A Ru-Sn-Co/AlO(OH) as a highly efficient catalyst for hydrogenation of dimethyl adipate to 1,6-hexanodiol in aqueous phase

Article abstract of DOI:10.1016/j.apcata.2012.09.041

A Ru-Sn-Co/AlO(OH) catalyst was prepared by co-impregnation, calcination, and hydrothermal reduction. The catalyst was well characterized by BET, XRD, SEM-EDX, XPS, H₂-TPR, and M?ssbauer spectroscopy, and then it was applied in the hydrogenation of dimethyl adipate to 1,6-hexanodiol in aqueous phase. It was found that the addition of cobalt could stabilize Sn(IV) and the cooperation between tin and cobalt could promote the reduction of ruthenium oxide. The γ-Al₂O₃ in the catalyst Ru-Sn-Co/γ-Al₂O₃ was transformed to AlO(OH) by the hydrothermal reduction. It was suggested that the stabilized Sn(IV) and the interaction between the hydroxyl group on the surface of AlO(OH) and the solvent water played a key role to improve the conversion of dimethyl adipate and the selectivity to 1,6-hexanodiol. The conversion of dimethyl adipate and the selectivity to 1,6-hexanodiol were up to 98% and 95%, respectively, under 5 MPa of H₂ and 493 K for 10 h.

Full text of DOI:10.1016/j.apcata.2012.09.041

Applied Catalysis A: General 447–448 (2012) 164–170
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
A Ru-Sn-Co/AlO(OH) as a highly efficient catalyst for hydrogenation of
dimethyl adipate to 1,6-hexanodiol in aqueous phase
Hong-Bin Jiang, Hai-Jun Jiang, Ke Su, De-Ming Zhu, Xue-Li Zheng, Hai-Yan Fu, Hua Chen,
Rui-Xiang Li∗
Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu, Sichuan 610064, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
A Ru-Sn-Co/AlO(OH) catalyst was prepared by co-impregnation, calcination, and hydrothermal reduction.
The catalyst was well characterized by BET, XRD, SEM-EDX, XPS, H2-TPR, and Mössbauer spectroscopy,
and then it was applied in the hydrogenation of dimethyl adipate to 1,6-hexanodiol in aqueous phase. It
was found that the addition of cobalt could stabilize Sn(IV) and the cooperation between tin and cobalt
could promote the reduction of ruthenium oxide. The ␥-Al2O3 in the catalyst Ru-Sn-Co/␥-Al2O3 was
transformed to AlO(OH) by the hydrothermal reduction. It was suggested that the stabilized Sn(IV) and
the interaction between the hydroxyl group on the surface of AlO(OH) and the solvent water played a key
role to improve the conversion of dimethyl adipate and the selectivity to 1,6-hexanodiol. The conversion
of dimethyl adipate and the selectivity to 1,6-hexanodiol were up to 98% and 95%, respectively, under
5 MPa of H2 and 493 K for 10 h.
Received 12 July 2012
Received in revised form
13 September 2012
Accepted 14 September 2012
Available online 9 October 2012
Keywords:
Ruthenium
Tin
Cobalt
© 2012 Elsevier B.V. All rights reserved.
Dimethyl adipate
1,6-hexanediol
Hydrogenation
1. Introduction
have been investigated to carry out the reaction in mild reaction
conditions [9–11]. Carvalho and Fraga’s group [12–15] has inves-
Diols are industrially important as they are used as raw
materials for a wide variety of fine chemicals, pharmaceuticals
and biodegradable polymers. Especially, 1,6-hexanediol is widely
applied in the synthesis of polyurethanes, varnishes, adhesives,
pharmaceuticals, plasticizers and so on [1]. Currently, the diols are
mainly produced by hydrogenating carboxylic acids or their esters.
The catalytic hydrogenation in both homogeneous and heteroge-
neous systems is one of the most significant synthetic methods.
In homogeneous systems, ruthenium–phosphine complexes have
been widely used as catalysts for ester hydrogenation [2,3]. How-
ever, it is a challenging problem to synthesize the organometallic
complexes and separate catalysts from products. In addition, it is
still difficult to avoid the catalyst being poisoned by carbon monox-
ide, which is generated by the decarbonylation of the carboxylic
acids or esters [4,5]. In heterogeneous system, this reaction is tra-
ditionally performed over the copper-chromium or zinc-chromium
catalysts under the severe reaction conditions (reaction tempera-
ture of 523–623 K and H2 pressure of 10–20 MPa) [6–8]. The use
of Cr is currently discouraged and severe reaction conditions are
undesirable, so many noble metallic (Ru, Pd, Pt and Rh) catalysts
tigated the activity and selectivity about monometallic catalysts
of Ru, Pd, Pt and Rh supported on alumina for the hydrogena-
tion of dimethyl adipate (DMA) to 1,6-hexanediol. Among these
monometallic catalysts, ruthenium-based catalyst showed a higher
activity, reaching about 80% conversion of DMA and 49% selectiv-
ity to 1,6-hexanediol at 528 K under 5 MPa of H2 [12]. Although
the hydrogenation conditions were much milder than that over
complex oxide catalysts, they gave a low selectivity to alcohol
[12,16]. Therefore, a second metal such as Sn, Re, Ge, or Zn had
to be introduced to modify the catalytic properties [11,17]. It was
found that Ru-Sn catalyst was the most effective for the hydrogena-
tion of DMA [17–19], likewise, ethyl lactate [16], adipic acid [1] and
1,4-cyclohexanedicarboxylic acid [11]. The addition of a second ele-
ment Sn obviously improved the 1,6-hexanediol selectivity up to
70% over Ru-Sn/TiO2 with Sn/Ru ratio of 2 [17].
Co, as a non-noble metal, is extensively used for catalytic
hydrogenation. For example, Coq et al. had demonstrated that Co
had a promising potential for acrolein hydrogenation [20]. It was
reported that the superficial structure of Co supported on silica
could influence the hydrogenation of C O bond in crotonalde-
hyde to crotyl alcohol [21–24]. Similarly, Kouachi et al. reported
the hydrogenation of citral in supported cobalt systems [25]. These
results also showed that Co particles did favor the hydrogenation
of the C O bond. In our previous works [5,26], it was observed
∗
Corresponding author. Tel.: +86 028 85412904; fax: +86 028 85412904.
E-mail address: liruixiang@scu.edu.cn (R.-X. Li).
0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2012.09.041

H.-B. Jiang et al. / Applied Catalysis A: General 447–448 (2012) 164–170
165
Table 1
that ␥-Al2O3 was transformed into AlO(OH) with a good crys-
The textural properties of different catalysts.
tallization under the hydrothermal condition. In the presence of
water, Ru-Pt/AlO(OH) showed the high activity and selectivity for
the hydrogenation of methyl propionate to 1-propanol under the
mild reaction condition. The excellent performance of the Ru-
Pt/AlO(OH) catalyst was assigned to the cooperation between the
hydroxyl group of supporter surface and the solvent water.
In this paper, we further investigated the catalytic perfor-
mance of Ru-Sn-Co/AlO(OH) for the hydrogenation of DMA to
1,6-hexanediol. An unprecedentedly high DMA conversion of 98%
and the selectivity of 95% to 1,6-hexanediol were shown over it in
aqueous phase.
Catalyst
surface area Pore volume
average pore
diameter (nm)
(m2/g)
(cm3/g)
Ru-Sn-Co/AlO(OH) 1:1:5
Ru-Sn-Co/AlO(OH) 1:1:3.5
Ru-Sn-Co/AlO(OH) 1:1:1
Ru-Sn/AlO(OH) 1:1
148.0
149.5
142.8
52.4
0.2607
0.2612
0.2491
0.1358
3.524
3.495
3.488
5.188
2.3. Catalytic tests
Catalytic hydrogenation of dimethyl adipate (99%, Alfa Aesar)
was performed in a 60 ml stainless autoclave equipped with a
pressure gauge, a magnetic stirrer, and an electric temperature con-
troller. The catalyst (200.0 mg), the solvent (4.0 ml) and dimethyl
adipate (1.0 ml) were introduced into the reactor, respectively. The
autoclave was purged with H2 three times, pressurized with H2 to
5 MPa, and started to stir. After the designated temperature was
reached, the reaction time began. The products were analyzed by
gas chromatography (GC-6890) with a FID detector and a HP-5
(25 m × 0.53 mm) capillary column. The vent gas was detected with
Agilent GC-1790T equipped with a thermal conductivity detec-
tor (TCD) and a plot carbon sieve column (25 m × 0.53 mm), and
hydrogen was used as a carrier gas. Reactants and products were
identified by comparison with the standard samples and GC MS.
2. Experimental
2.1. Catalyst preparation
All of the chemicals (A.R.) were commercially obtained and
used as received. The ␥-Al2O3 was calcined at 773 K for 4 h prior
to use. All catalysts were prepared as the ruthenium content of
5 wt.% and tin content of 5 wt.%; while the cobalt content was
varied to achieve different Co/Ru ratio. Typically, 1.0 g of ␥-Al2O3
was dispersed in 20 ml aqueous solution containing a certain
amount of RuCl3·2H2O (Kunming Institute of Precious Metals,
China), SnCl2·2H2O (98%, Tianjin Bodi Chemical Co., Ltd., China) and
CoCl2·6H2O (99%, Chengdu Kelong Chemical Co., Ltd., China). The
mixture was stirred at room temperature for 24 h, and then the
solvent was slowly removed by evaporation under vacuum. The
resulted solid was dried at 318 K overnight and calcined at 673 K for
2 h. Finally, the catalysts were reduced in a 60 ml stainless autoclave
in aqueous phase under the hydrogen pressure of 5 MPa, separated
by centrifugation, and dried under vacuum at 318 K for 10 h.
3. Results and discussion
3.1. Catalyst characterization
Table 1 shows the compositional and textural properties of the
catalysts containing the various cobalt loadings. The solids sur-
face area and pore structure are obviously improved with the
cobalt introduction. The XRD results of Ru-Sn-Co/␥-Al2O3 catalyst
(Ru:Sn:Co = 1:1:3.5) are shown in Fig. 1. Before reduction (Fig. 1a),
the catalyst diffraction peaks at 19.6◦, 37.6◦, 45.8◦ and 66.8◦ is
attributed to ␥-Al2O3 [5]. XRD profile of the Ru-Sn-Co/␥-Al2O3,
reduced in ethanol under 5 MPa of H2 at 453 K for 5 h, indicates
that ␥-Al2O3 does not convert into AlO(OH) (Fig. 1b). When the
catalyst was reduced in H2O for 3 h (Fig. 1c), a group of new diffrac-
tion peaks at 14.4◦, 28.1◦, 38.3◦, 48.9◦, 55.3◦ and 64.2◦ appears,
which are assigned to AlO(OH), and their intensity increases with
the extension of reduction time (Fig. 1d). These results indicate
that ␥-Al2O3 has transformed into AlO(OH) with a good crystalliza-
tion state under the hydrothermal condition. The diffraction peaks
of metallic ruthenium, tin oxide, and cobalt oxide species are not
2.2. Catalyst characterization
The content of metals in the catalyst was determined by induc-
tively coupled plasma (ICP, IRIS Intrepid). The textural properties
of the catalysts were obtained on a Quantachrome SI instrument.
Prior to the measurement, all samples were degassed under vac-
uum at 573 K for 3 h. X-ray diffraction (XRD) was recorded with
Philiphs X’Pert MPD with Cu K␣ as the radiation source at the
voltage of 40 kV and the current of 25 mA, and the samples were
scanned in the range of 2Â = 10–80◦ at a rate of 0.1◦/s. The surface
topography was determined by scanning electronic microscope
FEI Inspect F at 20 kV. X-ray photoelectron spectra (XPS, Kratos
XSAM800) were obtained by Al K␣ radiation (12 kV and 15 mA).
All the binding energy (BE) values were referenced to the C1s peak
(284.6 eV) of the contaminant carbon. The background pressure of
the analyzing chamber was 6.7 × 10−7 Pa. 119Sn Mössbauer Spectra
of the catalysts and a standard SnO2 were recorded by a con-
ventional Mössbauer spectrometer (Topologic, Japan) working in
a constant acceleration mode in transmission geometry [27,28]
with a Ca119SnO3 as the ␥-rays source. The velocity calibration
was performed by ␣-Fe and standard CaSnO3 measurements. The
spectra were fitted with appropriate superpositions of Lorentzian
lines using the MossWinn computer program. Temperature pro-
grammed reduction (TPR) experiments were carried out in a
conventional system equipped with a thermal conductivity detec-
tor. All samples (100 mg) were pretreated in a quartz U-tube in a
flow of N2 (20 ml/min) at 573 K for 1 h, and then cooled down to
room temperature. The reduction was carried out in a flow of H2
(5%) in N2 (20 ml/min) from 310 to 1273 K with a linear heating rate
of 8 K/min.
Fig. 1. XRD patterns of for Ru-Sn-Co/Al2O3 reduced with different ways. (a) The
catalyst was not reduced; (b) the catalyst was reduced in ethanol with 5 MPa H2 at
453 K for 5 h; (c) the catalyst was reduced in water with 5 MPa H2 at 453 K for 3 h;
(d) the catalyst was reduced in water with 5 MPa H2 at 453 K for 5 h.

166
H.-B. Jiang et al. / Applied Catalysis A: General 447–448 (2012) 164–170
Fig. 2. SEM images of Ru-Sn-Co/Al2O3 reduced in water for 5 h (a). EDX elemental maps Co, Ru and Sn (b–d).
observed due to the poor crystallinity and small metal particles over
the support. In addition, Fig. 2a shows the surface topography of the
Ru-Sn-Co/AlO(OH) (Ru:Sn:Co = 1:1:3.5) catalyst which was reduced
in water with 5 MPa of H2 at 453 K for 5 h. It explicitly revealed the
microscale flake-like boehmite. The EDX elemental map in Fig. 2b–d
clearly presents the well dispersion of the metal particles over the
supports, which is in accordance with the XRD pattern in Fig. 1.
And the molar ratio of Ru: Sn: Co was approximately 1:1:3.5 in the
representative sample with the EDX analysis.
Fig. 3 shows the XPS pattern of Ru 3d, Sn 3d, and Co 2p in Ru-
Sn-Co/Al2O3 (Ru:Sn:Co = 1:1:3.5) with different reduction time. As
the Ru 3d3/2 peak overlaps with that of C1s on contaminant car-
bon, Ru 3d5/2 peak is employed to determine the oxidation state of
Ru. As demonstrated in Fig. 3I, the binding energy values at 280 eV
and 281.6 eV represent Ru(0) and Ru(III), respectively [16,29]. Com-
pared curve b with curve a in Fig. 3I, it recovers that Ru(III) in
the Ru-Sn-Co system has been partly reduced to Ru(0) through
hydrothermal reduction with 5 MPa of H2 at 453 K for 3 h. Accord-
ing to the intensity ratio of the binding energy of Ru(0) to that of
Ru(III), the contents of Ru(0) and Ru(III) are about 45.8% and 54.2%,
individually. Extending the reduction time to 5 h, Ru(III) was
reduced completely and only Ru(0) was detected (Fig. 3Ic). How-
ever, the XPS pattern of Sn 3d and Co 2p shows that both Sn (BE
486.6–486.3 eV) and Co (BE 781.4–781.1 eV) species are maintained
as the oxidized species, and their binding energy values have little
decrease with the extension of reduction time. It indicates that the
low oxidation state species of Sn and Co are not formed at the low
reduction temperature of 453 K.
To study the reduction behavior of the catalysts, the 119Sn
Mössbauer spectra and the temperature programmed reduction
(TPR) experiment were conducted. Fig. 4 presents the 119Sn Möss-
bauer spectra of the catalysts with different cobalt content. These
results show that the Sn(IV) species is the dominant compo-
nent in the catalysts after being reduced in water with 5 MPa of
H2 at 453 K for 5 h. Comparing the spectral area of Sn(IV) with
that of Ru-Sn oxide species [9,17], it can be concluded that the
contents of Sn(IV) are about 87.6% in the catalyst with the molar
ratio of Ru:Sn:Co = 1:1:5 and 65.7% with Ru:Sn = 1:1, respectively.
Especially, it is found that the Sn(IV) formation is preferable at
Ru:Sn:Co = 1:1:3.5. Obviously, the appropriate molar ratio of Sn
Fig. 3. XPS patterns of Ru 3d,Sn 3d and Co2p for Ru-Sn-Co/Al2O3 reduced with different time. (a) The catalyst was not reduced; (b) the catalyst was reduced in water with
5 MPa H2 at 453 K for 3 h; (c) the catalyst was reduced in water with 5 MPa H2 at 453 K for 5 h.

H.-B. Jiang et al. / Applied Catalysis A: General 447–448 (2012) 164–170
167
100
90
80
70
60
50
40
30
20
10
0
conv[%]
sele [%]
1:1:5 1:1:4.5 1:1:4 1:1:3.5 1:1:3
1:1:2
1:1:1
1:1:0
mole ratio(Ru/Sn/Co)
Fig. 4. 119Sn Mössbauer spectra of Ru-Sn-Co/AlO(OH) with different content cobalt.
(a) Ru:Sn = 1:1; (b) Ru:Sn:Co = 1:1:3.5; (c) Ru:Sn:Co = 1:1:5.
Fig. 6. The effects of cobalt content in hydrogenation DMA.
hydrogenation of DMA to 1,6-hexanodiol over Ru-based catalysts.
A lot of research works have revealed that tin is the best modifier
among Ru-based catalysts for the carboxylic group hydrogenation
to the corresponding alcohol [16]. Santos and co-workers regarded
the introduction of Sn was a rather important factor for produc-
ing diol [12,17]. The selectivity to 1,6-hexanediol could reach 70%
at the 57% conversion of DMA over Ru-Sn/TiO2 [17]. In our case, a
similar result was obtained without the addition of cobalt. How-
ever, the introduction of the appropriate amount of cobalt into
Ru-Sn/␥-Al2O3 greatly promotes the hydrogenation of DMA to 1,6-
hexanodiol. An unprecedentedly high conversion and selectivity
were given over the Ru-Sn-Co/␥-Al2O3 (Ru:Sn:Co = 1:1:3.5) cata-
lyst. According to the textural property analysis of the catalysts,
the excellently catalytic performance is partially from their textural
property improvement. Furthermore, from the Mössbauer spec-
troscopy in Fig. 4, after Ru-Sn-Co/␥-Al2O3 (Ru:Sn:Co = 1:1:3.5) was
reduced, the oxidation state of tin is completely maintained in
Sn(IV) and no Ru-Sn oxide species is observed. While some Ru-Sn
oxide species appear in the other catalysts. Obviously, the introduc-
tion of the appropriate amount cobalt can stabilize Sn(IV). From the
TPR profile in the Fig. 5d, it can be seen that the synergic effect of Sn
and Co species is aroused. A broad peak of hydrogen consumption
in Ru-Sn-Co system (Ru:Sn:Co = 1:1:3.5) indicates that lots of the
low oxidation state active sites are formed at 434 K. In combination
with hydrogenation results, the Ru-Sn oxide species formation and
their content increase are not in favor to the catalysts activity and
selectivity improvement to the desirable product 1,6-hexanodiol.
And the introduction of a large amount of cobalt results in the cat-
alyst activity decrease owing to the formation of Ru-Sn oxide and
the active sites being covered by the excessive cobalt. The contribu-
tion of tin to the catalytic activity of ruthenium also depends on the
cobalt content. In other words, the existence of appropriate amount
of cobalt can inhibit the generation of the Ru-Sn oxide species and
stabilize Sn(IV). The Sn(IV), as a strong Lewis acid, can strongly act
with base oxygen atom of carboxylic group, so that it effectively
activates the C O bond in carboxylic group [37], facilitating the
hydrogen transfer from the adjacent ruthenium sites.
to Co is significant to stabilize Sn(IV) species. The TPR profiles of
bimetallic systems and trimetallic system are shown in Fig. 5. It is
well known that the peak area is a close correlation with the forma-
tion amount of the reductive species [30]. The peaks at 424–434 K
are assigned to the reduction peaks of ruthenium oxide [19,31].
The peak area, representing the hydrogenation consumption, is
in the order of Ru-Sn-Co/␥-Al2O3 (Ru:Sn:Co = 1:1:3.5) > Ru-Co/␥-
Al2O3 > Ru-Sn/␥-Al2O3 > Sn-Co/␥-Al2O3. The reduction peaks at
611–657 K are attributed to the reduction of tin oxide [32,33] and
cobalt oxide [34,35]. If the cobalt loading is further increased to
Ru:Sn:Co = 1:1:5 (Fig. 5e), the reduction of the ruthenium oxide
becomes difficult. A new reduction peak at 815 K indicates the pres-
ence of the strong interactions between cobalt oxide and the Al2O3
[35,36].
3.2. Hydrogenation of DMA
3.2.1. The effects of cobalt contents on the hydrogenation of DMA
Fig. 6 collects the hydrogenation results of DMA over the Ru-
Sn-Co/AlO(OH) with different cobalt contents. According to the
results in Fig. 6, both the conversion of DMA and selectivity to diol
increased with the increase of the Co content when the molar ratio
of Co to Ru was lower than 3.5:1. On the contrary, when the ratio
of Co to Ru was higher than 3.5:1, the conversion and the selectiv-
ity decreased with the increase of the Co content. The maximum
conversion of DMA was up to 98% and the selectivity to 1,6-
hexanodiol was over 95% in the molar ratio of Co:Sn:Ru = 3.5:1:1.
To the best of our knowledge, this is the best result for the
3.2.2. The effects of solvent on the hydrogenation of DMA
In our previous work about the hydrogenation of methyl pro-
pionate over Ru-Pt/␥-Al2O3 [5], it was found that the solvents, in
which the catalyst Ru-Pt/␥-Al2O3 was reduced and hydrogenation
was performed, had the dramatical influence on the catalyst per-
formance. Therefore, the effects of solvents on this hydrogenation
system were also investigated for the hydrogenation of DMA to 1,6-
diol and the results are shown in Table 2. It is interesting that the
catalyst dealt with water shows the higher catalytic activity and
selectivity than that disposed in other solvents. When the catalyst
Fig. 5. The TPR profile of different catalysts for: (a) Sn-Co/Al2O3 (1:3.5); (b)
Ru-Sn/Al2O3 (1:1); (c) Ru-Co/Al2O3 (1:3.5); (d) Ru-Sn-Co/Al2O3 (1:1:3.5) (e) Ru-Sn-
Co/Al2O3 (1:1:5).

168
H.-B. Jiang et al. / Applied Catalysis A: General 447–448 (2012) 164–170
Table 2
The effects of solvent on DMA hydrogenation.
Test
Reduction solvent
Reaction solvent
Conv (%)
Sele (%)
Diol
1-hexanol
OXE
HCA
MME
EMA
DEA
1
2
3
4
5
6
7
8
9
Water
Water
Water
Methanol
Methanol
Ethanol
Ethanol
1,4-dioxane
1,5-dioxane
Water
98.0
76.1
91.6
70.0
48.0
93.7
47.6
30.1
58.7
95.0
8.3
–
18.3
52.8
2.6
43.6
–
3.0
2.6
4.8
2.5
–
0.9
–
2.6
3.8
–
89.1
–
70.5
37.6
5.4
38.8
21.0
37.9
2.0
–
–
8.7
9.6
–
17.6
34.1
27.3
–
–
11.2
–
–
12.8
–
–
–
40.5
–
–
33.8
–
–
–
43.5
-
–
44.5
–
Methanol
Ethanol
Methanol
Water
Ethanol
Water
1,4-dioxane
Water
42.3
–
–
–
–
–
31.0
Diol: 1,6-hexanediol; OXE: 2,7-oxepanedione; HCA: 6-hydroxyl caproic acid; MME: adipic acid monomethyl ester; EMA: ethyl-methyl adipate; DEA: Diethyl adipate; reduction
time: 5 h; reaction time: 10 h; reduction temperature: 453 K; reaction temperature: 493 K; solvent: 4 ml; H2 pressure: 5.0 MPa
was reduced in water and the hydrogenation was also performed in
water, both the high conversion of 98% and selectivity of 95% were
given (Test 1). On the contrary, a low conversion and selectivity
were obtained if the catalyst precursor was reduced and hydro-
genation in organic solvent (Test 4 and 8). Especially, when the
hydrogenation was carried out in ethanol, the ethyl-methyl adipate
(EMA) and diethyl adipate (DEA) were predominantly detected in
the reaction products due to transesterification between substrate
and ethanol (Test 3 and 6). If the catalyst was reduced in organic
solvent and the hydrogenation was performed in aqueous phase,
the DMA conversion and the diol selectivity were still undesirable
(Test 5, 7 and 9). Hence, water plays a key role in the reduction
of catalyst as well as in hydrogenation process. On the one hand,
the ␥-Al2O3 was transformed into AlO(OH) in this hydrothermal
condition and many researchers have demonstrated that the Al
O
double layers of bohmite AlO(OH) are interconnected by hydrogen
bonds between the hydroxyl groups [38–40], so a lot of hydroxyl
groups are present on the surface of the AlO(OH). It is suggested that
the surface hydroxyl groups of AlO(OH) can interact with carboxylic
group in DMA by hydrogen bonds to promote the adsorption and
activation of the carboxylic group. On the other hand, the hydro-
gen bond between water as solvent and H2 adsorbed on active
sites can be formed and it is favorable for the heterolytic splitting
of H2 to get a ruthenium–hydride complex and protonated water
[41,42]. Therefore, the catalytic performance of Ru-Sn-Co/AlO(OH)
is further enhanced in aqueous phase.
Fig. 7. The kinetic curves for DMA hydrogenation over Ru-Sn-Co/AlO(OH) (1:1:3.5)
catalyst.
in water at a low temperature of 413 K and 433 K, respectively, the
catalyst not only gave a relative low selectivity to 1,6-hexanodiol
(Test 1 and 2) due to the formation of intermediate products 2,7-
oxepanedione (OXE), 6-hydroxyl caproic acid (HCA), and adipic acid
monomethyl ester (MME), but also showed a low conversion at the
reaction temperature of 493 K. The MME is formed by either hydro-
genation or hydrolysis of DMA. The existence of methane in the vent
gas proves that MME is partially generated by the hydrogenation of
DMA. If the hydrogenation temperature is kept and the reduction
temperature of the catalyst is raised to 453 K, a high conversion of
98% and high selectivity of 95% to diol are obtained (Test 3). Accord-
ing to the TPR and XRD profiles, when the reduction temperature is
3.2.3. The effects of temperature and time on the hydrogenation
of DMA
Table 3 collects the effects of the temperature and time on the
hydrogenation of DMA. When the catalyst precursor was reduced
Table 3
The effects of the temperature and time on DMA hydrogenation.
Test
Aa (K)
Bb (K)
Cc (h)
Dd (h)
Conv (%)
Sele (%)
Hexanediol
1-hexanol
OXE
HCA
MME
1
2
3
4
5
6
7
8
9
413
433
453
453
453
453
453
453
453
453
493
493
493
473
453
433
493
493
493
493
5
5
5
5
5
5
3
4
5
5
10
44.0
66.2
98.0
77.1
71.4
68.1
47.0
57.0
60.0
75.0
60.0
87.4
95.0
18.9
–
–
–
3.0
2.0
–
–
–
2.6
–
2.7
17.0
12.6
–
12.7
15.0
5.5
22.4
24.1
20.4
16.2
12.0
–
2.0
31.9
56.0
61.9
14.6
29.1
50.8
39.2
11.0
–
–
34.5
29.0
32.6
21.0
23.2
25.7
18.8
10
10
10
10
10
10
10
6
–
42.0
21.0
3.1
10
8
23.1
Diol: 1,6-hexanediol; OXE: 2,7-oxepanedione; HCA: 6-hydroxyl caproic acid; MME: adipic acid monomethyl ester; solvent: 4 ml H2O; H2 pressure: 5.0 Mpa.
a
Reduction temperature.
Reaction temperature.
Reduction time.
Reaction time.
b
c
d

H.-B. Jiang et al. / Applied Catalysis A: General 447–448 (2012) 164–170
169
Fig. 8. A simplified reaction routes for hydrogenation of dimethyl adipate.
lower than 434 K, the low oxidation state active ruthenium species
is difficult to be formed, so that the intermediates cannot convert
completely into the desired product. Similarly, if the catalyst was
fully reduced at 453 K and the hydrogenation was performed at a
low temperature for 10 h (Test 4, 5 and 6), both the catalyst activity
and selectivity were low as a great deal of intermediate products
did not transfer to diol. These results recovered that the hydro-
genation processing is step by step and the activation of substrates
can be achieved in a higher temperature. When the reduction tem-
perature is still kept at 453 K, the shorter the reduction time is,
the lower the catalyst activity is (Test 7 and 8). When the catalyst
reduction time was further extended to 5 h, the selectivity to diol
was up to 95% with almost the complete DMA conversion (Test 3).
The results show that a lot of low oxidation state active species are
formed in the enough reduction time. Similarly, if the catalyst was
fully reduced and the hydrogenation time was shorten, a low selec-
tivity to 1,6-hexanodiol was given due to the presence of the great
amount of intermediates. The relationship curves of the conversion
and selectivity vs reaction time are shown in Fig. 7. HCA was pro-
duced largely in the first 6 h and then its concentration begins to
decrease with the possible formations of diol and OXE. HCA can be
an intermediate product of DMA, where it is generated and rapidly
converted into other reaction products. It is also seen from Test 3, 4,
8 and 10 that 1-hexanol presented marginally due to dealcoholiza-
tion of 1,6-hexanodiol and partly decarboxylation of HCA. So the
reasonable hydrogenation routes are shown in Fig. 8.
Fig. 9. The proposed mechanism.
support. The hydrogen bond not only promoted the adsorption of
DMA on the catalyst surface, but also polarized the C O bond in
carboxyl group [5]. Due to the hydrogen bond role of water [41],
the H2 adsorbed on Ru(0) center undergoes heterolytic splitting to
get a ruthenium–hydride complex and protonated water. Subse-
quently, the hydride on ruthenium attacks the electron-deficient
carbon in carboxyl group to form the intermediate aldehyde which
is quickly hydrogenated into alcohol in this condition. The proposed
mechanism is shown in Fig. 9.
3.2.4. Suggested mechanism
4. Conclusions
The TPR profiles and XPS patterns show that the existence of
Co and Sn promotes the reduction of ruthenium oxide to Ru(0) as
the active species, and the Mössbauer spectra date indicates that
Co can stabilize Sn(IV) and inhibit the reduction of Sn(IV). Sn(IV),
as the high deficiency electron center, could strongly adsorb the
oxygen atom of C O in ester group and effectively polarize C O in
the carboxylic group [43–45], contributing to the addition reaction
between C O and hydrogen atoms which are from the split H2. So
cobalt and tin are the key modifiers in this catalyst for the hydro-
genation of DMA to 1,6-hexanedol. Undoubtedly, besides the action
of active metal species, the cooperation of water as solvent and
the hydroxyl group on the AlO(OH) surface also play an important
role in promoting the catalyst activity and selectivity. According
to the hydrogenation results of methyl propionate [5] and quino-
line [46], the surface hydroxyl group of the support can form the
hydrogen bond with oxygen atom of carboxylic group and nitrogen
atom in quinoline. Similarly, we suggest that the oxygen atom in
carboxylic group of DMA also forms the hydrogen bond with the
A Ru-Sn-Co/AlO(OH) catalyst was prepared, characterized, and
applied in the hydrogenation of dimethyl adipate to 1,6-hexanodiol
in aqueous phase. It was demonstrated that the Ru-Sn-Co/AlO(OH)
catalyst has highly catalytic activity and selectivity for the hydro-
genation of dimethyl adipate to 1,6-hexanodiol. The conversion and
selectivity improvement is attributed to the fact that Ru(0), inter-
acted with the tin and cobalt oxides, which are well dispersed over
the support, and the cooperation between the hydroxyl group on
the catalyst surface and the water as solvent.
Acknowledgements
We are grateful for the financial support by National Natural Sci-
ence Foundation of China (NSFC No. 21072138) and the Innovation
Foundation of Sichuan University.

170
H.-B. Jiang et al. / Applied Catalysis A: General 447–448 (2012) 164–170
References
[24] F. Djerboua, D. Benachour, R. Touroude, Appl. Catal., A 282 (2005) 123–133.
[25] K. Kouachi, G. Lafaye, C. Especel, O. Cherifi, P. Marécot, J. Mol. Catal. A 280 (2008)
52–60.
[26] Y.F. Zhou, J. Wei, G.Y. Fan, H.Y. Fu, R.X. Li, H. Chen, X.J. Li, Chem. Lett. 38 (2009)
1034–1035.
[27] Z. Klencsar, E. Kuzmann, A. Vertes, J. Radioanal. Nucl. Chem. 210 (1996)
105–118.
[28] M. Ji, X.F. Li, J.H. Wang, J.M. Zhu, L.M. Zhou, Catal. Today 173 (2011) 28–31.
[29] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray Photo-
electron Spectroscopy, PerkinElmer Corp., Norwalk,CT, 1992.
[30] Z.M. Liu, J.L. Wang, J.B. Zhong, Y.Q. Chen, S.H. Yan, M.C. Gong, J. Hazard. Mater.
149 (2007) 742–746.
[31] F.C.A. Figueiredo, E. Jordão, R. Landers, W.A. Carvalho, Appl. Catal., A 371 (2009)
131–141.
[32] E.A. Alarcón, L. Correa, C. Montes, A.L. Villa, Microporous Mesoporous Mater.
136 (2010) 59–67.
[33] Y.W. Chang, C.W. Liu, Y.C. Wei, K.W. Wang, Electrochem. Commun. 11 (2009)
2161–2164.
[34] A.Y. Khodakov, W. Chu, P. Fongarland, Chem. Rev. 107 (2007) 1692–1744.
[35] G. Jacobs, T.K. Das, Y.Q. Zhang, J.L. Li, G. Racoillet, B.H. Davis, Appl. Catal., A 233
(2002) 263–281.
[36] P. Arnoldy, J.A. Moulijn, J. Catal. 93 (1985) 38–54.
[37] T. Miyake, T. Makino, S. Taniguchi, H. Watanuki, T. Niki, S. Shimizu, Y. Kojima,
M. Sano, Appl. Catal., A 364 (2009) 108–112.
[38] R. Xalter, T.S. Halbach, R. Mu˝ lhaupt, Macromol. Symp. 236 (2006) 145–150.
[39] S.C. Shen, Q. Chen, P.S. Chow, G.H. Tan, X.T. Zeng, Z. Wang, R.B.H. Tan, J. Phys.
Chem. 111 (2007) 700–707.
[40] K. Okada, T. Nagashima, Y. Kameshima, A. Yasumori, T. Tsukada, J. Colloid Inter-
face Sci. 253 (2002) 308–314.
[41] K.A. Rashid, S.E. Clapham, A. Hadzovic, J.N. Harvey, A.J. Lough, R.H. Morris, J.
Am. Chem. Soc. 124 (2002) 15104–15118.
[42] S.E. Clapham, A. Hadzovic, R.H. Morris, Coord. Chem. Rev. 248 (2004)
2201–2237.
[43] P. Kluson, PhD Thesis, Inst. Chem. Technol., Prague, 1994.
[44] P. Kluson, L. Cerveny, Appl. Catal., A 128 (1995) 13–31.
[45] P. Louessard, J.P. Candy, J.P. Bourneville, J.M. Basset, Structure and Reactivity of
Surface, Elsevier, Amsterdam, 1989, p. 591.
[1] M. Toba, S. Tanka, S. Niwa, F. Mizukami, Z. Koppány, L. Guczi, K. Cheah, T.S. Tang,
Appl. Catal., A 189 (1999) 243–250.
[2] H. Imai, T. Nishiguchi, K. Fukuzumi, J. Org. Chem. 41 (1976) 665–671.
[3] F. Ozawa, I. Yamagami, A. Yamamoto, J. Organomet. Chem. 473 (1994) 265–272.
[4] G.Y. Fan, Y.F. Zhou, H.Y. Fu, X.L. Ye, R.X. Li, H. Chen, X.J. Li, Chin. J. Chem. 29
(2011) 229–236.
[5] Y.F. Zhou, H.Y. Fu, X.J. Zheng, R.X. Li, H. Chen, X.J. Li, Catal. Commun. 11 (2009)
137–141.
[6] T. Turek, D.L. Trimm, N.W. Cant, Catal. Rev. Sci. Eng. 36 (1994) 645–683.
[7] D.S. Brands, G. U-A-Sai, E.K. Poels, A. Blik, J. Catal. 186 (1999) 169–180.
[8] K. Fujitani, Y. Hayashi, EP 1566372 (2005).
[9] S.M. Santos, A.M. Silva, E. Jordão, M.A. Fraga, Catal. Today 107–108 (2005)
250–257.
[10] R.M. Deshpande, V.V. Buwa, C.V. Rode, R.V. Chaudhari, P.L. Mills, Catal. Com-
mun. 3 (2002) 269–274.
[11] Y. Hara, K. Endou, Appl. Catal., A 239 (2003) 181–195.
[12] S.M. Santos, A.M. Silva, E. Jordão, M.A. Fraga, Catal. Commun. 5 (2004) 377–381.
[13] F.C.A. Figueiredo, E. Jordão, W.A. Carvalho, Catal. Today 107–108 (2005)
223–229.
[14] A.M. Silva, O.A.A. Santos, M.A. Morales, E.M. Baggio-Saitovitch, E. Jordão, M.A.
Fraga, J. Mol. Catal. A 253 (2006) 62–69.
[15] F.C.A. Figueiredo, E. Jordão, W.A. Carvalho, Appl. Catal., A 351 (2008) 259–266.
[16] G. Luo, S.R. Yan, M.H. Qiao, J.H. Zhang, K.N. Fan, Appl. Catal., A 275 (2004)
95–102.
[17] A.M. Silva, M.A. Morales, E.M. Baggio-Saitovitch, E. Jordão, M.A. Fraga, Appl.
Catal., A 353 (2009) 101–106.
[18] J. Fontana, C. Vignado, E. Jordão, W.A. Carvalho, Chem. Eng. J. 165 (2010)
336–346.
[19] J. Fontana, C. Vignado, E. Jordao, F.C.A. Figueiredo, W.A. Carvalho, Catal. Today
172 (2011) 27–33.
[20] B. Coq, F. Figueras, P. Geneste, C. Moreau, P. Moreau, M. Warawdekar, J. Mol.
Catal. 78 (1993) 211–226.
[21] J.E. Bailie, G.J. Hutchings, H.A. Abdullah, J.A. Anderson, C.H. Rochester, Phys.
Chem. Chem. Phys. 2 (2000) 283–290.
[22] E.L. Rodrigues, C.E.C. Rodrigues, A.J. Marchi, C.R. Apesteguia, J.M.C. Bueno, Stud.
Surf. Sci. Catal. 130 (2000) 2087–2092.
[23] E.L. Rodrigues, J.M.C. Bueno, Appl. Catal., A 257 (2004) 201–211.
[46] Y.P. Sun, H.Y. Fu, D.L. Zhang, R.X. Li, H. Chen, X.J. Li, Catal. Commun. 12 (2010)
188–192.