Niobium-containing hydroxyapatites as amphoteric catalysts: Synthesis, properties, and activity

Article abstract of DOI:10.1021/cs4010453

We present here a new class of Nb-containing catalysts offering both acid and base functionalities thanks to the hydroxyapatite (HAP) structure, which acts as a host matrix for guest niobium species. Incorporation of Nb has led to samples with strengthened amphoteric properties with a prevalence of acid or basic sites depending on the synthesis procedure. Nb-containing HAP samples (12-14 wt % of Nb) have been synthesized by coprecipitation in water and in alcoholic solutions and by impregnation starting from different precursors. The intrinsic and effective acid and base properties have been measured, and a series of mass and surface characterizations have been performed to determine the Nb content, its oxidation state as well as its morphologic, structural, and electronic properties. The performances of the samples have been studied in several reactions of liquid-solid heterogeneous catalysis in batch or fixed tubular reactors working in the recirculation mode on the basis of the known catalytic features of Nb, including hydrolysis, acid-catalyzed isomerization, and selective oxidation reactions.

Full text of DOI:10.1021/cs4010453

Research Article
pubs.acs.org/acscatalysis
Niobium-Containing Hydroxyapatites as Amphoteric Catalysts:
Synthesis, Properties, and Activity
Paolo Carniti,† Antonella Gervasini,*,†,‡ Cristina Tiozzo,§ and Matteo Guidotti§
†Dipartimento di Chimica, Universita
̀
degli Studi di Milano, via Camillo Golgi, 19, I-20133 Milano, Italy
‡CIMaINa, Universita
̀
degli Studi di Milano, via Celoria 16, I-20133 Milano, Italy
§CNR-Istituto di Scienze e Tecnologie Molecolari, via Camillo Golgi 19, I-20133 Milano, Italy
S
* Supporting Information
ABSTRACT: We present here a new class of Nb-containing catalysts
offering both acid and base functionalities thanks to the hydroxyapatite
(HAP) structure, which acts as a host matrix for guest niobium species.
Incorporation of Nb has led to samples with strengthened amphoteric
properties with a prevalence of acid or basic sites depending on the synthesis
procedure. Nb-containing HAP samples (12−14 wt % of Nb) have been
synthesized by coprecipitation in water and in alcoholic solutions and by
impregnation starting from different precursors. The intrinsic and effective
acid and base properties have been measured, and a series of mass and
surface characterizations have been performed to determine the Nb content,
its oxidation state as well as its morphologic, structural, and electronic
properties. The performances of the samples have been studied in several
reactions of liquid−solid heterogeneous catalysis in batch or fixed tubular
reactors working in the recirculation mode on the basis of the known catalytic features of Nb, including hydrolysis, acid-catalyzed
isomerization, and selective oxidation reactions.
KEYWORDS: calcium hydroxyapatite, niobium catalysts, solid acidity, solid basicity, amphoteric properties of catalysts,
oxidation reactions, saccharide conversion
All of this evidence on the catalytic properties of the
exchanged HAPs convinced us that the surface properties of
these compounds can be modified and directed toward a
specific catalytic use. To our best knowledge, Nb species have
not been accommodated yet into the host HAP structure with
the final aim of obtaining amphoteric catalysts.
Niobium inserted in or deposited onto inorganic oxide
structures is an interesting catalytic species with particular
acidic and redox properties,16−20 which has been used in several
selective oxidation17,21−24 and acid-demanding reactions.22,25
Nb-containing catalytic systems, both mixed oxide25,26 and
zeolite systems,27,28 have shown interesting performances. In
addition, the literature reports several Nb-containing crystalline
solids with versatile catalytic features and well-defined
structures that can be applied in oxidation,29−33 insertion,34
and acid-catalyzed transformations.35,36
INTRODUCTION
■
Calcium hydroxyapatites (HAPs, Ca10(PO4)6(OH)2) obtained
by synthesis have been extensively used in a variety of
applications, such as biomaterials, ion exchangers, adsorbents,
and catalysts.1,2 All HAP compounds present a high chemical
and thermal stability and insolubility in water, properties that
are useful for their use in heterogeneous catalysis.
The applications of HAP in catalysts exploit its acid−base3
and redox4,5 properties, which can be easily adjusted taking
advantage of the flexibility of the apatite structure. With respect
to the acid−base properties, HAP has the unusual feature of
containing both acid sites and basic sites in a single crystal
lattice. Stoichiometric HAP, characterized by a Ca/P molar
ratio of 1.67, has been shown to possess a prevalence of basic
sites over the acid sites, and the nonstoichiometric HAP forms
(Ca/P ranging from 1.50 to 1.67) display more acid
characteristics.3,6 In general, many ionic elements and groups
can be exchanged if the charge balance is maintained. For
example, Ca2+ ions can be exchanged with divalent (Cu2+, Co2+,
Ni2+, Zn2+, Mn2+, Pb2+, or Sr2+)7−10 or trivalent (Fe3+, Cr3+,
Al3+, and La3+)8,11−13 metal cations, thus giving rise to effective
catalysts for various reactions. Also, the hydroxyl groups
localized in the HAP tunnel can be exchanged with other types
of anions, such as Cl−, F−, Br−, and the phosphate groups can
be substituted by anion groups, as vanadates.14,15
One of the distinguishing feature of the Nb catalysts is the
resistance to water of the isolated Nb Lewis acid sites
(LAS),28,37 which can maintain lively catalytic activity in polar
and protic solvents,38 differently from many conventional LAS-
containing catalysts. The effective acidity in water of some
Received: September 11, 2013
Revised: December 16, 2013
Published: December 20, 2013
© 2013 American Chemical Society
469
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479

ACS Catalysis
Research Article
catalysts containing Nb has been experimentally demonstrated
by various catalytic and analytical measurements,38,39 and it has
been exploited for reactions of biomass valorization realized in
water.25,40,41
Thus, our main purpose was to assess the effect of Nb
presence in/on the HAP matrix and to explore how the
synthesis method of the Nb/HAP structures can affect their
physical, chemical, and catalytic properties. Several liquid−solid
heterogeneous reactions have been chosen as probe tests to
enlighten the catalytic feature of niobium species, including
hydrolysis, dehydration, acid-catalyzed isomerization, and
oxidation reactions.
vacuum. Drying and calcination of the solid were performed as
reported above.
1.2. Catalyst Characterization. Inductively coupled
plasma (ICP) measurements were determined by ICP-OES
(plasma optic emission spectrometer, Horiba JOBIN YVON)
after solid digestion in acid mixtures (HClO4 + HNO3 + HF)
followed by evaporation and new dissolution in HNO3.
Thermogravimetric analysis (TGA) was performed on both
dried and calcined samples by using a TGA analyzer from
Perkin-Elmer (TGA7) in the temperature interval from 50 to
800 °C under flowing air with a ramp of 10 °C min−1.
BET specific surface area (m2·g−1), specific pore volume
(cm3·g−1), and BJH pore size distribution of the sample were
determined from N2 adsorption and desorption isotherms
measured at liquid nitrogen temperature with an automatic
surface area analyzer (Sorptomatic 1900 instrument). Prior to
the measurement, the sample (≈100 mg), crushed and sieved
as 0.250−0.354 mm particles, was thermally activated at 350 °C
for 16 h under vacuum.
This study provides the first evidence that Nb/HAP samples
can be successfully used as a new family of amphoteric catalysts
with tunable acid−base properties.
1. EXPERIMENTAL SECTION
1.1. Catalyst Preparation. All the chemical reactants,
diammonium hydrogen phosphate ((NH4)2HPO4, VWR
Prolabo AnalaR Normapur), calcium nitrate (Ca(NO3)2,
Merck ACS), ammonium hydroxide (NH4(OH), Fluka, 28 wt
%), and niobium chloride (NbCl5, Aldrich, 99%) were reagent-
grade chemicals, and they have been used without any further
purification. Ammonium niobium oxalate complex,
(NH4)NbO(C2O4)2·nH2O (ANBO), was kindly supplied
X-ray diffraction (XRD) of the powder samples was carried
out by a Philips PW1710 vertical goniometer diffractometer
using Ni-filtered CuKα radiation (λ = 1.54178 Å).
Scanning electron micrographs (SEM) were collected by a
LEO-1430 coupled with an energy-dispersive X-ray spectros-
copy (EDX) working with an accelerating voltage of 20 kV.
Transmission electron micrographs were collected on a Zeiss
LIBRA 200FE analytical HR-TEM microscope, equipped with
200 kV FEG, in-column second-generation omega filter for
energy selective spectroscopy (EELS) and for imaging (ESI),
HAADF STEM facility and EDS probe for chemical analysis.
UV−vis diffuse reflectance spectroscopy (UV−vis-DRS)
measurements were performed on fine powder samples put
in a cell with optical quartz walls of a Perkin-Elmer Lambda 35
instrument equipped with a Labsphere RSA-PE-20 integration
sphere and Spectralon as reference material.
X-ray photoelectron spectroscopy (XPS) analyses were
performed using a Kratos Analytical Axis Ultra DLD
spectrometer with a monochromatic Al Kα (hν = 1486.6 eV)
radiation source operated at 15 kV and 15 mA. The binding
energy (BE) was calibrated on the basis of the line position of
C 1s (285 eV). CasaXPS processing software was used to
estimate the relative abundance of the different species and to
deconvolute the photopeaks.
The acidity and basicity measurements of the samples in
liquid by using 2-phenylethylamine (PEA) as basic probe and
benzoic acid (BA) as acidic probe were carried out in
cyclohexane for the intrinsic acidity (I.A.) and intrinsic basicity
(I.B.), respectively, and in water for the effective acidity (E.A.)
determination (cyclohexane and water). The samples were
thermally activated at 350 °C for 16 h under air flow (8 mL·
min−1). A recirculation chromatographic line (HPLC)
comprising a pump (Waters 515) and a monochromatic UV
detector (Waters, model 2487, λ = 254 nm) was used as
described in the Supporting Information (SI) and refs 41 and
42. Previous tests of PEA and BA adsorption showed that both
probes are adsorbed a negligible amount on samples without
any acid and basic sites (sand sample, Fluka). In addition,
calibration tests by PEA and BA adsorption on acidic and basic
resins, Amberlite 200 and AmberlystTM A21, respectively,
(Carlo Erba and Dow) were performed with a certified amount
of acid and basic sites.
from Companhia Brasileira de Metallurgia e Mineraca̧ o
̃
(CBMM).
Stoichiometric calcium hydroxyapatite, Ca10(PO4)6(OH)2,
with a Ca/P molar ratio of 1.67 was synthesized by the
coprecipitation method, adopting the procedure reported by C.
Lamonier et al.3
Niobium was introduced during the HAP synthesis using
ammonium niobium oxalate, ANBO, or niobium chloride as
Nb precursors by the coprecipitation method (one-step
synthesis) or by the Nb deposition by impregnation of
presynthesized HAP sample. An amount of about 12−14 wt
% of Nb was introduced in the final sample.
Nb/HAPcw was synthesized using aqueous solutions of
calcium nitrate (0.372 M in 100 mL) and ammonium niobium
oxalate complex (0.083 M in 100 mL) simultaneously pumped
at a constant rate (0.55 mL·min−1) into an aqueous solution of
diammonium hydrogen phosphate (0.12 M in 250 mL) while
stirring and maintained at 90 °C under nitrogen atmosphere.
Then, basification of the solution at pH = 10, filtration,
washing, aging of precipitate, and solid drying (50 °C in vacuo)
and calcination (600 °C for 4 h) were carried out.
For the Nb/HAPca synthesis, an ethanol solution of niobium
chloride (0.083 M in 100 mL of ethanol) and a water solution
of calcium nitrate (0.3725 M in 100 mL) were simultaneously
pumped at constant rate (0.55 mL·min−1) into a hydroalcoholic
solution of diammonium hydrogen phosphate (0.12 M in 0.25
mL of 1:1 ethanol−water solution), while they were stirred and
maintained at 70 °C under nitrogen atmosphere. The
successive operations of basification of the solution, precipitate
filtration, washing, aging, and drying and calcination of the solid
were carried out as referred to above.
The sample labeled Nb/HAPi was prepared by impregnation
of ammonium niobium oxalate of a presynthesized HAP
sample. A weighted amount (5 g) of dried HAP was added to
ammonium niobium oxalate water solution (0.16 M in 50 mL),
and they were stirred and maintained at 30 °C for about 15 h.
Then, water was removed by evaporation at 50 °C under
1.3. Catalytic Activity. 1.3.1. α-Pinene Oxide Isomer-
ization. Tests of the catalytic isomerization of (−)-α-pinene
470
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479

ACS Catalysis
Research Article
Table 1. Composition of the Nb/HAP Samples
c
sample
elemental composition used for synthesis
actual elemental composition
wt %
Ca
atomic ratio
wt %
Ca
atomic ratio
Nb
P
Nb/P
Ca/P
Nb
P
Nb/P
Ca/P
a
HAP
40.0
27.7
27.7
18.5
17.2
17.2
1.67
1.25
1.25
1.67
37.84
30.2
26.5
31.7
18.4
13.84
16.4
1.59
1.69
1.25
1.61
Nb/HAPCW
Nb/HAPCA
14.2
14.2
12.1
0.28
0.28
0.26
9.7
13.1
7.7
0.23
0.26
0.17
b
Nb/HAPi
15.2
a
b
c
Chemical formula considered for the preparation: Ca10(PO4)6(OH)2. Prepared starting from finite HAP powder. Measured by ICP.
oxide (100 mg; 0.66 mmol; 97%, Aldrich) was performed in
liquid phase in a glass batch reactor (stirring rate of 1000 rpm)
at room temperature using toluene as solvent (8 mL; puriss.,
rpm magnetic stirring) under inert atmosphere. All the samples
were pretreated in vacuum at 120 °C and cooled to room
temperature under vacuum prior to use. Acetonitrile (Aldrich,
HPLC grade) was used as solvent. Aqueous hydrogen peroxide
(H2O2; 50% Aldrich) was used as oxidant agent, with an
oxidant to substrate molar ratio of 2:1. The total volume of the
mixture was 5 mL. Samples were analyzed by GC
chromatography (Agilent 6890 Series; HP-5 column, 30 m ×
0.25 mm; FID or MS detector). Neither significant auto-
oxidation nor support-catalyzed contributions to epoxidation
were recorded with any sample, limonene conversion being
about 10% with no remarkable epoxide formation. After all
tests, the presence of residual hydrogen peroxide was checked
and confirmed by iodometric assays.
Riedel-de-Haen), previously dried on molecular sieves (Silipor-
̈
ite, 3 Å). The solid catalyst (35 mg) was pretreated at 110 °C
for 3 h in vacuum. The reaction was performed for 6 h, and
samples were taken at regular intervals. Reaction products were
determined by GC analysis (HP5890; HP-5 column, 30 m ×
0.25 mm; FID or MS detectors, head pressure 160 kPa;
mesitylene internal standard). Standard deviation was
2%
and 3% in conversion and selectivity values, respectively.
1.3.2. Activity with Sugars in Aqueous Solvent. Sucrose, D-
(−) fructose, and D-(−) glucose (RPE, Carlo Erba) were used
as substrates and standards for the various analyses.
In the tests of catalyst recovery, the solid was separated by
filtration and thoroughly washed with fresh acetonitrile and
then with methanol (Fluka, HPLC grade). The filtered solid
was dried at 100 °C.
The reaction test of hydrolysis of sucrose was performed in
water with the powder catalyst put in a fixed catalytic bed
reactor working in a total recirculation reaction mode at
atmospheric pressure, as detailedly described in refs 42 and 43
and in the SI. The reaction solution was continuously circulated
(3 mL·min−1) through the catalyst (0.5 g, 0.250−0.354 mm
particles).
The catalytic test of glucose isomerization was carried out in
water in a total recirculation reaction line equipped with a
HPLC pump (Merck-Hitachi L6200 Intelligent Pump),
stainless steel preheater, and fixed-bed tubular flow reactor
(17 cm long, 0.5 cm internal diameter).40,42 The initial aqueous
glucose concentration (≈0.55 M) was continuously circulated
(3 mL·min−1) through the catalyst (3 mL·min−1) at constant
contact time (≈0.17 min·g·mL−1). The reaction was studied for
a total of 8 h. A sample (typically 1 mL) was withdrawn from
the reservoir every 2 h of reaction and analyzed.
The products were analyzed by HPLC, which was assembled
with an injector (Waters U6K), a pump (Waters 510), a heater
(Waters CHM) for the column, and a refractive index detector
(Waters 410). A SugarPak-I column (Waters) operating at 90
°C and eluted with a solution of Ca-EDTA 10−4 M in water
(0.5 mL min−1) was used.
1.3.3. Oxidation Catalytic Activity. The cyclo-octene
epoxidation reaction was carried out in the fixed catalytic bed
reactor working in a total recirculation reaction line at
atmospheric pressure, as described above. The powder catalyst
(0.6 g with 0.250−0.354 mm particles) was thermostatted at 55
0.1 °C, and the feeding solution was put in the reservoir
containing the cyclo-octene (0.45 M) plus H2O2 (0.45 M) in
methanol thermostatted at 17 °C. The solution volume in the
reservoir was 100 mL. The reaction solution was continuously
circulated (3 mL·min−1) through the catalyst. The reaction was
studied for a total of 10 h. At every hour, a sample (typically 1
mL) was withdrawn from the reservoir and analyzed.
2. RESULTS AND DISCUSSION
2.1. Preparation and Characteristics. The same
sequence of unit operations has been followed for the synthesis
of HAP and of Nb/HAPcw (by using aqueous ANBO
precursor) and Nb/HAPca (by using ethanolic NBE
precursor). The unripe solids were dried and calcined to
obtain the final samples. The other sample, Nb/HAPi, was
prepared by ANBO deposition on presynthesized HAP by a
conventional impregnation procedure.
The elemental amounts used for the synthesis and Ca/P and
Nb/P ratios of all the prepared samples are presented in Table
1. Niobium-free HAP was synthesized with the Ca/P atomic
ratio of 1.67, in accordance with the stoichiometric
hydroxyapatite (Ca10(PO4)6(OH)2) by coprecipitation. For
the Nb/HAP samples prepared by coprecipitation, in the
presence of aqueous ANBO solution or of ethanolic solution of
niobium chloride, an amount of Nb around 14 wt % was
introduced, leading to theoretical Nb/P and Ca/P ratios of 0.28
and 1.25, respectively. The impregnated sample, Nb/HAPi, was
prepared using the niobium-free HAP powder as support and
deposited an amount of ≈12 wt % of Nb on it by classical wet
impregnation. For HAP, the measured Ca/P ratio was lower
than that expected, indicating that a Ca-deficient hydroxyapatite
was formed with formation of cationic vacancies.7 Important
differences were found between Nb/HAPcw and Nb/HAPca
samples: a higher Nb amount was effectively introduced in Nb/
HAPca than in Nb/HAPcw, the higher amount of Nb caused
the lowering of the Ca/P ratio (Ca/P of 1.25 and 1.69 for Nb/
HAPca and Nb/HAPcw, respectively), with an influence on the
acidity properties of these two samples (see section 2.2, Surface
Acid and Base Properties). The real Nb concentration in Nb/
HAPi was much lower than the nominal one; the amount of Nb
The epoxidation tests on (R)-(+)-limonene (97% Aldrich;
98% e.e.) were carried out in a glass batch reactor (90 °C, 800
471
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479

ACS Catalysis
Research Article
not present on the HAP surface likely has segregated as a
separate NbOx phase.
dimension, covered by a layer of small particles with the
laminar-plate structure (Figure S.2, SI). The presence of Nb
makes the surface of the samples smoother, denser particles
have little protuberances at the surface layers transformed to
microsheets. EDX-microanalysis confirmed the trend in metal
content, as observed from ICP analysis. The measured Nb
content was 16.8, 19.7, and 14.3 wt % for Nb/HAPcw, Nb/
HAPca, and Nb/HAPi, respectively. By comparing EDX and
ICP values, niobium seems to be mainly located in the first
surface layers of the samples. This is particularly evident for
Nb/HAPi, as expected, since niobium has been deposited by
impregnation onto the HAP surface. HR-TEM analysis on the
Nb/HAPca sample revealed the presence of even crystalline
aggregates. At the HR-TEM level, at nanometric scale, a
homogeneous distribution of Nb, Ca, and P atoms was
observed. This is consistent with the formation of a definite
unique structure with no evident segregation of different phases
(EDX elemental mapping images in Figure S.3, SI)
The FT-IR spectra of all the samples confirmed the
formation of an apatite phase with the observation of several
fundamental vibration modes of PO43‑ groups.45 The Nb/HAP
samples in comparison with niobium-free HAP showed a
strong decrease of the intensity of ν(OH) stretching band at
3570 cm−1 (lattice OH− ions) and of the O−H− deformation
mode at 630 cm−1 (Figure S.4, SI). Such a decrease is
consistent with the substitution of Ca2+ by (NbO)3+, which
provokes the elimination of protons from the OH hydroxyls
located in the apatite tunnels to counterbalance the electron
excess due to the new guest species.12,46
TGA analyses revealed that Nb/HAPcw and Nb/HAPi,
prepared from ANBO, decomposed in three steps, about at
170, 250, and 700 °C. In addition, an evident mass loss was
observed at 450 °C that was due to water removal from
phosphate groups of the HAP structure in interaction with Nb.
Nb/HAPca, prepared from the niobium chloride precursor,
showed the mass loss at 450 °C along with a broad mass
diminution in the 100−200 °C interval, associated with water
and ethanol removal. The thermograms of all the samples after
calcination no longer presented defined peaks of mass loss;
only a light mass decrease (1−2 wt %) was observed at
temperatures around 700 °C (Figure S.1, SI).
The crystalline structure of niobium-free HAP and Nb/HAP
samples was studied by XRD analysis. It is known that low
crystallinity HAP particles have to be calcined to improve
thermal and chemical stability.43,44 All the samples exhibit
diffraction lines corresponding to the typical hydroxyapatite
structure (point group P63/m, JCPDS 01-086-0740), and the
poorer crystallization of the Nb/HAP samples in comparison
with bare HAP is clearly showed in Figure 1. HAP and Nb/
The textural results of all the samples are given in Table 2.
The HAP specific surface area (SBET) is 43 m2·g−1 with low
porosity. The isotherm belongs to IV type with little, but
visible, hysteresis between the adsorption and desorption
isotherms (Figure 2). Accordingly, a broad pore size
population, with two maxima centered around 18 and 30 nm
of pore diameter, was observed. The deposition of Nb on HAP
(Nb/HAPi) caused a shift of the two main pore size
populations with respect to HAP (Figure 2D). The
coprecipitated Nb/HAP samples showed more important
porosimetric differences. The Nb/HAPcw sample has a specific
surface area twice higher than HAP and a more defined pore
size distribution profile centered at 17 nm of size (Figure 2B).
The Nb/HAPca sample has a lower surface area than Nb/
HAPcw, with several distinct pore populations of different sizes
(Figure 2C). This could be due to the higher amount of Nb
introduced in the apatite structure, as confirmed by ICP
analyses.
Figure 1. XRD comparative patterns of all the studied samples.
HAP present the hexagonal system made of compact
3−
assemblage of Ca2+ ions and tetrahedral PO4 groups
delimiting the typical unconnected channels.7 Additional
diffraction lines were not observed in the samples containing
niobium, indicating the absence of Nb-separate phases in all the
Nb-containing samples. Unfortunately, due to the poor
crystalline feature of Nb/HAPcw and Nb/HAPca samples, it
was not possible to determine whether (NbO)3+ species had
been introduced in the apatite structure by the determination of
the unit-cell parameter values. Nevertheless, due to the large
ionic radius of (NbO)3+, the accommodation of the guest ions
in the apatite channels cannot be excluded.
The diffuse reflectance UV−vis spectra of HAP consists of
one broad low-intensity band located around 200 nm which can
be ascribed to charge transfer O2− → Ca2+. Following the
addition of Nb, new intense LMCT (ligand-to-metal charge-
transfer O2− → Nb5+) bands appeared in the UV region (Figure
S.5, SI), but it was impossible to discriminate between the
contribution of tetrahedral (230 nm) or five- and six-
The morphology of the HAP and Nb/HAP samples was
studied by direct observation of the surface images collected by
SEM-EDX. The powders consist of agglomerates of individual
particles with platelet-like morphology and 10−20 μm
Table 2. Morphologic and Structural Properties of the Samples
sample
HAP
BET surface (m2·g−1
)
pore volume (cm3·g−1
)
pore radius (nm)
42.7
98.5
55.0
47.7
0.24
0.20
0.19
0.21
9; 15
8.5
Nb/HAPCW
Nb/HAPCA
Nb/HAPi
6.5; 11; 15
11; 20
472
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479

ACS Catalysis
Research Article
Figure 2. Isotherms of N2 adsorption/desorption collected at −196 °C (left side) and cumulative pore volume and pore volume distribution as a
function of pore radius (right side) of all the studied samples: HAP (A), Nb/HAPCW (B), Nb/HAPCA (C), Nb/HAPi (D).
coordinated Nb species (250 nm).25,47 Interestingly, no band
has been observed at 320 nm, confirming the absence of niobia
(formed as segregated phase) in small nanodomains.47
Some information on the presence of Nb on the sample
surface has been collected by XPS analysis. The values of Nb 3d
binding energy and the shape of the peaks for all the samples
are typical for niobium with a formal oxidation state of +5
(3d5/2, values of 207.5−208.0 eV and 3d3/2, values of 210.2−
201.7 eV). In Figure 3 are reported the typical doublets (Nb 3d
photopeaks) observed in the 195−220 eV interval of binding
473
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479

ACS Catalysis
Research Article
between acid and basic sites of the sample, thus influencing its
catalytic performance. To our best knowledge, no information
about the acidity properties of hydroxyapatite hosting niobium
species has been reported until now.
The Nb/HAP samples have been investigated by measuring
the adsorption properties of basic (phenylethylamine, PEA)
and acid (benzoic acid, BA) probe molecules in an adsorption
reaction line previously reported.41 The measurements were
performed in two different liquids: an aprotic/apolar solvent,
cyclohexane, and a protic/polar one, water, in order to collect
information on the intrinsic and effective acidity, respec-
tively.25,41,50 Nb-containing catalysts are known to possess
water-tolerant properties, that is, the ability to maintain lively
acid properties in water and other polar and protic
liquids.37,39,40,42,51
Figure 4 shows all the obtained results of the intrinsic acidity
(I.A.) and intrinsic basicity (I.B.) for HAP and the three samples
containing Nb, collected in cyclohexane. On the right side of
Figure 4 are reported the first and the second isotherms of PEA
adsorption and on the left side the first and the second
isotherms of BA adsorption, which are related with the acidity
and basicity properties of the samples, respectively. The
determination of the total amount of acid(base) sites has
been made from the first run isotherm (Figure S.6, SI).
As expected, HAP shows a higher amount of intrinsic basic
sites than acid sites with a ratio between acid and basic sites
(I.A./I.B.) of 0.71 (Table 3). The situation changed when Nb
was presented into the HAP matrix, although a common-trend
behavior could not be observed. The amount of acid sites
increased for Nb/HAPcw and Nb/HAPca and deeply
decreased for Nb/HAPi, in comparison with Nb-free HAP.
With regard to basicity, a different situation was observed.
Nb/HAPcw and Nb/HAPi have a similar amount of basic sites
compared to HAP, while Nb/HAPca showed half of the basic
sites compared to HAP. Therefore, for the Nb/HAP samples,
the ratio between the acid and basic sites was observed to shift
from that of HAP; a higher value for Nb/HAPcw and, in
particular, for Nb/HAPca, and a lower value for Nb/HAPi have
been determined (Table 3). The lack of of acid sites on Nb/
HAPi was unexpected. Nb/HAPi has a similar amount of basic
sites than HAP (0.144 vs 0.155 mmol·g−1, respectively), but it
has a much lower content of acid sites (0.070 vs 0.110 mmol·
g−1, respectively). The amphoteric properties of all the samples
(although with a somehow more marked acid or basic
character) indicate that Nb was accommodated in/on different
sites of HAP as a function of the preparation mode.
The influence on the acid−base site strength of the samples
has been observed also, as it can be seen by comparing the
isotherms of the first and second runs of PEA and BA
adsorption (Table 3 and Figure 4). From the two-runs
isotherms, the percentage of the strong acid sites could be
calculated. The strong acid sites are 60% of the total acid sites
of HAP and Nb/HAPcw, while they increase up to 80% for
Nb/HAPca and deeply decrease for Nb/HAPi down to about
4%. In fact, the isotherms of the first and second runs of the
PEA adsorption are very distant from each other for Nb/
HAPca, and they are very close for Nb/HAPi (Figure 4C,D,
right side). With respect to the basic strength of the sites, only
Nb/HAPca shows an important decrease of the amount of
strong basic sites (only 19%) of the total ones, while for all the
other samples, the strong basic sites are around 50−59% of the
total basic sites possessed (Table 3 and Figure 4, left side).
Figure 3. XPS measurements: region of Nb 3d bands of Nb/HAPCW
(A), Nb/HAPCA (B), Nb/HAPi (C).
energy for Nb/HAPcw, Nb/HAPca, and Nb/HAPi. It is known
that it is possible to derive information on Nb dispersion from a
close inspection of the Nb 3d photopeak broadening.48 An
increase in the values of the full width at half-maximum (fwhm)
of Nb 3d5/2 and a reduction in the Nb 3d valley between the 5/
2 and 3/2 components indicates the presence of well-dispersed
surface Nb species, as reported by Dragone et al.49 and already
observed by some of us.25 Both the effects are clearly observed
for Nb/HAPcw (fwhm of 1.75 and 1.46 eV for Nb/HAPcw and
the two Nb/HAPca and Nb/HAPi samples) and are in
agreement with a better dispersion of Nb in the Nb/HAPcw
sample than in the other Nb-containing samples.
2.2. Surface Acid and Base Properties. The amphoteric
properties of HAP have been already showed by several groups
and exploited for catalytic purposes (Scheme S.1. displays the
acid and base functionalities of the HAP matrix).3,4,6 The
presence of guest ions in the HAP matrix can change the ratio
474
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479

ACS Catalysis
Research Article
Figure 4. Intrinsic acidity (I.A.) and intrinsic basicity (I.B.) measurements of HAP and Nb−HAP samples: isotherms of 2-phenyl-ethylamine (PEA)
adsorption, for acidity (right side), and of benzoic acid (BA) adsorption, for basicity (left side), in cyclohexane at 30 °C on HAP (A), Nb/HAPCW
(B), Nb/HAPCA (C), Nb/HAi (D) (1st run was carried and on fresh sample and 2nd run after flushing the PEA(BA) saturated sample with pure
solvent for 45 min).
Table 3. Surface Acidity and Basicity of HAP and Nb/HAP Samples Determined by Adsorption of Basic (2-Phenyl-Ethylamine,
a
PEA) and Acidic (Benzoic Acid, BA) Probes
b
b
c
b
sample
HAP
intrinsic acid sites (I.A.) (mmol g−1
)
intrinsic basic sites (I.B.) (mmol·g−1
)
I.A./I.B. ratio
effective acid sites (E.A.) (mmol·g−1
)
d
d
0.110 (62%)
0.155 (50%)
0.71
0.84
1.73
0.49
0.019
0.013
0.029
Nb/HAPCW
Nb/HAPCA
Nb/HAPi
0.164 (62%)
0.147 (80%)
0.070 (4.3%)
0.196 (59%)
0.085 (19%)
0.144 (53%)
0.019
a
b
Adsorption temperature, 30 °C; solvent, cyclohexane for I.A. and I.B. measurements, water for E.A. measurements. Determined at the intersection
point between the straight line of the low-concentration part of the adsorption isotherm with the asymptotic part of the isotherm (Figure S.5, SI).
c
d
Calculated from the relevant values in the 2nd and 3rd columns. Percent amount of strong acid and basic sites, determined by subtracting the
second from the first isotherm of PEA or BA probe adsorption, respectively.
In summary, the rather balanced population of acid and base
sites shown by HAP (I.A./I.B. = 0.71) was maintained for the
sample prepared by coprecipitation in aqueous solution, Nb/
HAPcw (I.A./I.B. = 0.84), with an increase in both acid and
base sites, which was due to the presence of Nb and to the
larger surface area of this sample in comparison with HAP. The
sample prepared by coprecipitation in alcoholic solution, Nb/
HAPca, had a higher amount of acid sites than basic sites; the
amount of acid sites were higher than those of HAP, due to the
presence of Nb. The addition of the guest Nb species to HAP
by impregnation led to a sample with a prevalent basic
character (Nb/HAPi with I.A./I.B. = 0.49); likely, Nb was
mainly accommodated on the acid functionality of HAP, thus
leaving behind the basic sites of the HAP matrix.
Effective acid properties of both HAP and Nb−HAP samples
were determined in water, and the results obtained are shown
in Table 3 and Figure 5. The PEA adsorption isotherms in
water on all the samples (HAP and Nb/HAP) do not have a
clear Langmuir-type shape, indicating low acid strength of the
sites. In water, a far lower amount of effective acid sites (E.A.)
than those measured in cyclohexane (I.A.) have been
determined in any case. In water, Nb/HAPca showed the
highest amount of effective acid sites. For the Nb/HAP
samples, the measurement of the effective basicity in water has
been performed, but no confident result has been determined
due to the extremely weak basic strength of the sites, which are
not able to manifest the basic property in water, a highly polar
and protic liquid.
475
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479

ACS Catalysis
Research Article
Table 4. Catalytic Performance of HAP and Nb−HAP
Materials in Liquid-Phase Isomerization of α-Pinene Oxide
a
c
selectivity (%)
b
conversion after 6 h of
reaction (%)
to
to 2 to 3 to 4 to 5 others
sample
HAP
25
35
33
68
23
38
39
22
9
2
5
19
10
20
59
30
36
29
4
11
10
13
Nb/HAPi
Nb/HAPCW
Nb/HAPCA
5
16
a
Conditions: toluene solvent; 35 mg cat; 0.66 mmol 1; 90 °C; batch
b
reactor. Conversion of α-pinene oxide (1) after 6 h of TOS.
c
Selectivity under iso-conversion conditions (typically 20%, for all
catalysts), to campholenic aldehyde (2), carveol (3), pinocarveol (4),
pinocamphone (5), and minor compounds (others).
Figure 5. Effective acidity (E.A.) measurements: isotherms of 2-
phenyl-ethylamine (PEA) adsorption in water on HAP and Nb/HAP
samples.
products for all the solids. This suggests that in low polar
aprotic solvents (the isomerization tests are run in toluene) the
nature of the acid-active sites on this class of materials is mainly
of Lewis type.
Then, due to the Nb-hosted in/on HAP structure, catalytic
activity of the Nb/HAP samples has been also investigated in
reactions involving saccharides (sucrose hydrolysis and glucose
isomerization) and in the selective oxidation reactions. For both
reactions, many Nb-containing systems have already been
recognized as good catalysts.39,42 In sucrose hydrolysis, when
inorganic solid catalysts are considered, the most active sites are
associated with Brønsted acid sites. Highly hydrated niobium
oxide48 and niobium phosphate57 have been recognized as
active and selective catalysts for sucrose hydrolysis.
2.3. Catalytic Activity. The acid character of HAP and Nb-
containing HAP samples was evaluated using the catalytic
rearrangement of α-pinene oxide in liquid phase, under batch
conditions, as a model test reaction (Scheme 1). In the
presence of acid sites, α-pinene oxide (1) typically isomerizes
into a series of different compounds: campholenic aldehyde
(2), carveol (3), pinocarveol (4), and pinocamphone (5), along
with other minor compounds. The strength and the nature of
the acid sites (either Lewis or Brønsted centers) have a direct
influence on the quality and the distribution of products
formed. Indeed, it is largely reported that the use of mild Lewis
acid sites favors the production of 2 with, as main side-
products, 4 and 5, while Brønsted acid sites enhance the
formation of 3, together with the main product 2.52−55
All the materials, HAP and Nb/HAP, showed a rather low
activity with respect to other heterogeneous acid materials
(Table 4). In fact, at 90 °C, the isomerization reaction is not
complete even after 24 h. On the contrary, for the sake of
comparison, a strong Brønsted acid catalyst, such as a protonic
synthetic saponite clay, and a Na-exchanged saponite clay have
shown, under fully comparable conditions, total (>98%) and
≈50% conversion of 1, respectively, at room temperature, in
only 5 min.56 In particular, the catalytic activity follows this
trend: Nb/HAPca > Nb/HAPi ≈ Nb/HAPcw > HAP. The
highest activity for Nb/HAPca is consistent with the highest
amount of effective acid sites, as reported in Table 4. The
isomerization of α-pinene oxide indeed can occur on either
Lewis or Brønsted acid sites, and it is roughly proportional to
the total amount of acid sites on the catalyst. With regard to the
selectivity, the formation of 3 was particularly remarkable over
Nb/HAPca, and this reveals the presence of some active
Brønsted acid sites, although not very strong, at the catalyst
surface. At the same time, however, 2 and 5 were the two major
HAP and Nb/HAP samples have been tested in this reaction,
performed under mild conditions (80 °C) in aqueous solvent.
All the samples are very poorly active (maximum conversion
observed ≈1%), indicating the absence of strong Brønsted acid
sites on both Nb-free HAP and Nb/HAP samples. The
obtained results are presented in Figure 6, in terms of product
concentration formed as a function of reaction time. The most
acidic sample (in terms of intrinsic and effective acidity, Nb/
HAPca) was also the most active one. The order of activity in
terms of reaction rate was Nb/HAPca (8.52 μmolsucrose·g−1·h−1)
> HAP (7.54 μmolsucrose·g−1·h−1) > Nb/HAPcw (5.01
μmolsucrose·g−1·h−1) > Nb/HAPi (3.46 μmolsucrose·g−1·h−1).
Glucose isomerization to fructose is a reaction that is
typically catalyzed by immobilized enzymes (xylose isomerase)
that generates an equilibrium mixture of fructose (42 wt %),
glucose (50 wt %), and other saccharides (8 wt %). Basic
alkaline-exchanged zeolites and hydrotalcites58,59 have been
proposed as alternative inorganic catalysts for this reaction, and
more recently, a solid Lewis acid catalyst in water was proposed
by Roman
́
-Leshkov et al.60 Catalysts with amphoteric proper-
ties, like HAP and Nb/HAP, are thus expected to show some
activity in glucose isomerization. Figure 7 summarizes the
Scheme 1. Acid-Catalyzed Isomerization of α-Pinene Oxide. α-Pinene Oxide (1), Campholenic Aldehyde (2), Carveol (3),
Pinocarveol (4) and Pinocamphone (5)
476
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479

ACS Catalysis
Research Article
dispersion in the matrix.24,62 The epoxidation reaction was
carried out under mild conditions (55 °C), and at the end of
the tests, aqueous hydrogen peroxide was not totally consumed
even on the most active catalyst. The calculated reaction rates
(calculated at 2 h TOS) are in the following ranking: Nb/
HAPcw (1.53 mmolC8·g−1·min−1) > Nb/HAPi (1.08 mmolC8·
g−1·min−1) > Nb/HAPca (0.76 mmolC8·g−1·min−1) ≈ HAP
(0.72 mmolC8·g−1·min−1). As expected, bare HAP was very
poorly active and selective. Nb/HAPcw, which has the highest
surface area and contained Nb highly dispersed, was the most
active sample. In consideration of selectivity to cyclo-octene
epoxide, the most acid strength sample, Nb/HAPca, showed
the most interesting selectivity (Figure 8).
Figure 6. Sucrose hydrolysis reaction: concentration profiles of
sucrose and formed products (glucose plus fructose) as a function of
reaction time over HAP and Nb/HAP samples (reaction solvent,
water; reaction temperature, 80 °C; catalyst mass, ≈0.5 g; sucrose
concentration, 0.2 M).
Figure 8. Comparison among samples in the cyclo-octene epoxidation
reaction; data have been calculated after 2 h of reaction (reaction
solvent, methanol; reaction temperature, 55 °C; catalyst mass, ≈0.6 g;
cyclo-octene concentration, 0.45 M, and H2O2, 0.45 M.
Analogously, the epoxidation of limonene with aqueous
hydrogen peroxide (Scheme 2) was also chosen as a benchmark
Figure 7. Glucose isomerization reaction: fructose selectivity against
glucose conversion over HAP and Nb/HAP samples (reaction solvent,
water; reaction temperature, 120 °C; catalyst mass, ≈0.5 g; glucose
concentration, 0.55 M).
Scheme 2. Epoxidation of Limonene to the Limonene Oxide
results obtained on all samples, in terms of fructose selectivity
against the glucose conversion. All the catalysts showed high
selectivity to fructose at low glucose conversion. At higher
conversion, a deactivation trend was observed, in particular on
Nb/HAPi and Nb/HAPcw, as already observed in the literature
starting from glucose conversion around 15%.53 On the
contrary, bare HAP was able to maintain very high fructose
selectivity when glucose conversion increased (above 20%).
The order of activity for the catalysts followed the trend: HAP
(6.54 mmolglucose·g−1·h−1) ≫ Nb/HAPcw (3.83 mmolglucose·g−1·
h−1) ≃ Nb/HAPi (3.41 mmolglucose·g−1·h−1) > Nb/HAPca
(1.03 mmolglucose·g−1·h−1). A balanced amount of acid and basic
sites seems to be necessary to guarantee a good activity, as
shown by HAP (I.A./I.B. = 0.71). The remarkable catalytic
performances of HAP seems promising and will merit attention
in our future experimental activity.
reaction to verify whether Nb(V) sites, in these Nb/HAP
solids, are accessible to a bulkier reactant. The catalytic features
are reported in Table 5 and in Figure 9, where the selectivity to
limonene epoxide is shown as a function of limonene
Table 5. Catalytic Performance of HAP and Nb−HAP
Samples in Liquid-Phase Epoxidation of Limonene in the
a
Presence of Aqueous H2O2
b
c
b
c
sample
conversion (%) yield (%) conversion (%) yield (%)
after 1 h of reaction after 6 h of reaction
Cyclo-octene epoxidation by aqueous hydrogen peroxide is
an oxidation reaction that gives rise to a stable peroxide which
does not isomerize easily. For this aim, isolated Nb metal
centers show interesting oxidation properties once incorporated
in/on inorganic oxide matrices.18,61 Mesoporous niobio-
silicates have been found to be effective catalysts for cyclo-
octene epoxidation, and their catalytic activity was found to be
strongly influenced by the Nb acid properties and Nb
HAP
12
29
19
20
3
25
39
19
40
10
Nb/HAPi
19
8
24
8
Nb/HAPCW
Nb/HAPCA
9
7
a
Conditions: CH3CN solvent; 100 mg cat; 1.0 mmol limonene; 2.0
mmol aq H2O2; reflux temperature; batch reactor. Limonene
conversion. Yield to limonene epoxide.
b
c
477
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479

ACS Catalysis
Research Article
presence of niobium is mainly associated with Lewis acid sites,
which gives a promising catalytic activity not only in acid-
promoted isomerization reactions but also in selective oxidation
reactions as well, in the presence of aqueous hydrogen
peroxide.
The obtained results presented here can open new routes
toward the application of this class of solid materials in various
reactions of applicative interest.
ASSOCIATED CONTENT
* Supporting Information
■
S
Scheme of the acid sites and basic sites on the HAP matrix;
experimental details on thermogravimetric analysis (TGA) of
the dried and calcined samples with all the thermograms
obtained; SEM and STEM images and EDX maps of calcium,
niobium, and phosphorus atom distributions; experimental
details on FT-IR spectroscopic analysis with spectra of typical
zones of IR absorption; experimental details on UV−vis-DRS
analysis with spectra of the samples (Kubelka−Munk function
vs energy); experimental details on surface acid and base
property determination and data interpretation; experimental
details on the reactions of sugars in aqueous solvent (sucrose
hydrolysis and glucose isomerization); Table with direct
comparison of the specific activity of the catalysts; results of
the recovery and reuse of the catalyst in the limonene
epoxidation test. This material is available free of charge via
the Internet at http://pubs.acs.org.
Figure 9. Selectivity vs conversion trend for the liquid-phase
epoxidation of limonene over HAP and Nb/HAP samples.
conversion. The role of Nb in promoting the epoxidation
reaction was evident, since Nb-free HAP showed the lowest
conversion value, after 1 h of reaction, with respect to the other
Nb/HAP catalysts. Nevertheless, in this reaction as well, the
presence of the niobium centers is not crucial in order to have
an effective epoxidation catalyst. Then, Nb/HAPi displayed the
most interesting performance, and the results were better, on
average, than the ones obtained on the other samples. Over
such catalyst, selectivity values as high as ≈70% were obtained
at substrate conversion values of ≈40%, with yields in limonene
oxide close to 30%. In all cases, however, the (mainly Lewis)
acid character of the catalysts was a disadvantage, since the
formed epoxide can easily undergo acid-catalyzed ring opening
and rearrangement. The epoxide selectivity values were, in fact,
poorer than the ones obtained, for instance, over other widely
used Nb−silica-based materials on the same substrate. For the
sake of comparison, yields in limonene oxide up to 75% can be
obtained, under the same conditions, in the presence of grafted
Nb−SiO2 catalysts.19,20The formation of undesired secondary
products was, in this case, particularly marked for Nb/HAPca,
and such behavior is consistent with the observation that Nb/
HAPca is the sample with the highest amount of effective acid
sites, as from Table 3 and from the direct comparison of the
specific activity of the catalysts (Table S.1).
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: antonella.gervasini@unimi.it. Fax: 0039 02 50314300.
Tel.: 0039 02 50314254.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
All the authors gratefully thank Dr. Matteo Marzo from
Dipartimento di Chimica of UNIMI for the experimental
support and Dr. Claudio Evangelisti from CNR-ISTM for the
HR-TEM analysis. A.G. acknowledges the UCSS de Lille-1,
́
France, Unite de Catalyse et Chimie du Solide, for performing
Finally, Nb/HAP catalysts were recovered, thoroughly
washed with the fresh solvent, pretreated again, and reused.
During the second catalytic run, the initial reaction rate was
slower, but then limonene epoxide yields as high as those of the
fresh catalyst were attained after 2−4 h of reaction (Figure S.7,
SI). In addition, hot-filtration tests have shown no further
reaction activity of the solid-free reaction mixture. No leaching
of homogeneously active Nb species was thus present.
the XPS analyses. C.T. and M.G. acknowledge the Italian
Ministry of Education, University and Research through the
Project “ItalNanoNet” (Rete Nazionale di Ricerca sulle
Nanoscienze; Prot. No. RBPR05JH2P) for economic support.
REFERENCES
■
(1) Elliott, J. C. Structure and Chemistry of the Apatites and Other
Calcium Orthophosphates; Elsevier: Amsterdam, 1994; Vol. 18, pp
111−189.
3. CONCLUDING REMARKS
(2) Hench, L. L. J. Am. Chem. Soc. 1991, 74, 1487−1510.
(3) Lamonier, C.; Lamonier, J.-F.; Aellach, B.; Ezzamarty, A.; Leglise,
J. Catal. Today 2010, 164, 124−130.
This study provides the first evidence that Nb centers can be
successfully added into (or onto) a hydroxyapatite scaffold. The
resulting Nb/HAP samples can be used as a new family of
amphoteric catalysts with tunable acid−base properties. The
preparation method used for the synthesis affects, to a large
extent, the surface properties of the Nb/HAP solids, and it has
a major influence, in particular, on the dispersion of niobium
species and on the acid to basic sites ratio. Marked differences
have been determined between the intrinsic acid properties of
the samples and their effective acid properties measured in
water that governs the catalytic activity in aqueous solvent. The
(4) Matsumura, Y.; Moffat, J. B. J. Chem. Soc., Faraday Trans. 1996,
92, 1981−1984.
(5) Dasireddy, V. D. B. C.; Singh, S.; Friedrich, H. B. Appl. Catal. A:
Gen 2013, 456, 105−117.
(6) Tsuchida, T.; Kubo, J.; Yoshihoka, T.; Sakuma, S.; Takeguchi, T.;
Ueda, W. J. Catal. 2008, 259, 183−189.
(7) Aellach, B.; Ezzamarty, A.; Leglise, J.; Lamonier, C.; Lamonier, J.-
F. Catal. Lett. 2010, 135, 197−206.
(8) Wakamura, M.; Kandori, K.; Ishikawa, T. Colloids Surf., A 1998,
142, 107−116.
478
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479

ACS Catalysis
Research Article
(9) Ogo, S.; Onda, A.; Yanagisawa, K. Appl. Catal. A: Gen 2011, 402,
188−195.
(40) Carniti, P.; Gervasini, A.; Biella, S.; Auroux, A. Catal. Today
2006, 118, 373−378.
(41) Carniti, P.; Gervasini, A.; Marzo, M. Catal. Today 2010, 152,
42−47.
(10) Matsumura, Y.; Moffat, J. B.; Sugiyama, S.; Hayashi, H.;
Shigemoto, N.; Soitoli, K. J. Chem. Soc., Faraday Trans. 1994, 90,
2133−2140.
(42) Gervasini, A.; Carniti, P.; Marzo, M.; Auroux, A. J. Catal. 2012,
296, 143−155.
(11) Wakamura, M.; Kandori, K.; Ishikawa, T. Colloids Surf., A 2000,
164, 297−305.
(43) Landi, E.; Tampieri, A.; Celotti, G.; Sprio, S. J. Eur. Ceram. Soc.
2000, 20, 2377−2387.
(12) Khachani, M.; Kacimi, M.; Ensuque, A.; Piquemal, J. Y.;
Connan, C.; Bozon-Verduraz, F.; Ziyad, M. Appl. Catal. A: Gen 2010,
388, 113−123.
(44) Bernache-Assolant, D.; Ababoua, A.; Champion, E.;
Heughebaert, M. J. Eur. Ceram. Soc. 2003, 23, 229−241.
(45) Lazic,
2001, 374, 13−22.
(46) Suguyama, S.; Hayashi, H. Int. J. Mod. Phys. B 2003, 17, 1476−
1481.
́ ́ ́
S.; Zec, S.; Miljevic, N.; Milonjic, S. Thermochim. Acta
(13) Boucetta, C.; Kacimi, M.; Ensuque, A.; Piquemal, J. Y.; Bozon-
Verduraz, F.; Ziyad, M. Appl. Catal. A: Gen 2009, 356, 201−210.
(14) Ogo, S.; Onda, A.; Yanagisawa, K. Appl. Catal. A: General 2008,
348, 129−134.
(47) Zhu, H.; Zheng, Z.; Gao, X.; Huang, Y.; Yan, Z.; Zou, J.; Yin, H.;
Zou, Q.; Kable, S. H.; Zhao, J.; Xi, Y.; Martens, W. N.; Frost, R. L. J.
Am. Chem. Soc. 2006, 128, 2373−2384.
(15) Onda, A.; Ogo, S.; Kajiyoshi, K.; Yanagisawa, K. Mater. Lett.
2008, 62, 1406−1409.
(16) Ziolek, M. Catal. Today 2003, 78, 47−64.
(17) Wachs, I. E.; Jehng, J.-M.; Deo, G.; Arora, N. Catal. Today 1996,
28, 199−205.
(18) Lacerda, V., Jr.; Araujo dos Santos, D.; da Silva-Filho, L. C.;
Greco, S. J.; Bezerra dos Santos, R. Adrichim. Acta 2012, 45, 19−27.
(19) Gallo, A.; Tiozzo, C.; Psaro, R.; Carniato, F.; Guidotti, M. J.
Catal. 2012, 298, 77−83.
(20) Tiozzo, C.; Bisio, C.; Carniato, F.; Gallo, A.; Scott, S. L.; Psaro,
R.; Guidotti, M. Phys. Chem. Chem. Phys. 2013, 15, 13354−13362.
(48) Sumiya, S.; Oumi, Y.; Sadakane, M.; Sano, T. Appl. Catal. A: Gen
2009, 365, 261−267.
(49) Dragone, L.; Moggi, P.; Predieri, G.; Zanoni, R. Appl. Surf. Sci.
2002, 187, 82−88.
(50) Carniti, P.; Gervasini, A.; Marzo, M.; Auroux, A. Thermochim.
Acta 2013, 567, 51−56.
(51) Marzo, M.; Gervasini, A.; Carniti, P. Catal. Today 2012, 192,
89−95.
(52) Kaminska, J.; Schwegler, M. A.; Hoefnagel, A. J.; van Bekkum,
H. Recl. Trav. Chim. Pays-Bas 1992, 111, 432−437.
(53) Wilson, K.; Renson, A.; Clark, J. H. Catal. Lett. 1999, 61, 51−55.
(54) Ravindra, D. B.; Nie, Y. T.; Jaenicke, S.; Chuah, G. K. Catal.
Today 2004, 96, 147−153.
(21) Guerrero-Per
243−251.
(22) Trejda, M.; Kujawa, J.; Ziolek, M.; Mrowiec-Bialon
Today 2008, 139, 196−201.
́
ez, M. O.; Banares, M. A. Catal. Today 2009, 142,
̃
́
, J. Catal.
(23) Coelho, J. V.; Oliveira, L. C. A.; Moura, F. C. C.; de Souza, P. P.;
Silva, C. A.; Batista, K. B.; Silva, M. J. Appl. Catal. A: Gen 2012, 419−
420, 215−220.
(55) Ravasio, N.; Zaccheria, F.; Guidotti, M.; Psaro, R. Topics Catal.
2004, 27, 157−168.
(56) Ostinelli, L.; Recchia, S.; Bisio, C.; Carniato, F.; Guidotti, M.;
Marchese, L.; Psaro, R. Chem. Asian J. 2012, 7, 2394−2402.
(57) Marzo, M.; Gervasini, A.; Carniti, P. Carbohydr. Res. 2012, 347,
23−31.
(24) Aronne, A.; Turco, M.; Bagnasco, G.; Ramis, G.; Santacesaria,
E.; Di Serio, M.; Marenna, E.; Bevilacqua, M.; Cammarano, C.; Fanelli,
E. Appl. Catal. A: Gen 2008, 347, 179−185.
(25) Carniti, P.; Gervasini, A.; Marzo, M. J. Phys. Chem. C 2008, 112,
14064−14074.
(58) Moreau, C.; Durand, R.; Roux, A.; Tichit, D. Appl. Catal. A: Gen
2000, 193, 237−264.
(26) Gao, X.; Wachs, I. E.; Wong, M. S.; Ying, J. Y. J. Catal 2001,
203, 18−24.
(59) Lecomte, J.; Finiels, A.; Moreau, C. Starch/Starke 2002, 54, 75−
̈
79.
(60) Roman
(27) Ziolek, M.; Sobczak, I.; Lewandowska, A.; Nowak, I.; Decyk, P.;
Renn, M.; Janowska. Catal. Today 2001, 70, 169−181.
́
-Leshkov, Y.; Moliner, M.; Labinger, J. A.; Davis, M. E.
Angew. Chem., Int. Ed. 2010, 49, 8954−8957.
(61) Nowak, I.; Ziolek, M. Chem. Rev. 1999, 99, 3603−3624.
(62) Gallo, J. M. R.; Paulino, I. S.; Schuchardt, U. Appl. Catal. A: Gen
2004, 266, 223−227.
́
(28) Corma, A.; Llabres i Xamena, X.; Prestipino, C.; Renz, M.;
Valencia, S. J. Phys. Chem. C 2009, 113, 11306−11315.
(29) Tanaka, M.; Shima, H.; Yokoi, T.; Tatsumi, T.; Kondo, J. N.
Catal. Lett. 2011, 141, 283−292.
(30) Havecker, M.; Wrabetz, S.; Krohnert, J.; Csepei, L.-I.; Nauman
̈
̈
d’Alnoncourt, R.; Kolen’ko, Y. V.; Girgsdies, F.; Schlogl, R.;
̈
Trunschke, A. J. Catal. 2012, 285, 48−60.
(31) Roussel, M.; Bouchard, M.; Bordes-Richard, E.; Karim, K.; Al-
Sayari, S. Catal. Today 2005, 99, 77−87.
(32) Glaum, R.; Welker-Nieuwoudt, C.; Dobner, C. K.; Eichelbaum,
M.; Gruchow, F.; Heine, C.; Karpov, A.; Kniep, R.; Rosowski, F.;
Schlogl, R.; Schunk, S. A.; Titlbach, S.; Trunschke, A. Chem. Ing. Tech.
̈
2012, 84, 1766−1779.
(33) Nguyen, T. T.; Aouine, M.; Millet, J. M. M. Catal. Commun.
2012, 21, 22−26.
(34) Dibenedetto, A.; Aresta, M.; Angelini, A.; Ethiraj, J.; Aresta, B.
M. Chem.Eur. J. 2012, 18, 10324−10334.
(35) Okumura, K.; Tomiyama, T.; Shirakawa, S.; Ishida, S.; Sanada,
T.; Arao, M.; Niwa, M. J. Mater. Chem. 2011, 21, 229−235.
(36) de La Cruz, M. H. C.; Rocha, M. S.; Lachter, E. R.; Forrester, A.
M. S.; Reis, M. C.; San Gil, R. A. S.; Caldarelli, S.; Farias, A. D.;
Gonzalez, W. A. Appl. Catal. A: Gen 2010, 386, 60−64.
(37) Okuhara, T. Chem. Rev. 2002, 102, 3641−3666.
(38) Nakaijma, K.; Baba, Y.; Noma, R.; Kitano, M.; Kondo, J. N.;
Hayashi, S.; Hara, M. J. Am. Chem. Soc. 2011, 133, 4224−4227.
(39) Carniti, P.; Gervasini, A.; Biella, S.; Auroux, A. Chem. Mater.
2005, 17, 6128−6136.
479
dx.doi.org/10.1021/cs4010453 | ACS Catal. 2014, 4, 469−479