Hydrogenation of citral over carbon supported iridium catalysts

Article abstract of DOI:10.1007/s10562-012-0809-1

Hydrogenation of citral (3,7-dimethyl-2,6-octadienal) was carried out in a batch reactor at atmospheric hydrogen pressure and 70 °C over Sibunit (mesoporous carbon) supported iridium catalysts with different metal loadings. The major products were isopropyl ethers of geraniol and nerol (2-propanol was used as solvent), but also geraniol, nerol, citronellal di-isopropyl acetal and citronellal were formed. The metal particle size was determined by CO-chemisorption and varied from 2 to 6 nm. The turnover frequency for hydrogenation reactions increased with the metal particle size.

Full text of DOI:10.1007/s10562-012-0809-1

Catal Lett (2012) 142:690–697
DOI 10.1007/s10562-012-0809-1
Hydrogenation of Citral Over Carbon Supported Iridium
Catalysts
•
Heidi Bernas Irina Simakova Igor P. Prosvirin
•
•
•
Paivi Maki-Arvela Reko Leino Dmitry Yu. Murzin
•
¨
¨
Received: 27 January 2012 / Accepted: 19 March 2012 / Published online: 5 April 2012
Ó Springer Science+Business Media, LLC 2012
Abstract Hydrogenation of citral (3,7-dimethyl-2,6-oct-
adienal) was carried out in a batch reactor at atmospheric
hydrogen pressure and 70 °C over Sibunit (mesoporous
carbon) supported iridium catalysts with different metal
loadings. The major products were isopropyl ethers of
geraniol and nerol (2-propanol was used as solvent), but
also geraniol, nerol, citronellal di-isopropyl acetal and
citronellal were formed. The metal particle size was
determined by CO-chemisorption and varied from 2 to
6 nm. The turnover frequency for hydrogenation reactions
increased with the metal particle size.
citronellal, geraniol, nerol, and citronellol. Hydrogenation
of citral has been studied using Ru [4–9], Pt [7, 10–15], Pd
[7, 16, 17], Rh [6, 7, 11], Au [18], Ni [6, 7, 19, 20], Ir [7,
21–23], Co [7], Os [7], Ag [24] and bimetallic [12, 21, 25–
29] catalysts. Citral contains three double bonds of which
two are conjugated (C=O and C=C) and the third is iso-
lated. Hydrogenation of the C=C bond is thermodynami-
cally more favourable than the C=O bond making the
selective hydrogenation of the C=O bond challenging and a
topic of continuous research. Selective hydrogenation of
the C=O bond can be achieved using homogeneous cata-
lysts but heterogeneous catalysts are more environmentally
friendly and easier to separate and re-use than their
homogeneous counterparts [1]. Different factors influence
chemoselectivities in selective hydrogenations, such as the
metal type, metal particle size and morphology, the use of
second metal, electronic and steric influence of the support,
metal-support interactions with partially reducible sup-
ports, steric effects of the substituents at the conjugated
double bond, reaction conditions, and use of additives or
promoters (Lewis acids or bases) [24].
Keywords Catalysis Á Hydrogenation Á Citral Á Iridium
1 Introduction
Hydrogenation of a,b-unsaturated aldehydes is an impor-
tant reaction, which has recently been reviewed [1–3].
Citral (3,7-dimethyl-2,6-octadienal) belongs to this class of
compounds and is also valuable in industrial processes for
the synthesis of fragrances, flavours and pharmaceuticals.
Hydrogenation of citral gives valuable products, including
In this work, iridium catalysts supported by Sibunit were
used in the hydrogenation of citral in a batch reactor under
atmospheric hydrogen pressure. Sibunit is a class of porous
carbon–carbon composite materials combining advantages
of graphite (chemical stability and electric conductivity)
and active carbons (high specific surface area and adsorp-
tion capacity) [30, 31]. Sibunit is produced in the form of
powders with particle sizes of 5–200 lm, spherical gran-
ules of 0.2–4 mm diameter, extrudes in the shape of cyl-
inders, tubes, trefoils, and quatrefoils, and blocks of
honeycomb structure. The aim was to hydrogenate citral
selectively to either citronellal or geraniol and nerol.
Savchenko et al. [23] reported 100 % yield of geraniol
using catalysts supported on activated carbon with iridium
¨
H. Bernas Á I. Simakova Á P. Maki-Arvela Á D. Yu. Murzin (&)
Laboratory of Industrial Chemistry and Reaction Engineering,
˚
Process Chemistry Centre, Abo Akademi University,
˚
Biskopsgatan 8, FI-20500 Turku/Abo, Finland
e-mail: Dmitry.Murzin@abo.fi
I. Simakova Á I. P. Prosvirin
Boreskov Institute of Catalysis, Novosibirsk, Russia 630090
R. Leino
Laboratory of Organic Chemistry, Abo Akademi University,
˚
˚
Biskopsgatan 8, FI-20500 Turku/Abo, Finland
123

Hydrogenation of Citral
691
content of 4.7 and 4.9 wt% and boron content of 0.08 and
0.1 wt% (no reaction conditions were given). Singh and
Vannice [7] used Ir/SiO2 and obtained 52 % selectivity to
citronellal, but no geraniol and nerol at 27 °C, atmospheric
hydrogen pressure and hexane as a solvent. Reyes et al.
[21] used TiO2 and SiO2 supported monometallic and
bimetallic iridium catalysts reduced at low (200 °C) and
high temperatures (500 °C) at 67 °C, 4.1 bar hydrogen
pressure with ethanol as a solvent. It was observed that the
selectivity to nerol and geraniol varied with the catalyst
reduction temperature. The selectivity to geraniol was
100 % at 37 % conversion for Ir–Ge/TiO2 reduced at high
temperature and the selectivity to nerol was 100 % at 3 %
conversion for Ir/TiO2 reduced at low temperature. The
selectivity to citronellal was zero for most of the catalysts
and at highest 12 % at 14 % conversion (Ir–Ge/TiO2
reduced at low temperature).
with the aqueous solution of Na2CO3 was washed, dried
overnight at 120 °C and reduced by hydrogen at tempera-
ture ramp 2 °C/min up to 350 °C.
Iridium catalyst (Ir/C-2) was prepared by incipient
wetness impregnation where the precursor solution was
poured down onto the support by drops in a single step
using the appropriate concentration to obtain required
metal content. The sample was kept sealed at room tem-
perature overnight, then dried and reduced by hydrogen
similarly to Ir/C-3 sample.
2.2 Catalyst Characterization
To determine the metal particle size and dispersion, sam-
ples were analysed by CO pulse chemisorption (Autochem
2910, Micromeritics) where 0.1–0.3 g of the sample was
placed into a quartz sample tube containing silica wool.
The tube was inserted to the system and the sample was
treated in a stream of helium (AGA) gas at 50 °C for
30 min. After that, the sample was reduced at 400 °C for
1 h in hydrogen (AGA) using helium as a carrier gas, after
which it was flushed with helium (still at 400 °C) for 1 h,
cooled to room temperature, placed on water bath, and
subsequently the CO pulses were introduced (10 % CO in
helium, helium as a carrier gas) until adsorption was
complete. The dispersion was calculated from the amount
of CO consumed, assuming the CO:Ir stoichiometry to be
unity [32].
2 Experimental
2.1 Catalyst Synthesis
Mesoporous graphite-like carbon material of the Sibunit
family was used as a support material to prepare the irid-
ium catalysts. The preparation procedures and properties of
Sibunit carbons are described elsewhere [30]. Textural
characteristics of the mesoporous graphite-like carbon
material Sibunit (fr. 40–100 lm, SBET = 350 m2/g, total
pore volume 0.6 cm3/g, mean pore diameter 7 nm) were
obtained from the data on nitrogen adsorption measured at
77 K with an automatic volumetric device ASAP 2400
(Micrometritics).
Quantitative analysis of iridium concentration was per-
formed using wavelength dispersive X-ray fluorescence
(WDXRF) spectrometry with pressed powder pellet
method. Test samples (0.1 g) were diluted with cellulose
(0.9 g). Resulting samples were ground in a ball mill for
30 min and pelletized in 20 mm diameter die using hand-
press. Calibration samples with iridium content from 0.5 to
3 % were prepared utilizing the same technique by mixing
an appropriate amount of cellulose with iridium containing
substance with known iridium content. The intensities of
iridium La1 lines in samples were measured on an ARL
Advant’X spectrometer equipped with a rhodium anode
X-ray tube. Excitation conditions were as follows: tube
voltage of 50 kV; current of 40 mA; collimator with a
divergence of 0.25°; LiF200 crystal was used as a mono-
chromator; scintillation counter was used as a detector;
counting time was 12 s.
Prior to use, the support was washed to remove any
impurities, first by boiling Sibunit with distilled water to
remove suspended coal dust, then washing with HCl (2 M)
under reflux for 4 h, and finally washed with distilled water
and dried at 120 °C. The iridium catalysts used in the
experiments were prepared using solutions of reagent grade
IrCl4ÁnH2O in HCl (0.5 M) (Ir content 52.9 %, JSC ‘‘The
Gulidov Krasnoyarsk non-ferrous metals plant’’) by two
different procedures: incipient wetness impregnation (IWI)
and deposition–precipitation (DP). Iridium catalysts (Ir/C-1,
Ir/C-3) were prepared by DP using the following proce-
dure. To prepare Ir/C-1 catalyst sample the carbon support
was dispersed in water. An appropriate amount of an
aqueous solution of H2IrCl6 was added dropwise under
moderate stirring, followed by addition of Na2CO3 (1 M)
aqueous solution, reduction by NaBH4, separation by fil-
tration, washing with water until the pH of the rinsing
waters reached a value of 6, and finally dried overnight at
120 °C. The procedure of Ir/C-3 sample preparation was
similar to Ir/C-1 except that catalyst sample after mixing
The SEM analysis was performed with a LEO Gemini
1530 with a Thermo Scientific UltraDry Silicon Drift
Detector (SDD), equipped with SE (secondary electron),
BSE backscattered electron) detectors, as well as an
In-Lens detector.
X-ray photoelectron spectra (XPS) were recorded using
SPECS spectrometer with PHOIBOS-150 hemispherical
energy analyzer and MgKa X-ray irradiation (hm = 1253.6 eV,
123

692
H. Bernas et al.
100 W). Binding energy scale was preliminarily calibrated
by the position of the peaks of Au4f7/2 (84.00 eV) and
Cu2p3/2 (932.67 eV) core levels. For spectra recording the
samples were supported to the conductive copper scotch
tape. The binding energy of peaks was corrected to take
into account the sample charging by referencing to the C1s
(284.5 eV) (internal standard). The ratio of surface atomic
concentrations of the elements was calculated from the
integral intensities of photoelectron peaks corrected by
corresponding atomic sensitivity factors (ASF) [33]. In
addition to the survey photoelectron spectra, more narrow
spectral regions C1s, Ir4f, Cl2p, and O1s have been
recorded. For the survey spectra the pass energy of the
analyzer was 50 eV, while for the narrow spectral regions
the pass energy was 20 eV.
results are presented in Table 1. Ir/C-1 had 0.55 wt%
iridium (will be denoted as 0.55 wt% Ir/Sibunit hereon), Ir/
C-2 had 2.97 wt% iridium (denoted 3 wt% Ir/Sibunit), and
Ir/C-3 had 0.99 wt% (denoted 1 wt% Ir/Sibunit).
The catalysts were characterized by CO chemisorptions
and SEM-EDX. The results from CO chemisorptions are
presented in Table 2. For the Sibunit supported catalysts
the metal dispersion varied from 18 % for 1 wt% Ir/Sibunit
to 37 and 53 % for 0.55 and 3 wt% Ir/Sibunit, respectively.
Figure 1 shows the results from SEM, confirming that the
catalyst particles are spherical. EDX showed that no other
metals than iridium were present.
The XP spectra of 3 wt% Ir/Sibunit are shown in Fig. 2.
They were recorded after (1) the catalyst reduction has
been performed at 300 °C, (2) heating at 500 °C in vacuum
for 1 h, (3) heating at 500 °C in O2 for 1 h. One should
note that the intensities of Ir4f were normalized to the
intensity of C1s level of Sibunit support. The value of
binding energy of Ir4f level in the sample treated with H2 is
61.5 eV, while heating of the sample in vacuum shifts this
line to 61.2 eV, which is typical for Ir0 in the bulk
(BE = 60.9 eV) [34]. One the contrary sample heating in
O2 shifts the peak to higher values of BE (62.0 eV) char-
acteristic for Ir in the oxide. It can be concluded therefore
that in the initial 3 % Ir/Sibunit highly dispersed species
have been formed with Ir in the metallic state. Heating of
the sample in vacuum results in sintering which is evi-
denced by a decrease of Ir/C ratio (Table 3) and a sub-
sequent shift of Ir 4f level towards value characteristic for
bulk metal.
2.3 Experimental Procedure
The experiments were performed in a stainless steel
shaking reactor with a shaking frequency of 4 double
movements per second. Citral and 2-propanol were from
Sigma-Aldrich ([99.8 %) and hydrogen from AGA
(99.999 %). The reactant concentration was 0.02 M and
the amount of catalyst in the experiments was 0.2 g
(\40 lm). The iridium catalysts were reduced ex-situ at
400 °C (heating rate 5 °C/min) for 1 h in hydrogen flow.
The catalyst was pre-treated additionally in situ at 70 °C in
70 ml solvent for 45 min in hydrogen flow. The experi-
ments were conducted at 70 °C under atmospheric hydro-
gen pressure. The reactant in 30 ml solvent was injected to
a bubbling unit and hydrogen was bubbled through for
5 min. The reactant solution was led to the reactor, the
reaction time was set to zero and the reaction was started.
Samples were taken at different time intervals and ana-
lyzed by a gas chromatograph (GC). The samples were
prepared for GC using 0.5 ml of the sample taken from the
reactor and to this 0.5 ml of internal standard, consisting of
0.02 M cyclohexanone in cyclohexane, was added. An ali-
quot of 1 ll of the sample was injected with an autosampler
to the GC. The injector temperature was 220 °C and the split
ratio 50:1. A Supelco B-Dex-225 column was used and the
helium gas flow rate was 0.9 ml/min and the temperature
program 78 °C for 85 min, 1 °C/min to 90 °C and 10 °C/min
to 165 °C for 10 min. The flame ionization detector (FID)
was kept at 300 °C. The products were identified by GC–MS.
The Ir 4f spectra for 0.55 wt% Ir/Sibunit, 1 wt% Ir/
Sibunit, and 3 wt% Ir/Sibunit are displayed in Fig. 3. Two
of these catalysts, namely 0.55 wt% Ir/Sibunit and 3 wt%
Ir/Sibunit contain predominately Ir(0) with a small amount
of Ir(III). Contrary to it 1 wt% Ir/Sibunit contains mainly
Ir(III) with minor quantities of Ir(0), indicating that during
Table 1 Ir content in catalysts samples determined by XFA
Sample
Ir content (%)
Ir/C-1
Ir/C-2
Ir/C-3
0.55
2.97
0.99
Table 2 Catalyst characterization data
3 Results and Discussion
Catalyst
Ir dispersion (%) Ir particle size (nm)
0.55 wt% Ir/Sibunit
(Ir/C-1)
37
3
3.1 Catalyst Characterization
1 wt% Ir/Sibunit (Ir/C-3) 18
3 wt% Ir/Sibunit (Ir/C-2) 53
6
2
The Ir contents of Ir/C-1, Ir/C-2, Ir/C-3 catalysts were
measured using X-ray fluorescence (XRF) technique. The
123

Hydrogenation of Citral
693
The intensities of Ir4f line in 0.55 and 3 wt% Ir/Sibunit
catalysts are concomitant indicating similar content of Ir
species in the same electronic state [Ir (0) and Ir(III)]. As
expected, significant differences in the Ir/C atomic con-
centration ratios were detected for 0.55 and 3 wt% Ir/
Sibunit contrary to 1 wt% Ir/Sibunit (Table 3). In the
1 wt% catalyst, Ir/C ratio on the surface is lower than that
in the bulk as compared with 0.55 wt% catalyst suggesting
a lower dispersion of the Ir particles in agreement with CO
chemisorption results. At higher Ir loading (3 wt% pre-
pared by impregnation) this Ir/C ratio remains almost the
same as for Ir 0.55 wt% indicating approximately the same
Ir dispersion as well as predominate core–shell distribution
of Ir in 0.55 wt% Ir/Sibunit with lower bulk Ir concentra-
tion. It is worth to note that the surface Ir/C ratio is inde-
pendent on the preparation method.
Fig. 1 Scanning electron micrograms of 3 wt% Ir/Sibunit 925,000
3.2 Hydrogenation of Citral
Ir4f
62.0
61.5
61.2
The reaction scheme for the hydrogenation of citral is
shown in Fig. 4. The starting material consists of E-citral
(geranial) and Z-citral (neral) with about 50 % of each (the
ratio changed slightly during the reaction). When the C=O
bond is hydrogenated the reaction products are geraniol
and nerol. When the conjugated C=C bond is hydroge-
nated, the reaction product is citronellal. Citronellol is
formed when the double bond in geraniol and nerol or
when the C=O bond in citronellal is hydrogenated. Citro-
nellol can be further hydrogenated to 3,7-dimethyloctanol,
but this did not take place in our case. Side reactions such
as etherification of geraniol and nerol as well as acetal-
ization of citronellal also took place. Moreover, decarb-
onylation and cyclization can occur, but we did not notice
these.
3%-Ir/Sibunit
Oxygen-500°C
3%-Ir/Sibunit-300°C
3%-Ir/Sibunit-Vacuum-500°C
56
58
60
62
64
66
68
70
72
74
Binding energy (eV)
Fig. 2 Ir4f core-level photoelectron spectra of the 3 wt% Ir/Sibunit
catalyst
The conversion profiles as a function of time for dif-
ferent Sibunit catalysts are given in Fig. 5. The conversion
was 31 % in 4 h for 0.55 wt% Ir/Sibunit, 58 % for 1 wt%
Ir/Sibunit and 99 % for 3 wt% Ir/Sibunit. The selectivity to
citronellal (CAL) as a function of conversion is shown in
Fig. 6. The selectivity is highest in the beginning, but
decreases with conversion for 3 and 1 wt% Ir/Sibunit, but
is relatively constant around 12 % for 0.55 wt% Ir/Sibunit.
the preparation procedure treatment with hydrogen was not
sufficient, thus it was completed with ex-situ pre-reduction
prior to experiments at 400 °C. Moreover, analysis of the
samples for chlorine showed that 1 wt% Ir/Sibunit had
much more chlorine than the two other, which is in
agreement with the presence of Ir(III) in 1 wt% Ir/Sibunit.
Table 3 XPS binding energies
Sample
Ir4f7/2
C1s
Ir/C
and Ir/C ratios
0.55 wt% Ir/Sibunit (Ir/C-1)
61.6
62.6
61.5
61.2
62.0
284.5
284.5
284.5
284.5
284.5
0.0026
0.0013
0.0027
0.0021
0.0027
1 wt% Ir/Sibunit (Ir/C-3)
3 wt% Ir/Sibunit (Ir/C-2)
3 wt% Ir/Sibunit, vacuum, T = 500 °C
3 wt% Ir/Sibunit, oxygen, T = 500 °C
123

694
H. Bernas et al.
Ir4f
61.5
62.6
3%-Ir/Sibunit
1%-Ir/Sibunit
0.55%-Ir/Sibunit
Fig. 5 Hydrogenation of citral, conversion versus time, filled
diamond 3 wt% Ir/Sibunit (2 nm), filled square 1 wt% Ir/Sibunit
(6 nm), and filled triangle 0.55 wt% Ir/Sibunit (3 nm)
56
58
60
62
64
66
68
70
72
74
Binding energy (eV)
3 wt% Ir/Sibunit the selectivity decreased with conversion
from around 14 % to ca. 11 %. The selectivity to nerol was
mostly 4 % for 0.55 wt% Ir/Sibunit, and for 3 wt% Ir/Sib-
unit the selectivity was constant ca. 4 % decreasing slightly
after 80 % conversion. The selectivity to nerol for 1 wt% Ir/
Sibunit was constant (ca. 2 %).
Fig. 3 Ir4f core-level photoelectron spectra of the Ir/Sibunit catalysts
The highest selectivity was 18 % for 1 wt% Ir/Sibunit at
around 20 % conversion.
The selectivity to geraniol (G) and nerol (N) as a function
of conversion is shown in Figs. 7 and 8, respectively. As
can be seen, the selectivity was lower for 1 wt% Ir/Sibunit
with 6 nm metal particles compared to the other catalysts
(with 2 and 3 nm). This suggests that the selectivity to
geraniol and nerol is depending on the metal particle size.
Galvagno et al. [5] found no influence of particle size on the
selectivity to geraniol and nerol when Ru/C catalyst was
used, which is different from the current data for iridium.
The selectivity to geraniol increased with conversion for
0.55 wt% Ir/Sibunit from around 11 to 15 %. For 1 wt% Ir/
Sibunit the selectivity to geraniol was around 6 % and for
The selectivity to isopropyl ethers of nerol (NE) and
geraniol (GE) (identified by GC–MS that gave a molecular
weight of Mw 196 and the isopropyl mass fragment M/e? 59)
is presented in Figs. 9 and 10, respectively. It can be noticed,
that the selectivities were close to each other in the case of
0.55 and 1 wt% Ir/Sibunit, but much higher for 3 wt% Ir/
Sibunit. The explanation could be that 3 wt% has much
more metal and this could have an influence on the selec-
tivity. Etherification is mostly done using acidic catalysts
without metal [35, 36]. However, previously etherification
was observed on carbon nanofiber supported palladium
Fig. 4 The reaction scheme for
hydrogenation of citral
123

Hydrogenation of Citral
695
Fig. 9 Selectivity to nerol isopropylether as a function of conversion,
filled diamond 3 wt% Ir/Sibunit (2 nm), filled square 1 wt% Ir/
Sibunit (6 nm), and filled triangle 0.55 wt% Ir/Sibunit (3 nm)
Fig. 6 Selectivity to citronellal as a function of conversion, filled
diamond 3 wt% Ir/Sibunit (2 nm), filled square 1 wt% Ir/Sibunit
(6 nm), and filled triangle 0.55 wt% Ir/Sibunit (3 nm)
Fig. 7 Selectivity to geraniol as a function of conversion, filled
diamond 3 wt% Ir/Sibunit (2 nm), filled square 1 wt% Ir/Sibunit
(6 nm), and filled triangle 0.55 wt% Ir/Sibunit (3 nm)
Fig. 10 Selectivity to geraniol isopropylether as a function of
conversion, filled diamond 3 wt% Ir/Sibunit (2 nm), filled square
1 wt% Ir/Sibunit (6 nm), and filled triangle 0.55 wt% Ir/Sibunit
(3 nm)
Fig. 8 Selectivity to nerol as a function of conversion, filled diamond
3 wt% Ir/Sibunit (2 nm), filled square 1 wt% Ir/Sibunit (6 nm), and
filled triangle 0.55 wt% Ir/Sibunit (3 nm)
Fig. 11 Selectivity to citronellal di-isopropylacetal as a function of
conversion, filled diamond 3 wt% Ir/Sibunit (2 nm), filled square
1 wt% Ir/Sibunit (6 nm), and filled triangle 0.55 wt% Ir/Sibunit
(3 nm)
catalysts where the main reaction was hydrogenolysis [37].
In [37] more ether was formed with less acidic catalyst (the
selectivity to hydrogenolysis product was lower for less
acidic catalyst). We excluded the acidity function since the
pH of the materials were the same, 7.1, when dispersed in
water, which suggests that the materials were not acidic after
reduction at 400 °C. The selectivity to nerol isopropyl ether
was around 7 % for 0.55 wt% Ir/Sibunit, from 8 % in the
beginning at lower conversion to 12 % at higher conversions
for 1 wt% Ir/Sibunit and decreasing from 23 % at lower
123

696
H. Bernas et al.
An overview of TOF for hydrogenation reactions and
selectivity to different products at 30 % conversion is
given in Table 4. The selectivity to nerol and geraniol
ethers was high, as high as 72 % in the case of 3 wt% Ir/
Sibunit. Since the amount of ether is so high, that would
explain why the selectivity to geraniol and nerol was
moderate as they reacted further to form ethers. However,
the lumped selectivity to geraniol, nerol and their ethers
was 90 % for 3 wt% Ir/Sibunit, 38 % for 1 wt% Ir/Sib-
unit and 45 % for 0.55 wt% Ir/Sibunit. It seems that the
amount of surface metal affects the C=O group hydro-
genation; the more surface metal is available the more of
the C=O bond is hydrogenated (the dispersion for 1 wt%
Ir/Sibunit is low compared to 0.55 wt% Ir/Sibunit thus
that the amount of surface metal is actually higher for
0.55 wt%). Moreover, several unknown products, which
were not isopulegol, menthol, citronellol or 3,7-dimeth-
yloctanol, were formed when 0.55 and 1 wt% Ir/Sibunit
were used.
Fig. 12 Turnover frequency of hydrogenation reaction (excluding
acetalization and etherification) at 30 min as a function of Ir particle
size
conversion to 19 % at higher conversion for 3 wt% Ir/Sib-
unit. The selectivity to geraniol isopropyl ether was around
17 % for 0.55 wt% Ir/Sibunit, from 17 % up to 28 % for
1 wt% Ir/Sibunit. For 3 wt% Ir/Sibunit initial selectivity to
this compound was 48 % at lower conversion and decreased
to 43 % at higher conversion. 3 wt% Ir/Sibunit also behaved
differently from the other catalysts regarding formation of
citronellal di-isopropyl acetal (CALacetal, Fig. 11), dis-
playing lower selectivity. The selectivity varied from 24 %
at lower conversion to 32 % at higher conversion for
0.55 wt% Ir/Sibunit and around 33 % for 1 wt% Ir/Sibunit.
3 wt% Ir/Sibunit exhibited just 4 % selectivity towards
citronellal acetals in the beginning and 14 % in the end of the
experiment. Note that acetalization is an acid catalyzed
homogeneous reaction [6] that does not require metal.
However, the materials were not acidic since the pH of the
catalyst slurries were as high as 7.1.
4 Conclusions
The hydrogenation of citral was carried out in a batch
reactor over Sibunit (mesoporous carbon) supported cata-
lysts with 0.55, 1 and 3 wt% iridium. Isopropyl ethers of
geraniol and nerol (2-propanol was used as a solvent),
geraniol, nerol, citronellal di-isopropyl acetal and citro-
nellal were obtained as products. The iridium particle size
was determined by CO-chemisorption and varied from 2 to
6 nm. The turnover frequency for hydrogenation reactions
increased when the metal particle size increased. The for-
mation of geraniol and nerol was lower for 1 wt% Ir/Sib-
unit, which had larger metal particles, higher chlorine
content and different oxidation state than the other two
catalysts. The etherification was the highest for the catalyst
with the highest metal loading (3 wt%), while the least
amount of acetals was formed with the same catalyst. The
results cannot be attributed to acidity function, since the pH
of the catalyst slurries was 7.1, which means that the cat-
alyst was not acidic.
The metal particle size had an influence on turnover
frequency, as seen from Fig. 12 where the turnover fre-
quency (TOF) of hydrogenation reactions (excluding ace-
tals and ethers) is plotted as a function of metal particle
size, showing TOF increase as the metal particle size
increased. Reyes and Rojas [15] used Pt and Pt–Fe/SiO2
catalysts and investigated the influence of metal particle
size in the range of 1.6 to 3.8 nm. The metal particle size
did not affect the TOF in their case.
Table 4 Hydrogenation of citral using carbon supported catalysts
Catalyst
TOF (h-1 a
)
Sel CAL (%)b Sel G (%)b Sel N (%)b Sel GE ? NE (%)b Sel CALacetal (%)b Sel unknown (%)b
0.55 wt% Ir/Sibunit 117.0
11
10
6
15
5
5
1
4
25
32
72
31
33
4
13
19
0
1 wt% Ir/Sibunit
174.8
37.0
3 wt% Ir/Sibunit
14
a
Moles hydrogenated (excluding acetalization and etherification) in 30 min per mole of surface metal
Selectivity at 30 % conversion
b
123

Hydrogenation of Citral
697
˚
Acknowledgments This work is part of the activities at the Abo
Akademi University Process Chemistry Centre within the Finnish
Centre of Excellence Programme (2000–2011). Financial support
from European Union through the Seventh Framework Programme
(Project 246095-2) is gratefully acknowledged.
20. Salmi T, Ma¨ki-Arvela P, Toukoniitty E, Kalantar Neyestanaki A,
Tiainen L-P, Lindfors L-E, Sjo¨holm R, Laine E (2000) Appl
Catal A 196:93
21. Reyes P, Rojas H, Pecchi G, Fierro JLG (2002) J Mol Catal A
179:293
´
22. Rojas H, Borda G, Reyes P, Martınez JJ, Valencia J, Fierro JLG
(2008) Catal Today 133–135:699
23. Savchenko VI, Makaryan IA, Dorokhov VG (1997) Platin Met
Rev 41:176
24. Steffan M, Jakob A, Claus P, Lang H (2009) Catal Commun
10:437
References
´
¨
1. Maki-Arvela P, Hajek J, Salmi T, Murzin DY (2005) Appl Catal
A 292:1
2. Claus P (1998) Top Catal 5:51
3. Gallezot P, Richard D (1998) Catal Rev Sci Eng 40:81
´
25. Asedegbega-Nieto E, Bachiller-Baeza B, Guerrero-Ruız A,
´
Rodrıguez-Ramos I (2006) Appl Catal A 300:120
26. Lafaye G, Ekou T, Micheaud-Especel C, Montassier C, Marecot
P (2004) Appl Catal A 257:107
´
4. Alvarez-Rodrıguez J, Rodrıguez-Ramos I, Guerrero-Ruiz A
´
(2009) Appl Catal A 366:114
´
27. Liu R, Yu Y, Yoshida K, Li G, Jiang H, Zhang M, Zhao F, Fuijita
S, Arai M (2010) J Catal 269:191
28. Recchia S, Dossi C, Poli N, Fusi A, Sordelli L, Psaro R (1999) J
Catal 184:1
5. Galvagno S, Milone C, Donato A, Neri G, Pietropaolo R (1993)
Catal Lett 18:349
¨
6. Maki-Arvela P, Tiainen L-P, Kalantar Neyestanaki A, Sjoholm R,
Rantakyla T-K, Laine E, Salmi T, Murzin DY (2002) Appl Catal
¨
¨
´
29. Zgolicz PD, Rodrıguez VI, Vilella IMJ, de Miguel SR, Scelza OA
(2011) Appl Catal A 392:208
A 237:181
7. Singh UK, Vannice MA (2001) J Catal 199:73
8. Neri G, Mercadante L, Donato A, Visco AM, Galvagno S (1994)
Catal Lett 29:379
30. Yermakov YI, Surovikin VF, Plaksin GV, Semikolenov VA,
Likholobov VA, Chuvilin AL, Bogdanov SV (1987) React Kinet
Catal Lett 33:435
31. Fenelonov VB, Likholobov VA, Derevyankin AY, Mel’gunov
MS (1998) Catal Today 42:341
32. McVicker GB, Baker RTK, Garten RL, Kugler EL (1980) J Catal
65:207
33. Scofield JHJ (1976) Electron Spectrosc Relat Phenom 8:129
34. Moudler J, Stickle W, Sobol P, Bomben K (1992) Handbook of
X-ray photoelectron spectroscopy. Perkin-Elmer Corp., Eden
Priarie
´ˇ ´
35. Klepacova K, Mravec D, Kaszonyi A, Bajus M (2007) Appl Catal
A 328:1
36. Ruppert AM, Parvulescu AN, Arias M, Hausoul PJC, Bruijnincx
PCA, Klein Gebbink RJM, Weckhuysen BM (2009) J Catal
268:251
¨
9. Tiainen L-P, Maki-Arvela P, Kalantar Neyestanaki A, Salmi T,
Murzin DY (2003) React Kinet Catal Lett 78:251
´
10. Asedegbega-Nieto E, Guerrero-Ruiz A, Rodrıguez-Ramos I
(2006) Carbon 44:799
11. Ekou T, Ekou L, Vicente A, Lafaye G, Pronier S, Especel C,
´
Merecot P (2011) J Mol Catal A 337:82
12. Malathi R, Viswanath RP (2001) Appl Catal A 208:323
13. Mukherjee S, Vannice MA (2006) J Catal 243:108
¨
¨
14. Maki-Arvela P, Kumar N, Eranen K, Salmi T, Murzin DY (2006)
Chem Eng J 122:127
15. Reyes P, Rojas H (2006) React Kinet Catal Lett 88:363
¨
16. Aumo J, Lilja J, Maki-Arvela P, Salmi T, Sundell M, Vainio H,
Murzin DY (2002) Catal Lett 84:219
17. Virtanen P, Salmi TO, Mikkola J-P (2010) Top Catal 53:1096
18. Milone C, Tropeano ML, Gulino G, Neri G, Ingoglia R, Galv-
agno S (2002) Chem Comm 8:868
¨
37. Markus H, Plomp AJ, Maki-Arvela P, Bitter JH, Murzin DY
(2007) Catal Lett 113:141
19. Tang Y, Yang D, Qin F, Hu J, Wang C, Xu HJ (2009) Solid State
Chem 182:2279
123