Reductive elimination of
ethylbenzene would afford the 16-electron species that
can re-enter the cycle by reaction with dihydrogen.
In conclusion, we have provided evidence that is not
inconsistent with the existence of a dihydrogen complex
in the catalytic hydrogenation of styrene using ruthe-
nium(II)-arene precursors. The arene ligand may un-
dergo exchange, but its loss is not a prerequisite for
catalysis as in many reactions involving ruthenium(II)-
arene precatalysts, which are finding ever-increasing
applications in organic synthesis.
Exp er im en ta l Section
Gen er a l P r oced u r es. All organic solvents used were
analytical grade. Deionized water was used for the compound
preparation and doubly distilled water for the catalytic reac-
tions. RuCl3‚xH2O (Apollo), diphenylphosphinomethane (dppm),
diphenylphosphinoethane (dppe), diphenylphosphinopropane
(dppp) (Acros Organics), 1,2-bis(di-4-sulfonatophenylphosphi-
no)benzene (dppbts) tetrasodium salt (Strem Chemicals),
hydrogen 45 (Carbagas), and deuterium 99.6% (Cambridge
Isotope Laboratories Inc.) were used as received. The p-
cymene, benzene, and [2.2]paracyclophane dimers, [RuCl2(η6-
arene)]2, were prepared as previously described.55,64,65 NMR
spectra were recorded on either a Bruker Avance 200 or a
Bruker Avance DPX 400 spectrometer. 1H and 13C chemical
shifts were referenced to residual solvent resonances, and 31P
chemical shifts were referenced to an external 85% aqueous
solution of H3PO4. High gas pressure NMR experiments were
carried out using sapphire NMR tubes.38 ESI mass spectra
were recorded on a Thermo Finnigan LCQ DECA XPPlus
using conditions described previously.24
F igu r e 6. ESI-MS spectra of [H(D)Ru(dppbts)(η6-C6H5-
CH2(HD)CH3(H2D))]3- formed during catalytic hydroge-
nation of styrene with 5 in (A) D2-D2O, (B) D2-H2O, (C)
H2-D2O, and (D) H2-H2O. Reaction conditions: 100 °C,
p(H2 or D2) ) 30 bar, 4 h. The frames correspond to
calculated isotopic patterns for mixtures of products.
the H2-D2O reaction, only two species corresponding
to M (17%) and M+1 (83%) are present. The latter must
be the deuteride complex according to the 1H NMR
spectrum, which shows a rather well-defined triplet at
δ 0.57 and a quartet at δ 1.95 (3J HH ) 7.5 Hz) attributed
to the fully hydrogenated ethyl group of the coordinated
ethylbenzene. In contrast, the product distribution
resulting from the reaction in D2-H2O is more compli-
cated, with a mixture of species of the three different
masses M+1 (40%), M+2 (46%), and M+3 (14%), indi-
cating that other compounds are present.
What is clear from these experiments is that the
incorporation of hydrogen (or deuterium) from H2O (or
D2O) into styrene is more efficient with the D2-H2O
reactants than with H2-D2O. Under the latter condi-
tions, the catalytic hydrogenation of styrene must take
place faster on the η2-H2 complex than on the η2-HD
one and H/D exchange is only observable because the
deuteride species is formed. Conversely, with the D2-
H2O reactants, the H/D exchange has an effect on the
isotopic distribution in the coordinated ethylbenzene;
the reaction must be slower on the η2-D2 complex than
on the η2-HD one, leading to incorporation of hydrogen
into the substrate.
P r ep a r a tion of [(η6-p-Cym en e)(η2-d p p m )Ru Cl]Cl (1).
Compound 1 was prepared as reported earlier.45 The tetrafluo-
roborate analogue of 1 was formed by anion exchange in an
aqueous solution of 1 (ca. 10 mg in 20 mL) saturated with
ammonium tetrafluoroborate (ca. 50 mg). The resulting yellow-
white precipitate was washed with water and diethyl ether.
Crystals suitable for X-ray analysis were grown by slow
evaporation of acetone.
1H NMR (CDCl3) δ (ppm): 7.60-7.37 (m, 20H, C(Ph)H), 6.21
3
3
(d, J HH ) 7 Hz, 2H, CH), 6.11 (d, J HH ) 7 Hz, 2H, CH), 4.77
2
2
(dt, J HH ) 15 Hz, J HP ) 10 Hz, 1H, PCHHP), 4.55 (dt,
2J HH ) 15 Hz, J HP ) 10 Hz, 1H, PCHCHP), 2.48 (sept,
2
3J HH ) 7 Hz, 1H, CH(CH3)2), 1.52 (s, 3H, CH3), 1.05 (d, 3J HH
7 Hz, 6H, CH(CH3)2). 31P NMR (CDCl3) δ (ppm): +0.44.
)
P r ep a r a tion of [(η6-p-Cym en e)(η2-d p p e)Ru Cl]Cl (2). A
suspension of the ruthenium p-cymene dimer (200 mg, 0.327
mmol) in ethanol (40 mL) was added to a solution of dppe (560
mg, 1.41 mmol) in ethanol (100 mL). The mixture was heated
at 50 °C for ca. 20 h. The solution was cooled to room
temperature, filtered, and concentrated to 2 mL. Addition of
diethyl ether (ca. 15 mL) allows formation of a precipitate. The
crude product was then washed with diethyl ether (15 mL),
filtered, and dried to finally yield a yellow solid (337 mg, 73%).
The tetrafluoroborate analogue was formed using the same
method as for the dppm chelate described above. X-ray suitable
These data provide further evidence for an acid-base
dynamic equilibrium between the dihydrogen and the
hydride complexes. The kinetics and thermodynamics
of this equilibrium, which have not been studied in this
work, certainly play an important part in the catalytic
process since the active species is believed to be the
short-lived dihydrogen complex. Based on the results
described herein, the mechanism illustrated in Scheme
2 is proposed.
The loss of the chloride ligand would generate from 4
a 16-electron complex, which might be stabilized by the
formation of an aqua species, and leads to the active
(64) Bennett, M. A.; Huang, T.-N.; Matheson, T. W.; Smith, A. K.
Inorg. Synth. 1982, 21, 74.
(65) Zelonka, R. A.; Baird, M. C. Can. J . Chem. 1972, 50, 3063.

Hydrogenation of Alkenes
Organometallics, Vol. 23, No. 21, 2004 4857
crystals of the tetrafluoroborate analogue were obtained by
slow evaporation of an acetone solution.
extracted with methanol, and then filtered. The filtrate was
evaporated to dryness to yield a light-yellow-gray solid (6 mg,
66%).
1H NMR (CDCl3) δ (ppm): 7.73-7.23 (m, 20H, C(Ph)H), 6.10
3
3
1H NMR (CD3OD) δ (ppm): 7.99-7.37 (m, 20H, C(Ph)H),
(d, J HH ) 6 Hz, 2H, CH), 6.01 (d, J HH ) 6 Hz, 2H, CH), 2.92
3
3
3
(m, 2H, PCH2), 2.48 (m, 2H, PCH2), 2.32 (sept, J HH ) 7 Hz,
5.90 (d, J HH ) 6.4 Hz, 2H, CH), 5.77 (d, J HH ) 6.4 Hz, 2H,
3
CH), 2.21 (sept, 3J HH ) 7 Hz, 1H, CH(CH3)2), 1.76 (s, 3H, CH3),
1H, CH(CH3)2), 1.14 (s, 3H, CH3), 0.78 (d, J HH ) 7 Hz, 6H,
CH(CH3)2). 31P NMR (CDCl3) δ (ppm): +69.4.
3
2
0.87 (d, J HH ) 6.8 Hz, 6H, CH(CH3)2), -11.38 (t, J HP ) 35
P r ep a r a tion of [(η6-p-Cym en e)(η2-d p p p )Ru Cl]Cl (3). A
hot solution of ruthenium p-cymene dimer (100 mg, 0.164
mmol) in ethanol (30 mL) was added to a boiling solution of
dppp (290 mg, 0.703 mmol) in ethanol (40 mL). The reaction
mixture was stirred at reflux temperature for 4 h and then
concentrated to ca. 2 mL. The residue was triturated in diethyl
ether (15 mL) and the precipitate filtered and dried to yield a
yellow-orange solid (145 mg, 60%).
Hz, 1H, RuH). 31P NMR (CD3OD) δ (ppm): 78.0 (d, J PH ) 35
2
Hz). ESI-MS negative ion: m/z 333 [M]3-, 511 [M + Na]2-
.
Ca ta lysis. All catalytic experiments were conducted using
a home-built multicell autoclave at 100 °C pressurized with
either hydrogen or deuterium. To each reaction vessel the
catalyst (ca. 4 × 10-3 mmol) and the substrate (1 mL) were
weighted and the solvent (1.5 mL) was added via syringe. The
catalytic runs were carried out using varying lengths of time
and the percent conversions determined by GC analysis of the
organic phase performed with a Varian Chrompack CP-3380
gas chromatograph. The mercury poisoning experiments were
conducted in the same way with the addition of about 100 µL
of Hg(0) to the reaction vessel.
Da ta An a lysis. Conversions after catalysis were deter-
mined by integration of the corresponding peaks on the
chromatograms using Varian software. Integrations of the
NMR signals for the kinetics studied by high gas pressure
NMR measurements were performed with a Levenberg-
Marquard fitting procedure on the NMRICMA program work-
ing on a Matlab platform.66 Initial rates were calculated from
the parameters determined with a pseudo-first-order kinetics
model by a least-squares fit of the data using Scientist for
Windows.
X-r a y Ch a r a cter iza tion of 1‚BF 4 a n d 2‚BF 4. Details
about the crystals and their structure refinement are listed
in Table 1, and relevant geometrical parameters, including
bond lengths and angles are included in the figure captions
(see Figures 1 and 2). Data collection for 1‚BF4 was performed
at room temperature on a mar345 IPDS and for 2‚BF4 at 140
K on a four-circle goniometer having kappa geometry and
equipped with an Oxford Diffraction KM4 Sapphire CCD. Data
reduction was carried out on both data collections with
CrysAlis RED, release 1.7.0.67 Absorption corrections have
been applied to both data sets. Structure solution and refine-
ment as well as molecular graphics and geometrical calcula-
tions were performed for both structures with the SHELXTL
software package, release 5.1.68 The structures were refined
using full-matrix least-squares on F2 with all non-H atoms
anisotropically defined. H atoms were placed in calculated
positions using the “riding model”. Some disorder (see Figure
1) has been encountered during the refinement of 1‚BF4 and
has been treated splitting the phenyl ring into two different
orientations (named A and B), having the same occupancy
factor and the same regular and fixed geometry.
1H NMR (CDCl3) δ (ppm): 7.66-7.25 (m, 20H, C(Ph)H), 6.18
(d, 3J HH ) 6 Hz, 2H, CH), 5.79 (d, 3J HH ) 6 Hz, 2H, CH), 3.11-
3.03 (m, 2H, CH2), 2.25-2.19 (m, 4H, PCH2), 2.09 (sept,
3J HH ) 7 Hz, 1H, CH(CH3)2), 1.55 (s, 3H, CH3), 0.83 (d, 3J HH
)
7 Hz, 6H, CH(CH3)2). 31P NMR (CDCl3) δ (ppm): +23.0, +22.8.
P r ep a r a tion of [(η6-p-Cym en e)(η2-d p p bts)Ru Cl]Na3 (4).
A suspension of the ruthenium p-cymene dimer (35 mg, 0.057
mmol) in ethanol (5 mL) was added to a solution of dppbts
(130 mg, 0.15 mmol) in water (5 mL). The mixture was stirred
at room temperature for 30 min and then concentrated to ca.
1 mL. Addition of ethanol leads to precipitation of the product,
which was washed by decantation with ethanol and diethyl
ether and then dried to yield a yellow-orange solid (100 mg,
80%).
Anal. Calcd for RuC40H34S4O12P2ClNa3‚4H2O: C, 40.91; H,
3.60. Found: C, 41.34; H, 3.69. 1H NMR (CD3OD) δ (ppm):
8.07-7.23 (m, 20H, C(Ph)H), 6.27 (s, 4H, CH), 2.45 (sept,
3J HH ) 7 Hz, 1H, CH(CH3)2), 1.37 (s, 3H, CH3), 0.87 (d, 3J HH
)
7 Hz, 6H, CH(CH3)2). 31P NMR (CD3OD) δ (ppm): +67.1.
ESI-MS negative ion: m/z 344 [M]3-, 516 [M + H]2-, 528
[M + Na]2-
.
P r ep a r a tion of [(η6-Ben zen e)(η2-d p p bts)Ru Cl]Na 3 (5).
A suspension of the ruthenium benzene dimer (30 mg, 0.060
mmol) in ethanol (5 mL) was added to a solution of dppbts
(130 mg, 0.15 mmol) in water (5 mL). The mixture was stirred
at room temperature for ca. 20 min and then concentrated to
1-2 mL. Addition of ethanol leads to precipitation of the
product, which was washed by decantation with ethanol and
then dried to yield a yellow-orange solid (70 mg, 52%).
Anal. Calcd for RuC36H26S4O12P2ClNa3‚4H2O: C, 38.66; H,
3.06. Found: C, 39.16; 3.27. 1H NMR (CD3OD) δ (ppm): 8.07-
7.14 (m, 20H, C(Ph)H), 6.22 (s, 6H, CH). 31P NMR (CD3OD)
δ (ppm): +65.1. ESI-MS negative ion: m/z 326 [M]3-, 489
[M + H]2-, 500 [M + Na]2-
.
P r ep a r a tion of [(η6-p-Cyclop h a n e)(η2-d p p bts)Ru Cl]-
Na 3 (6). A suspension of the ruthenium [2.2]paracyclophane
dimer (30 mg, 0.039 mmol) in ethanol (4 mL) was added to a
solution of dppbts (95 mg, 0.12 mmol) in water (4 mL). The
mixture was stirred at room temperature for ca. 30 min,
filtered, and concentrated to 1 mL. Addition of ethanol leads
to precipitation of the product, which was washed by decanta-
tion with ethanol and then dried to yield a yellow-orange solid
(46 mg, 47%). Anal. Calcd for RuC46H36S4O12P2ClNa3‚5H2O:
Ack n ow led gm en t. We thank Novartis, the EPFL,
and Swiss National Science Foundation for financial
support.
Su p p or tin g In for m a tion Ava ila ble: Crystallographic
information for 1‚BF4 and 2‚BF4 in CIF format, 2D 13C-1H
NMR data for 4-6, and Figure S1. This material is available
free of charge via the Internet at http://pubs.acs.org.
1
C, 43.62; H, 3.66. Found: C, 43.28; 3.71. H NMR (CD3OD) δ
(ppm): 7.96-6.92 (m, 20H, C(Ph)H), 6.90 (s, 4H, CH), 5.21 (s,
4H, CH), 3.28-3.24 (m, 4H, CH2), 2.94-2.90 (m, 4H, CH2).
31P NMR (CD3OD) δ (ppm): +72.9. ESI-MS negative ion: m/z
369 [M]3-, 554 [M + H]2-, 565 [M + Na]2-
.
OM049665Q
P r ep a r a tion of [(η6-p-Cym en e)(η2-d p p bts)Ru H]Na 3 (7).
In a stainless autoclave, [(η6-p-cymene)(η2-dppbts)RuCl]Na3‚
4H2O (10 mg, 0.0085 mmol) was added to a 50 mM phosphate
buffer pH 8 (2.5 mL). The system was pressurized at 45 bar
with dihydrogen and heated to 100 °C for 2 h. The gray
reaction mixture was concentrated under reduce pressure,
(66) Helm, L.; Borel, A. NMRICMA 2.8; Lausanne, Switzerland,
2001.
(67) CrysAlis RED, release 1.7.0; Oxford Diffraction Ltd.: Abingdon,
Oxfordshire, UK, 2003.
(68) Sheldrick, G. M. SHELXTL; University of Go¨ttingen: Germany,
1997; Bruker AXS, Inc., Madison, WI, 1997.