Facile Formation of Lactic Acid from a Triose Sugar in Water over Niobium Oxide with a Deformed Orthorhombic Phase

Article abstract of DOI:10.1021/acscatal.7b03003

Hydrothermal treatment of NH₄[NbO(C₂O₄)₂(H₂O)₂]·nH₂O in water at 448 K for 3 days produced crystalline Nb₂O₅ with a deformed orthorhombic structure and a high surface area (208 m² g^-1). Fourier-transform infrared spectroscopy measurements of pyridine adsorption revealed that the Nb₂O₅ catalyst has both high densities of Br?nsted and Lewis acid sites that can work in the presence of water. One feature of the Nb₂O₅ catalyst is its high density of water-compatible Lewis acid sites (0.21 mmol g^-1), which is much larger than that of Nb₂O₅·nH₂O (0.03 mmol g^-1). The Nb₂O₅ catalyst was studied as a solid acid catalyst for the formation of lactic acid from 1,3-dihydroxyacetone and pyruvaldehyde in water at 373 K, and was determined to be a highly active and selective catalyst, compared with typical acid catalysts (H₂SO₄, Sc(OTf)₃, and Nb₂O₅·nH₂O). A high Lewis acid density with moderate acid strength is a crucial factor for the high catalytic performance exhibited for the former reaction. High densities of both Br?nsted and Lewis acid sites in the catalyst promote the fast and selective production of lactic acid in the latter reaction. In addition, Such Lewis acidity of the Nb₂O₅ is also effective over conventional acid catalysts in xylose dehydration to furfural in water, with respect to reaction rate and furfural selectivity. The combination of aqueous-phase dehydration of xylose over the Nb₂O₅ and the continuous extraction of furfural with an immiscible organic solvent resulted in a high selectivity toward furfural of ～78.5% with high xylose conversion (97%).

Full text of DOI:10.1021/acscatal.7b03003

Subscriber access provided by READING UNIV
Article
Facile Formation of Lactic Acid from a Triose Sugar in Water
over Niobium Oxide with a Deformed Orthorhombic Phase
Kiyotaka Nakajima, Jun Hirata, Minjune Kim, Navneet Kumar Gupta, Toru
Murayama, Akihiro Yoshida, Norihito Hiyoshi, Atsushi Fukuoka, and Wataru Ueda
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b03003 • Publication Date (Web): 22 Nov 2017
Downloaded from http://pubs.acs.org on November 22, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
Facile Formation of Lactic Acid from a Triose Sugar
in Water over Niobium Oxide with a Deformed
Orthorhombic Phase
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†,‡,
†
†
†,η
ζ
Kiyotaka Nakajima, * Jun Hirata, Minjune Kim, Navneet Kumar Gupta, Toru Murayama,
§
ξ
†
§
Akihiro Yoshida, Norihito Hiyoshi, Atsushi Fukuoka and Wataru Ueda ,*
†
Institute for Catalysis, Hokkaido University, Kita 21 Nishi 10, Kitaꢀku, Sapporo 001ꢀ0021,
Japan
‡
Advanced Low Carbon Technology Research and Development (ALCA) program, Japan
Science and Technology (JST) Agency, 4ꢀ1ꢀ8 Honcho, Kawaguchi 332ꢀ0012, Japan
η
Central Institute of Mining and Fuel Research (CIMFR), Digwadih Campus, PO FRI 828108,
Dhanbad, Jharkhand, India
ζ
Research Center for Gold Chemistry, Tokyo Metropolitan University, 1ꢀ1ꢀ203, MinamiꢀOsawa,
Hachioji, 192ꢀ0397, Japan
§
Department of Material and Life Chemistry, Faculty of Engineering, Kanagawa University, 3ꢀ
2
7, Rokkakubashi, Kanagawaꢀku, Yokohama 221ꢀ8686, Japan
ACS Paragon Plus Environment
1

ACS Catalysis
Page 2 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ξ
Research Institute for Chemical Process Technology, National Institute of Advanced Industrial
Science and Technology (AIST), 4ꢀ2ꢀ1 Nigatake, Miyagino, Sendai 983ꢀ8551 Japan
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ABSTRACT Hydrothermal treatment of NH [NbO(C O ) (H O) ]·nH O in water at 448 K for
4
2
4 2
2
2
2
three days produced crystalline Nb O with a deformed orthorhombic structure and a high
2
5
2
ꢀ1
surface area (208 m g ). Fourierꢀtransform infrared spectroscopy measurements of pyridine
adsorption revealed that the Nb O catalyst has both high densities of Brønsted and Lewis acid
2
5
sites that can work in the presence of water. One feature of the Nb O catalyst is a high density
2
5
ꢀ1
of waterꢀcompatible Lewis acid sites (0.21 mmol g ), which is much larger than that of
ꢀ1
Nb O ·nH O (0.03 mmol g ). The Nb O catalyst was studied as a solid acid catalyst for the
2
5
2
2
5
formation of lactic acid from 1,3ꢀdihydroxyacetone and pyruvaldehyde in water at 373 K, and
was determined to be a highly active and selective catalyst compared with typical acid catalysts
(
H SO , Sc(OTf) , and Nb O ·nH O). A high Lewis acid density with moderate acid strength is a
2 4 3 2 5 2
crucial factor for the high catalytic performance exhibited for the former reaction. High densities
of both Brønsted and Lewis acid sites in the catalyst promote the fast and selective production of
lactic acid in the latter reaction. In addition, Such Lewis acidity of the Nb O is also effective
2
5
over conventional acid catalysts in xylose dehydration to furfural in water with respect to
reaction rate and furfural selectivity. The combination of aqueousꢀphase dehydration of xylose
over the Nb O and the continuous extraction of furfural with an immiscible organic solvent
2
5
resulted in a high selectivity toward furfural of approximately 78.5% with high xylose
conversion (97%).
ACS Paragon Plus Environment
2

Page 3 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
KEYWORDS: Niobium oxide, Waterꢀtolerant Lewis acid, Biomass conversion, isomerization,
Lactic acid
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
3

ACS Catalysis
Page 4 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
. INTRODUCTION
Chemical production using an easily separable and reusable solid acid catalyst has been of
th
increasing importance in the 21 century to realize a sustainable society. The use of conventional
homogeneous liquid acids requires energyꢀinefficient or environmentally hazardous workꢀup
processes, such as distillation and neutralization, for separation of the catalyst from the reaction
mixture. Zeolites were recognized as effective solid acid catalysts for a variety of reactions as
replacements for liquid Brønsted acids, and have been used practically for the reforming of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1ꢀ3
hydrocarbons derived from fossil fuel resources. Amorphous niobium oxide hydrate, also
referred to as niobic acid (Nb O ·nH O), has been widely studied as an insoluble solid acid
2
5
2
4
ꢀ6
catalyst for various waterꢀmediated reactions, such as esterification, hydrolysis, and hydration.
The unique catalysis of Nb O ·nH O is attributed to strong Brønsted acidity, which is
2
5
2
4
,6
comparable to 70% sulfuric acid solution. This unique functionality of Nb O ·nH O has led to
2
5
2
versatile applications as a solid Brønsted acid in the chemical industry.
Unsaturated coordination sites on some metal oxides including zeolite and Nb O ·nH O were
2
5
2
recently reported to function as waterꢀtolerant Lewis acid sites in water or in water/organic
7
ꢀ15
solvent mixtures for the valorization of biomassꢀderived carbohydrates.
A Snꢀcontaining
zeolite with betaꢀtype topology (SnꢀBEA) possesses SnO tetrahedra as Lewis acid sites that
4
16,17
selectively catalyze the isomerization of glucose to fructose in water.
A combination of Snꢀ
BEA zeolite with a liquid Brønsted acid enabled a oneꢀpot process for the conversion of glucose
to 5ꢀ(hydroxymethyl)furfural (HMF), an intermediate for the production of various biomassꢀ
7
derived plastics. Methyl lactate formation from triose sugars in methanol is an excellent
example of utilizing the Lewis acidity of the SnꢀBEA zeolite for selective biomass conversion;
SnO tetrahedra alone catalyze the dehydration and subsequent isomerization of triose sugars in
4
ACS Paragon Plus Environment
4

Page 5 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
18ꢀ25
methanol to produce methyl lactate.
Incorporation of carbon nanoparticles bearing Brønsted
acidic COOH groups into the mesopores of mesoporous Snꢀcontaining silica produces a
composite catalyst that can convert hexose and triose sugars into lactic acid in water, and alkyl
2
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
lactates in corresponding alcohols. The composite catalyst with both Brønsted and Lewis
acidity is clearly superior to SnꢀBEA with Lewis acidity in terms of reaction efficiency. Despite
the high catalytic activity and durability of the composite catalyst in the formation of alkyl
lactates, a continuous decrease in activity was observed during lactic acid formation in water. A
24
similar deactivation phenomenon was also reported for the same reaction with SnꢀBEA. This
deactivation can be explained simply by the leaching of the active Sn species in an acidic
reaction mixture; an increase in the lactic acid concentration is accompanied by a decrease in the
24
pH of the reaction solution. Heat treatment of heteroatomꢀcontaining zeolites in an acidic
solution generally results in removal of the metal species from the zeolite framework; therefore,
these Snꢀbased catalysts lose their original activities during the course of lactic acid formation.
The development of a stable and highly active solid acid catalyst for the selective production of
26
lactic acid directly from triose sugars is thus a significant subject in this field.
TiO and NbO tetrahedra on TiO and Nb O surfaces have already been demonstrated to
4
4
2
2
5
exhibit waterꢀcompatible Lewis acidity and catalyze the isomerization of Pyruvaldehyde (PVA)
8,10
to lactic acid.
The combination of Brønsted and Lewis acidity in Nb O ·nH O is potentially
2 5 2
useful for the formation of lactic acid from triose sugars because dehydration of triose sugars to
2
2
PVA catalyzed by a Brønsted acid contributes largely to an increase in the overall reaction rate.
This insight prompted us to study acid catalysis over niobium oxide for the selective production
of lactic acid from 1,3ꢀdihydroxyacetone (DHA), a triose sugar, in water. Here, we focused our
attention on crystalline Nb O with a deformed orthorhombic structure (orthoꢀNb O ) prepared
2
5
2
5
ACS Paragon Plus Environment
5

ACS Catalysis
Page 6 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
27
by hydrothermal treatment of ammonium niobium oxalate in water at 448 K for 3 days. Orthoꢀ
Nb O consists of NbO octahedra and Nb O pentagonal units forming a layered structure.
2
5
6
6
21
While orthoꢀNb O exhibited higher catalytic performance for the FriedelꢀCrafts alkylation of
2
5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
anisole with benzyl alcohol than amorphous Nb O ·nH O, Brønsted and Lewis acid catalysis in
2
5
2
water has not been studied to date. In this study, the Brønsted and Lewis acidity of orthoꢀNb O5
2
in the presence or absence of physisorbed water was studied systematically using Fourierꢀ
transform infrared (FTIR) spectroscopy with pyridine as a basic probe molecule. Acid catalysis
of orthoꢀNb O was mainly examined with respect to two aqueousꢀphase reactions, the
2
5
conversion of DHA and PVA to lactic acid in water (Scheme 1). Owing to poor reusability of
Snꢀcontaining silicates toward lactic acid formation in water, we used homogeneous Brønsted
acid (H SO ) and Lewis acid (Scandium trifluoromethanesulfonate) as reference catalysts in this
2
4
study. In addition, orthoꢀNb O was applied to furfural formation from xylose in water and
2
5
toluene/water mixture because water tolerant Lewis acidity was reported to be effective for this
2
8
reaction.
Isomerization
O
OH
Dehydration
O
O
OH
OH
OH
O
HO
OH
O
Scheme 1. Catalytic conversion of DHA into lactic acid in water.
ACS Paragon Plus Environment
6

Page 7 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
2
. EXPERIMENTAL SECTION
2
.1 Preparation of ortho-Nb O . OrthoꢀNb O was prepared by a conventional hydrothermal
2 5 2 5
treatment of an aqueous ammonium niobium oxalate (NH (NbO(C O ) (H O) ·nH O;
4
2
4 2
2
2
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
27
Companhia Brasileira de Metalurgia
e
Mineração (CBMM), Brazil) solution.
NH (NbO(C O ) (H O) ·nH O powder containing 6 mmol of niobium atoms was completely
4
2
4 2
2
2
2
dissolved in 40 mL of deionized water. The solution was transferred to a Teflonꢀlined stainlessꢀ
steel autoclave (60 mL) and then heated at 448 K for three days. The resulting precipitate was
isolated by filtration, washed repeatedly with deionized water (300ꢀ500 mL), and dried at 353 K
overnight. After calcination of the dry powder at 673 K for 4 h, orthoꢀNb O was obtained with
2
5
the yield of 90ꢀ95%.
2
.2 Characterization. Powder Xꢀray diffraction (XRD) pattern was obtained with a
diffractometer (Ultima IV, Rigaku) using Cu Kα radiation (40 kV, 40 mA) over the 2θ range of
5−80°. Nitrogen adsorptionꢀdesorption isotherm was measured at 77 K with a surface area
analyzer (Novaꢀ4200e, Quantachrome). Prior to measurement, the sample was heated at 473 K
for 1 h under vacuum to remove physisorbed water. The BrunauerꢀEmmettꢀTeller (BET) surface
area was estimated over a relative pressure (P/P ) range of 0.05−0.30. A pore size distribution
0
curve was obtained from the adsorption branch of the isotherm using the BarrettꢀJoynerꢀHalenda
(BJH) method. Transmission electron microscopy (TEM) and aberrationꢀcorrected scanning
transmission electron microscopy (STEM) observations were made with an electron microscope
(JEMꢀARM200F, JEOL Ltd, Japan) at an acceleration voltage of 200 kV. FTIR spectra were
ꢀ1
obtained at a resolution of 4 cm using a spectrometer (FT/IRꢀ6100, Jasco) equipped with an
extended KBr beam splitting device and a mercury cadmium telluride (MCT) detector. A total of
64 scans were averaged for each spectrum. The samples were pressed into selfꢀsupporting disks
ACS Paragon Plus Environment
7

ACS Catalysis
Page 8 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
20 mm diameter, ca. 20 mg) and placed in an IR cell attached to a closed circulation system.
The disks were dehydrated at 473 K for 1 h under vacuum to remove physisorbed water. The
dehydrated disks were then exposed to saturated water vapor (ca. 20 Torr) at room temperature
for 1 h, followed by evacuation for 5 min to remove weakly adsorbed water, which produces
hydrated samples. Pyridine was adsorbed as a basic probe molecule on the dehydrated or
hydrated sample disks in the IR cell at room temperature. Each IR spectrum was measured after
adsorbed pyridine and pyridine in the gas phase reached equilibrium. IR spectra of the samples
before pyridine adsorption were used as backgrounds for difference spectra by subtraction from
the spectra measured for the pyridineꢀadsorbed samples.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2.3 Catalytic reactions. The Lewis acid catalysis of orthoꢀNb O was examined by the
2
5
formation of lactic acid from DHA and PVA in water. In a typical reaction, a mixed solution of
DHA/PVA (10 mL, 0.1 M) and the catalyst (0.1 g) was heated in a sealed Pyrex glass tube at 373
K where reaction temperature was controlled by an oil bass. The solutions after reaction were
analyzed using highꢀperformance liquid chromatography (HPLC; LCꢀ2000 plus, Jasco) with
refractive index (RI) and photodiode array (PDA) detectors. The HPLC apparatus was equipped
2
with an Aminex HPXꢀ87H column (BioꢀRad Laboratories) and used at 308 K with 5 mM H SO4
ꢀ1
as an eluent at a flow rate of 0.5 mL min .
OrthoꢀNb O was also tested by furfural formation from xylose in water and toluene/water
2
5
2
8
mixture. An aqueousꢀphase reaction was performed in a sealed Pyrex glass tube using 20 mg of
solid catalyst, 0.5 mmol of xylose, and 5 mL of water at 393 K where reaction temperature was
controlled by an oil bass. After the reactions, the resultant solutions were diluted 10ꢀ30 times
with water and then analyzed with HPLC. The rate for furfural formation was estimated from the
slope of the linear time course line. Toluene was used as an extraction solvent for furfural
ACS Paragon Plus Environment
8

Page 9 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
produced in the reaction mixture in order to suppress the formation of byꢀproducts. Typically, the
mixture of catalyst (100 mg), xylose (0.5 mmol), water (2 mL), and toluene (3 mL) was stirred in
a sealed Pyrex glass tube at 393 K for 3 h where reaction temperature was controlled by an oil
bass. The resulting water and organic phases were analyzed separately with HPLC and gas
chromatography (GC; GCꢀ2025, Shimadzu) with a capillary column (DBꢀFFAP, GL Sciences)
and chlorobenzene as an external standard.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
. RESULTS AND DISCUSSION
Figure 1. (A) XRD pattern and proposed crystalline structure of orthoꢀNb O (B) in the aꢀb
2
5
plane and (C) along the cꢀaxis.
ACS Paragon Plus Environment
9

ACS Catalysis
Page 10 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. (A) TEM and (B) aberrationꢀcorrected STEM images of orthoꢀNb O .
2
5
Figure 1(A) shows an XRD pattern of orthoꢀNb O . It was reported that hydrothermal
2
5
treatment of an aqueous ammonium niobium oxalate solution results in the formation of a
crystalline Nb O that has a deformed orthorhombic structure in the aꢀb plane (Figure 1(B)) and
2
5
27
a layered structure along the c axis (Figure 1(C)). The sharp and intense diffractions at 22.7°
and 45.2° in Figure 1(A) are assignable to (001) and (002), respectively, and are derived from the
layered structure along the c axis with an interval distance of 0.39 nm. Although the deformed
orthorhombic structure consisting of niobium octahedra (NbO ) and Nb O pentagonal units
6
6
21
cannot be confirmed by XRD measurement, previous studies that used Raman spectroscopy, N2
adsorption measurements, and TEM analysis revealed that orthoꢀNb O closely resembles
2
5
2
9,30
hydrothermally synthesized Mo VO in the crystalline structure of the aꢀb plane.
The local
3
x
structure of orthoꢀNb O can be evaluated in detail by TEM and aberrationꢀcorrected STEM
2
5
analyses. Figure 2A shows a typical TEM image of orthoꢀNb O , where many rodꢀlike particles
2
5
with an average diameter of 5ꢀ10 nm are evident. Fringes perpendicular to the rod represent a
ACS Paragon Plus Environment
10

Page 11 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
layered structure, and the average distance between fringes is estimated to be 0.40 nm, which is
consistent with the value calculated from the XRD pattern. In addition, most of the rod particles
appear to have channels running along the c axis, the average interlayer distance of which is 0.62
nm. In agreement with the previous report on Mo VO , such micropores are formed by the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
x
assembly of pentagonal (Nb O ) units in the same manner (Figures 1(B) and 1(C)). The
6
21
arrangement of pentagonal units at the crossꢀsection can be observed clearly in the STEM image
Figure 2B). The pentagonal units are connected to each other by octahedral NbO linkers, which
(
6
results in the formation of 7ꢀmembered ring micropores with an average diameter of 0.6 nm. No
diffractions derived from an ordered structure of pentagonal units were detected in XRD profile
because the crystal domains were too small. Another important feature of orthoꢀNb O is that the
2
5
2
ꢀ1
2
ꢀ1
BET surface area (208 m g ) is larger than that of amorphous Nb O ·nH O (171 m g ). In
2
5
2
addition to original micropores, aggregation of the rodꢀshape particles at the nanometer scale
produces interparticle mesoꢀscale voids, which contribute largely to the increase in the effective
surface area (Figure S1).
Total number of acid site has been already estimated by temperatureꢀprogramed desorption of
ꢀ1 27
NH (NH ꢀTPD) to be 0.15 mmol g . However, high temperature treatment in He and
3
3
subsequent adsorption of NH at 373 K tend to reduce total number of original acid site, due to
3
thermal condensation of oxide network and elimination of NH molecule adsorbing on weakly
3
acidic site as well as physisorbed NH . In addition, a conventional NH ꢀTPD measurement
3
3
cannot estimate the amounts of Brønsted acid and Lewis acid separately. Here, the Brønsted and
31
Lewis acidity of orthoꢀNb O was evaluated by pyridine adsorption measurements with FTIR.
2
5
It can be expected that orthoꢀNb O could possess a high density of Lewis acid sites, due to the
2
5
large surface area and defectꢀrich crystalline structure. Figure 3 shows difference FTIR spectra of
ACS Paragon Plus Environment
11

ACS Catalysis
Page 12 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
pyridineꢀadsorbed orthoꢀNb O where the sample was first heated at 473 K for 1 h under vacuum
2
5
to remove physisorbed water and then exposed to gaseous pyridine with various vapor pressures.
ꢀ1
Specific bands assignable to pyridinium ions formed on Brønsted acid sites (1540 cm ),
ꢀ1
ꢀ1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
coordinated pyridine on Lewis acid sites (1445 cm ), and physisorbed pyridine (1440 cm ) were
all observed in Figures 3(a)ꢀ3(f), which confirms that orthoꢀNb O has both Brønsted and Lewis
2
5
+
acid sites. Brønsted acid sites can be formed after the removal of NH species by calcination of
4
27
asꢀsynthesized orthoꢀNb O . As for Nb O ·nH O, the Lewis acid sites of orthoꢀNb O can also
2
5
2
5
2
2
5
8
be regarded as NbO tetrahedra. The band intensities of Brønsted and Lewis acid sites
4
continuously increased with the pyridine pressure up to 4.0 Torr (Figures 3(a)ꢀ3(f)). These bands
remained intact, even after evacuation of the pyridineꢀsaturated sample at room temperature for 4
h (Figure 3(g)), while the band of physisorbed pyridine disappeared. The amounts of Brønsted
and Lewis acid sites for dehydrated orthoꢀNb O can be estimated from their band areas and
2
5
ꢀ2
ꢀ2
molecular absorption coefficients (Brønsted acid, 1.20 ꢀmol·cm ; Lewis acid, 3.15 ꢀmol·cm )
ꢀ1
ꢀ1
8
to be 0.14 ± 0.01 mmol g and 0.21 ± 0.01 mmol g , respectively. Error margins were
estimated by the standard deviation of three replicates using dehydrated orthoꢀNb O samples
2
5
ꢀ1
with different batches. Total number of Brønsted and Lewis acid sites (0.35 mmol g ) is clearly
ꢀ1 27
larger than that estimated by NH ꢀTPD (0.15 mmol g ). This can be explained mainly by room
3
temperature adsorption and desorption of pyridine molecule, which enables accurate estimation
of strong, medium, and weak acid sites.
ACS Paragon Plus Environment
12

Page 13 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
445 cm-1
1
540 cm-1
O
N
O
N
Nb
Physisorbed
pyridine
1
0.2
H+
O
O
440 cm-1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
g
f
e
d
c
b
a
1
600
1500
1400
Wavenumber / cm-
1
Figure 3. Difference FTIR spectra for pyridineꢀadsorbed dehydrated orthoꢀNb O . Gasꢀphase
2
5
pyridine pressure: (a) 0.06, (b) 0.15, (c) 0.5, (d) 1.0, (e) 2.0, and (f) 4.0 Torr. Spectrum (g) was
obtained by evacuation at room temperature for 4 h after (f).
ACS Paragon Plus Environment
13

ACS Catalysis
Page 14 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
445 cm-1
1540 cm-1
O
N
O
Nb
Physisorbed
pyridine
N
O
O
1440 cm-1
H+
0
.2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c
b
a
1
700
1600
Wavenumber / cm
1500
1400
-1
Figure 4. FTIR spectra for pyridineꢀadsorbed hydrated orthoꢀNb O : (a) sample saturated with
2
5
physisorbed water, (b) sample obtained by the introduction of pyridine (5.0 Torr) after (a), and
(
c) sample obtained by evacuation at room temperature for 4 h after (b).
The pyridine adsorption technique was also applied to hydrated orthoꢀNb O5 for
2
characterization of the water compatibility of these acid sites (Figure 4). When the sample was
exposed to saturated water vapor at room temperature, an intense and broad band due to the
ꢀ1
bending mode of physisorbed water was observed at 1620 cm (Figure 4(a)). After the
introduction of large amounts of pyridine vapor (5.0 Torr) to the hydrated sample, two bands for
ꢀ
1
ꢀ1
adsorbed pyridine were obtained at 1540 cm and 1443 cm (Figure 4(b)), in addition to the
ꢀ1
band for the OH bending mode of physisorbed water (1620 cm ). The two former bands are
ACS Paragon Plus Environment
14

Page 15 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
assignable to pyridinium ions formed on Brønsted acid sites, and a combined band for
ꢀ1
ꢀ1
coordinated pyridine on Lewis acid sites (1445 cm ) and physisorbed pyridine (1440 cm ). Two
ꢀ1
ꢀ1
bands at 1540 cm (pyridinium ion) and 1445 cm (coordinated pyridine) remained intact after
evacuation of the pyridineꢀsaturated hydrated sample at room temperature for 4 h, while the
bands for physisorbed water and physisorbed pyridine almost disappeared. Brønsted and Lewis
acid sites covered with large amounts of physisorbed water can thus interact with pyridine
molecules and successfully form corresponding adducts, which suggests that these acid sites are
able to work in water. The amounts of waterꢀtolerant Brønsted and Lewis acid sites were
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ1
ꢀ1
estimated to be 0.09 ± 0.01 mmol g and 0.20 ± 0.01 mmol g , respectively. Error margins
estimated here were also based on the standard deviation of three replicates using hydrated
orthoꢀNb O samples with different batches. Table 1 summarizes the surface areas and acid site
2
5
densities of orthoꢀNb O and Nb O ·nH O. The pyridine adsorption measurements revealed that
2
5
2
5
2
8
both samples have waterꢀtolerant Brønsted and Lewis acid sites. A unique feature of orthoꢀ
Nb O is a high density of waterꢀtolerant Lewis acid sites: the density of orthoꢀNb O is
2
5
2
5
8
approximately seven times that of Nb O ·nH O. This large difference may be attributed to the
2
5
2
strong Lewis acidity of Nb O ·nH O, which induces partial deactivation to form stable Lewis
2
5
2
acidꢀwater adducts. As a result, approximately 80% of the Lewis acid sites on Nb O ·nH O are
2
5
2
not available as catalytically active sites. In contrast, almost all the Lewis acid sites on orthoꢀ
Nb O retain their Lewis acidity, even in a large amount of physisorbed water, which suggests
2
5
that the Lewis acidity of orthoꢀNb O is relatively weak compared with that of Nb O ·nH O.
2
5
2
5
2
ACS Paragon Plus Environment
15

ACS Catalysis
Page 16 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Structural properties of orthoꢀNb O and Nb O ·nH O.
2
5
2
5
2
a
ꢀ1
SBET
Acid site density (mmol g )
Catalyst
Crystalline structure
2
ꢀ1
(
m g ) Brønsted acid
Lewis acid
0.20 ± 0.01
(0.21 ± 0.01)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0.09 ± 0.01
OrthoꢀNb O
Deformed Orthorhombic 208
171
2
5
b
b
(0.14 ± 0.01)
b
b
Nb O ·nH O Amorphous
0.14 (0.17)
0.03 (0.15)
2
5
2
a
The density for waterꢀtolerant acid sites was estimated by pyridineꢀadsorption measurements
b
using FTIR in reference 8. Values in parentheses denote the density for dehydrated samples.
1
00
80
60
4
0
0
0
2
0
50
100
Reaction time / min
Figure 5. Time courses for lactic acid formation from PVA over various acid catalysts at 373 K.
Open circles, orthoꢀNb O ; closed triangles, Nb O ·nH O; open squares, Sc(OTf) ; closed circles,
2
5
2
5
2
3
H SO ; open triangles, no catalyst. Error bars represent the standard deviation of three replicates.
2
4
OrthoꢀNb O was used as a catalyst for the conversion of PVA to lactic acid at 373 K. This
2
5
reaction is one of the isomerization reactions catalyzed by a Lewis acid through the Meerweinꢀ
3
2
PonndorfꢀVerley (MPV) reduction mechanism. Density functional theory calculation indicated
ACS Paragon Plus Environment
16

Page 17 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
that PVA in the monoꢀhydrated form (ꢀ7.2 eV) has higher HOMO energy than H O (ꢀ8.8 eV),
2
3
3
which suggests that waterꢀtolerant Lewis acids react with PVA in preference to H O. Here,
2
3
4,35
sulfuric acid (H SO ) and scandium trifluoromethanesulfonate (Sc(OTf) )
were used as
2
4
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
homogeneous Brønsted and Lewis acid catalysts, and Nb O ·nH O was also used for comparison.
2
5
2
Figure 5 shows time courses for lactic acid formation from PVA in water over orthoꢀNb O and
2
5
the reference catalysts. Although H SO is conventionally used as an active catalyst for various
2
4
acidꢀcatalyzed reactions, lactic acid was not formed at all within a reaction time of 2 h. The same
result was obtained in a blank experiment, which implies that the Brønsted acid does not function
as an acid catalyst for this reaction. In contrast, catalysts with waterꢀtolerant Lewis acid sites
(orthoꢀNb O , Nb O ·nH O, and Sc(OTf) ) are able to produce lactic acid effectively; lactic acid
2
5
2
5
2
3
yields of these three catalysts continuously increased as a function of reaction time. OrthoꢀNb O5
2
exhibited the highest catalytic activity among these catalysts with a lactic acid yield that reached
approximately 90% at 120 min, which was twice those of Sc(OTf) and Nb O ·nH O. The
3
2
5
2
combination of homogeneous Brønsted acid (H SO , 50 mg) and Lewis acid (Sc(OTf) , 50 mg)
2
4
3
gave poor activity (PVA conversion, 14%; Lactic acid yield, 5%) at 120 min under the same
reaction conditions. Such catalysis is probably due to the decomposition of Lewis acidic
Sc(OTf) with H SO . Table 2 summarizes the rates for lactic acid formation estimated for PVA
3
2
4
ꢀ1
ꢀ1
conversions of 20ꢀ30%. OrthoꢀNb O had a larger reaction rate (0.15 mmol g min ) than
2
5
ꢀ1
ꢀ1
ꢀ1
ꢀ1
Sc(OTf) (0.08 mmol g min ) and Nb O ·nH O (0.09 mmol g min ). Sc(OTf) is completely
3
2
5
2
3
dissolved in water, and all the Lewis acid centers can be easily accessible to reactant molecules.
Therefore, Lewis acid catalysis over these niobium catalysts should be much more effective than
that of Sc(OTf) for the isomerization of PVA in water. The high density of Lewis acid sites on
3
orthoꢀNb O resulted in higher catalytic activity for the reaction than Nb O ·nH O. The
2
5
2
5
2
ACS Paragon Plus Environment
17

ACS Catalysis
Page 18 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
selectivity of Nb O ·nH O (63.8%) was lower than that of orthoꢀNb O (98.4%), which may be
2
5
2
2
5
explained by the acid strength of the Lewis acid, as shown in Tables 1 and 2. The strong Lewis
acid sites on Nb O ·nH O promote the isomerization reaction in parallel with side reactions for
2
5
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
byꢀproduct formation, which results in low selectivity toward lactic acid. Moderate strength
Lewis acid sites in high density on orthoꢀNb O are suitable for the selective production of lactic
2
5
acid from PVA through the isomerization reaction.
Table 2. Acid site densities and catalytic activities for the isomerization of PVA into lactic acid
a
in water.
ꢀ
1
d
e
f
Acid site density (mmol g )
Time Conv.
(min) (%)
Yield Selec.
g
Rate (mmol
Catalyst
ꢀ1
ꢀ1
g min )
Brønsted acid Lewis acid
(%) (%)
b
b
OrthoꢀNb O 0.09 ± 0.01
0.20 ± 0.01
20
31± 2
28± 2
20± 3
0.0
31± 2 98± 1 0.15
18± 2 64± 2 0.09
16± 3 80± 2 0.08
2
5
c
c
Nb O ·nH O 0.14
0.03
20
2
5
2
Sc(OTf)3
H SO
ꢀ
2.0
20
22.4
ꢀ
ꢀ
ꢀ
120
120
0.0
0.0
ꢀ
ꢀ
ꢀ
ꢀ
2
4
No catalyst
0.0
a
b
Reaction conditions: 10 mL aqueous PVA solution (0.1 M), 100 mg catalyst, T=373 K. The
density of waterꢀtolerant acid site was estimated by pyridineꢀadsorption measurements with
c
d
e
FTIR. The density of waterꢀtolerant acid sites reported in reference 8. PVA conversion. Lactic
f
g
acid yield. Lactic acid selectivity. Rate for lactic acid formation. Error margins of all
conversion, yields, and selectivities were calculated from the standard deviation of three
replicates.
Next, orthoꢀNb O was applied to a oneꢀpot synthesis of lactic acid from DHA, a triose sugar,
2
5
in water at 373 K. Figure 6 shows the time courses for DHA conversion and the product yields
ACS Paragon Plus Environment
18

Page 19 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
over orthoꢀNb O . In this profile, DHA conversion and PVA yield increased as a function of the
2
5
reaction time during the initial stage. The PVA yield reached a maximum at approximately 30
min and then started to decrease with an increase in the lactic acid yield. This is a consecutive
reaction that consists of the dehydration of DHA to PVA and subsequent isomerization to lactic
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
31
acid. The lactic acid yield continuously increased with time and finally approached
approximately 80% at 180 min. The catalytic activity of orthoꢀNb O was systematically
2
5
compared with those of the reference catalysts at 120 min (Table 3).
100
DHA
80
Lactic acid
60
40
PVA
20
0
0
50
100
150
Reaction time / min
Figure 6. Time course for DHA conversion and product yields over orthoꢀNb O at 373 K. Error
2
5
bars represent the standard deviation of three replicates.
ACS Paragon Plus Environment
19

ACS Catalysis
Page 20 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 3. Acid site densities and catalytic activities for the conversion of DHA into lactic acid in
a
water.
ꢀ
1
d
Acid site density (mmol g )
Product yield (%)
Conversion
%)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Catalyst
(
Brønsted acid Lewis acid
Lactic acid
56 ± 2
27 ± 2
20 ± 2
0.0
PVA
Others
18 ± 2 17 ± 2
50 ± 2 19 ± 2
14 ± 2 26 ± 2
44 ± 2 0
b
b
OrthoꢀNb O 0.09 ± 0.01
0.20 ± 0.01 91 ± 3
2
5
c
c
Nb O ·nH O 0.14
0.03
2.0
ꢀ
96 ± 3
60 ± 2
44 ± 2
2
5
2
Sc(OTf)3
ꢀ
H SO
2
22.4
4
a
Reaction conditions: 10 mL aqueous DHA solution (0.1 M), 100 mg catalyst, T=373 K, t=120
b
min. The density of waterꢀtolerant acid site was estimated by pyridineꢀadsorption measurements
c
d
with FTIR. The density of waterꢀtolerant acid sites reported in reference 8. DHA conversion.
Error margins of all conversion, yields, and selectivities were calculated from the standard
deviation of three replicates.
H SO quantitatively converted DHA into PVA as a sole product by dehydration to give a 44%
2
4
yield of PVA at 120 min (Figure S2). Sc(OTf) can produce both PVA and lactic acid
3
simultaneously under the reaction conditions (Figure S3). The difference between H SO and
2
4
Sc(OTf) is clearly derived from Lewis acid catalysis, whereby the isomerization of the evolved
3
PVA is promoted. While a Lewis acid alone catalyzed both dehydration and isomerization,
Sc(OTf) formed large amounts of byꢀproducts (26% yield) at 60% DHA conversion. Lewis acid
3
catalysts are generally effective for the activation of carbonyl groups and induce CꢀC bond
forming reactions, such as aldolꢀtype condensation. DHA and PVA would be involved in the
formation of waterꢀsoluble condensation products over Sc(OTf) , whereas they are not detected
3
by HPLC measurement. Both niobium catalysts exhibited larger DHA conversion (>90%) than
Sc(OTf) and H SO (Figures 6 and S4) because the Brønsted and Lewis acid sites on niobium
3
2
4
ACS Paragon Plus Environment
20

Page 21 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
oxide catalysts function individually as active sites for the dehydration of PVA. Although these
niobium catalysts also produced certain amounts of byꢀproducts, orthoꢀNb O is significantly
2
5
superior to Nb O ·nH O and Sc(OTf) with respect to lactic acid yield. This can again be
2
5
2
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
explained solely by Lewis acid catalysis over orthoꢀNb O where PVA is selectively converted
2
5
to lactic acid, as shown in Table 2.
A reuse experiment was conducted with orthoꢀNb O for the formation of lactic acid from
2
5
DHA in water. Table S1 shows DHA conversion and lactic acid yield over fresh and reused
orthoꢀNb O at 373 K for 180 min. After the reaction, the catalyst was isolated from reaction
2
5
mixture by suction filtration, washed with distilled water (100 mL), and dried completely at 353
K in an oven. The recovered catalyst was used for the subsequent reaction under the same
reaction conditions. There was no obvious decrease in activity, even after the catalyst was reused
four times, which indicates that orthoꢀNb O is a highly active and reusable catalyst for the direct
2
5
formation of lactic acid from DHA in water.
Finally, orthoꢀNb O was applied to aqueous phase conversion of xylose to furfural. We
2
5
reported that Nb O ·nH O with waterꢀtolerant Lewis acid facilitates furfural formation from
2
5
2
2
8
xylose in water through stepꢀwise dehydration mechanism. Furfural is one of the important
furans synthesized from a biomassꢀderived sugar and can be used as a raw material for the
production of industrially important chemicals including tetrahydrofuran, furfuryl alcohol, and
3
6
1
,5ꢀpentanediol. We successfully identified Nb O ·nH O as a more active and selective catalyst
2 5 2
for xylose dehydration than typical homogeneous Brønsted and Lewis acid catalysts including
28
HCl and Sc(OTf) in terms of reaction rate and furfural selectivity. Therefore, catalytic activity
3
of orthoꢀNb O was simply compared with that of Nb O ·nH O in this study.
2
5
2
5
2
ACS Paragon Plus Environment
21

ACS Catalysis
Page 22 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 4. Acid site densities and catalytic activities of orthoꢀNb O and Nb O ·nH O for xylose
2
5
2
5
2
dehydration in water and in toluene/water mixture.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
b
Aqueous reaction
Biphasic reaction
e
f
g
Catalyst
Brønsted acid
Lewis acid
Rate
Conversion
(%)
Yield
(%)
ꢀ1
ꢀ1
ꢀ1 ꢀ1
(
mmol g )
(mmol g )
(mmol gꢀcat h )
c
c
OrthoꢀNb O
0.09 ± 0.01
0.20 ± 0.01
5.1
2.1
97.0
>99
78.5
72
2
5
d
d
Nb O ·nH O
0.14
0.03
2
5
2
a
Reaction conditions: 75 mg xylose (0.5 mmol), 5 mL water, 20 mg catalyst, T = 393 K.
b
Reaction conditions: 75 mg xylose (0.5 mmol), 2 mL water, 3 mL toluene, 100 mg catalyst, T =
c
3
93 K, t = 3 h. The density of waterꢀtolerant acid site was estimated by pyridineꢀadsorption
d
measurements with FTIR. The density of waterꢀtolerant acid sites reported in reference 8.
e
f
Reaction rates for furfural formation in water estimated from Figure S5. Xylose conversion.
g
furfural yield
Table 4 summarizes acid site densities and catalytic activities of orthoꢀNb O and Nb O ·nH O.
2
5
2
5
2
When xylose dehydration was conducted in water (Figure S5), the rates of orthoꢀNb O and
2
5
ꢀ1
ꢀ1
ꢀ1
ꢀ1
Nb O ·nH O were estimated to be 5.1 mmol gꢀcat min and 2.1 mmol gꢀcat min ,
2
5
2
ꢀ1
ꢀ1
respectively. These rates are much larger than those of Sc(OTf) (0.54 mmol gꢀcat min ) and
3
ꢀ1
ꢀ1
28
HCl (0.18 mmol gꢀcat min ) under the same reaction conditions. Therefore, acid catalysis of
niobium oxide catalysts is more profitable than typical homogeneous Brønsted and Lewis acid
2
8
catalysts for this reaction. Brønsted acid of Nb O ·nH O cannot catalyze the reaction, which
2
5
2
indicates that the high activity of orthoꢀNb O is attributed to the high density of water tolerant
2
5
Lewis acid site. Next, inꢀsitu extraction of furfural from aqueous solution into toluene, a waterꢀ
ACS Paragon Plus Environment
22

Page 23 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
28
immiscible solvent, was used to avoid catalytic degradation and polymerization of the products.
There was no large difference in furfural selectivity between orthoꢀNb O (78.5%) and
2
5
Nb O ·nH O (72%) with similar conversion in the biphasic reaction system, which means that
2
5
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Lewis acid catalysis of orthoꢀNb O toward xylose dehydration is similar to that of Nb O ·nH O.
2
5
2
5
2
With respect to reaction rate and furfural selectivity, orthoꢀNb O with a large amount of waterꢀ
2
5
tolerant Lewis acid site is clearly advantageous over Nb O ·nH O and typical homogeneous
2
5
2
acids for the conversion of xylose to furfural in a biphasic reaction system.
CONCLUSIONS
OrthoꢀNb O was demonstrated as an effective catalyst for the oneꢀpot formation of lactic acid
2
5
from DHA in water. The moderate acidity of the Lewis acid sites on orthoꢀNb O means they
2
5
can keep their original acidity, even in the presence of water, although most of the Lewis acid
sites (ca. 80%) on Nb O ·nH O were deactivated by the formation of stable water adducts. A
2
5
2
high density of waterꢀtolerant Lewis acid sites can be identified as the decisive factor to promote
the isomerization of PVA, the rateꢀdetermining step for the conversion of DHA to lactic acid.
Such Lewis acidity was also useful for fast and selective conversion of xylose in water to furfural.
The high activity and durability of orthoꢀNb O is beneficial for aqueousꢀphase conversion of
2
5
hydrophilic organic compounds including sugars to valueꢀadded raw materials.
ASSOCIATED CONTENT
ACS Paragon Plus Environment
23

ACS Catalysis
Page 24 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Supporting Information. N adsorptionꢀdesorption isotherm and pore size distribution of orthoꢀ
2
Nb O ; Time courses for lactic acid formation from DHA over H SO , Sc(OTf) , and
2
5
2
4
3
Nb O ·nH O; A reuse experiment of orthoꢀNb O in lactic acid formation from DHA.
2
5
2
2
5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
AUTHOR INFORMATION
Corresponding Author
*
Eꢀmail: nakajima@cat.hokudai.ac.jp (K. N.), uedaw@kanagawaꢀu.ac.jp (W. U.)
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
This work was partly supported by the European UnionꢀJapan joint project, NOVACAM (Novel
cheap and abundant materials for catalytic biomass conversion, FP7ꢀNMPꢀ2013ꢀEUꢀJapanꢀ
6
04319).
REFERENCES
(
(
(
1)
2)
3)
Corma, A. Catal. Lett. 1993, 22, 33−52.
Li, Y.; Yu, J. Chem. Rev. 2014, 114, 7268–7316.
Moliner, M.; Martinez, C.; Corma, A. Angew. Chem. Int. Ed. 2015, 54, 3560−3579.
Tanabe, K. Mater. Chem. Phys. 1987, 17, 217−25.
(4)
ACS Paragon Plus Environment
24

Page 25 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
(
5)
6)
Nowak, I.; Ziolek, M. Chem. Rev. 1999, 99, 3603–3624.
(
Tanabe, K. Catal. Today, 2003, 78, 65−77.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(7)
Nikolla, E.; RománꢀLeshkov, Y.; Moliner, M.; Davis, E. M. ACS Catal. 2011, 1,
4
08−410.
(8)
Nakajima, K.; Baba, Y.; Noma, R.; Kitano, M.; Kondo, N. J.; Hayashi, S.; Hara, M. J.
Am. Chem. Soc. 2011, 133, 4224−4227.
(9)
PagánꢀTorres, J. Y.; Wang, T.; Gallo, R. M. J.; Shanks, H. B.; Dumesic, A. J. ACS Catal.
2
012, 2, 930−934.
(10) Nakajima, K.; Noma, R.; Kitano, M.; Hara, M. J. Phys. Chem. C 2013, 117,
16028−16033.
(11) Choudhary, V.; Mushrif, H. S.; Ho, C.; Anderko, A.; Nikolakis, V.; Marinkovic, S. N.;
Frenkel, I. A.; Sandler, I. S.; Vlachos, G. D. J. Am. Chem. Soc. 2013, 135, 3997−4006.
(12) Nakajima, K.; Noma, R.; Kitano, M.; Hara, M. J. Mol. Catal. A: Chem. 2014, 388,
1
00−105.
(13) Dijkmans, J.; Dusselier, M.; Janssens, W.; Trekels, M.; Vantomme, A.; Breynaert, E.;
Kirschhock, C.; Sels, B. F. ACS Catal. 2016, 6, 31−46.
(14) Yue, C.; Li, G.; Pidko, E. A.; Wiesfeld, J. J.; Rigutto, M.; Hensen, E. J. M.
ChemSusChem 2016, 9, 2421−2429.
(15) Li, G.; Pidko, E. A.; Hensen, E. J. M. ACS Catal. 2016, 6, 4162−4169.
ACS Paragon Plus Environment
25

ACS Catalysis
Page 26 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
16) RománꢀLeshkov, Y.; Moliner, M.; Labinger, A. J.; Davis, E. M. Angew. Chem. Int. Ed.
2
010, 49, 8954−8957.
(17) Yang, G.; Pidko, E. A.; Hensen, E. J. M. ChemSusChem 2013, 6, 1688−1696.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(18) Taarning E.; Saravanamurugan, S.; Holm, M. S.; Xiong, J.; West, R. M.; Christensen, C.
H. ChemSusChem 2009, 2, 625−627.
(19) Holm, M.S.; Saravanamurugan, S.; Taarning E. Science, 2010, 328, 602−605.
(20) Pescarmona, P. P.; Janssen, K. P. F.; Delaet, C.; Stroobants, C.; Houthoofd, K.;
Philippaerts, A.; De Jonghe, C.; Paul, J. S.; Jacobs, P. A.; Sels, B. F. Green Chem. 2010, 12,
083−1089.
1
(21) Li, L.; Stroobants, C.; Lin, K.; Jacobs, P. A.; Sels, B. F.; Pescarmona, P. P. Green Chem.
011, 13, 1175−1181.
2
(22) de Clippel, F.; Dusselier, M.; Van Rompaey, R.; Vanelderen, P.; Dijkmans, J.; Makshina,
E.; Giebeler, L.; Oswald, S.; Baron, G. V.; Denayer, J. F. M.; Pescarmona, P. P.; Jacobs, P. A.;
Sels, B. F. J. Am. Chem. Soc. 2012, 134, 10089−10101.
(23) Guo, Q.; Fan, F.; Pidko, E. A.; van der Graaff, W. N. P.; Feng, Z.; Li, C.; Hensen, E. J.
M. ChemSusChem 2013, 6, 1352−1356.
(24) Dusselier, M.; Van Wouwe, P.; Dewaele, A.; Makshina, E.; Sels, B. F. Energy Environ.
Sci. 2013, 6, 1415−1442.
(25) MäkiꢀArvela, P.; Simakova, I. L.; Salmi, T.; Murzin, D. Y. Chem. Rev. 2014, 114,
1909−1971.
ACS Paragon Plus Environment
26

Page 27 of 28
ACS Catalysis
1
2
3
4
5
6
7
8
9
(
26) Wang, X.; Liang, F.; Huang, C.; Li, Y.; Chen, B. Catal. Sci. Technol. 2016, 6,
6
551−6560.
(27) Murayama, T.; Chen, J.; Hirata, J.; Matsumoto, K.; Ueda, W. Catal. Sci. Technol. 2014,
, 4250−4257.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
(28) Gupta, N.K.; Fukuoka, A.; Nakajima, K. ACS Catal. 2017, 7, 2430–2436.
(29) Sadakane, M.; Watanabe, N.; Katou, T.; Nodasaka Y.; Ueda, W. Angew. Chem. Int. Ed.
2
007, 46, 1493−1496.
(30) Sadakane, M.; Kodato, K.; Kuranishi, T.; Nodasaka, Y.; Sugawara, K.; Sakaguchi, N.;
Nagai, T.; Matsui, Y.; Ueda, W. Angew. Chem. Int. Ed. 2008, 47, 2493–2496.
(31) Buzzoni, R.; Ricchiardi, G.; Lamberti, C.; Zecchina, A. Langmuir 1996, 12, 930−940.
(32) West, R. M.; Holm, M. S.; Saravanamurugan, S.; Xiong, J.; Beversdorf, Z.; Taarning, E.;
Christensen, C.H. J. Catal. 2010, 269, 122−130.
(33) Koito, Y.; Nakajima, K.; Kobayashi, H.; Hasegawa, R.; Kitano, M.; Hara, M. Chem. Eur.
J. 2014, 20, 8068−8075.
(34) Nagayama, S.; Kobayashi, S. J. Org. Chem. 1996, 61, 2256−2257.
(35) Nagayama, S.; Kobayashi, S. Angew. Chem. Int. Ed. 2000, 39, 567−569.
(36) Delidovich, I.; Palkovits, R. ChemSusChem 2016, 9, 547−561.
ACS Paragon Plus Environment
27

ACS Catalysis
Page 28 of 28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
SYNOPSIS
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
28