A noble-metal free Cu-catalyst derived from hydrotalcite for highly efficient hydrogenation of biomass-derived furfural and levulinic acid

Article abstract of DOI:10.1039/c3ra22158j

A Cu-catalyst derived from hydrotalcite precursor was highly selective for the hydrogenation of biomass-derived furfural and levulinic acid. 90% yield of furfuryl alcohol and 51% yield of 2-methylfuran were achieved selectively in the hydrogenation of furfural. 91% yield of γ-valerolactone was achieved in the hydrogenation of levulinic acid.

Full text of DOI:10.1039/c3ra22158j

RSC Advances
View Article Online
View Journal | View Issue
COMMUNICATION
A noble-metal free Cu-catalyst derived from
hydrotalcite for highly efficient hydrogenation of
biomass-derived furfural and levulinic acid3
Cite this: RSC Advances, 2013, 3,
853
3
Received 13th September 2012,
Accepted 21st January 2013
Kai Yan,{* Jiayou Liao, Xu Wu and Xianmei Xie*
DOI: 10.1039/c3ra22158j
www.rsc.org/advances
A Cu-catalyst derived from hydrotalcite precursor was highly
selective for the hydrogenation of biomass-derived furfural and
levulinic acid. 90% yield of furfuryl alcohol and 51% yield of
genation of furfural to furfuryl alcohol and/or tetrahydrofurfuryl
alcohol (THFA) has been investigated using RANEY1 Ni, Ru, Pd,
11
Cu supported metal catalysts with good performance achieved.
2
-methylfuran were achieved selectively in the hydrogenation of
For the hydrogenation of furfural to MF, the reaction appeared to
be relatively difficult, possibly due to the long chain, whereas the
reaction firstly requires selective hydrogenation of the aldehyde
group (–CLO) without the breaking of the conjugated ring system
(–CLC–CLC–), and secondly, hydrogenolysis of the C–O bond of FA
furfural. 91% yield of c-valerolactone was achieved in the
hydrogenation of levulinic acid.
With decreasing fossil resources and the fast increasing need for
energy from economic development, worldwide demand for fuels,
climate concerns about the use of fossil-based energy carriers, and
political focus have forced and sparked the utilization of renew-
8a
to form MF, where additional energy is needed. Poliakoff et al.
reported a twin catalyst system in combination with supercritical
CO in a continuous-flow reactor coupled with multi-steps to yield
2
1–3
able energy resources.
Biomass is an abundant and carbon-
90% MF in the hydrogenation of furfural. This catalytic strategy
neutral renewable energy resource for the promising production of
creates a flexible route for the transformation of furfural into the
liquid fuel MF, which is promising for the transportation sector.
Although the above-mentioned methods were advanced, the
catalysts are primarily based on noble-metals and the cost of
noble-metals and/or the multiple steps of the catalytic system was
an issue to some extent.
The hydrogenation of LA to GVL involves the hydrogenation
followed by intramolecular cyclisation, with the loss of one
2,4
biofuels and value-added chemicals. During past decade, huge
effort has been devoted to the selective transformations of a set of
feedstocks (cellulose, lignin, carbohydrates) into the platform
intermediates furfural, 5-hydroxymethylfurfural (HMF) and levu-
linic acid (LA), and their conversion into potential biofuels and
12
4–6
value-added chemicals. Schemes 1 and 2 show a diverse range
of highly attractive fuel components and value-added chemicals
can be produced from furfural and LA, respectively, which can be
molecular water; good yields were often achieved using homo-
7,8
obtained from the green lignocellulosic biomass. Among the
potential fuel components, 2-methylfuran (MF) and c-valerolac-
tone (GVL) are highly attractive due to their unique fuel
7a,7c,13
geneous catalysts.
Heterogeneous systems, which are
promising for recycling and separation, are more economic and
relatively easy to scale-up. Subsequently, supported Pd, Ru, Au
8
a,9
properties.
lubricity. In addition, they display higher energy densities, higher
boiling points and low solubility in water (see Table S1 in the ESI ).
Besides, they are also advocated as an alternative green solvent in
Compared with ethanol fuel, they display better
8a,14
catalysts
were employed and good performances were
obtained. Although the hydrogenations using noble-metal cata-
3
10
the pharmaceutical industry. Concerning their properties and
transformations, several pioneering works have been published by
the groups of Horv ´a th, Leitner and Poliakoff. The hydro-
7
c
7a
8a
Institute of Chemistry and Chemical Engineering, Taiyuan University of Technology,
Yingze Street No. 79, Taiyuan, 030024, China. E-mail: yankai0553@sina.com;
xxmsxty@sina.com; Tel: +86-351-6018564
3
Electronic supplementary information (ESI) available: XRD analysis of fresh and
spent catalysts, EDX, TEM and details of catalytic reactions. See DOI: 10.1039/
c3ra22158j
{
Present address: Department of Chemistry, Lakehead University, P7B 5E1, ON,
Canada. Tel.: +1-807-6273059.
Scheme 1 Hydrogenation of furfural.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 3853–3856 | 3853

View Article Online
Communication
RSC Advances
hydrotalcite precursor, whereas Cu(II) in octahedral coordination
cannot be easily accommodated in the layers of the hydrotalcite
structure during the coprecipitation synthesis and thus a slight
amount of Cu(II) would go into solution, leading to a slight
16
leaching of Cu(II). The X-ray diffraction (XRD) patterns of the
three resulting Cu-catalysts (Fig. S2 ) revealed that the phases were
primarily composed of CuO and spinel (CuAl , CuFe
CuCr O ) phases, which was further confirmed by TEM analysis
3
2
O
4
2 4
O ,
2
4
3
(Fig. S3 ). Subsequently, their catalytic performances were
Scheme 2 Hydrogenation of levulinic acid.
evaluated in the hydrogenation of furfural and levulinic acid. To
our satisfaction, our hypothesis was confirmed that all the
resultant Cu-catalysts display good performance in both hydro-
genation tests.
lysts are advanced, however, the high cost of the expensive noble-
metals would limit their applications.
In the case of the hydrogenation of furfural, the resulting Cu-
catalysts worked very well (Table 1), which further confirmed the
copper sites were important for the hydrogenation. In terms of the
conversion of furfural, the best result was achieved with the Cu–Fe
catalyst (entry 3), followed by the Cu–Cr catalyst (entry 2) and then
by the Cu–Al catalyst (entry 1). Alternatively, in terms of selectivity,
the Cu–Al and Cu–Cr catalysts display very high selectivity toward
the formation of FA (entries 1 and 2), while the Cu–Fe displays
much higher selectivity for the formation of MF. Besides, one
comparative reaction using a mechanical mixture of CuO and
CuFe O as the catalyst was performed (entry 4); 51.4% yield of FA
In this work, we report that cheap and noble-metal free Cu-
catalysts (Cu–Al, Cu–Cr, Cu–Fe) derived from their hydrotalcite
precursors were efficient for the hydrogenation of furfural and LA,
whereby the hydrotalcite precursors were simply synthesized by a
modified coprecipitation method using the cheap metal nitrates.
In the first hydrogenation of furfural (Scheme 1), FA (90% yield)
and MF (51% yield) can be selectively produced using the Cu–Fe
catalyst under mild conditions. Besides, the resultant Cu-catalysts
(Cu–Al, Cu–Cr, Cu–Fe) were also highly selective for the hydro-
genation of LA to GVL (Scheme 2). The best performance was
achieved with the Cu–Cr catalyst, where 91% yield of GVL was
achieved at close to perfect conversion of LA.
2
4
and 10.6% yield of MF was obtained at 88.5% conversion.
Although good performance was obtained in this case, unfortu-
Vannice et al. studied copper chromite catalysts for the
hydrogenation of furfural and reported that the active centre for
0
nately, partial Cu was found to be deposited on the surface of the
reactor after hydrogenation, which would make the recovery of the
mixed catalyst very difficult. Based on all these data, it was clearly
shown that the metal oxide was highly dispersed in the fabricated
Cu-catalysts, which further confirmed the advantage of the
preparation method developed in this work. The electronegativ-
15
the hydrogenation was the copper site. Based on this finding, we
hypothesized that Cu-based catalysts derived from their hydro-
talcite precursors would also be efficient. Thus we initially began
our study through the synthesis of Cu-catalysts (Cu–Al, Cu–Cr, Cu–
Fe) derived from their hydrotalcite precursors. Scanning electron
microscopy (SEM) coupled with the energy-dispersive X-ray (EDX)
analysis was initially employed with the representative Cu–Fe
3
+
3+
3+
3+
ities of M (Al of 1.61, Cr of 1.66 and Fe of 1.83) may also
17
play a role in the catalytic activities; it was supposed that the
electron-deficient Cu and M could attract the oxygen in the
carbonyl group of furfural via a side-bond interaction, where the
2+
3+
2+
catalyst and the analysis results indicated the molar ratio of Cu /
18
3+
Fe was y1.6 (Fig. S1
3
). The relatively precise value of 1.73 was
copper site would dissociate the adsorbed hydrogen, resulting in
calculated from ICP measurements. Both of the detected values
18a
3+
2+
3+
the hydrogenation of the CLO bond.
The Fe , having the
were less than the original Cu /Fe ratio of 2, which was possibly
due to the rigid Jahn–Teller effect in the synthetic process of their
highest electronegativity, would present the highest ability to
Table 1 Initial hydrogenation of furfural and levulinic acid (LA) using the resultant Cu-catalysts
a
b
Hydrogenation of furfural/%
Hydrogenation of LA/%
Entry
Catalyst
Cu–Al
3+
Conv.
32.6
94.0
96.4
88.5
Y
FA
Y
MF
Conv.
98.3
.99
.99
—
Y
GVL
Y
PA
Y
MTHF
1
2
3
31.5
83.2
15.9
51.4
0
87.3
90.7
81.5
—
1.2
0.8
0.6
0.8
0.1
3.5
2+
(
Cu /Al = 2)
Cu–Cr
0.3
2+
3+
(Cu /Cr = 2)
Cu–Fe
36.0
10.6
2+
3+
(
Cu /Fe = 2)
CuO–CuFe
1 : 1)
c
4
2
O
4
—
—
(
a
Reaction conditions: V(furfural) = 2.1 mL, V(octane) = 5 mL, T = 200 uC, t = 4 h, m(catalyst) = 0.2 g, p(H
2
) = 60 bar, stirring speed = 1000
O) = 5 mL, T = 200
2
) = 70 bar, stirring speed = 1000 rpm. YGVL: the yield of GVL, YPA: the yield of PA, YMTHF: the yield of
b
rpm. YFA: the yield of furfuryl alcohol (FA), YMF: the yield of 2-methylfuran (MF). Reaction conditions: m(LA) = 1.02 g, V(H
uC, t = 10 h , m(catalyst) = 0.1 g, p(H
MTHF. Weight ratio.
2
c
3
854 | RSC Adv., 2013, 3, 3853–3856
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Communication
further attract electrons from the hydroxyl group of FA and
promoted further hydrogenolysis to MF at 200 uC.
hydrogen pressure was used (entries 5 and 7), the total hydrogen
(45 bar) was simply not enough to reduce the furfural amount
(y25 mmol) completely into MF, and FA was mainly formed.
Furthermore, reaction time was prolonged in an endeavour to
improve the yield of MF under the standardized conditions.
Interestingly, the best yield of MF achieved was 51% at 99%
conversion of furfural when the reaction time was 14 h. Reviewing
Table 2, we find that MF was produced more at high temperature
(over 200 uC) and under high pressure. This was possibly due to
that fact that the high temperature would enhance the faster mass
8b,c
For the hydrogenation of LA, all the resulting Cu-catalysts
continued to display good performances. The best performance
was achieved using the Cu–Cr catalyst, with which 91% yield of
GVL was achieved (entry 2, Table 1) at close to perfect conversion
of LA. Through comparison with work from the literature (Table
S2
of 81.5% yield of GVL was achieved using the Cu–Fe catalyst (entry
). The catalytic difference was possibly due to the electronegativity
3
), we find that our results are superior. The worst performance
3
3+
3+
3+
3+
17
of M (Al of 1.61, Cr of 1.66 and Fe of 1.83), where the
highest electronegativity of Fe would present a strong ability to
attract electrons and promote further hydrogenolysis to
transfer and greater H adsorption on the Cu–Fe catalyst, where
2
3+
the support CuFe O would contact with more H , resulting in the
2
4
2
rapid hydrogenolysis of FA to MF. As far as we are concerned, our
results are among the top-list comparing with the representative
1
1b,15
2-MTHF.
Based on the good performance in the hydrogenation of
work from the literature (Table S4 ). Moreover, on prolonging the
3
furfural, we subsequently focused on the production of FA and
MF. We envisaged that both yields could be further improved
through the optimization of the reaction parameters. In set of
experimental plans shown in Table 2, as expected, the yield of FA
and MF could be greatly enhanced through rational design, where
the best yield of FA of 89.5% was achieved at 91% conversion of
reaction time to 18 h, no clear change in the yield of MF was
found. However, when the reaction time was further prolonged to
22 h, the yield of MF decreased by 8%. This was due to competitive
side reactions occurring, such as the formation of THFA, which
occurred on the conjugated ring system of furfural. As detected
and verified by our GC analysis, the yield of THFA (not shown
here) was increased slightly from 4.1% to 7.8% on prolonging the
time from 10 to 22 h.
In order to explore the recyclability, the spent catalyst was
separated from the reaction solution through centrifugation for 30
min and then dried at 80 uC in an air oven for 12 h. The spent
catalyst was then used for the hydrogenation of furfural at 220 uC,
furfural at 160 uC, 5 h and 90 bar H
knowledge, these values are among the top results in comparison
with the representative works from the literatures (Table S3 ). On
2
(entry 2). To our limited
3
the other hand, it was interesting to find the low temperature (160
uC) was good for achieving higher yields of FA (entries 1 to 4),
which verified that the hydrogenation at the carbonyl group CLO
bond of furfural was preferred over the CLC bond in the
conjugated ring system. Besides, at the same temperature of 160
uC, the hydrogen pressure (e.g. entry 1 vs. 2) presented a higher
influence on the yield of FA than the reaction time (e.g. entry 2 vs.
5
2
h and 90 bar H ; the catalytic result (Run 0) is shown in Fig. 1.
36% yield of FA was achieved at a conversion of 84%, which was
much lower than the fresh catalyst’s performance of 58% yield of
FA at a conversion of 99% (entry 6, Table 2). This was possibly due
to carbon deposits from the reactant as well as products. In the
hydrogenation of furfural in octane solvent, the reactant furfural
3
). This was possibly because Cu sites were responsible for the
dissociative activation of H . When a higher pressure of H (90
bar) was employed (entries 2 and 3), more H would be adsorbed
2
2
2
2 4
was absorbed on the surface of the Cu–Fe catalyst (CuO/CuFe O ).
on the surface and would enhance the catalytic activities. This
finding was also verified at higher temperature (220 uC) in the
subsection. Subsequently, when the reaction temperature was
raised to 220 uC, a higher hydrogen pressure (entries 6 and 8) was
good for achieving a higher yield of MF. In the cases where lower
The electron-deficient Cu and Fe could attract the oxygen in the
carbonyl group through a side-bond interaction, which would
18
activate the CLO bond p-complexed to the surface of CuO/CuFe O4
2
a
Table 2 Optimization of the hydrogenation of furfural using Cu–Fe catalyst
Hydrogenation of furfural
Entry T/uC Time/h Pressure/bar Conv./%
YFA/%
YMF/%
1
2
3
4
5
6
7
8
9
0
1
160
160
160 10
160 10
220
220
220 10
220 10
220 14
220 18
220 22
5
5
45
90
90
45
45
90
45
90
90
90
90
82.9
91.0
95.9
83.1
98.2
99.2
98.5
99.6
99.4
99.6
99.3
81.2
89.5
82.2
75.6
81.1
57.9
82.6
58.7
41.7
40.1
39.8
0.3
0.5
2.5
0.2
6.2
17.0
16.1
29.6
51.1
50.1
43.1
5
5
1
1
a
Note: other conditions were the same as for Table 1. Time
influence was investigated from entries 8 to 11.
Fig. 1 Catalytic activities of the spent Cu–Fe catalyst.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 3853–3856 | 3855

View Article Online
Communication
RSC Advances
and promote the hydrogenation of the CLO bond on the one
6 (a) J. Q. Bond, D. M. Alonso, D. Wang, R. M. West and J.
A. Dumesic, Science, 2010, 327, 1110–1114; (b) H. Zhao, J.
E. Holladay, H. Brown and Z. C. Zhang, Science, 2007, 316,
18a
hand. On the other hand, the strong adsorption of reactant and
products, the low volatility (in our gas-phase reaction) and the low
solubility in octane solvent would present longer retention on the
1597–1600; (c) Y. Roman-Leshkov, C. J. Barrett, Z. Y. Liu and J.
A. Dumesic, Nature, 2007, 447, 982–985.
19
surface. At the reaction temperature (220 uC), the condensation
7
(a) F. M. A. Geilen, B. Engendahl, A. Harwardt, W. Marquardt,
J. Klankermayer and W. Leitner, Angew. Chem., Int. Ed., 2010,
of furfural would possibly produce potential carbon deposit on the
19a
surface of the catalyst. To remove the potential carbon deposit,
the spent catalyst was calcined at 550 uC for 3 h and then used for
the next run. To our delight, very good performance (y50% yield
of FA with y18% yield of MF at y98% conversion) was achieved
49, 5510–5514; (b) L. Deng, J. Li, D. M. Lai, Y. Fu and Q. X. Guo,
Angew. Chem., Int. Ed., 2009, 48, 6529–6532; (c) H. Mehdi,
V. F ´a bos, R. Tuba, A. Bodor, L. T. Mika and I. T. Horv ´a th, Top.
Catal., 2008, 48, 49–54; (d) R. A. Bourne, J. G. Stevens, J. Ke and
M. Poliakoff, Chem. Commun., 2007, 4632–4634.
(Run 1, Fig. 1). In the next two runs, fortunately, the yields of FA
8
(a) J. G. Stevens, R. A. Bourne, M. V. Twigg and M. Poliakoff,
Angew. Chem., Int. Ed., 2010, 49, 8856–8859; (b) H. Y. Zheng, Y.
L. Zhu, B. T. Teng, Z. Q. Bai, C. H. Zhang, H. W. Xiang and Y.
W. Li, J. Mol. Catal. A: Chem., 2006, 246, 18–23; (c) S. Sitthisa,
T. Sooknoi, Y. G. Ma, P. B. Balbuena and D. E. Resasco, J. Catal.,
and MF remained stable (Runs 2 and 3). To check the structural
evolvement, XRD was further used to characterize the fresh, spent
0
and calcined catalyst (Fig. S4
and delafossite CuFeO
further confirmed that the Cu sites were responsible for the
dissociative activation of H and governed the catalytic activities.
3
); it could be clearly seen that Cu
2
presented in the spent catalyst. This
2011, 277, 1–13; (d) L. Hu, G. Zhao, W. W. Hao, X. Tang, Y. Sun,
2
L. Lin and S. J. Liu, RSC Adv., 2012, 2, 11184–11206.
In conclusion, biomass-derived platform chemical furfural and
levulinic acid can be selectively hydrogenated under mild reaction
conditions using Cu-catalysts derived from hydrotalcite precur-
sors. In the hydrogenation of furfural, furfuryl alcohol (90% yield)
and 2-methylfuran (51% yield) can be selectively produced on the
Cu–Fe catalyst. Besides, the Cu-catalysts were also efficient for the
hydrogenation of levulinic acid to c-valerolactone, whereby 91%
yield of c-valerolactone was achieved at close to perfect conversion.
The Cu sites functioned to activate the H2 and governed the
catalytic reaction. The Cu-catalysts presented stable activity over
several consecutive runs through a simple calcination procedure.
The present work might find general application in the hydro-
genation of other biomass-derived monomers containing aldehyde
functional groups.
9 (a) I. T. Horv ´a th, H. Mehdi, V. Fabos, L. Boda and L. T. Mika,
Green Chem., 2008, 10, 238–242; (b) V. F ´a bos, G. Kocz ´o ,
H. Mehdi, L. Boda and I. T. Horv ´a th, Energy Environ. Sci., 2009,
2
, 767–769.
0 (a) D. H. B. Ripin and M. Vetelino, Synlett, 2003, 2353; (b) T.
W. Rudolph and J. J. Thomas, Biomass, 1988, 16, 33–49.
1
1
1 (a) B. J. Liu, L. H. Lu, B. C. Wang, T. X. Cai and I. Katsuyoshi,
Appl. Catal., A, 1998, 171, 117–122; (b) S. Sitthisa and D.
E. Resasco, Catal. Lett., 2011, 141, 784–791.
12 (a) K. J ¨a hnisch, V. Hessel, H. L o¨ we and M. Baerns, Angew.
Chem., Int. Ed., 2004, 43, 406–446; (b) B. P. Mason, K. E. Price, J.
L. Steinbacher, A. R. Bogdan and D. T. McQuade, Chem. Rev.,
2007, 107, 2300–2318; (c) D. M. Roberge, B. Zimmermann,
F. Rainone, M. Gottsponer, M. Eyholzer and N. Kockmann, Org.
Process Res. Dev., 2008, 12, 905–910.
13 L. Deng, Y. Zhao, J. Li, Y. Fu, B. Liao and Q. X. Guo,
ChemSusChem, 2010, 3, 1172–1175.
Acknowledgements
14 (a) L. E. Manzer, US Patent 0,055,270, 2003; (b) Z. P. Yan, L. Lin
and S. J. Liu, Energy Fuels, 2009, 23, 3853–3858; (c) X. L. Du,
L. He, S. Zhao, Y. M. Liu, Y. Cao, H. Y. He and K. N. Fan, Angew.
Chem., Int. Ed., 2011, 50, 1–6.
This work is supported by the National Natural Science
Foundation of China (50872086) and the Natural Science
Foundation of Shanxi Province (2008011018).
1
5 R. Rao, A. Dandekar, R. T. K. Baker and M. A. Vannice, J. Catal.,
997, 171, 406–419.
6 (a) S. Kannan, A. Dubey and H. Knozinger, J. Catal., 2005, 231,
81–392; (b) C. F. Schwenk and B. M. Rode, J. Am. Chem. Soc.,
004, 126, 12786–12787.
7 (a) F. Cavani, F. Trifir o` and A. Vaccari, Catal. Today, 1991, 11,
73–301; (b) W. L. Jolly, Modern Inorganic Chemistry, 2nd edn,
McGraw-Hill, New York, 1991, pp. 71–76.
1
1
Notes and references
3
2
1
G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106,
044–4098.
4
1
2
J. C. Serrano-Ruiz, R. M. West and J. A. Dumesic, Annu. Rev.
Chem. Biomol. Eng., 2010, 1, 79–100.
1
3
4
5
M. St o¨ cker, Angew. Chem., Int. Ed., 2008, 47, 9200–9211.
S. Dutta, RSC Adv., 2012, 2, 12575–12593.
Top value added chemicals from biomass, ed. T. Werpy and G.
Petersen, 2004, http://www1.eere.energy.gov/biomass/pdfs/
18 (a) F. Delbecq and P. Sautet, J. Catal., 1995, 152, 217–236; (b)
H. Li, H. Luo, L. Zhuang, W. Dai and M. Qiao, J. Mol. Catal. A:
Chem., 2003, 203, 267–275.
19 (a) M. Guisnet and P. Magnoux, Appl. Catal., A, 2001, 212,
83–96; (b) C. H. Bartholomew, Appl. Catal., A, 2001, 212, 17–60.
35523.pdf.
3
856 | RSC Adv., 2013, 3, 3853–3856
This journal is ß The Royal Society of Chemistry 2013