Catalytic conversion of biomass-derived levulinic acid to valerate esters as oxygenated fuels using supported ruthenium catalysts

Article abstract of DOI:10.1039/c3gc40927a

The development of the catalytic conversion of biomass-based platform molecules into oxygenated fuel molecules is of great significance in order to reduce the dependence on fossil resources and to solve environmental problems. Alkyl valerate esters were proven to have the potential to be renewable additives of gasoline and diesel. In this work, we studied the hydrogenation of levulinic acid (LA) to valerate esters over supported Ru catalysts, and found that the acidity was an important factor for the catalyst performance. A bifunctional catalyst Ru/SBA-SO₃H was developed as an active catalyst, and a highest yield of 94% to ethyl valerate (EV) was achieved. The catalyst was characterized by nitrogen adsorption/desorption methods, X-ray power diffraction (XRD), transmission electron spectroscopy (TEM) and X-ray photoelectron spectroscopy (XPS). The effects of reaction conditions were comprehensively investigated and probable reaction pathways were proposed and verified. The conversion of LA to various alkyl valerate esters can also be catalyzed by the bifunctional catalyst. In addition, supported Cu and Ni catalysts were also screened under similar reaction conditions as Ru-based catalysts, and the combination of Ni/SBA-15 and SBA-SO₃H exhibited activity for the conversion of LA to EV.

Full text of DOI:10.1039/c3gc40927a

Green Chemistry
View Article Online
View Journal
PAPER
Catalytic conversion of biomass-derived levulinic acid
to valerate esters as oxygenated fuels using supported
ruthenium catalysts†
Cite this: DOI: 10.1039/c3gc40927a
Tao Pan, Jin Deng, Qing Xu, Yang Xu, Qing-Xiang Guo and Yao Fu*
The development of the catalytic conversion of biomass-based platform molecules into oxygenated fuel
molecules is of great significance in order to reduce the dependence on fossil resources and to solve
environmental problems. Alkyl valerate esters were proven to have the potential to be renewable addi-
tives of gasoline and diesel. In this work, we studied the hydrogenation of levulinic acid (LA) to valerate
esters over supported Ru catalysts, and found that the acidity was an important factor for the catalyst
3
performance. A bifunctional catalyst Ru/SBA-SO H was developed as an active catalyst, and a highest
yield of 94% to ethyl valerate (EV) was achieved. The catalyst was characterized by nitrogen adsorption/
desorption methods, X-ray power diffraction (XRD), transmission electron spectroscopy (TEM) and X-ray
photoelectron spectroscopy (XPS). The effects of reaction conditions were comprehensively investigated
and probable reaction pathways were proposed and verified. The conversion of LA to various alkyl
valerate esters can also be catalyzed by the bifunctional catalyst. In addition, supported Cu and Ni
catalysts were also screened under similar reaction conditions as Ru-based catalysts, and the combination
Received 18th May 2013,
Accepted 13th August 2013
DOI: 10.1039/c3gc40927a
www.rsc.org/greenchem
3
of Ni/SBA-15 and SBA-SO H exhibited activity for the conversion of LA to EV.
6
7
acid (LA) and sorbitol ) into liquid hydrocarbon fuels and
1
Introduction
8
other oxygenated fuels.
Since the increasing consumption of fossil resources has
Among these platform molecules serving as “bridges”
caused many problems such as environmental pollution and which connect the biomass feedstock with liquid fuels and
carbon dioxide emissions, the development of renewable chemicals, levulinic acid occupies a prominent position since
resources is one part of the driving initiative to reduce the it can be cost-effectively produced from lignocellulosic
1
9
dependence on fossil resources. Compared with existing materials via an acid hydrolysis process in high yields.
renewable energy sources such as solar power, wind power and Additionally, the LA derivatives which feature proper combus-
water power, carbon-neutral biomass is an attractive energy tion properties and energy density have the potential to serve
2
10
and organic carbon source for our industrial society. To this as fuel additives (e.g. ethyl levulinate (EL), γ-valerolactone
1
1
12
end, many efforts have been made to convert lignocellulosic (GVL), and 2-methyl tetrahydrofuran (MTHF) in Scheme 1).
biomass, which is a kind of non-food-derived biomass, into However, the performance of these molecules as fuels in an
3
liquid fuels and chemicals.
In the same manner, within the petrochemical industry, a
few building blocks which can be derived from lignocellulosic
biomass and its derivatives with low cost have been identified
4
and proposed for the construction of bio-refinery processes.
In the last decade, a variety of strategies have been proposed to
convert these building blocks (also called “platform mole-
5
cules”, such as 5-hydroxymethyl furfural (5-HMF), levulinic
Anhui Province Key Laboratory of Biomass Clean Energy, Department of Chemistry,
University of Science and Technology of China, Hefei 230026, China.
E-mail: fuyao@ustc.edu.cn; Fax: (+86)-551-63606689
†
Electronic supplementary information (ESI)
0.1039/c3gc40927a
available. See DOI:
1
Scheme 1 Conversion of LA to fuel additives.
This journal is © The Royal Society of Chemistry 2013
Green Chem.

View Article Online
Paper
Green Chemistry
internal combustion engine was not satisfactory due to their (P123, Aldrich) used as a template was added to 125 mL of
polarity, expansion coefficient or dissolubility to the 0.8 M HCl in a 500 mL round bottom flask and stirred at 313 K
1
3
traditional polymer materials.
for 2 h. After P123 was completely dissolved, 6.8 g of tetra-
Lange et al. proposed valerate esters as novel oxygenated ethoxysilane (TEOS, 32.8 mmol, SCRC) was added slowly and
1
3
fuels which could be blended in both gasoline and diesel.
the mixture was stirred for an additional 3 h at 313 K. After
They developed a value chain containing the multi-step cataly- pre-hydrolysis, the required amount of 3-mercaptopropyl tri-
tic conversion of levulinic acid to alkyl valerate esters over sup- methoxysilane (MPTMS, SCRC) and 15 mL of 30 wt% hydrogen
ported Pt catalysts. Besides, a comprehensive investigation of peroxide (H O ) were added and stirred at 313 K for 20 h. Then
2
2
fuel properties of valeric biofuels in gasoline or diesel was the resulting mixture was transferred to the hydrothermal reac-
carried out. Dayma et al. further studied the combustion pro- tion kettle with polytetrafluoroethylene (PTFE) lining and aged
1
4
perties of ethyl valerate. The results showed that valerate for 24 h at 373 K. The solid was filtered, washed with de-
esters were better than other cellulose-based candidate fuel ionized water and dried overnight at 363 K under vacuum. The
molecules both in terms of the production and fuel properties. copolymer template P123 was removed by refluxing the sample
Recently, Weckhuysen et al. employed Ru/H-ZSM-5 as a hetero- with 300 mL of ethanol for 24 h in a Soxhlet extractor. Finally,
geneous catalyst to directly convert LA to valeric acid (VA) and the extracted solid acid was collected and dried at 353 K under
1
5
its esters. A yield of 45.8% of VA was achieved in dioxane as vacuum. Here we reported four kinds of solid acids that are
solvent. Ravasio et al. reported a bifunctional catalyst Cu-SiZr designated SBA-SO H-x with increasing amounts of sulfonic
3
1
6
to convert GVL to valerate ester. Pentyl valerate was obtained acid, where x varies from 1 to 4. Accordingly, the amounts of
with 90% conversion and 83% selectivity. Therefore, the devel- MPTMS added were 0.2, 0.4, 0.8 and 2 g.
opment of a novel catalytic system for the conversion of levu-
linic acid to valerate esters in one pot with high efficiency is of ing the same procedure as SBA-SO
great significance for the application of valeric biofuels. MPTMS and H . AC-SO H was synthesized as follows: 5 g of
In this paper, we have studied the influence of catalyst sup- activated carbon was added to 100 mL of fuming sulfuric acid
ports and acidity on the hydrogenation of levulinic acid to (20 wt%) and stirred at 353 K for 12 h under N atmosphere.
Synthesis of a pure SBA-15 support was performed follow-
3
H without the addition of
2
O
2
3
2
alkyl valerate esters over supported Ru catalysts. A bifunctional After cooling to room temperature, the mixture was slowly
catalyst Ru/SBA-SO H was developed as the active catalyst and poured into the ice water and filtered to get the black powder.
3
9
4% yield to ethyl valerate (EV) was achieved. The effects of Finally, the solid acid was washed with hot water until no
reaction conditions were investigated comprehensively. Also, sulfate ions remained and subsequently dried at 353 K under
probable reaction pathways were proposed and verified. vacuum.
2
.2 Catalyst characterization
2
Experimental
The catalyst was characterized by nitrogen adsorption/
desorption methods, X-ray power diffraction (XRD), trans-
2
.1 Catalyst preparation
The reagents used in catalyst preparation and reactions were mission electron spectroscopy (TEM) and X-ray photoelectron
commercially available. Supported Ru catalysts were prepared spectroscopy (XPS). The amount of acid sites on the catalysts
by incipient wetness impregnation using an aqueous solution was determined by the sulphur content and XPS of S 2p,
of Ru(NO)(NO ) (Alfa Aesar) with metal loading of 5 wt%. The Hammett titration and acid–base titration. The detailed
3
3
supports used were MnO
x
(as prepared in ref. 17), MgO methods were presented in the ESI.†
(
(
(
Aladdin Reagent Inc.), CeO
x
(as prepared in ref. 18), ZSM-5
Si/Al = 50, Nankai University Catalyst Co., Ltd), HZSM-5
Si/Al = 50, Nankai University Catalyst Co., Ltd), ZrO (50 nm,
(40 nm, rutile, Studies of levulinic acid hydrogenation were performed in a
Aladdin Reagent Inc.), SiO2 (50 nm, Aladdin Reagent Inc.), 25 mL pressure vessel (Hastelloy, Parr Instrument). In a typical
Nb (Aladdin Reagent Inc.), C-SO H and SBA-SO H (as pre- procedure, 4 mmol of levulinic acid and 0.2 g of catalyst were
2.3 Reaction studies
2
hydrophilic, Aladdin Reagent Inc.), TiO
2
2
O
5
3
3
pared in ref. 19 and the details were as follows). After impreg- added into 10 mL of ethanol. In some experiments, a certain
nation, these catalysts were dried at 383 K overnight and amount of sulfuric acid (H SO ) or p-toluenesulfonic acid
2
4
reduced under the flow of H
2
(5%) and N
2
(95%) at 553 K for (p-TsOH) was added to the mixture. The reactor was sealed,
three times, and then pressurized with H to
−1
4
h with a heating rate of 2 K min . Ru/C (5 wt% of Ru) was purged with H
2
2
purchased from Aladdin Reagent and used directly. Ru-black 4 MPa. The mixture was heated to 523 K within 30 min and
was purchased from Alfa Aesar and used after reduction at held for a specified period of time. After reaction completion,
5
2
53 K for 4 h and passivation by using 2% O
98 K.
2
in helium at the reactor was cooled in an ice bath. The products were fil-
tered and analyzed by GC (Shimadzu GC-2014, FID, DM-WAX
3
Sulfonic acid-functionalized mesoporous silica (SBA-SO H) column) and GC-MS (Thermal Trace GC Ultra with a Polaris Q
1
9
inspired by Stucky and Shanks et al., was synthesized using a ion trap mass spectrometer equipped with a TR-5MS capillary
one pot procedure. In a typical synthesis, 4 g of Pluronic 123 column).
Green Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
unsupported Ru (Ru-black) also exhibited high selectivity to
GVL (Table 1, entries 6 and 7). Interestingly, similar results
were obtained using several typical oxide-based supports
3
Results and discussion
3
.1 Support effects with ruthenium catalysts
The hydrogenation of levulinic acid was performed with sup- (ZrO
ported Ru catalysts. In this study, the supports were carbon, to GVL remained almost constant. The supports with Lewis
various oxides, zeolite and mesoporous silica materials. acid sites (Nb ) and mesoporous structure (SBA-15) also pro-
2
2 2
, TiO and SiO , Table 1, entries 8–10) and the selectivity
2 5
O
Ethanol was employed as the reaction solvent because it could duced GVL as the main product (Table 1, entries 11 and 12).
be used as a reactant for the production of ethyl valerate (EV) Surprisingly, sulfonic acid-functionalized mesoporous silica
at the same time. The reaction temperature was 523 K because (SBA-SO
levulinic acid was mainly hydrogenated to γ-valerolactone conversion of LA (Table 1, entry 13), indicating that the
GVL) at low temperature. support with Brønsted acid sites had high activity
Product distributions for levulinic acid hydrogenation with to promote the conversion of LA to VA and EV. The detailed
3
H) exhibited high selectivity to VA and EV with 100%
(
a
3
Ru supported on various supports were shown in Table 1. information of catalyst characterization of SBA-SO H and
Levulinic acid was completely converted when catalyzed by Ru. Ru/SBA-SO H was shown in the ESI.†
3
Firstly, the blank experiment was performed and levulinic acid
almost remained unconverted (Table 1, entry 1). The results
3.2 The effects of acid intensity
indicated that the Hastelloy reactor had no catalytic activity for Based on the results that the acidity is an important factor in
the hydrogenation of levulinic acid. Subsequently, three cata- the conversion of LA to VA and EV, we explored the effects of
lysts consisting of Ru supported on metal oxide supports con- acid intensity (Table 2). To this end, a variety of supported
taining basic sites, MnO , MgO, CeO , were screened (Table 1, Ru catalysts with various solid acids and different amounts of
x
x
entries 2–4). The main products, EL, GVL, MTHF, valeric acid acid sites were synthesized.
VA) and EV which could be obtained from the hydrogenation
Initially, the controlled experiment was conducted to obtain
(
of levulinic acid were determined and presented in Table 1. an expected selectivity of EL (86%, Table 2, entry 1). Then a
It was interesting that the reaction in ethanol led to the co- series of Ru/SBA-SO H-x catalysts with different amounts of
3
production of VA and EV. This undesirable result may be attri- acid sites were used and the selectivity to GVL, VA and EV
buted to the equilibrium reaction of VA esterification to EV in varied considerably for the different catalysts (Table 2, entries
ethanol. The explanation was verified by the control experi- 2–5). Of these sulfonic acid-functionalized mesoporous silica
ments. EV and several active catalysts for the conversion of LA catalysts, the supports with high acidity produced VA and EV at
to EV (e.g. Ru/MnO , Ru/ZSM-5, Ru/SBA-SO H) were mixed in high selectivities from 54 to 90%. The selectivities to GVL
x
3
ethanol under the reaction conditions for 6 h and the mixture decreased with increasing the amount of acid sites.
of EV and VA were detected. The ratio of EV to VA was main-
To elucidate the effects of the structure of the supports, Ru/
tained within 2.5 : 1 and 3 : 1. Of the above oxide-based sup- C-SO H with microporous and mesoporous structure and Ru/
3
ports, the predominant product was GVL and Ru/MnO
low selectivity to VA and EV (15%).
x
had a HZSM-5 with microporous structure were performed (Table 2,
entries 6 and 7). Moderate selectivities to VA and EV were
Ru supported on zeolite ZSM-5 was tested and the main achieved and Ru/C-SO H got a higher selectivity to EV and VA
3
products were GVL (62%) with increasing the selectivity to VA due to its higher amount of acid sites than Ru/HZSM-5. Fur-
and EV to 29% (Table 1, entry 5). Ru supported on carbon and thermore, the hydrogenation catalyst combined with liquid
Table 1 Product distributions for LA hydrogenation with various supports. Reaction conditions: 4 mmol of LA, 0.2 g of catalyst, 10 mL of ethanol, 6 h, 523 K and
4
MPa H
2
Selectivity (%)
EL
a
Entry
Catalyst
LA conversion (%)
GVL
MTHF
VA + EV
Total
1
2
3
4
5
6
7
8
9
1
1
1
1
Blank
Ru/MnO
Ru/MgO
Ru/CeO
Ru/ZSM-5
Ru/C
4
100
100
100
100
100
99
100
100
100
99
87
1
0
1
0
0
1
2
3
0
82
92
88
62
91
92
81
80
73
66
70
4
0
1
0
1
8
1
2
5
4
0
15
3
6
29
1
2
4
6
10
3
87
99
95
96
99
93
97
92
93
95
89
89
95
x
x
Ru black
Ru/ZrO
Ru/TiO
Ru/SiO
Ru/Nb
Ru-SBA-15
Ru/SBA-SO
2
2
0
1
2
3
2
3
9
10
1
2
O
5
10
16
0
99
100
2
90
3
H-4
1
a
Ru loading was 5 wt% except Ru black.
This journal is © The Royal Society of Chemistry 2013
Green Chem.

View Article Online
Paper
Green Chemistry
Table 2 The effects of the acidity on LA hydrogenation to EV. Reaction conditions: 4 mmol of LA, 0.2 g of catalyst, 10 mL of ethanol, 6 h, 523 K and 4 MPa H
2
Selectivity (%)
a
Amount of acid sites (mmol g−1)
b
Entry
Catalyst
LA conversion (%)
EL
GVL
MTHF
VA + EV
Total
1
2
3
4
5
6
7
SBA-SO
3
H-4
1.20
0.17
0.39
0.46
1.14
0.66
0.28
1.50
3.00
1.50
93
99
99
100
100
99
99
99
100
100
86
1
1
1
0
1
0
45
36
19
0
35
29
14
4
34
43
26
33
31
0
1
0
1
1
2
1
5
4
4
0
54
63
76
90
55
49
8
86
91
93
92
95
92
93
84
82
70
Ru/SBA-SO
Ru/SBA-SO
Ru/SBA-SO
Ru/SBA-SO
Ru/C-SO
Ru/HZSM-5
Ru/C + H
Ru/C + H
Ru/C + p-TsOH
3
3
3
3
H-1
H-2
H-3
H-4
3
H
c
8
9
2
SO
SO
4
d
2
4
9
16
e
10
a
b
c
d
Ru loading was 5 wt%. The amount of acid sites (entries 1–7) were determined by Hammett titration. 0.15 mmol of H
2 4
SO .
0.30 mmol of
e
2 4
H SO . 0.3 mmol of p-TsOH.
acids (H SO , p-toluenesulfonic acid (p-TsOH)) at considerable
2
4
acidity were investigated (Table 2, entries 8–10). Unfortunately,
the major products were EL and GVL while the selectivity to VA
and EV was low. This may be attributed to the hydrogenation
of liquid acid catalyzed by Ru.
Clearly, the conversion of LA to VA and EV with supported
Ru catalysts in ethanol did not rely heavily on the structure of
the supports, rather on suitable acid strength. The most active
3
catalyst Ru/SBA-SO H-4 was chosen as the suitable catalyst for
further studies.
3
3
.3 Effects of reaction conditions with Ru/SBA-SO H
The effects of reaction conditions including reaction
temperature, initial hydrogen pressure, solvent and Ru loading
were investigated. A factorial design of experimental con-
ditions and corresponding results was shown in Table S4 in
the ESI.† The results indicated that the reaction temperature
and Ru loading had more significant effects on the
selectivity to EV and VA than initial hydrogen pressure and
solvent.
The product selectivities for the hydrogenation of LA as a
function of the reaction temperature for Ru/SBA-SO H catalyst
3
was shown in Fig. 1A. Complete conversion of LA was achieved
under the reaction conditions. The major product was GVL
under 483 K and the selectivity to GVL deceased with the reac-
tion temperature, especially in the range of 483–503 K. The
2
Fig. 1 Effects of temperature (A, 4 MPa H ) and initial hydrogen pressure (B,
selectivity to EV and VA improved as a function of temperature reaction temperature of 513 K) on the product selectivities for the hydrogen-
ation of LA at full conversion. Reaction conditions: 4 mmol of LA, 0.2 g of Ru
from 423 to 513 K and the highest selectivity of 94% was
3
(5 wt%)/SBA-SO H-4, 10 mL of ethanol and 6 h.
obtained at 513 K with 100% LA conversion. Therefore, 513 K
was selected as the optimized temperature in the following
study. These changes in selectivities to GVL, EV and VA indi-
cated that GVL was further converted into EV and VA with the added in 10 mL of ethanol and the mixture was heated to
increase of reaction temperature.
553 K for 6 h. After cooling, the recovery of EV and VA was only
However, the selectivity to EV and VA decreased at higher 51%. This result was consistent with that of hydrogenation of
temperature (from 523 to 553 K), presumably due to the LA at 553 K in Fig. 1A. Moreover, Ru/SBA-SO H used at 553 K
3
decomposition of EV and VA and the destruction of the struc- was analyzed by power XRD (2θ range was 0.5–10°) and the
ture of the catalyst at high temperature. In order to verify this (100) diffraction peaks between 0.8 and 0.9° disappeared
hypothesis, a decomposition experiment of EV in ethanol at (Fig. S8, ESI†), demonstrating that the structure of the support
5
53 K was conducted. 4 mmol of EV and 0.2 g of catalyst was was destroyed under the reaction conditions. Additionally,
Green Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
Ru/SBA-SO
in Fig. 1A.
3
H had low selectivities to EL and MTHF as shown 0.2 g of catalyst were added to 10 mL of water, THF or dioxane
and the mixture was heated to 513 K for 6 h. After the reaction,
Studies of the effects of initial hydrogen pressure were also the recoveries of VA were 90, 79 and 70%, respectively, which
carried out (Fig. 1B). At a hydrogen pressure of 2 MPa, com- were in agreement with the changes of the selectivities to VA in
plete conversion of LA was achieved. The primary products the three solvents.
formed were EV and VA (68% selectivity) and GVL (18% selecti-
Studies were also conducted to probe the influence on the
vity). The increase of hydrogen pressure from 2 to 4 MPa gave selectivity for the hydrogenation of LA by changing the loading
a rise of the selectivity to EV and VA, accompanied by a of Ru on the support SBA-SO H. The results were shown in
3
decrease of the selectivity to GVL, suggesting the full conver- Fig. 2B. After a reaction time of 6 h, almost all of LA was con-
sion of GVL to EV and VA under high hydrogen pressure. The verted in all reactions. Decreasing the loading of Ru to 1%
highest selectivity to EV and VA was 94% when initial hydrogen decreased the amount of EV and VA produced to a selectivity of
pressure was 4 MPa. After increasing the hydrogen pressure to 66%. Although a lower metal loading may lead to a higher
6
MPa, the selectivity to EV and VA did not change significantly dispersion of Ru on the support, the activity of the catalyst
and the other products remained unchanged.
decreased obviously. On the other hand, a decrease in the
selectivity to EV and VA (from 94 to 41%) was accompanied by
3
Fig. 2A showed the catalytic performance of Ru/SBA-SO H
in the different solvents. Three typical solvents (water, tetra- an increase in the Ru loading from 5 to 10 wt%, probably due
hydrofuran (THF) and dioxane) in the process of hydrogen- to an aggregation of Ru particles on the support based on the
ation were employed. LA was completely converted in each of results of 17% selectivity to GVL. Interestingly, 23% selectivity
the reactions. VA was the major product in three solvents. It to MTHF was observed, indicating over-hydrogenation of LA at
can be seen that the selectivity to VA decreased from 78 to 58% high Ru loading.
with the increase of the selectivity to the unidentified products
from 11 to 24% when the solvent was changed from water to 3.4 Probable reaction pathways for the conversion of LA
dioxane. The changes may be explained by the decomposition to EV
experiments of VA in the three solvents. 4 mmol of VA and
In order to elucidate the reaction pathways for the conversion
of LA to EV over Ru/SBA-SO
hydrogenation as a function of reaction time using Ru/
SBA-SO H as catalyst were shown in Fig. 3. At a reaction time
3
H, product distributions for LA
3
of 30 min, complete conversion of LA was achieved, and the
primary products observed were GVL, EV and VA at selectivities
of 66 and 20%, respectively. The increase in selectivity to EV
and VA with reaction time was almost linear from 30 to
360 min as the selectivity to GVL gradually decreased, and the
amount of EL and MTHF remained low. It was suggested that
the formation of EV and VA was based on the ring opening of
GVL over an acidic catalyst. The findings were consistent with
the product selectivities as a function of the reaction
temperature.
Probable reaction pathways for the conversion of LA to EV
over Ru/SBA-SO H were shown in Scheme 2. The reaction
3
Fig. 2 Effects of solvents (A) and Ru loading (B) on the product selectivities for
the hydrogenation of LA at full conversion. Others corresponded to the uniden-
tified products. Reaction conditions: 4 mmol of LA, 0.2 g of Ru/SBA-SO
5 wt% Ru for A), 10 mL of solvent (ethanol as solvent for B), 6 h, 4 MPa H
13 K.
3
H-4
and
Fig. 3 Formation of EL, GVL, EV, VA and MTHF as a function of reaction time
for the hydrogenation of LA using Ru/SBA-SO H-4. Reaction conditions: 4 mmol
of LA, 0.2 g of Ru/SBA-SO H-4, 10 mL of ethanol, 4 MPa H and 513 K.
(
2
3
5
3
2
This journal is © The Royal Society of Chemistry 2013
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 4 The reusability of Ru/SBA-SO
3
H-4 at full conversion of LA. Reaction con-
ditions: 4 mmol of LA, 0.2 g of used Ru/SBA-SO
MPa H and 513 K. An additional 0.2 g of SBA-SO
Runs 2–5.
3
H-4, 10 mL of ethanol, 6 h,
4
2
3
H-4 was added in
was employed under the optimized reaction conditions as
Run 1. It can be seen that the selectivity to EV and VA
decreased significantly. The result may be explained by the
leaching study of Ru and S (Table S5 in the ESI†). The loss of
Ru to the liquid phase was only 3.1% of the original weight of
Scheme 2 Probable reaction pathways for the conversion of LA to EV over
3
Ru/SBA-SO H. Pathways based on those reported in refs 13 and 16.
3
Ru in Ru/SBA-SO H-4. However, 45.7% of S was lost by leach-
ing, leading to a drastical decrease of the acidity (Table S2 in
pathways for the formation of EV and VA were based on the
mechanism for the conversion of GVL to VA proposed by
Lange et al., and Ravasio et al. Initially, LA was converted
into GVL over the Ru catalyst and this process was fast. Then
GVL proceeded acid-catalyzed ring opening to ethyl pentenoate
the ESI†). Thus, an additional 0.2 g of SBA-SO H-4 was added
3
in Runs 2–5 to provide the acidity. These results from Run 2 to
Run 5 showed that the selectivity to EV and VA decreased
slightly in 4 runs.
1
3
16
(
EP), followed by hydrogenation to EV and VA over supported
3
.6 Preparation of various alkyl valerate esters
Various alkyl valerate esters with the alkyl chain range from C1
to C5 were prepared over Ru/SBA-SO H under the optimized
Ru catalyst. The conversion of LA to EV and VA was affected by
acid intensity and hydrogenation functionalities of Ru catalyst.
A small amount of acid sites increased the co-production of
GVL (Table 2) and high Ru loading favored the over-hydrogen-
ation to MTHF (Fig. 2B). The proposed reaction pathways were
studied by controlled experiments. It was undoubted that
there was no need to verify the conversion of LA to GVL cata-
3
reaction conditions. Methyl valerate was produced with a yield
of 93% using methanol as the solvent (Scheme 3). Substituting
methanol with other aliphatic alcohols (n-propanol, n-butanol
and n-pentanol) led to the problem of low yields of target pro-
ducts. Increasing the amount of catalyst to 0.4 g (double the
amount of the previous catayst) and extending the reaction
time to 12 h resulted in considerable yields of 88, 86 and 82%,
respectively.
lyzed by Ru/SBA-SO
ring opening of GVL to EP was also reported by Lange et al.
3
H. A subsequent reaction of acid-catalyzed
2
0
The key intermediate EP was not detected in the product
mixture, suggesting that the hydrogenation of EP to EV was a
fast step and the formed EP from GVL was converted immedi-
ately over Ru catalyst. In this respect, the controlled experi-
ments were performed using EP as a substrate, catalyzed by
3.7 Effect of other metal catalysts
A further study was conducted to investigate the effect on the
hydrogenation of LA by changing the metal element in the
3
Ru/SBA-15 and Ru/SBA-SO H. Under the optimized reaction
conditions for the conversion of LA (513 K, 4 MPa H , ethanol
2
as solvent), EP was completely converted over both Ru/SBA-15
and Ru/SBA-SO H. High selectivities to EV and VA were
3
obtained (95 and 96%, respectively). These results gave
support to the proposed reaction pathways for the formation
of EV and VA.
3
.5 Reusability of the catalyst
In order to assess the practical application of the catalysts, the
reusability of Ru/SBA-SO H-4 was studied and the results are
3
3
shown in Fig. 4. The Ru/SBA-SO H-4 used once at 513 K for 6 h Scheme 3 Preparation of various alkyl valerate esters over Ru/SBA-SO H.
3
Green Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
and Fok Ying Tung Education Foundation for the financial
support.
References
1
N. Armaroli and V. Balzani, Angew. Chem., Int. Ed., 2007,
6, 52.
4
2
B. Kamm, P. R. Gruber and M. Kamm, Biorefineries-
industrial processes and products, Wiley-VCH, Weinheim,
2006.
3
Fig. 5 LA hydrogenation over Cu/SBA-15, Cu/SBA-SO H-4, Ni/SBA-15 and
3
4
(a) G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006,
106, 4044; (b) A. Corma, S. Iborra and A. Velty, Chem. Rev.,
Ni/SBA-15 & SBA-SO
3
H-4. Metal loading was 5%. Reaction conditions: 4 mmol
and 513 K.
of LA, 0.2 g of catalyst, 10 mL of ethanol, 6 h, 4 MPa H
2
2007, 107, 2411.
(a) T. Werpy and G. Petersen, Top Value-Added Chemicals
from Biomass Vol. I—Results of Screening for Potential Candi-
dates from Sugars and Synthesis Gas, U. S. Department of
Energy (DOE) by the National Renewable Energy Labora-
tory, 2004; (b) J. J. Bozell and G. R. Petersen, Green Chem.,
catalysts. Supported Cu or Ni on SBA-15 and SBA-SO H with a
3
metal loading of 5 wt% were studied in ethanol under the
same reaction conditions as Ru. As shown in Fig. 5, the
majority of LA was converted into GVL, when either Cu or
Ni as the catalyst was used. Some EV and VA were detected and
LA was almost fully consumed. Compared with the reported
catalyst Cu-SiZr which was employed to convert GVL to valerate
2010, 12, 539.
5
6
7
8
(a) R.-J. van Putten, J. C. van der Waal, E. de Jong,
C. B. Rasrendra, H. J. Heeres and J. G. de Vries, Chem. Rev.,
1
6
ester in high yield, the poor selectivity to EV and VA may
be attributed to the weak acidity of Cu/SBA-15 and
2
013, 113, 1499; (b) A. A. Rosatella, S. P. Simeonov,
R. F. M. Fradea and C. A. M. Afonso, Green Chem., 2011, 13,
54.
Cu/SBA-SO H-4 (SiZr: H ≤ −8.2, SBA-SO H-4: −8.2 ≤ H
≤
3
0
3
0
7
+
0.8). It is also worth noting that 33% yield of EV and VA was
(a) B. V. Timokhin, V. A. Baransky and G. D. Eliseeva, Russ.
Chem. Rev., 1999, 68, 73; (b) J. J. Bozell, L. Moens,
D. C. Elliott, Y. Wang, G. G. Neuenscwander,
S. W. Fitzpatrick, R. J. Bilski and J. L. Jarnefeld, Resour.,
Conserv. Recycl., 2000, 28, 227.
(a) G. W. Huber, R. D. Cortright and J. A. Dumesic, Angew.
Chem., Int. Ed., 2004, 43, 1549; (b) J. Zhang, S. B. Wu, B. Li
and H. D. Zhang, ChemCatChem, 2012, 4, 1230;
produced when using a combination of Ni/SBA-15 and SBA-SO H,
indicating that a proper hydrogenation activity of the catalyst
played a key role in the conversion of LA to valerate esters.
3
4
Conclusions
In summary, the hydrogenation of levulinic acid to valerate
esters over supported Ru catalysts was affected by the acidity of
the supports. A bifunctional catalyst Ru/SBA-SO H was devel-
(
c) Q. Zhang, T. Jiang, B. Li, T. J. Wang, X. H. Zhang,
Q. Zhang and L. L. Ma, ChemCatChem, 2012, 4, 1084.
3
(a) D. M. Alonso, J. Q. Bond and J. A. Dumesic, Green
Chem., 2010, 12, 1493; (b) J. C. Serrano-Ruiz and
J. A. Dumesic, Energy Environ. Sci., 2011, 4, 83; (c) S. Van de
Vyver, J. Geboers, P. A. Jacobs and B. F. Sels, ChemCatChem,
oped as the active catalyst and characterized. The increase in
the amount of acid sites on the support improved the selecti-
vity to EV and VA, and the highest yield of 94% to EV and VA
was achieved. The reaction temperature, initial hydrogen
pressure, solvent and Ru loading were also important factors.
Probable reaction pathways were proposed and verified.
γ-Valerolactone obtained from levulinic acid proceeded by
acid-catalyzed ring opening to ethyl pentenoate, followed by
hydrogenation to EV over supported Ru catalyst. Various alkyl
valerate esters with the alkyl chain range from C1 to C5 were
also prepared. 33% yield of EV and VA was produced catalyzed
2011, 3, 82; (d) J. A. Geboers, S. Van de Vyver, R. Ooms,
B. Op de Beeck, P. A. Jacobs and B. F. Sels, Catal. Sci.
Technol., 2011, 1, 714; (e) C.-H. Zhou, X. Xia, C. X. Lin,
D.-S. Tong and J. Beltramini, Chem. Soc. Rev., 2011, 40,
5588; (f) P. Gallezot, Chem. Soc. Rev., 2012, 41, 1538;
(
g) H. Kobayashi, H. Ohta and A. Fukuoka, Catal. Sci.
Technol., 2012, 2, 869; (h) R. Palkovits, Angew. Chem., Int.
Ed., 2010, 49, 4336.
S. W. Fitzpatrick, World patent WO 8910362, 1990;
S. W. Fitzpatrick, World patent WO 9640609, 1997.
by the combination of Ni/SBA-15 and SBA-SO H. In summary,
the suitable acidity and proper hydrogenation activity of the
catalyst were the key factors for the reaction.
3
9
10 (a) B. C. Windom, T. M. Lovestead, M. Mascal, E. B. Nikitin
and T. J. Bruno, Energy Fuels, 2011, 25, 1878;
(
b) E. Christensen, A. Williams, S. Paul, S. Burton and
Acknowledgements
R. L. McCormick, Energy Fuels, 2011, 25, 5422.
This work was supported by the National Basic Research 11 (a) H. Mehdi, V. Fabios, R. Tuba, A. Bodor, L. T. Mika and
Program of China (2012CB215305), NSFC (21172209), FRFCU
WK2060190025), SRFDP (20123402130008), CAS (KJCX2-EW-J02)
I. T. Horváth, Top. Catal., 2008, 48, 49; (b) I. T. Horváth,
H. Mehdi, V. Fabos, L. Boda and L. T. Mika, Green Chem.,
(
This journal is © The Royal Society of Chemistry 2013
Green Chem.

View Article Online
Paper
Green Chemistry
2
008, 10, 238; (c) D. M. Alonso, S. G. Wettstein and 18 E. I. Gürbüz, E. L. Kunkes and J. A. Dumesic, Appl. Catal.,
J. A. Dumesic, Green Chem., 2013, 15, 584. B, 2010, 94, 134.
2 F. M. A. Geilen, B. Engendahl, A. Harwardt, W. Marquardt, 19 (a) D. Margolese, J. A. Melero, S. C. Christiansen,
1
J. Klankermayer and W. Leitner, Angew. Chem., Int. Ed.,
010, 49, 5510.
3 J.-P. Lange, R. Price, P. M. Ayoub, J. Louis, L. Petrus,
B. F. Chmelka and G. D. Stucky, Chem. Mater., 2000, 12,
2448; (b) J. A. Melero, G. D. Stucky, R. van Grieken and
G. Morales, J. Mater. Chem., 2002, 12, 1664; (c) D. M. Lai,
L. Deng, J. Li, B. Liao, Q. X. Guo and Y. Fu, ChemSusChem,
2011, 4, 55; (d) D. M. Lai, L. Deng, Q. X. Guo and Y. Fu,
Energy Environ. Sci., 2011, 4, 3552; (e) D. Reyes-Luyanda,
J. Flores-Cruz, P. J. Morales-Pérez, L. G. Encarnación-
Gómez, F. Shi, P. M. Voyles and N. Cardona-Martínez,
Top. Catal., 2012, 55, 148; (f) Y. Tang, S. J. Miao,
B. H. Shanks and X. M. Zheng, Appl. Catal., A, 2010,
375, 310.
2
1
L. Clarke and H. Gosselink, Angew. Chem., Int. Ed., 2010,
4
9, 4479.
1
1
1
1
4 G. Dayma, F. Halter, F. Foucher, C. Togbé, C. Mounaim-
Rousselle and P. Dagaut, Energy Fuels, 2012, 26, 4735.
5 W. H. Luo, U. Deka, A. M. Beale, E. R. H. van Eck, P. C. A.
Bruijnincx and B. M. Weckhuysen, J. Catal., 2013, 301, 175.
6 C. E. Chan-Thaw, M. Marelli, R. Psaro, N. Ravasio and
F. Zaccheria, RSC Adv., 2013, 3, 1302.
7 B. Zhang, Y. Zhu, G. Ding, H. Zheng and Y. Li, Green 20 J.-P. Lange, J. Z. Vestering and R. J. Haan, Chem. Commun.,
Chem., 2012, 14, 3402. 2007, 33, 3488.
Green Chem.
This journal is © The Royal Society of Chemistry 2013