Reduction of citral in water under typical transfer hydrogenation conditions-Reaction mechanisms with evolution of and hydrogenation by molecular hydrogen

Article abstract of DOI:10.1016/j.molcata.2012.10.010

The reduction of an α,β-unsaturated aldehyde, citral, was investigated over a 10 wt% Pd catalyst under transfer hydrogenation (TH) conditions in a closed system with microwave assistance. Surprisingly, it was found that hydrogen was produced quite fast under the microwave irradiation during the reaction, and the reduction of citral was proved to go mainly through consecutive pathways of hydrogen production - hydrogenation rather than those commonly considered for TH reactions. Similar reaction pathways were also observed with a homogeneous catalyst of [RuCl₂(C₆H ₆)]₂ and other typical hydrogen donors like formate salts and isopropanol, which are usually used in the typical transfer hydrogenations.

Full text of DOI:10.1016/j.molcata.2012.10.010

Journal of Molecular Catalysis A: Chemical 366 (2013) 315–320
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Reduction of citral in water under typical transfer hydrogenation
conditions—Reaction mechanisms with evolution of and hydrogenation by
molecular hydrogen
Ruixia Liua,b, Yu Wangb, Haiyang Chenga, Yancun Yua, Fengyu Zhaoa,∗, Masahiko Araib,∗∗
a State Key Laboratory of Electroanalytical Chemistry, Laboratory of Green Chemistry and Process, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun
130022, PR China
b Division of Chemical Process Engineering, Faculty of Engineering, Hokkaido University, Sapporo 060-8628, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 April 2012
Received in revised form 12 October 2012
Accepted 13 October 2012
Available online 22 October 2012
The reduction of an ␣,␤-unsaturated aldehyde, citral, was investigated over a 10 wt% Pd catalyst under
transfer hydrogenation (TH) conditions in a closed system with microwave assistance. Surprisingly, it was
found that hydrogen was produced quite fast under the microwave irradiation during the reaction, and
the reduction of citral was proved to go mainly through consecutive pathways of hydrogen production
– hydrogenation rather than those commonly considered for TH reactions. Similar reaction pathways
were also observed with a homogeneous catalyst of [RuCl2(C6H6)]2 and other typical hydrogen donors
like formate salts and isopropanol, which are usually used in the typical transfer hydrogenations.
© 2012 Elsevier B.V. All rights reserved.
Keywords:
Transfer hydrogenation
Citral
Mechanism
Microwave
Pd/C
1. Introduction
the TH reactions with hydrogen donors, it is still not clear how
does the hydrogen transfer. For example, formic acid was pro-
Reduction of organic compounds is significantly important for
organic synthetic chemistry both in laboratory and in industry
[1–4]. Among all the methods of reduction of organic functional
groups, hydrogenation is the most commonly used one. It is a
process including the addition of hydrogen or replacement of
a functional group by hydrogen. Hydrogenation with molecular
hydrogen has gained a great success in understanding the reaction
mechanisms and has been widely used in industry. It is gener-
ally regarded that many catalytic hydrogenations with molecular
hydrogen actually involve the action of atomic hydrogen over the
surface of catalysts. For transfer hydrogenation (TH), a great deal of
progress has been achieved [5–11], since the report on reduction
of ketones and aldehydes to secondary alcohols with aluminum
isopropylate catalysis in isopropanol in 1925 [7,12]. However, TH
mechanisms are still not understood well, in particular for the het-
erogeneous catalytic TH [13–21]. Heterogeneous catalytic TH is not
a simple reaction in which hydrogen is generated from a hydro-
gen donor and then an acceptor is reduced by hydrogen formed. In
posed as a hydrogen donor to give a proton and a hydride or two
hydrogen atoms [3]. Until now, the mechanisms are still contra-
dictory. Earlier, Wieland suggested that the donor reacted initially
with palladium catalyst to form a palladium hydride intermediate,
which was added to the acceptor, and then decomposed [22]. Later
researchers preferred a mechanism in which both hydrogen donor
and acceptor were co-adsorbed onto the palladium surface fol-
lowed by direct transfer of hydrogen without formation of a hydride
[23]. According to the results of deuterium labeling experiments,
hydrogen can be transferred directly from a donor to an acceptor
[24]. However, some authors pointed out that the transfer of hydro-
gen as a hydride species was not unreasonable [2]. Certainly, some
evidence was presented for the formation of a hydride species in
the decomposition of formic acid over Pd [25]. More recently, it was
proposed that transfer of hydrogen between adsorbed species was
more available than transfer of hydrogen from metal to alkene [26].
The adsorption of alkene should give an intermediate that could
transfer hydrogen directly to an adjacent adsorbed species.
In the present work, the reduction of C C bonds in an ␣,␤-
unsaturated aldehyde (citral) was investigated over a commercial
10 wt% Pd catalyst under TH conditions with HCOONa as hydrogen
donor in a closed system with microwave assistance. Although the
reactions were carried out under typical TH conditions, these were
∗
Corresponding author. Tel.: +86 0431 85262410; fax: +86 0431 85262410.
Corresponding author. Tel.: +81 011 7066594; fax: +81 011 7066594.
E-mail addresses: zhaofy@ciac.jl.cn (F. Zhao), marai@eng.hokudai.ac.jp (M. Arai).
∗∗
1381-1169/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2012.10.010

316
R. Liu et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 315–320
indicated to mainly go through the formation of molecular hydro-
gen. That is, the reduction of citral occurs through the formation
of molecular hydrogen followed by hydrogenation of the substrate
with hydrogen formed. The influence of other hydrogen donors and
solvents on the reaction outcome and mechanisms has also been
investigated.
of citral reached 68%, citronellal (hydrogenation of conjugated
bond) and dihydrocitronellal (hydrogenation of conju-
C
C
gated C C bond and isolated C C bond) were produced as the
main products with selectivity of 86% and 9%, respectively,
and traces of isopulegol, menthol, citronellol and verbenol
were also detected. So consumption of hydrogen during the
reaction is about n = 2 mmol × 68% × (86% + 9% × 2) = 1.41 mmol.
The pressure of the system after the reaction is 3.7 bar, while
vapor pressure of the mixture of citral (0.6 mmol) and citro-
nellal (1.4 mmol) in 5 mL H2O is 3 bar under the reaction
conditions. So the quantity of hydrogen left after reaction is
n = pV/RT = (3.7 − 3.0) bar × 10−5 × 5 mL × 10−6/8.314472 J K−1
mol−1 × 353 K = 0.12 mmol. So the total quantity of hydro-
gen produced in the presence of acceptor (2 mmol citral)
n = (0.12 + 1.41) mmol = 1.53 mmol, which is in agreement to
the quantity of hydrogen calculated by the amount of removing
water suggesting there are no CO2 formation during the reac-
tion. Therefore, we can say that the present reaction is mainly
going through the pathway of hydrogen production and then
hydrogenation with hydrogen molecules.
2. Experimental
2.1. Materials
All microwave experiments were performed using a Biotage
Initiator 2.0 microwave synthesizer (Uppasla, Sweden). A com-
mercial 10 wt% Pd/C (Shanghai Reagent Ltd.) and [RuCl2(C6H6)]2
(Aldrich) were used as catalysts. Citral (trans and cis, purity 95%)
was purchased from Aldrich and toluene, poly(ethylene glycol)
(PEG), isopropanol, K2CO3, HCOONa·2H2O, HCOOK, HCOONH4 and
HCOOH from Beijing Chemical Reagent Company.
2.2. General procedures for the reduction of citral with
microwave irradiation
3. Results and discussion
The reduction of citral was carried out in a quartz tube (10 mL)
under microwave irradiation. The order of addition of the reagents
plays an important role in the reactions [27]. We selected the fol-
lowing standard protocol. Hydrogen donor was first dissolved in
a solvent in the reactor, catalyst was then added, and finally the
substrate was added. Then, the reaction vessel was sealed and the
reaction was carried out under microwave irradiation at 300 W
with a stirring speed of 900 r/min. The reaction time was started
to count when the reaction mixture reached the desired tempera-
ture. After the reaction, the mixture was extracted with n-hexane
and the resulting solution was analyzed with gas chromatography
(GC-Shimadzu-14C, FID, Capillary column Rtx-Wax 30 m-0.53 mm-
0.25 mm) and gas chromatography/mass spectrometry (GC/MS,
Agilent 5890). The gas phases were analyzed by Shimadzu GC-14C
with TCD and a TDX-01 packed column. The reactions in the auto-
clave (50 mL) were also carried out in a water-bath with the same
procedures.
3.1. Hydrogen evolution
Citral was hydrogenated in a sealed microwave reactor with
HCOONa as a hydrogen donor with 10 wt% Pd/C catalyst in water at
80 ◦C. The total conversion of citral was 68% under the conditions
used. In the organic liquid phase, citronellal and dihydrocitronellal
were detected in the selectivity values of 86% and 9%, respec-
tively, along with such very minor products as isopulegol, menthol,
citronellol, and verbenol. The system pressure was continuously
measured with a pressure sensor during the reaction. Fig. 1 (line (a))
shows that the pressure increased sharply to 9.6 bar within an ini-
tial 30 s and decreased gradually to 4 bar in about 30 min. After the
reactor was cooled down to room temperature, the pressure was
still 1 bar (relative to atmospheric pressure). The pressure increase
should be caused by the production of molecular hydrogen but not
by vaporization of water and organic compounds with an increase
in the temperature (Fig. 1B).
2.3. The hydrogen formation and consumption during the
reaction
To examine the implication of the above-mentioned obser-
vations, we further measured the changes of pressure and
temperature with time for several selected reaction mixtures
(Fig. 1). When only water was used (line (b)), the pressure increased
but less sharply, compared with the above-mentioned line (a), up
to 4.5 bar, this pressure maintained unchanged during 30 min, and
it decreased to atmospheric pressure after the reactor was cooled to
room temperature. When organic compounds of citral and/or citro-
nellal were added, similar pressure changes were seen (lines (c)
and (d)) but the increased pressure (3.5 bar) was smaller than that
observed in water alone. Note that such a large pressure increase as
observed during the reaction (line (a)) did not appear for the mix-
tures of water and/or organic compounds. The addition of catalyst
into water did not influence the pressure change of the water alone
(lines (b) and (e)). But, for the mixture containing water, HCOONa,
and Pd/C catalyst, the pressure increased promptly up to 13.8 bar in
a few seconds and then maintained unchanged (line (f)). After the
reactor was cooled, the gas phase was collected and analyzed by GC,
only hydrogen gas was detected as we expected, and moreover, the
gas phase sample is flammable and could burn in air, confirming
again it was hydrogen gas. In previous works, however, the forma-
tion of hydrogen was not detected in the TH reactions, whenever
it occurred in open or closed reactors under either conventional
oil or microwave heating [28–33]. The aqueous phase indicated a
pH value of 8.9–9.3, which was between NaHCO3 solution (pH 8.3)
2.3.1. Demonstration of the production of hydrogen
The reaction gas phase was analyzed by GC and the conventional
test. After reaction, the gas phase was collected and analyzed by GC
using the TCD detector, the results show that hydrogen was formed.
Moreover, the collected gas can be kindled in the air and so it was
confirmed to be hydrogen again.
2.3.2. Estimation of the quantity of hydrogen produced
The hydrogen produced during the reaction in the presence of
HCOONa has been collected by water removing method and cal-
culated by the ideal gas equation. We carried out the test under
the reaction conditions with Pd/C catalyst, water, and HCOONa,
but without the reactant citral. After the pressure goes to the
unchanged level, the reactor was cooled to room temperature then
the gas was transferred to a volumetric cylinder immersed into
water and then the volume of the gas produced was calculated by
sum volume of V1 + V2, V1 volume of reactor excluding the volume
of reaction solution; V2 volume of the amount of removing water.
Finally, the hydrogen amount was calculated by ideal gas equation
pV = nRT.
When the selective reduction of citral (2 mmol) were carried
out with HCOONa aqueous solution (1.2 M, 5 mL) as hydrogen
donor over 10% Pd/C catalyst at 80 ◦C in 30 min, the conversion

R. Liu et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 315–320
317
Scheme 1. Reaction pathways for the reduction of citral over 10%Pd/C in H2O with
HCOONa as a hydrogen donor under the microwave assistance.
H2 formed through the decomposition of HCOONa and both the
decomposition and the hydrogenation are promoted by Pd/C cat-
alyst as shown in Scheme 1. The TH should also occur but less
significant in the present reaction system. Such a decomposition
reaction of HCOONa with noble metal catalysts was reported, but
no hydrogen was detected to form in the presence of acceptor in
the literature [34]. In addition, some researchers suggested that the
acceptor could not or less be reduced with the produced hydrogen
under the conditions, even though hydrogen was produced in their
experiments [35,36]. For example, it was observed that formic acid
decomposed to H2 and CO2 to a larger extent during RuII-catalyzed
asymmetric TH of ketones; however, the gaseous hydrogen was
claimed to contribute very less to the formation of alcohols [35].
In another example, hydrogen was reported to be released from
p-menthane and decalin in xylene at 144 ◦C but no hydrogenation
reaction occurred [36].
No hydrogen was detected in the most TH reactions over Pd/C,
this may be ascribed to (1) the hydrogenation of hydrogen accep-
tor is much faster than the decomposition of hydrogen donor, (2)
the quantity of hydrogen, if produced, is so small that it cannot
be detected, which may be significant for open systems (in the
present work, a closed reaction system was used). HCOONa decom-
poses rapidly under the microwave irradiation, yielding molecular
hydrogen. Then, the formed hydrogen was consumed for the hydro-
genation of citral in a lower rate; so, some hydrogen remains
un-reacted in the gas phase and can be detected in the present
work.
Fig. 1. In situ pressure (A) and temperature (B) changes for various mixtures on
microwave heating at 80 ◦C. Conditions: H2O 5 cm3, 10% Pd/C 40 mg, HCOONa·2H2O
6 mmol, citral 2 mmol, 0.6 mmol citral + 1.4 mmol citronellal.
and Na2CO3 solution (pH 11.6) under the same conditions. Those
results indicate that HCOONa decomposes to H2 and NaHCO3 in the
aqueous phase assisted by 10 wt% Pd/C.
3.3. Reaction with different hydrogen donors
3.2. A case study of reaction pathway
In the TH reactions with hydrogen donors, it is still not clear how
does the hydrogen transfer. For example, formic acid was proposed
as a hydrogen donor to give a proton and a hydride or two hydro-
gen atoms [3]. Until now, the mechanisms are still contradictory.
Earlier, Wieland suggested that the donor reacted initially with pal-
ladium catalyst to form a palladium hydride intermediate, which
was added to the acceptor, and then decomposed [22]. Herein the
influence of hydrogen donor was studied also. Note that, in the
presence of HCOONa and Pd/C catalyst, citral was hydrogenated
mainly to citrenollal and dihydrocitonellal along with the forma-
tion of molecular hydrogen. Further we considered the significance
of two kinds of hydrogen donors, H2 gas and HCOONa, and mea-
sured and compared with the reactions conducted in an autoclave
(50 mL) under conventional heating because the decomposition
of HCOONa to hydrogen was quite fast (less than 30 s) under
microwave irradiation. The results are shown in Table 1, implying
that when H2 gas was used as hydrogen donor, the TOF (turnover
frequency) was 14.0 mmol g−1 min−1 (entry 1). The reaction was
also run with an adjusted amount of HCOONa (6 mmol) in which the
quantity of H atoms included was very similar as that in H2 (4 bar in
entry 1), and a smaller TOF of 0.04 mmol g−1 min−1 was obtained
for the reaction with HCOONa (entry 3). That is, the TOF of the
We calculated and compared the amount of H2 evolved with
that of citral consumed for the reaction of citral (2 mmol), HCOONa
(6 mmoL), H2O (5 mL), and 10% Pd/C catalyst (21 mg) operated at
80 ◦C for 30 min under the microwave irradiation. The results of
lines (a) and (f) in Fig. 1 strongly suggested that the H2 formed
was consumed for the hydrogenation of citral, because that the
system pressure kept a stable level in the absence of citral, indi-
cating that hydrogen was produced (a) and it decreased smoothly
with the progress of the reaction in the presence of citral, suggest-
ing the hydrogen formed was consumed during the reaction (f).
The present reaction gave a total citral conversion of 68% and pro-
duced citronellal in a selectivity of 86% and dihydronellal in 9%, as
mentioned above. The quantity of H2 needed for these conversion
and selectivity values was 1.41 mmol as calculated in Section 2.
Considering the 0.12 mmol hydrogen left in the system after reac-
tion, the total quantity of H2 formed was estimated to be 1.53 mmol
from the results of line (a) of Fig. 1. That is similar to the results of
line (f), in which 1.56 mmol H2 was formed through the decom-
position of HCOONa in water over Pd catalyst in the absence of
citral. Therefore, for the present reaction using HCOONa as a hydro-
gen donor, the reaction should mainly take place with gaseous

318
R. Liu et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 315–320
Table 1
Comparison of the results of citral reduction with H2 gas or HCOONa as hydrogen donor under the conventional heating mode.
Hydrogen donor
+
O
O
O
Citral
Citronellal
(A)
Dihydrocitronellal
(B)
.
Entry
Hydrogen donor
Additive
Catalyst
Time (min)
Conversion (%)
Selectivity (%)
TOF (mmol g−1 min−1
)
A
B
C
1
2
3
H2 (4 bar)
H2 (1 bar)
HCOONa (6 mmol)
10 wt% Pd (5 mg)
10 wt% Pd (5 mg)
10 wt% Pd (40 mg)
10
10
240
35
45
20
72
93
56
6
2
13
22
5
31
14
18
0.04
NaHCO3
Citral 2 mmol; NaHCO3 6 mmol; H2O 5 mL, 80 ◦C. C: citronellol, 3,7-dimethyloctanol, geraniol, nerol, and menthol.
reaction with H2 was about 350 times larger than that of the reac-
tion with HCOONa in spite of the similar quantities of H atoms
involved. According to the result of Fig. 1 (line (a)), about 26% of
the initial amount of HCOONa used was decomposed to H2 and
NaHCO3. So, we conducted another reaction in the presence of
H2 and NaHCO3 in the quantities corresponding to the extent of
HCOONa decomposition of 26% and a TOF of 18.0 mmol g−1 min−1
was observed (entry 2). These results show that the direct hydro-
genation of citral with H2 should be faster than the TH with HCOONa
under the conditions used, and such results should be suitable
to the reactions under the microwave assistance. Note that lit-
tle amount of hydrogen was left in the system after reaction (as
calculated and discussed above), and, in addition, the hydrogena-
tion with hydrogen gas was much faster than the TH. Therefore,
the reduction of citral with the HCOONa hydrogen donor should
mainly go through the hydrogenation with the produced hydrogen.
Furthermore, Table 1 also indicates that the product distribution
depends on the reaction systems and the selectivity to citronellal is
larger in the presence of H2 and NaHCO3 (entry 3); the selectivity
to citronellal is 93%, which is comparable to 86% observed in the
above-mentioned microwave-assisted reaction in the presence of
HCOONa. This can also suggest that the hydrogenation is predom-
inating reaction in the presence of HCOONa under the microwave
irradiation, but the TH should not be ignored under the reaction
conditions used.
The reduction of citral was further studied with other com-
mon hydrogen donors to compare with HCOONa. Many studies
demonstrate that formate salt as a hydrogen donor is superior to
formic acid in the TH. Formate salt is usually utilized as a univer-
sal hydrogen carrier and the decomposition mechanism of formate
acid/salt to produce hydrogen has attracted increasing interest as
hydrogen is the cleanest energy source [37–40]. Table 2 gives the
results with different hydrogen donors. When HCOOK was used,
the total conversion and the selectivity to citronellal were 62% and
76%, respectively, which were both lower than those with HCOONa
(entries 1 and 2). However, Wiener et al. observed that HCOOK was
more effective than HCOONa and HCOOH in TH of nitrotoluene
over 10 wt% Pd/C [34]. They suggested that the KHCO3 produced
from HCOOK was more soluble than NaHCO3 from HCOONa and
did not cover the surface of catalyst, leading to the higher activ-
ity. The results with HCOONH4 were quite different from those
with HCOONa and HCOOK (entry 3). This is because it can supply
molecular hydrogen (decomposing to hydrogen, ammonia and car-
bon dioxide) even in the absence of H2O [27,41–43]. However, the
total conversion was small (44%) and the selectivity to citronellal
was only 53% due to that NH3 reacted with aldehydes (citral and
citronellal) to produce corresponding hydroxylamines and imines.
HCOOH was practically inactive for hydrogenation of citral but
effective for acid catalyzed reactions (entry 4). The system pres-
sure (4.8 bar) did not change during the whole course of reaction
(Fig. 2d), which indicated that HCOOH did not decompose to hydro-
gen. Similarly, when a mixture of HCOONa and HCOOH was used
(entries 5 and 6), the main product came from acid-catalyzed reac-
tions rather than the hydrogenation. So, the presence of HCOOH
might retard the decomposition of HCOONa and HCOOH could not
be activated over Pd/C as reported in transfer hydrogenolysis of
2-chlorotoluene [43].
In addition, isopropanol was tested for the reduction of citral
with a heterogeneous catalyst of 10 wt% Pd/C and a homogeneous
one of [RuCl2(C6H6)]2 (entries 7 and 8). Very similar pressure
changes were seen in the both cases (Fig. 2, lines (g) and (h)). It
was confirmed that isopropanol produced hydrogen and acetone
and then citral was hydrogenated with the in situ released hydro-
gen, in the same manners as for the other hydrogen donors. But, the
conversion was quite low (12%) over 10 wt% Pd/C and the selectivity
to hydrogenation was low due to the aldolization reaction between
citral and isopropanol. It was also found that no reaction occurred
in the absence of K2CO3 base. For [RuCl2(C6H6)]2 catalyst, the con-
version was enhanced up to 57% and the selectivity to citronellal
increased to 54%. Those results indicate that the selective reduction
of citral mainly goes through consecutive reactions of hydrogen
Table 2
Results of citral reduction with different hydrogen donors.a
Entry
Hydrogen donor
Conversion (%)
Selectivity (%)
A
B
C
1
2
3
4
5
6
7
8
HCOONa
HCOOK
HCOONH4
HCOOH
68
62
44
91
13
7
86
76
53
9
13
6
5f
11f
4g
100h
75h
80h
91i
HCOOH/HCOONab
HCOOH/HCOONac
Isopropanold
Isopropanole
17
16
6
8
4
3
1
12
57
54
45i
A: citronellal; B: dihydrocitronellal; C: others.
a
Citral 2 mmol, formic salt or formic acid 6 mmol, 10% Pd/C 21 mg, H2O 5 mL,
80 ◦C, 30 min, microwave heating 300 W.
b
The mol ratio of HCOOH to HCOONa is (1:1).
The mol ratio of HCOOH to HCOONa is (1:5).
Isopropanol 5 mL, K2CO3 0.1 g, without H2O.
Isopropanol 5 mL, [RuCl2(C6H6)]2 10 mg, without H2O.
c
d
e
f
Traces of citronellol, 3,7-dimethyloctanol, geraniol, nerol, and menthol.
g
3,7-Dimethylocta-2,6-dienylidene
hydroxylamine,
3,7-dimethylocta-2,6-
dienylidene imine and others formed through the reaction between NH3 and
aldehydes.
h
Benzene, ethylbenzene, isopropylbenzene, and others formed through acid-
catalyzed reactions.
i
3,7-Dimethyl-2,6-octadienal diisopropanol acetal.

R. Liu et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 315–320
319
Table 3
Results for the reduction of citral in different solvents.a
Solvent
Catalyst
Time (min)
Conversion(%)
Selectivity (%)
A
B
C
H2O
10% Pd/C (40 mg)
10% Pd/C (40 mg)
10% Pd/C (40 mg)
10% Pd/C (21 mg)
10% Pd/C (21 mg)
[RuCl2(C6H6)]2 (20 mg)
[RuCl2(C6H6)]2(5 mg)
30
30
30
20
20
30
5
68
75
9
86
87
59
96
2
9
7
2
5d
6d
H2O/tolueneb
Toluene
PEGc
39e
4f
9
H2O/PEGc
H2O
90
15
82
98f
5g
49
16
H2O/PEGc
84h
A: citronellal; B: dihydrocitronellal; C: others.
a
Citral 2 mmol, solvent 5 mL, HCOONa·2H2O 6 mmol, 80 ◦C, microwave heating 300 W.
b
c
d
e
f
H2O 3 mL, toluene 2 mL.
H2O 3 mL, PEG-400 2 mL, 180 ◦C.
Citronellol, 3,7-dimethyloctanol, geraniol, nerol, menthol, and 6-methyl-5-hepten-2-one.
Citronellol, 3,7-dimethyloctanol, geraniol, nerol, menthol, 3,7-dimethyloctanol and some other unidentified products.
6-Methyl-5-hepten-2-one.
Citronellol.
g
h
Unsaturated alcohols of geraniol and nerol.
Fig. 2. In situ pressure changes for the reduction of citral with different hydrogen
donors on microwave heating. Reaction conditions: citral 2 mmol, formic salt/formic
acid 6 mmol, 10% Pd/C 21 mg, H2O 5 mL, 80 ◦C, 30 min, microwave heating 300 W. (a)
Mixture hydrogen donor and the mol ratio of HCOOH and HCOONa is 1:1; (b) mixture
hydrogen donor and the mol ratio of HCOOH and HCOONa is 1:5; (c) isopropanol
5 mL, K2CO3 0.1 g, without H2O; (d) isopropanol 5 mL, [RuCl2(C6H6)]2 10 mg, without
H2O.
Fig. 3. In situ pressure changes for the reduction of citral in different solvents on
microwave heating. Reaction conditions: citral 2 mmol, microwave heating 300 W,
HCOONa·2H2O 6 mmol, (a) H2O 3 mL, 10% Pd/C 5 mg; (b) H2O 3 mL, toluene 2 mL,
10% Pd/C 5 mg; (c) toluene 2 mL, 10% Pd/C 5 mg; (d) H2O 3 mL, PEG-400 2 mL, 10%
Pd/C 5 mg; (e) H2O 3 mL, [RuCl2(C6H6)]2 5 mg; (f) [RuCl2(C6H6)]2 5 mg, PEG-400
2 mL, H2O 3 mL.
production and hydrogenation with the formed hydrogen gas irre-
spective of the hydrogen donors and the catalysts used under the
present conditions.
molecular weight of 400. Fig. 3 and Table 3 show the results
obtained at 80 ◦C. When a H2O/toluene mixture was used, the pres-
sure increased rapidly to 10.5 bar in an initial 2 min and dropped
to 3 bar in the following 7 min (Fig. 3, line (b)). The pressure drop
was much faster than that observed with pure H2O (line (a)). With
pure toluene, in contrast, the pressure change was very small (line
(c)). The reaction results in these different media are presented in
3.4. Reaction in different green media
The reduction was further studied with different media using
green solvents of water and poly(ethylene glycol) (PEG) with a
Scheme 2. Reaction pathway of H2O and citral in H2O/PEG solvent with microwave assistance at elevated temperature.

320
R. Liu et al. / Journal of Molecular Catalysis A: Chemical 366 (2013) 315–320
Table 3. The mixed solvent medium of H2O/toluene gave a total
conversion of 75% and a selectivity to citronellal of 87%, which
were larger and similar to those obtained with H2O alone, respec-
tively. In contrast, very smaller conversion of 9% and selectivity of
59% were obtained in toluene alone. The low conversion in this
medium should result from the absence of one of two components
of H2O giving molecular hydrogen (Scheme 1). Furthermore, PEG
was also checked because of its lower vapour pressure and good
absorption of microwave [44]. It was observed that no reaction
occurred at temperatures <120 ◦C; at 180 ◦C, a small conversion
of 9% was obtained. When H2O was also added, the total con-
version increased significantly but 6-methyl-5-hepten-2-one was
produced in a selectivity of 93%, through the addition–elimination
reaction of citral and H2O, as shown in Scheme 2. It was reported
that H2O could behave differently at elevated temperatures; the
bond angle widens and its dielectric properties become similar
to those of organic solvents [45]. It is thus assumed that H2O is
activated at 180 ◦C and added to the citral molecule, yielding 6-
methyl-5-hepten-2-one and acetaldehyde (Scheme 2).
References
[1] R.E. Harmon, S.K. Gupta, D.J. Brown, Chem. Rev. 73 (1973) 21.
[2] G. Brieger, T.J. Nestrick, Chem. Rev. 74 (1974) 567.
[3] J.R.W. Johnstone, A.H. Wilby, I.D. Entwistle, Chem. Rev. 85 (1985) 129.
[4] B.M. Trost, I. Fleming (Eds.), Comprehensive Organic Synthesis, vol. 8, Pergmon,
Oxford, 1991.
[5] M. Kitamura, R. Noyori, in: S.-I. Murahashi (Ed.), Ruthenium in Organic Synthe-
sis, Wiley-VCH, Weinheim, 2004, pp. 31–32.
[6] J.M. Brunel, Tetrahedron 63 (2007) 3899.
[7] D. Klomp, U. Hanefeld, J.A. Peters, in: J.G. de Vries, C.J. Elsevier (Eds.), The Hand-
book of Homogeneous Hydrogenation, vol. 1, Wiley-VCH, Weinheim, 2007
(Chapter 20).
[8] C. Wang, X. Wu, J. Xiao, Chem. Asian J. 3 (2008) 1750.
[9] F. Alonso, P. Riente, M. Yus, Acc. Chem. Res. 44 (2011) 379.
[10] Y. Masashi, Y. Issaku, N. Ryoji, Angew. Chem. Int. Ed. 40 (2001) 2818.
[11] M.K. Ikariya, T.R. Noyori, Org. Biomol. Chem. 4 (2006) 393.
[12] P. Wolfgang, Angew. Chem. 39 (1926) 138.
[13] F. Alonso, M. Yus, Tetrahedron Lett. 38 (1997) 149.
[14] F. Alonso, M. Yus, Tetrahedron Lett. 37 (1996) 6925.
[15] F. Alonso, I. Osante, M. Yus, Synlett (2006) 3017.
[16] F. Alonso, I. Osante, M. Yus, Tetrahedron 63 (2007) 93.
[17] F. Alonso, P. Riente, M. Yus, Tetrahedron 64 (2008) 1847.
[18] F. Alonso, P. Riente, M. Yus, Tetrahedron 65 (2009) 10637.
[19] J.F. Quinn, D.A. Razzano, K.C. Golden, B.T. Gregg, Tetrahedron Lett. 49 (2008)
6137.
4. Conclusion
[20] Y. Kume, K. Qiao, D. Tomida, C. Yokoyama, Catal. Commun. 9 (2008) 369.
[21] A.M.R. Galletti, C. Antonetti, A.M. Venezia, G. Giambastiani, Appl. Catal. A: Gen.
386 (2010) 124.
Under the microwave irradiation, the reaction pathway for the
selective reduction of citral under typical transfer hydrogenation
(TH) conditions over a heterogeneous 10 wt% Pd/C catalyst was
studied. The system pressure increased promptly within the first
3 min, the GC analysis indicated the production of a large amount
of hydrogen from HCOONa in aqueous phase over Pd/C catalyst.
The results suggest that the reaction mainly goes through consec-
utive pathways of hydrogen production and then hydrogenation
reaction, rather than those commonly considered of TH reactions.
Moreover, the proposed reaction pathway was also demonstrated
for other catalysts under TH conditions like a homogeneous cat-
alyst of [RuCl2(C6H6)]2, and typical hydrogen donors such as
formate salts and isopropanol. However, acid-catalyzed reaction
took place when formate acid was used as hydrogen donor, and
addition–elimination reaction of citral and H2O occurred at high
temperature in the case of H2O/PEG mixed solvent was used.
[22] H. Wieland, Chem. Ber. 45 (1912) 484.
[23] E.A. Braude, R.P. Linstead, P.W.D. Mitchell, J. Am. Chem. Soc. (1954) 3578.
[24] I. Fujii, K. Ryu, K. Hayakawa, K. Kanematsu, J. Chem. Soc. Chem. Commun. (1984)
844.
[25] J.P. Neilan, R.M. Laine, N. Cortese, R.F. Heck, J. Org. Chem. 41 (1976) 3455.
[26] S.J. Thomson, G. Webb, J. Chem. Soc. Chem. Commun. (1976) 526.
[27] S. Rajagopal, A.F. Spatola, J. Org. Chem. 60 (1995) 1347.
[28] H. Wiener, J. Blum, Y. Sasson, J. Org. Chem. 56 (1991) 6145.
[29] H. Wiener, J. Blum, Y. Sasson, J. Org. Chem. 56 (1991) 4481.
[30] B. Devi, M.S.L. Karuna, K.N. Rao, P.S. Saiprasad, R.B.N. Prasad, J. Am. Oil Chem.
Soc. 80 (2003) 1003.
[31] Y.M. Ma, X.Y. Wei, X. Zhou, K.Y. Cai, Y.L. Peng, R.L. Xie, Y. Zong, Y.B. Wei, Z.M.
Zong, Energy Fuels 23 (2009) 638.
[32] A. Sharma, V. Kumar, A.K. Sinha, Adv. Synth. Catal. 348 (2006) 354.
[33] B.K. Banik, K.J. Barakat, D.R. Wagle, M.S. Manhas, A.K. Bose, J. Org. Chem. 64
(1999) 5746.
[34] H. Wiener, Y. Sasson, J. Blum, J. Mol. Catal. 35 (1986) 277.
[35] A. Fujii, S. Hashiguchi, N. Uematsu, T. Ikariya, R. Noyori, J. Am. Chem. Soc. 118
(1996) 2521.
[36] K. Kindler, K. Luhrs, Justus Liebigs Ann. Chem. 685 (1965) 36.
[37] D.J. Morris, G.J. Clarkson, M. Wills, Organometallics 28 (2009) 4133.
[38] C. Fellay, N. Yan, P.J. Dyson, G. Laurenczy, Chem. Eur. J. 15 (2009) 3752.
[39] C. Fellay, P.J. Dyson, G. Laurenczy, Angew. Chem. Int. Ed. 47 (2008) 3966.
[40] B. Loges, A. Boddien, H. Junge, M. Beller, Angew. Chem. Int. Ed. 47 (2008) 3962.
[41] M.K. Anwer, D.B. Sherman, J.G. Roney, A.F. Spatola, J. Org. Chem. 54 (1989) 1284.
[42] M.K. Anwer, A.F. Spatola, Tetrahedron Lett. 26 (1985) 1381.
[43] M.K. Anwer, A.F. Spatola, Synthesis (1980) 929.
Acknowledgements
The authors gratefully acknowledge the financial support from
the One Hundred Talent Program and KJCX2,YW.H16 from CAS,
the NSFC 21273222. This work was also supported by the CAS-JSPS
international joint project GJHZ05.
[44] G.D. Yadav, B.G. Motirale, Org. Process. Res. Dev. 13 (2009) 341.
[45] R. Breslow, Acc. Chem. Res. 24 (1991) 159.