Mononuclear ruthenium complexes containing two different phosphines in trans position: II. Catalytic hydrogenation of C{double bond, long}C and C{double bond, long}O bonds

Article abstract of DOI:10.1016/j.jorganchem.2006.11.036

Bis(acetate) ruthenium(II) complexes of the general formula Ru(CO)₂(OAc)₂(PⁿBu₃) [P(p-XC₆H₄)₃] (OAc = acetate, X = CH₃O, CH₃, H, F or Cl), containing different phosphine ligands trans to PⁿBu₃, have been employed as catalyst precursors for the hydrogenation of 1-hexene, acetophenone, 2-butanone and benzylideneacetone. For comparative purposes, analogous reactions have been performed using the homodiphosphine precursors Ru(CO)₂(OAc)₂(PⁿBu₃)₂ and Ru(CO)₂(OAc)₂(PPh₃)₂. The catalytic activity of the heterodiphosphine complexes depends on the basicity of the triarylphosphine trans to PⁿBu₃ as this factor controls, inter alia, the rate of formation of hydride(acetate), Ru(CO)₂(H)(OAc)(PⁿBu₃)[P(p-XC₆ H₄)₃], or dihydride, Ru(CO)₂(H)₂(PⁿBu₃)[(p-XC ₆H₄)₃], complexes, by hydrogenation of the bis(OAc) precursors. The catalytic hydrogenation of the C{double bond, long}C double bond is best accomplished by homodiphosphine dihydride catalysts, while heterodiphosphine monohydrides are more efficient catalysts than the homo- and heterodiphosphine dihydrides for the reduction of the keto C{double bond, long}O bond.

Full text of DOI:10.1016/j.jorganchem.2006.11.036

Journal of Organometallic Chemistry 692 (2007) 1442–1450
www.elsevier.com/locate/jorganchem
Mononuclear ruthenium complexes containing two different phosphines
in trans position: II. Catalytic hydrogenation of C@C and C@O bonds
a
a,
a
b
Luca Salvi , Antonella Salvini *, Francesca Micoli , Claudio Bianchini ,
b
Werner Oberhauser
a
Department of Organic Chemistry, University of Florence, Via della Lastruccia 13, 50019 Sesto Fiorentino, Florence, Italy
b
ICCOM-CNR, Area di Ricerca CNR, Via Madonna del Piano 10, 50019 Sesto Fiorentino Florence, Italy
Received 24 May 2006; received in revised form 31 October 2006; accepted 22 November 2006
Available online 1 December 2006
Abstract
Bis(acetate) ruthenium(II) complexes of the general formula Ru(CO)2(OAc)2(PnBu3)[P(p-XC6H4)3] (OAc = acetate, X = CH3O, CH3,
H, F or Cl), containing different phosphine ligands trans to PnBu3, have been employed as catalyst precursors for the hydrogenation of 1-
hexene, acetophenone, 2-butanone and benzylideneacetone. For comparative purposes, analogous reactions have been performed using
the homodiphosphine precursors Ru(CO)2(OAc)2(PnBu3)2 and Ru(CO)2(OAc)2(PPh3)2. The catalytic activity of the heterodiphosphine
complexes depends on the basicity of the triarylphosphine trans to PnBu3 as this factor controls, inter alia, the rate of formation of
hydride(acetate), Ru(CO)2(H)(OAc)(PnBu3)[P(p-XC6H4)3], or dihydride, Ru(CO)2(H)2(PnBu3)[(p-XC6H4)3], complexes, by hydrogena-
tion of the bis(OAc) precursors. The catalytic hydrogenation of the C@C double bond is best accomplished by homodiphosphine dihy-
dride catalysts, while heterodiphosphine monohydrides are more efficient catalysts than the homo- and heterodiphosphine dihydrides for
the reduction of the keto C@O bond.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Hydrogenation; Phosphines; Trans effect; Ruthenium; Ketones; Alkenes
1. Introduction
plexes Ru(CO)2(H)2(PnBu3)[(p-XC6H4)3] (2a–e) is faster [2]
and occurs at lower temperature as compared to the known
homodiphosphine complexes Ru(CO)2(OAc)2(PPh3)2 (4)
and Ru(CO)2(OAc)2(PnBu3)2 (5) [4].
The formation of the dihydride complexes 2a–e upon
hydrogenation of the bis(OAc) precursors is apparently
preceded by that of the monohydride compounds Ru(CO)2
(H)(OAc)(PnBu3)[P(p-XC6H4)3] (3a–e), but bis(OAc), dihy-
drides and monohydrides are in equilibrium, unless acetic
acid is removed by treatment of the solution with an appro-
priate base such as Na2CO3 [2].
Since the homodiphosphine complexes 4 and 5 share
many reactivity aspects with 1a–e, especially as regards
the reactions with H2 [4], and have also been employed
with success as catalysts for the homogeneous hydrogena-
tion of alkynes [5], alkenes [5] and ketones [6], we decided
to study analogous reactions using 1a–e. The presence of
two trans phosphines with different basicity in the latter
Some of us have recently reported the synthesis of
mononuclear ruthenium complexes of the formula
Ru(CO)2(Y)(Y0)(PnBu3)[P(p-XC6H4)3] containing a PnBu3
ligand in trans position to triarylphosphines differently
substituted in the para position of the aromatic ring
(X = CH3O, CH3, H, F or Cl), and where Y and Y0 may
be equal (OAc or H) or different from each other (OAc
or H) (Scheme 1) [1,2]. The triarylphosphine ligands are
kinetically labile in all complexes, due to an effective trans
effect [3] caused by the greater basicity of the trans PnBu3
ligand. In particular, the existence of a trans effect was
proved for the reactions of 1a–e with hydrogen in benzene
[2] (Scheme 1). Indeed, the formation of the dihydride com-
*
Corresponding author. Tel.: +39 055 4573455; fax: +39 055 4573531.
E-mail address: antonella.salvini@unifi.it (A. Salvini).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.11.036

L. Salvi et al. / Journal of Organometallic Chemistry 692 (2007) 1442–1450
1443
HOAc
PnBu3
PnBu3
CO
PnBu3
H2
P(p-XC6H4)3
AcO
H
AcO
AcO
AcO
H
P(
p-XC6H4)3
CO
CO
CO
CO
Ru
Ru
Ru
CO
H2
P(p-XC6H4)3
P(p-XC6H4)3
P(p-XC6H4)3
HOAc
3a-e
P(
p
-XC6H4)3
H2
1a-e
H2
HOAc
HOAc
H2
X = OCH3, a; CH3, b; H,c; F,d; Cl,e
PnBu3
HOAc
H
CO
Ru
H
CO
-XC6H4)3
P(p
2a-e
Scheme 1.
complexes and the possibility to vary the nucleophilicity of
the P(p-XC6H4)3 ligands [7], by changing the X substituent
without affecting the steric properties (the cone angle does
not vary within the triarylphosphines examined [8]), would
create suitable conditions to evaluate the role of the trans
effect in the catalytic reactions. This study would also pro-
vide information on the reaction mechanism, in particular
as regards the occurrence of phosphine dissociation for
substrate activation. This question is still a matter of
debate for the hydrogenation of ketones or a,b-unsaturated
ketones catalyzed by Ru(CO)2(H)2(PPh3)2 that has been
reported to involve substrate-association by some authors
[6] and phosphine-dissociation by other authors [9,10].
upon heating solutions of the heterodiphosphine precur-
sors [2]. In an attempt of evaluating the phosphine equili-
bration process under the catalytic conditions of the
present hydrogenation reactions, 1c was heated in benzene
at different temperatures and H2 pressures. The products
obtained were analyzed by 1H and 31P{1H} NMR
spectroscopy.
2.1.1. Thermal stability of 1c
After a C6D6 solution of 1c was heated under nitrogen
at 80 ꢁC for 3 h, a 31P{1H} NMR spectrum was acquired,
showing the presence of the homodiphosphine products 4
and 5 in trace amount (<2%). The formation of the latter
compounds became important (45%) only after heating at
100 ꢁC for 24 h.
2. Results and discussion
2.1. Reactivity of Ru(CO)2(OAc)2(PnBu3)[P(C6H5)3]
(1c)
2.1.2. Reactivity of 1c with H2
The reactions of 1c with H2 were carried out in C6D6
and the products obtained were identified by NMR spec-
troscopy (Table 1). In most reactions, anhydrous sodium
carbonate was added to the reaction mixtures to neutralize
the evolved acetic acid and ultimately make irreversible the
formation of the dihydride 2c (Scheme 2) [2]. The monohy-
A previous study has shown that the P(p-XC6H4)3
ligands in 1a–1e are kinetically labile in solution so that
small amounts of the homodiphosphine complexes Ru
(CO)2(OAc)2[P(p-XC6H4)3]2 and 5 are invariably formed
Table 1
Product distribution with time and temperature upon hydrogenation of 1c at 100 bar H2
Entry
Reaction conditions
Products (%)a
T (ꢁC)
Reaction time (h)
1c
2c
3c
4
5
6
7
8
1
2
3
4
a
20b
40b
60b
60c
24
16
16
24
100
0
0
0
84
80
60
0
0
0
0
1
0
0
0
6
8
0
7
10
9
0
2
2
0
0
0
0
2
0
18
11
Calculated by 31P NMR integration.
In the presence of Na2CO3.
Without Na2CO3.
b
c

1444
L. Salvi et al. / Journal of Organometallic Chemistry 692 (2007) 1442–1450
Na2CO3, H2
reaction parameters such as reaction time, temperature
NaOAc, CO2, H2O
and H2 pressure was evaluated for selected catalyst precur-
sors. The substrate/catalyst ratio was fixed to 100:1, while
the substrate conversion was determined by GC and GC–
MS techniques.
Ru(CO)2(OAc)2(PnBu3)(PPh3)
Ru(H)2(CO)2(PnBu3)(PPh3)
2c
1c
Scheme 2.
All tests were performed twice in order to confirm the
reproducibility of the experiments.
dride RuH(CO)2(OAc)(PnBu3)(PPh3) (3c) was intercepted
only in the absence of Na2CO3.
At 20 ꢁC, even with added Na2CO3, 1c was fully stable
for at least 24 h under 100 bar H2 (entry 1). Only by heat-
ing to 40 ꢁC, the complete conversion of 1c was achieved to
give 2c in 84% yield together with the homodiphosphine
acetates (7%) and the dihydrides Ru(CO)2(H)2(PPh3)2 (6)
(7%) and Ru(CO)2(H)2(PnBu3)2 (7) (2%) (entry 2). Increas-
ing the temperature to 60 ꢁC for 16 h did not improve
appreciably the selectivity; a lower production of the dihy-
dride 2c was observed in fact, while all formed 4 was con-
verted to the corresponding dihydride 6 (entry 3). These
results are in agreement with the previously reported phos-
phine-redistribution reaction exhibited by 1c to give 4 and
5.
In the absence of Na2CO3 (entry 4), 2c was still the
major product (60%), but a significant amount (18%) of
the monohydride 3c was also formed together with 5 and
6 as well as an appreciable amount of the homodiphos-
phine monohydride RuH(CO)2(OAc)(PPh3)2 (8). On leav-
ing this reaction mixture at room temperature for 6 h
under nitrogen, the amount of 3c increased to 58% due
to the reaction of 2c with the free acetic acid present in
solution.
2.2.1. Hydrogenation of 1-hexene
The hydrogenation of 1-hexene catalyzed by either 1c or
2c was carried out in toluene at 50 bar H2 for a reaction
time of 3 h in the temperature range from 40 to 80 ꢁC.
The results obtained are reported in Table 2.
As a general trend, the conversion to n-hexane increased
with the temperature and the dihydride 2c proved invari-
ably better than the bis(OAc) precursor 1c, which is consis-
tent with previous results obtained with the bis(PnBu3)
couple 5/7 [5]. In order to observe a significant conversion
to n-hexane, the temperature was increased to 80 ꢁC (25.4%
with 1c and 40.8% with 2c). At this temperature, however,
extensive isomerization of 1-hexene occurred with a preva-
lence of trans- and cis-2-hexenes. Again, the dihydride cat-
alyst was more efficient than the bis(OAc) precursor,
yielding 55.8% of isomerized alkenes. It is noteworthy that
the ratio between n-hexane and isomerized products
increased with the temperature, consistent with a higher
energy barrier to hydrogenation of internal double bonds
as compared to the terminal one. An independent reaction
at 80 ꢁC in the presence of 1 equiv. of PPh3 showed in
either case a dramatic drop of the conversion (from
54.5% to 2.5% and from 96.6% to 33.2% for 1c and 2c,
respectively). The same trend was observed for the isomer-
ization reaction yielding trans-2-hexene.
2.2. Catalytic hydrogenation reactions
The catalytic performance of selected heterodiphosphine
bis(acetate) complexes was investigated in the hydrogena-
tion of 1-hexene, two ketones (acetophenone and butan-
2-one) and the a,b-unsaturated ketone trans-4-phenyl-3-
buten-2-one. No yield optimization was attempted, the
study being exclusively centered on a comparison of the
activity of the different catalyst precursors, in an attempt
of relating catalytic activity and selectivity with the nature
of the phosphine trans to PnBu3. The influence of some
A comparison of the catalytic activity of 1c and 2c ver-
sus that of the homodiphosphine complexes 4, 5, 6 and 7
under comparable experimental conditions is reported in
Table 3.
Irrespective of the phosphine ligands, the catalytic activ-
ity of the dihydride complexes was higher than that of the
corresponding bis(OAc) precursors. Among the latter, the
bis(PPh3) derivative 4 showed the highest conversion of
1-hexene, yet this was largely due to isomerized products,
Table 2
Hydrogenation of 1-hexene catalyzed by 1c and 2c at different temperatures
Catalyst
T (ꢁC)
Conversion (%)
Reaction products composition (%)
n-Hexane
trans-3-Hexene
trans-2-Hexene
cis-3-Hexene
1-Hexene
cis-2-Hexene
1c
2c
1c
2c
1c
1ca
2c
2ca
40
40
60
60
80
80
80
80
0.9
12.2
15.2
59.0
54.5
2.5
0.9
4.7
5.9
8.1
25.4
1.3
0
0
2.4
3.9
28.2
14.9
0.7
29.0
3.0
0
99.1
87.8
84.8
41.0
45.5
97.5
3.4
0
3.1
3.3
19.3
11.0
0.5
13.3
2.8
0.9
1.3
2.5
2.2
0
1.1
0.8
0.9
1.0
0
96.6
33.2
40.8
25.6
11.2
1.0
2.3
0.8
66.8
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; p(H2), 50 bar at 20 ꢁC; reaction time, 3 h.
a
PPh3 in excess: PPh3/Ru molar ratio 2:1.

L. Salvi et al. / Journal of Organometallic Chemistry 692 (2007) 1442–1450
1445
Table 3
Hydrogenation of 1-hexene catalyzed by selected hetero- and homodiphosphine complexes
Catalyst
Conversion (%)
Reaction products composition (%)
n-Hexane
trans-3-Hexene
trans-2-Hexene
cis-3-Hexene
1-Hexene
cis-2-Hexene
5
41.8
86.1
54.5
96.6
92.1
97.8
19.1
45.4
25.4
40.8
21.0
77.3
2.8
3.5
2.2
11.2
13.3
1.6
10.1
24.7
14.9
29.0
40.0
11.1
0.8
1.0
1.0
2.3
3.4
2.8
58.2
13.9
45.5
3.4
7.9
2.2
9.0
11.5
11.0
13.3
14.4
5.0
7
1c
2c
4
6
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; p(H2), 50 bar at 20 ꢁC; reaction time, 3 h; T, 80 ꢁC.
while 1c was the most effective catalysts for the hydrogena-
tion of the C@C double bond. It is worth noticing, how-
ever, that the experimental conditions of the reactions
reported in Table 3 do not allow the complete transforma-
tion of the bis(OAc) complexes into the corresponding
dihydrides (2c, 6 and 7) (vide infra). Therefore, the highest
activity of 1c may be simply related to the faster formation
of the dihydride 2c according to the mechanism shown in
Scheme 1. Consistent with this hypothesis, the use of the
isolated dihydrides 2c, 6 and 7 showed a reverse trend,
the homodiphosphine catalysts being more active than 2c,
up to a 77.3% conversion into n-hexane obtained with 6.
The influence of the X substituent in P(p-XC6H4)3 on
the hydrogenation of 1-hexene was evaluated using the
bis(OAc) catalyst precursors 1a–1e for a reaction lasting
6 h at 80 ꢁC under 100 bar H2 (Table 4). The catalytic
activity of the heterodiphosphine complexes was compared
to that of the homodiphosphine analogues 4 and 5. The
results obtained are reported in Table 4.
ability of the corresponding catalysts to hydrogenate 1-hex-
ene. Indeed, the lowest yield in n-hexane (53.7%) was
obtained using 1e that is the catalyst with the least basic
P(p-XC6H4)3 ligand [7]. On the other hand, the latter com-
plex and its close member 1d were the most efficient for the
isomerization of 1-hexene to trans-2-hexene, which sug-
gests a preferential anti-Markovnikov hydride migration
for the catalyst precursors containing the least basic P(p-
XC6H4)3 ligand.
The results reported in Table 4 suggest a dual role for
the P(p-XC6H4)3 ligand in the hydrogenation of 1-hexene.
Provided that phosphine unfastening, facilitated by less
basic phosphines, is required for H2 activation (Table 2),
product removal (hydride-alkyl reductive elimination)
would be best accomplished by more basic phosphines.
Within this picture, the best performance of the catalyst
1c might simply be due to the specific properties of PPh3.
The influence of reaction parameters such as hydrogen
pressure and reaction time, in the 1-hexene hydrogenation,
was evaluated using 1a as catalyst precursor (Table 5).
According to this study, the best experimental condi-
tions for the conversion of 1-hexene to n-hexane catalyzed
by 1a are a H2 pressure of 100 bar and a temperature of
80 ꢁC. Interestingly, these are also the best conditions for
the complete transformation of 1a into the dihydride 2a.
The reaction conditions chosen for these experiments
guarantee the complete conversion of the bis(OAc) precu-
sors into the corresponding dihydrides.
Irrespective of the catalyst precursor, a nearly complete
conversion of 1-hexene was obtained, the only difference
being the n-hexane/isomerized hexenes ratio. The highest
yield in n-hexane (98.5%) was obtained with the bis(PPh3)
complex 4, followed by 1c with a 81.1% conversion. In
the heterodiphosphine complexes series, the production
of n-hexane increased in the order 1e < 1d < 1b < 1a < 1c,
which reflects, at least for the first four members, a direct
correlation between the basicity of P(p-XC6H4)3 and the
2.2.2. Hydrogenation of ketones
The hydrogenation of acetophenone was studied mak-
ing use of the catalyst precursors 1c and 2c. For compara-
tive purposes, reactions under identical experimental
conditions were carried out with the homodiphosphines
Table 4
Hydrogenation of 1-hexene catalyzed by 1a-e and 4–5
Catalyst
Conversion (%)
Reaction products composition (%)
n-Hexane
trans-3-Hexene
trans-2-Hexene
cis-3-Hexene
1-Hexene
cis-2-Hexene
5
99.3
99.6
99.3
99.5
99.3
99.0
99.9
58.3
78.4
70.8
81.1
60.8
53.7
98.5
7.8
4.5
6.2
4.8
7.8
6.4
0.3
24.1
12.5
16.4
10.5
22.3
28.3
0.9
1.9
1.0
1.3
0.3
1.9
2.1
0.0
0.7
0.4
0.7
0.5
0.7
1.0
0.1
7.2
3.2
4.6
2.8
6.5
8.5
0.2
1a
1b
1c
1d
1e
4
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; p(H2), 100 bar at 20 ꢁC; reaction time, 6 h; T, 80 ꢁC.

1446
L. Salvi et al. / Journal of Organometallic Chemistry 692 (2007) 1442–1450
Table 5
Hydrogenation of 1-hexene catalyzed by 1a: influence of the hydrogen pressure and the reaction time
p(H2) (bar)
Reaction time (h)
Conversion (%)
Reaction products composition (%)
n-Hexane
trans-3-Hexene
trans-2-Hexene
cis-3-Hexene
1-Hexene
cis-2-Hexene
50
3
6
24
54.0
98.8
99.0
29.0
41.7
41.7
0.9
10.6
11.3
15.3
33.3
33.4
0.0
2.9
2.7
46.0
1.2
1.0
8.8
10.3
9.9
100
3
6
24
99.7
99.6
99.9
72.3
78.4
97.6
5.9
4.5
0.7
16.3
12.5
1.3
1.2
1.0
0.0
0.3
0.4
0.1
4.0
3.2
0.3
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; T, 80 ꢁC.
bis(OAc)/dihydride couples 4/6 and 5/7. Initially a reaction
temperature of 80 ꢁC was selected, just to have a reliable
comparison with the hydrogenation reactions of 1-hexene
catalyzed by the same complexes (vide infra). The results
obtained are reported in Table 6.
28.3% to 2.7%. These results and the higher catalytic activ-
ity of 1c respect to 4 and 5 suggest that also the reduction
of ketones requires the decoordination of a phosphine
ligand to start the catalytic cycle.
A screening of the catalytic properties of the various
heterodiphosphine complexes studied in this work was per-
formed for 2-butanone. The results obtained are reported
in Table 7.
From a perusal of Table 6, one may readily realize that
1c is by far the best catalyst, yielding a conversion of 28.3%
to 2-phenylethanol at 80 ꢁC. As a general trend, the
bis(OAc) precursors were more efficient catalysts than the
dihydrides, which contrasts with the results of 1-hexene
hydrogenation. It is therefore probable that the two reac-
tions proceed via different active species and eventually dif-
ferent catalytic mechanisms. The higher conversion
observed for 1c as compared to 2c is a clear indication that
the active species along the hydrogenation of acetophenone
is intermediate between the bis(acetate) and the dihydride.
In view of the reactivity of 1c and related homo- and hete-
rodiphosphines acetates with H2 [2,4a,4b] (see Scheme 1),
the likely candidate is the monohydride 3c. An active role
of the latter species in the catalytic cycle is also suggested
by catalytic reactions performed at lower temperature at
which the dihydrides are barely formed [2]. Indeed,
decreasing the temperature first to 60 ꢁC, then to 40 ꢁC,
did not change the order of reactivity as 1c was still more
efficient than 2c. Like for 1-hexene hydrogenation, the
addition of an excess of PPh3 to the reaction mixture cata-
lyzed by 1c caused a dramatic drop of the conversion from
Besides confirming the results obtained for acetophe-
none (Table 6), the activity trend exhibited by 1a–e for 2-
butanone reduction showed that the catalytic activity
increases with the basicity of the phosphine trans to PnBu3,
which seemingly contrasts with P(p-XC6H4)3 dissociation
as a preliminary step to H2 addition.
In an attempt of obtaining further information on the
possible reaction mechanisms involved in ketone hydroge-
nation by the present heterodiphosphine complexes, a
number of reactions were carried out with the catalyst pre-
cursor 1a, varying systematically experimental parameters
such as temperature, hydrogen pressure and reaction time.
Interestingly, decreasing the hydrogen pressure from 100
to 20 bar, at even temperature (80 ꢁC) and time (3 h),
increased the conversion to 2-phenyl ethanol from 24.3%
to 33.4% (Fig. 1). This result is consistent with our assump-
tion that the monohydrides rather than the dihydrides are
the preferred catalysts for the reduction of ketones,
although a catalytic role of the dihydrides cannot be ruled
out (Table 6).
The influence of the temperature on the productivity
was evaluated both at 20 and 50 bar H2 (Fig. 2). From
60 to 80 ꢁC, the conversion increased with the temperature,
Table 6
Hydrogenation of acetophenone by selected homo- and hetereodiphos-
phine complexes
Catalyst
Conversion (%)
Table 7
40 ꢁC
60 ꢁC
80 ꢁC
Hydrogenation of 2-butanone by hetereodiphosphine complexes
1c
1cb
2c
4
7.5
18.1
28.3
2.7
10.3
9.7
3.0a
16.8
2.1
Catalyst
Conversion (%)
1a
1b
1c
1d
1e
4
28.1
27.5
24.6
24.5
19.6
4.9
2.4
6.5
6
5
7
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; reac-
tion time, 3 h; p(H2), 50 bar at 20 ꢁC.
5
11.3
a
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; reac-
tion time, 3 h; p(H2), 50 bar at 20 ꢁC; T, 80 ꢁC.
Taken from the literature [6].
PPh3 in excess: PPh3/Ru molar ratio 2:1.
b

L. Salvi et al. / Journal of Organometallic Chemistry 692 (2007) 1442–1450
1447
Table 8
40
35
30
25
20
15
10
5
Hydrogenation of 2-butanone catalyzed by 1a or 2a: influence of the
temperature and reaction time
Catalyst
Time (h)
Conversion (%)
60 ꢁC
80 ꢁC
100 ꢁC
1a
1a
1a
2a
3
6
24
3
12.7
23.0
36.0
5.7
28.1
38.3
46.2
6.3
21.4
13.0
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; p(H2),
50 bar at 20 ꢁC.
0
0
20
40
60
80
100
120
genation of 2-butanone is best accomplished by the
bis(OAc) precursor, while the productivity trend suggested
the hydroacetate 3a as the catalytically active species
(Scheme 1).
Hydrogen pressure (bar)
Fig. 1. Hydrogenation of 2-butanone with 1a: dependence on the
hydrogen pressure. Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol;
toluene 4 ml; T, 80 ꢁC; reaction time, 3 h.
The formation of 3a in catalytic hydrogenation condi-
tions as well as its disappearance with time and with
increasing H2 pressure was proved by 31P{1H} NMR spec-
troscopy. Two different reactions were performed using 1a
in C6D6. At the end of each reaction, Na2CO3 was added to
neutralize the acetic acid formed. After 3 h at 80 ꢁC with a
H2 pressure of 20 bar, the ruthenium complexes 1a, 2a and
3a were produced in a 3:3:1 ratio. After 24 h at 80 ꢁC with a
H2 pressure of 100 bar, only 2a and 3a were present in a 3:1
ratio.
The results obtained and the complicated chemistry of
the present catalytic systems do not allow one to draw
out any unambiguous conclusion on ketone hydrogena-
tion. Having said this, the better performance of the hete-
rodiphosphine catalysts 1a–e as compared to the
homodiphosphine catalysts 4–5 may be ascribed to the
more facile decoordination of the trans triarylphosphine
ligands. However, the basicity of the latter must be finely
tuned to get the best performance: when the nucleophilicity
of the P(p-XC6H4)3 ligand is reduced too much as is the
case of 1e (Table 7), then the conversion of the more active
monohydride complexes into the less active dihydrides is
favored (Scheme 1) with a detrimental effect on the overall
catalytic activity. On the other hand, the removal of the
alcohol product from the metal center, which is an impor-
tant catalysis step, may be favored by more nucleophilic
phosphines.
whereas it decreased at 100 ꢁC. Again, this finding is con-
sistent with a higher concentration of the less active dihy-
dride 2a at 100 ꢁC. The detrimental effect of a too large
concentration of 2a would oppose the positive effect of
the higher temperature, in fact.
Finally, when the reaction time was increased from 3 to
6 h and then to 24 h for reactions carried out at 60 or 80 ꢁC
under 50 bar H2, the productivity did not increase propor-
tionally with time like if other less reactive ruthenium spe-
cies would form (Table 8).
In the absence of kinetic measurements, it would be
highly speculative to draw out any conclusion from these
experiments, yet the results do not contrast the hypothesis
that the concentration of the dihydride 2a increases with
time at the expense of the more active monohydride RuH(-
CO)2(OAc)(PnBu3)[P(p-CH3OC6H4)3] (3a).
Experiments in the temperature range from 60 to 100 ꢁC
with 1a and isolated 2a (Table 8) confirmed that the hydro-
40
20 bar H2
35
30
25
20
15
10
5
50 bar H2
2.2.3. Hydrogenation of trans-4-phenyl-3-buten-2-one
(BZA)
Intrigued by the results obtained for the hydrogenation
of C@C and C@O double bonds, we decided to investigate
the reduction of a substrate containing both functional
groups [11]. trans-4-Phenyl-3-buten-2-one (BZA) was
selected as model substrate and its catalytic hydrogenation
was attempted using 1c and 2c as heterodiphosphine pre-
cursors and 4, 5, 6 or 7 as homodiphosphine precursors.
The results obtained are reported in Table 9.
0
40
60
80
Temperature (˚C)
100
120
The highest conversion (25.5%) and selectivity (91.3% of
trans-4-phenylbut-3-en-2-ol (D)) were obtained with the
heterodiphosphine bis(OAc) complex 1c, while the corre-
Fig. 2. Hydrogenation of 2-butanone with 1a; dependence on the
temperature. Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; tolu-
ene 4 ml; reaction time, 3 h.

1448
L. Salvi et al. / Journal of Organometallic Chemistry 692 (2007) 1442–1450
Table 9
Table 10
Hydrogenation of trans-4-phenyl-3-buten-2-one by hetero- and homodi-
Hydrogenation of trans-4-phenyl-3-buten-2-one catalyzed by 1c or 2c:
phosphine complexes
influence of the temperature and of the reaction time
Catalyst
Conversion (%)
Selectivity (%)
Catalyst
T (ꢁC)
Time
(h)
Conversion
(%)
Selectivity (%)
A
B
C
D
A
B
C
D
5
1.8
5.4
25.5
13.0
2.1
11.1
37.0
6.7
17.7
23.8
81.9
11.1
13.0
1.6
3.1
4.8
11.1
0.0
0.4
1.5
4.8
4.5
66.7
50.0
91.3
77.7
66.6
0.0
1c
1c
1c
40
40
40
3
24
48
9.6
28.8
36.3
5.2
4.2
6.9
0.0
1.0
4.1
2.1
0.4
0.8
92.7
94.4
88.2
7
1c
2c
4
1c
1c
1c
60
60
60
3
24
48
25.5
35.8
37.0
6.7
9.5
9.2
1.6
0.8
0.8
0.4
2.5
1.1
91.3
87.2
88.9
6
4.4
13.6
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; p(H2),
50 bar at 20 ꢁC; reaction time, 3 h; T, 60 ꢁC. A: 4-phenylbutan-2-one; B:
trans-4-phenyl-3-buten-2-one; C: 4-phenylbutan-2-ol; and D: trans-4-
phenyl-3-buten-2-ol.
1c
1c
1c
80
80
80
3
24
48
30.5
50.1
82.4
12.4
12.4
30.7
1.3
0.6
0.2
2.0
3.0
20.3
84.3
84.0
48.8
1c
1c
100
100
3
24
62.0
98.7
41.3
34.0
0.5
0.0
11.6
49.3
46.6
16.7
2c
2c
40
40
3
24
4.3
15.0
16.3
12.7
9.3
4.0
2.3
1.3
72.1
82.0
sponding dihydride 2c was less active yielding only 13.0%
of hydrogenated products and also less selective (77.7%
D). Almost negligible conversions of BZA were achieved
with the homodiphosphine compounds 4–7. However, the
results obtained with latter catalysts deserve a comment
as they can contribute to rationalize the different behavior
of the homo- and heterodiphoshime complexes. Unlike the
heterodiphosphine compounds, the homodiphosphine
dihydride compounds 6 and 7 proved more active than
the bis(OAc) derivatives. A net change in selectivity was
observed with the bis(PPh3) dihydride 6 that gave 81.9%
of the saturated ketone 4-phenylbutan-2-one (A).
2c
2c
60
60
3
24
13.0
40.6
17.7
34.7
3.1
1.2
1.5
9.6
77.7
54.5
2c
2c
80
80
3
24
21.2
57.1
30.7
29.6
0.5
0.7
7.5
15.4
61.3
54.3
2c
2c
100
100
3
24
37.9
99.9
57.3
38.2
1.3
0.0
8.2
61.8
33.2
0.0
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; p(H2),
50 bar at 20 ꢁC. A: 4-phenylbutan-2-one; B: trans-4-phenyl-3-buten-2-one;
C: 4-phenylbutan-2-ol; and D: trans-4-phenyl-3-buten-2-ol.
Overall, the results obtained for BZA confirm the data
obtained for alkenes and ketones, with the heterodiphos-
phine precursors being more selective for C@O reduction
and the homodiphosphine precursors, in particular com-
pound 6, being more selective for C@C reduction. The
combined activity/selectivity data reported in Table 9 also
confirm that the hydrogenation of the heterodiphosphine
complex 1c leads to the formation of two catalytically
active species: the monohydride 3c, suited for the reduction
of C@O bonds, and the dihydride 2c, suited for the reduc-
tion of the C@C bond.
The influence of temperature and reaction time on the
hydrogenation of BZA was evaluated for 1c and 2c.
Indeed, the highest chemoselectivity for the C@O reduction
(94.4%) was obtained in the presence of 1c after 24h at
40 ꢁC with a conversion of 28.8% (Table 10).
to 28.2%, yet the best selectivity in D was obtained in the
hydrogen pressure range from 5 to 50 bar, likely due to a
smaller concentration of the dihydride 2c.
Two catalytic reactions, with 1c and 2c, were carried out
in the presence of a twofold excess of PPh3 (PPh3/Ru com-
plex = 2:1). As shown in Table 12, a remarkable decrease
in the productivity was observed with either catalyst in
the presence of added PPh3. It is worth noticing, however,
that the added PPh3 did not remarkably affect the selectiv-
ity of 1c, whereas the selectivity in D decreased from 77.7%
to 30.0% in the case of 2c, with a concomitant increase in
the production of the saturated ketone A. Again, these
results are consistent with an initial and mandatory phos-
phine-dissociation for H2 activation and with a different
Increasing the temperature of the reactions catalyzed by
1c, the conversion to hydrogenated products increased, but
also the selectivity in D decreased, as expected for higher
concentrations of the dihydride 2c in the reaction mixtures.
Consistently, when isolated 2c was used as catalyst, a low
selectivity was observed in the temperature range from 40
to 80ꢁ C even after 24 h. Only at 100 ꢁC, the system became
quite active (100% conversion after 24 h) and fairly selec-
tive, yielding exclusively 61.8% of saturated alcohol C
together with 38.2% of saturated ketone A.
Table 11
Hydrogenation of trans-4-phenyl-3-buten-2-one catalyzed by 1c: influence
of the hydrogen pressure
p(H2) (bar)
Conversion (%)
Selectivity (%)
A
B
C
D
5
20
50
17.4
21.4
25.5
28.2
5.7
6.5
6.7
2.3
2.8
1.6
1.4
0.0
1.4
0.4
1.8
92.0
89.3
91.3
84.4
100
12.4
The influence of the H2 pressure was evaluated using 1c
in reactions lasting 3 h at 60 ꢁC (Table 11). As expected,
increasing the pressure to 100 bar the conversion increased
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; reac-
tion time, 3 h; T, 60 ꢁC. A: 4-phenylbutan-2-one; B: trans-4-phenyl-3-
buten-2-one; C: 4-phenylbutan-2-ol; and D: trans-4-phenyl-3-buten-2-ol.

L. Salvi et al. / Journal of Organometallic Chemistry 692 (2007) 1442–1450
1449
Table 12
4. Experimental
Hydrogenation of trans-4-phenyl-3-buten-2-one catalyzed by 1c and 2c:
influence of the PPh3 excess
4.1. Materials
Catalyst
Conversion (%)
Selectivity (%)
A
B
C
D
All non-catalytic reactions and manipulations were per-
formed under dry nitrogen in Schlenk tubes. C6D6 (99.8%),
sodium carbonate (99.5%), 2-butanone and trans-4-phenyl-
but-3-en-2-one were reagent grade and used without fur-
ther purification. Toluene was purified and stored using
the method reported in the literature [13]. Acetic acid was
distilled under nitrogen prior to use (b.p. 118 ꢁC). 1-Hexene
was purified by elution through a neutral Al2O3 (70-230
mesh) chromatographic column, then distilled and stored
under nitrogen. Acetophenone was distilled prior to use
(b.p. 82 ꢁC/15 mm Hg).
1c
25.5
3.5
13.0
2.0
6.7
1.6
2.8
3.1
0.4
2.8
1.5
91.3
88.6
77.7
30.0
1ca
2c
5.8
17.7
40.0
2ca
15.0
15.0
Conditions: Substrate, 1.50 mmol; catalyst, 15.0 lmol; toluene 4 ml; p(H2),
50 bar at 20 ꢁC; reaction time, 3 h; T, 60 ꢁC. A: 4-phenylbutan-2-one; B:
trans-4-phenyl-3-buten-2-one; C: 4-phenylbutan-2-ol; and D: trans-4-
phenyl-3-buten-2-ol.
a
PPh3/Ru molar ratio 2:1.
selectivity of the active species formed from 1c or 2c in
The complexes 1a–1e [2], 4 [14] and 5 [15] were synthe-
sized according to the literature methods. The complexes
2a–2e [2], 6 [4c] and 7 [5] were similarly prepared following
known procedures, except for using toluene as solvent.
these reaction conditions.
3. Conclusions
The ruthenium(II) heterodiphosphine complexes
4.2. Instruments
Ru(CO)2(OAc)2(PnBu3)(PR3) (R = p-XC6H4) and the
homodiphosphine
complexes
Ru(CO)2(OAc)2(PR3)2
1H NMR spectra were recorded at 399.92 MHz on
either a Varian Mercury 400 spectrometer or at
(R = Ph, nBu) catalyze the hydrogenation of C@C and
C@O groups, showing different activity (and chemoselec-
tivity when both groups are present in the substrate). Irre-
spective of the P(p-XC6H4)3 ligand, the hydrogenation of
1-hexene by the heterodiphosphine precursors is slow and
incomplete even under very high H2 pressure, likely due
to effective isomerization to internal alkenes. For C@C
hydrogenation, the homodiphosphine catalysts are defi-
nitely much more efficient. The reverse trend is observed
for the hydrogenation of the C@O group from ketones
or a,b-unsaturated ketones, the heterodiphosphine precur-
sors being more active than the homodiphosphine ones.
Notably, the hydrogenation of trans-4-phenyl-3-buten-2-
one by the heterodiphosphine catalysts can occur with
excellent selectivity in the corresponding allylic alcohol,
trans-4-phenyl-3-buten-2-ol (>90%), which is surprisingly
high for ruthenium complexes [11a,12].
199.985 MHz on a Varian VXR 200 spectrometer, using
the solvent residual peak as reference. 31P{1H} NMR spec-
tra were registered at 121.421 MHz on a Varian VXR 300
instrument, using H3PO4 (85%) as external standard:
downfield values were taken as positive. All 31P{1H}
NMR spectra were acquired using a broad-band decoupler.
4.3. Catalytic reactions – general procedure
In a glass vial placed in a stainless-steel autoclave a tol-
uene solution (4.0 ml) containing the catalyst
(1.50 · 10ꢀ5 mol) and the substrate (1.50 · 10ꢀ3 mol) was
introduced under dry nitrogen. The autoclave was sealed
and pressurized with H2 to the desired pressure. The reac-
tor was heated at the desired temperature with stirring. At
the end of each reaction, the autoclave was cooled to room
temperature and the residual gas was vented off. The prod-
uct composition was determined by GC using pure com-
pounds as standards.
The different activity and selectivity exhibited by the
heterodiphosphine and homodiphosphine precursors is
difficult to interpret as in either system the decoordination
of a phosphine ligand is mandatory for any catalytic activ-
ity. However, it has been established that the catalytic
hydrogenation of the C@C double bond is best accom-
plished by homodiphosphine dihydride catalysts, while
heterodiphosphine monohydrides are more efficient cata-
lysts than both homo- and heterodiphosphine dihydrides
for the reduction of the keto C@O bond. The formation
of mono- and dihydride species is faster for the heterodi-
phosphine complexes than for the homodiphosphine ones,
which reflects the trans effect exerted by PnBu3.
The following GC methods were employed:
1-Hexene: a Chrompack capillary column Al2O3/
Na2SO4 PLOT (length: 50 m, diameter: 0.45 mm) was kept
at 130 ꢁC for 25 min, then heated up to 200 ꢁC at a rate of
30 ꢁC minꢀ1. Finally, it was maintained at this temperature
for 50 min.
The other GC analyses were performed with packed col-
umns (length: 2 m, diameter: 1/800).
Acetophenone: a CW 20M column was kept at 60 ꢁC for
10 min, then heated to 160 ꢁC at a rate of 5 ꢁC minꢀ1
.
Current studies in our laboratories are presently aimed
at finding the proper conditions for the selective formation
of either monohydride or dihydride intermediates as either
species controls activity and substrate/product selectivity.
Finally, it was maintained at this temperature for 30 min.
2-Butanone: a PPG column was heated at 50 ꢁC.
trans-4-Phenylbut-3-en-2-one: a FFAP column (‘free
fatty acids phase’ supported on Chromosorb G AW-

1450
L. Salvi et al. / Journal of Organometallic Chemistry 692 (2007) 1442–1450
(b) R.K. Pomeroy, J. Organomet. Chem. 177 (1979) C27;
(c) R.K. Pomeroy, K.S. Wijesekera, Inorg. Chem. 19 (1980) 3729;
(d) M.M. Crozat, S.F. Watkins, J. Chem. Soc., Dalton Trans. (1972)
2512;
DMCS 5%) was kept at 50 ꢁC for 2 min, then heated to
140 ꢁC at a rate of 10 ꢁC minꢀ1. Finally, it was maintained
at this temperature for 40 min.
(e) R.D. Holmes-Smith, S.R. Stobart, R. Vefghi, M.J. Zaworotko,
K. Jochem, T.S. Cameron, J. Chem. Soc., Dalton Trans. (1987) 969;
(f) V.K. Dioumaev, L.J. Procopio, P.J. Carroll, D.H. Berry, J. Am.
Chem. Soc. 125 (2003) 8043.
4.4. Thermal stability of 1c
A
C6D6 (0.75 ml) solution of 1c (22.193 mg,
3.0 · 10ꢀ5 mol) was introduced into a NMR tube and
heated at the desired temperature. The conversion of 1c
[4] (a) P. Frediani, C. Faggi, A. Salvini, M. Bianchi, F. Piacenti, Inorg.
Chim. Acta 272 (1998) 1417;
(b) P. Frediani, M. Bianchi, A. Salvini, F. Piacenti, S. Ianelli, M.
Nardelli, J. Chem. Soc., Dalton Trans. (1990) 165;
(c) P. Frediani, C. Faggi, S. Papaleo, A. Salvini, M. Bianchi, F.
Piacenti, S. Ianelli, M. Nardelli, J. Organomet. Chem. 123 (1997) 536.
[5] A. Salvini, P. Frediani, M. Bianchi, F. Piacenti, Inorg. Chim. Acta
227 (1994) 247.
[6] A. Salvini, P. Frediani, S. Gallerini, Appl. Organomet. Chem. 14
(2000) 570.
[7] T. Allman, R.G. Goel, Can. J. Chem. 60 (1982) 716.
[8] C.A. Tolman, Chem. Rev. 77 (1977) 313.
[9] R.A. Sanchez-Delgado, J. Bradley, G. Wilkinson, J. Chem. Soc.,
Dalton Trans. (1976) 399.
[10] R.A. Sanchez-Delgado, N. Valencia, R.L. Marquez-Silva, A. Andri-
ollo, M. Medina, Inorg. Chem. 25 (1986) 1106.
[11] (a) T. Ohkuma, H. Ooka, S. Hashiguchi, T. Ikariya, R. Noyori, J.
Am. Chem. Soc. 117 (1995) 2675;
was monitored by 31P{1H} and H NMR spectroscopy.
1
4.5. Reactivity of 1c with H2
A
C6D6 (0.75 ml) solution of 1c (22.193 mg,
3.0 · 10ꢀ5 mol) was introduced under nitrogen into a glass
contained in a stainless-steel autoclave (150 ml). In some
reactions anhydrous sodium carbonate (150 mg, 1.45 mmol)
was added. Then the autoclave was closed and pressurized
with hydrogen (100 bar). The autoclave was placed in a ther-
mostatic oil bath set at the desired temperature and rocked
for the desired time. At the end, the reactor was cooled, the
gases vented off and the solution, recovered under nitrogen,
was filtered to eliminate any suspended solid, and analyzed
by 1H and 31P{1H} NMR spectroscopy.
(b) T. Ohkuma, H. Ooka, T. Ikariya, R. Noyori, J. Am. Chem. Soc.
117 (1995) 10417;
(c) T. Ohkuma, H. Ooka, M. Yamakawa, T. Ikariya, R. Noyori, J.
Org. Chem. 61 (1996) 4872;
(d) M. Hernandez, P. Kalck, J. Mol. Catal. A 116 (1997) 131;
(e) C. Bianchini, E. Farnetti, M. Graziani, M. Peruzzini, A. Polo,
Organometallics 12 (1993) 3753;
Acknowledgments
The authors thank the University of Florence and the
CNR of Italy for financial support. The Ente Cassa di
Risparmio di Firenze (Italy) is gratefully acknowledged
for granting a 400 MHz NMR spectrometer to the Depart-
ment of Organic Chemistry of the University of Florence as
well as for supporting the ICCOM-CNR through the pro-
ject HYDROLAB. The European Commission is gratefully
acknowledged for funding the Network of Excellence
IDECAT.
(f) S. Bolano, L. Gonsalvi, F. Zanobini, F. Vizza, V. Bertolasi, A.
˜
Ramerosa, M. Peruzzini, J. Mol. Catal. A 224 (2004) 61;
(g) J. Hannedouche, J.A. Kenny, T. Walsgrove, M. Wills, Synlett 2
(2002) 263;
(h) M.J. Burk, W. Hems, D. Herzberg, C. Malan, A. Zanotti-Gerosa,
Org. Lett. 2 (26) (2000) 4173;
(i) E. Lindner, I. Warad, K. Eichele, H.A. Mayer, Inorg. Chim. Acta
350 (2003) 49.
[12] (a) S. Bhaduri, K. Sharma, J. Chem. Soc., Chem. Commun. (1988)
173;
(b) C. Bianchini, E. Farnetti, M. Graziani, M. Peruzzini, A. Polo,
Organometallics 12 (1993) 3753;
References
(c) G. Zassinovich, G. Mestroni, S. Gladiali, Chem. Rev. 92 (1992)
1051.
[13] A. Vogel, Vogel’s Textbook of Practical Organic Chemistry, fourth
ed., Longman, London, 1978, p. 353.
[14] S.D. Robinson, M.F. Uttley, J. Chem. Soc., Dalton Trans. (1973)
1912.
[15] M. Bianchi, P. Frediani, U. Matteoli, G. Menchi, F. Piacenti, G.
Petrucci, J. Organomet. Chem. 259 (1983) 207.
[1] A. Salvini, P. Frediani, C. Giannelli, L. Rosi, J. Organomet. Chem.
690 (2005) 371.
[2] F. Micoli, L. Salvi, A. Salvini, P. Frediani, C. Giannelli, J.
Organomet. Chem. 690 (2005) 4867.
[3] (a) For a review on the trans effect in octahedral complexes, see: R.K.
Pomeroy, R.S. Gay, G.O. Evans, W.A.G. Graham, J. Am. Chem.
Soc. 94 (1972) 272;