Catalytic studies with ruthenium clusters substituted with diphosphines: Part I. Studies with Ru3(CO)10(Ph2PCH2PPh2)

Article abstract of DOI:10.1016/S1381-1169(99)00195-8

The catalytic precursor Ru₃(CO)₁₀(Ph₂PCH₂PPh₂)(dppm) (dppm = Ph₂PCH₂PPh₂) isomerizes 1-hexene mainly to the kinetic product cis-2- hexene at low H₂ pressure

Full text of DOI:10.1016/S1381-1169(99)00195-8

Ž
.
Journal of Molecular Catalysis A: Chemical 149 1999 75–85
www.elsevier.comrlocatermolcata
Catalytic studies with ruthenium clusters substituted with
diphosphines
ž / ž
/
Part I. Studies with Ru3 CO 10 Ph2 PCH2 PPh2
)
Bernardo Fontal , Marisela Reyes, Trino Suarez, Fernando Bellandi, Juan Carlos Dıaz
´
´
Laboratorio Organometalicos, Departamento de Quımica, Facultad de Ciencias, La Hechicera, UniÕersidad de Los Andes, Merida,
´
´
´
Venezuela
Received 23 November 1998; accepted 17 March 1999
Abstract
Ž
. Ž
.Ž
. Ž
.
The catalytic precursor Ru3 CO Ph2PCH2PPh2 dppm dppmsPh2PCH2PPh2 isomerizes 1-hexene mainly to the
10
kinetic product cis-2-hexene at low H2 pressure, and to n-hexane at higher hydrogenation conditions. The overall first order
rate constant on substrate is 0.11 miny1. The compound hydrogenates other unsaturated groups in good yield. Turnover
frequency studies indicate cluster catalysis. q 1999 Elsevier Science B.V. All rights reserved.
Keywords: Ruthenium cluster; Bidentate phosphines; Hydrogenation and isomerization catalysts
Ž
.
CO
12
w x
6 . Cyclohexene is hydrogenated to cy-
1. Introduction
clohexane with the following substituted Ru
clusters: Ru3 CO PPh3 3, Ru3 CO PPh3 2
Ž
PCH2 PPh2 H4Ru4 CO dppm 8 ,H2Ru3-
Ž .Ž
of ligands capable of stabilizing the cluster is
considered important, especially during catalytic
Catalytic studies with metal clusters is an
area of recent research 1,2 . Terminal olefin
isomerization to a mixture of internal olefins
has been done with Ru3 CO
substituted with monodentate phosphines 2 .
Ž
. Ž
.
Ž
. Ž
.
9
Ž
7
w
x
.
w x
. Ž
. Ž
C6H4 7 , H4Ru4 CO dppm , dppmsPh2-
10
.
Ž
. Ž
.
w x
10
Ž
.
and the cluster
. Ž
. Ž
. w x
12
E CO dppm EsO, S 9 . The presence
5
2
w x
in
Ž
.
12
Isomerization of 1-pentene with Ru3 CO
refluxing hexane gave mainly trans-2-pentene
w x
reactions 1 . Polydentate phosphine ligands
w x
Ž
.
Ž
.
Ž
3 . H4Ru4 CO and H4Ru4 CO 11L Lsmo-
12
could be useful for cluster stabilization, but
little work has been reported on the effect of
these ligands on cluster catalysis. The diphos-
phine dppm has been used in the synthesis of
many mononuclear and cluster complexes a few
of which has been used in homogenous catalysis
.
nodentate phosphine clusters gave similar re-
w x
Ž
.
12
sults 4 . Under photolytic conditions Ru3 CO
Ž
.
Ž
.
and Ru3 CO
PPh3 3 gave different isomer
9
w x
ratios 5 . Sanchez-Delgado et al. have reported
1-hexene hydrogenation to n-hexane with Ru3
w
x
10–14 . In our laboratory we have been study-
ing the catalytic behavior of Ru complexes with
)
Corresponding author. Tel.: q58-074-401380; Fax: q58-
w
x
polydentate phosphines 15–20 . In other stud-
074-401286; E-mail: fontal@ciens.ula.ve
1381-1169r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S1381-1169 99 00195-8

(
)
76
B. Fontal et al.rJournal of Molecular Catalysis A: Chemical 149 1999 75–85
.
Ž
ies we reported the catalytic reactions of the Ru
CH2; JPH s11 Hz ; Mass Spectra solid, HP
Ž
.
carbonyl cluster in the presence of bis-1,3 di-
5988A GC-MS molecular ion at 969.8 amu,
.
Ž
phenylphosphino propane PPh2CH2CH2CH2-
PPh2 sdppp 21 . In this paper we report the
reactions of Ru3 CO dppm under hydrogen
decomposition pattern of consecutive CO loss
and ligand fragmentation. The compound de-
composes between 174 and 1768C.
.
w x
Ž
. Ž
.
10
Ž .
pressure.
b Synthesis via thermal reaction. The com-
plex was prepared by a modified version of the
w
x
reported method 24 . In a typical synthesis,
Ž
.
Ž
. Ž
2. Experimental
Ru3 CO
150 mg, 0.234 mmol , dppm 90
12
.
Ž
.
mg, 0.234 mmol and xylene Aldrich, 5 ml in
a round bottom flask were heated to 508C with
stirring and then refluxed for 5 h under Ar,
giving a red–brown solution. After solvent
evaporation, the residue is separated on silica
gel column chromatography, eluted with a 3r1
CH2Cl2rhexane mixture, giving several bands.
The larger red–brown band is recrystallized in
(
) ( )
2.1. Synthesis of Ru3 CO 10 dppm
Several synthetic methods were tried to com-
pare the percent yield and the ease of prepara-
tion. All the gases used are high purity, reagents
are analytical grade and all solvents were appro-
priately dried before use.
Ž .
Ž
.
a
Preparation via diphenylketyl radical,
CH2Cl2rhexane to give red–brown Ru3 CO
10
yP
Ž
.
Ž
.
Ž
.
Ph2C2O . The preparation of Ru3 CO
dppm by this method has been reported by
Bruce et al. 22 . Ru3 CO was prepared from
RuCl3 P 3H2O Strem Chemicals and CO
Matheson, UAP grade according to the litera-
ture method 23 . Sodium diphenylketyl was
prepared prior to use reacting benzophenone 91
mg, 0.5 mmol in very dry THF Aldrich, 20
dppm crystals 210 mg, 94% yield . Identifica-
10
Ž
.
Ž .
tion similar to a .
w
x
Ž
.
Ž .
c Synthesis via photolysis. In a typical syn-
12
Ž
.
Ž
.
Ž
.
thesis, Ru3 CO
150 mg, 0.234 mmol , dppm
12
Ž
.
Ž
.
Ž
.
90 mg, 0.234 mmol in THF 15 ml are heated
to 408C with stirring for several minutes under
Ar. The red orange solution is transferred to a
Schlenk tube with a water jacket, covered with
w
x
Ž
.
Ž
.
Ž
ml and very finely divided metallic sodium
Al foil under Ar; a pen Hg UV lamp Ultra-
under Ar with stirring for 2 h. The very intense
violet solution had a concentration near 0.025
Violets Products, 114 V, 60 cycle, low inten-
.
sity is introduced and photolyzed for about 5 h,
P
mmolrml of Ph2C2Oy. In a typical synthesis,
the progress of the reaction followed by disap-
Ž
.
Ž
.
Ž
.
Ru3 CO
Ž
150 mg, 0.234 mmol and the dppm
pearance of Ru3 CO metal carbonyl IR bands.
12
12
.
Ž
.
Strem Chemicals ligand 90 mg, 0.234 mmol
After the reaction is complete, the solvent is
reduced under vacuum and silica gel column
chromatographed using a 50r50 THFrhexane
eluant. The red–brown band is recrystallized in
Ž
.
are dissolved in dry THF 20 ml with stirring at
408C in Schlenk glassware. The diphenylketyl
solution is added dropwise with syringe and the
reaction followed by the disappearance of
Ž
.
CH2Cl2rhexane giving red–brown Ru3 CO
10
Ž
.
Ž
Ž
. Ž .
dppm crystals 216 mg, 95% yield . Identifica-
Ru3 CO metal carbonyl IR bands P.E. 1725-
12
.
Ž .
X FTIR . After solvent evaporation, the residue
tion similar to a .
Ž .
Ž
.
is separated on silica gel column chromatogra-
phy, eluted with a 50r50 THFrhexane mixture.
The brown–red fraction is recrystallized from
CH2Cl2rhexane to give red–brown crystals of
d Synthesis via CH3 3NO activation. In a
Ž
.
Ž
150 mg, 0.234
12
typical experiment Ru3 CO
.
Ž
.
mmol is dissolved in dry CH2Cl2 40 ml in a
Schlenk tube with constant Ar bubbling during
the whole reaction. The system is cooled to
Ž
. Ž
. Ž
.
Ru3 CO dppm 214 mg, 94.2% yield . IR:
Ž
. Ž 10
.
Ž
.
.
Ž
.
Ž .
Ž .
n CO CH2Cl2 2080 m , 2040 sh , 2010 s ,
1998 sh , 1958 m cm
Varian T-60 d 7.1–7.6 m, C6H5 , 4.3 t,
y788C in a Dewar flask with a CO2 s racetone
1
y
1
Ž
.
Ž
.
Ž
.
Ž
.
Ž
H NMR CDCl3
bath. Trimethylammine oxide, CH3 3NO Al-
Ž
.
Ž
.
Ž
drich, sublimed three times, 52.6 mg, 0.701

(
)
B. Fontal et al.rJournal of Molecular Catalysis A: Chemical 149 1999 75–85
77
.
Ž
.
mmol completely dissolved in CH2Cl2 15 ml
reactor is pressurized to the desired H2 pres-
sure, introduced in an oil bath at the desired
temperature allowing 5 min for thermal stabi-
lization. At the end of the run the reactor is
rapidly cooled and the liquid analyzed by gas
was added dropwise via canula to the Ru car-
bonyl solution at low temperature, the progress
of the reaction followed by disappearance of
Ru3 CO metal carbonyl IR bands. The dppm
ligand 90 mg, 0.234 mmol dissolved in
Ž
.
Ž12
.
Ž
chromatography P.E. AutoSystem 900, PE Nel-
Ž
.
CH2Cl2 15 ml was added dropwise via canula
at low temperature. The Schlenk tube is re-
moved from the cold bath and allowed to warm
up slowly to room temperature, heating then to
408C during 30 min. After evaporating the sol-
vent, the residue is chromatographed in a silica
gel column and the separated fraction recrystal-
lized in a 1r1 CH2Cl2rhexane mixture, giving
son software; stainless steel column, 15% tricre-
syl phosphate on Chromosorb P, 60–80 mesh, 3
m long, 0.6 cm diameter or capillary column
Plot fused silica, PoraPlot U, 12.5 m long, 0.53
.
Ž
mm i.d., 0.70 mm o.d. or GC-MS HP 5988A
.
GC-MS . Olefin ammination was carried out
using a 10r1 ArrNH3 mixture. Test reactions
with all the other unsaturated substrates were
carried out under similar conditions, adjusting
the H2 pressure, solvent or reaction time as
required.
Ž
. Ž
.
Ž
red–brown Ru3 CO dppm crystals 141 mg,
10
.
Ž .
62% yield . Identification similar to a . In Table
1, a comparison of the four synthetic methods is
presented.
Kinetic runs: In a 250 ml stainless steel high
Ž
pressure reactor Parr Instruments, glass liner,
internal stirring, internal thermocouple and sam-
(
) ( )
2.2. Catalytic trials with Ru3 CO 10 dppm
y
2
.
Ž
. Ž
. Ž
ple outlet Ru3 CO dppm 20 mg, 2=10
.
Ž 10
mmol , 1-hexene 10 ml, 80 mmol; 4000r1
. Ž
substratercatalyst ratio , ethanol solvent, 50
In a typical hydrogenation or isomerization
trial, Ru3 CO dppm 2 mg, 2=10 mmol ,
y
3
Ž
Ž
. Ž
. Ž
.
.
Ž
.
ml , cyclohexane 10 ml, internal standard were
mixed. After purging with H2 three times, hy-
drogen pressure adjusted to 100 psi and 5 min
10
1-hexene Aldrich, 1 ml, 8 mmol; 4000r1 sub-
.
Ž
.
stratercatalyst ratio , 5 ml solvent e.g., ethanol
Ž
.
are mixed in a high pressure stainless steel
allowed for thermal stabilization 908C , sam-
ples were taken every 5 min during 0.5 h. The
samples were immediately quenched and ana-
lyzed by gas chromatography.
Ž
reactor 10 ml, glass liner, internal magnetic
stirrer, 20–2000 psi manometer, Parr Instru-
ments . After purging with H2 three times, the
.
Table 1
Ž
.
Ž
.
Synthesis methods for Ru3 CO dppm
10
Ž
.
Synthesis
method
Radical
Thermal
Photochemical
Assisted with CH3 3NO
Ž
.
Yield %
94.2
73
95
62
Solvent
THF
Xylene
THF
CH2Cl2
General
characteristics
Good yield
and selectivity.
Less clean
reaction.
Preparation
and handling
of radical very
air and moisture
sensitive.
Acceptable
yield. Less
selectivity,
several
Good yield and
selectivity.
Very clean reaction.
Product separation
is simpler.
Acceptable yield and good
selectivity. Intermediates
are very sensitive.
Separation of products
more complex.
products
are obtained,
that requires
separation.

(
)
78
B. Fontal et al.rJournal of Molecular Catalysis A: Chemical 149 1999 75–85
Table 2
Hydrogenation and isomerization of 1-hexene
Solvent
Percent n-hexane
Percent 1-hexene
Percent cis-2-hexene
Percent trans-2-hexene
Ethanol
2-ethoxyethanol
THF
Toluene
Benzene
Heptane
7.1
20.9
12.0
16.1
10.4
8.4
1.5
1.9
34.9
58.6
66.9
66.5
72.4
59.9
36.5
16.5
13.7
15.5
19.1
17.3
17.2
8.8
9.0
9.7
Ž
.
Ž
.
Ž
.
Ž
. Ž
.
Solvent effect. Reaction conditions: solvent 5 ml , 1-hexene 1 ml, 0.67 g , Ru3 CO dppm 2 mg , PH s100 psi, Ts908C, ts2 h.
10
2
Ž .
A systematic study of several reaction condi-
tions were carried out.
e Amount of catalyst effect: Amount of
catalyst change results are shown in Fig. 4.
Ž .
Ž .
a Solvent effect: The following solvents
Reaction conditions: Ethanol 5 ml , 1-hexene
Ž
.
1 ml, 0.67 g , PH s100 psi, Ts908C, ts2
were tried: ethanol, 2-ethoxyethanol, THF,
tolueno, benzene, heptane. Reaction conditions:
solvent 5 ml , 1-hexene 1 ml, 0.67 g , Ru3-
2
Ž
h. A catalyst amount of 2 mg about 4000r1
Ž
.
. Ž
Ž
.
.
substratercatalyst ratio was selected for other
Ž
. Ž
.
CO dppm 2 mg , PH s100 psi, Ts908C,
catalytic studies.
10
ts2 h. The results are2 shown in Table 2.
Ethanol was the solvent selected for the other
catalytic studies.
f Hydrogenation of 2-hexene 11% cisr89%
Ž .
Ž
.
trans thermodynamic isomer mixture : A trial
reaction was run with the isomer mixture. Reac-
Ž .
Ž . Ž
isomer mixture , cyclohexane 1 ml, internal
b
PH effect: H2 pressure change results
tion conditions: Ethanol 5 ml , 2-hexene 1 ml
are show2n in Fig. 1. Reaction conditions:
.
Ž
Ž
.
Ž
.
.
Ž
. Ž
. Ž
.
Ethanol 5 ml , 1-hexene 1 ml, 0.67 g ,
standard , Ru3 CO dppm 2 mg , PH s100
10
2
Ž
. Ž
. Ž
.
Ru3 CO dppm 2 mg , Ts908C, ts2 h. A
psi, Ts908C. Product distribution: after 6 h:
10
Ž . Ž .
cis-2-hexene 15% , trans-2-hexene 50% , 1-
100 psi H2 pressure was selected for other
Ž
.
Ž
.
catalytic studies. H2 pressure effects were also
analyzed by the following experiments: Exp. 1
similar conditions as above, but use pressure
hexene 35% ; after 24 h: cis-2-hexene 73% ,
trans-2-hexene 15% , n-hexane 12% .
Ž
.
Ž
.
Ž
.
Ž
. Ž
.
changes of Ar no H2 pressure ; Exp. 2 simi-
lar conditions as above, but use pressure changes
Ž
.
of Arq15 psi H2; Exp. 3 similar conditions
as above, but use pressure changes of ArrH2
Ž
.
50r50 . The results are shown in Table 3.
Ž .
c Temperature effect: Reaction temperature
change results are shown in Fig. 2. Reaction
Ž
.
Ž
conditions: Ethanol 5 ml , 1-hexene 1 ml, 0.67
.
Ž
. Ž
. Ž
.
g , Ru3 CO dppm 2 mg , PH s100 psi,
10
2
ts2 h. A 908C temperature was selected for
other catalytic studies.
Ž .
d
Reaction time effect: Reaction time
change results are shown in Fig. 3. Reaction
Ž
.
Ž
conditions: Ethanol 5 ml , 1-hexene 1 ml, 0.67
.
Ž
. Ž
. Ž
.
g , Ru3 CO dppm 2 mg , PH s100 psi,
10
Ts908C. A 2 h reaction time was 2selected for
Fig. 1. Hydrogenation and isomerization of 1-hexene. H2 pres-
Ž
.
Ž
sure effect. Reaction conditions: Ethanol 5 ml , 1-hexene 1 ml,
. Ž
the other catalytic studies.
.
Ž
.
Ž
.
0.67 g , Ru3 CO dppm 2 mg , T s908C, ts2 h.
10

(
)
B. Fontal et al.rJournal of Molecular Catalysis A: Chemical 149 1999 75–85
79
3. Results and discussion
tion of polar andror coordinatively unsaturated
intermediates in the catalytic cycle. The ten-
dency to favor the cis-2-isomer clearly indicates
an important steric requirement during the cat-
alytic cycle.
(
) ( )
3.1. Synthesis of Ru3 CO 10 dppm
The several synthetic methods tried for the
Ž
. Ž
.
Ž .
synthesis of Ru3 CO dppm and shown in
b PH effect: The isomerization reaction is
10
2
Table 1, indicate that the complex can be ob-
tained in good yield using the diphenylketyl
radical and photochemical activation. Lesser
yields are observed with the thermal reaction,
since this method is less selective and various
products are obtained, and with CH3 3NO there
is good selectivity, but decomposition of the
trimethylamine Ru carbonyl intermediate lowers
the overall yield. The first two methods are thus
recommended.
greatly favored at low H2 pressure, giving pre-
Ž
.
dominantly the cis-2-hexene isomer Fig. 1 . At
100 psi H2 pressure, all the 1-hexene substrate
has been consumed, mainly to give the internal
isomers, and some hydrogenation product, n-
hexane, its concentration increasing as H2 pres-
sure increase; the concentration of isomers de-
creases rapidly and at 1000 psi H2 pressure, the
main product is the saturated alkane. To test
further the H2 pressure behavior, especially the
low pressure region, several reactions in the
Ž
.
Ž
.
3.2. Catalytic results
presence of Ar were carried out Table 3 . In
the absence of H2, only the isomerization reac-
tion favoring the cis-2-hexene isomer is ob-
served in low yield; this process does not con-
sume hydrogen and increases slightly with in-
creasing Ar pressure to a maximum overall
yield of about 30% when the total pressure
reaches 1000 psi. When a 15 psi partial H2
pressure is included, the overall yield improves
and increases slightly with increasing total pres-
Ž .
a Solvent effect: As shown in Table 2,
under the reaction conditions used, more polar
and better coordinating solvents like ethanol,
THF and 2-ethoxyethanol give better overall
yields, favoring the isomerization reaction with
cis-2-hexene in greater percentage. Less polar
solvents with lesser coordinating ability like
toluene, benzene and heptane, give less overall
yields and less marked product selectivity. This
solvent behavior indicates a possible stabiliza-
Ž
.
sure Arq15 psi H2 , still favoring isomeriza-
tion of 1-hexene mainly to the cis-2-hexene
Table 3
Hydrogenation and isomerization of 1-hexene
Ž
.
Ar pressure psi
Percent n-hexane
Percent 1-hexene
Percent cis-2-hexene
Percent trans-2-hexene
0
15
100
500
0.3
0.3
0.3
0.3
0.4
99.8
90.4
81.8
76.6
71.0
0
6.5
12.3
17.0
20.8
0
2.8
5.6
6.0
7.9
1000
Arq15 psi H2
100
2.4
2.5
2.5
13.5
11.0
5.5
58.2
65.2
70.3
25.9
22.4
21.7
500
1000
Ž
.
ArqH2 50r50
15
4.6
35.6
47.7
48.9
6.8
2.6
1.8
0.7
61.7
43.4
42.1
41.8
19.1
16.3
8.4
100
500
1000
8.6
Ž
.
Ž
.
Ž
.
Ž
.
Ž
. Ž
.
ArqH2 pressure effect. Reaction conditions: Ethanol 5 ml , 1-hexene 1 ml, 0.67 g , Ru3 CO dppm 2 mg , Ts908C, ts2 h.
10

(
)
80
B. Fontal et al.rJournal of Molecular Catalysis A: Chemical 149 1999 75–85
Fig. 2. Hydrogenation and isomerization of 1-hexene. Tempera-
Ž
.
Ž
ture effect. Reaction conditions: Ethanol 5 ml , 1-hexene 1 ml,
. Ž
.
Ž
.
Ž
.
0.67 g , Ru3 CO dppm 2 mg , PH s100 psi, ts2 h.
10
2
Fig. 4. Hydrogenation and isomerization of 1-hexene. Amount of
Ž
.
catalyst effect. Reaction conditions: Ethanol 5 ml , 1-hexene
Ž
.
1 ml, 0.67 g , PH s100 psi, T s908C, ts2 h.
2
isomer, and very little hydrogenation product.
When a 50r50 ArrH2 mixture is used, the
product distribution is the expected isomeriza-
tion favoring cis-2-hexene at lower partial hy-
drogen pressure and increases in the hydrogena-
tion product, n-hexane, as the hydrogen partial
pressure increases. There is a greater proportion
of isomers at the higher H2 pressures when
using an ArrH2 mixture than when an equiva-
lent pure hydrogen pressure is used. It becomes
apparent than the isomerization route in the
catalytic cycle can be activated in the absence
of hydrogen pressure, probably via hydrogen
abstraction from the p-bonded olefin to form a
p-allyl intermediate; this route gets enhanced
under low hydrogen pressure, or a different
mechanism, sharing a common s-alkyl interme-
diate with the hydrogenation path could start
dominating the reaction.
Ž .
c Temperature effect: The overall yield is
low at 608C and increases rapidly with increas-
Ž
.
ing temperature Fig. 2 . The isomerization re-
action increases more rapidly, greatly favoring
the cis-2-hexene isomer and reaching a maxi-
mum percent conversion at about 908C. At this
temperature all the substrate has been trans-
formed to products. This is indicative that some
minimum activation energy is required for the
isomerization reaction to proceed, probably as-
sociated to ligand loss, like a CO group or
phosphorus dissociation, to produce the coordi-
natively unsaturated species required for sub-
strate substitution in the catalytic cycle. The
hydrogenation reaction increases slowly with
the temperature rise, and more rapidly after
908C with a concomitant decrease in isomeriza-
tion products, n-hexane being the dominant
product at 1308C, the highest temperature tried.
The higher temperature required for the hydro-
genation reaction to proceed indicates a higher
activation energy for this reaction path, most
likely associated with ligand dissociation and
H2 activation necessary for H2 oxidative addi-
tion in the catalytic cycle.
Fig. 3. Hydrogenation and isomerization of 1-hexene. Reaction
Ž
.
Ž
time effect. Reaction conditions: Ethanol 5 ml , 1-hexene 1 ml,
. Ž
.
Ž
.
Ž
.
0.67 g , Ru3 CO dppm 2 mg , PH s100 psi, T s908C.
10
2

(
)
B. Fontal et al.rJournal of Molecular Catalysis A: Chemical 149 1999 75–85
81
Ž .
d
Reaction time effect: The reaction
to the terminal olefin is faster than trans–cis
isomerization; after 24 h, the cis-2-hexene iso-
mer predominates, and some hydrogenation
products have already appeared. This result in-
dicates that the internal olefins can enter the
catalytic cycle and be equilibrated with the other
products in reversible reactions. The fact that
the times required for reaction are longer than
when the initial substrate is 1-hexene indicate
that the internal olefins are more difficult to
coordinate due to steric reasons and that the
reverse reaction to p-olefin dissociation in the
catalytic cycle is slower.
progress with time is shown in Fig. 3. After 30
min, the percent overall conversion is already
near 40%, with greater proportion of isomeriza-
tion and favoring the cis-2-hexene internal iso-
mer. After 2 h the reaction is practically com-
plete, with the maximum production of cis-2-
hexene. After this time complete hydrogenation
to n-hexane increases rapidly, the internal iso-
mers likewise being hydrogenated as the reac-
tion proceeds. These results indicate that the
isomerization reaction is faster than hydrogena-
tion, the initial alkane being produced by hydro-
genation of the terminal 1-hexene olefin and
when this is consumed, the internal olefins that
are more difficult to hydrogenate get reduced to
alkane.
Ž .
Ž
. Ž
.
g Stability studies of Ru3 CO dppm : In
10
homogenous catalysis it is important to verify
that the reaction is carried out by the cluster
complex rather than by another species derived
from cluster fragmentation or metallic particles
Ž .
e Amount of catalyst effect: A reaction
Ž
. Ž
.
Ž
.
from complex decomposition. Ru3 CO
12
using 1 mg of Ru3 CO dppm as catalyst
was
10
precursor, under the reaction conditions used,
produces more than 90% conversion, predomi-
nating the cis-2-hexene isomerization product
used in a test reaction under similar reaction
conditions. The results are shown in Table 4.
Ž
.
12
The Ru3 CO cluster gives a similar catalytic
Ž
.
Ž
.
behavior but is less active than Ru3 CO -
10
Fig. 4 . As the amount of Ru complex is
Ž
.
dppm , and readily decomposes to metallic par-
increased, more hydrogenation product is ob-
served, being the predominant product when 20
mg of compound are used. As the substrater
catalyst ratio decreases, more active sites are
introduced, increasing the probability for com-
plete olefins hydrogenation, including the inter-
nal olefins produced more rapidly during the
isomerization process.
ticles above 300 psi H2 pressure. A test reaction
carried out under similar reaction conditions,
Ž
.
Ž
.
2 mg
100 psi H2 but containing Ru3 CO
12
Ž
.
and dppm ligand 1.20 mg gives a higher per-
cent total yield, with the following product dis-
tribution: 68.4% cis-12-hexene, 16.9% trans-2-
hexene and 5.9% n-hexane; this indicates that a
dppm substituted Ru cluster is formed in situ,
providing greater thermal stability to the ruthe-
nium carbonyl cluster. The dppm substituted
cluster does not show decomposition signs up to
1000 psi H2 pressure, indicating that the biden-
tate ligand stabilizes the trinuclear cluster. This
is verified by observing the final reaction solu-
Ž .
f Hydrogenation of 2-hexenes: The results
obtained when the thermodynamic isomer mix-
ture of trans-2-hexene and cis-2-hexene is
placed under the reaction conditions for isomer-
ization of 1-hexene, demonstrates that after suf-
ficient time the kinetic mixture is equilibrated.
The results after 6 h, indicate that isomerization
Table 4
Ž
.
12
Catalysis study of 1-hexene with Ru3 CO
Ž
.
H2 pressure psi
Percent n-hexane
Percent 1-hexene
Percent cis-2-hexene
Percent trans-2-hexene
100
300
500
12.6
55.1
66.4
27.0
1.0
0.0
42.0
35.4
27.7
18.3
8.3
5.9
Ž
.
Ž
.
Ž
.
Ž
.
H2 pressure effect. Reaction conditions: Ethanol 5 ml , 1-hexene 1 ml, 0.67 g , Ru3 CO
2 mg , Ts908C, ts2 h.
12

(
)
82
B. Fontal et al.rJournal of Molecular Catalysis A: Chemical 149 1999 75–85
and benzene with reflux at high temperatures in
noncoordinating solvents to give species as
Ž
Ž
.
Ž
.Ž
..Ž
.
Ru3 m3-P C6 H5 CH 2 P C6 H5 C6 H 4 CO
8
w
x
28–30 . Our reaction conditions do not favor
cluster decomposition and the solution color
remains after reaction, but the presence of
species as the thermolysis product above cannot
be discarded as possible intermediates. This
species is presently being investigated under our
reaction conditions to compare its possible cat-
alytic behavior.
Ž .
h Kinetic results: The first order plot for the
kinetic studies in the isomerization and hydro-
genation of 1-hexene, following the disappear-
ance of substrate is shown in Fig. 6. The results
indicate first order on substrate with an apparent
Fig. 5. Hydrogenation and isomerization of 1-hexene. Turnover
Ž
Ž
.
.Ž
..
frequency millimoles productr millimoles catalyst minutes vs.
Ž
Ž
Amount of catalyst millimoles . Reaction conditions: Ethanol 5
.
Ž
.
Ž
ml , 1-hexene 1 ml, 0.67 g , cyclohexane 1 ml, internal stan-
.
dard PH s100 psi, T s908C, ts0.5 h.
1
rate constant value of 0.11 miny for the total
2
reaction.
tion under the microscope and by running a test
reaction in the presence of metallic mercury
3.3. Catalysis with other substrates
Ž
Ž
.
with similar conditions ethanol 5 ml , 1-hexene
Ž
.
Ž
. Ž
. Ž
.
Ž
. Ž
.
1 ml, 0.67 g , Ru3 CO dppm 2 mg , PH
The Ru3 CO dppm complex was tried
10
10
2
.
w
x
s100 psi, Ts908C, ts2 h 25,26 . When Hg
is used the change in percent yield for the
different products is less than 5%, discarding
metallic Ru as the active catalyst.
with other organic substrates to explore the
extent of its catalytic activity. The results are
summarized in Table 5. Terminal olefins like
1-hexene and styrene get hydrogenated almost
completely at 1000 psi H2 pressure and 90 min
Several reactions were carried out increasing
Ž
. Ž
.
the number of millimoles of Ru3 CO dppm
10
and the products analyzed before reaching 40%
Ž
total conversion reaction conditions: ethanol
Ž
Ž
.
Ž
.
5 ml , 1-hexene 1 ml, 0.67 g , cyclohexane
.
1 ml, internal standard , PH s100 psi, T
2
.
s908C, ts0.5 h . A plot of turnover frequency
Ž
Ž
Ž
.
millimoles of productr millimoles catalyst
..
minutes vs. millimoles of catalyst gives an
indication of cluster catalysis, possible fragmen-
tation or catalysis by higher nuclearity species
w
x
27 . The plot is shown in Fig. 5 for isomeriza-
tion and hydrogenation results. Both reactions
show an increase in turnover frequency as the
amount of Ru complex increases, but the slopes
are rather different. The increasing tendency is
interpreted as indicative of cluster catalysis
rather than produced by active fragments of
Fig. 6. Hydrogenation and isomerization of 1-hexene. Kinetic
Ž
.
.
Ž
results. Reaction conditions: Ethanol 50 ml , 1-hexene 10 ml,
w
x
Ž
. Ž
.
lower nuclearity 27 . The Ru3 CO dppm
10
.
Ž
Ž
.
Ž
.
6.7 g , cyclohexane 10 ml, internal standard , Ru3 CO dppm
10
complex can be thermally induced to lose CO
Ž
.
20 mg , PH s100 psi, T s908C.
2

Table 5
.
Catalytic reactions of unsaturated organic substrates with Ru3 CO dppm
Ž
.
Ž
10
Ž
.
Ž .
Time h
Substrate
Product %
Conditions
( )
a Hydrogenation of unsaturated organic substrates
Ž
.
Ž
.
Ž
.
1-hexene
styrene
1-hexyne
n-hexane 91.9 , cis-2-hexene 6.5 , trans-2-hexene 1.6
ethanol, 908C, 1000 psi H2
ethanol, 908C, 1000 psi H2
ethanol, 908C, 1500 psi H2
1.5
1.5
5
Ž
.
ethylbenzene 96.3
Ž . Ž . Ž .
n-hexane 52.4 , 1-hexene 33.4 , trans-2-hexene 5.9 ,
Ž
.
cis-2-hexene 6.3
Ž
.
cyclohexene
benzene
acetone
cyclohexane 96.2
ethanol, 908C, 1000 psi H2
ethanol, 908C, 1500 psi H2
THF, 908C, 1600 psi H2
THF, 908C, 1600 psi H2
2
24
18
18
Ž
.
Ž
.
Ž
.
cyclohexadiene 1.3 , cyclohexene 5.2 , cyclohexane 7.8
Ž
.
isopropanol 32.1
Ž .
cyclohexanol 37.9
cyclohexanone
( )
b Hydrogenation of a -b unsaturated aldehydes
Ž
.
Ž
.
Ž
.
.
acrolein
crotonaldehyde
cinnamaldehyde
propanaldehyde 52.9 , propanol 27.3 , 2-propenol 4.9
THF, 908C, 1600 psi H2
THF, 908C, 1600 psi H2
THF, 908C, 1600 psi H2
8
8
8
Ž
.
Ž
.
Ž
butyraldehyde 36.3 , butanol 22.1 , 2-butenol 21.4
3-phenylpropanal 12.2 , 3-phenylpropanol 17.6 ,
Ž
.
Ž
.
Ž
.
3-phenyl-2-propenol 46.3
( )
c Isomerization of unsaturated organic substrates
Ž
.
Ž
.
Ž
.
1-hexene
1-hexyne
n-hexane 7.1 , cis-2-hexene 72.4 , trans-2-hexene 19.1
ethanol, 908C, 100 psi H2
ethanol, 908C, 1500 psi H2
2
5
Ž
.
Ž
.
Ž
.
n-hexane 52.4 , 1-hexene 33.4 , trans-2-hexeno 5.9 ,
Ž
.
cis-2-hexeno 6.3
methylcyclohexene 15.6 , 2-methylcyclohexene 5.4 ,
Ž
.
Ž
.
1-methylcyclohexene
allylic alcohol
THF, 908C, 500 psi H2
THF, 908C, 200 psi H2
4
2
Ž
.
Ž
3-methylcyclohexene 2.3
Ž
.
.
propanal 33.3 , propanol 26.0
( )
d Hydroformylation of 1-hexene
Ž
.
Ž
.
Ž
.
1-hexene
2-methylhexanal 8.4 , heptanaldehyde 34.7
THF, 1308C, 1000 psi H2 rCO 1r1
6
( )
e Amine synthesis. Hydrogenation
Ž
.
acetonitrile
benzonitrile
nitrobenzene
ethylamine, diethylamine, triethylamine 20 total
ethanol, 908C, 1600 psi H2
ethanol, 908C, 1600 psi H2
ethanol, 908C, 1600 psi H2
24
24
24
Ž
.
benzylamine 66.6
Ž
.
aniline 42.1
Condensation of NH3 with alcohol and olefins
Ž
.
Ž
.
CH3OHrNH3
CH3NH2 15
methanol, 908C, 1000 psi NH3 rAr 10r1 24
Ž
.
.
Ž
.
CH3CH2OHrNH3
PropylenerNH3
CH3CH2 NH2 23
propylamine 13.3 , isopropylamine 5.8
ethanol, 908C, 1000 psi NH3 rAr 10r1
24
16
Ž
Ž
.
THF, 1258C, 30 mg cat,
1000 psi propylenerNH3 rAr,
5r10r150
Ž
.
Ž
.
1-hexenerNH3
hexylamine 16.4 , 2-hexylamine 7.2
THF, 1258C, 30 mg cat, 1-hexeno,
1000 psi ArrNH3 15r1
16
Ž
.
Ž
.
Ž
.
Ž
. Ž
.
Reaction conditions: Solvent 5 ml , substrate 1 ml , Ru3 CO dppm 2 mg .
10

(
)
84
B. Fontal et al.rJournal of Molecular Catalysis A: Chemical 149 1999 75–85
reaction; cyclic olefins like cyclohexene are
harder to hydrogenate and take longer to get
high percent conversions. Alkynes as 1-hexyne
require stronger hydrogenation conditions and
more reaction time to attain complete reduction,
important yield of the intermediate 1-hexene is
observed and small amounts of isomerization
products. Aromatic substrates like benzene and
ketones as acetone and cyclohexanone are more
difficult to hydrogenate, requiring stronger reac-
tion conditions and longer reaction time to get
partial reduction. For substrate with two func-
tional groups susceptible of reduction as in a,b
unsaturated aldehydes, the hydrogenation results
indicate that both the carbon–carbon double
bond and the carbonyl group can be hydro-
genated, under stronger reaction conditions and
longer times. Complete reduction is observed in
important yield for acrolein and crotonaldehyde,
but for cinnamaldehyde, carbonyl reduction pre-
dominates. This results indicate that the Ru
complex studied does not show strong regiose-
lectivity in its hydrogenating activity. The iso-
merization reaction observed with 1-hexene,
which favors the cis-2-hexene isomer, was ex-
tended to other substrates. With 1-hexyne, the
stronger reduction conditions required for the
first hydrogenation, only allows accumulation
of a small percentage of the expected isomers
derived from 1-hexene before complete hydro-
genation occurs. With 1-methylcyclohexene,
only small amounts of isomerization to the 2
and 3 positions are observed. With allylic alco-
hol, double bond migration to form the carbonyl
group occurs in good proportion, but the com-
peting complete reduction reaction is also very
prevalent. A test reaction under hydroformyla-
tion conditions with 1-hexene, favors the linear
aldehyde, with some branched aldehyde pro-
duced, but with intermediate overall percent
yield. Our research group is interested in the
catalytic activity of ruthenium complexes for
synthesis of amines. Hydrogenation of nitriles,
such as acetonitrile and benzonitrile and nitro
groups as in nitrobenzene, to give the expected
amines requires higher H2 pressure and longer
reaction times. With acetonitrile, the scrambling
reaction is apparent, and this reaction is being
further pursued in our laboratory. The condensa-
tion reaction between NH3 and small alcohols
such a methanol and ethanol provides reason-
able amine yields under the reaction conditions
tried. The reactions between NH3 and olefins
such as propylene and 1-hexene under the reac-
tion conditions tried, give the N–H addition
products, with a slight preference for the termi-
nal amine.
Ž
. Ž
.
The Ru3 CO dppm complex has shown
10
good hydrogenation activity for several unsatu-
rated organic groups, and a strong isomerization
tendency for 1-hexene at low hydrogen pres-
sures, favoring the cis-2-hexene isomer, indica-
tive of an important steric requirement that could
have eventual application in regioselective or
stereoselective hydrogenation reactions. The
complex also shows an interesting N–H activa-
tion capacity that is being explored for specific
amine synthesis.
Acknowledgements
The present work was financed by Bid-Con-
Ž
.
Ž
icit Proy. QF-04 Also Conicit Proy. F-124 for
.
FTIR and Proy. S1 2338 for GC .
References
w x
Ž
.
1
D. Shriver, H.D. Kaesz, R. Adams Eds. , The Chemistry of
Metal Cluster Complexes, VCH Publishers, New York, 1990.
G. Suss-Fink, G. Meister, Advances in Organometallic
Chemistry, Academic Press, 35, 1993, 41.
w x
2
w x
3
M. Castiglioni, L. Milone, D. Osella, G. Vaglio, Inorg.
Ž
.
Chem. 15 1976 394.
w x
Ž
.
.
4
M. Valle, D. Osella, G. Vaglio, Inorg. Chim. Acta 20 1976
213.
w x
5
Ž
J.L. Graff, R.D. Sanner, M.S. Wrighton, JACS 101 1979
273.
w x
6
R. Sanchez-Delgado, I. Duran, J. Manfort, E. Rodriguez, J.
Ž
.
Mol. Catal. 11 1981 193.
w x
7
A. Basu, H. Bhaduri, H. Khwaja, P.G. Jones, T. Schroeder,
Ž
.
G.M. Sheldrick, J. Organomet. Chem. 290 1985 C19.
w x
8
C. Bergounhou, J.J. Bonnet, P. Fompeyrine, G. Lavigne, N.
Ž
.
Logun, F. Mansilla, Organometallics 5 1985 60.

(
)
B. Fontal et al.rJournal of Molecular Catalysis A: Chemical 149 1999 75–85
85
w x
w
x
9
C. Bergounhou, P. Fompeyrine, G. Commenges, J.J. Bonnet,
21 B. Fontal, R. Hernandez, T. Suarez, M. Reyes, F. Bellandi,
Revista Ciencia, Fac. Ingenieria, Univ. Zulia, submitted for
publication, 1998.
Ž
.
J. Mol. Catal. 48 1988 285.
w
w
w
x
10 M. Bruce, J. Hinchliffe, R. Surynt, B. Skelton, A. White, J.
Ž
.
w
x
Organomet. Chem. 469 1994 89.
22 M. Bruce, J. Mathisons, B. Nicholson, J. Organomet. Chem.
x
Ž
.
Ž .
247 1983 321.
11 H. Mirza, J. Vittal, J. Puddephatt, Inorg. Chem. 32 1993
w
x
1327.
23 M. Bruce, J. Mathisons, R. Wallis, J. Patrick, B. Skelton, J.
x
12 A. Colombie, G. Lavigne, J.J. Bonnet, J. Chem. Soc. Dalton.
Chem. Soc. Dalton Trans., 1983, 2365.
Ž
.
w
x Ž .
24 A. Cotton, B. Hanson, Inorg. Chem. 16 1977 3369.
Trans. 1986 899.
w
w
x
Ž
.
w
x
Ž
.
.
13 G. Lavigne, J.-J. Bonnet, Inorg. Chem. 20 1981 2713.
25 P. Foley, R. DiCosimo, G.M. Whitesides, JACS 102 1980
x
14 C. Adams, M. Bruce, P. Durkworth, P. Humphrey, O. Kuhl,
¨
6713.
w
w
w
x Ž .
E. Tiekink, W. Cullen, P. Braunstein, C. Cea, B. Skelton, A.
26 P. Garrow, Chem. Rev. 81 1981 229.
x Ž .
27 R. Laine, J. Mol. Catal. 14 1982 137.
28 N. Lugan, J.J. Bonnet, J.A. Ibers, Organometallics 7 1988
Ž
.
Ž
White, J. Organomet. Chem. 467 1994 251.
w
w
w
w
x
.
x
Ž
15 T. Suarez, B. Fontal, J. Mol. Catal. 32 1985 19.
x
Ž
.
16 T. Suarez, B. Fontal, J. Mol. Catal. 45 1988 335.
1538.
x
Ž
.
w x
29 N. Lugan, J.A. Ibers, J.J. Bonnet, J. Am. Chem. Soc. 107
17 T. Suarez, B. Fontal, D. Garcia, J. Mol. Catal. 34 1986 163.
x
Ž
.
Ž
.
18 T. Suarez, B. Fontal, H. Medina, J. Mol. Catal. 50 1989
1985 4484.
w
x
355.
30 M.I. Bruce, E. Horn, O.B. Shawkataly, M.R. Snow, E.R.T.
w
x
Ž
.
19 T. Suarez, B. Fontal, D. Garcia, Acta Cient. Venez. 34
Tiekink, M.L. Williams, J. Organomet. Chem. 316 1986
Ž
.
1983 198.
187.
w
x
Ž
.
20 B. Fontal, Acta Cient. Venez. 33 1982 202.