Hydrogenation of citral using monometallic Pt and bimetallic Pt-Ru catalysts on a mesoporous support in supercritical carbon dioxide medium

Article abstract of DOI:10.1002/adsc.200303160

Supercritical carbon dioxide was shown to be a suitable reaction medium for the highly efficient hydrogenation of citral using monometallic Pt and bimetallic Pt-Ru supported on a mesoporous material, MCM-48, as catalyst. A remarkable change in the product distribution was observed after the addition of Ru to the monometallic Pt catalyst in supercritical carbon dioxide. The monometallic Pt catalyst was found to be highly selective to the unsaturated alcohol (geraniol + nerol) at a temperature of 323 K within 2 h whereas the bimetallic catalyst becomes selective to the partially saturated aldehyde (citronellal) under the same reaction conditions. Phase behavior plays an important role in the product distribution. Highest conversion and high selectivity to citronellal were achieved in the homogeneous phase for the Pt-Ru catalyst while on the other hand the unsaturated alcohol (geraniol + nerol) was produced in the heterogeneous phase for the monometallic Pt catalyst. An XPS study offers strong evidence of the electronic modification of Pt after the addition of Ru in the bimetallic catalyst. The change in product distribution on the Pt-Ru bimetallic catalyst may be explained by the appreciable interaction between the medium and the metal particles promoted by the presence of metallic Ru.

Full text of DOI:10.1002/adsc.200303160

FULL PAPERS
Hydrogenation of Citral using Monometallic Pt and Bimetallic
Pt-Ru Catalysts on a Mesoporous Support in Supercritical
Carbon Dioxide Medium
M. Chatterjee,a F. Y. Zhao,a Y. Ikushimaa, b,*
a
Supercritical Fluid Research Center, National Institute of Advanced Industrial Science and Technology, 4-2-1 Nigatake,
Miyagino-ku, Sendai 983 8551, Japan
Phone: ( ‡ 81)-22-237-5211; fax: ( ‡ 81)-22-237-5224, y-ikushima@aist.go.jp
b
CREST, Japan Science and Technology Corporation (JST), Kawaguchi 332 0012, Japan
Received: October 10, 2003; Accepted: February 11, 2004
Abstract: Supercritical carbon dioxide was shown to to citronellal were achieved in the homogeneous
be a suitable reaction medium for the highly efficient phase for the Pt-Ru catalyst while on the other hand
hydrogenation of citral using monometallic Pt and the unsaturated alcohol (geraniol ‡ nerol) was pro-
bimetallic Pt-Ru supported on a mesoporous materi- duced in the heterogeneous phase for the monome-
al, MCM-48, as catalyst. Aremarkable change in the
tallic Pt catalyst. An XPS study offers strong evidence
product distribution was observed after the addition of the electronic modification of Pt after the addition
of Ru to the monometallic Pt catalyst in supercritical of Ru in the bimetallic catalyst. The change in
carbon dioxide. The monometallic Pt catalyst was product distribution on the Pt-Ru bimetallic catalyst
found to be highly selective to the unsaturated may be explained by the appreciable interaction
alcohol (geraniol ‡ nerol) at a temperature of 323 K between the medium and the metal particles promot-
within 2 h whereas the bimetallic catalyst becomes ed by the presence of metallic Ru.
selective to the partially saturated aldehyde (citro-
nellal) under the same reaction conditions. Phase Keywords: citral; hydrogenation; monometallic Pt;
behavior plays an important role in the product phase behavior; selective bimetallic Pt-Ru catalysts;
distribution. Highest conversion and high selectivity supercritical carbon dioxide
Introduction
hydrogenated most easily followed by the conjugated
C C and C O bonds. Several authors have reported the
ˆ
ˆ
Selective hydrogenation of a,b-unsaturated aldehydes is liquid-phase hydrogenation of citral over noble metals
an important research area because of the relevance of on different supports,[2] bimetallic catalysts such as Rh-
these compounds in the pharmaceutical and fine Ge,[3] Ru-Sn,[4] and basic catalysts [5] in various organic
chemical industries.[1] Citral is an unsaturated aldehyde, solvents. The effectiveness of supported noble metals
hydrogenated products of which have been widely used for the selective hydrogenation of citral can also be
ˆ
in the fragrance industry. It possesses an isolated C C enhanced by the addition of promoters such as Sn, Ge,
ˆ
ˆ
bond besides the conjugated C O and C C bonds, an Fe, etc., whereas by altering the operating conditions
interesting reactant from the scientific point of view. such as temperature, pressure, and especially solvent,
Scheme 1 displays the reaction network of citral hydro- the product distribution can be varied significantly.[6]
genations. The hydrogenation of the conjugated C O However, in hydrophobic solvents (such as cyclohexane,
bond leads to cis- and trans-isomers of the unsaturated alcohols) and in the presence of quite acidic catalysts,
ˆ
alcohol (geraniol and nerol), while hydrogenation of the different types of side reaction occur (acetal formation,
ˆ
ˆ
conjugated C C bond or the isolated C C bond yields cyclization etc), leading to a drop in selectivity for the
either the saturated aldehyde (citronellal). Moreover, desired product.[7]
the resulting unsaturated alcohol or partially saturated
Supercritical carbon dioxide (scCO2) is receiving
aldehyde can by hydrogenated at its remaining unsatu- considerable and growing interest as an environmen-
rated bond to yield the partially saturated alcohol tally benign reaction medium for organic synthesis. The
(citronellol). In addition, there are the possibilities of high miscibility of scCO2 with many reactant gases can
cis-trans isomerization or several types of side reactions help to avoid mass-transfer limitations,[8] which are
such as cyclization, acetalization, etc. Thermodynamic frequently encountered in gas-liquid phase reactions.
calculations indicate that the isolated C C bond is The high compressibility of scCO2,[9] especially in the
ˆ
Adv. Synth. Catal. 2004, 346, 459 466
DOI: 10.1002/adsc.200303160
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
459

M. Chatterjee et al.
FULL PAPERS
Scheme 1.
near critical region could result in a significant change Pt-Ru catalyst also show better visualization of the
in selectivity. Applications of scCO2 to hydrogenation metal particle after long-term exposure. Massive ag-
reactions involving heterogeneous catalysts[10 12] are glomerations of smaller particles were observed on the
particularly attractive. Furthermore, the careful design surface of the mesoporous support. These agglomerates
of solid catalysts to capitalize the unique solvent consisted of many small crystallites clustered together to
properties of scCO2 may afford greater advantages, form larger aggregates and were distributed randomly
resulting in enhanced activity and selectivity. Consid- on the surface of the support. Mainly larger particles are
ering our previous work[13] on the hydrogenation of seen on the surface of the bimetallic catalyst and they
cinnamaldehyde in scCO2 using a Pt-Ru bimetallic were in different shape.
catalyst, leading to the exclusive formation of unsatu-
To investigate the oxidation states of Pt and Ru, XPS
rated alcohol, we have now attempted to achieve the analyses were performed. Table 1 shows the binding
hydrogenation of citral using monometallic Pt and Pt- energies for the photoelectron peaks of Pt 4f7/2, 4f5/2,
Ru bimetallic catalysts supported on MCM-48 in Ru3d5/2, 3d3/2, C 1s and Si 2p peaks for Pt-Ru as well as Pt
scCO2.
catalysts. As shown in Table 1 it can be observed that, for
the monometallic catalyst, the bindingenergies of Pt 4f7/2
and 4f5/2 are 70.4 and 73.7, respectively, which are very
close to the value of Pt foil[14] (Pt 4f7/2 and 4f5/2 70.7 and
74.0 eV, respectively). Therefore, the presence of only
metallic Pt in the monometallic Pt-MCM-48 catalyst has
been suggested.
Results and Discussions
Catalyst Characterization
In the bimetallic catalyst, the spectrum can be fitted by
The XRD patterns of the mesoporous silica support and two pairs of overlapping Lorentzian curves. The Pt 4f7/2
the resulting materials either of the monometallic or the and 4f5/2 lines appearing at 71.0 and 74.3 eV are
bimetallic type clearly implied the structural order for attributed to the metallic Pt. The second pair appearing
the cubic crystallographic space group Ia3d with distinct as Pt 4f7/2 and 4f5/2 peaks with binding energies of 72.5
(2 1 1) and (2 2 0) reflections. The mesoporosity of the and 75.8, respectively, are very closely related to those of
material was shown to remain unaltered after the in situ Pt2‡. So, regarding the state of Pt, it can be inferred that
addition of Pt or Pt-Ru. For the monometallic and the bimetallic catalyst contains Pt2‡ as well as metallic
bimetallic catalysts, no line corresponding to Pt or the Pt. Analysis of Ru (3d5/2 and 3d3/2) is less straightforward
Pt-Ru alloy was observed.
as peaks occur together with the carbon peak (Table 1);
TEM photographs also directly reveal the intact hence, much care has been taken to avoid misinter-
mesoporosity of the support after the encapsulation of preting the results. The Ru 3d spectra for Pt-Ru
Pt and Pt-Ru. For the monometallic Pt-MCM-48 bimetallic catalyst exhibit a distinctive spin orbit dou-
catalyst, the images of the calcined material indicate a blet of metallic Ru (280.1 eV and 284.3 eV) along with
good Pt distribution represented by black dots and the the peak of oxidized Ru (282.5 eV and 286.6 eV,
particles were mainly spherical in shape. TEM images of respectively, Table 1).
460
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 459 466

Hydrogenation of Citral
FULL PAPERS
Table 1. XPS binding energies (B. E.) of monometallic Pt and bimetallic Ru-Pt catalysts.
Catalyst
Pt 4f7/2
Pt 4f5/2
Ru 3d5/2
Ru 3d3/2
Si 2p
C 1s
B. E. (eV)
B. E. (eV)
B. E. (eV)
B. E. (eV)
Pt-MCM-48
70.4
70.7
73.7
74.0
104.3
Pt-Foil[a]
Pt-Ru-MCM-48
70.8 (72.5)
71.6
74.3 (75.8)
74.6
280.1 (282.5)
282.9
284.3 (286.6)
287.2
104.3
104.3
284.4 and 285.7
284.6 and 286.0
Pt-Ru-MCM-48[b]
Data in the parenthesis indicate the higher valence state that was observed in the bimetallic catalysts
[a]
Ref.: J. F. Moulder, W. F. Sticklle, P. E. Sobol, K. D. Bomben, Handbook of X-ray Photoelectron Spectroscopy,
(Eds.: J. Chastain, R. C. King Jr.), Physical Electronics Inc., Minnesota, USA, 1995.
After on-stream operation.
[b]
Catalytic Activity
Monometallic Pt-MCM-48 Catalyst
Fig. 1a shows the influence of the CO2 pressure on the
performance of the monometallic Pt catalyst for the
hydrogenation of citral at 323 K. Under the studied
reaction conditions (reaction time: 2 h, PH : 4 MPa) the
2
hydrogenation reaction path from citral to unsaturated
alcohol (path I) is more active than citral to citronellal
(path II; Scheme 1). The selectivity to the unsaturated
alcohol increases from ꢀ 15% to ꢀ 90% with an
increase in pressure from 6 to 10 MPa, and then
decreases. From the previous work[15] on the selective
hydrogenation of cinnamaldehyde, it was found that Pt-
supported catalysts are selective to the hydrogenation of
ˆ
C O in scCO2 medium. Citral also shows a similar trend
in selectivity in the presence of the Pt catalyst. The
selectivity to unsaturated alcohol (geraniol ‡ nerol)
changes with CO2 pressure (or with CO2 density). For
instance, a change in pressure from 8 to 10 MPa
corresponds to density changes from 520.8 to
708.3 kg mÀ3.[16] Therefore, the CO2-pessure-dependent
ˆ
selectivity to the C O bond could be appropriately
related to the density of the medium. The real reason for
this effect may be a debatable issue but there is no doubt
that the pressure effect has a strong influence on the
selectivity. In addition, the presence of the metallic
platinum, inferred from the XPS study (as the surface
Figure 1. CO2 pressure vs. conversion and selectivity for the
hydrogenation of citral over (a) monometallic Pt-MCM-48
and (b) bimetallic Pt-Ru-MCM-48. Reaction conditions:
Tˆ 323 K; reaction time ˆ 2 h; PH ˆ 4 MPa; amount of cata-
lyst ˆ 0.1 g. (c) Ru/Pt atomic ratio 2vs. selectivity to citronellal.
ˆ
exposing metallic Pt is more favorable for C O adsorp-
ˆ
tion) favors the hydrogenation of C O followed by the
formation of unsaturated alcohol. No other by-product
(isopulegol, cyclization product) has been found under
the studied reaction conditions.
pressure (Fig. 1b). The addition of Ru to the Pt changes
the reaction path from I to II (Scheme 1) and the
conversion and selectivity increases with the Ru/Pt
atomic ratio and reaches a maximum at ꢀ 30% Ru
(Fig. 1c). Thus, it may be concluded that the Ru plays an
Bimetallic Pt-Ru Catalyst
Fig. 1b shows the results of the hydrogenation of citral important role in the formation of the partially saturated
using bimetallic Pt-Ru catalyst in scCO2 medium at aldehyde. From the XPS study (Table 1), it was found
323 K for 2 h. The addition of Ru significantly changed that the Pt-Ru bimetallic catalyst contains metallic Ru
the product selectivity to the partially saturated alde- along with Pt2‡. Furthermore, this result is in contrast to
hyde (citronellal). Aselectivity of 80% to the citronellal
our earlier observation on the hydrogenation of cinna-
has been achieved in 90% conversion at 12 MPa of CO2 maldehyde, [13], and so it can be suggested that citral may
Adv. Synth. Catal. 2004, 346, 459 466
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
461

M. Chatterjee et al.
FULL PAPERS
be adsorbed on the metal surface in such a manner that
ˆ
the hydrogenation of the C C bond occurs smoothly
ˆ
since it is well known that hydrogenation of C C is
ˆ
kinetically favored over C O.
For the bimetallic Pt-Ru catalyst, the conversion and
the selectivity to citronellal rises from ꢀ 8% to 80% and
ꢀ 70% to ꢀ 90%, respectively, with an increase in CO2
pressure from 6 to 12 MPa, reaching a maximum at
12 MPa (Fig. 1b). It should be mentioned that for both
catalysts the corresponding product selectivity passes
through a maximum. This phenomenon could be
associated with tuning of the average distance between
the solute molecules and the transition state, accom-
panied with the density, varies with pressure. That is, for
a compressed liquid such as supercritical fluid, variation
in density causes a change in chemical equilibrium and
affects the activity and selectivity of a reaction.[17]
Figs. 2a and 2b, respectively, represent the variation of
conversion and product distribution with time for the
monometallic Pt and the bimetallic Pt-Ru catalysts at
323 K in the hydrogenation of citral. It can be seen that
the rate of reaction is faster at short reaction times and
almost a steady state has been reached within a period of
2 h of reaction. The conversion of citral is higher for the
Pt-Ru catalyst compared to the monometallic Pt catalyst
at the same reaction time. Fig. 2a also reveals that for the
Pt catalyst the reaction proceeds through the reaction
path I but for the bimetallic catalyst the path changes to
Figure 2. Variation of conversion and selectivity with time in
the hydrogenation of citral over (a) monometallic Pt-MCM-
48 and (b) bimetallic Pt-Ru-MCM-48. Reaction conditions:
Tˆ 323 K; PH ˆ 4 MPa; PCO ˆ 10 MPa for monometallic
2
2
catalyst; PCO ˆ 12 MPa for bimetallic catalyst; amount of
catalyst ˆ 0.12g in each case.
II (Scheme 1). Hence, in scCO2, the Pt catalyst favors heterogeneous phase but not in the homogeneous phase
ˆ
ˆ
the hydrogenation of the C O bond instead of the C C above 12 MPa. Although the exact reason is not clear,
bond.
the pressure-dependence of the appearance of the
maximum peaks (Fig. 1a and Fig. 1b) may lead to
some direct interaction between the metal particles
and the medium; addition of Ru to Pt changes the nature
of the metal and the types of interaction with the
Phase Behavior Studies
The phase behavior of the reaction mixture is of major medium.[16] To get some primordial insight about the
concern for the design and understanding of the type of interaction between the medium and the metal,
catalytic systems operating in scCO2. From naked eye XPS and FTIR have been done after on-stream
observations, it has been found that the solubility of operation of the catalyst. Adistinctive spin orbit doublet
citral increases with increasing CO2 pressure, and a of oxidized Ru at 282.9 eV (3d5/2) and 287.2 eV (3d3/2),
homogeneous system was attained above 11 MPa of respectively (Table 1), indicates a significant change in
CO2 and 4 MPa of hydrogen. Apronounced effect of
the oxidation state of Ru. These binding energy values
phase behavior was observed on the catalytic perform- are in between those of the two oxidation states of RuO3
ance during hydrogenation of citral. For both catalysts, (282.5) and RuO4 (283.3). Hence it may be suggested
at the lower CO2 pressures below 7 MPa the conversion that CO2 could be dissociated on the metal particles and
is lower due to the mass-transfer limitation in the liquid produces some oxygen fragments. The oxygen frag-
phase, but with increasing pressure the increase in the ments may be partially replaced by surface hydrogen as
reactant concentration in scCO2 results in higher con- observed in the primary data obtained from FTIR. This
version. Consequently, around 17 MPa of CO2 pressure, could be a reason but further studies are required to
additional amounts of CO2 lower the reactant concen- clarify the actual surface species during the reaction,
tration and so reduce the conversion (Figs. 1a and 1b). which is beyond the scope of this article.
Regarding the selectivity, for the bimetallic catalyst a
maximum in the citronellal selectivity corresponds well
to the appearance of the homogeneous phase around Effect of Hydrogen Pressure
12 MPa, whereas for the monometallic Pt catalyst a
maximum in the selectivity to the unsaturated alcohol We have further examined the effect of hydrogen
(geraniol ‡ nerol) was observed around 10 MPa in the pressure on the conversion and selectivity for the
462
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 459 466

Hydrogenation of Citral
FULL PAPERS
hydrogenation of citral. Figures 3a and 3b indicate an Effect of Temperature
increase in conversion with the hydrogen pressure for
the monometallic and bimetallic catalysts, respectively. Table 2 shows that the product distribution for the citral
The high solubility of hydrogen in scCO2 compared to hydrogenation is strongly dependent on the reaction
that in conventional organic solvents is responsible for temperature for the Pt-Ru catalyst. The reactions were
the increase in conversion. On the other hand, Figure 3b conducted at 308, 323 and 343 K at 10 and 12 MPa of
indicates a change in selectivity of unsaturated alcohol CO2 with the constant hydrogen partial pressure of
(geraniol ‡ nerol) from 5 to 33.4% with H2 pressure 4 MPa using monometallic Pt (Entries 1 3) and bimet-
from 2 to 6 MPa along with the increasing conversion. allic Pt-Ru catalysts (Entries 4 6). With an increase in
As the hydrogenation of citral consists of a series of temperature, the selectivity to the unsaturated alcohol
reaction pathways, the change in hydrogen pressure (geraniol ‡ nerol) was increased from 0 to 23.4% for the
could affect the individual rate constants due to the bimetallic catalysts; however, for the monometallic
difference in activation volume, and hence the increased catalyst it remained constant (Table 2). Either the
pressure may favor one reaction over others, resulting in physical properties of scCO2 such as density vary widely
an improved selectivity for the unsaturated alcohol with change in temperature or the effect of temperature
(geraniol ‡ nerol) in scCO2.
on the reaction network could change the product
selectivity.
Comparison with Organic Solvents
The advantage of using scCO2 as a reaction medium in
the hydrogenation is further exemplified by comparing
the conversion and selectivity in organic solvents and
under solvent-less conditions. As shown in Figure 4, in
the organic solvent, the monometallic Pt catalyst was
found to be selective to citronellal, whereas unsaturated
alcohol (geraniol ‡ nerol) is the main product in scCO2
(Figure 1a). This differentiated behavior of the mono-
metallic catalyst in scCO2 is due to the difference in the
availability of hydrogen. In the liquid-phase hydro-
genation the SiO2-supported catalyst exhibits first-order
dependences on hydrogen pressure.[18] In the scCO2
medium the better availability of hydrogen leads to
faster and more consecutive hydrogenation of the
unsaturated aldehyde to the unsaturated alcohol.
Figure 3. Influence of hydrogen pressure on conversion and
selectivity in the hydrogenation of citral over (a) monome-
tallic Pt-MCM-48 and (b) bimetallic Pt-Ru-MCM-48. Reac-
For the bimetallic catalysts, although citronellal was
the main product for scCO2 and organic solvents, but
when compared to scCO2, the selectivity has been
decreased significantly. Figure 4 also reveals that, in the
solvent-less conditions, the conversion is 1.3% and 1.8%
tion conditions: Tˆ 323 K; reaction time ˆ 2 h; PCO ˆ 10 MPa
2
for monometallic catalyst; PCO ˆ 12 MPa for bimetallic
2
catalyst; amount of catalyst ˆ 0.1 g in each case.
Table 2. Effect of temperature on the hydrogenation of citral over Pt-MCM-48 and Pt-Ru-MCM-48 catalyst under scCO2
medium.[a]
Entry
Catalyst
Temp. (K)
(%) Conv.
Selectivity (%)
Citronellal
Citronellol
Geraniol ‡ Nerol
1
2
3
4
5
6
Pt/MCM-48[b]
308
323
343
308
323
343
32.7
71.3
75.8
21.4
89.8
91.2
23.2
2.1
1.4
68.7
69.8
66.6
22.8
7.5
10.1
31.2
28.3
10.0
54.0
90.4
88.5
0.1
1.9
23.4
Pt -Ru/MCM-48[c]
[a]
Reaction conditions: citral, 6.5 mmol; catalyst, 0.1 g; H2 ˆ 4 MPa; reaction time ˆ 2 h.
[b]
[c]
CO2: 10 MPa.
CO2: 12 MPa.
Adv. Synth. Catal. 2004, 346, 459 466
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
463

M. Chatterjee et al.
FULL PAPERS
Table 3. Catalyst deactivation studies during the hydrogena-
tion of citral.[a]
Conversion
[%][b]
Conversion
[%][c]
Fresh
89.8
89.3
89.3
89.2
71.3
71.3
71.0
71.1
Recycle 2
Recycle 3
Recycle 4
[a]
Reaction conditions: citral, 6.5 mmol; catalyst, 0.1 g; H2,
4 MPa; reaction time, 2 h.
Monometallic, 10 MPa of CO2.
Figure 4. Effect of organic solvent in selectivity and con-
version in the hydrogenation of citral. Reaction conditions:
Tˆ 323 K; reaction time ˆ 2 h; amount of solvent ˆ 10 mL;
[b]
[c]
Bimetallic, 12 MPa of CO2.
amount of catalyst ˆ 0.1 g; PH ˆ 4 MPa.
2
for the mono- and bimetallic catalysts, respectively. The and a maximum in the selectivity was found at the
remarkable reduction in conversion and selectivity can homogeneous conditions. We speculate that this specific
be attributed to the greater mass-transfer resistances behavior could be due to the modification of the metal×s
existing in the liquid phase and solvent-less conditions surface upon interaction with the medium.[16] Further
relative to the scCO2 medium. Therefore, the enhance- research should be required to elucidate the actual
ment of conversion and selectivity for the above- reason. The deactivation of catalyst as well as the
mentioned reaction must be an advantage of scCO2 formation of several kinds of by-products that usually
medium.
occur in organic solvents can be avoided by using scCO2
as reaction medium. Hence, from the practical view-
point, scCO2 as reaction medium can provide several
advantages over organic solvents. In addition to that, the
solubility of hydrogen also plays a key role in the rate
Deactivation of Catalysts
Deactivation of the catalyst is a commonly occurring enhancement (phase behavior). So, scCO2 cannot be
problem for the hydrogenation of citral in organic considered merely as a simple ™substitute solvent∫ to the
solvents, which is mainly associated with the decarbon- conventional organic solvents. The simple catalyst
ylation reaction.[5,19 21] This is due to the accumulation of methodology presented here could be highly relevant
inhibiting carbonaceous species on the catalyst sur- for the further development of clean chemical processes
face.[22,23] To check the probability of leaching of Ru, we based on heterogeneous catalysts in scCO2 medium.
used the liquid product as the catalyst but it was inactive.
Moreover, in scCO2 hydrogenation of citral can be
carried out for a longer time and also for successive runs
without facing any problems of deactivation such as a
Experimental Section
change in conversion and selectivity (Table 3). The
reason could be a limitation of the inhibiting species in
scCO2 as an effect of the extraction power of scCO2,
which prevented the deposition of the inhibiting mate-
rials over the catalyst surface.[24] Therefore, this result
highly implied the possible advantage of supercritical
medium in heterogeneous catalysis over conventional
processes for the hydrogenation of citral.
Catalyst Synthesis
The bimetallic catalyst was prepared by an in situ addition of
1 wt % solution of Ru and Pt salt in a mixture of surfactant and
silica precursor. The sources of silicon and Pt are tetraethyl
orthosilicate (95%, TEOS) (Wako, Japan) and chloroplatinic
acid (98.5%, Aldrich), respectively. As the Ru source, ruthe-
nium acetylacetonate (97%, Aldrich) was chosen to prevent
acetal formation.[25]
.
Cetyltrimethylammonium bromide
(99%, Merck) was used as template to obtain the mesoporous
structure. Under stirring conditions, the solutions of different
Ru /Pt ratios followed by the TEOS (5.0 g) were added to the
starting gel, containing template (2.39 g) and sodium hydrox-
ide (0.33 g) in deionized water (27.0 g). The stirring was
continued for 2 h. Finally, the resultant gel was autoclaved at
413 K for 48 h.0 The molar composition of the gel was: 1TEOS:
0.27 CTABr: 0.34 Na2O: 62.5 H2O. The resultant material was
filtered, dried and then calcined at 823 K for 10 h in air. The
monometallic catalysts were also prepared in the same way as
Conclusions
It was demonstrated that scCO2 is an attractive medium
to carry out the selective hydrogenation of citral. Simple
tuning ofthe reactioncondition leads tothe formation of
the desired product. The monometallic Pt catalyst was
found to predominantly form the unsaturated alcohol
(geraniol ‡ nerol), whereas the addition of Ru to the Pt
considerably changes the selectivity to the citronellal described above. The total amounts of metal loading were
464
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 459 466

Hydrogenation of Citral
FULL PAPERS
ꢀ 3.5% and 1% for bimetallic and monometallic catalysts,
Phase Behavior Studies
respectively.
The phase behavior of the system under the reaction condition
was observed at 323 Kusing a 10 mL high-pressure viewing cell
containing citral, CO2 and H2. The same pressure controller as
used for the hydrogenation experiment was used to regulate
the pressure inside the viewing cell. Citral (0.2 g) was
introduced into the viewing cell at a constant hydrogen
pressure of 4 MPa and the CO2 pressure was in the range of
7.0 to 17.0 MPa. Amagnetic stirrer stirred the content inside
the viewing cell.
Catalyst Characterization Techniques
All the catalysts were primarily characterized by X-ray
diffraction (XRD), transmission electron microscopy (TEM)
and X-ray photoelectron spectroscopy (XPS). The X-ray
diffraction pattern was recorded on a Rigaku-RAD-X system
using monochromatized Cu Ka radiation (l ˆ 1.542 ä). In
general, the diffraction data were collected using the contin-
uous scan mode with a scan speed of 2 deg/min over the scan
range 2q ˆ 2 158. Transmission electron microscopy (TEM)
of the calcined material was performed on a JEOL JEM-2000
EX (II) operating at 200 kV. Calcined samples were deposited
on a grid with a holey carbon copper film. The powder was
suspended in ethanol by the ultrasonic method and a drop of
this solution was placed on the support film. X-ray photo-
electron spectra (XPS) were obtained with an Ulvac Phi 5601
spectrometer using Mg Ka (1253.6 eV) radiation of twin
anodes in the constant analyzer mode with a pass energy of
50 eV. The X-ray power was limited to 14.0 kVand 400 W. The
binding energy was determined by referencing to the Si 2p
binding energy of 104.3 eV. The spectrum was fitted to the
Gaussian spectra of Pt 4f7/2, Pt 4f5/2 and Ru3d3/2, Ru 3d5/2 by
optimizing the peak heights and positions assuming the fixed
peak-to-peak difference at 3.30 eV.
References
[1] K. Bauer, D. Garbe, Common Fragrances and Flavor
Materials, VCH, Weinheim, 1985.
[2] a) U. K. Singh, M. A. Vannice, J. Catal. 2001, 199, 73 84;
b) P. Reyes, H. Rojas, G. Pecchi, J. L. G. Fierro, J. Mol.
Catal. A 2002, 179, 293 299; c) C. Milone, M. L.
Tropeano, G. Gulino, G. Neri, R. Ingoglia, S. Galvagno,
Chem. Commun. 2002, 868 869.
[3] G. Lafaye, C. Micheaud-Especel, C. Montassier, P.
Marecol, Appl. Catal. A 2002, 230, 19 30.
[4] A. M. Silva, O. A. A. Santos, M. J. Mendes, E. Joradao,
M. A. Fraga, Appl. Catal. A 2003, 241, 155 164.
¬
[5] M. A. Aramendia, V. Borau, C. Jimenez, J. M. Marinas,
J. R. Ruiz, F. J. Urbano, J. Mol. Catal. A 2001, 171, 153
158.
[6] U. K. Singh, M. A. Vannice, J. Catal. 2000, 191, 165 180.
[7] A. Wismeijer, A. Kieboom, H. van Bekkum, Appl. Catal.
1986, 25, 181.
[8] a) P. G. Jessop, T. Ikariya, R. Noyori, Nature 1994, 368,
231 233; b) P. G. Jessop, Y. Hsiano, T. Ikariya, R.
Noyori, J. Am. Chem. Soc. 1996, 118, 344 355.
[9] A. F¸rstner, D. Koch, K. Langemann, W. Leitner, C. Six,
Angew. Chem. Int. Ed. 1997, 36, 2466 2469.
[10] a) A. Baiker, Chem. Rev. 1999, 99, 453 473. b) M. G.
Hitzler, F. R. Smail, S. K. Ross, M. Poliakoff, Org.
Process Res. Dev. 1998, 2, 137 146.
[11] G. I. Childs, A. I. Cooper, T. F. Nolan, M. J. Carrott,
M. W. George, M. Poliakoff, J. Am. Chem. Soc. 2001,
123, 6857 6866.
[12] U. R. Pillai, E. Sahle-Demessie, Ind. Eng. Chem. Res.
2003, 42, 6688 6696.
General Procedure for the Hydrogenation of Citral in
scCO2
The hydrogenation of citral was carried out in a stainless steel
batch reactor (50 mL). 0.1 g of catalyst and the reactant
(6.5 mmol) were loaded into the reactor. Then the reactor was
sealed and flushed two times with 2 MPa of CO2 to remove the
air. After flushing, the reactor was maintained at a reaction
temperature of 323 K. Aprescribed amount of hydrogen was
first loaded into the reactor. Liquid CO2 was charged into the
reactor using a high-pressure liquid pump, and then com-
pressed to the desired pressure. Pressure was kept constant by a
back-pressure regulator. The hydrogen and the reaction
mixture were stirred continuously with a Teflon-coated
magnetic bar during the reaction. After the reaction, the
reactor was cooled in ice/water and depressurized carefully by
the backpressure regulator. The liquid mixture was identified
by GC/MS and analyzed quantitatively by GC (HP 5890)
equipped with a flame ionization detector. Quantification of
the products was obtained by a multipoint calibration curve for
each product. For recycle studies the catalyst was separated
from the reactant and product then recharged with fresh
reactant. The hydrogenation in organic solvents has also been
performed by the above method using conventional organic
solvents (10 mL) instead of CO2. The selectivity to each
product was calculated by the following expression
Si ˆ Ci/SCp, where Ci is the concentration of the product −i×
and SCp is the total concentration of the product.
[13] M. Chatterjee, Y. Ikushima, FY. Zhao, New Journal of
Chemistry 2003, 27, 510 513.
[14] a) J. F. Moulder, W. F. Sticklle, P. E. Sobol, K. D. Bomb-
en, Handbook of X-ray Photoelectron Spectroscopy,
(Eds.: J. Chastain, R. C. King Jr.), Physical Electronics
Inc., Minnesota, USA, 1995; b) K. S. Kim, N. Winograd,
R. E. Davis, J. Am. Chem. Soc. 1971, 93, 6296 6297.
[15] a) M. Chatterjee, Y. Ikushima, FY. Zhao, Catal. Lett.
2002, 82, 141 144; b) B. M. Bhanage; Y. Ikushima; M.
Shirai, M. Arai, Catal. Lett. 1999, 62, 175 177; c) F.
Zhao, Y. Ikushima, M. Shirai, T. Ebina, M. Arai, J. Mol.
Catal. 2002, 180, 259 265.
[16] M. Arai, Y. Nishiyama, Y. Ikushima, J. Supercrit. Fluids
1998, 13, 149 153.
Adv. Synth. Catal. 2004, 346, 459 466
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
465

M. Chatterjee et al.
FULL PAPERS
[17] A. A. Clifford, K. Pople, W. J. Gaskill, K. D. Bartle,
[22] R. Shekhar, M. A. Barteau, Surf. Sci. 1994, 319, 298 314.
[23] P. Gallezot, A. G. Fendler, D. Richard, Chemical Indus-
try (Catalysis of Organic reactions), Vol. 47, 1992.
[24] a) H. Tiltscher, H. Hofmann, Chem. Eng. Sci. 1987, 42,
959 977; b) S. Saim, D. M. Ginosur, B. Subramanium,
ACS Symp. Ser. 1989, 406, 301.
C. M. Rayner, J. Chem. Soc. Faraday Trans. 1998, 94,
1451 1456.
[18] U. K. Singh, M. A. Vannice, J. Mol. Catal. 2000, 163,
233 250.
[19] C. J. Houtman, M. A. Barteau, J. Catal. 1991, 130, 528
546.
[25] G. Neri, L. Mercadante, A. Donato, A. M. Visco, S.
Galvagno, Catal. Lett. 1994, 29, 379 386.
[20] J. L. Davis, M. A. Barteau, J. Am. Chem. Soc. 1989, 111,
1782 1792.
[21] A. Wagray, D. G. Blackmond, J. Phys. Chem. 1993, 97,
6002 6006.
466
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 459 466