Enhanced hydrogenation of ethyl levulinate by Pd-AC doped with Nb 2O5

Article abstract of DOI:10.1039/c4gc00972j

Gamma-valerolactone (GVL), as sustainable feedstock for high-value chemicals and fuel, has been produced by hydrogenation of levulinic acid (LA) or ethyl levulinate (EL). In this work, Pd nanoparticles supported on Nb ₂O₅-doped activated carbon (AC) were prepared via wet incipient impregnation with PdCl₂ followed by reduction with NaBH₄. The dispersed niobia plays a bifunctional role in the catalytic process: stabilizing Pd nanoparticles and acting as an acidic co-catalyst. This synergistic effect between niobia and Pd leads to an unprecedented high activity of supported Pd catalysts in EL hydrogenation. The synergy is correlated with both the niobia loading and calcination temperature, with 10 wt% Nb₂O₅ calcined at 500 °C showing the best performance (EL conversion of 87% and GVL selectivity of 97%) under mild reaction conditions (100 °C and 0.5 MPa H₂). This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4gc00972j

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Enhanced hydrogenation of ethyl levulinate by
Pd–AC doped with Nb2O5†
Cite this: Green Chem., 2014, 16,
3951
Feiyang Ye, Damin Zhang, Teng Xue, Yimeng Wang and Yejun Guan*
Gamma-valerolactone (GVL), as sustainable feedstock for high-value chemicals and fuel, has been pro-
duced by hydrogenation of levulinic acid (LA) or ethyl levulinate (EL). In this work, Pd nanoparticles sup-
ported on Nb2O5-doped activated carbon (AC) were prepared via wet incipient impregnation with PdCl2
followed by reduction with NaBH4. The dispersed niobia plays a bifunctional role in the catalytic process:
stabilizing Pd nanoparticles and acting as an acidic co-catalyst. This synergistic effect between niobia and
Pd leads to an unprecedented high activity of supported Pd catalysts in EL hydrogenation. The synergy is
correlated with both the niobia loading and calcination temperature, with 10 wt% Nb2O5 calcined
at 500 °C showing the best performance (EL conversion of 87% and GVL selectivity of 97%) under mild
reaction conditions (100 °C and 0.5 MPa H2).
Received 27th May 2014,
Accepted 18th June 2014
DOI: 10.1039/c4gc00972j
www.rsc.org/greenchem
noticeably the most active one which gave a LA conversion of
80%. By further optimization, complete LA conversion was
Introduction
Cellulosic biomass is a vast and renewable resource that serves obtained with the GVL selectivity exceeding 95%. Compared
as feedstock to produce both chemicals and fuels.1,2 In the with Ru, Pt and Pd were reported to be much less active for
near future, catalytic conversion of biomass to a variety of GVL production, giving a moderate LA conversion.
chemicals will drastically change the energy-base from fossil
LA is generally produced from lignocellulose in aqueous
to renewable feedstock. In this context, many carbohydrate- acid medium, wherein polymeric humin is always formed by
derived compounds from lignocellulosic biomass are highly the thermal decomposition of intermediates, therefore leading
applicable and could be widely utilized. For example, it is to a low yield of LA.34–37 The humin formation can be signifi-
possible to produce ethanol and other platform chemicals, cantly inhibited in the presence of alcohols, resulting in the
such as levulinate ester, levulinic acid and 5-hydroxymethyl- formation of the corresponding levulinic esters in high yield
furfural (HMF) from glucose monomers obtained from ligno- (95%) compared with LA (25%) from C6 sugars.38 A 75% yield
cellulose.3 Further hydrogenation of levulinate ester or of methyl levulinate could be obtained directly from cellulose
levulinic acid then leads to the formation of γ-valerolactone with a mixed-acid system consisting of both Lewis and
(GVL), which is a highly stable and low toxic compound.4 This Brønsted acids.39 Li and coworkers40 reported an economical
compound has been widely used as a fuel,5 a fuel additive,6 an pathway for the production of pure levulinate esters from cellu-
intermediate for the production of chemicals,7–11 and a renew- lose. An optimized yield of around 50% was achieved at 210 °C
able solvent.6,12,13 Therefore, the production and application for 120 min with a very low sulfuric acid concentration
of GVL have attracted much attention in catalysis.4,11,13–27 (0.01 M) and an initial cellulose concentration below
Both homogeneous28 and heterogeneous29–32 catalysts have 100 g L−1. In fact, the easier separation of levulinate esters
been extensively explored.
from the reaction mixture and developing techniques to
To date, most studies have mainly focused on the conver- increase the recovery efficiency of alcohols during the syn-
sion of LA as a precursor. Manzer and coworkers systematically thesis will make the production of levulinate esters economi-
investigated the catalytic activity of various metals (e.g. Ir, Rh, cally favorable. Therefore the catalytic hydrogenation of
Pd, Ru, Pt, Re and Ni) supported on AC at 150 °C under levulinic esters to GVL may show great commercial potential
800 psi H2 for 2 h.11,33 Of the catalysts studied, Ru–C was due to suppression of active metal leaching of hydrogenation
catalysts caused by free carboxyl of LA.39,40
Nadgeri et al.41 reported the liquid-phase hydrogenation of
methyl levulinate over graphite- and zeolite-supported ruthe-
nium catalysts and the mixture of graphite-supported ruthe-
nium metal catalysts and zeolite in water. Two reaction
Shanghai Key Laboratory of Green Chemistry and Chemical Processes, East China
Normal University, North Zhongshan Road 3663, 200062 Shanghai, China.
E-mail: yjguan@chem.ecnu.edu.cn; Fax: +86-21 32530334
†Electronic supplementary information (ESI) available. See DOI:
pathways were found: direct hydrogenation to γ-valerolactone
10.1039/c4gc00972j
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3951–3957 | 3951

View Article Online
Paper
Green Chemistry
synthesis (3Pd–10Nb–500-AC), 0.291 g of the 10Nb–500-AC
support was distributed in 75 ml of H2O with stirring for 2 h.
An appropriate amount of H2PdCl4 aqueous solution was
added to the solution. After being stirred for 2 h, 0.6 ml of 1 M
NaOH aqueous solution was added to the mixture. Then, a
NaBH4 aqueous solution was added into the suspension as the
reducing reagent (NaBH4/Pd = 10, molar ratio) with stirring for
5 h. The slurry was filtered, washed with deionized water, and
finally dried at 60 °C for 24 h. For comparison, the catalytic
activity of a commercial catalyst, i.e., 5 wt% Pd–C from Alfa-
Aesar was tested.
Preparation of the Pd–niobia catalyst
For comparison, pure niobia supported Pd catalysts were pre-
pared by the same deposition–precipitation method. The
niobia supports included a commercial niobia (Nb2O5-COM)
purchased from Alfa-Aesar, and two other niobia samples pre-
pared with calcination of niobium oxalate at 300 and 500 °C
for 2 h (denoted Nb2O5-300 and Nb2O5-500, respectively).
Scheme 1 Conversion of ethyl levulinate to ethyl pentanoate.
and hydrogenation of carbonyl groups to methyl 4-hydroxy-
valerate. The latter compound undergoes dealcoholization to
form γ-valerolactone. From Scheme 1, one can clearly see that
the presence of zeolites may enhance the dealcoholization
step. This promotion effect of acidic materials has also been
observed in LA hydrogenation using USY, Amberlyst-15,
HZSM-5, and β-zeolite.
Recently, Xiong et al. developed a Nb2O5/C composite,
which showed high activity in the conversion of glucose to
5-hydroxymethylfurfural.42 Moreover, the carbon-loaded oxide
composites showed improved hydrothermal stability,43 which
is of great advantage regarding biomass conversion. When Pd
was supported on the obtained Nb2O5/C composite, GVL can
be successfully hydrogenated to pentanoic acid. Inspired by
these findings, herein we systematically studied the synergy
between Pd and Nb2O5 in hydrogenation of EL to GVL and PE.
We demonstrated that the addition of niobia significantly
improved the acidity of Pd/AC catalysts, leading to a fast
hydrogenation and dealcoholization in the overall process.
Consequently, the EL can be converted to GVL under mild
reaction conditions.
Catalyst characterization
Nitrogen adsorption–desorption isotherms at −196 °C were
obtained using BELSORP-Max equipment. Prior to the
measurement, the samples were first degassed at 150 °C under
vacuum for 6 h. Specific surface areas were calculated accord-
ing to the BET-method using five relative pressure points at an
interval of 0.05–0.30. The pore size distribution was obtained
by the BJH model applied to the adsorption branch of the
isotherm.
Pulse CO chemisorption was performed on a Micromeritics
AutoChem 2910 to determine the metal dispersion of the
reduced catalysts. Prior to measurement, 100 mg catalyst was
reduced in a flow of 100 mL min−1 10 vol% H2 in Ar at 200 °C
for 2 h and then flushed in He for 2 h. After being cooled to
30 °C in He, the CO gas pulses (5 vol% in He) were introduced
in a flow of 100 ml min−1. The changes in the CO gas phase
concentration were followed by a thermal conductivity detector
(TCD).
The acidities of the supports were characterized by NH3
temperature-programmed desorption (TPD) recorded on a
TP-5080 (Tianjin Xianquan, China) equipped with TCD. Before
the TPD experiments, the samples were pretreated under a
helium flow of 25 mL min−1 at 500 °C for 2 h. The physisorbed
NH3 was removed by flowing helium at 100 °C until a stable
signal was obtained. Desorption of strongly adsorbed NH3 was
performed by heating the catalyst at a rate of 10 °C min−1
under flowing helium (25 mL min−1), from room temperature
to 500 °C.
The powder X-ray diffraction (XRD) patterns were collected
on a Rigaku Ultima IV X-ray diffractometer using Cu Kα radi-
ation (λ = 1.5405 Å) operated at 35 kV and 25 mA. Scanning
electron microscopy (SEM) equipped with an energy dispersive
X-ray analysis (EDX) unit (Oxford, UK) was performed on a
Hitachi S-4800 microscope. Transmission electron microscopy
Experimental
Preparation of niobia–carbon supports
Niobia–carbon supports with various loading from 0 to 20 wt%
Nb2O5 were prepared by incipient impregnation. Active carbon
was purchased from STREM Chemicals and was dried before
use. A typical preparation of supports (10 wt% niobia loading)
involved the use of 2.61 g of active carbon which was mixed
with niobium oxalate hydrate (0.607 g) dissolved in 5 ml of
H2O. After being dried at 60 °C for 12 h, the sample was cal-
cined at 500 °C for 2 h with a ramp rate of 3 °C min−1 under a
N2 flow. The sample was denoted xNb–y-AC (x stands for
niobia loading, y stands for the calcination temperature).
Preparation of the Pd–niobia–carbon catalyst
The Nb–AC composites were used to load Pd with a loading of (TEM) images were taken on a FEI Tecnai G2 F30 microscope
3 wt% via a deposition–precipitation method. In a typical operating at 300 kV.
3952 | Green Chem., 2014, 16, 3951–3957
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
Catalytic activity measurements
Table 1 N2 adsorption data of the xNb–y-AC materials
Catalytic reactions for conversion of ethyl levulinate to γ-velero-
lactone were conducted in a Teflon-lined (100 mL) steel batch
reactor at a stirring speed of 800 rpm for 5 h. The typical
hydrogenation conditions were: 100 °C; 10 ml of 0.25 M ethyl
levulinate aqueous solution; 25 mg of catalyst; and H2 pressure
0.5 MPa. Once the reactor was cooled down to room tempera-
ture, the products were analyzed with an FID detector and a
capillary column DB-Wax. The following temperature program
method was used for GC analysis: 120 °C for 16 min and then
heated to 220 °C with a ramp rate of 20 °C min−1 and kept at
220 °C for another 9 min.
SBET
Vtotal
Pore
TNH Acid density
3
(m2 g−1
)
(cm3 g−1
)
size (nm) (°C) (mmol gcat )
−1
Sample
AC
1204
0.79
0.75
0.73
0.71
0.57
0.60
0.79
2.6
2.5
2.5
2.5
2.4
2.5
2.6
—
—
n.d.a
—
2.5Nb–500-AC 1184
5Nb–500-AC 1121
10Nb–500-AC 1116
—
—
245
230
200
—
0.097
0.17
0.29b
n.d.
20Nb–500-AC
10Nb–300-AC
926
956
10Nb–700-AC 1205
a Not detected. b Pretreated at 300 °C.
Results and discussion
Catalyst characterization
Fig. 1 shows the XRD patterns of the niobia–carbon supports
calcined at 300, 500 and 700 °C. Some unidentified diffraction
peaks were observed on the commercial AC, which might be
ascribed to impurities. Except for these peaks, no extra peak
was seen for the supported niobia species when calcined at
300 or 500 °C, meaning that Nb2O5 were highly dispersed in
the amorphous phase. Previous studies have shown that
niobium oxalate can be completely decomposed by calcination
at around 300 °C.47,48 In contrast, the 10Nb–700-AC catalyst
showed several diffraction peaks at 22 and 28° corresponding
to the diffractions of the hexagonal TT-Nb2O5 phase (JCPDS
[07-0061]).29
Fig. 2 NH3-TPD profiles of the supports.
The textural properties of AC and Nb2O5 doped AC, includ-
ing BET specific surface area (SSA), total pore volume (Vtotal
)
respectively (Fig. S2 in ESI†). We surmise that the highly dis-
persed amorphous Nb2O5 might block the mesopores to some
extent and this blockage effect disappears when Nb2O5 turns
to bulk material. This phenomenon has also been observed by
the Datye group.44
and the pore size are summarized in Table 1. Similar N2
adsorption isotherms (Type I) were observed for all AC based
materials (Fig. S1 in ESI†), which are characteristic of meso-
porous materials. Loading Nb2O5 led to a gradual decrease of
surface area from 1204 (AC) to 926 m2 g−1 (20 wt% Nb2O5–AC).
It should be noted that the calcination temperature also
affected the surface area. The surface areas of 10Nb–y-AC (y =
The acid densities of xNb–y-AC materials calculated accord-
ing to NH3-TPD experiments are also shown in Table 1 (Fig. 2).
AC did not show any NH3 desorption at temperatures higher
than 100 °C. The Nb2O5-COM also showed very weak NH3 de-
sorption. The 10Nb–AC supports calcined at 300 and 500 °C
gave acid densities of 0.29 and 0.097 mmol g−1, respectively.
This means that the acid density of Nb2O5 decreased during
the calcination. In line with this finding, the 10Nb–AC calcined
at 700 °C showed negligible desorption of NH3. Another
important feature of acidity is the acidic strength which is
revealed by the desorption peak of NH3. The maximum de-
sorption peaks of 10Nb–AC calcined at 300 and 500 °C are
about 200 and 245 °C, respectively. This result indicates that
the niobia species calcined at 500 °C showed stronger acidity
than that calcined at 300 °C as reported by Iizuka and co-
workers.45,46 The acidity of niobia is mainly attributed to its
Brønsted acid sites and medium-strong Lewis acid sites due to
its coordinatively unsaturated Nb5+ species.47 For the Nb–AC
catalyst calcined at 500 °C, the 20Nb–500-AC contained more
amounts of acid site (0.17 mmol gcat−1) due to higher niobia
300, 500, and 700 °C) supports are 956, 1116, 1205 m2 g−1
,
Fig. 1 XRD patterns of 10Nb–y-AC supports calcined at different temp-
eratures (y = 300, 500, and 700 °C).
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3951–3957 | 3953

View Article Online
Paper
Green Chemistry
Fig. 4 XRD patterns of pure AC and Pd–xNb–500-AC catalysts with
various niobia loadings: (a) 3Pd–AC; (b) 3Pd–2.5Nb–500-AC; (c) 3Pd–
5Nb–500-AC; (d) 3Pd–10Nb–500-AC; (e) 3Pd–20Nb–500-AC; and
(f) AC.
Fig. 3 SEM images of 10Nb–y-AC supports calcinated at different
temperatures: (a) 300 °C; (b) 500 °C; (c) 700 °C.
loading. The maximum NH3 desorption temperature of
20Nb–500-AC was about 230 °C, slightly lower than that of
10Nb–500-AC.
SEM images (Fig. 3) show that morphologies of supports
were similar to each other when the calcination temperature
did not exceed 500 °C. The EDX data (Table S1 in ESI†)
recorded in different zones of the 10Nb–500-AC support
(Fig. S3 in ESI†) suggested that the Nb2O5 was indeed well-
dispersed on AC without any aggregation. Interestingly, slice-
like and rod-like structures were clearly observed for the
sample calcined at 700 °C. Nowak et al.47 reported that Nb2O5
had strong mobility, which will explain the crystallization of
niobia on the 10Nb–700-AC support.
Fig. 4 displays the XRD patterns of the Pd–xNb–y-AC cata-
lysts. The diffraction peak at 40° is due to the (111) reflection
of Pd nanoparticles. The peak intensity became weaker with
the increase of Nb2O5 content, indicating that Pd nanoparti-
cles are well dispersed on Nb2O5 attributed to the so-called
strong metal–support interaction (SMSI). Fig. 5 shows the
typical TEM images of supported Pd catalysts. No visible Pd
particles were found, consistent with that of XRD. Thus the
accurate size of Pd cannot be determined by XRD and TEM.
The supported Pd catalysts were subjected to CO chemisorp-
tion measurements and the results are shown in Table 2. The
dispersions of Pd nanoparticles on 3Pd–AC, 3Pd–2.5Nb–500-
AC, 3Pd–5Nb–500-AC, 3Pd–10Nb–500-AC, 3Pd–20Nb–500-AC,
3Pd–10Nb–300-AC and 3Pd–10Nb–700-AC are 53%, 56%, 59%,
65%, 78%, 35%, and 52%, respectively. Apparently the pres-
ence of Nb2O5 indeed benefits the loading of Pd and results in
very fine Pd particles with a diameter of about 1.4 nm. Further-
more, the calcination temperature of Nb2O5 greatly influences
Fig. 5 TEM image of 3Pd–10Nb–500-AC (scale bar: 20 nm).
Table 2 CO chemisorption for the Pd–xNb–y-AC catalysts
Sample
Dispersion (%)
DPd (nm)
3Pd–AC
53(32a)
56
2.1(3.5a)
2.0
3Pd–2.5Nb–500-AC
3Pd–5Nb–500-AC
3Pd–10Nb–500-AC
3Pd–20Nb–500-AC
3Pd–10Nb–300-AC
3Pd–10Nb–700-AC
59
1.9
65(59b)
78
1.7(1.9b)
1.4
35
52
3.2
2.2
a Used for five times. b Used for six times.
the dispersion of Pd nanoparticles, with 500 °C being the best
one.
Catalyst performance of EL hydrogenation with different
niobia contents
The hydrogenation activities of ethyl levulinate to γ-valero-
lactone on 3Pd–xNb–500-AC are listed in Table 3. Commercial
Pd–AC exhibited the lowest activity with a GVL yield of 21%.
Similarly, 3 wt% Pd–AC also gave a low yield of 28%. It should
be noted that two side-products, namely LA and ethyl 4-hydroxy-
pentanoate (EHP) were also detected in varying amounts
3954 | Green Chem., 2014, 16, 3951–3957
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
3Pd–10Nb–500-AC catalyst was 87%, which was higher than
those of 3Pd–10Nb–300-AC (61%) and 3Pd–10Nb–700-AC
(68%). This result emphasized the importance of calcination
temperature in governing the catalytic activity of the Pd–Nb–
AC catalyst. The niobia phase changed from amorphous
to crystalline when calcination temperature increased from
300 to 700 °C. This change may affect the electronic properties
of supported Pd and also the acidic properties of the catalysts.
The lower EL conversion on Pd–10Nb–300-AC might be due
to the larger Pd particles (3.4 nm), whereas the presence of
bulk Nb2O5 (Fig. 2) may be responsible for the activity decrease
of Pd–10Nb–700-AC which is consistent with the activities of
Pd supported on bulk Nb2O5 catalysts (Table S3 in ESI†). On
the bulk Nb2O5 support, no XRD diffraction of Pd was
observed (Fig. S4 in ESI†), meaning the Pd species were well
dispersed. To shed more light on the interaction of Pd and
niobia on the hydrogenation activity, we have compared the
reactivity of physical mixtures of Pd–AC and 10Nb–y-AC
(Table 3). The EL conversion on the Pd–AC catalyst increased
from 38% to 57% in the presence of 10Nb–y-AC (y = 300 and
500). The promotion effect is weakened for 10Nb–700-AC.
These results suggest that the 10Nb–y-ACs were indeed
involved in the hydrogenation process. However the activities
of the physical mixtures were lower than their counterparts
with close interactions between Pd and niobia, namely,
Pd–10Nb–y-ACs, pointing to the synergistic effect of Pd and
niobia on AC. It is worth noting that the bulk Nb2O5 is
probably not in favor of strong Pd–Nb2O5 interaction.
It is clearly seen that the Nb–y-AC supports were involved in
the overall hydrogenation process. To clarify the role of Nb–y-
AC in the process, we compared the performance of 10Nb–y-AC
and AC in catalytic hydrolysis of EL and dealcoholization of
EHP, which are two possible reaction pathways involved in the
process. (Table S4 in ESI†). The results suggest that a minute
increase in the hydrolysis rate of EL was observed when niobia
was loaded. In another experiment, the trend of the deal-
coholization activity from EHP to GVL follows: 10Nb–500-AC
(90%) > 10Nb–300-AC (86%) > 10Nb–700-AC (56%) > AC (51%).
This trend is very similar to that of the acidity determined by
NH3-TPD, meaning that the surface acidity is essential for the
GVL formation. It should be noted that though 10Nb–700-AC
showed negligible NH3 desorption in TPD measurement, it
still showed excellent activity in dealcoholization of EHP and
as a co-catalyst in EL hydrogenation in aqueous solution. This
unexpected behavior in aqueous solution of Nb–700-AC is
probably related to its NbO4 tetrahedra of Nb2O5·xH2O with
water-tolerant acidic properties as reported by Hara and co-
workers.50 We consider that the acidity of niobia hydrate may
not be concisely justified by NH3-TPD and further studies
should be carried out to clarify the acid catalyzed mechanism
of niobia-involved reactions in aqueous solution.
Table 3 EL hydrogenation on supported Pd catalysts
Sel. (%)
Catalysta
Conv. (%)
LA
GVL
EHP
TOFb
5Pd–ACc
3Pd–AC
30
38
68
74
87
83
61
68
57
57
53
7
3
3
3
2
4
4
3
4
2
3
70
74
93
94
93
92
92
79
94
92
85
23
23
4
3
5
4
4
18
2
6
0.19
0.25
0.44
0.48
0.57
0.53
0.39
0.43
—
3Pd–2.5Nb–500-AC
3Pd–5Nb–500-AC
3Pd–10Nb–500-AC
3Pd–20Nb–500-AC
3Pd–10Nb–300-AC
3Pd–10Nb–700-AC
3Pd–AC + 10Nb–300-ACd
3Pd–AC + 10Nb–500-ACd
3Pd–AC + 10Nb–700-ACd
—
—
12
a Reaction conditions: 10 ml of 0.25 M EL aqueous solution; 0.5 MPa
H2; 100 °C; 25 mg catalyst (3 wt% Pd); 5 h; 800 rpm. b The TOF is
defined as molGVL+EHP gPd−1 h−1 c 5Pd–AC catalyst was purchased from
.
Alfa-Aesar, 15 mg. d Physical mixture of 25 mg 3Pd–AC and 25 mg
10Nb–y-AC.
depending on the nature of the AC used. The selectivity to
EHP on commercial 5 wt% Pd–AC was as high as 23%, mean-
while it was also observed as the main side-product (selectivity
of 23%) on home-made 3 wt% Pd–AC. The very similar selec-
tivity to EHP is probably determined by the thermodynamics of
the dealcoholization from EHP to GVL in the absence of acid
catalysts. The formation of LA might be ascribed to the hydro-
lysis of EL in aqueous solution. With 2.5 wt% Nb2O5 loaded on
AC, an EL conversion of 68% and a GVL selectivity of 93% were
achieved on 3Pd–2.5Nb–500-AC. The activities went up along
with the increase of niobia content. The highest yield of GVL
(81%) was obtained on the 3Pd–10Nb–500-AC catalyst. The
−1
TOF based on Pd loading is calculated to be 0.57 (molGVL gPd
h−1). This value is close to previous reports on Pd/SiO2 and
Pd/CNT catalysts under a much harsher condition for the
hydrogenation of LA, wherein higher temperature (140–200 °C)
and higher H2 pressure (30–90 bar) were employed.48,49 This
result clearly demonstrates the advantage of using EL as the
starting compound to produce GVL in biomass transform-
ation. To clarify the difference in supported Pd catalysts in EL
and LA hydrogenation, we have compared the catalytic activity
of Pd–AC and Pd–Nb–AC in LA and EL hydrogenation
(Table S2 in ESI†). The conversion of LA and EL on the Pd–AC
catalyst was 20% and 38%, respectively. The relative lower
hydrogenation activity of LA might be ascribed to the strong
adsorption ability of the carboxylate group on Pd nanoparti-
cles, which has been reported in some hydrogenation reac-
tions involving HCOOH.23 On the Pd–Nb–AC catalyst, the EL
and LA conversion was 87% and 62%, respectively. This result
may be explained by the effect of Nb2O5 as an acidic promoter
in hydrogenating carbonyl bonds.16
Catalyst performance of EL hydrogenation with different
Catalyst reusability
calcination temperatures of niobium oxide
Given the excellent performance of 3Pd–10Nb–500-AC in
Also shown in Table 3 are the EL conversions on Pd–10Nb– hydrogenation, its reusability was also of interest from the
y-AC (y = 300, 500, and 700). The GVL conversion over the viewpoint of practical application. After the first run, the cata-
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3951–3957 | 3955

View Article Online
Paper
Green Chemistry
Fig. 7 Recycle study of 3Pd–AC in EL hydrogenation. Reaction con-
ditions: 10 ml of 0.25 M EL aqueous solution; 0.5 MPa H2; 100 °C; 25 mg
catalyst (3 wt% Pd); 5 h; 800 rpm. The TOF is defined as molGVL+EHP
Fig. 6 Recycle study of 3Pd–10Nb–500-AC in EL hydrogenation. Reac-
tion conditions: 10 ml of 0.25 M EL aqueous solution; 0.5 MPa H2;
100 °C; 25 mg catalyst (3 wt% Pd); 5 h; 800 rpm. The TOF is defined as
gPd−1 h−1
.
molGVL+EHP gPd−1 h−1
.
The reaction was conducted in a 50 mL autoclave at 250 °C for
10 h using a hydrogen pressure of 20 bar and a stirring speed
of 300 rpm. A 70% yield of PE was obtained. This result
suggested that Nb doped Pd nanoparticle showed similar reac-
tivity to Pt and Ru in PE production from GVL or EL, thereby
showing promising practical application in future biomass
conversion.
lyst was separated through centrifugation, thoroughly washed
and then dried at 50 °C in a vacuum oven before the next use.
The activity in terms of EL conversion is shown in Fig. 6. It can
be clearly seen that the 3Pd–10Nb–500-AC catalyst showed very
stable reactivity in the first 4 runs and starts to deactivate in
the next two runs. EL conversion of 80% was achieved in the
6th run. Chemisorption measurement suggested that the Pd
dispersion was 59% for the spent sample, slightly decreased in
comparison with the fresh one (65%, Table 2). An SEM-EDX
study on the spent catalyst (not shown) was conducted and a
negligible change of Pd content was observed. For comparison,
Conclusions
the reusability of Pd–AC was also tested and the results are We demonstrated that the hydrogenation of aqueous ethyl
shown in Fig. 7. EL conversion on 3Pd–AC remained constant levulinate to γ-valerolactone can be successfully performed
in the first 3 runs and then significantly decreased in the next under mild conditions over the Pd–AC catalyst doped with
2 runs. CO chemisorption result suggested a remarkable Nb2O5. Unlike bulk Nb2O5, the homogeneously dispersed
decrease of Pd dispersion 53% (fresh) to 32% after 5 runs Nb2O5 plays a bifunctional role in stabilizing the Pd nanoparti-
(Table 2). This result demonstrates the importance of cles in the range of 1.4–2.5 nm and acting as an acidic promo-
niobia species in stabilizing Pd during the hydrogenation ter enhancing the formation of GVL from both EL and LA. The
process.29,44,51
promoting effect of Nb2O5 is dependent on the surface loading
and calcination temperature during the preparation. The
highest reactivity was found for 3 wt% Pd supported on 10 wt%
Nb2O5–AC calcined at 500 °C, with GVL yield of 81% at
100 °C under a substrate–Pd mole ratio of 355. Furthermore,
this catalyst can catalyze the conversion of GVL to PE in
ethanol. Our studies showed important implications for the
development of catalysts for bio-renewable conversions.
Ring-opening of GVL
Lange et al.40 disclosed the production of pentanoate (PA)
based biofuels from LA via hydrogenation in different alcohols.
To this end, a multifunctional Pt catalyst was employed,
wherein Pt-TiO2 catalyzing LA hydrogenation to GVL and
zeolite (i.e. H-ZSM-5) catalyzing the GVL ring-opening to PA
and esterification of PA to give ethyl pentanoate (PE). In this
case, a 70% yield of PA was achieved with a Pt-H-ZSM-5-SiO2
catalyst.52 This process could also be performed in a one-pot
reaction from ethyl levuliante (EL) over the Ru-H-β-silica cata-
Acknowledgements
lyst, with PA and PE yield of 62%.53 In our work, ring-opening We acknowledge the financial support from the National
reaction of GVL to PE was performed with GVL (10 mmol) in Natural Science Foundation of China (grant no. 21203065) and
EtOH (100 mmol) over 3Pd–10Nb–500-AC catalysts (100 mg). the Fundamental Research Funds for the Central Universities.
3956 | Green Chem., 2014, 16, 3951–3957
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
27 J. P. Lange, J. Z. Vestering and R. J. Haan, Chem. Commun.,
Notes and references
2007, 3488.
1 E. I. Gurbuz, D. M. Alonso, J. Q. Bond and J. A. Dumesic, 28 M. Chalid, A. A. Broekhuis and H. J. Heeres, J. Mol. Catal.
ChemSusChem, 2011, 4, 357. A: Chem., 2011, 341, 14.
2 A. J. J. E. Eerhart, W. J. J. Huijgen, R. J. H. Grisel, J. C. van 29 R. Buitrago-Sierra, J. C. Serrano-Ruiz, F. Rodriguez-Reinoso,
der Waal, E. de Jong, A. de Sousa Dias, A. P. C. Faaij and
M. K. Patel, RSC Adv., 2014, 4, 3536.
A. Sepulveda-Escribano and J. A. Dumesic, Green Chem.,
2012, 14, 3318.
3 H. Rasmussen, H. R. Sorensen and A. S. Meyer, Carbohydr. 30 S. G. Wettstein, J. Q. Bond, D. M. Alonso, H. N. Pham,
Res., 2014, 385, 45.
4 D. M. Alonso, S. G. Wettstein and J. A. Dumesic, Green
Chem., 2013, 15, 584.
5 J. Q. Bond, D. M. Alonso, D. Wang, R. M. West and
J. A. Dumesic, Science, 2010, 327, 1110.
A. K. Datye and J. A. Dumesic, Appl. Catal., B, 2012,
117–118, 321.
31 P. P. Upare, J.-M. Lee, D. W. Hwang, S. B. Halligudi,
Y. K. Hwang and J.-S. Chang, J. Ind. Eng. Chem., 2011, 17,
287.
6 I. T. Horvath, H. Mehdi, V. Fabos, L. Boda and L. T. Mika, 32 H. Heeres, R. Handana, D. Chunai, C. Borromeus Rasren-
Green Chem., 2008, 10, 238.
7 D. Cerniauskaite, J. Rousseau, A. Sackus, P. Rollin and
A. Tatibouet, Eur. J. Org. Chem., 2011, 2293.
dra, B. Girisuta and H. Jan Heeres, Green Chem., 2009, 11,
1247.
33 L. E. Manzer, WO 2002074760A1, 2002.
8 R. H. Guo, Q. Zhang, Y. B. Ma, X. Y. Huang, J. Luo, 34 M. J. Climent, A. Corma and S. Iborra, Green Chem., 2014,
L. J. Wang, C. A. Geng, X. M. Zhang, J. Zhou, Z. Y. Jiang
and J. J. Chen, Bioorg. Med. Chem., 2011, 19, 1400.
9 Z. Ma, Y. Hong, D. M. Nelson, J. E. Pichamuthu,
16, 516.
35 L. Peng, L. Lin, J. Zhang, J. Shi and S. Liu, Appl. Catal., A,
2011, 397, 259.
C. E. Leeson and W. R. Wagner, Biomacromolecules, 2011, 36 B. C. Windom, T. M. Lovestead, M. Mascal, E. B. Nikitin
12, 3265. and T. J. Bruno, Energy Fuels, 2011, 25, 1878.
10 I. van der Meulen, E. Gubbels, S. Huijser, R. Sablong, 37 B. Kim, J. Jeong, S. Shin, D. Lee, S. Kim, H. J. Yoon and
C. E. Koning, A. Heise and R. Duchateau, Macromolecules,
2011, 44, 4301.
11 L. E. Manzer, Appl. Catal., A, 2004, 272, 249.
12 P. G. Jessop, Green Chem., 2011, 13, 1391.
J. K. Cho, ChemSusChem, 2010, 3, 1273.
38 X. Hu and C. Z. Li, Green Chem., 2011, 13, 1676.
39 K.-i. Tominaga, A. Mori, Y. Fukushima, S. Shimada and
K. Sato, Green Chem., 2011, 13, 810.
13 D. Fegyverneki, L. Orha, G. Láng and I. T. Horváth, Tetra- 40 J. P. Lange, R. Price, P. M. Ayoub, J. Louis, L. Petrus,
hedron, 2010, 66, 1078.
14 Z. Yang, Y. B. Huang, Q. X. Guo and Y. Fu, Chem. Commun.,
2013, 49, 5328.
15 M. G. Al-Shaal, W. R. H. Wright and R. Palkovits, Green
Chem., 2012, 14, 1260.
16 A. M. R. Galletti, C. Antonetti, V. De Luise and
M. Martinelli, Green Chem., 2012, 14, 688.
L. Clarke and H. Gosselink, Angew. Chem., Int. Ed., 2010,
49, 4479.
41 J. M. Nadgeri, N. Hiyoshi, A. Yamaguchi, O. Sato and
M. Shirai, Appl. Catal., A, 2014, 470, 215.
42 H. Xiong, T. Wang, B. H. Shanks and A. K. Datye, Catal.
Lett., 2013, 143, 509.
43 H. Xiong, H. N. Pham and A. K. Datye, J. Catal., 2013, 302, 93.
17 W. R. Wright and R. Palkovits, ChemSusChem, 2012, 5, 44 H. Xiong, M. Nolan, B. H. Shanks and A. K. Datye, Appl.
1657.
Catal., A, 2014, 471, 165.
18 A. M. Hengne and C. V. Rode, Green Chem., 2012, 14, 1064.
19 W. Li, J.-H. Xie, H. Lin and Q.-L. Zhou, Green Chem., 2012,
14, 2388.
45 G. S. Nair, E. Adrijanto, A. Alsalme, I. V. Kozhevnikov,
D. J. Cooke, D. R. Brown and N. R. Shiju, Catal. Sci.
Technol., 2012, 2, 1173.
20 M. Chia and J. A. Dumesic, Chem. Commun., 2011, 47, 46 T. Iizuka, K. Ogasawara and K. Tanabe, Bull. Chem. Soc.
12233. Jpn., 1983, 56, 2927.
21 H. N. Pham, Y. J. Pagan-Torres, J. C. Serrano-Ruiz, 47 I. Nowak and M. Ziolek, Chem. Rev., 1999, 99, 3603.
D. Wang, J. A. Dumesic and A. K. Datye, Appl. Catal., A, 48 K. Yan, T. Lafleur and J. Liao, J. Nanopart. Res., 2013, 15,
2011, 397, 153.
22 X. L. Du, Q. Y. Bi, Y. M. Liu, Y. Cao and K. N. Fan, Chem- 49 K. Yan, T. Lafleur, G. Wu, J. Liao, C. Ceng and X. Xie, Appl.
SusChem, 2011, 4, 1838. Catal., A, 2013, 468, 52.
23 L. Deng, Y. Zhao, J. Li, Y. Fu, B. Liao and Q. X. Guo, Chem- 50 K. Nakajima, Y. Baba, R. Noma, M. Kitano, J. N. Kondo,
SusChem, 2010, 3, 1172. S. Hayashi and M. Hara, J. Am. Chem. Soc., 2011, 133, 4224.
24 L. Deng, J. Li, D. M. Lai, Y. Fu and Q. X. Guo, Angew. Chem., 51 W. Luo, U. Deka, A. M. Beale, E. R. H. van Eck,
1906.
Int. Ed., 2009, 48, 6529.
25 H. Mehdi, V. Fabos, R. Tuba, A. Bodor, L. T. Mika and
I. T. Horvath, Top. Catal., 2008, 48, 49.
P. C. A. Bruijnincx and B. M. Weckhuysen, J. Catal., 2013,
301, 175.
52 J.-P. Lange, US Patent 2011/0112326, 2011.
26 R. A. Bourne, J. G. Stevens, J. Ke and M. Poliakoff, Chem. 53 P. J. van den Brink, K. L. Von Hebel, J.-P. Lange and
Commun., 2007, 4632.
L. Petrus, WO/2006/[067171], 2006.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3951–3957 | 3957