Characterization of HNbMoO6, HNbWO6 and HTiNbO 5 as solid acids and their catalytic properties for esterification reaction

Article abstract of DOI:10.1016/j.apcata.2012.07.036

HTiNbO₅, HNbMoO₆·H₂O and HNbWO ₆·1.5H₂O were prepared by proton-exchange of the precursors KTiNbO₅, LiNbMoO₆ and LiNbWO₆, respectively, which were synthesized by a solid-state reaction method. The morphology and the micro-structure were characterized by means of scanning electron microscope (SEM), transmission electron microscope (TEM), powder X-ray diffraction (XRD) and Fourier-transform infrared spectroscopy (FT-IR). NH ₃ temperature-programmed desorption (NH₃-TPD) was used to measure the acidic properties of the samples. The Sanderson electronegativity and Kataoka′s bond strength theory were also applied to investigate the acid properties of the as-prepared samples. Finally, their catalytic activities were evaluated through catalyzing the liquid-phase esterification reaction of acetic acid and n-butanol to form n-butyl acetate. It has been revealed that H⁺ ions existed in different forms and interaction modes with MO bond in the solid acids. The acid strength for these as-prepared samples follows the order HNbMoO₆·H₂O > HNbWO₆·1. 5H₂O > HTiNbO₅. In this work, HNbMoO ₆·H₂O presents an excellent catalytic activity for the esterification reaction, while HNbWO₆·1.5H₂O and HTiNbO₅ have little catalytic activity. The result suggested that the catalytic activity of the layered solid acid for the esterification reaction was mainly controlled by gallery height.

Full text of DOI:10.1016/j.apcata.2012.07.036

Applied Catalysis A: General 443–444 (2012) 145–152
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Characterization of HNbMoO6, HNbWO6 and HTiNbO5 as solid acids and their
catalytic properties for esterification reaction
J. He∗, Q.J. Li, Y. Tang, P. Yang, A. Li, R. Li, H.Z. Li
School of Chemical Engineering, Anhui University of Science and Technology, Huainan 232001, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 5 April 2012
Received in revised form 27 July 2012
Accepted 28 July 2012
Available online 7 August 2012
HTiNbO5, HNbMoO6·H2O and HNbWO6·1.5H2O were prepared by proton-exchange of the precursors
KTiNbO5, LiNbMoO6 and LiNbWO6, respectively, which were synthesized by a solid-state reaction
method. The morphology and the micro-structure were characterized by means of scanning electron
microscope (SEM), transmission electron microscope (TEM), powder X-ray diffraction (XRD) and Fourier-
transform infrared spectroscopy (FT-IR). NH3 temperature-programmed desorption (NH3-TPD) was used
to measure the acidic properties of the samples. The Sanderson electronegativity and Kataokaꢀs bond
strength theory were also applied to investigate the acid properties of the as-prepared samples. Finally,
their catalytic activities were evaluated through catalyzing the liquid-phase esterification reaction of
acetic acid and n-butanol to form n-butyl acetate. It has been revealed that H+ ions existed in different
forms and interaction modes with M O bond in the solid acids. The acid strength for these as-prepared
samples follows the order HNbMoO6·H2O > HNbWO6·1.5H2O > HTiNbO5. In this work, HNbMoO6·H2O
presents an excellent catalytic activity for the esterification reaction, while HNbWO6·1.5H2O and HTiNbO5
have little catalytic activity. The result suggested that the catalytic activity of the layered solid acid for
the esterification reaction was mainly controlled by gallery height.
Keywords:
Titanoniobate
Niobium molybdate
Niobium tungstate
Acidic characteristics
Esterification reaction
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
the interlayer proton activity and the laminate negative charges
make the layered solid acids have the shape-selectivity for dif-
Growing concerns regarding environmental and human health
issues have necessitated the development of new industrial reac-
tions with minimal environmental impact. Solid acid catalyst has
drawn more and more attentions because of several advantages,
such as no corrosion to equipment, environmental-friendliness
and easy separation from reactants [1–5]. As a kind of solid acid
material, proton-exchanged form of layered transition metal oxide
possesses the merits of strong acidity, the adjustable surface
properties and interlayer channel, the high surface area and
excellent selectivity for the reactant molecules [6–19]. Niobium
can combine with neighboring elements to form the layered
compounds due to the good affinity. Recently, niobate has shown
promising application in some water-containing reactions, such
as dehydration, condensation and hydrolysis [20,21]. It has been
reported that HNbMoO6 has good catalytic activity for the reac-
tions of Friedel–Crafts alkylation, esterification, condensation,
hydration and hydrolysis [2,10,20–24]. Due to strong acid sites
presenting in the interlayer, the catalytic activity of HNbMoO6
is significantly higher than the molecular sieve and niobate in
the process of esterification and hydrolysis. The difference of
ferent reactants. In addition, the catalytic activity would change
significantly if the reactants vary during esterification reaction
process [18]. By exfoliation and aggregation of the nanosheets
which made up of the layered material, the catalytic activity of the
layered transition metal oxides can be improved dramatically [25].
For example, the layered HTiNbO5 shows high catalytic activity
on esterification, cumene cracking and isopropanol dehydration
reactions after exfoliation and aggregation of the nanosheets
[26–28]. Though the acidic characteristic and the formation mech-
anism of HTiNbO5, HNbMoO6·H2O and HNbWO6·1.5H2O have
been explored detailedly [28], the influence of the composition,
structure and bonding configurations of H+ ions on the acidic
characteristics still need to be explored from the theory.
Herein, the layered transition metal oxides HTiNbO5,
HNbMoO6·H2O and HNbWO6·1.5H2O were obtained with
a
simple proton-exchange reaction. After discussion of the acid
characteristics according to the Sanderson electronegativity and
Kataoka’s theory, the as-prepared solid acids were used to catalyze
the esterification reaction of acetic acid and n-butanol. Under the
same conditions, the catalytic activity of these three kinds of solid
acids were investigated and compared carefully. On the other
hand, the effect of the composition, structure and acidic character-
istics of the as-prepared materials on the catalytic activity for the
esterification reaction was studied specially.
∗
Corresponding author. Tel.: +86 554 6668520; fax: +86 554 6668520.
E-mail address: jhe@aust.edu.cn (J. He).
0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2012.07.036

146
J. He et al. / Applied Catalysis A: General 443–444 (2012) 145–152
2. Experimental
2.1. Preparation of solid acids
Layered HTiNbO5, HNbMoO6 and HNbWO6 were prepared
by proton-exchange of the precursors KTiNbO5, LiNbMoO6 and
LiNbWO6, respectively, which were synthesized through a solid-
state reaction method. The detailed procedure has been described
as follows:
KTiNbO5 was synthesized by heating a stoichiometric mixture
of Nb2O5, K2CO3 and TiO2 in Alumina crucible at 1373 K for 24 h
with the heating rate at 5 K min−1. Considering the volatilization of
potassium, 15% excessive potassium carbonate was needed. Before
calcining, this mixture was ultrasonicated at ambient tempera-
ture for 30 min. The precursor was washed with distilled water to
remove the excess carbonate, and dried in air at 373 K for 10 h.
The protonated form HTiNbO5 was obtained by the treatment of
KTiNbO5 with 2 M HNO3 at 323 K for 48 h under continuous stir-
ring, with the one intermediate replacement of the HNO3. Then the
product was washed with distilled water and dried in air at 353 K
over night.
The host LiNbMoO6/LiNbWO6 was synthesized by calcining
the mixture of the stoichiometric Li2CO3, Nb2O5 and MoO3/WO3
at 853 K/1033 K for 24 h/72 h, respectively. The proton-exchange
reaction was performed by shaking 2.0 g of LiNbMoO6/LiNbWO6
power in 40 mL of 2 M HNO3 aqueous solution at 323 K for four days,
exchanging the acid solution every 24 h over that period. Finally,
the products were washed with distilled water and dried in air at
353 K over night.
2.2. Characterization
The X-ray diffraction (XRD) patterns were collected using a
XD-3 powder diffractometer (Beijing Purkinje General Instrument
Co., Ltd.) with a curved graphite-monochromatied CuK␣ radiation
˚
(ꢀ = 1.5418 A) operated at 40 kV and 30 mA under room tempera-
ture. The TEM microphotographies and SEM images were obtained
by a Hitachi TM-1000 scanning electron microscope (SEM) (KYKY
Technology Development Ltd.) with an acceleration voltage of
25 kV and a JEOL 2010 transmission electron microscopy (HRTEM)
(JEOL Ltd.) with an acceleration voltage of 200 kV, respectively. The
FT-IR absorption spectra were acquired on a Bruker Vector 33 FT-IR
spectrophotometer (DTGS detector) with a resolution of 4 cm−1
.
The NH3-TPD experiments were carried out using a continuous
flow fixed-bed reactor system with thermal conduction detector
(TCD, Tianjin Xianquan, Industry & Trade Development Co., Ltd.).
Before NH3-TPD experiment, 100 mg of the sample was placed in
a quartz tube and heated at 453 K in a flow of N2 (30 mL min−1
,
99.999%) for 1 h. The catalyst was cooled to 333 K and then NH3
adsorption was performed by pulsing NH3 to the sample until it
was saturated. Then the sample was purged using N2 (30 mL min−1
,
99.999%) to remove excessive NH3 which came from the physically
adsorption on the surface of catalyst. Finally, the NH3-TPD pro-
file was recorded in a flow of N2 (30 mL min−1, 99.999%) and the
Fig. 1. SEM images of the as-prepared samples: (a) KTiNbO5, (b) LiNbMoO6 and (c)
LiNbWO6.
temperature was raised from 333 K to 973 K at a rate of 10 K min−1
.
the mixture before and after the reaction, the consumed volume of
NaOH aqueous solution was represented with V1 and V2, respec-
tively. The esterification yield is equal to (1 − V2/V1) × 100%.
2.3. Catalytic experiment
The catalytic activities of the samples were evaluated by the
esterification reaction of acetic acid and n-butanol, which was per-
formed in the flask equipped with a thermometer, a water separator
and a reflux device. After putting 0.4 g catalyst, 104 mmol n-butanol
and 80 mmol acetic acid in the flask, the mixture was maintained
at 387 K until no water evaporated out. Then the separation was
performed for removing the catalyst from the mixture after cool-
ing. 0.1 M NaOH aqueous solution was used to titrate the acid in
3. Results
3.1. Morphology and structure
The SEM images of KTiNbO5, LiNbWO6 and LiNbMoO6 are shown
in Fig. 1. As shown in Fig. 1, one can find that all these three lay-
ered metal oxides were made up of tabular particles with the size

J. He et al. / Applied Catalysis A: General 443–444 (2012) 145–152
147
of 2–5 ␮m. However, it should be noted that layered LiNbWO6 and
LiNbMoO6 samples consist of smaller particles than the KTiNbO5
sample, which could be ascribed to the different calcination condi-
tions employed in the preparation of the products.
The TEM microphotographies of KTiNbO5, LiNbWO6 and
LiNbMoO6 are illustrated in Fig. 2, in which the perfect layered
structure could be observed obviously.
Fig. 3 shows the XRD patterns of the precursors and the corre-
sponding proton-exchanged samples. It is witnessed that KTiNbO5
has diffraction peaks at 2ꢁ = 9.58◦, 19.28◦ and 29.12◦, indicat-
ing a well-ordered layered structure, which could be indexed to
orthorhombic phase KTiNbO5(JCPDS 72-1076). After reaction with
HNO3 aqueous solution, the peak of the sample corresponding
to (0 0 2) plane shifted to 10.47◦, which could be attributed to
the proton-exchange of K+ ions in KTiNbO5. For LiNbMoO6 and
LiNbWO6, the results showed that they own tetragonal structure
referring to JCPDS (50-0884) and JCPDS (84-1764), respectively. The
distinct peak at low angle was assigned to (0 0 1) plane.
Since d-spacing is an important parameter for the layered com-
pounds, the d values of all investigated samples were calculated
based on the Bragg equation ꢀ = 2d sin ␪. Here, the laminate thick-
ness of the layered compounds h and the gallery height I were
calculated with the following equations: (1) h = d − 2r (r is the radius
of interlayer ions); (2) I = d − h (assume the thickness of the lami-
nate is constant before and after proton-exchange). The parameters
(d, h, I) of the precursors and the corresponding solid acids were all
summarized in Table 1.
Compared with the corresponding precursors, three features
could be drawn out for the as-synthesized samples after proton-
exchange. Firstly, the d-spacing significantly changed. For HTiNbO5,
the d-spacing was decreased, while the opposite changes existed
in HNbMoO6 and HNbWO6. This result is associated with the rel-
ative size of K+, Li+ and H+ ions and the hydration degree of
interlayer protons. Referring to the literatures [29,30], the form
of protons in HNbMoO6 and HNbWO6 was most likely to exist as
H3O+ ions. So the solid acids could be formulated as HNbMoO6·H2O
and HNbWO6·1.5H2O, and the corresponding schematic struc-
tures are shown in Fig. 4. Secondly, the lessened diffraction peak
intensity and widened FWHM of the samples demonstrated the
decrease in crystallinity. It could be deduced that the periodic lay-
ered structure of the corresponding precursor has been partially
destroyed in the process of proton-exchange. However, the pres-
ence of diffraction peaks due to in-plane diffractions (e.g., (2 0 0))
for HTiNbO5 indicates that the two-dimensional structure is pre-
served after proton-exchange. Thirdly, the main plane is (0 0 2) for
HNbMoO6·H2O and HNbWO6·1.5H2O, while in the corresponding
precursors the main plane is (0 0 1). We can conclude that the main
plane have changed, in condition of maintaining the original unit
cell, during the process of proton-exchange.
In order to further investigate the effect of proton-exchange
reaction on structure and properties, the IR spectra of samples
before and after proton-exchange were collected (Fig. 5). Accord-
ing to the literatures [31–36], the attribution of typical absorption
peaks are listed in Table 2.
For layered niobium-based composite oxides, although Nb ele-
ment exists in the form of NbO6 unit, the vibration absorption
of terminal Nb O bond is significantly influenced by adjacent
metal elements, such as in LiNbMoO6 and LiNbWO6, the Nb
O
band was located at 871.7 and 879.4 cm−1, respectively. Simi-
larly, the Nb O band is sensitive to the interlayer cations. When
the cations of interlayer (K+ or Li+) in KTiNbO5, LiNbMoO6 and
LiNbWO6 were exchanged by H+ ions, the Nb O bands shifted
from 921.4, 871.7 and 879.4 cm−1 to higher wavenumbers of 944.9,
Fig. 2. TEM microphotographies of the as-prepared samples: (a) KTiNbO5, (b)
LiNbMoO6 and (c) LiNbWO6.
908.3 and 912.2 cm−1, respectively. At the same time, the Mo
O
and W O also blue-shifted to 962.3 and 985.4 cm−1 from 830.0
and 958.4 cm−1, respectively. In addition, although LiNbMoO6 and

148
J. He et al. / Applied Catalysis A: General 443–444 (2012) 145–152
Table 1
The parameters of the as-prepared samples: crystal face, d-spacing, laminate thickness, gallery height and the radius of ions.
Samples
Crystal face
d-spacing (d)/nm
Laminate thickness (h)/nm
The radius of ions (r)/nm
Gallery height (I)/nm
KTiNbO5
HTiNbO5
LiNbMoO6
(0 0 2)
(0 0 2)
(0 0 1)
(0 0 2)
(0 0 1)
(0 0 2)
0.922
0.834
0.958
1.359
0.929
1.265
0.646
0.646
0.806
0.806
0.777
0.777
R(K+) 0.138
r(Ti4+) 0.060
r(Li+) 0.076
r(Mo6+) 0.065
r(Li+) 0.076
r(W6+) 0.062
0.276
0.188
0.152
0.553
0.152
0.488
HNbMoO6·H2O
LiNbWO6
HNbWO6·1.5H2O
LiNbWO6 have similar structure, the vibration frequency of Nb
O
in the interlayer [19,30]. For HNbMoO6·H2O and HNbWO6·1.5H2O,
the same phenomena was not observed, which indicated that the
state and distribution of H+ ions were rather uniform.
bands for LiNbMoO6 and LiNbWO6 also blue-shifted 36.6 and
32.8 cm−1, respectively. From above, it could be inferred that the
hydration degree of the proton has some influence on the M
O
Table 2 also showed that the vibrations of the O Nb O,
O Mo O, O W O, Mo O Nb and W O Nb bridging bonds
shifted in some degree after proton-exchange, which suggested
that these bridging bonds were all influenced by interlayer
cations.
(M = Nb, Mo,W) bands. Moreover, the band 921.4 cm−1, corre-
sponding to vibration of terminal Nb O, is split into two bands
at 944.9 and 906.4 cm−1 for KTiNbO5 (Fig. 5a). This result may be
induced by the two types of hydrogen ions (H+ and H3O+) existing
Fig. 3. XRD patterns of the as-prepared samples.

J. He et al. / Applied Catalysis A: General 443–444 (2012) 145–152
149
Fig. 4. Schematic structures of layered compounds: (a) KTiNbO5, (b) HTiNbO5, (c) LiNbWO6/LiNbMoO6, (d) HNbMoO6·H2O and (e) HNbWO6·1.5H2O.
Table 2
Infrared band assignments of the precursors and the corresponding proton-exchanged samples.
KTiNbO5
HTiNbO5
LiNbMoO6
HNbMoO6·H2O
LiNbWO6
879.4
HNbWO6·1.5H2O
ꢂs (Ti O)
ꢂs (Nb O)
900.0
921.4
981.0
944.9
906.4
871.7
930.0
908.3
962.3
912.2
985.4
ꢂs (Mo O)
ꢂs (W O)
958.4
ꢂs (O Ti O)
ꢂas (O Nb O)
ꢂs (O Nb O)
ꢂs (O Mo O)
632.0
543.0
710
630.6
503.3
781
482.1
434.1
615.2
742
511.5
449.3
634.5
755
472.5
439.6
457.0
457.0
ꢂs (Mo
ꢂs (O
O
W
Nb)
O)
Nb)
632.5
752
649.9
757
ꢂs (W
O
3.2. The characteristics of solid acids
subtraction inference of blank sample, and the result is shown in
Table 3.
NH3-TPD is one of the basic methods to measure the acid
amount and acid strength from the desorbed NH3 quantity
and desorbed temperature. The NH3-TPD profiles of HTiNbO5,
HNbMoO6·H2O and HNbWO6·1.5H2O are shown in Fig. 6. Tak-
ing into account the possible interference of dehydration of
hydroxyl from the laminates, TPD reference profiles without
adsorption NH3 (marked blank) are also shown in Fig. 6 under
the same condition. The peak area of NH3-TPD was integrated
in the corresponding temperature range, expressing the con-
centration of acid sites in corresponding acid intensity. The
total acid amount was calculated from the peak area after
The results clearly suggested that two types of acid sites
were presented on these samples. The NH3-TPD profile for
Table 3
The acid strength and acid amount (which was calculated by subtraction inference
of blank sample) of the as-prepared samples.
Samples
Peak position of NH3-TPD/K
Acid amount/mmol g−1
HTiNbO5
481, 591
598, 700–755
535–583, 690–740
0.439
0.185
0.167
HNbMoO6·H2O
HNbWO6·1.5H2O

150
J. He et al. / Applied Catalysis A: General 443–444 (2012) 145–152
Fig. 7. Catalytic activities for esterification of acetic acid and butanol catalyzed by
HTiNbO5, HNbMoO6·H2O and HNbWO6·1.5H2O.
3.3. Catalytic activity for the esterification
The catalytic activity of the as-prepared samples was evalu-
ated by catalyzing the esterification reaction of acetic acid and
n-butanol, and the results are shown in Fig. 7.
The esterification reaction of acetic acid and n-butanol is a
typical acid-catalyzed reaction. Fig. 7 shows that the esterifica-
tion yield is 41.3% in the absence of catalyst, which could be
attributed to the self-catalysis effect of acetic acid. Among the
present catalysts, HNbMoO6·H2O exhibited the highest esterifica-
tion yield (67.8%) and HNbWO6·1.5H2O had relatively low activity
for this reaction (44.6%). However, when HTiNbO5 (42.0%) was
used as the catalyst, the esterification yield is similar to the blank.
Therefore, it can be concluded that HTiNbO5 almost has no cat-
alytic activity for the esterification reaction of acetic acid and
n-butanol.
Fig. 5. FT-IR spectra of the as-prepared samples: (a) KTiNbO5, (b) HTiNbO5, (c)
LiNbMoO6, (d) HNbMoO6·H2O, (e) LiNbWO6 and (f) HNbWO6·1.5H2O.
HTiNbO5 exhibits two sharp peaks at 481 and 591 K. Fur-
thermore, more NH3 desorption occurred at low temperature
(481 K). However, the NH3-TPD plots for HNbMoO6·H2O and
HNbWO6·1.5H2O exhibit broad TPD profiles with peaks at 598 K
and 700–755 K, 535–583 K and 690–740 K, respectively. This result
indicated that there are more strength acid sites have been gen-
erated on HNbMoO6·H2O and HNbWO6·1.5H2O than on HTiNbO5.
The acid strength for as-prepared samples followed the order
HNbMoO6·H2O > HNbWO6·1.5H2O > HTiNbO5.
4. Discussion
The layered oxides usually contain at least one of Lewis acid
(L acid) sites and Brønsted acid (B acid) sites. L acid sites come
from central metal ions of MO6 octahedral structure unit of the lay-
ered compounds, while B acid sites come from interlayer H+ ions
or the protonated hydroxyl groups. The strength of L acid sites on
the surface of layered compounds could be explained by Sanderson
electronegativity [37] which is a chemical property that describes
the tendency of an atom or a functional group to attract electrons
(or electron density) towards itself. An atom’s electronegativity is
affected by both its atomic number and the distance that its valence
electrons reside from the charged nucleus. The higher the associ-
ated electronegativity value, the more an element or a functional
group attracts electrons towards it, and the stronger the L acid
strength of oxide.
The Sanderson electronegativity of composite oxide could be
calculated according to Eq. (4.1) [38]
ꢀ
ꢁ
1/(p+q+m+n)
Sm
=
XapXqXcmXdn
(4.1)
b
where Xa, Xb and Xc represented the corresponding values of
Sanderson electronegativity of H, Nb and O elements, respectively,
Xd represented the value of Sanderson electronegativity of Ti, Mo or
W element. The letter of p, q, m and n represented the stoichiometric
number of corresponding element.
Fig. 6. NH3-TPD profiles of the as-prepared samples: (a) HTiNbO5, (b) HTiNbO5
(blank), (c) HNbMoO6·H2O, (d) HNbMoO6·H2O (blank), (e) HNbWO6·1.5H2O and (f)
HNbWO6·1.5H2O (blank).
The Sanderson charge fraction ıi of i atom could be calcu-
lated by electronegativity Sm and the initial electronegativity value

J. He et al. / Applied Catalysis A: General 443–444 (2012) 145–152
151
Table 4
a contradiction in the Kataoka’s bond strength analysis and the
experiments conducted by Takagaki et al. However, it is true that
Kataoka’s bond strength theory can be used to explain the strength
of B acid sites qualitatively, either supported oxide catalysts or
layered composite metal oxide catalysts. The acidic characteristics
of the HTiNbO5, HNbMoO6·H2O and HNbWO6·1.5H2O nanosheet
aggregates were also evaluated with 31P MASNMR by Takagaki
et al. [28] and Tagusagawa et al. [24]. The results showed that the
strength of B acid sites in these nanosheet aggregates is closely
depend on the combination of metal cations, decreasing in the order
The bond strength of hydroxyl groups in HTiNbO5, HNbMoO6·H2O and
HNbWO6·1.5H2O (v.u.).
HTiNbO5 HNbMoO6·H2O HNbWO6
·1.5H2O
Ti OH
Nb OH
Mo OH
1.33
1.17
Terminal hydroxyls
Bridge hydroxyls
1.17
1.00
1.17
1.00
W
OH
Ti(OH)Ti
0.67
0.50
0.33
Ti(OH)Nb
Nb(OH)Nb
Mo(OH)Mo
Nb(OH)Mo
W(OH)W
Nb(OH)W
of Nb5+(OH)Mo6+ > Nb5+(OH)W6+ > Nb5+(OH)Nb5+ > Ti4+(OH)Nb5+
,
0.33
0
0.33
which is consistent with the order that estimated by Kataoka’s bond
strength theory.
0.17
0
0.17
As a typical acid-catalyzed reaction, acetic acid can act as a
catalyst to catalyze the esterification reaction of acetic acid and
n-butanol. In accordance with the mechanism of esterification reac-
tion, only the B acid sites in the layered transition metal oxides
have catalytic activity. Although HNbWO6·1.5H2O has the similar
acid strength to HNbMoO6·H2O, their catalytic activity presented
significant difference and HNbMoO6·H2O exhibited the highest cat-
alytic activity. Referring to Table 1, one can deduce that the gallery
height may be the main factor that influenced the catalytic activity
of layered solid acids.
Si of i atom in the compounds, which has been described as
Eq. (4.2) [38].
Sm − Si
1.57 Si
ı =
(4.2)
ꢂ
For H+, Ti4+, Nb5+, O2−, Mo6+ and W6+ in the layered solid acids,
whose Sanderson electronegativity are 2.59, 1.50, 1.42, 3.65, 2.22
and 1.67 [39], respectively. According to Eq. (4.2), the Sanderson
electronegativity values of 2.78, 2.99 and 2.90 could be obtained
for HTiNbO5, HNbMoO6·H2O and HNbWO6·1.5H2O, respectively.
The electronegativity values of metal atoms in MO6 octahe-
The kinetic diameters of acetic acid and n-butanol are 0.44 nm
and 0.50 nm, respectively. Nevertheless, the gallery height of
HTiNbO5 is only 0.188 nm. The esterification reaction of acetic
acid and n-butanol can not occur because the reactant molecules
failed to diffuse into the interlayer channel. Although the gallery
height of HNbWO6·1.5H2O is larger than the kinetics diameter
of acetic acid, the n-butanol molecules are too larger to diffuse
into the interlayer, so that the reaction can’t happen because the
reactant molecules are hard to contact the Brønsted acid sites.
However, the HNbMoO6·H2O not only has large enough gallery
height (0.553 nm) to allow the reactant molecules to contact
the Brønsted acid sites in interlayer, but also own the strongest
acid sites, resulting in the highest catalytic activity for the reac-
tion.
dral units made up of the laminate structure are ı
= 0.6657,
4+
ıNb = 0.7269 for HTiNbO5, ı
= 0.8413, ıMo T=i 0.3309 for
5+
5+
6+
Nb
HNbMoO6 and ıNb = 0.7911, ıW = 0.6026 for HNbWO6. Nb5+
5+
6+
shows the strongest
L
acidity because it has the strongest
ability to capture electrons in these solid acids. The results
indicated that the ability of Nb5+ to capture electrons is differ-
ent due to the changes of its surrounding environment. Thus,
the Lewis acid intensity of Nb5+ in these solid acids decreased
in the order of HNbMoO6·H2O > HNbWO6·1.5H2O > HTiNbO5,
which is consistent with the electronegativity sequence of the
compounds.
5. Conclusion
Although HTiNbO5, HNbMoO6·H2O and HNbWO6·1.5H2O are
all Nb5+-containing compounds, it seems to exist a definitive dif-
ference in catalytic activity as well as Brønsted acidity among
these solid acids, which suggested that the appearance of Brønsted
acid sites is largely structure-dependent. It has been well known
that the terminal hydroxyl groups M OH and bridge hydroxyl
groups M1(OH)M2 could generate the B acid sites in the lay-
ered transition metal oxides. According to the view of Kataoka
and Dumesic [40], the acid strength of terminal hydroxyl groups
In conclusion, a simple proton-exchange method was applied
to synthesize three kinds of solid acids: HTiNbO5, HNbMoO6·H2O
and HNbWO6·1.5H2O. Several works have been done for these
solid acids. On the one hand, an obvious interaction could be
observed between the interlayer cations of the layered transi-
tion metal oxides and the terminal metals-oxygen bonds (M O).
The interaction force not only related to transition metal ele-
ment M and its neighbor transition metal in laminate, but was
also affected by the hydrated degree of H+ ions, which decreased
with the increase of hydrated degree due to a smaller Coulomb
field. On the other hand, the Lewis acid strength of the layered
transition metal oxides has been described based-on Sander-
son electronegativity Sm and Sanderson charge fraction ıi. We
concluded that the Brønsted acid sites on the as-prepared sam-
ples resulted from the bridge hydroxyl groups M1(OH)M2. The
acid strength of these solid acids decreased in the order of
HNbMoO6·H2O > HNbWO6·1.5H2O > HTiNbO5, which is in agree-
ment with the order of the results from NH3-TPD and Kataoka’s
theory. Finally, the solid acids were used to catalyze the esteri-
fication reaction of acetic acid and n-butanol and their catalytic
activities were mainly affected by the gallery height. Featuring
with good shape-selectivity for reactant molecules, large enough
interlayer channel and strong surface acidity, the present solid acid
HNbMoO6·H2O may be used as an effective catalyst for the esteri-
fication reaction.
M
OH and bridge hydroxyl groups M1(OH)M2 are closely related
to the value of the bond strength. The hydroxyl groups could pro-
vide protons and display B acidity only when the value of the
bond strength is in the range of 0.1–0.4 vu (valence unit). The
closer the bond strength closes to 0.1 vu, the stronger the strength
of B acid site becomes. The possible terminal hydroxyl groups
and bridge hydroxyl groups on these solid acids are all listed in
Table 4.
Table 4 shows that all the terminal hydroxyl groups have lit-
tle B acidity. Takagaki et al. [41] also proved that the Nb OH in
HSr2Nb3O10 cannot present strong B acidity. Only bridge hydroxyl
groups Nb5+(OH)Nb5+, Nb5+(OH)W6+ and Nb5+(OH)Mo6+ presented
a certain strength of B acid sites. The order of acid strength fol-
lows Nb5+(OH)Mo6+ = Nb5+(OH)W6+ > Nb5+(OH)Nb5+. According to
Kataoka’s bond strength analysis and our early result [42], the
Ti4+(OH)Nb5+ has not emerge the strong B acid sites. Although
Takagaki et al. [41] thought that HTiNbO5 has strong B acid sites
which come from bridge hydroxyl Ti4+(OH)Nb5+. It seems to be

152
J. He et al. / Applied Catalysis A: General 443–444 (2012) 145–152
Acknowledgments
[18] C. Tagusagawa, A. Takagaki, S. Hayashi, K. Domen, J. Am. Chem. Soc. 130 (2008)
7230–7231.
[19] C.R. Kumar, P.S. Prasad, N. Lingaiah, J. Mol, Catal. A: Chem. 350 (2011) 83–90.
[20] Y.J. Pagán-Torres, J.M.R. Gallo, D. Wang, H.N. Pham, J.A. Libera, C.L. Marshal, J.W.
Elam, A.K. Datye, J.A. Dumesic, ACS Catal. 1 (2011) 1234–1245.
[21] K. Nakajima, Y. Baba, R. Noma, M. Kitano, J.N. Kondo, S. Hayashi, M. Hara, J. Am.
Chem. Soc. 133 (2011) 4224–4227.
The authors acknowledge the financial supports from Natural
Science Foundation of China (21071004) and Natural Science Foun-
dation of Anhui Province of China (11040606M38).
[22] A. Takagaki, C. Tagusagawa, K. Domen, Chem. Commun. (2008) 5363–5365.
[23] A. Takagaki, R. Sasaki, C. Tagusagawa, K. Domen, Top. Catal. 52 (2009) 592–596.
[24] C. Tagusagawa, A. Takagaki, K. Takanabe, K. Ebitani, S. Hayashi, K. Domen, J.
Phys. Chem. C 113 (2009) 17421–17427.
[25] A. Takagaki, M. Sugisawa, D. Lu, J.N. Knodo, M. Hara, K. Domen, S. Hayashi, J.
Am. Chem. Soc. 125 (2003) 5479–5485.
[26] A. Takagaki, T. Yoshida, D.L. Lu, J. Phys. Chem. B 108 (2004) 11549–11555.
[27] C. Tagusagawa, A. Takagaki, K. Takanabe, K. Ebitani, S. Hayashi, K. Domen, J.
Catal. 270 (2010) 206–212.
[28] A. Takagaki, C. Tagusagawa, S. Hayashi, M. Hara, K. Domen, Energ. Environ. Sci.
3 (2010) 82–93.
References
[1] J.H. Clark, Green Chem. 1 (1999) 1–8.
[2] K. Tanabe, W.F. Holderich, Appl. Catal. A: Gen. 181 (1999) 399–434.
[3] Y. Jiang, J. Huang, W. Dai, M. Hunger, Solid State Nucl. Magn. Reson. 39 (2011)
116–141.
[4] A. Takagaki, D.L. Lu, J.N. Kondo, M. Hara, S. Hayashi, K. Domen, Chem. Mater. 17
(2005) 2487–2489.
[5] C. Tagusagawa, A. Takagaki, S. Hayashi, K. Domen, Catal. Today 142 (2009)
267–271.
[29] V. Bhat, J. Gopalakrishnan, Solid State Ionics 26 (1988) 2–32.
[30] N.S.P. Bhuvanesh, J. Gopalakrishnan, Inorg. Chem. 34 (1995) 3760–3764.
[31] O.C. Compton, F.E. Osterloh, J. Phys. Chem. C 113 (2009) 479–485.
[32] Y.J. Lu, R. Lalancette, R.H. Beer, Inorg. Chem. 35 (1996) 2524–2529.
[33] K. Saruwatari, H. Sato, J. Kameda, A. Yamagishi, A. Takagaki, K. Domen, J. Phys.
Chem. B 109 (2005) 12410–12416.
[34] Y.F. Chen, X.H. Zhao, H. Ma, S.L. Ma, G.L. Huang, Y.J. Makita, X.J. Yang, J. Solid
State Chem. 181 (2008) 1684–1694.
[35] Z. Zhang, B.J.M. Goodall, S. Brown, R.J.H. Clark, Dalton Trans. 39 (2010) 711–714.
[36] E. Dussauze, E.I. Kamitsos, E. Fargin, V. Rodriguez, J. Phys. Chem. C 111 (2007)
14560–14566.
[37] G. Connell, J. Dumesic, J. Catal. 105 (1987) 285–298.
[38] R.T. Sanderson, Poler Covalence [M], Academic Press, New York, 1983, p. 17.
[39] http://www.knowledgedoor.com/2/elements handbook/sanderson
electronegativity.html.
[40] T. Kataoka, J.A. Dumesic, J. Catal. 112 (1988) 66–79.
[41] A. Takagaki, M. Sugisawa, D. Lu, J.N. Kondo, K. Domen, S.J. Hayashi, S. J. Am.
Chem. Soc. 125 (2003) 5479–5485.
[6] K. Maeda, M. Eguchi, W. Youngblood, T.E. Mallouk, Chem. Mater. 20 (2008)
6770–6778.
[7] K. Katsumata, S. Okazaki, C.E. Cordonier, T. Shichi, T. Sasaki, A. Fujishima, ACS
Appl. Mater. Interfaces 2 (2010) 1236–1241.
[8] A. Takagaki, M. Sugisawa, D. Lu, J.N. Kondo, M. Hara, K. Domen, S. Hayashi, J.
Am. Chem. Soc. 125 (2003) 5479–5485.
[9] B.H. Xu, B.Z. Lin, Q.Q. Wang, X.T. Pin, O. Zhang, L.M. Fu, Micropor. Mesopor.
Mater. 147 (2012) 79–85.
[10] J. Fu, G.G. Li, F.N. Xi, X.P. Dong, Chem. Eng. J. 180 (2012) 330–336.
[11] N. Teruyuki, J. Porous. Mater. 11 (2004) 79–86.
[12] N. Teruyuki, N. Koichiro, S. Yasuhiro, S. Yuko, H. Takeshi, J. Phys. Chem. C 115
(2011) 8934–8939.
[13] K.G.S. Ranmohotti, E. Josepha, J. Choi, J. Zhang, J.B. Wiley, Adv. Mater. 23 (2011)
442–460.
[14] J.F. Colin, V. Pralong, V. Caignaert, M. Hervieu, B. Raveau, Inorg. Chem. 45 (2006)
7217–7223.
[15] X.G. Xu, X. Ding, Q. Chen, L.M. Peng, Phys. Rev. B 75 (2007) 035423.
[16] A. Riss, M.J. Else, J. Bernardi, O. Diwald, J. Am. Chem. Soc. 131 (2009) 6198–6206.
[17] D. Li, J. Yang, L. Zhang, X. Wang, L. Lu, X. Yang, J. Mater. Chem. 12 (2002)
1796–1799.
[42] J. He, Y.N. Fan, J.H. Qiu, Y. Chen, Acta Chim. Sin. 62 (2004) 1311–1317.