Hydrogenolysis of biomass-derived sorbitol to glycols and glycerol over Ni-MgO catalysts

Article abstract of DOI:10.1016/j.catcom.2013.05.012

Ni-MgO catalysts with varying Ni/Mg ratios were prepared by co-precipitation and tested in sorbitol hydrogenolysis. At 473 K and 4 MPa H₂, the best catalyst with Ni/Mg ratio of 3:7 exhibited 67.8% conversion and 80.8% total selectivity of ethylene glycol, 1,2-propylene glycol and glycerol. These catalysts were characterized by XRD, CO₂-TPD and H₂ chemisorption, revealing that the activity depended strongly on the basicity and Ni surface area. The reaction conditions were optimized, which were relatively mild for this chemoselective conversion.

Full text of DOI:10.1016/j.catcom.2013.05.012

Catalysis Communications 39 (2013) 86–89
Contents lists available at SciVerse ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Hydrogenolysis of biomass-derived sorbitol to glycols and glycerol over
Ni-MgO catalysts
Xinguo Chen a,b, Xicheng Wang a, Shengxi Yao a, Xindong Mu a,
⁎
a
Key Laboratory of Biobased Materials, Chinese Academy of Sciences, Qingdao 266101, PR China
b
University of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Ni-MgO catalysts with varying Ni/Mg ratios were prepared by co-precipitation and tested in sorbitol
hydrogenolysis. At 473 K and 4 MPa H2, the best catalyst with Ni/Mg ratio of 3:7 exhibited 67.8% conversion
and 80.8% total selectivity of ethylene glycol, 1,2-propylene glycol and glycerol. These catalysts were character-
ized by XRD, CO2-TPD and H2 chemisorption, revealing that the activity depended strongly on the basicity and
Ni surface area. The reaction conditions were optimized, which were relatively mild for this chemoselective
conversion.
Received 14 March 2013
Received in revised form 10 May 2013
Accepted 15 May 2013
Available online 21 May 2013
Keywords:
Sorbitol
© 2013 Elsevier B.V. All rights reserved.
Hydrogenolysis
Glycols
Ni-MgO
Basic carrier
1. Introduction
Herein, the bifunctional Ni-MgO catalysts were prepared by co-
precipitation and used for sorbitol hydrogenolysis under mild condi-
Glycerol and C2–C3 glycols containing ethylene glycol (EG) and
propylene glycol (PG) are important commodity chemicals widely
used in the manufacture of polyesters resins, surfactants, pharmaceu-
ticals and functional fluids. Sugars and sugar alcohols, which can be
derived largely from the renewable lignocellulosic biomass instead
of fossil resources, are considered to be potential feedstocks for the
production of lower glycols through hydrogenolysis because of their
rich oxygen-containing functional groups [1]. Previously, the catalytic
hydrogenolysis of sorbitol and other polyols was usually conducted
using a metal catalyst and a basic promoter, and those promoters
like NaOH and Ca(OH)2 have proved a positive influence on the C–C
scission where a widely accepted mechanism was the retro-aldol
condensation [2–6]. However, such problems as accelerated degrada-
tion and glycol products separation would occur when the alkali was
dissolved in the reaction solution. Without alkalis, MgO supported Cu,
Co and Pt catalysts showed bifunctional effects and excellent perfor-
mances in the aqueous-phase hydrogenolysis of glycerol to glycols
[7–9], which should be also efficient for sorbitol hydrogenolysis. Con-
sidering the high cost of noble metal catalysts, some non-noble metal
catalysts with excellent C–C bond breaking ability, like Ni, have
attracted great attention in polyols hydrogenolysis [5,10,11].
tions and without alkaline additives. The catalysts were characterized
to study the relationship between the structure and their catalytic per-
formance. We also discussed the effects of temperature, H2 pressure and
reaction time.
2. Experiment
2.1. Catalyst preparation and characterization
A series of Ni-MgO catalysts with varying Ni/Mg molar ratios were
prepared by co-precipitation [12,13]. A mixed solution (1 M total
metal) of Ni(NO3)2 · 6H2O and Mg(NO3)2 · 6H2O was deposited by a
Na2CO3 solution (1.2 M) at an adding rate of 1 mL/min at room temper-
ature (RT). After aged overnight, isolated and washed by vacuum filtra-
tion until pH b 8, the precipitates were dried at 383 K for 12 h and
calcined at 773 K for 3 h in static air, followed by reduction with pure
H2 at 773 K for 3 h at a ramp rate of 3 K/min. Ni-Al2O3 catalyst was
also prepared by the same procedures but with reduction temperature
of 1073 K. The NiOx sample was obtained by co-precipitation and using
only Ni(NO3)2 · 6H2O as precursor. All the catalysts were identified
as Ni-Mg(x:y) or Ni-Al(x:y), where x:y referred to the atomic ratio.
The catalysts were characterized by ICP-AES, N2-TPD, CO2-TPD, XRD,
H2-TPR and H2 chemisorption to study their physico–chemical proper-
ties and the technical details were provided in the Supporting Informa-
tion (SI).
⁎
Corresponding author at: Key Laboratory of Biobased Materials, Qingdao Institute of
Bioenergy and Bioprocess Technology, Chinese Academy of Sciences, Qingdao 266101, PR
China. Tel.: +86 532 80662723; fax: +86 532 80662724.
E-mail address: muxd@qibebt.ac.cn (X. Mu).
1566-7367/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.catcom.2013.05.012

X. Chen et al. / Catalysis Communications 39 (2013) 86–89
87
2.2. Catalytic test and product analysis
MgO & MgNiO
♦ Ni
∇
2
(111)
♦
(200)
∇
♦
(220)
Sorbitol hydrogenolysis was carried out in a batch system with a
100 mL 316 L stainless steel autoclave reactor. After 50 mL aqueous
solution of 20% sorbitol and 0.5 g catalyst were loaded, the reactor
was purged with hydrogen for four times, aerated to the desired pres-
sure and then heated to the defined temperature for a given period at
600 rpm stirring speed. All the experiments were carried out in trip-
licate. After reaction, the unconsumed sorbitol and products were
quantified by both gas chromatography (GC) and ion chromatogra-
phy (IC). The byproducts were analyzed by GC coupled with a mass
spectrometer (MS), and the total carbon balance (TC) was also mea-
sured. The analytical methods were shown in the SI [13].
(111)
(220)
(311)
(200)
♦
(222)
∇
∇
∇
∇
f
e
d
c
b
a
3. Results and discussion
3.1. Chemical and physical properties of catalysts
40
50
60
70
80
Table 1 summarized the physico–chemical properties of Ni-MgO
catalysts with Ni/Mg molar ratio ranging from 1/9 to 5/5. A high sur-
face area accompanied with a strong basicity was obtained for MgO
sample. As Ni was introduced and increased, both the BET surface
areas and the basicities of catalysts continuously decreased. The actu-
al content of Ni was very close to the theoretical loading, which in-
creased with the increasing Ni/Mg ratio. Based on H2-TPR results,
only a small fraction of nickel oxides could be reduced at the temper-
ature below 700 K for Ni-MgO catalysts, while the unsupported NiOx
sample could be completely reduced at 700 K. The Ni oxides loaded
on MgO were more difficult to reduce than the unsupported NiOx
and the reduction temperature rose with the increase of Ni/Mg ratio
(Fig. S1 in the SI). When the Ni/Mg ratio increased from 1/9 to 3/7,
the Ni surface area reached a maximum of 6.2 m2/g, and subsequent-
ly dropped as the ratio continuously increased. This decrease could be
attributed to the decline in catalyst reducibility and the aggregation
of Ni particles [12]. As identified in Fig. 1, the diffraction lines at
44.5°, 51.8° and 76.4° for 2θ were attributed to the Ni(111), Ni(200)
and Ni(220) crystal planes (JCPDS No. 87-0712), which showed
stronger intensity as Ni content increased. On the other hand, the
MgO characteristic diffraction peaks at 36.9°, 42.9°, 62.3°, 74.7° and
78.6° for 2θ (JCPDS No. 77-2179) became weaker when Ni was intro-
duced. The angles of MgNiO2 characteristic diffraction peaks (JCPDS
No. 24-0712) were a little higher than those of MgO and NiO (JCPDS
No. 89-7130), and a gradual shift of peaks at about 37°, 43° and 62°
for 2θ to the higher diffraction angles was observed after increasing
partial substitution of Mg with Ni [14,15]. The phenomenon indicated
that the amount of MgNiO2 solid solution increased, which brought
about higher reduction temperature and poorer reducibility in Ni-MgO
catalysts with the increasing Ni content [12,16]. Ni nanoparticles grew
slowly as Ni content increased. The degree of reduction was rapidly
2 Theta (degree)
Fig. 1. XRD patterns of MgO (a), Ni-Mg(1:9) (b), Ni-Mg(2:8) (c), Ni-Mg(3:7) (d), Ni-Mg(4:6)
(e) and Ni-Mg(5:5) (f) samples.
lessened from 52.4% to 18.9% and the average size of Ni particles was
about 10 nm.
3.2. Hydrogenolysis of sorbitol on Ni-MgO catalysts
Table 2 presented the results of sorbitol hydrogenolysis over
Ni-MgO catalysts under relatively mild conditions of 473 K and
4 MPa H2, and sorbitol was mostly converted to glycerol and lower gly-
cols mainly containing EG and 1,2-PG. Other compounds like sorbitans,
mannitol, acetylacetone and lactic acid were also observed. The total
carbon balance ranged from 92.3% to 100.3% for all the tests. When
the Ni-Mg(1:9) catalyst was used, a 47.4% sorbitol conversion with a
66.2% total selectivity of glycols and glycerol was obtained. With Ni con-
tent increasing, the sorbitol conversion was improved to the biggest
value of 67.8% over Ni-Mg(3:7) catalyst, and then dropped as the Ni
loading continued to increase. The total selectivity of EG, 1,2-PG and
glycerol, also reached the maximum of 80.8% on Ni-Mg(3:7) catalyst.
The activity deterioration for the Ni-Mg(4:6) and Ni-Mg(5:5) catalysts
compared with the catalysts with lower Ni-Mg ratios could be ascribed
to the decrease of basicity and Ni surface area which were critical in the
C–C bond cleavage and the subsequent hydrogenation [3,6].
From the reaction mechanism, the strong basicity of catalyst was
conducive to the C–C scission in the retro-aldol condensation. Sorbitol
and intermediates might also be abundantly adsorbed on the plentiful
Table 1
Table 2
Characterization results of Ni-MgO catalysts.
Hydrogenolysis of sorbitol over Ni-MgO catalysts.a
b
c
d
Catalyst
Nia
(wt.%) (%)
Dreduc. SBET
(m2 g−1
SNi
(m2 g−1
Basic.e
PSf (nm)
Enter Catalyst
Sorbitol
Conv. (%) EG
Product selectivity (%)
TCd
)
)
(×10−6 mol g−1
)
Ni
MgO
1,2
Glycerol Gas
productsb
Othersc (%)
-PG
Ni-Mg(1:9) 14.6
Ni-Mg(2:8) 28.5
Ni-Mg(3:7) 38.4
Ni-Mg(4:6) 47.2
Ni-Mg(5:5) 56.7
52.4
40.1
36.2
27.3
18.9
–
193.5
141.6
106.9
99.2
79.0
262.0
2.3
4.8
6.2
5.6
3.8
–
362.2
285.7
257.9
229.9
175.8
380.8
7.0 7.1
8.5 6.8
9.1 6.7
10.5 6.8
10.8 6.4
1
2
3
4
5
Ni-Mg(1:9) 47.4
16.6 24.5 25.1
18.6 30.8 23.7
26.0 33.7 21.1
15.6 34.1 11.7
10.3 28.6 10.9
15.2
12.9
9.7
6.3
10.4
14.0
11.9
8.1
25.8
29.9
97.2
98.3
99.6
94.7
93.4
Ni-Mg(2:8) 61.4
Ni-Mg(3:7) 67.8
Ni-Mg(4:6) 41.9
Ni-Mg(5:5) 33.0
MgO
–
–
9.8
a
a
Ni loading, determined by ICP analysis.
Degree of reduction, quantified from H2-TPR data.
BET surface area, measured by N2-TPD.
Ni surface area, characterized with H2 chemisorption.
Basicity, measured by CO2-TPD.
Reaction conditions: 20% aqueous sorbitol solution, 50 mL; catalyst amount, 0.5 g;
b
c
d
e
f
reaction time, 4 h; H2 pressure, 4 MPa; temperature, 473 K; stirring speed, 600 rpm.
b
Gas products include CH4, CO2 and hexane.
Others include acetone, methanol, ethanol, lactic acid, hydroxyacetone, butanediol,
c
hexanediol, acetonyl acetone, mannitol, erythritol, sorbitans, isosorbide, etc.
d
Particle size, calculated in XRD characterization from the Scherrer equation.
Total carbons, includes total organic carbons and total inorganic carbons.

88
X. Chen et al. / Catalysis Communications 39 (2013) 86–89
alkaline sites to improve the reaction rate [17]. The addition of MgO
or NaOH promoted the sorbitol conversion, however, the glycol selec-
tivities decreased in spite of the extra alkaline sites introduced by
MgO added (Table S1 in the SI). It could be concluded that the basicity
of Ni-MgO was strong enough to perfectly facilitate this chemoselective
reaction and such alkaline additives were hardly necessary [3]. For the
Ni-Al(3:7) catalyst, NaOH significantly enhanced the sorbitol conver-
sion and the product selectivity, though the resultant conversion was
only 48.1%. The result indicated that the acidic carrier Al2O3 supported
metal catalysts should proceed under harsher conditions and with an
appropriate amount of basic additives to reduce the byproducts in sor-
bitol conversion [5,18].
As shown in Table 2, the Ni-Mg(3:7) catalyst with the highest
nickel surface area exhibited the best catalytic performance, which
suggested that the available active Ni sites were also essential for
this conversion. During the recycling process, MgO was partly eroded
by H2O, reacted with CO2 and transformed into MgCO3. Therefore, the
Ni nanoparticles aggregated (Fig. S3 in the SI) [9,19]. Based on the
XRD and H2 chemisorption analysis, the recycled Ni-Mg(3:7) catalyst
in the second run possessed 19.4 nm average-sized Ni clusters with
an only 3.5 m2/g Ni surface area, over which the sorbitol conversion
dropped to 58.7% along with a 63.5% total selectivity of EG, 1,2-PG
and glycerol (Fig. S2 in the SI). Besides, the products with chelating
groups would facilitate leaching and poisoning of Ni catalyst, which
also led to the loss of catalytic activity [20]. Over Ni-MgO catalysts,
the 1,2-PG selectivity increased and glycerol selectivity decreased
with the Ni/Mg ratio ranging from 1/9 to 4/6, revealing that the cata-
lysts with higher metal loading also enhanced the glycerol dehydra-
tion with C–O bond breaking [3,5,19,21]. The optimized Ni/Mg
atomic ratio was 3/7 and the Ni-Mg(3:7) catalyst was chosen to in-
vestigate the influences of temperature, H2 pressure and reaction
time on the sorbitol hydrogenolysis.
Sorbitol Conversion
EG Selectivity
1,2-PG Selectivity
Glycerol Selectivity
100
90
80
70
60
50
40
30
20
10
60
50
40
30
20
10
433
453
493
473
Reaction Temperature (K)
Fig. 2. Effect of reaction temperature on the sorbitol hydrogenolysis over Ni-Mg(3:7)
catalyst under 4 MPa H2 for 4 h.
4. Conclusions
Ni-MgO catalysts were very efficient for sorbitol hydrogenolysis
to lower glycols, even under mild conditions and without basic pro-
moters. Characterization results suggested that catalytic performance
largely hinged upon the basicity and active metal sites. The poor hydro-
thermal stability of Ni-MgO catalyst would lead to the loss of activity in
repeated runs. Under harsh conditions, the deep degradation was accel-
erated resulting in more byproducts and lower yield of glycols although
the sorbitol conversion increased.
3.3. The effects of reaction conditions on sorbitol hydrogenolysis
As revealed in Fig. 2, the sorbitol conversion increased rapidly with
temperature increasing from 433 K to 453 K and then remained at
about 70%, whereas the total selectivity of EG, 1,2-PG and glycerol
reached 80.8% at 473 K followed by a significant decrease. So, polyols
could be considerably converted under a lower temperature, while the
further hydrogenation of intermediates into lower glycols would be
favorable at the higher appropriate temperature [4]. Product distribu-
tion showed that the selectivity towards glycerol was reduced, but the
total selectivity of C3 products including glycerol and 1,2-PG remained
almost unchanged with reaction temperature increasing to 473 K. The
increasing temperature could accelerate the glycerol hydrogenolysis
to 1,2-PG with promoted C–O bond scission [19]. At the temperature
higher than 473 K, further degradation of all the glycols was strongly
enhanced, resulting in a sharp decline of product selectivity.
Fig. 3 showed that the sorbitol conversion increased significantly
with higher H2 pressure, together with a steady growth of glycerol se-
lectivity. In contrast, the selectivity of EG and 1,2-PG reached the
maxima of 33.7% and 26.0% under 4 MPa, and then dropped with
the continuously increasing H2 pressure. Thus, the higher H2 pressure
could promote the hydrogenation of unsaturated intermediates as
well as the further degradation of glycols [21,22]. To our knowledge,
the optimized H2 pressure of 4 MPa was the lowest, suggesting that
the Ni-MgO catalysts were indeed efficient.
Acknowledgements
This work was supported by the Natural Science Foundation of
China (No. 21273260) and the Key Science and Technology Program of
Shandong Province (No. 2007GG2QT07006 and No. 2008GG20002038).
Sorbitol Conversion
EG Selectivity
1,2-PG Selectivity
Glycerol Selectivity
100
90
80
70
60
50
40
30
20
10
60
50
40
30
20
10
As the reaction proceeded, both the sorbitol conversion and the
glycerol selectivity gradually increased (Fig. 4). After 8 h, sorbitol
was completely converted and the highest glycerol selectivity of
28.1% was acquired. However, the total selectivity of EG, 1,2-PG and
glycerol reached the maximum when the reaction time was extended
to 4 h, and subsequently dropped with time. GC analysis indicated
that the deep degradation products generally increased over a prolonged
period of time, like alcohols, CO2, CH4, and other alkanes.
2
4
6
8
H2 Pressure (MPa)
Fig. 3. Effect of H2 pressure on the sorbitol hydrogenolysis over Ni-Mg(3:7) catalyst at
473 K for 4 h.

X. Chen et al. / Catalysis Communications 39 (2013) 86–89
89
References
Sorbitol Conversion
EG Selectivity
1,2-PG Selectivity
Glycerol Selectivity
100
90
80
70
60
50
40
30
20
10
60
[1] J.N. Chheda, G.W. Huber, J.A. Dumesic, Angewandte Chemie, International Edition
46 (2007) 7164.
[2] M.A. Andrews, S.A. Klaeren, Journal of the American Chemical Society 111 (1989)
4131.
[3] L. Zhao, J.H. Zhou, Z.J. Sui, X.G. Zhou, Chemical Engineering Science 65 (2010) 30.
[4] J. Sun, H. Liu, Green Chemistry 13 (2011) 135.
[5] L. Ye, X. Duan, H. Lin, Y. Yuan, Catalysis Today 183 (2012) 65.
[6] K. Wang, M.C. Hawley, T.D. Furney, Industrial & Engineering Chemistry Research
34 (1995) 3766.
[7] X. Guo, Y. Li, R. Shi, Q. Liu, E. Zhan, W. Shen, Applied Catalysis A: General 371
(2009) 108.
[8] Z. Yuan, P. Wu, J. Gao, X. Lu, Z. Hou, X. Zheng, Catalysis Letters 130 (2009) 261.
[9] Z. Yuan, J. Wang, L. Wang, W. Xie, P. Chen, Z. Hou, X. Zheng, Bioresource Technology
101 (2010) 7099.
[10] U. Saxena, N. Dwivedi, S.R. Vidyarthi, Industrial
Research 44 (2005) 1466.
[11] L. Zhou, A. Wang, C. Li, M. Zheng, T. Zhang, ChemSusChem 5 (2012) 932.
[12] Y. Echegoyen, I. Suelves, M.J. Lázaro, M.L. Sanjuán, R. Moliner, Applied Catalysis A:
General 333 (2007) 229.
50
40
30
20
10
& Engineering Chemistry
[13] X.C. Wang, L.Q. Meng, F. Wu, Y.J. Jiang, L. Wang, X.D. Mu, Green Chemistry 14
(2012) 758.
[14] T. Nakayama, N. Ichikuni, S. Sato, F. Nozaki, Applied Catalysis A: General 158
(1997) 185.
[15] L. Chuanming, C. Donghua, T. Wanjun, P. Yuhua, Journal of Analytical and Applied
Pyrolysis 75 (2006) 240.
2
4
6
8
Reaction Time (h)
[16] V. Mohan, C.V. Pramod, M. Suresh, K.H. Prasad Reddy, B.D. Raju, K.S. Rama Rao,
Catalysis Communications 18 (2012) 89.
Fig. 4. Effect of reaction time on the sorbitol hydrogenolysis over Ni-Mg(3:7) catalyst
under 4 MPa H2 at 473 K.
[17] J.X. Li, W.L. Dai, K. Fan, Journal of Physical Chemistry C 112 (2008) 17657.
[18] K.L. Deutsch, D.G. Lahr, B.H. Shanks, Green Chemistry 14 (2012) 1635.
[19] M. Balaraju, K. Jagadeeswaraiah, P.S.S. Prasad, N. Lingaiah, Catalysis Science &
Technology 2 (2012) 1967.
[20] R. Rinaldi, F. Schüth, Energy & Environment Science 2 (2009) 610.
[21] C. Montassier, J.C. Menezo, L.C. Hoang, C. Renaud, J. Barbier, Journal of Molecular
Catalysis 70 (1991) 99.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.catcom.2013.05.012.
[22] N. Li, G.W. Huber, Journal of Catalysis 270 (2010) 48.