Isomerization of olefins by phosphine-substituted ruthenium complexes and influence of an 'additional gas' on the reaction rate

Article abstract of DOI:10.1016/S0022-328X(01)00669-6

Phosphine-substituted ruthenium carbonyls have often been used as catalytic precursors in reactions such as the hydrogenation or the hydroformylation of olefins. To collect evidence on the coordination of the olefin as a preliminary step of these reactions we have investigated the isomerization of hex-1-ene, in hydrocarbon solvent, in the presence of the phosphine-substituted ruthenium carbonyls Ru(CO)₃(PR₃)₂, Ru₃(CO)₉(PR₃)₃ and Ru(CO)₂(OAc)₂(PR₃)₂ [R=Bu, Ph]. When using Ru(CO)₃(PPh₃)₂ the rate of the reaction shows a partial first order with respect to the concentration of the catalyst and of the substrate. The activation parameters were also evaluated and the activation entropy is negative. A reaction scheme involving the displacement of a carbonyl ligand with formation of a π-olefin-ruthenium complex is suggested. The rate of the reaction significantly changes if an alcohol is used as solvent. This behaviour is attributed to a modification of the catalytic precursor with formation of a ruthenium hydride. This hypothesis is confirmed by the identification of an alkoxy ruthenium hydride. The isomerization of olefins by phosphine-substituted ruthenium carbonyls is retarded by the presence of an 'additional gas' such as dinitrogen. This influence is more evident than the analogous one reported in the hydroformylation reaction: the same pressure of the 'additional gas' present in the reaction vessel reduces the rate of the isomerization to a larger extent, i.e. the presence of 1000 bar of nitrogen decreases in otherwise identical experiments the isomerization conversion of hex-1-ene from 95.6% to 25.8%. An analogous effect is also caused by the presence of argon and xenon. Helium, on the other hand, does not display any influence. These data are an indication of an interaction between the 'additional gas' and a catalytically active transition metal complex.

Full text of DOI:10.1016/S0022-328X(01)00669-6

Journal of Organometallic Chemistry 625 (2001) 255–267
www.elsevier.nl/locate/jorganchem
Isomerization of olefins by phosphine-substituted ruthenium
complexes and influence of an ‘additional gas’ on the reaction rate
Antonella Salvini, Franco Piacenti, Piero Frediani *, Andrea Devescovi,
Maria Caporali
Department of Organic Chemistry, Uni6ersity of Florence, Via Gino Capponi, 9, I-50121, Florence, Italy
Received 14 November 2000; accepted 3 January 2001
Abstract
Phosphine-substituted ruthenium carbonyls have often been used as catalytic precursors in reactions such as the hydrogenation
or the hydroformylation of olefins. To collect evidence on the coordination of the olefin as a preliminary step of these reactions
we have investigated the isomerization of hex-1-ene, in hydrocarbon solvent, in the presence of the phosphine-substituted
ruthenium carbonyls Ru(CO)3(PR3)2, Ru3(CO)9(PR3)3 and Ru(CO)2(OAc)2(PR3)2 [R=Bu, Ph]. When using Ru(CO)3(PPh3)2 the
rate of the reaction shows a partial first order with respect to the concentration of the catalyst and of the substrate. The activation
parameters were also evaluated and the activation entropy is negative. A reaction scheme involving the displacement of a carbonyl
ligand with formation of a p-olefin–ruthenium complex is suggested. The rate of the reaction significantly changes if an alcohol
is used as solvent. This behaviour is attributed to a modification of the catalytic precursor with formation of a ruthenium hydride.
This hypothesis is confirmed by the identification of an alkoxy ruthenium hydride. The isomerization of olefins by phosphine-sub-
stituted ruthenium carbonyls is retarded by the presence of an ‘additional gas’ such as dinitrogen. This influence is more evident
than the analogous one reported in the hydroformylation reaction: the same pressure of the ‘additional gas’ present in the reaction
vessel reduces the rate of the isomerization to a larger extent, i.e. the presence of 1000 bar of nitrogen decreases in otherwise
identical experiments the isomerization conversion of hex-1-ene from 95.6% to 25.8%. An analogous effect is also caused by the
presence of argon and xenon. Helium, on the other hand, does not display any influence. These data are an indication of an
interaction between the ‘additional gas’ and a catalytically active transition metal complex. © 2001 Elsevier Science B.V. All rights
reserved.
Keywords: Ruthenium; Olefin; Isomerization; Nitrogen; Argon; Xenon
1. Introduction
reaction of an olefin with appropriate reagents to pro-
duce different chemicals on the basis of the conditions
employed. The coordination of the olefin to the catalyst
plays an important role in the reaction. If the coordi-
nated olefin isomerizes, different products may be
formed [6].
Different mechanisms are reported for the isomeriza-
tion process connected with the use of different cata-
lytic systems or reaction conditions. The usual
hypothesis to explain the migration of a double bond
through the alkyl chain of an olefin involves the inter-
vention of a metal (i.e. Rh, Ru, Co) hydride as catalyst.
Through several addition–elimination steps a thermo-
dynamically stable mixture of olefins is obtained [7].
A different mechanism may be suggested when the
starting catalyst is not a metal hydride or when an
hydride cannot be formed in a preliminary step of the
The chemistry of the olefin coordination to transition
metal complexes covers a large field of organometallic
chemistry and a large part of this subject involves
homogeneous catalysis. Some industrial processes that
involve interactions between an olefin and a catalyst
are: the hydroformylation of olefins [1], the polymeriza-
tion of ethylene by a one-site organometallic catalyst
[2], the stereospecific polymerization of propene [3],
metathesis of olefins [4], Wacker–Hoechst oxidation of
ethylene [5], and many others (Scheme 1).
The above processes, and other syntheses, involve the
* Corresponding author. Tel.: +39-055-2757649; fax +39-055-
2757660.
E-mail address: frediani@chimorg.unifi.it (P. Frediani).
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)00669-6

256
A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
2. Results and discussion
reaction. The formation of a p-olefin metal complex is
suggested, followed by the formation of a p-allyl system
coordinated to a metal hydride complex. The system
evolves and isomeric olefins are obtained [8].
Other different isomerization mechanisms are re-
ported in the literature to explain the results obtained
in catalytic reactions such as the hydroformylation of
olefins [9–11].
The isomerizing activity of a metal complex may
affect the regioselectivity of the reaction (as reported in
the hydroformylation of olefins) or may give rise to a
less reactive regioisomer.
Other examples are reported where the olefin isomer-
ization plays an important role in synthetic chemistry
to afford products from more easily synthesized inter-
mediate as reported for the synthesis of N-protected
1,2,3,4-tetrahydropyridazine [12,13] or the formation of
1,4-dicyanobutane from butadiene and HCN [14].
In this paper we report the results of an investigation
on the isomerization of an olefin such as hex-1-ene in
the presence of an organometallic ruthenium complex.
The catalysts tested were mononuclear or cluster
phosphine-substituted Ru(0) carbonyl complexes
Ru(CO)3(PnBu3)2
(1),
Ru3(CO)9(PnBu3)3
(2),
Ru(CO)3(PPh3)2 (3), Ru3(CO)9(PPh3)3 (4) or mononu-
clear Ru(II) compounds Ru(CO)2(OAc)2(PnBu3)2 (5)
and Ru(CO)2(OAc)2(PPh3)2 (6). These complexes were
chosen in consideration of their activity in hydrogena-
tion processes of CꢀC and CꢀO double bonds in which
the activation of the double bond is one of the prelimi-
nary steps. Their solubility in organic solvents and the
presence of the phosphine and carbonyl ligands render
these complexes appropriate to investigate the forma-
tion of intermediates under reaction conditions through
spectroscopic methods. The kinetic and the activation
parameters were also evaluated. Complexes with no
hydridic hydrogen atoms were chosen because we want
only fair catalytic activity in order to be able to evalu-
ate, as a final goal of this investigation, the influence of
an ‘additional gas’ on the rate of the reaction.
Particular attention was devoted to evaluating the
influence of ‘additional gases’, such as N2, He, Ar or
Xe, on the isomerization rate. Our aim was to evaluate
if, as reported for the hydroformylation of an olefin in
the presence of a cobalt complex [15], the ‘additional
gas’ affects the rate of olefin isomerization in the pres-
ence of a ruthenium catalyst.
2.1. Isomerization of an olefin in the presence of
ruthenium complexes
The influence of temperature, reaction time, catalyst
concentration, and substrate concentration on the cata-
lytic activity of ruthenium complexes was tested in the
isomerization of hex-1-ene. These variables were evalu-
ated by changing each one of the parameter at a time
while the other conditions were kept constant. In each
test, the conversion of the substrate and the composi-
tion of the reaction products were determined. No
optimization of the reaction was performed at this stage
of the investigation.
Scheme 1. Olefins as raw materials in the synthesis of chemicals.
Table 1
Hex-1-ene isomerization in the presence of ruthenium catalysts: influ-
2.1.1. Influence of the catalytic precursor
ence of catalysta
The data on the catalytic activity of complexes 1–6
are reported in Table 1. Among the mononuclear Ru(0)
complexes, 1 shows a lower catalytic activity than 3.
The catalytic activity of the ruthenium clusters 2 and 4
is very different from that of 3. The increase of the
catalytic activity of 2 as a function of temperature is
higher than that of 3, giving a conversion of 19.1% at
353 K and 97.1% at 393 K.
At 373 K, in the presence of 2, the amount of hex-2-
enes [(Z)+(E)] is largely preponderant with respect to
hex-3-enes [(Z)+(E)] with a molar ratio 20:1. This
ratio is higher than that (3:1) obtained in the presence
of 3.
Catalyst
Code
Conversion (%)
353 K
373 K
393 K
Ru(CO)3(PnBu3)2
1
2
3
4
5
6
5.9
19.1
25.1
0.0
34.8
69.4
53.4
97.0
0.0
94.9
97.1
91.7
97.6
6.5
Ru3(CO)9(PnBu3)3
Ru(CO)3(PPh3)2
Ru3(CO)9(PPh3)3
Ru(CO)2(OAc)2(PnBu3)2
Ru(CO)2(OAc)2(PPh3)2
0.0
0.0
0.0
0.0
a Catalyst: 28.4 mg (0.040 mmol); hex-1-ene 490 ml (4.00 mmol);
substrate/catalyst (molar ratio)=100; toluene 5.0 ml; p(He): 5 bar at
293 K; reaction time: 3 h.

A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
257
Table 2
Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2 (3): influence of temperaturea
T (K)
Conversion (%)
Reaction products composition (%)
Hex-1-ene
(Z)-Hex-2-ene
(E)-Hex-2-ene
(Z)-Hex-3-ene
(E)-Hex-3-ene
333
353
373
393
7.5
25.1
53.4
91.7
92.5
74.9
46.6
8.3
3.3
10.6
24.2
37.7
2.6
8.4
17.1
32.0
0.0
2.8
6.0
1.6
3.3
6.1
10.4
11.6
a Catalyst: 28.4 mg (0.040 mmol); hex-1-ene 490 ml (4.00 mmol); substrate/catalyst (molar ratio)=100; toluene 5.0 ml; p(He): 5 bar at 293 K;
reaction time: 3 h.
Table 3
Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2 (3): influence of reaction timea
Reaction time (h)
T (K)
Conversion (%)
Reaction products composition (%)
Hex-1-ene
(Z)-Hex-2-ene
(E)-Hex-2-ene
(Z)-Hex-3-ene
(E)-Hex-3-ene
3
333
333
333
333
353
353
353
353
353
353
353
353
7.5
25.8
39.1
52.3
25.8
35.6
56.6
70.0
93.9
97.4
97.9
98.1
92.5
74.2
60.9
47.7
74.2
64.4
43.4
30.0
6.1
3.3
10.8
15.8
20.3
10.9
14.6
24.2
29.8
36.3
27.1
18.6
18.6
2.6
9.5
13.2
17.7
8.9
11.3
17.5
22.1
36.9
49.1
55.6
55.9
0.0
1.6
4.9
7.4
2.2
4.6
7.1
8.5
9.6
5.5
3.8
3.7
1.6
3.9
5.2
6.9
3.8
5.1
7.8
9.6
11.1
15.7
19.9
19.9
14
24
48
1.5
3
6
9
18
56
122
146
2.6
2.1
1.9
a Catalyst: 28.4 mg (0.040 mmol); hex-1-ene 490 ml (4.00 mmol); substrate/catalyst (molar ratio)=100; toluene 5.0 ml; p(He): 5 bar at 293 K.
Complex 4 is catalytically inactive at 353 K (Table 1),
but at 373 K it provides an almost quantitative isomer-
ization of hex-1-ene (97.0%). At this temperature the
ratio between hex-2-enes [(Z)+(E)] and hex-3-enes
[(Z)+(E)], as reported for 2, is very high, i.e. 10:1,
with (E)-hex-2-ene as the largely predominant
stereoisomer (64.0%).
2.1.2.1. Reaction temperature. The influence of tempera-
ture on the catalytic activity of 3 is reported in Table 2.
A temperature of 333 K is necessary to obtain a
detectable conversion of the substrate. The conversion
increases from 7.5% at 333 K up to 91.7% at 393 K. All
linear isomers of hex-1-ene were formed and their
amount increases almost ten times when the tempera-
ture is raised from 333 to 393 K.
The Ru(II) complex Ru(CO)2(OAc)2(PnBu3)2 (5) is a
poor catalyst even at 393 K: only 6.5% conversion was
obtained (Table 1). Let us recall, however, that, at this
temperature, complex 5 is partially transformed into
Ru2(CO)4(OAc)2(PnBu3)2 (7), catalytically active in the
isomerization of hex-1-ene [16]. As a consequence, the
isomerization noticed in the presence of 5 may be
attributed to the formation of 7.
2.1.2.2. Reaction time. At 333 K, the conversion in-
creases from 7.5 to 52.3% when the reaction time
increases from 3 to 48 h (Table 3). Taking into account
a first-order reaction rate, the integrated equation
kt=ln[ao/ao(1−x)]
(1)
Compound 6 is not catalytically active for the iso-
merization of hex-1-ene in the range of temperatures
tested.
can be obtained, where ao is the initial concentration of
the substrate, t is the reaction time and x is the percent-
age conversion. Plotting ln[100/(100−x)] as a function
of time, a linear relation was obtained with k=4.28×
2.1.2. Influence of reaction parameters
10−6 −1. The linear relation is in agreement with a
s
The data reported above show that complex 3 is a
fairly good catalyst and, therefore, the influence of
different reaction parameters in the isomerization of
linear hexenes was tested in the presence of 3.
partial first-order reaction rate with respect to the sub-
strate concentration.
The amount of (Z)-hex-2-ene increases from 3.3%
after 3 h to 20.3% after 48 h; similarly, (E)-hex-2-ene

258
A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
tetrahydrofurane, 1,4-dioxane, n-heptane, 2,2,2-tri-
fluoroethanol, 1,1,1,3,3,3-hexafluoropropan-2-ol, pro-
pan-2-ol, propan-1-ol, ethanol, methanol and butan-1-
ol. The results are reported in Table 4.
Complex 3 is only partially soluble in n-heptane,
methanol and 2,2,2-trifluoroethanol, whereas it is com-
pletely soluble in the other solvents employed.
The nature of the solvent greatly affects the isomer-
ization rate of hex-1-ene. A substrate conversion of
9.5% was obtained in a coordinating solvent such as
acetone, whereas an 87.7% conversion was obtained in
a hydrogen donor solvent such as propan-2-ol. The
higher value obtained in propan-2-ol may be tentatively
attributed to a transformation of the catalytic precursor
in the reaction conditions adopted, and consequently
the mechanism changes. In fact a ruthenium hydride
complex may be formed in a preliminary step through
an oxidative addition of the alcohol to 3, and this new
hydridic complex (Scheme 2) is the catalytically active
species having a very high isomerizing activity. IR and
NMR spectroscopies, however, do not support the
Fig. 1. Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2
(3). Rate costant kc at 353 K as a function of time (data from Table
3). x: conversion %. For the linear regression y=ax+b, a=4.22×
10−5, r2=0.992. The specific rate is kc=4.22×10−5 s−1
.
increases from 2.6 to 17.7%, and (Z)-hex-3-ene in-
creases from 0.5 to 7.4%, whereas (E)-hex-3-ene
changes from 1.5 to 6.9%.
Another set of tests was performed at 353 K, and the
data are reported in Table 3. Using Eq. (1), a first-order
reaction rate has been obtained with a specific rate of
k=4.22×10−5 s−1 (Fig. 1). The specific rate increases
ten times as the temperature increases by 20 K. The
thermodynamic equilibrium mixture of olefins (Table 3)
was obtained after a reaction time of 122 h.
presence
of
the
hypothetical
intermediate
[Ru(H)(CO)3(PPh3)2][(CH3)2CHO] (8a), probably be-
cause it is very reactive, or it is formed in a very low
concentration (see Section 2.3.1).
The data reported were also confirmed by using a
higher substrate concentration in
a
series of
The high catalytic activity of hydrido ruthenium
complexes [17] supports this hypothesis. A mechanism
involving subsequent addition–elimination steps of a
CꢀC double bond on an MꢁH bond is more efficient
than that of one involving the formation of a p-allyl
intermediate.
isochronous experiments carried out at 353 K with a
substrate concentration in the range between 0.40 and
1.60 M, while the other reaction conditions remained
unchanged. The conversion is a linear function of the
initial concentration of the substrate, confirming the
first-order reaction rate.
The conversion of the olefin is 46.5% when using
ethanol as solvent, a value higher than that obtained
using toluene (34.6%). Almost the same value is ob-
tained using propan-1-ol (46.1%) or butan-1-ol (39.3%).
The low conversion (9.5%) obtained in a coordinat-
ing solvent such as acetone may be attributed to a
competition between the solvent and the substrate for a
coordinative place available on the catalyst. By using
The data collected between 333 and 353 K show that
the (E)-hex-3-ene/(Z)-hex-3-ene ratio increases with
temperature, and at higher temperatures the thermody-
namic equilibrium is obtained rapidly.
2.1.2.3. Sol6ent. The following solvents were tested in
consideration of their polarity and acidity: acetone,
Table 4
Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2 (3): influence of solventa
Solvent
Conversion (%)
Reaction products composition (%)
Hex-1-ene
(Z)-Hex-2-ene
(E)-Hex-2-ene
(Z)-Hex-3-ene
(E)-Hex-3-ene
(CF3)2CHOH
Acetone
Tetrahydrofurane
1.4-Dioxane
Toluene
Butan-1-ol
Propan-1-ol
Ethanol
8.4
9.5
91.6
90.5
89.0
76.3
65.4
60.7
53.9
53.5
12.4
3.8
2.0
4.4
4.0
5.6
3.9
12.6
10.8
13.6
16.5
17.6
31.8
0
0
0
0.6
1.9
2.7
2.0
4.3
5.4
6.4
6.0
12.0
11.0
23.7
34.6
39.3
46.1
46.5
87.6
9.1
0
14.7
15.0
17.2
18.6
33.2
4.8
5.3
6.0
4.3
10.6
Propan-2-ol
a Catalyst: 28.4 mg (0.040 mmol); hex-1-ene 490 ml (4.00 mmol); substrate/catalyst (molar ratio)=100; solvent 5.0 ml; p(He): 5 bar at 293 K;
reaction time: 3 h.

A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
259
As a consequence, the formation of (Z)- or (E)-hex-
3-ene from hex-1-ene takes place through several reac-
tion steps that do not involve the displacement of the
intermediate hex-2-ene from the catalyst.
2.1.2.5. Catalyst concentration. An increase of the cata-
lyst concentration (Table 5) affects the rate of the
reaction: the conversion increases from 25.0 to 70.5%.
The kinetic equation for a first-order reaction rate, with
respect to catalyst concentration, is kt[cat]=ln[100/
(100−x)] where [cat] is the molar concentration of 3
and the other variables have the meaning given in Eq.
(1). Because a linear relation between ln[100/(100−x)]
and [cat] was obtained, a partial first-order of the
reaction with respect to the catalyst concentration is
obtained with a specific rate of kcat=0.025 dm3 mol−1
s−1
.
2.1.2.6. Carbon monoxide. Carbon monoxide (5 bar)
suppresses the isomerization of hex-1-ene at 393 K.
2.1.2.7. Free triphenylphosphine. The data for the tests
performed at 393 K, using free triphenylphosphine, are
reported in Table 6. When the P/Ru molar ratio in-
creases from 0 to 5 the isomerization of hex-1-ene
decreases from 66.0% to 26.8% respectively. A further
increase of the P/Ru ratio from 5 up to 20 results in the
opposite trend in the rate of isomerization; it is now
increasing, and the conversion reaches a value of
69.7%.
Scheme 2.
aprotic solvents or solvents having a lower polarity,
such as tetrahydrofurane or 1,4-dioxane, respectively,
the conversion increases to 11.0 and 23.7%.
2.1.2.4. Substrate. The catalytic activity of 3 in the
isomerization of different olefinic substrates has been
tested using all linear isomers of hex-1-ene. The isomer-
ization of the linear regioisomers of hex-1-ene after 3 h
is undetectable. If the reaction is carried out for a much
longer time (122 h), however, a thermodynamic mixture
of hexenes may be obtained (Table 3). It is evident that
the steric hindrance around the double bond of an
internal olefin plays a fundamental role; hex-1-ene in
fact may be easily coordinated to the catalyst to form a
p-olefin ruthenium complex and isomerized. (Z)-Hex-2-
ene, the isomer having a lower steric hindrance among
the regioisomers tested, is much too crowded and is
thus less efficiently coordinated to 3, thereby giving a p
complex in lower concentration.
The increase of the reaction rate in the presence of a
large excess of free phosphine, may be explained by
assuming that catalyst 3 is transformed into the more
reactive species Ru(CO)2(PPh3)3 (9) [18]. Spectroscopic
evidence for the presence of 9 supports this hypothesis.
It is not easy to discriminate between Ru(CO)3(PPh3)2
(w=1896 cm−1) and Ru(CO)2(PPh3)3 (w=1905 cm−1
)
by IR spectroscopy, because only a strong stretching is
present in both cases. The same reaction was therefore
carried out in an NMR sample tube (see Section 4.5.4).
The 31P-NMR spectrum in C6D6, after 6 h shows the
presence of the starting complex 3 (l=56.8 ppm), and
Table 5
Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2 (3): influence of catalyst concentrationa
[cat] (mM)
Conversion (%)
Reaction products composition (%)
Hex-1-ene
(Z)-Hex-2-ene
(E)-Hex-2-ene
(Z)-Hex-3-ene
(E)-Hex-3-ene
0.1
0.2
0.4
0.8
1.6
3.2
25.0
33.5
43.8
50.5
60.0
70.5
75.0
66.5
56.2
49.5
40.0
29.5
8.1
15.1
20.8
23.3
28.0
34.1
13.3
12.6
13.2
15.9
17.4
19.8
1.6
2.9
5.9
6.5
8.7
9.9
2.0
2.9
3.9
4.8
5.9
6.7
a Hex-1-ene 490 ml (4.00 mmol); toluene 25.0 ml; substrate/catalyst (molar ratio)=100; p(He): 5 bar at 293 K; T=393 K.

260
A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
Table 6
Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2 (3): influence of the presence of free phosphinea
PPh3/Ru molar ratio
Conversion (%)
Reaction products composition (%)
Hex-1-ene
(Z)-Hex-2-ene
(E)-Hex-2-ene
(Z)-Hex-3-ene
(E)-Hex-3-ene
0.0
0.5
1
2.5
5
7.5
10
20
60.0
41.4
42.0
28.4
26.8
42.0
46.8
69.7
40.0
58.6
58.0
71.6
73.2
58.0
53.2
30.3
28.0
22.8
27.0
17.3
17.7
27.8
32.6
47.4
17.4
11.1
12.7
8.1
8.7
3.6
5.9
3.9
2.3
1.5
1.1
1.3
1.8
1.8
Trace
1.5
Trace
Trace
Trace
1.4
8.0
12.9
12.4
19.1
a Catalyst: 28.4 mg (0.040 mmol); hex-1-ene 490 ml (4.00 mmol); substrate/catalyst (molar ratio)=100; toluene 25.0 ml; p(He): 5 bar at 293 K;
T=393 K.
free triphenylphosphine (l= −3.99 ppm) together with
a little amount of 9 (l=50.9 ppm) (conversion 2.5%).
The signals of both free triphenylphosphine and 3
decrease when the reaction time increases, whereas the
intensity of the resonance of 9 increases.
identified, to which the [Ru(H)(CO)3(PPh3)2]-
[(CF3)2CHO] formulation may be attributed. The IR
spectrum shows in fact that 3 is completely transformed
into the new compound 8b, having two absorptions
Further support to this hypothesis was obtained by
carrying out an isomerization experiment in the condi-
tions reported in Table 6 using 9 as catalyst. The
conversion of hex-1-ene was 93.6%, whereas in the
presence of 3 it was 60.0%.
2.1.2.8. Acti6ation parameter. The data on the influence
of temperature (Table 2) were employed to evaluate the
activation parameters DG**, DH** and DS**. Plotting
the DG** values as a function of T (Fig. 2) a straight
line was obtained from which the DS** (−167.3 J
Fig. 2. Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2
(3). Activation parameter (data from Table 2). Linear regression
y=ax+b, a=167.29, b=58 883.73, r2=0.998. DS**: −167.3
mol−1
K K
−1= −39.9 cal mol−1 −1) and DH** (58.9
J mol−1 K−1. DH**: 58.9 kJ mol−1
.
kJ mol−1=14.05 kcal mol−1) values were obtained.
These data are related to the rate-determining step of
the reaction and the negative value of DS** is an
indication that the transition state involves the forma-
tion of a complex having a higher steric hindrance than
the starting catalyst. A reaction mechanism is suggested
(Scheme 3) to rationalize the data collected and the
rate-determining step is assumed to be the coordination
of the olefin to the metal forming a p olefin–metal
complex. This hypothesis is in agreement with the
observed influence of the structure of the substrate on
the reaction rate. In fact, whereas hex-1-ene is isomer-
ized, hex-2-enes and hex-3-enes are unchanged in the
same conditions, after 3 h.
2.2. Reacti6ity of Ru(CO)3(PPh3)2 (3)
2.2.1. With an alcohol
The reaction of 3 with 1,1,1,3,3,3-hexafluoropropan-
2-ol was investigated. The complex 3 is insoluble in this
alcohol at room temperature, but it dissolves at 333 K
and the new complex 8b, analogous to 8a, has been
Scheme 3.

A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
261
at 2078(s) and 2056(s) cm−1 attributable to w(CꢀO) and
2.3. Influence of an ‘additional gas’ on the
hydroformylation of olefins
another band at 1485 cm−1 attributed to w(CꢁH) of a
ruthenium alkoxide. The formation of 8b (Scheme 2) is
1
An additional gas, that is not involved in the cata-
lytic cycle of the reaction but which may obstruct it,
has recently been used to single out the elementary
steps of a catalytic reaction. An appropriate amount of
an ‘additional gas’ such as N2, Ar or Xe, has been
shown to reduce considerably the rate of hydroformyla-
tion of olefins in the presence of dicobalt octacarbonyl
[15]. The influence of such ‘additional gas’ has been
explained on a molecular basis [15]: the ‘additional gas’
may compete with a reactant for a coordinatively un-
saturated position on the catalyst to form a complex
that is kinetically inactive. The amount of catalyst
available for the catalytic reaction is therefore reduced
and, as a consequence, the reaction rate is decreased.
This role may be easily explained when the ‘additional
gas’ is nitrogen if, for instance, we consider the
equilibrium:
confirmed by H-NMR: a triplet at l= −6.1 ppm may
be unequivocally attributed to a hydridic proton cou-
pled with two equivalent phosphine ligands with a
2JPꢁH=15.1 Hz. The spin system (AX2) and the cou-
pling constant are in agreement with a hydride in a cis
position with respect to the phosphorus atoms. Only
one singlet, at l=36.7 ppm, is present in the 31P-
NMR, confirming that the two phosphinic ligands are
equivalent. The 13C-NMR spectrum shows a triplet at
l=191.9 ppm (2JCꢁP=13 Hz) in agreement with two
equivalent carbonyl groups coupled with two cis phos-
phinic ligands. Another resonance attributable to a
carbonyl group is present at l=189.8 ppm. The reso-
nances of the alkoxy group are not detected because the
complex is formed in a solution of 1,1,1,3,3,3-hexa-
fluoropropan-2-ol.
A 1 mM solution of 8b shows the same conductivity
as an equimolar solution of sodium acetate. The mass
spectrum of the complex shows a cluster of peaks
attributed to [RuH(CO)3(PPh3)2]+ (m/z 711).
The reaction between 3 and 1,1,1,3,3,3-hexafluoro-
propan-2-ol is reversible: when the solvent is removed
the starting complex 3 is completely restored.
The higher stability of compound 8b, compared with
that of 8a, is in agreement with its lower catalytic
activity in the isomerization of hex-1-ene: using 3 as
catalytic precursor and 1,1,1,3,3,3-hexafluoropropan-2-
ol as solvent the olefin conversion is only 6.5%. It is
known that perfluoroalkoxy complexes are more stable
than the corresponding non-fluorinated compounds
[19]. For instance Rh(H)(PPh3)4 reacts with 1,1,1,3,3,3-
hexafluoropropan-2-ol to give the corresponding
rhodium alkoxide [19] and Ru(H2)(PPh3)4 reacts with
2,2,2-trifluoroacetophenone to form the stable
Ru(H)[OCH(CF3)Ph](PPh3)4 complex [19].
LnM(H2)+N2 X LnM(N2)+H2.
It is completely shifted to the right in the presence of a
high concentration of nitrogen, as reported by Sacco
and Rossi [20]. When the reaction is carried out in the
presence of Ar or Xe an effect analogous to that
reported for nitrogen has been noticed: the rate of the
reaction is considerably reduced [15]. These results may
be attributed to an interaction between Ar or Xe with
the catalytically active cobalt species in agreement with
the explanation reported for the influence of nitrogen.
Ar- or Xe-containing transition metal complexes have
been identified, usually by spectroscopic methods, in a
solid matrix at low temperature [21], in a solution of
liquefied noble gas [21b,22] or in the gaseous state
[21b].
We have now tested the eventual influence of an
‘additional gas’ on the isomerization rate of hex-1-ene
in the presence of 3 as catalyst.
The behaviour of 3 in 2,2,2-trifluoroethanol is
analogous to that shown in 1,1,1,3,3,3-hexafluoro-
propan-2-ol. The complex 3 dissolved is completely
transformed into the new complex [Ru(H)(CO)3-
(PPh3)2][CF3CH2O] (8c), as shown by the IR spectrum
of the solution: only two absorptions are present, in the
w(CO) region, at 2066 and 2010 cm−1. Owing to the
stability of 8c, and to the low solubility of 3 in this
solvent, the isomerization of hex-1-ene is 8.4%.
2.3.1. Nitrogen
The influence of nitrogen (pressures between 0 and
1000 bar) on the rate of hex-1-ene isomerization was
tested in isochronous experiments performed at 373 K,
in the presence of 3. The results are reported in Table 7.
A 55.9% conversion was obtained in a reference test;
the conversion is reduced to 44.1% when adding nitro-
gen at a pressure of 244 bar and it decreases progres-
sively to 9.4% when the nitrogen pressure is increased
to 1032 bar.
The reduction of the isomerization rate of hex-1-ene
may be attributed to the formation of a ruthenium
complex containing nitrogen (Scheme 4) that, in the
reaction conditions tested, is unable to coordinate the
olefin. As a consequence, the amount of catalytically
active ruthenium complex is reduced. Using an increas-
2.2.2. With PPh3
Using 31P-NMR spectroscopy, the conversion of 3
into 9 was tested on a solution of 3, hex-1-ene and
PPh3. After 6 h the amount of 9 is 2.5%, and its
conversion increases to 13.3% after 24 h, 21.1% after 89
h and 30.3% after 144 h.

262
A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
Table 7
Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2 (3): influence of nitrogen as additional gasa
Nitrogen (bar)
Conversion (%)
Reaction products composition (%)
Hex-1-ene
(Z)-Hex-2-ene
(E)-Hex-2-ene
(Z)-Hex-3-ene
(E)-Hex-3-ene
0
55.9
44.1
29.6
17.7
9.4
44.1
55.9
70.4
82.3
90.6
24.4
19.8
13.4
8.8
19.2
14.6
9.0
6.5
3.4
4.8
4.7
4.3
Trace
Trace
7.5
5.0
2.9
2.4
1.7
244
504
740
1032
4.3
a Catalyst: 28.4 mg (0.040 mmol); hex-1-ene 490 ml (4.00 mmol); substrate/catalyst (molar ratio)=100; toluene 5.0 ml; T=373 K; reaction time:
3 h.
ing pressure of nitrogen, the amount of the nitrogen-
Scheme 3 the transition state may be approached with
a molar volume lower than the sum of the volumes of
the starting complex and the olefin, and, as a conse-
quence, it is facilitated by increasing the total pressure.
However, this is only a suggestion, and further data
would be necessary to confirm this hypothesis.
containing ruthenium complex increases and this re-
duces the concentration of the ruthenium complex
catalytically active in the hex-1-ene isomerization. An
analogous behaviour was reported by Pennella and
Banks [23] in the isomerization of pent-1-ene catalysed
by [Ru(H4)(PPh3)3] or [Ru(H2)(N2)(PPh3)3] complex.
The presence of nitrogen does not affect the isomeric
distribution of the olefins formed, i.e. the (Z)/(E) ratio
remains constant (Table 7). The hex-3-enes obtained
from hex-1-ene must be formed directly from the start-
ing olefin coordinated to the ruthenium complex and,
therefore, this process is not influenced by the presence
of nitrogen. These data support the hypothesis that the
catalyst active for olefin isomerization is not modified
by the presence of nitrogen, but that there is only a
competition between nitrogen and the olefin to a coor-
dinatively unsaturated position of the catalyst.
2.3.4. Xenon
The influence of xenon as ‘additional gas’ has been
evaluated using different experimental conditions with
respect to those adopted with the other additional gases
previously reported due to the different critical temper-
ature of this gas. The results are reported in Table 10.
The behaviour of xenon is in line with that reported for
nitrogen and argon, even if the total pressure of this
experiment is very low. The amount of xenon in solu-
tion affects the conversion of the olefinic substrate that
is reduced from 47.0% in the absence of xenon to 41.7%
when 2.2 g of xenon is added and to 32.8% when the
amount of xenon is 9.4 g.
2.3.2. Argon
The results for the influence of argon are reported in
Table 8. The conversion of the olefinic substrate is
reduced to 48.8% in the presence of argon at a pressure
of 240 bar, and to 8.6% with a pressure of 910 bar. This
behaviour of argon is analogous to that reported for
nitrogen, and is in agreement with the almost identical
ionization potentials of these two gases, even if, at the
same partial pressure, argon shows a higher solubility
in toluene [24]. As a consequence, a simple correlation
between the data obtained with these two gases is not
possible due to their different solubility and compress-
ibility factors [24].
2.3.3. Helium
The results for the influence of helium as ‘additional
gas’ are reported in Table 9. Almost no influence was
observed up to a pressure of 669 bar of He as addi-
tional gas. By increasing the pressure of helium to 835
bar, however, the conversion increases (83.9%), and a
further increase is obtained at 976 bar (89.5%). The
data obtained between 835 and 976 bar of helium are
not easily rationalized. In the mechanism reported in
Scheme 4.

A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
263
Table 8
Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2 (3): influence of argon as additional gasa
Argon (bar)
Conversion (%)
Reaction products composition (%)
Hex-1-ene
(Z)-Hex-2-ene
(E)-Hex-2-ene
(Z)-Hex-3-ene
(E)-Hex-3-ene
0
55.9
48.8
32.2
15.2
8.6
44.1
51.2
67.8
84.8
91.4
24.4
22.8
14.1
6.9
19.2
14.4
10.8
5.3
4.8
6.4
3.2
ꢀ0
ꢀ0
7.5
5.2
4.1
3.0
1.8
240
487
766
910
3.7
3.1
a Catalyst: 28.4 mg (0.040 mmol); hex-1-ene 490 ml (4.00 mmol); substrate/catalyst (molar ratio)=100; toluene 5.0 ml; T=373 K; reaction time:
3 h.
Table 9
Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2 (3): influence of helium as additional gasa
Helium (bar)
Conversion (%)
Reaction products composition (%)
Hex-1-ene
(Z)-Hex-2-ene
(E)-Hex-2-ene
(Z)-Hex-3-ene
(E)-Hex-3-ene
5
55.9
61.0
61.1
62.6
83.0
89.5
44.1
39.0
38.9
37.4
17.0
10.5
24.4
26.2
27.8
26.7
35.9
37.1
19.2
20.2
20.0
20.9
27.3
30.8
4.8
7.6
6.6
7.6
9.9
7.5
7.0
6.7
7.4
9.9
257
509
669
835
976
10.0
11.6
a Catalyst: 28.4 mg (0.040 mmol); hex-1-ene 490 ml (4.00 mmol); substrate/catalyst (molar ratio)=100; toluene 5.0 ml; T=373 K; reaction time:
3 h.
Table 10
Hex-1-ene isomerization in the presence of Ru(CO)3(PPh3)2 (3); influence of xenon as additional gasa
Xenon (g)
Conversion (%)
Reaction products composition (%)
Hex-1-ene
(Z)-Hex-2-ene
(E)-Hex-2-ene
(Z)-Hex-3-ene
(E)-Hex-3-ene
0
47.0
41.7
38.8
38.1
37.9
32.8
53
21.9
19.3
19.3
18.5
19.2
16.8
14.6
12.4
11.0
10.9
11.1
9.4
6.0
5.5
3.9
5.0
3.2
3.0
4.5
4.5
4.6
3.7
4.4
3.6
2.2
3.5
4.5
5.9
9.4
58.3
61.2
61.9
62.1
67.2
a Catalyst: 28.4 mg (0.040 mmol); hex-1-ene 490 ml (4.00 mmol); substrate/catalyst (molar ratio)=100; toluene 5.0 ml; T=353 K; reaction time:
3 h.
1. the steric hindrance of the PBu3 ligands in 1 is
higher than the PPh3 phosphine in 3 and, as a
consequence, the coordination of the olefin is more
difficult;
2. the higher basicity of tri-n-butylphosphine, if com-
pared with triphenylphosphine, increases the stabil-
ity of complex 1 with respect to 3 and, as a
consequence, 1 is less prone to release a carbonyl
ligand to start a catalytic cycle.
In the presence of 3 a pseudo-first-order reaction rate
is obtained with respect to the substrate and catalyst
concentrations. The large influence of carbon monoxide
on the reaction rate suggests that the activation of the
3. Conclusions
The Ru(0) complexes Ru(CO)3(PnBu3)2 (1),
Ru3(CO)9(PnBu3)3 (2), Ru(CO)3(PPh3)2 (3), Ru3(CO)9-
(PPh3)3 (4) are catalytically active in the isomerization
of hex-1-ene to hex-2-enes and hex-3-enes, whereas the
Ru(II) complexes Ru(CO)2(OAc)2(PnBu3)2 (5) and
Ru(CO)2(OAc)2(PPh3)2 (6) show a low or negligible
isomerization activity. The reaction shows a low re-
giospecificity: all possible linear isomers are formed.
The higher activity of 3 with respect to 1 may be
explained taking into account that:

264
A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
olefin may involve the displacement of a carbon
monoxide molecule present in 3 with formation of a
p-olefin–ruthenium complex. This compound may
evolve towards the formation of a p-allyl hydride ruthe-
nium complex, in agreement with the suggestion re-
ported by Valle et al. [25], from which it is possible to
restore the starting p-olefin–ruthenium complex or an
analogous complex containing the isomerized olefin.
Owing to its higher steric hindrance the internal olefin
is displaced by the terminal one, and the complex
formed starts another catalytic cycle.
An alcohol used as solvent affects the catalytic activ-
ity of complex 3 because it causes the formation of a
hydridic ruthenium complex that changes the catalytic
cycle. In fact, a ruthenium hydride is formed in the
reaction between 1,1,1,3,3,3-hexafluoropropan-2-ol and
3.
In consideration of the fact that hex-2-ene and hex-3-
ene are not isomerized after a reaction time of 3 h in
the presence of 3, the (E)- and (Z)-hex-3-enes formed in
the isomerization of hex-1-ene are not obtained through
several addition–elimination steps. These olefins must
be formed through a multistep isomerization process
that does not involve the displacement of the intermedi-
ate isomerized olefin bound to the catalytic intermedi-
ate. To isomerize internal olefins a very long reaction
time is necessary.
An additional gas, such as nitrogen, present in ap-
propriate concentration in the reaction medium reduces
the rate of the reaction. In consideration of the mecha-
nism suggested, the role played by an ‘additional gas’ is
an interaction with the ruthenium complex forming an
intermediate that, in the condition tested, is unable to
activate an olefin. Whereas helium has no influence on
the reaction, at least up to 650 bar, the presence of
argon causes an effect analogous to that reported for
nitrogen. The different behaviour observed in the pres-
ence of helium and argon rules out the hypothesis that
the different reaction rate is attributable to the high
pressure employed. Furthermore, the influence of
xenon, having an effect analogous to that of nitrogen
or argon, but at a low pressure, confirms that a high
pressure of the ‘additional gas’ is only necessary to
obtain the appropriate concentration of the gas in
solution.
analogous effect in the hydroformylation of olefin
catalysed by Co2(CO)8 [15].
The following considerations may be formulated:
“ an interaction between a ruthenium carbonyl com-
plex and an ‘additional gas’ is possible, in agreement
with the hypothesis reported for the hydroformyla-
tion of cyclohexene in the presence of a cobalt
complex, even if different reaction conditions are
employed;
“ competition between an ‘additional gas’ and an
olefin is possible in addition to those between the
‘additional gas’ and hydrogen or carbon monoxide
[15] previously reported;
“ the methodology to carry out experiments in the
presence of an ‘additional gas’ is very useful for
showing, in a catalytic cycle, weak interactions be-
tween the substrate and the transition metal
complex.
4. Experimental
A Shimadzu GC14 chromatograph equipped with a
PPG column (Ucon oil LB 550X on Chromosorb W
15%, i.d. 1/8¦, length 4 m) and with a Perkin–Elmer
Model 8320 capillary gas-chromatograph, equipped
with a Chrompack Al2O3 PLOT column (Al2O3 deacti-
vated with KCl, i.d. 0.32 mm, length 50 m) were used
for quantitative analyses of the reaction product. The
instruments had FID detectors. In consideration of the
analogy of the products examined, no correction fac-
tors were introduced. The identity of the products was
confirmed by GC–MS using a Shimadzu apparatus
with a GC14A capillary column chromatograph and a
QP2000 mass detector. The chromatograph was
equipped with a CP-Sil8 50 m capillary column.
The conductivity determination was carried out using
an Analytical Instrument 111 conductivity meter with
Orion 99.01.01 cell having a cell constant 1.00 cm−1
.
Electrospray mass was carried out using a Perkin–
Elmer Sciex Api 365 system.
IR spectra were recorded with a Perkin–Elmer mod.
1760 FTIR spectrometer.
1H-, 13C- and 31P-NMR spectra were recorded using
a Varian VXR300 spectrometer operating at 299.987
The reduction of the reaction rate cannot be at-
tributed to an increase of the volume of the solution
due to the ‘additional gas’ dissolved in the solvent. The
amount of nitrogen dissolved in toluene, when 1000 bar
of nitrogen is employed, increases the volume of the
solution of 20% and this dilution of the solution does
not explain the data obtained.
The reduction of the reaction rate in the presence of
an ‘additional gas’ may be ascribed to an interaction
between the ‘additional gas’ and the catalytic specie in
agreement with the hypothesis formulated to explain an
1
MHz for H-NMR, at 75.429 MHz for 13C-NMR and
at 121.421 MHz for 31P-NMR, using solutions in ap-
propriate solvents. SiMe4 was used as external standard
for 1H- and 13C-NMR, H3PO4 (85%) for 31P-NMR
(signals reported as positive downfield to the standard).
13C- and 31P-NMR spectra were acquired using a broad
band decoupler.
Hex-1-ene was purified by elution through a neutral
Al2O3 (70–230 mesh) chromatographic column, then
distilled under nitrogen, and had a b.p. of 335 K.

A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
265
(Z)-Hex-2-ene, (E)-hex-2-ene, (Z)-hex-3-ene and (E)-
hex-3-ene were purified as reported for hex-1-ene.
Tri-n-butylphosphine (Aldrich) was distilled prior to
use; b.p. 431–433 K/60 mmHg
Solvents: toluene, methanol, ethanol, propan-2-ol,
butan-1-ol, and n-heptane were dried according to Vogel
[26]. 2,2,2-Trifluoroethanol and 1,1,1,3,3,3-hexafluoro-
propan-2-ol were dried on molecular sieves and distilled
prior to use.
having a capacity of 125 ml. A toluene solution (25 ml)
containing 3 (0.040 mmol), hex-1-ene (4.0 mmol) and
triphenylphosphine was introduced by suction in the
vessel, then helium was added up to 5 bar. The autoclave
was heated at the prefixed temperature and the solution
stirred. After a rapid cooling to room temperature,
helium was vented and the solution collected and
analysed by GC as indicated above. The data obtained
are reported in Table 6.
Other reagents and solvents were commercial products
and used without further purification. The following
gases were used: nitrogen, Sol Centro product with a
purity of 99.996%; argon, Sol Centro product with a
purity of 99.998%; xenon, Rivoira product with a purity
of 99.997%; helium, Sol Centro product with a purity of
99.998%.
All experiments, unless otherwise specified, were car-
ried out under a nitrogen atmosphere using the Schlenk
technique.
The following catalysts were prepared according to the
literature; their spectroscopic characteristics were in
agreement with the data reported: Ru(CO)3(PnBu3)2 (1)
[27], Ru3(CO)9(PnBu3)3 (2) [28], Ru(CO)3(PPh3)2 (3) [29],
Ru3(CO)9(PPh3)3 (4) [30], Ru(CO)2(OAc)2(PnBu3)2 (5)
[31,32], and Ru(CO)2(OAc)2(PPh3)2 (6) [33].
4.3. Catalytic isomerization in the presence of CO
The experiment was carried out in a Parr autoclave
having a capacity of 125 ml. A toluene solution (25 ml)
containing 3 (0.040 mmol), hex-1-ene (4.0 mmol) was
introduced by suction into the vessel, then CO was added
up to 5 bar. The autoclave was heated and rocked at
393 K. After a rapid cooling to room temperature, CO
was vented and the solution collected and analysed by
GC as indicated above; only hex-1-ene was present.
4.4. Catalytic isomerization in the presence of an
additional gas
The tests were carried out in a stainless steel rocking
autoclave having a capacity of 27.5 ml. Air was evacu-
ated and, in a nitrogen atmosphere, a toluene solution
containing the catalyst and the olefin was introduced.
The autoclave was sealed and the ‘additional gas’ was
introduced, up to the pressure required. The autoclave
was electrically heated at the prefixed temperature and
rocked for the established time. After a rapid cooling to
room temperature, the ‘additional gas’ was vented and
the solution collected and analysed by GC as indicated
above. The results are reported in Tables 7–10.
Ru(CO)2(PPh3)3 (9) was synthesized according to the
following procedure: the acetato (6) was converted in the
dihydride complex H2Ru(CO)2(PPh3)2 (10) as reported
by Frediani et al. [34]. On adding pentane a white cream
solid was obtained and the solution was separated by
filtration. A solution of complex 10 (0.263 mmol) and
PPh3 (1.841 mmol) in methanol (7 ml) was refluxed for
2 h. The yellow solid was filtered, washed with anhy-
drous ethanol and anhydrous n-hexane and dried under
1
vacuum. The IR, H- and 31P-NMR spectra of 9 are in
agreement with the data reported in the literature [35,36].
4.5. Reacti6ity of ruthenium complexes
4.1. Catalytic isomerization
4.5.1. Ru(CO)3(PPh3)2 and hex-1-ene
The tests were carried out in a 50 ml glass vial placed
in a stainless steel rocking autoclave having a capacity
of 150 ml. Air was evacuated and a solution containing
the catalyst and the olefin was introduced in the glass vial
under a nitrogen atmosphere. The autoclave was sealed
and helium was then added up to 5 bar. The autoclave
was heated in an thermostatic oil bath at the prefixed
temperature and rocked for the established time. After
a rapid cooling to room temperature, the gases were
vented and the solution collected and analysed by GC as
indicated above. The results of the catalytic experiments
are reported in Tables 1–5.
A solution of Ru(CO)3(PPh3)2 (3) (0.042 mmol) in
C6D6 (0.6 ml) was introduced in an NMR sample tube
and the 31P-NMR spectrum was collected. Hex-1-ene
(1.43 mmol) was added and the sample tube heated at
333 K for 14 h. 31P- and 1H-NMR spectra were recorded
at hourly intervals. The 31P-NMR spectra showed the
presence of only one signal at 56.3 ppm, in keeping with
the presence of only 3. The H-NMR spectra collected
at various time intervals showed the presence of an
increasing amount of isomeric olefins.
1
4.5.2. Ru(CO)3(PPh3)2 and (CF3)2CHOH
A suspension of Ru(CO)3(PPh3)2 (3) (35.1 mmol) in
(CF3)2CHOH (1 ml) was introduced in an NMR sample
tube in which a capillary tube containing C6D6 was
present for lock purposes. The sample tube was heated
4.2. Catalytic isomerization in the presence of free
triphenylphosphine
1
The experiments were carried out in a Parr autoclave
at 333 K for 8 h. The 31P-, H-NMR, and IR spectra

266
A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
were recorded every 30 min. A new product 8b was
References
obtained, after 30 min, having the following character-
1
istic data: 31P-NMR l=36.7 ppm; H-NMR l= −6.1
[1] (a) G. Parshall, Homogeneous Catalysis, Wiley–Interscience,
New York, 1980. (b) C.D. Frohning, Ch.W. Kohlpaintner, in: B.
Cornils, W.A. Hermann (eds.), Applied Homogeneous Catalysis
with Organometallic Compounds, VCH, Weinheim, 1996, p. 29.
[2] (a) H.G. Alt, A. Ko¨ppl, Chem. Rev. 100 (2000) 1205. (b) S.D.
Ittel, L.K. Johnson, M. Brookhart, Chem. Rev. 100 (2000) 1169.
[3] W. Keim, A. Behr, M. Ro¨per, in: G. Wilkinson, F.G.A. Stone,
E.W. Abel (Eds.), Comprehensive Organometallic Chemistry,
vol. 8, Pergamon, New York, 1982, p. 371.
[4] K.J. Ivin, Olefin Metathesis, Academic, New York, 1983.
[5] K. Weissermel, H.J. Arpe, Industrial Organic Chemistry, III
edition, VCH, Weinheim, 1997, p. 163.
[6] W.A. Hermann, in: B. Cornils, W.A. Hermann (Eds.), Applied
Homogeneous Catalysis with Organometallic Compounds,
VCH, Weinheim, 1996, p. 980.
[7] O.A. Hougen, K.M. Watson, R.A. Ragatz, Principi dei processi
chimici, vol. II, CEA, Milano, 1966.
[8] M.A. Bennett, T.W. Matheson, in: G. Wilkinson, F.G.A. Stone,
E.W. Abel (Eds.), Comprehensive Organometallic Chemistry,
vol. 4, Pergamon, NewYork, 1982, p. 944.
[9] P. Pino, F. Piacenti, M. Bianchi, in: I. Wender, P. Pino (Eds.),
Organic Syntheses via Metal Carbonyls, vol. II, Wiley, New
York, 1977, p. 118.
2
(t, 1H, JPꢁH=15.1 Hz) ppm; 13C-NMR l=191.9 (t,
CO, 2JCꢁP=13 Hz), 189.8 (s, CO) ppm. No further
changes were shown. The IR spectrum [(CF3)2CHOH
as solvent] in the 2200–1400 cm−1 region displays ab-
sorptions at 2078(s), 2056(s), 1485(s) cm−1. An electro-
spray mass spectrum of the complex shows cluster of
peaks centred at m/z (intensity% in parentheses): 711
(20) [RuH(CO)3(PPh3)2]+, 654 (50) [Ru(CO)(PPh3)2]+,
626 (100) [Ru(PPh3)3]+, and 363 (16) [Ru(PPh3)2]+.
A 1 mM solution of this complex shows the same
conductivity (0.58×10−4
equimolar solution of sodium acetate.
V
−1 cm−1 at 289 K) as an
The solvent was distilled at reduced pressure and the
residue dissolved at room temperature in C6D6. The
1
31P-, H-, 13C-NMR and IR spectra confirm the pres-
ence of Ru(CO)3(PPh3)2 as the only complex in
solution.
[10] M. Bianchi, F. Piacenti, P. Frediani, U. Matteoli, J. Organomet.
Chem. 135 (1977) 387.
[11] F. Piacenti, M. Bianchi, P. Frediani, G. Menchi, U. Matteoli, J.
Organomet. Chem. 417 (1991) 77.
[12] G. Menchi, U. Matteoli, A. Scrivanti, S. Paganelli, C. Botteghi,
J. Organomet. Chem. 354 (1988) 215.
[13] M. Fetizon, M. Golfier, R. Milcent, I. Papadakis, Tetrahedron
31 (1975) 165.
[14] J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke, Principles
and Applications of Organotransition Metal Chemistry, Univer-
sity Science Books, Mill Valley, CA, USA, 1987, p. 570.
[15] (a) F. Piacenti, F. Calderazzo, M. Bianchi, L. Rosi, P. Frediani,
Organometallics 16 (1997) 4235. (b) L. Rosi, F. Piacenti, M.
Bianchi, P. Frediani, A. Salvini, Eur. J. Inorg. Chem. (1999) 67.
[16] A. Salvini, P. Frediani, F. Piacenti, J. Mol. Catal. A: Chem. 159
(2000) 185.
[17] M.A. Bennett, T.W. Matheson, in: G. Wilkinson, F.G.A. Stone,
E.W. Abel (Eds.), Comprehensive Organometallic Chemistry,
vol. 4, Pergamon, New York, 1982, p. 942.
[18] F. Porta, S. Cenini, S. Giordano, M. Pizzotti, J. Organomet.
Chem. 150 (1978) 261.
4.5.3. Ru(CO)3(PPh3)2 and CF3CH2OH
A suspension of Ru(CO)3(PPh3)2 (3) (35.1 mmol) in
CF3CH2OH (1 ml) was introduced in a vial under
nitrogen. The IR spectrum (CF3CH2OH as solvent) in
the 2200–1800 cm−1 region displays absorptions at
2066(s), 2010(s) cm−1. The solvent was distilled at re-
duced pressure and the residue dissolved at room tem-
1
perature in C6D6. The 31P-, H-NMR, and IR spectra
confirm the presence of Ru(CO)3(PPh3)2 as the only
complex in solution.
4.5.4. Ru(CO)3(PPh3)2 and PPh3
A solution of Ru(CO)3(PPh3)2 (3) (7.2 mmol), hex-1-
ene (0.72 mmol), and PPh3 (0.072 mmol) in C6D6 (1 ml)
was introduced in an NMR sample tube and analysed
by 31P-NMR. The spectrum in C6D6, after 6 h at 373 K,
showed the presence of the starting complex 3 (l=
56.8 ppm), and free triphenylphosphine (l=
−3.99 ppm) together with 2.5% of 9 (l=50.9 ppm).
The signals of free triphenylphosphine and 3 decrease
in the course of the reaction, while the signal due to 9
increases. The following conversions were obtained:
13.3% after 24 h, 21.1% after 89 h and 30.3% after
144 h.
[19] Y. Hayashi, S. Komiya, T. Yamamoto, A. Yamamoto, Chem.
Lett. (1984) 1363.
[20] A. Sacco, M. Rossi, J. Chem. Soc. Chem. Commun. (1967) 316.
[21] (a) R.L. Sweany, F.N. Russell, Organometallics 7 (1988) 719. (b)
R.N. Perutz, in: J.D. Atwood (Ed.), Inorganic Reactions and
Methods, vol. 14, Wiley–VCH, 1998, p. 107. (c) D.M. Heinekey,
W.J. Oldham, Chem. Rev. 93 (1993) 913. (d) P.G. Jessop, R.H.
Morris, Coord. Chem. Rev. 121 (1992) 155. (e) J.R. Wells, E.J.
Weitz, J. Am. Chem. Soc. 114 (1992) 2873. (f) B.H. Weiller, J.
Am. Chem. Soc. 114 (1992) 10 910.
[22] M.B. Simpson, M. Poliakoff, J.J. Turner, W.B. Maier II, J.G.
McLaughlin, J. Chem. Soc. Chem. Commun. (1983) 1355.
[23] (a) F. Pennella, R.L. Banks, J. Catal. 35 (1974) 73. (b) F.
Pennella, Coord. Chem. Rev. 16 (1975) 51.
[24] E. Wilhelm, R. Battino, Chem. Rev. 73 (1973) 1.
[25] M. Valle, D. Osella, G.A. Vaglio, Inorg. Chim. Acta 20 (1976)
213.
Acknowledgements
The authors wish to thank the University of Florence
and the Ministero della Ricerca Scientifica e Tecnolog-
ica (MURST), Programmi di Ricerca Scientifica di
Notevole Interesse Nazionale, Cofinanziamento
MURST 1998–99, for financial support.
[26] A. Vogel, Vogel’s Practical Organic Chemistry, Longman, Lon-
don, 1978.

A. Sal6ini et al. / Journal of Organometallic Chemistry 625 (2001) 255–267
267
[27] A. Salvini, P. Frediani, D. Rovai, M. Bianchi, F. Piacenti, J.
Mol. Catal. 89 (1994) 77.
[32] P. Frediani, U. Matteoli, G. Menchi, M. Bianchi, F. Piacenti,
Gazz. Chim. Ital. 115 (1985) 365.
[28] F. Piacenti, M. Bianchi, E. Benedetti, G. Sbrana, J. Inorg. Nucl.
Chem. 29 (1967) 1389.
[33] B.F.G. Johnson, R.D. Johnson, J. Lewis, I.G. Williams, J.
Chem. Soc. A (1971) 689.
[29] J. Collman, W.R. Roper, J. Am. Chem. Soc. 87 (1965)
4008.
[30] F. Piacenti, M. Bianchi, E. Benedetti, G. Braca, Inorg. Chem. 7
(1815) 1968.
[31] M. Bianchi, P. Frediani, U. Matteoli, G. Menchi, F. Piacenti, G.
Petrucci, J. Organomet. Chem. 259 (1983) 207.
[34] P. Frediani, C. Faggi, S. Papaleo, A. Salvini, M. Bianchi, F.
Piacenti, S. Ianelli, M. Nardelli, J. Organomet. Chem. 536–537
(1997) 123.
[35] B.E. Cavit, K.R. Grundy, W.R. Roper, J. Chem. Soc. Chem.
Commun. (1972) 60.
[36] T.R. Gaffney, J.A. Ibers, Inorg. Chem. 21 (1982) 2851.
.