Esterification of levulinic acid with ethanol over sulfated mesoporouszirconosilicates: Influences of the preparation conditions on thestructural properties and catalytic performances

Article abstract of DOI:10.1016/j.cattod.2013.11.008

Levulinic acid is considered one of the most important biomass-derived chemicals owing to its poten-tial as a versatile building block to synthesize valuable fuels and chemicals. Levulinate esters, such asmethyl levulinate and ethyl levulinate obtained via esterification of levulinic acid with alcohols, can inparticular be used as fuel additives and plasticizers, and thus have a potential to replace a significantamount of petroleum-derived chemical feedstocks. In this article, sulfated zirconosilicates having P6mmhexagonal mesoporous structure were applied as solid acid catalysts to the esterification of levulinic acidwith ethanol to produce ethyl levulinate, and the influences of preparation conditions on the structuralproperties and catalytic performances were investigated. A distinct correlation was observed betweenthe catalytic activity and the density of acid sites, showing that dispersibility of the acid sites and theassociated accessibility of the organic reactants play an important role in determining the overall activ-ity. Among the catalysts tested, sulfated Zr-SBA-15 with optimum Zr content (Si/Zr ratio of 10.7) wasfound to be the best catalyst, the activity of which was far superior to that of conventional sulfated ZrO₂.In addition, direct conversion of cellulosic sugars (glucose and fructose) into levulinate esters was alsoexamined.

Full text of DOI:10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
Catalysis Today xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Esterification of levulinic acid with ethanol over sulfated mesoporous
zirconosilicates: Influences of the preparation conditions on the
structural properties and catalytic performances
Yasutaka Kuwahara , Tadahiro Fujitani , Hiromi Yamashitab
a,∗
a
a
Research Institute for Innovation in Sustainable Chemistry, National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Higashi,
Tsukuba, Ibaraki 305-8565, Japan
b
Division of Materials and Manufacturing Science, Graduate School of Engineering, Osaka University, 2-1 Yamadaoka, Suita, Osaka 565-0871, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Levulinic acid is considered one of the most important biomass-derived chemicals owing to its poten-
tial as a versatile building block to synthesize valuable fuels and chemicals. Levulinate esters, such as
methyl levulinate and ethyl levulinate obtained via esterification of levulinic acid with alcohols, can in
particular be used as fuel additives and plasticizers, and thus have a potential to replace a significant
amount of petroleum-derived chemical feedstocks. In this article, sulfated zirconosilicates having P6mm
hexagonal mesoporous structure were applied as solid acid catalysts to the esterification of levulinic acid
with ethanol to produce ethyl levulinate, and the influences of preparation conditions on the structural
properties and catalytic performances were investigated. A distinct correlation was observed between
the catalytic activity and the density of acid sites, showing that dispersibility of the acid sites and the
associated accessibility of the organic reactants play an important role in determining the overall activ-
ity. Among the catalysts tested, sulfated Zr–SBA-15 with optimum Zr content (Si/Zr ratio of 10.7) was
found to be the best catalyst, the activity of which was far superior to that of conventional sulfated ZrO2.
In addition, direct conversion of cellulosic sugars (glucose and fructose) into levulinate esters was also
examined.
Received 1 August 2013
Received in revised form 30 October 2013
Accepted 14 November 2013
Available online xxx
Keywords:
Biomass conversion
Levulinic acid
Esterification
Solid acid catalyst
Mesoporous zirconosilicate
©
2013 Elsevier B.V. All rights reserved.
1
. Introduction
miscible biofuel in regular diesel car engines without modification
owing to their physicochemical properties similar to the biodiesel
fatty acid methyl esters (FAME), and thus has a potential to reduce
the consumption of petroleum-derived fossil fuels [10,16].
Generally, levulinate esters are produced by esterification of LA
Driven by increasing concern over the depletion of petroleum
resources, use of biomass as renewable natural resources and its
conversion into valuable biofuels and feedstock chemicals have
received considerable attention in recent years [1–4]. Among var-
ious chemicals synthesized from cellulose and cellulosic biomass,
levulinic acid (LA) and its esters are gaining increasing attention
as promising platform chemicals for a range of derivatives in the
biofuel, polymer and specialty chemicals markets [5–10]. LA is a
molecule generally produced by the acid-catalyzed hydrolysis of
cellulose [11–13], and can catalytically be converted into levulinate
esters [10,14–16], ␥-valerolactone [5–7,17–20], 1,4-pentanediol
with several alcohols using mineral acids such as HCl, H SO4 and
2
H PO (Scheme 1), which leads to a high yield of corresponding
3
4
products within a short reaction period. However, such homo-
geneous acid catalysts are unrecyclable and always suffer from
operational problems (e.g., use of a large volume of base for neu-
tralization and corrosion of equipment). Therefore, substitution of
homogeneous catalysts by heterogeneous analogues that are easily
separable and reusable over repeated cycles is highly desirable.
Several kinds of solid acid catalysts have been reported so
far for the esterification of LA with alcohols. Baronetti et al.
[
5] and 5-nonanone (via pentanoic acid) [21] as well as diphenolic
acid as an intermediate for the synthesis of epoxy resins and poly-
carbonates [22,23]. Levulinate esters are also useful compounds
that can be used as fuel additives, solvents and plasticizers. In par-
ticular, ethyl levulinate can directly be used up to 5 wt% as a diesel
◦
examined the esterification of LA with ethanol at 78 C using
silica-stabilized Wells–Dawson heteropolyacids [14]. Bokade et al.
used montmorillonite-supported heteropolyacids for the esterifi-
◦
cation of LA with n-butanol to obtain n-butyl levulinate at 120 C
[
15]. These precedent works indicate that LA can be transformed
∗ Corresponding author. Tel.: +81 29 861 2790.
E-mail address: y.kuwahara@aist.go.jp (Y. Kuwahara).
to levulinate esters in excellent yields over solid acid catalysts
which possess as strong acid strength as that of heteropolyacids.
0
920-5861/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2013.11.008
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
2
Y. Kuwahara et al. / Catalysis Today xxx (2013) xxx–xxx
Scheme 1. Esterification of levulinic acid with ethanol.
◦
Recently, Silva et al. screened several types of solid acid catalysts
e.g., zeolites and sulfated mixed oxides) through the esterification
of LA with ethanol and found that sulfated mixed oxides (sulfated
100 C for another day under static conditions. The resulting prod-
uct was filtered, washed with deionized water, dried overnight
(
◦
and finally calcined in air at 600 C for 6 h to remove the organic
template. The thus obtained solids were subsequently sulfated by
suspending 2.0 g of solid in 30 mL of 1.0 M of sulfuric acid solutions
for 1 h, followed by vacuum filtration, drying at 100 C overnight
and calcination in air at 600 C for 3 h. The obtained samples were
2
−
2−
2−
zirconia (SO4 /ZrO ), niobia (SO
/Nb O5), titania (SO4 /TiO2)
2
4
2
2−
and stannia (SO4 /SnO )) bearing a number of strong acid sites
2
◦
are the most promising candidates providing high catalytic activity
within an appropriate period of time [16]. Nevertheless, the
reaction rates are likely dependent on the number of acid sites and
the preparation conditions applied, still leaving room for a further
study; there might be a possibility that the activity of sulfated
mixed oxides for this reaction can further be improved by optimiz-
ing preparation conditions, the number of acid sites and dispersion
state of the sulfate species by introducing more elaborated catalyst
preparation techniques. One possible approach is the introduc-
tions of mesopores in the mixed oxides via a surfactant-induced
self-assembly approach. Mesoporosity is especially effective in
liquid-state catalysis for easy transport of the reactant molecules,
and the resulting high surface area provides better dispersion of
the acid sites and an increased number of acid sites [24,25].
Herein, we report the esterification of levulinic acid with ethanol
to produce ethyl levulinate over sulfated mesoporous zirconosili-
cate materials. Among a number of choices of mesoporous silicate
structure, SBA-15 type structure consisting of two-dimensional
cylindrical pores arranged in a hexagonal order (P6mm symmetry)
was chosen, because of its pore diameter larger than that of MCM-
◦
denoted as S-ZrSBA15(X), where X is the atomic ratio of Si/Zr in the
final solid.
For comparison, a pure siliceous SBA-15 sample was prepared
the same way as the above procedure except for the addition of
zirconium source. A sulfated zirconia (SO42 /ZrO ) was also pre-
−
2
pared by the conventional wet impregnation method [29,30]; a
total of 5.8 g of ZrOCl ·8H O was dissolved in 200 mL of deionized
2
2
water, followed by precipitation of zirconium hydroxide at pH 9.0
using 10% NH3 solution, aging at room temperature for 2 days and
◦
calcination in air at 600 C for 3 h. The resulting ZrO2 was then sus-
pended in 1.0 M of sulfuric acid solutions (15 mL per gram of solid)
for 1 h, followed by vacuum filtration, drying at 100 C overnight
and calcination in air at 600 C for 3 h.
◦
◦
2
.2. Preparation of sulfated mesoporous zirconosilicate via a
direct-sulfation procedure
Sulfated mesoporous zirconosilicates with larger Zr contents
were also synthesized via a direct-sulfation route according to the
methods previously reported by Mou et al. [24]. In a typical synthe-
sis, 3.0 g of Pluronic P123 was dissolved in 100 mL of HCl aqueous
solution with vigorous stirring for 2 h in a flask. To this micellu-
lar solution, an amount of ammonium sulfate ((NH ) SO , Wako
Pure Chemical Industries, 99.5%) and a mixture solution of TEOS
and zirconium tetrapropoxide (Zr(O Pr) , Aldrich, 70% w/w in 1-
propanol) were sequentially added and stirred at 40 C for 1 day.
The molar ratio of the initial gel was adjusted to P123: Si: Zr: HCl:
H O = 0.017: 1.0: (0.25–1.0): 4.7: 180, and the SO4 /Zr molar ratio
was fixed to 1.0. The solution was transferred to an oven and then
hydrothermally reacted at 100 C for another day under static con-
ditions. The resulting product was filtered, washed with deionized
water, dried overnight at 100 C, and finally calcined at 600 C for
h to remove the organic template. The synthesized samples were
4
1 analogues. Larger pore diameter provides a wider space for
mass transportation, which may allow reactants to more efficiently
access to the active sites dispersed inside the pore channels. A series
of sulfated zirconosilicates having P6mm hexagonal mesoporous
structure with varied Zr content were synthesized by sol–gel pro-
cess using Pluronic P123 block copolymer as a pore-directing agent
and tetraethylorthosilicate as a silicon source either via a post-
sulfation or via a direct-sulfation procedure. The influences of
preparation conditions and Zr content on structural properties and
catalytic performances were investigated in detail. The activities of
the catalysts were examined by the esterification of LA with ethanol
4
2
4
n
4
◦
2−
2
◦
at 70 C and were compared with that of the conventional sulfated
◦
zirconia. Furthermore, conversions of cellulosic sugars (glucose and
fructose) directly into levulinate esters by acid-catalyzed thermo-
lysis in methanol were also examined using a few selected samples.
◦
◦
6
named as S-mesoZS(X), where X is the atomic ratio of Si/Zr in the
final solid.
2
. Experimental
2.1. Preparation of sulfated mesoporous zirconosilicate via a
2.3. Characterization
post-sulfation procedure
Powder X-ray diffraction (XRD) patterns were collected using a
Bruker AXS D8 Advance X-ray diffractometer equipped with a scin-
The typical synthetic procedure for the post-sulfation route is
as follows [26–28]. An amount of 3.0 g of Pluronic P123 (PEO20
-
tillation counter for low-angle region and a LynxEye detector for
PPO70-PEO20, Aldrich, MW = 5800) and 1.77 g of NaCl was dissolved
in 120 mL of deionized water with vigorous stirring for 2 h in a
flask. To this micellular solution, a designated amount of zirconium
oxychloride octahydrate (ZrOCl ·8H O, Aldrich, >99%) and 6.44 g
high-angle region with CuK␣ radiation (ꢀ = 1.54056 A˚ ). Scans were
◦
◦
performed at step size 0.02 over the 2ꢁ range of 0.5–80 . Nitro-
◦
gen adsorption–desorption isotherms were measured at-196 C
by using Micromeritics ASAP2020. The samples were degassed at
2
2
◦
of tetraethylorthosilicate (TEOS, Wako Pure Chemical Industries,
300 C under vacuum for 4 h prior to the measurements to vaporize
◦
9
5%) were sequentially added and stirred at 40 C for 1 day. The
physisorbed water. The specific surface area was calculated by the
BET (Brunauer–Emmett–Teller) method by using adsorption data
molar ratio of the initial gel was adjusted to P123: Si: Zr: NaCl:
H O = 0.017: 1.0: (0–0.2): 1.0: 220. The solution was then trans-
2
ranging from P/P = 0.05 to 0.30. The pore size distributions were
0
ferred to a pressure bottle, sealed, and hydrothermally treated at
obtained from the adsorption branch of the nitrogen isotherms by
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
Y. Kuwahara et al. / Catalysis Today xxx (2013) xxx–xxx
3
the BJH (Barret–Joyner–Halenda) method. Infrared spectra were
recorded with a JASCO FTIR-6300 instrument in the spectral range
−
1
−
1
2
000–400 cm under vacuum with a resolution of 4 cm using
samples diluted with KBr. Diffuse reflectance UV–vis spectra were
collected using a Shimadzu UV-2600 spectrophotometer equipped
with an integrating sphere in the spectral range of 200–400 nm.
Absorption spectra were calculated using the Kubelka–Munk func-
tion. Inductively coupled plasma atomic emission spectrometry
(
ICP-AES) was used to quantify sulfur contents in the samples.
The acidity of catalysts was studied by temperature pro-
grammed desorption of NH3 (NH –TPD) by using a BELCAT-B
3
system (BEL Japan, Inc.). Approximately 100 mg of each sample was
◦
preheated under a He flow (50 mL/min) at 600 C for 1 h, allowed to
◦
cool to 50 C, and subsequently exposed to flowing 5% NH /He gas
3
◦
mixture (50 mL/min) for 1 h. Then, the system was purged at 50 C
for 0.5 h with He to eliminate weakly adsorbed NH . NH –TPD was
3
3
◦
carried out between 50 and 600 C under a He flow (30 mL/min)
with a ramping rate of 10 C/min, and the desorbed NH was quan-
◦
3
tified by an on-line thermal conductivity detector.
2.4. Esterification of levulinic acid with ethanol
Levulinic acid esterification was carried out in a quartz glass
reactor with a reflux condenser. The reactions were typically
◦
performed at 70 C for up to 24 h under stirring of 700 rpm
using 5.0 wt% of solid acid catalysts and with the molar ratio
of LA:EtOH = 1:10. A portion of the reaction mixture was with-
drawn at appropriate intervals, filtered and analyzed by using a
gas chromatography (GC; Shimadzu GC-14B) with a flame ioniza-
tion detector equipped with a capillary column (ULBON HR-20 M;
0
.53 mm × 30 m; Shinwa Chemical). The levulinic acid conversion
(
or ethyl levulinate yield) was calculated based on the levulinic
acid/ethyl levulinate area ratio. To examine the catalyst recycla-
bility, the spent catalyst was recovered from the reaction solution
◦
by simple filtration, washed with acetone, calcined at 600 C in air
for 3 h to remove the strongly adsorbed organic species and then
subjected to next catalytic run.
2.5. Direct conversion of cellulosic sugars into methyl levulinate
Direct conversion of sugars (either glucose or fructose) into
methyl levulinate was performed in a 60 mL cylindrical stainless
steel autoclave reactor (EYELA, Inc.) [31]. 450 mg of sugars (Aldrich,
2
.5 mmol), 20 mL of methanol (Aldrich, reagent grade) and 2.5 wt%
of catalyst were introduced into the reactor, which was then sealed,
◦
purged and pressurized with 0.5 MPa of N and heated at 200 C for
2
3
h with magnetic stirring. After cooling, the reaction mixture was
filtered and analyzed with a high performance liquid chromatogra-
phy (HPLC; Shimadzu). The yields of products and unreacted sugars
were quantified by an internal standard method.
3
. Results and discussion
3.1. Structural analysis
The structural characteristics of a series of sulfated mesoporous
zirconosilicates were investigated by XRD measurement and N2
physisorption. Fig. 1 shows XRD patterns of S-ZrSBA15 with var-
ied Si/Zr atomic ratios (corresponding to 0.38–17.8 mol% Zr/Si). The
Si/Zr atomic ratio in the final solids were higher than the ones in the
initial gel solutions (see Table 1), indicating that some fraction of Zr
species remained unincorporated into the silicate particles under
the synthesis conditions applied in this study. Low angle XRD pat-
terns of S-ZrSBA15 with the Si/Zr atomic ratio above 15.5 exhibited
◦
defined single peaks at around 2ꢁ = 0.8–0.9 and two adjacent small
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
4
Y. Kuwahara et al. / Catalysis Today xxx (2013) xxx–xxx
(
A)
(B)
d100 = 11.8 nm
d100 = 10.3 nm
×
×
2
2
(d)
(c)
(d)
(c)
(b)
(a)
d100 = 10.0 nm
d100 = 9.9 nm
(b)
(100)
(110)
(200)
(a)
0
1
2
3
4
5
10
20
30
40
50
60
70
80
2θ
/ degree
2θ / degree
Fig. 1. (A) Small-angle and (B) wide-angle XRD patterns of S-ZrSBA15 with varied Si/Zr atomic ratios (Si/Zr = (a) 26.6, (b) 15.5, (c) 10.7, and (d) 5.6 in the final solid). Several
diffraction planes associated with 2D hexagonal mesoporous structure (P6mm) are indicated for S-ZrSBA15(26.6).
diffractions that are assigned to the (1 0 0), (1 1 0) and (2 0 0) reflec-
tions, respectively, showing that these samples have 2D hexagonal
mesoporous structures with P6mm symmetry (Fig. 1A(a) and (b)).
On the other hand, S-ZrSBA15 with the Si/Zr atomic ratio below
The textural parameters obtained from XRD measurement and N2
adsorption isotherms are summarized in Table 1. Along with the
decrease in the Si/Zr ratio from 26.6 to 5.6, the specific surface areas
2
and total pore volumes decreased from 585 to 454 m /g and 0.88
3
1
0.7 exhibited less resolved, broad diffraction peaks in this region,
to 0.36 cm /g, respectively. The pore size distribution curves cal-
suggesting a decrease in the ordered pore arrangement for these
samples.
This disordered arrangement of mesopores can be elucidated by
nitrogen adsorption isotherms. S-ZrSBA15(26.6) displayed a type
culated by the BJH method show that the average pore diameter
decreased from 8.9 to 2.2 nm (Fig. 2B), being coincided with the
increases in d-spacing values and the corresponding lattice param-
eter a0 (pore to pore distance) as well as pore wall thickness Tw
calculated from XRD data. These combined results reveal that the
ordered structure of SBA-15 silica appears to degrade at higher per-
centages of Zr (more than 5.0 mol% per Si), and accordingly results
in reduced mesoporosity [25–28,32–34]. Nonetheless, the materi-
als still had large surface areas and pore volumes as to be regarded
IV isotherm with clear capillary condensation at around P/P = 0.7,
0
reflecting the presence of defined mesopore channels with narrow
pore size distributions (Fig. 2A(a)). The adsorbed nitrogen quan-
tities appreciably decreased and the hysteresis in the isotherms
gradually disappeared as the Zr content increased (Fig. 2A(b)–(d)).
0
0
0
.8
.7
.6
1
1
200 (A)
(B)
(d)
(
d) +900
Si/Zr=5.6
Si/Zr=10.7
Si/Zr=15.5
Si/Zr=26.6
000
(
c) +600
0.5
8
6
4
2
00
00
00
00
0
(c)
(b)
(a)
0
0
0
0
.4
.3
.2
.1
(
b) +300
a)
(
0.0
0
.0
0.2
0.4
0.6
0.8
1.0
0
10
20
30
40
50
P/P0
Pore diameter / nm
Fig. 2. (A) Nitrogen adsorption–desorption isotherms and (B) the corresponding pore size distribution curves of S-ZrSBA15 with varied Si/Zr atomic ratios (Si/Zr = (a) 26.6,
(
b) 15.5, (c) 10.7 and (d) 5.6 in the final solid). Filled and empty symbols in the left represent adsorption and desorption branches, respectively.
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
Y. Kuwahara et al. / Catalysis Today xxx (2013) xxx–xxx
5
M : Monoclinic ZrO2
T : Tetragonal ZrO2
(
A)
(B)
M
M
d100 = 9.3 nm
d100 = 9.3 nm
T
T
M
M(T)
M
T
(
d)
M
M
M
(d)
(c)
(b)
(a)
M
M
M
T
M
M
T
M
M
M
M(T)
M(T)
M(T)
T
(
c)
T
d100 = 9.9 nm
d100 = 10.3 nm
M
T
T
(b)
(100)
T
M M
(
110)
(
T
M
200)
(a)
T
0
1
2
3
4
5
10
20
30
40
50
60
70
80
2θ / degree
2θ / degree
Fig. 3. (A) Small-angle and (B) wide-angle XRD patterns of S-mesoZS with varied Si/Zr atomic ratios (Si/Zr = (a) 7.9, (b) 6.6, (c) 3.7 and (d) 1.7 in the final solid). Several
diffraction planes associated with 2D hexagonal mesoporous structure (P6mm) are indicated for S-mesoZS(7.9), and diffraction planes assigned to monoclinic (M) and
tetragonal (T) ZrO2 crystals are indicated in the right.
as mesoporous materials. In high angle XRD patterns, no crystalline
phase could be observed at all Zr loading levels, except for broad
peaks attributable to amorphous silicate phase, suggesting an iso-
morphorous substitution of Zr atoms within the silica matrix in the
S-ZrSBA15 samples [28,33,34].
Fig. 3 shows XRD patterns of S-mesoZS with varied Si/Zr atomic
ratios (corresponding to 12.7–58.8 mol% Zr/Si). While S-mesoZS
materials contain much higher Zr contents than S-ZrSBA15 materi-
als, they exhibited well-defined diffraction planes associated with
Si/Zr ratio decreased from 7.9 to 1.7; however, the pore-associated
parameters, such as d-spacing value, average pore diameter and
wall thickness, scarcely changed, indicating a retention of the
ordered mesoporous structure. It should be noted that a series of
S-mesoZS materials have relatively larger pore volumes and wider
pore channels compared to those of S-ZrSBA15 materials, despite
the incorporation of larger Zr content.
The most obvious difference between S-mesoZS and S-ZrSBA15
samples was observed in their crystalline structure. In high angle
XRD patterns, S-mesoZS samples exhibited diffraction planes asso-
2
D hexagonal mesoporous structure (P6mm) in all Zr loading levels
◦
(Fig. 3A). In N2 adsorption, a series of S-mesoZS materials showed
ciated with both monoclinic ZrO2 (2ꢁ = 24.2, 28.2, 31.4 and 34.4 )
◦
typical type IV isotherms with clear inflections at around P/P = 0.7
and tetragonal ZrO2 (2ꢁ = 30.2, 49.6, 50.4 and 60.1 ) (Fig. 3B),
0
and with narrow pore size distributions centered at around 10 nm
evidencing the aggregation of Zr species. This difference in the
crystalline phase is due to the different zirconium precursors used
(
Fig. 4). The specific surface areas and total pore volumes decreased
2
3
n
from 695 to 392 m /g and 1.25 to 0.72 cm /g, respectively, as the
(ZrOCl ·8H O for S-ZrSBA15 and Zr(O Pr) for S-mesoZS) and the
2
2
4
0
0
0
.8
.7
.6
2
1
1
1
1
1
000
800
600
400
200
000
(
A)
(B)
(
d) +1500
Si/Zr=1.7
Si/Zr=3.7
Si/Zr=6.6
(d)
(
c) +1000
0.5
.4
0.3
(c)
(b)
(a)
0
(
b) +500
a)
8
6
4
2
00
00
00
00
0
0
0
.2
.1
(
Si/Zr=7.9
30
0.0
0
.0
0.2
0.4
0.6
0.8
1.0
0
10
20
40
50
P/P0
Pore diameter / nm
Fig. 4. (A) Nitrogen adsorption–desorption isotherms and (B) the corresponding pore size distribution curves of S-mesoZS with varied Si/Zr atomic ratios (Si/Zr = (a) 7.9, (b)
6
.6, (c) 3.7 and (d) 1.7 in the final solid). Filled and empty symbols in the left represent adsorption and desorption branches, respectively.
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
6
Y. Kuwahara et al. / Catalysis Today xxx (2013) xxx–xxx
(A)
(B)
968
968 752
580
501
Si/Zr = 5.6
Si/Zr = 1.7
Si/Zr = 3.7
Si/Zr = 6.6
Si/Zr = 7.9
Si/Zr = 10.7
Si/Zr = 15.5
Si/Zr = 26.6
2
000
1600
1200
800
400
2000
1600
1200
800
400
-
1
-1
Wavenumber / cm
Wavenumber / cm
Fig. 5. FTIR spectra of (A) S-ZrSBA15 samples and (B) S-mesoZS samples with different Si/Zr atomic ratios.
groups [32,35]. The bands at 460, 806, 1085 cm 1 are assigned
to the bending, symmetric stretching and asymmetric stretching
vibration of the Si–O–Si bond, respectively [32,35]. The intensity
−
different synthetic conditions applied (almost neutral condition for
S-ZrSBA15 and strong acidic condition for S-mesoZS). Considering
the fact that the adsorption–desorption branches were essentially
parallel and the average pore diameter scarcely changed, most Zr
species in the S-mesoZS materials are likely to be present inside
−
1
of the band positioned at around 968 cm can be a good indicator
of the formation of the Si–O–Zr bond [32]. In the case of S-ZrSBA15
samples, the intensity of this band quantitatively increased as
the Zr content in the samples increased (Fig. 5A), indicating the
incorporation of larger Zr atoms into the silica framework. Contrary
to this, in the IR spectra of the S-mesoZS samples, the intensity of
the mesoporous silica wall as nano-crystalline ZrO particles with-
2
out significant pore blocking (extra-framework aggregated ZrO2
particles were not observed in the TEM images).
3.2. Investigation of chemical environment of Zr species
−
1
the band at 968 cm
gradually diminished as increasing the Zr
content, simultaneously with which specific bands assignable to
In order to verify the chemical environment of the incorpo-
−1
contiguous Zr–O–Zr bond appeared at 501, 580, 752 cm (Fig. 5B),
indicating the formation of aggregated ZrO2 species at high Zr
loading levels [35]. It should be noted that no appreciable differ-
ence was observed in higher wavenumber region of the IR spectra,
in which only a broad band corresponding to the O–H stretching
rated Zr atoms, the samples were characterized with FTIR and
UV–vis techniques. Usually, pure siliceous mesoporous silica
shows several infrared absorption bands at around 460, 806, 1085,
−1
−1
1
235, 1630 cm . The band at 1630 cm is associated with O–H
bending vibration of adsorbed H O molecules and surface silanol
2
3
2
2
1
1
0
0
.0
.5
.0
.5
.0
.5
6
(
A)
(B)
5
2
06 nm
Si/Zr = 5.6
Si/Zr = 1.7
4
Si/Zr = 10.7
Si/Zr = 15.5
Si/Zr = 26.6
Si/Zr = 3.7
Si/Zr = 6.6
3
2
1
Si/Zr = 7.9
.0
2
0
200
00
250
300
350
400
250
300
350
400
Wavelength / nm
Wavelength / nm
Fig. 6. Diffuse reflectance UV–vis spectra of (A) S-ZrSBA15 samples and (B) S-mesoZS samples with different Si/Zr atomic ratios.
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
Y. Kuwahara et al. / Catalysis Today xxx (2013) xxx–xxx
7
(
A)
Si/Zr = 5.6
Si/Zr = 10.7
Si/Zr = 15.5
Si/Zr = 26.6
(B)
Si/Zr = 1.7
Si/Zr = 3.7
Si/Zr = 6.6
Si/Zr = 7.9
1
30
118
117
112
1
15
1
03
5
107
9
1
00 200 300 400 500 600
Temperature / ºC
100 200 300 400 500 600
Temperature / ºC
Fig. 7. NH3–TPD profiles of (A) S-ZrSBA15 samples and (B) S-mesoZS samples with different Si/Zr atomic ratios. TCD signals were normalized by the weight of the samples
used in each experiment.
vibration associated with physisorbed H O molecules and Si–OH
on the other hand, synthesis by using zirconium alkoxide precursor
and under strong acid conditions yielded high-quality mesoporous
zirconosilicates with larger Zr contents embedding crystalline ZrO2
particles inside their silica walls.
2
(
or Zr–OH) terminal groups was observed in all the samples.
.
The above result has been corroborated by UV–vis measurement
Fig. 6). In the diffuse reflectance UV–vis spectra, the S-ZrSBA15
(
samples showed sole absorption bands at 206 nm. This band is
attributed to the presence of monomeric Zr atoms surrounded by
four oxygen-bridged Si atoms, revealing that the incorporated Zr
atoms are dominantly present as monodispersed Zr atoms within
the silica matrix [24,27,28]. On the contrary, the S-mesoZS samples
exhibited absorption bands extended towards longer wavelength
region (∼245 nm). This is the absorption band typical of bulk
3.3. Acidity measurement
The acidity of the catalysts that substantially affects their cat-
alytic performances was investigated by TPD measurement using
NH3 as a probe molecule. Normalized NH –TPD profiles of S-
3
ZrSBA15 and S-mesoZS samples are shown in Fig. 7A and B,
respectively. All samples exhibited desorption peaks positioned at
◦
ZrO [24,27,28], proving that Zr in the S-mezoZS samples are
100–130 C which are assigned to weak acid sites and desorption
2
◦
mainly present in the form of crystalline ZrO2 (possibly contain-
ing monomeric/polymeric zirconium oxide moieties as well, due
to the presence of specific absorption at 200–220 nm).
Thus, the coordination geometry of the incorporated Zr atoms,
as well as the resulting quality of mesoporous structure, alter
depending on the kind and amount of the Zr precursors used and
the synthetic conditions applied; synthesis by using zirconium
oxychloride and under mild conditions afforded mesoporous zir-
conosilicates having monodispersed Zr within the silica matrix, and
peaks distributed in the range of 150–400 C which correspond to
medium acid sites. In the case of S-ZrSBA15 samples, the inten-
sity of desorption peak was enhanced and the desorption peak was
◦
slightly shifted to higher temperature region (from 107 to 130 C) in
accordance with the decrease in the Si/Zr atomic ratio from 26.6 to
5.6. As summarized in Table 2, a quantitative analysis demonstrated
that a higher sulfur content was provided by S-ZrSBA15 having
larger Zr content, and accordingly led to an increased amount of
acid sites. Similarly, S-mesoZS also provided a higher sulfur content
Table 2
Textural properties of sulfated mesoporous zirconosilicate materials.
Catalytic activity d (%)
Sample
Sulfur contenta (wt%)
Acidity
Acid amount b (mmol/g)
Acid density c (␮mol/m2)
Unmodified SBA-15
S-ZrSBA15(26.6)
S-ZrSBA15(15.5)
S-ZrSBA15(10.7)
S-ZrSBA15(5.6)
S-mesoZS(7.9)
S-mesoZS(6.6)
S-mesoZS(3.7)
S-mesoZS(1.7)
–
0.172
0.567
0.689
0.810
0.935
0.631
0.632
0.594
0.527
0.534
0.20
0.97
1.31
1.57
2.06
0.91
0.98
1.07
1.34
10.27
6.2
0.58
0.95
0.97
0.98
0.14
0.28
0.33
0.37
0.60
58.3
66.2
79.0
68.5
43.8
49.6
59.6
63.3
31.8
2–
SO4 /ZrO2
a
Determined by the combination of ICP and EDX analyses.
Amount of NH3 chemisorbed per gram of sample.
b
c
2
2
Calculated by the equation: [acid density (␮mol/m )] = [acid amount (mmol/g)]/[specific surface area of sample (m /g)] × 1000.
d
◦
Yield of ethyl levulinate after 24 h of reaction under the standard reaction conditions (5.0 wt% of catalyst, LA:EtOH = 1:10, 70 C, 700 rpm).
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
8
Y. Kuwahara et al. / Catalysis Today xxx (2013) xxx–xxx
1
00
100
S-ZrSBA15(5.6)
S-ZrSBA15(10.7)
S-ZrSBA15(15.5)
S-ZrSBA15(26.6)
SO4 /ZrO2
SBA15
no catalyst
S-mesoZS(1.7)
S-mesoZS(3.7)
S-mesoZS(6.6)
S-mesoZS(7.9)
(
A)
(B)
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
90
80
70
60
50
40
30
20
10
0
2-
0
5
10
15
20
25
0
5
10
15
20
25
Reaction time / h
Reaction time / h
Fig. 8. Esterification of levulinic acid with ethanol over sulfated mesoporous zirconosilicate catalysts: (A) S-ZrSBA15 with varied Si/Zr atomic ratios, sulfated zirconia
2
−
◦
(
SO4 /ZrO2), unmodified SBA-15 and without catalyst. (B) S-mesoZS with varied Si/Zr atomic ratios. Reaction conditions: 5.0 wt% of catalyst, LA:EtOH = 1:10, 70 C, 700 rpm.
as increasing the Zr content in the samples; however, the sul-
fur contents of S-mesoZS samples were overall smaller than those
found in S-ZrSAB15 samples, indicating that SO4 species are more
easily attached on zirconia via a post-sulfation route. It is also inter-
esting that, in the S-mesoZS samples, the total amount of acid sites
decreased as opposed to the increase in the Zr (and sulfur) content
unmodified SBA-15 materials. The reaction profiles are shown in
Fig. 8, and the yields of ethyl levulinate after 24 h of the reaction
are listed in Table 2.
2
−
The esterification of levulinic acid with alcohols proceeds even
in absence of catalyst, because LA itself shows strong acidity capable
of catalyzing the reaction [16]; a blank experiment demonstrated
that 3.6% of ethyl levulinate can be yielded in the absence of cat-
(
see Table 2). This result clearly indicates that some fraction of sul-
2−
◦
fate anions (SO4 ) in the S-mesoZS samples are inaccessible by
NH3 molecules probably due to the existence of intra-framework
SO4 species, while those attached on the S-ZrSBA15 samples are
most likely surface-exposed and accessible for reactant molecules.
This difference is probably the consequence of the different prepa-
ration method applied (i.e., post- and direct-sulfation procedure).
alyst at 70 C after 24 h of reaction. Unmodified SBA-15 gave an
absolutely low yield of ethyl levulinate (6.2%) after 24 h of reaction,
although it is slightly higher than that in the blank experiment.
As shown in Fig. 8, the reaction was significantly accelerated in
the presence of the solid acid catalysts. When a series of sulfated
Zr-SBA15 were used as catalysts, catalytic activity significantly
increased in the order of: S-ZrSBA15(26.6) < S-ZrSBA15(15.5) < S-
ZrSBA15(5.6) < S-ZrSBA15(10.7), demonstrating that sulfate anions
2
−
3.4. Esterification of levulinic acid with ethanol over sulfated
(SO42 ) act as main active sites for this reaction. Interestingly, this
order is not simply consistent with the order of sulfur content or the
amount of acid sites, implying a possibility of other factors dom-
inating the catalytic activity. In fact, S-ZrSBA15(26.6) afforded an
activity (58.3% after 24 h) far superior to that of the conventional
−
mesoporous zirconosilicates
The esterification of levulinic acid with ethanol was carried out
◦
at 70 C using 5.0 wt% of catalyst and with LA:EtOH = 1:10 over
the sulfated mesoporous zirconosilicates, sulfated zirconia and
100
100
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
90
(
A)
(B)
80
70
60
50
40
30
20
10
0
0.0
0.2
0.4
0.6
0.8
1.0
0.0
0.4
0.8
1.2
1.6
2.0
-2
-1
Acid amount / mmol g
Acid density /
µ
mol m
Fig. 9. (A) Correlation between catalytic activity and acid amount and (B) correlation between catalytic activity and acid density over sulfated mesoporous zirconosilicate
catalysts: (᭹) S-ZrSBA15 samples, (ꢀ) S-mesoZS samples and (ꢁ) unmodified SBA-15.
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
Y. Kuwahara et al. / Catalysis Today xxx (2013) xxx–xxx
9
sulfated zirconia without mesoporosity (31.8% after 24 h), while
these two samples possess almost similar acid amounts of 0.567
100
S-ZrSBA15(10.7)
S-mesoZS(1.7)
90
and 0.534 mmol-NH /g, respectively (see Table 2). Although
3
S-ZrSBA15(5.6) showed the highest acid amount (0.935 mmol-
80
2-
SO /ZrO
NH /g), its activity was lower than that of S-ZrSBA15(10.7) bearing
4
2
3
70
60
50
40
30
20
a smaller amount of acid sites (0.810 mmol-NH /g). Improved cat-
3
alytic activities were also observed over a series of S-mesoZS
catalysts, albeit to a lesser extent in comparison with a series of
S-ZrSBA15 catalysts (cf. Fig. 8A and B), where a higher activity was
obtained basically as increasing the Zr content. Among the solid
acid catalysts examined in this study, S-ZrSBA15 with modest Zr
content, S-ZrSBA15(10.7), was found to be the most effective cat-
◦
alyst, affording 79.0% yield of ethyl levulinate at 70 C after 24 h of
reaction.
To elucidate what is the most dominant factor controlling the
reaction rate, a quantitative study was performed by plotting the
obtained catalytic activities as a function of the amount of the acid
sites Fig. 9A or as a function of the acid density of the catalysts
10
0
1
st
2nd
3rd
4th
5th
Number of cycles
(Fig. 9B), the latter is defined as the amount of chemisorbed NH3
per unit of surface area. As shown in Fig. 9A and B it was found
that the reaction rate in the esterification of levulinic acid corre-
lates better with the density of acid sites, rather than the amount
of acid sites (neither the specific surface area nor the total pore
volume provided any reasonable correlations against the activity).
As an outlying case, S-ZrSBA15(5.6) deviated from this linear cor-
relation, which is probably due to its subpar pore characteristics
Fig. 10. Catalytic activity of S-ZrSBA15(10.7), S-mesoZS(1.7) and SO42 /ZrO2
−
over 5 repeated cycles in the levulinic acid esterification (5.0 wt% of catalyst,
LA:EtOH = 1:10, 70 ◦C, 24 h, 700 rpm).
To examine the reusability of the catalyst, several selected cata-
lysts were subjected to multiple catalytic cycles under the standard
reaction conditions (i.e., 5.0 wt% of catalyst, LA:EtOH = 1:10, 70 C,
◦
(
see Table 1), indicating that the ordered mesoporous structure
2
4 h). After each catalytic run, the catalyst had been recovered
is essential for achieving a high catalytic efficiency. The conven-
tional sulfated zirconia without porous structure largely deviated
from the correlation as well (plots are not shown). Considering
these facts, it is presumable that dispersibility of the acid sites
and the associated accessibility of the organic reactants play an
important role in determining the overall activity [36]. It is believed
that the mesoporosity-derived large surface area allows the acid
◦
by filtration, washed with acetone and then calcined at 600 C in
air for 3 h, thereby discounting the possibility of catalyst deacti-
vation by the strongly adsorbed organic species. Fig. 10 compares
catalytic activities of S-ZrSBA15(10.7), S-mesoZS(1.7) and the con-
ventional sulfated zirconia over five repeated cycles. As can be
seen in Fig. 10, all the catalysts showed a continual activity reduc-
tion during moderate cycling; for example, S-mesoZS(1.7) and
the conventional sulfated zirconia eventually lost 58% and 60% of
their initial activities, respectively, after five catalytic cycles. This
considerable activity reduction can primarily be attributed to the
substantial loss of acid sites during their repeated use, which is
2−
sites (mainly SO4
sites) to be highly dispersed on a solid sur-
face and also enlarges the reaction interface, and thus most of the
acid sites become accessible by reactants and catalytically active.
Furthermore, this analysis also demonstrated that post-sulfation
route is more effective way for achieving high acid density (direct-
2−
caused by elution of sulfate anions (SO4 ) into the alcohol media
for detailed reaction profiles, see parts A and B). In fact, the sul-
fur content in S-mesoZS(1.7) decreased from 0.37 to 0.31 wt%, and
that in the conventional sulfated zirconia decreased from 0.60 to
sulfation route results in the formation of intra-framework SO42
−
(
species) and that the catalytic activity does not so much rely on the
crystalline structure of the incorporated Zr species.
0
.44 wt% after five repeated use. It needs to be addressed that the
3
.5. Reusability of catalyst
activity reduction is not solely attributed to the loss of sulfate
anions, but is also attributed to the structural change of the catalyst.
Nitrogen adsorption–desorption measurement of the spent cata-
lyst revealed that the BET surface area of the conventional sulfated
zirconia decreased from 52 to 29 m /g. This may be the secondary
cause of the continual activity reduction, because the reduced sur-
face area restricts the reaction interface, thereby some of the active
In addition to the catalytic activities, reusability and stability
of the catalyst are of key importance for practical applications,
because such a biomass conversion reaction is supposed to be oper-
ated in a large scale, thereby the lifetime of the catalysts strongly
affects the cost of the overall process.
2
Scheme 2. Synthesis of methyl levulinate from glucose and fructose.
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
1
0
Y. Kuwahara et al. / Catalysis Today xxx (2013) xxx–xxx
acid sites become inaccessible by reactants. On the other hand, S-
ZrSBA15(10.7) catalyst lost only 31% of its initial activity after the
same number of cycles, demonstrating its superior reusability. This
difference in the catalyst durability might be the consequence of
the different chemical environment of Zr species; i.e., the monodis-
persed Zr species incorporated within the silica matrix may have
an ability to strongly stabilize the sulfate anions, whereas the crys-
talline ZrO2 binds the sulfate anions more weakly.
Additionally, structural analysis of the used zirconosilicate cat-
alysts confirmed that they are crystallographically unchanged
without any appreciable porosity reductions even after five
repeated catalytic cycles (for XRD patterns and N2 adsorption
isotherms of the spent catalysts, see parts C and D, respectively).
Regarding the stability of mesoporous zirconosilicate materials, a
number of publications have commonly described that they are
thermally/hydrothermally more stable than a pure siliceous ana-
logue [37–39]. These facts indicate that the sulfated mesoporous
zirconosilicates, especially the sulfated Zr–SBA-15, can potentially
be used as more efficient, reusable and stable solid acid catalysts
for the levulinic acid esterification compared to the conventional
sulfated zirconia. Although a leaching of some fractions of sulfate
anions is likely unavoidable under the reaction conditions applied,
it may be successful to retain the initial activity by re-sulfating the
catalyst before its reuse.
3.6. Direct conversion of sugars over sulfated mesoporous
zirconosilicates
Among the conversion technologies for biomass resources,
direct conversion of lignocellulosic biomass, such as cellulose, glu-
cose and fructose, into levulinate esters in alcohol media using acid
catalysts is one of the most attractive approaches ever reported.
The advantages of this conversion pathway are; a high yield of
product compared to the reaction in aqueous media, minimized
water generation and easy isolation of the product by distillation
[
31,40–43].
To expand the availability of the sulfated mesoporous zir-
conosilicate catalysts, a selected sample, S-mesoZS(1.7), was
applied to the above reaction to directly synthesize levulinate
esters, and was compared to the conventional sulfated zirconia.
The reaction was performed in a stainless steel autoclave reactor
using 2.5 wt% of catalyst and methanol as an alcohol medium at
◦
2
00 C under 0.5 MPa pressure of nitrogen (Scheme 2). The results
of the reactions and some related properties of the catalysts are
summarized in Table 3. More than 99% conversions of sugars were
◦
obtained after 3 h of reaction at 200 C in each case, indicating
that the acidities of the catalysts are strong enough to catalyze
monosaccharides. Over the conventional sulfated zirconia, 50.9%
and 66.4% methyl levulinate yields were obtained from glucose
and fructose, respectively. It should be noted that methyl lactate,
a side reaction product, was scarcely formed under the conditions
examined in this study, but an equivalent mole of methyl formate,
a cleaved counterpart of methyl levulinate, was produced as a
byproduct of methyl levulinate [31,40–43]. The remaining fraction
of yields is considered to be the carbonaceous species deposited
on the catalyst (such as insoluble polysaccharides and humin
compounds), otherwise, the intermediate species such as methyl-
D-glucopyranoside [43]. Unexpectedly, the sulfated mesoporous
zirconosilicate, S-mesoZS(1.7), provided a quite lower methyl lev-
ulinate yield compared to the conventional sulfated zirconia under
the same reaction conditions, while these two catalysts having
almost similar amount of acid sites (0.527 and 0.534 mmol-NH /g,
3
respectively). S-mesoZS(1.7) transformed glucose and fructose into
methyl levulinate in 4.7% and 33.3% yield, respectively, suggesting
that major part of sugars were deposited on the catalyst surface.
This result can be attributed to the mesoporous structure with 2D
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008

GModel
CATTOD-8728; No. of Pages11
ARTICLE IN PRESS
Y. Kuwahara et al. / Catalysis Today xxx (2013) xxx–xxx
11
straight pore channels of S-mesoZS(1.7), in which the monosac-
charide molecules are considered to be easily condensed and
acid-polymerized to form polysaccharides or humin compounds.
The thus formed carbonaceous species cause a contamination and
block the pore channels, and thus result in the low methyl levuli-
nate yields. The above catalytic experiments demonstrated that a
sulfated mesoporous zirconosilicate is not as suitable catalyst as the
conventional sulfated zirconia for direct conversion of sugars into
levulinate esters due to the significant deposition of carbonaceous
species inside the pores, whereas it effectively works as a solid
acid catalyst in the levulinic acid esterification, allowing the high
dispersion of acid sites and facilitating the access of the organic
reactants.
References
[
[
1] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
2] J.N. Chheda, G.W. Huber, J.A. Dumesic, Angew. Chem. Int. Ed. 46 (2007)
7164–7183.
[
[
3] A.A. Peterson, F. Vogel, R.P. Lachance, M. Fröling, M.J. Antal Jr., J.W. Tester,
Energy Environ. Sci. 1 (2008) 32–65.
4] J. Zakzeski, P.C.A. Bruijnincx, A.L. Jongerius, B.M. Weckhuysen, Chem. Rev. 110
(2010) 3552–3599.
[5] F.M.A. Geilen, B. Engendahl, A. Harwardt, W. Marquardt, J. Klankermayer, W.
Leitner, Angew. Chem. Int. Ed. 49 (2010) 5510–5514.
[
6] J.-P. Lange, R. Price, P.M. Ayoub, J. Louis, L. Petrus, L. Clarke, H. Gosselink, Angew.
Chem. Int. Ed. 49 (2010) 4479–4483.
[7] D.J. Braden, C.A. Henao, J. Heltzel, C.C. Maravelias, J.A. Dumesic, Green Chem.
3 (2011) 1755–1765.
1
[
[
8] M. Vasiliu, K. Guynn, D.A. Dixon, J. Phys. Chem. C 115 (2011) 15686–15702.
9] J.C. Serrano-Ruiz, A. Pineda, A.M. Balu, R. Luque, J.M. Campelo, A.A. Romero, J.M.
Ramos-Fernández, Catal. Today 195 (2012) 162–168.
[
[
10] J. Zhang, S. Wu, B. Li, H. Zhang, ChemCatChem 4 (2012) 1230–1237.
11] J. Hegner, K.C. Pereira, B. DeBoef, B.L. Lucht, Tetrahedron Lett. 51 (2010)
4
. Conclusions
2356–2358.
[
[
[
12] Z. Wu, S. Ge, C. Ren, M. Zhang, A. Yip, C. Xu, Green Chem. 14 (2012) 3336–3343.
13] T. Runge, C. Zhang, Ind. Eng. Chem. Res. 51 (2012) 3265–3270.
14] G. Pasquale, P. Vázquez, G. Romanelli, G. Baronetti, Catal. Commun. 18 (2012)
115–120.
15] S. Dharne, V.V. Bokade, J. Nat. Gas Chem. 20 (2011) 18–24.
16] D.R. Fernandes, A.S. Rocha, E.F. Mai, C.J.A. Mota, V. Teixeira da Silva, Appl. Catal.,
A 425–426 (2012) 199–204.
A series of sulfated zirconosilicates having P6mm hexagonal
mesoporous structure with varied Zr content were synthesized by
sol–gel process and via a post-sulfation or via a direct-sulfation
procedure, and were applied to the esterification of levulinic acid
with ethanol to produce ethyl levulinate. A distinct correlation
was observed between the catalytic activity and the density of
acid sites, although less ordered porous solids deviated from this
correction, showing that dispersibility of the acid sites and the
associated accessibility of the organic reactants play an impor-
tant role in determining the overall activity. Among the catalysts
examined, sulfated Zr–SBA-15 with modest Zr content (Si/Zr ratio
of 10.7), prepared via a post-sulfation procedure, was found to
be the most promising catalysts combining a high catalytic effi-
ciency, recyclability and stability, which far outperformed those
of the conventional sulfated zirconia. Recyclability tests demon-
strated that the catalysts suffer from the elution of sulfate anions
into the alcohol media, and thus cause a continual activity reduc-
tion during their repeated use. It was demonstrated that the
sulfated mesoporous zirconosilicates are potential solid acid cat-
alysts for the production of levulinate esters from levulinic acid
and alcohols, but further optimization and improvement of the
catalytic performance are clearly needed for the advanced appli-
cation in direct conversion of cellulosic sugars into levulinate
esters.
[
[
[17] Z.-P. Yan, L. Lin, S. Liu, Energy Fuels 23 (2009) 3853–3858.
[18] X.-L. Du, L. He, S. Zhao, Y.-M. Liu, Y. Cao, H.-Y. He, K.-N. Fan, Angew. Chem. Int.
Ed. 50 (2011) 7815–7819.
[
19] J.M. Tukacs, D. Király, A. Strádi, G. Novodarszki, Z. Eke, G. Dibó, T. Kégl, L.T. Mika,
Green Chem. 14 (2012) 2057–2065.
[20] W.R.H. Wright, R. Palkovits, ChemSusChem 5 (2012) 1657–1667.
[
[
[
21] J.C. Serrano-Ruiz, D. Wang, J.A. Dumesic, Green Chem. 12 (2010) 574–577.
22] K. Li, J. Hu, W. Li, F. Ma, L. Xu, Y. Guo, J. Mater. Chem. 19 (2009) 8628–8638.
23] S. Van de Vyver, J. Geboers, S. Helsen, F. Yu, J. Thomas, M. Smet, W. Dehaen, B.F.
Sels, Chem. Commun. 48 (2012) 3497–3499.
24] X.-R. Chen, Y.-H. Ju, C.-Y. Mou, J. Phys. Chem. C 111 (2007) 18731–18737.
25] L. Fuxiang, Y. Feng, L. Yongli, L. Ruifeng, X. Kechang, Microporous Mesoporous
Mater. 101 (2007) 250–255.
[
[
[26] S.-Y. Chen, L.-Y. Jang, S. Cheng, Chem. Mater. 16 (2004) 4174–4180.
[
27] B.L. Newalkar, J. Olanrewaju, S. Komarneni, J. Phys. Chem. B 105 (2001)
356–8360.
28] S.-Y. Chen, J.-F. Lee, S. Cheng, J. Catal. 270 (2010) 196–205.
8
[
[29] G.D. Yadav, J.J. Nair, Microporous Mesoporous Mater. 33 (1999) 1–48.
[
30] M.L. Grecea, A.C. Dimian, S. Tanase, V. Subbiah, G. Rothenberg, Catal. Sci. Tech-
nol. 2 (2012) 1500–1506.
31] K. Tominaga, A. Mori, Y. Fukushima, S. Shimada, K. Sato, Green Chem. 13 (2011)
810–812.
[
[32] D.-M. Do, S. Jaenicke, G.-K. Chuah, Catal. Sci. Technol. 2 (2012) 1417–1424.
[
33] Y. Kuwahara, D.-Y. Kang, J.R. Copeland, N.A. Brunelli, S.A. Didas, P. Bollini, C.
Sievers, T. Kamegawa, H. Yamashita, C.W. Jones, J. Am. Chem. Soc. 134 (2012)
10757–10760.
[
[
34] Y. Kuwahara, D.-Y. Kang, J.R. Copeland, P. Bollini, C. Sievers, T. Kamegawa, H.
Yamashita, C.W. Jones, Chem.—Eur. J. 18 (2012) 16649–16664.
35] B. Tyagi, K.B. Sidhpuria, B. Shaik, R.V. Jasra, J. Porous Mater. 17 (2009) 699–709.
Acknowledgeents
This work was financially supported by the Grant-in-Aid
awarded to Y.K. from the Institute of Advanced Industrial Science
and Technology (AIST).
[36] J. Iglesias, M.D. Gracia, R. Luque, A.A. Romero, J.A. Melero, ChemCatChem 4
2012) 379–386.
37] Y. Du, S. Liu, Y. Zhang, F. Nawaz, Y. Ji, F.-S. Xiao, Microporous Mesoporous Mater.
21 (2009) 185–193.
38] A.V. Ivanov, S.V. Lysenko, S.V. Baranova, A.V. Sungurov, T.N. Zangelov, E.A.
Karakhanov, Microporous Mesoporous Mater. 91 (2006) 254–260.
39] J.-M. Liu, S.-J. Liao, G.-D. Jiang, X.-L. Zhang, L. Petrik, Microporous Mesoporous
Mater. 95 (2006) 306–311.
(
[
[
[
1
Appendix A. Supplementary data
[
[
[
40] L. Peng, L. Lin, J. Zhang, J. Shi, S. Liu, Appl. Catal. A 397 (2011) 259–265.
41] X. Hu, C.-Z. Li, Green Chem. 13 (2011) 1676–1679.
42] M. Mascal, E.B. Nikitin, ChemSusChem 3 (2010) 1349–1351.
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.cattod.
2
013.11.008.
[43] S. Saravanamurugan, A. Riisager, Catal. Commun. 17 (2012) 71–75.
Please cite this article in press as: Y. Kuwahara, et al., Esterification of levulinic acid with ethanol over sulfated mesoporous zir-
conosilicates: Influences of the preparation conditions on the structural properties and catalytic performances, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.11.008