A sulfated ZrO2 hollow nanostructure as an acid catalyst in the dehydration of fructose to 5-hydroxymethylfurfural

Article abstract of DOI:10.1002/cssc.201300416

Mesoporous hollow colloidal particles with well-defined characteristics have potential use in many applications. In liquid-phase catalysis, in particular, they can provide a large active surface area, reduced diffusion resistance, improved accessibility to reactants, and excellent dispersity in reaction media. Herein, we report the tailored synthesis of sulfated ZrO ₂ hollow nanostructures and their catalytic applications in the dehydration of fructose. ZrO₂ hollow nanoshells with controllable thickness were first synthesized through a robust sol-gel process. Acidic functional groups were further introduced to the surface of hollow ZrO ₂ shells by sulfuric acid treatment followed by calcination. The resulting sulfated ZrO₂ hollow particles showed advantageous properties for liquid-phase catalysis, such as well-maintained structural integrity, good dispersity, favorable mesoporosity, and a strongly acidic surface. By controlling the synthesis and calcination conditions and optimizing the properties of sulfated ZrO₂ hollow shells, we have been able to design superacid catalysts with superior performance in the dehydration of fructose to 5-hydroxymethyfurfural than the solid sulfated ZrO₂ nanocatalyst. Core of the matter: Sulfated ZrO₂ hollow shells with favorable mesoporosity, excellent dispersity in liquid media, and a strongly acidic surface show significantly enhanced catalytic activity toward the dehydration of fructose to 5-hydroxymethylfufural (see picture). Copyright

Full text of DOI:10.1002/cssc.201300416

CHEMSUSCHEM
FULL PAPERS
DOI: 10.1002/cssc.201300416
A Sulfated ZrO Hollow Nanostructure as an Acid Catalyst
2
in the Dehydration of Fructose to 5-Hydroxymethylfurfural
Ji Bong Joo, Austin Vu, Qiao Zhang, Michael Dahl, Minfen Gu, Francisco Zaera, and
[a]
Yadong Yin*
Mesoporous hollow colloidal particles with well-defined char-
acteristics have potential use in many applications. In liquid-
phase catalysis, in particular, they can provide a large active
surface area, reduced diffusion resistance, improved accessibili-
ty to reactants, and excellent dispersity in reaction media.
of hollow ZrO shells by sulfuric acid treatment followed by
2
calcination. The resulting sulfated ZrO2 hollow particles
showed advantageous properties for liquid-phase catalysis,
such as well-maintained structural integrity, good dispersity, fa-
vorable mesoporosity, and a strongly acidic surface. By control-
ling the synthesis and calcination conditions and optimizing
Herein, we report the tailored synthesis of sulfated ZrO hollow
2
nanostructures and their catalytic applications in the dehydra-
the properties of sulfated ZrO hollow shells, we have been
2
tion of fructose. ZrO hollow nanoshells with controllable thick-
able to design superacid catalysts with superior performance
in the dehydration of fructose to 5-hydroxymethyfurfural than
2
ness were first synthesized through a robust sol–gel process.
Acidic functional groups were further introduced to the surface
the solid sulfated ZrO nanocatalyst.
2
Introduction
[
2b,e,f,8]
The replacement of liquid acids by solid alternatives has seen
considerable attention recently because solid-acid catalysis is
highly desirable for designing green chemical processes with
significantly reduced pollution in comparison to the cases with
and dehydration.
Their advantageous catalytic character-
istics stem from not only the strength of the acid, but also the
type of acidity (Brønsted and Lewis); therefore, enabling en-
hanced activity and selectivity. As a result, much effort was
[
1]
typical liquid acids. Solid superacids, defined as catalytic ma-
terials with an acidic strength stronger than 100% sulfuric
acid, are potentially useful for many chemical reactions that re-
previously devoted to synthesizing highly active sulfated ZrO
2
[9]
catalyst with well-controlled properties.
Because chemical reactions typically occur on the surface of
catalysts, one can rationally design ideal catalysts that consist
[2]
quire strong acidic properties under mild reaction conditions.
In general, several solid superacids, including heteropolyacids
of very small ZrO nanocrystals with a strongly acidic surface
2
[
3]
(
acidic polyoxometalate), anion-modified metal oxides (TiO2,
to maximize the catalytically active surface area. However, to
[
4]
ZrO2), and binary metal oxides (TiO ÀSiO , Al O ÀZrO , TiO À maintain colloidal stability, nanocrystals synthesized by solu-
2
2
2
3
2
2
[
1a,5]
ZrO , and others),
have been extensively studied in funda-
tion-phase processes are usually covered with organic surfac-
tants, which significantly hinder diffusion of reactant molecules
to the nanocrystal surface, resulting in poor catalytic efficien-
2
mental research as well as for industrial applications.
Zirconium dioxide (ZrO2, zirconia) has been used in
a number of applications, such as hard ceramics, solid electro-
[10]
cy. To make a clean surface on the ZrO nanocrystals, direct
2
[
6]
lytes, sensors, drug-delivery systems, and catalysis. Since Hino
et al. first demonstrated the highly acidic or superacidic prop-
heat treatment at high temperatures may be applied to burn
off the organic surfactants. Such calcination processes, howev-
er, often cause small nanocrystals to sinter into larger ones
and/or reconstruct into new shapes, resulting in the loss of the
unique catalytic properties seen in the original small nanocrys-
tals. Another challenge to the use of colloidal nanocrystals in
solution-phase catalysis is that they are too small to be effi-
ciently recovered from liquid media by either conventional
centrifugation or filtration after running the reaction. To ad-
dress this issue, catalytic materials can be synthesized in the
form of a mesoporous structure by templating strategies fol-
lowed by calcination at high temperatures, which ensures
[
7]
erties of sulfate-modified ZrO surfaces, the synthesis and cat-
2
alysis of this class of materials have been intensely studied.
As a new type of acid catalysts, these materials have been
used for practical applications in many chemical reactions,
such as alkylations, isomerization, condensation, etherification,
[a] Dr. J. B. Joo, A. Vu, Dr. Q. Zhang, M. Dahl, Dr. M. Gu, Prof. F. Zaera,
Prof. Y. Yin
Department of Chemistry, University of California Riverside
9
00 University Ave., Riverside, CA, 92521 (USA)
E-mail: yadong.yin@ucr.edu
a clean surface, a high surface area, and large pores within the
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201300416. This manuscript is part of
a Special Issue on synthesis and use of shape-controlled particles. The
Table of Contents will appear here. This text will be updated once the
Special Issue is assembled.
[2c,9a–c]
framework.
Over the past few decades, much effort has
been devoted to synthesizing mesoporous ZrO and/or ZrO -
2
2
supported mesoporous materials. It has also been shown that
producing mesoporous ZrO in the form of colloidal particles
2
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 9
&
1
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
could further improve mass transfer during chemical reac-
[
11]
tions. An additional benefit of making catalysts in colloidal
form is that they can usually be stabilized by the electrostatic
charges carried on the surface, so that no additional stabilizer
is needed to maintain dispersion stability. The relatively larger
size of particles relative to typical nanocrystals also makes
them easier to recycle from the reaction media by, for exam-
ple, centrifugation. In many cases, the core region of a mesopo-
rous microsphere is denser than the surface. Additional im-
provements may be achieved by removing the core portion to
[
6c,d,12]
produce porous hollow shells,
which are expected to
show more active sites per unit mass, reduced diffusion resist-
ance, and improved accessibility of the reactant to the active
[
12b,13]
surface.
Scheme 1. Schematic illustration of the procedure for the synthesis of sulfat-
ed ZrO hollow nanoshells.
2
Herein, we report the synthesis of sulfated ZrO2 hollow
nanostructures with controllable physical and chemical proper-
ties and their catalytic application in the dehydration of fruc-
tose to 5-hydroxymethyfurfural (HMF). More specifically, we
first synthesized SiO @ZrO core–shell particles by coating an
[16]
ZrO shells. In this work, we mainly optimized the amount of
2
surfactant and ZBOT to control the thickness of the ZrO layer
2
2
2
amorphous ZrO layer on the surface of colloidal silica tem-
and prevent particle agglomeration. It is well known that sur-
factants play an important role by providing electrostatic and/
or steric hindrance to prevent particle aggregation during col-
2
plates through a sol–gel process, removed the silica cores by
NaOH etching to form hollow ZrO shells, introduced sulfate
2
[16,17]
functional groups to ZrO surface by treatment with sulfuric
loidal synthesis.
HPC appears to be an effective surfactant
2
acid, and finally calcined the samples to produce sulfated ZrO2
in this synthesis. Figure 1 compares the as-etched amorphous
hollow nanoshells. The resulting sulfated ZrO shells show ad-
2
vantageous characteristics towards catalysis, including well-
maintained structural integrity, good dispersity in polar sol-
vents, favorable mesoporous structure, and strongly acidic sur-
face. The flexible synthetic procedure allows easy structural op-
timization and produce hollow sulfated ZrO shells with signifi-
2
cantly higher catalytic performance than solid particles. Our
work represents the first report on the controllable synthesis
of sulfated ZrO mesoporous nanoshells for acid-catalyzed con-
2
version of fructose to HMF.
Results and Discussion
Amorphous ZrO hollow nanoshells were synthesized by a con-
2
ventional templating method that combined sol–gel coating
[
14]
and selective etching processes. With the introduction of sul-
fate ions to the surface of as-etched ZrO shells followed by
2
calcination, sulfated ZrO2 hollow nanostructures containing
both Brønsted and Lewis acid sites can be prepared
(Scheme 1). Specifically, a typical synthesis involves four main
steps: 1) preparation of uniform colloidal silica spheres as tem-
[
15]
plates by using the Stçber method, 2) deposition of a ZrO2
layer on the silica surface to form SiO @ZrO core–shell struc-
2
Figure 1. TEM images of as-etched amorphous ZrO hollow nanoshells when
different amounts of HPC were used during the synthesis: a) 0, b) 2, c) 10,
and d) 20 mg. The precursor (ZBOT) was fixed at 0.4 mL.
2
2
tures through the sol–gel reaction of zirconium butoxide
ZBOT) with hydroxypropyl cellulose (HPC) as a surfactant,
) etching to remove the silica template by using an aqueous
(
3
solution of NaOH, and 4) sulfuric acid treatment followed by
calcination to introduce acidic functional groups to the sur-
face.
hollow ZrO shells prepared by varying the amount of HPC.
2
Without HPC, the as-etched hollow ZrO shells are severely ag-
2
gregated with a relatively thick ZrO shell layer. With the addi-
2
The hydrolysis and condensation of the ZBOT precursor are
highly influenced by several factors, such as the concentrations
of water, template, ZBOT, and surfactant. Fine-tuning of these
factors allows convenient control over the structure of the
tion of increasing amounts of HPC, the tendency for particle
aggregation decreases dramatically and perfectly isolated indi-
vidual particles of hollow ZrO shells can eventually be pro-
2
duced (Figure 1c and d). The increase in the concentration of
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 9
&
2
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
surfactant also affects the thickness of the ZrO2 shell layer,
which in fact reduces continuously. The HPC polymer contains
many hydroxyl and ether groups that can form hydrogen
bonds with the ZrO surfaces. The hydrolyzed ZrO species are
As shown in Figure 4a–c, we also confirm the structural in-
tegrity and stability of sulfated ZrO hollow samples calcined
2
at high temperatures. The sulfated hollow shells can retain
small ZrO crystal grains and a highly porous structure, even
2
2
2
stabilized by HPC molecules, making it difficult for them to be
after calcination at a high temperature of 8008C (Figure 4c).
deposited to the growing ZrO surfaces, and therefore, produc-
Figure 4d shows the XRD patterns of the sulfated ZrO hollow
2
2
[
16]
ing thinner shells. On the other hand, the binding of HPC ef-
fectively prevents close contact of two ZrO2 surfaces, and
therefore, suppresses interparticle agglomeration. When an
excess of HPC is used, the products show no clear core–shell
morphology.
samples calcined at different temperatures. The sample cal-
cined at 6008C (SZH-600) displays broad diffraction patterns,
indicating the amorphous nature of the sol–gel-derived ZrO2.
Consistent with many other syntheses, the coating of a ZrO2
layer can also be controlled by varying the concentration of
[
12c]
the precursor.
As shown in Figure 2a, when 0.2 mL of ZBOT
was used, the as-synthesized hollow ZrO sample shows a shell
2
about 21 nm thick without apparent particle aggregation.
As the amount of ZBOT increases, the shell thickness continu-
ously increases (Figure 2b–e). The average thicknesses of the
ZrO shells are estimated to be approximately 34, 40, 48, and
2
6
1
2 nm for the samples prepared by using 0.4, 0.6, 1.0, and
.4 mL of ZBOT, respectively (Figure 2 f). Importantly, the use of
a large amount of ZBOT (1.4 mL) does not lead to aggregation:
all of the hollow ZrO shells remain as isolated particles, as evi-
2
denced by the low-magnification TEM images given in Fig-
ure 2e.
To produce solid-acid catalysts, the hollow shells were sub-
jected to sulfuric acid treatment and then calcination at differ-
ent temperatures. Because the sample in Figure 2c shows
a suitable shell thickness and no aggregation, we focus our
subsequent discussion on hollow shells prepared under the
same conditions. The calcined hollow shell samples are denot-
ed SZH-x, in which x is the calcination temperature. Table 1
summarizes the notations, preparation conditions, and particle
morphologies of the ZrO samples employed in the catalysis
2
studies.
The pore properties and crystalline structures of the sulfated
ZrO hollow nanostructures (SZH-x) were investigated by N2
2
adsorption and XRD. Figure 3 shows the N adsorption–desorp-
2
tion isotherms and the corresponding Barrett–Joyner–Halenda
(BJH) pore size distributions obtained for the sulfated ZrO2
hollow shells. All samples display type IV isotherms with a hys-
teresis loop, indicating mesoporous characteristics. All of the
sulfated samples show large adsorption volumes in the mono-
layer region, indicating a porous structure with a high BET sur-
face area. The surface areas for samples calcined at 600, 650,
2
Figure 2. TEM images of as-etched amorphous hollow ZrO particles when
different amounts of ZBOT were used during the synthesis: a) 0.2, b) 0.4,
c) 0.6, d) 1.0, and e) 1.4 mL. f) The relationship between the amount of ZBOT
and shell thickness of the hollow samples. The amount of HPC was fixed at
10 mg.
2
À1
700, and 8008C were 134, 111, 114, and 115 m g , respective-
ly; these are much larger values
than that of the control sample
of sulfated ZrO2 solid particles
2
Table 1. Notations, preparation conditions, and morphology of ZrO samples.
2
À1
(SZS-650, 50 m g ). Although
Sample
Preparation method
Morphology
there are no sharp distribution
peaks in the BJH pore size distri-
bution curves, all sulfated hollow
samples show pores with sizes
in the range of 2 to 10 nm (Fig-
ure 3b).
SZH-600
SZH-650
SZH-700
SZH-800
ZH-650
SZS-650
ZS-650
etching SiO
etching SiO
etching SiO
2
2
2
from SiO
from SiO
from SiO
2
2
2
@ZrO
@ZrO
@ZrO
2
2
2
, sulfuric acid treatment, then calcination at 6008C
, sulfuric acid treatment, then calcination at 6508C
, sulfuric acid treatment, then calcination at 7008C
hollow
hollow
hollow
hollow
hollow
solid
etching SiO2 from SiO @ZrO , sulfuric acid treatment, then calcination at 8008C
etching SiO
sulfuric acid treatment of solid ZrO
calcination of solid ZrO
2
2
2
from SiO
2
@ZrO
2
, then direct calcination at 6508C
particles, then calcination at 6508C
particles at 6508C
2
2
solid
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 9
&
3
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
The sulfated hollow shells calcined at 6508C (SZH-650) displays
clear peaks for the tetragonal phase at 2q=30.4, 34.8, and
50.3, which can be attributed to the (111), (200), and (220)
planes, respectively. As the calcination temperature increased
from 650 to 7008C, the peak intensity was further enhanced.
However, calcination at an even higher temperature (8008C)
makes no significant difference to the crystallinity. The average
grain sizes of the sulfated ZrO samples, as estimated from the
2
widths of the tetragonal (220) peak in the XRD pattern by
using the Scherrer formula, are approximately 6.1, 6.9, and
7
nm for samples calcined at 650, 700, and 8008C, respectively.
The surface functional groups and their thermal decomposi-
tion behavior were investigated by FTIR spectroscopy and
thermogravimetric analysis (TGA), respectively. Figure 5a shows
Figure 3. a) N
sulfated ZrO
2
isotherms and b) corresponding BJH pore size distributions of
hollow samples (SZH-x) calcined at different temperatures.
2
2
Figure 5. a) FTIR and b) TGA results of ZrO samples employed herein.
the FTIR spectra of the sulfated ZrO hollow samples calcined
2
at different temperatures (SZH-x) and the non-sulfated hollow
ZrO shells calcined at 6508C (ZH-650). All samples display an
2
À1
absorption peak in the range of 500–650 cm , which can be
[18]
assigned to ZrÀO stretching. The IR spectra of the sulfated
hollow sample calcined at 6008C (SZH-600) shows a pattern of
four bands (dashed lines in Figure 5a) appearing at about
À1
1
220, 1140, 1059, and 993 cm , which are assigned to biden-
Figure 4. Representative TEM images of final sulfated ZrO
calcined at 600 (a), 650 (b), and 8008C (c) for 2 h. d) XRD patterns of sulfated
ZrO hollow samples (SZH-x) calcined at different temperatures.
2
hollow samples
2À
tate SO4 ions coordinated to ZrO in C n symmetry with a n3
2
2
[19]
2
vibration, indicating a strongly superacidic surface. A band
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 9
&
4
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
À1
at about 1380 cm also appears, which may be attributed to
[20]
the formation of sulfones or sulfonyl groups on the surface.
When the calcination temperature increases to 6508C, the rep-
resentative peaks can still be observed, which indicates the sta-
bility of the surface sulfate groups. When the samples are cal-
cined at higher temperatures, for example, 700 and 8008C, the
À1
four IR bands in the range 990–1250 cm become progres-
sively weaker until they finally disappear. When the as-etched
hollow ZrO sample is directly calcined at 6508C without intro-
2
ducing any sulfate ions (ZH-650), it shows only broad absorp-
À1
tion bands in the range of 500–650 and 990–1250 cm with-
out any clear peaks related to the sulfate groups.
Figure 5b shows TGA data of sulfated ZrO hollow samples
2
calcined at different temperatures and sulfated ZrO solid parti-
2
cles calcined at 6508C (SZS-650). All samples display weight
loss in the range of room temperature to 6008C, which is at-
tributed to desorption of physically and chemically adsorbed
water molecules and the dehydroxylation process on the sur-
face of ZrO . In addition, significant weight loss is clearly ob-
2
served in the temperature range of 650 to 8508C, which is at-
tributed to the decomposition of sulfate species with the evo-
[
2e,21]
lution of sulfur oxide.
The sulfated ZrO hollow sample cal-
2
cined at 6008C (SZH-600) displays the biggest weight loss in
all temperature ranges. When the calcination temperature is
not high enough (e.g., 6008C), the particle surface contains
a considerably large amount of hydroxyl groups, as well as
water molecules attracted by the surface sulfate groups, and
therefore, the sample shows significant weight loss across the
whole temperature range during TGA. For samples calcined at
higher temperatures, the weight loss continuously decreases,
which indicates enhanced condensation of surface hydroxyl
groups and a decreased amount of sulfate species. When the
sample was calcined at 8008C, most sulfate species decom-
posed, resulting in no clear weight loss in the range of 650–
Figure 6. NH
sulfated ZrO
calcined at 6508C; and b) sulfated ZrO
3
desorption results for a) sulfated ZrO
solid particles (SZS-650), and bare ZrO
2
hollow shells (SZH-650),
2
2
solid particles (ZS-650)
2
hollow nanoshells calcined at differ-
ent temperatures.
8
508C during TGA. These results are consistent with the
tion signal, indicating the largest amount of active acidic sites
of the three catalysts. The area ratio of ammonia desorption is
about 1:0.43:0.06 for SZH-650, SZS-650, and ZS-650, respective-
ly. Although these ratio values cannot reflect their exact cata-
lytic performance, the amount of active acidic sites is directly
related to the catalytic activity; this is further discussed in the
next section.
above-mentioned FTIR experimental data. We also analyzed
the thermogravimetric behavior on a control sample of sulfat-
ed ZrO solid particles calcined at 6508C (SZS-650). It also dis-
2
plays two major weight losses in the temperature ranges from
room temperature to 6008C and 650 to 8508C; this indicates
similar behavior, although the degree of weight loss is much
smaller than that of hollow samples due to the smaller surface
area.
We also performed ammonia desorption studies for sulfated
ZrO hollow samples calcined at different temperatures (SZH-
2
To compare the amount of active sulfate functional groups
on the surface of the sulfated ZrO2 catalyst, we performed
temperature-programmed desorption (TPD) experiments by
using NH as the adsorbate (NH -TPD). Figure 6a shows the
x). The results are shown in Figure 6b. All samples display clear
NH desorption signals in the range from 100 to 5008C due to
3
acidic surface properties. As expected, sulfated ZrO2 hollow
catalysts calcined at relatively low temperatures (600 and
6508C) desorbed a larger amount of ammonia than those cal-
cined at higher temperatures (700 and 8008C). As discussed
previously, at high calcination temperatures, the surface sulfate
groups may be partially decomposed and the surface acidity
3
3
ammonia desorption behavior of a sulfated ZrO2 hollow
sample (SZH-650), a sulfated ZrO solid sample (SZS-650), and
2
a bare ZrO solid sample (ZS-650); all calcined at 6508C. As ex-
2
pected, the bare ZrO solid sample calcined at 6508C (ZS-650)
2
showed a very small NH3 desorption signal, indicating few
markedly decreases. The sulfated ZrO hollow sample calcined
2
acidic sites, whereas the sulfated ZrO solid sample calcined at
at 6508C shows the highest desorption peak and the relative
desorption peak signal of the sulfated hollow samples follows
the order SZH-650>SZH-600>SZH-700>SZH-800. Although
SZH-600 showed the largest weight loss in TGA, SZH-650 de-
sorbs a larger amount of ammonia. This inconsistency has
2
6
508C (SZS-650) showed a clear increase in the NH desorption
3
signal in the temperature range from 100 to 5008C, indicating
superacidic properties. The sulfated ZrO hollow sample cal-
2
cined at 6508C (SZH-650) displays a much higher NH desorp-
3
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 9
&
5
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
[
22]
been explained previously by Ward and Ko. They pointed
out that, in amorphous sulfated ZrO , sulfate was embedded in
2
the bulk of the solid framework and had a predominantly ionic
character. This explains the relatively high sulfate content in
TGA, but low number of acid sites observed in TPD studies for
sample SZH-600. Upon crystallization into the tetragonal
phase, as in the case of SZH-650, sulfate is displaced from the
bulk of the solid and moves to the surface; this gives rise to
more “active” covalent sulfate species that are responsible for
strong Brønsted acidity.
The catalytic activity of our sulfated ZrO catalysts was evalu-
2
ated by catalyzing the dehydration reaction of fructose to
HMF. HMF is a versatile biomass-derived chemical that can be
used as a platform to synthesize a broad range of high-value
chemicals currently obtained from petroleum industry. In addi-
tion, liquid fuels derived from HMF by using chemical process-
es are potential alternatives to ethanol-based fuel, which is
[
23]
currently obtained by fermentation processes. To obtain sus-
tainable liquid fuels and high-value compounds, an efficient
conversion process from cellulosic biomass, including cellulose,
glucose, and fructose, to HMF is highly desirable. The HMF
molecule can be obtained by dehydrating three water mole-
cules from fructose in the presence of an acid catalyst.
Figure 7a shows the production yield of HMF versus time by
using various catalysts, including the sulfated ZrO2 hollow
sample (SZH-650), the sulfated ZrO2 solid sample (SZS-650),
and the bare ZrO solid particles (ZS-650). In a blank test (with-
2
out catalyst), the HMF yield was about 5.7% after reaction at
1208C for 60 min. The ZrO2 solid particle catalyst prepared
without sulfuric acid treatment (ZS-650) shows no clear activity
in terms of HMF yield; this indicates low acidic activity. Produc-
tion of HMF is significantly enhanced by using the sulfated
Figure 7. HMF yield versus time when various ZrO
in the dehydration of fructose: a) sulfated ZrO hollow nanoshells (SZH-650),
sulfated ZrO solid particles (SZS-650), and bare ZrO solid particles (ZS-650)
calcined at 6508C, and b) sulfated ZrO hollow nanoshells calcined at differ-
2
based catalysts are used
2
2
2
2
ZrO solid catalyst, which shows about 44% HMF yield. The
2
ent temperatures. All scale bars in a) are 50 nm.
sulfated ZrO hollow sample calcined at 6508C (SZH-650) dis-
2
plays the highest HMF yield (ca. 64%) among the catalysts em-
ployed herein. Consistent with previous characterization data,
the high catalytic activity of the hollow sample should be at-
tributed to the large surface area, large number of active
acidic sites, and improved accessibility of the reactants to the
catalyst surface.
brown after the dehydration reaction; this suggests that the
organic products may adsorb to the catalyst surface. Also, as
shown in the TGA measurements given in Figure S2 in the
Supporting Information, the SZH-650 catalyst after four reac-
tion cycles displays different behavior of weight loss compared
with as-synthesized SZH-650 catalyst (Figure 4b). Additional
weight loss in the range from 300 to 5008C can be observed
for the catalyst after reaction; this should be attributed to the
decomposition of adsorbed organic compounds on the cata-
lyst surface. We have also confirmed that the active sulfate
In addition, we investigated the deactivation behavior and
reusability of the SZH-650 catalyst. During the cycling tests on
the dehydration of fructose to HMF, the yield gradually de-
creased (Figure S1 in the Supporting Information). It is well
known that the deactivation of sulfated ZrO catalysts can be
2
caused by either changes of sulfur oxidation resulting from the
formation of organosulfur complexes or the blockage of active
sites by organic compounds formed during the reactions.
Because sulfur with an oxidation state of +6 is known to be
the most active in sulfated zirconia catalyst, partial reduction
of this oxidation state during catalytic reactions by either the
formation of an organosulfur complex or the complete block-
age of active sites may result in a substantial loss in catalytic
groups are strongly bonded to the surface of ZrO and they do
2
not leach out in reaction conditions (Figure S3 in the Support-
ing Information). To verify the intrinsic activities of our sulfated
ZrO catalysts, we compared HMF production activity per unit
2
surface area by using a small amount of catalyst to rule out
the limitation of molecular diffusion and maximize the utiliza-
tion degree of the catalyst surface. Although the sulfated ZrO2
hollow sample (SZH-650) showed an overall larger production
yield than the solid catalyst (SZS-650), the HMF production ac-
tivity per unit surface area is the same for both catalysts (Fig-
ure S4 in the Supporting Information). It is, therefore, conclud-
[
24]
activity. We believe, in our case, that deactivation is most
likely to be due to adsorption of organic compounds, which
lead to blockage of active sites. One piece of evidence is the
color change of the catalyst from originally white to light
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 9
&
6
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
ed that the better overall performance of the hollow catalyst is
mainly due to its higher surface area.
ized water (0.1 mL). After stirring for 30 min, a desired amount of
ZBOT (80% in butanol) in ethanol (4.5 mL) was injected into the
mixture by using a syringe pump at a rate of 0.25 mLmin . Then
À1
We also investigated the catalytic activities of sulfated
hollow samples calcined at different temperatures. The results
the mixture was stirred at 600 rpm for 20 h at room temperature.
The precipitate was isolated by centrifugation, washed with etha-
nol, and redispersed in water (20 mL) to give SiO @ZrO core–shell
are shown in Figure 7b. As expected, the sulfated ZrO hollow
2
2
2
sample calcined at 6508C shows the highest HMF yield and
the catalytic activities of the sulfated hollow samples in terms
of HMF yield follow the order SZH-650>SZH-600>SZH-700>
composites. To produce the hollow structures, the SiO @ZrO dis-
2
2
persion was etched by an aqueous solution of NaOH (4 mL, 2.5m).
After etching for 6 h, the hollow ZrO particles were finally isolated
2
SZH-800. This trend is consistent with the NH desorption re-
by centrifugation, washed with deionized water and ethanol, and
dried under vacuum.
3
sults. For the dehydration reaction of fructose to HMF, the cat-
alytic activities of this class of ZrO -based catalysts are ulti-
2
Sulfuric acid treatment and calcination: The as-etched hollow ZrO2
particles (typically 0.2 g) were dispersed in an aqueous solution of
H SO (0.1m, 5 mL) and stirred for 1 h. The resulting samples were
mately determined by the active acid sites, which can be af-
fected by several factors, including surface area, crystalline
phase, and surface properties. When the calcination tempera-
ture is too low, in the case of SZH-600, a large portion of sul-
fate species is embedded in the bulk solid, so they do not con-
tribute to the catalysis. Calcination of the sample at a higher
temperature not only produces the metastable tetragonal
2
4
isolated by centrifugation, washed with deionized water and etha-
nol, and dried under vacuum. The dried ZrO samples were cal-
2
cined at the desired temperature for 2 h in air to obtain sulfated
ZrO hollow samples (SZH-x).
2
Control experiment: Solid ZrO sample was prepared by using
2
a process similar to that for synthesizing hollow shells, except the
use of silica templates. Specifically, HPC (10 mg) was dissolved in
a mixture of ethanol (25 mL) and deionized water (0.1 mL). After
stirring for 30 min, ZBOT (0.6 mL) in ethanol (4.5 mL) was injected
into the mixture by using a syringe pump. After finishing the injec-
tion, the mixture was stirred at 600 rpm for 20 h at room tempera-
ture. The precipitate was isolated by centrifugation, washed with
ethanol, and dried under vacuum. Sulfuric acid treatment and calci-
nations were conducted by the same method as that described
above.
phase, which is regarded as the most active form of ZrO -
2
[
2b,4a,6a]
based catalysts,
but also drives the active sulfate species
to the surface; therefore, significantly enhances the catalytic
activity, as observed in sample SZH-650. Calcination at even
higher temperatures (700 and 8008C) partially decomposes the
sulfate species on the surface, and therefore, leads to less
active catalysts.
Conclusions
Characterization: The morphology of the samples was character-
ized by TEM (Tecnai 12). The nitrogen adsorption isotherms were
obtained at 77 K by using a Quantachrome NOVA 4200e surface
area and pore size analyzer. The crystal phase was determined by
XRD analysis using a Bruker D8 advance diffractometer with CuKa
radiation (l=1.5406 ꢁ). FTIR spectroscopic measurements were
conducted by using a Bruker Equinox-55 FTIR spectrometer.
We have demonstrated the tailored synthesis of sulfated ZrO2
hollow nanostructures with well-defined characteristics and en-
hanced catalytic activities in the dehydration of fructose to
HMF. The synthetic method involves the formation of a ZrO2
layer on the surface of a SiO template through a sol–gel pro-
2
cess, base etching to remove the SiO core, sulfuric acid treat-
To measure the acidity of the samples, NH -TPD was performed by
3
2
using a conventional TPD machine (AutoChem II, Micrometritics)
equipped with a thermal conductivity detector (TCD). TGA was
conducted by using a Seiko 220U TG-DTA instrument.
ment to introduce sulfate ions, and finally calcination to pro-
duce sulfated ZrO hollow shells. A robust procedure has been
2
developed for coating ZrO onto colloidal silica with controlla-
2
ble thickness and minimum aggregation. Sulfuric acid treat-
ment followed by calcination introduced acidic species to the
Catalytic activity tests: The catalytic activity was evaluated by the
dehydration reaction of fructose to HMF. Typically, a catalyst
(
(
15 mg) was added to a solution of fructose dissolved in DMSO
10 mL, 1.665 mm) in a 25 mL glass flask, and well dispersed by
surface of the hollow ZrO shells, making them excellent solid-
2
acid catalysts. With advantageous characteristics, including
well-maintained hollow morphologies, an active tetragonal
crystalline phase, good dispersity in liquid media, well-devel-
oped mesoporosity, and a strongly acidic surface, the sulfated
ultrasonication for 5 min. The reaction was initiated by putting the
reactor cell into a preheated oil vessel (1208C) controlled by an
electric heater equipped with a PID controller. The concentration
of HMF was analyzed by means of a HPLC (Waters 2790) system
equipped with a UV detector (UVIS-201, Perseptive Biosystem). The
mobile phase was an aqueous solution of 12 vol% acetonitrile
ZrO hollow catalysts showed significantly enhanced catalytic
2
activity in the dehydration of fructose to HMF.
À1
with a flow rate of 0.4 mLmin . The target product was separated
and quantified by using a Kromasil C18 column connected to a UV
detector (l=285 nm).
Experimental Section
Synthesis of hollow ZrO2 shells: Colloidal silica particles as tem-
plates were prepared through a modified Stçber method by
mixing tetraethyl orthosilicate (TEOS, 99%, 0.86 mL) with deionized
water (4.3 mL), ethanol (23 mL), and an aqueous solution of ammo-
nia (28%, 0.62 mL). After 4 h, the silica particles were separated by
centrifugation, washed with ethanol, and then redispersed in etha-
nol (5 mL). The above silica solution was dispersed in a solution
containing HPC of a desired amount, ethanol (20 mL), and deion-
Acknowledgements
Financial support for this project was provided by the U.S. De-
partment of Energy (DE-FG02-09ER16096). Y.Y. also thanks the Re-
search Corporation for Science Advancement for the Cottrell
Scholar Award and DuPont for a Young Professor Grant. F.Z. ac-
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 9
&
7
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
knowledges additional funding from the U.S. National Science
Foundation.
[10] a) J. Joo, T. Yu, Y. W. Kim, H. M. Park, F. Wu, J. Z. Zhang, T. Hyeon, J. Am.
Chem. Soc. 2003, 125, 6553–6557; b) S. Zhou, G. Garnweitner, M. Nie-
derberger, M. Antonietti, Langmuir 2007, 23, 9178–9187; c) N. Zhao, D.
Pan, W. Nie, X. Ji, J. Am. Chem. Soc. 2006, 128, 10118–10124.
[
[
[
11] a) P. M. Arnal, F. Schꢃth, F. Kleitz, Chem. Commun. 2006, 1203–1205;
b) D. Chen, J. Liu, P. Wang, L. Zhang, J. Ren, F. Tang, W. Wu, Colloids Surf.
A 2007, 302, 461–466; c) X. Sun, J. Liu, Y. Li, Chem. Eur. J. 2006, 12,
Keywords: dehydration · mesoporous materials · sulfated
zirconia · superacidic systems · nanostructures
2
039–2047.
12] a) N. C. Strandwitz, S. Shaner, G. D. Stucky, J. Mater. Chem. 2011, 21,
0672–10675; b) Q. Zhang, I. Lee, J. B. Joo, F. Zaera, Y. Yin, Acc. Chem.
[
1] a) J. Navarrete, T. Lopez, R. Gomez, F. Figueras, Langmuir 1996, 12,
385–4390; b) R. M. de Almeida, L. K. Noda, N. S. GonÅalves, S. M. P. Me-
1
4
Res. 2013, 46, 1816–1824; c) J. B. Joo, Q. Zhang, M. Dahl, F. Zaera, Y.
Yin, J. Mater. Res. 2013, 28, 362–368.
neghetti, M. R. Meneghetti, Appl. Catal. A 2008, 347, 100–105.
[2] a) F. Lꢂnyi, J. Valyon, J. Engelhardt, F. Mizukami, J. Catal. 1996, 160,
13] a) X. Liang, J. Li, J. B. Joo, A. Gutiꢄrrez, A. Tillekaratne, I. Lee, Y. Yin, F.
Zaera, Angew. Chem. 2012, 124, 8158–8160; Angew. Chem. Int. Ed.
279–289; b) G. D. Yadav, J. J. Nair, Microporous Mesoporous Mater. 1999,
33, 1–48; c) L. Yang, J. B. Joo, N. Kim, K. Jung, J. Yi, Korean J. Chem. Eng.
2
2
012, 51, 8034–8036; b) I. Lee, J. B. Joo, Y. Yin, F. Zaera, Angew. Chem.
011, 123, 10390–10393; Angew. Chem. Int. Ed. 2011, 50, 10208–10211;
2010, 27, 1695–1699; d) L. Yang, J. B. Joo, Y. Kim, S. Oh, N. Kim, J. Yi,
Korean J. Chem. Eng. 2008, 25, 1014–1017; e) B. M. Reddy, M. K. Patil,
Chem. Rev. 2009, 109, 2185–2208; f) T. Yamaguchi, Appl. Catal. 1990, 61,
c) J. Yin, X. Qian, J. Yin, M. Shi, J. Zhang, G. Zhou, Inorg. Chem. Commun.
003, 6, 942–945.
14] a) J. B. Joo, Q. Zhang, I. Lee, M. Dahl, F. Zaera, Y. Yin, Adv. Funct. Mater.
012, 22, 166–174; b) J. B. Joo, Q. Zhang, M. Dahl, I. Lee, J. Goebl, F.
2
1
–25.
[
[
[
3] D. S. Park, B. K. Kwak, N. D. Kim, J. R. Park, J.-H. Cho, S. Oh, J. Yi, Chem-
CatChem 2012, 4, 836–843.
4] a) C. Morterra, G. Cerrato, F. Pinna, M. Signoretto, J. Catal. 1995, 157,
2
Zaera, Y. Yin, Energy Environ. Sci. 2012, 5, 6321–6327; c) J. B. Joo, I. Lee,
M. Dahl, G. D. Moon, F. Zaera, Y. Yin, Adv. Funct. Mater. 2013, DOI:
1
1
09–123; b) D. F a˘ rca s¸ iu, J. Q. Li, S. Cameron, Appl. Catal. A 1997, 154,
73–184; c) D. F a˘ rca s¸ iu, J. Q. Li, Appl. Catal. A 1998, 175, 1–9; d) K. Sar-
10.1002/adfm.201300255.
[
[
15] W. Stçber, A. Fink, E. Bohn, J. Colloid Interface Sci. 1968, 26, 62–69.
16] a) S. Shukla, S. Seal, R. Vij, S. Bandyopadhyay, J. Nanopart. Res. 2002, 4,
avanan, B. Tyagi, H. C. Bajaj, Catal. Sci. Technol. 2012, 2, 2512–2520.
[5] a) T. Y. Kim, D. S. Park, Y. Choi, J. Baek, J. R. Park, J. Yi, J. Mater. Chem.
553–559; b) S. Shukla, S. Seal, R. Vanfleet, J. Sol–Gel Sci. Technol. 2003,
2012, 22, 10021–10028; b) H. Yan, Y. Yang, D. Tong, X. Xiang, C. Hu,
2
7, 119–136.
Catal. Commun. 2009, 10, 1558–1563; c) Z. Li, R. Wnetrzak, W. Kwapin-
ski, J. J. Leahy, ACS Appl. Mater. Interfaces 2012, 4, 4499–4505; d) G. X.
Yu, X. L. Zhou, F. Liu, C. L. Li, L. F. Chen, J. A. Wang, Catal. Today 2009,
[
[
17] J. W. Lee, K. Hong, W. S. Kim, J. Kim, J. Ind. Eng. Chem. 2005, 11, 609–
14.
18] E. Rodrꢅguez-Castellꢂn, A. Jimꢄnez-Lꢂpez, P. Maireles-Torres, D. J. Jones,
J. Roziꢆre, M. Trombetta, G. Busca, M. Lenarda, L. Storaro, J. Solid State
Chem. 2003, 175, 159–169.
6
1
48, 70–74.
6] a) R. Srinivasan, T. Watkins, C. Hubbard, B. H. Davis, Chem. Mater. 1995,
, 725–730; b) F. Gao, Z. Xu, Q. Wang, Z. Hu, G. Gu, J. Dispersion Sci.
[
7
[19] G. D. Yadav, A. D. Murkute, J. Catal. 2004, 224, 218–223.
[20] X. Li, K. Nagaoka, R. Olindo, J. A. Lercher, J. Catal. 2006, 238, 39–45.
[21] a) X. Qi, M. Watanabe, T. M. Aida, R. L. Smith, Jr., Catal. Commun. 2009,
Technol. 2009, 30, 178–184; c) S. Tang, X. Huang, X. Chen, N. Zheng,
Adv. Funct. Mater. 2010, 20, 2442–2447; d) Y. Jiang, S. Yang, X. Ding, Y.
Guo, H. Bala, J. Zhao, K. Yu, Z. Wang, J. Mater. Chem. 2005, 15, 2041–
1
0, 1771–1775; b) J. R. Satam, R. V. Jayaram, Catal. Commun. 2008, 9,
2
1
046; e) Y. Wang, A. V. Biradar, T. Asefa, ChemSusChem 2012, 5, 132–
39.
1033–1039.
[
[
22] D. A. Ward, E. I. Ko, J. Catal. 1994, 150, 18–33.
23] a) K.-I. Shimizu, R. Uozumi, A. Satsuma, Catal. Commun. 2009, 10, 1849–
[
7] a) M. Hino, S. Kobayashi, K. Arata, J. Am. Chem. Soc. 1979, 101, 6439–
441; b) M. Hino, K. Arata, J. Chem. Soc. Chem. Commun. 1979, 0, 1148–
149; c) M. Hino, K. Arata, J. Chem. Soc. Chem. Commun. 1980, 0, 851–
52.
6
1
8
1853; b) A. S. Amarasekara, L. D. Williams, C. C. Ebede, Carbohydr. Res.
2
008, 343, 3021–3024; c) J. N. Chheda, Y. Roman-Leshkov, J. A. Dumesic,
Green Chem. 2007, 9, 342–350; d) Y. Su, H. M. Brown, X. Huang, X.-d.
Zhou, J. E. Amonette, Z. C. Zhang, Appl. Catal. A 2009, 361, 117–122.
24] B. Li, R. D. Gonzalez, Appl. Catal. A 1997, 165, 291–300.
[
[
8] X. Song, A. Sayari, Catal. Rev. 1996, 38, 329–412.
9] a) S. K. Das, M. K. Bhunia, A. K. Sinha, A. Bhaumik, J. Phys. Chem. C 2009,
[
113, 8918–8923; b) D. J. McIntosh, R. A. Kydd, Microporous Mesoporous
Mater. 2000, 37, 281–289; c) Y. Sun, L. Zhu, H. Lu, R. Wang, S. Lin, D.
Jiang, F.-S. Xiao, Appl. Catal. A 2002, 237, 21–31; d) H. Althues, S.
Kaskel, Langmuir 2002, 18, 7428–7435; e) M. A. Al-Daous, A. Stein,
Chem. Mater. 2003, 15, 2638–2645.
Received: May 3, 2013
Revised: July 4, 2013
Published online on && &&, 0000
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 9
&
8
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
J. B. Joo, A. Vu, Q. Zhang, M. Dahl, M. Gu,
F. Zaera, Y. Yin*
&& – &&
A Sulfated ZrO Hollow Nanostructure
2
as an Acid Catalyst in the Dehydration
of Fructose to 5-
Hydroxymethylfurfural
Core of the matter: Sulfated ZrO2
nificantly enhanced catalytic activity
toward the dehydration of fructose to
5-hydroxymethylfufural (see picture).
hollow shells with favorable mesoporos-
ity, excellent dispersity in liquid media,
and a strongly acidic surface show sig-
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 9
&
9
&
ÞÞ
These are not the final page numbers!