Effect of the niobium state on the properties of NbSiBEA as bifunctional catalysts for gas- and liquid-phase tandem processes

Article abstract of DOI:10.1016/j.molcata.2016.06.024

NbSiBEA zeolites contained isolated framework mononuclear Nb(V) (Nb_0.7SiBEA) and a mixture of framework mononuclear and extra-framework polynuclear Nb(V) (Nb_2.0SiBEA) were prepared by two-step postsynthesis method as evidenced by XRD, NMR, DR UV–vis and FTIR. DR UV–vis showed that two types of framework mononuclear Nb(V) are present in Nb_0.7SiBEA, while Nb_2.0SiBEA mainly contained isolated mononuclear Nb(V) in the framework of zeolite and polynuclear Nb(V) in the extra-framework position. FTIR with pyridine and 2,6-di-tert-butylpyridine as probe molecules showed that major amount Lewis and weak Br?nsted acidic sites are formed by incorporation of niobium in the framework of zeolites as mononuclear Nb(V). The catalytic properties of Nb-containing zeolites were investigated in tandem processes of ethanol conversion into 1,3-butadiene (gas-phase) and synthesis of unsymmetrical ethers from aromatic aldehyde and aliphatic alcohol (liquid-phase). It has been found that Nb_0.7SiBEA catalyst, containing only isolated framework mononuclear Nb(V) is more active than Nb_2.0SiBEA in the conversion of ethanol and ethanol/acetaldehyde mixture into 1,3-butadiene, MPV reduction of crotonaldehyde with ethanol and etherification of 4-methoxybenzyl alcohol with 2-butanol. The higher specific activity (turnover number/frequency) of Nb_0.7SiBEA than Nb_2.0SiBEA catalyst has been revealed for gas- and liquid-phase tandem processes.

Full text of DOI:10.1016/j.molcata.2016.06.024

Accepted Manuscript
Title: Effect of the niobium state on the properties of
NbSiBEA as bifunctional catalysts for gas- and liquid-phase
tandem processes
Author: Pavlo I. Kyriienko Olga V. Larina Nataliia O.
Popovych Sergiy O. Soloviev Yannick Millot Stanislaw
Dzwigaj
PII:
S1381-1169(16)30246-1
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2016.06.024
MOLCAA 9930
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
18-4-2016
16-6-2016
25-6-2016
Please cite this article as: Pavlo I.Kyriienko, Olga V.Larina, Nataliia O.Popovych,
Sergiy O.Soloviev, Yannick Millot, Stanislaw Dzwigaj, Effect of the niobium
state on the properties of NbSiBEA as bifunctional catalysts for gas- and
liquid-phase tandem processes, Journal of Molecular Catalysis A: Chemical
http://dx.doi.org/10.1016/j.molcata.2016.06.024
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

1
Effect of the niobium state on the properties of NbSiBEA as bifunctional
catalysts for gas- and liquid-phase tandem processes
Pavlo
I.
Kyriienkoa,*,
Olga
V.
Larinaa,
Nataliia
O.
Popovycha,
Sergiy O. Solovieva, Yannick Millotb, Stanislaw Dzwigajb,*
a L.V.Pisarzhevsky Institute of Physical Chemistry of the NAS of Ukraine
31 Prosp. Nauky, 03028 Kyiv, Ukraine
b Sorbonne Universités, UPMC Univ Paris 06, CNRS, UMR 7197, Laboratoire de Réactivité de
Surface, F-75005, Paris, France
* Corresponding authors:
Pavlo I. Kyriienko, e-mail address: pavlo_kyriienko@ukr.net
Stanislaw Dzwigaj, e-mail address: stanislaw.dzwigaj@upmc.fr
1

2
Graphical Abstract
2

3
Highlights
Mononuclear and polynuclear Nb(V) were identified in NbSiBEA zeolites.
The state of niobium in NbSiBEA zeolite determined its catalytic properties.
Higher TOF and TON values were achieved on NbSiBEA with framework mononuclear Nb(V).
Bifunctional catalyst for tandem processes was designed using two-step postsynthesis method.
Abstract
NbSiBEA zeolites contained isolated framework mononuclear Nb(V) (Nb0.7SiBEA) and a
mixture of framework mononuclear and extra-framework polynuclear Nb(V) (Nb2.0SiBEA) were
prepared by two-step postsynthesis method as evidenced by XRD, NMR, DR UV-vis and FTIR.
DR UV-vis showed that two types of framework mononuclear Nb(V) are present in Nb0.7SiBEA,
while Nb2.0SiBEA mainly contained isolated mononuclear Nb(V) in the framework of zeolite
and polynuclear Nb(V) in the extra-framework position. FTIR with pyridine and 2,6-di-tert-
butylpyridine as probe molecules showed that major amount Lewis and weak Brønsted acidic
sites are formed by incorporation of niobium in the framework of zeolites as mononuclear
Nb(V). The catalytic properties of Nb-containing zeolites were investigated in tandem processes
of ethanol conversion into 1,3-butadiene (gas-phase) and synthesis of unsymmetrical ethers from
aromatic aldehyde and aliphatic alcohol (liquid-phase). It has been found that Nb0.7SiBEA
catalyst, containing only isolated framework mononuclear Nb(V) is more active than
Nb2.0SiBEA in the conversion of ethanol and ethanol/acetaldehyde mixture into 1,3-butadiene,
MPV reduction of crotonaldehyde with ethanol and etherification of 4-methoxybenzyl alcohol
with 2-butanol. The higher specific activity (turnover number/frequency) of Nb0.7SiBEA than
Nb2.0SiBEA catalyst has been revealed for gas- and liquid-phase tandem processes.
Keywords: niobium, zeolite, acidic sites, framework sites, tandem processes
3

4
1. Introduction
Niobium-containing materials are widely studied as solid catalysts for heterogeneous
processes including sustainable organic synthesis [1]. Nb2O5/SiO2 catalysts are active in sugar
dehydration [2–5], methanol oxidation [6], glycerol dehydration to acrolein [7] and oxidation
reactions with aqueous H2O2 [8–11]. Catalysts based on niobium oxides were also used in the
process of ethanol (EtOH) conversion to 1,3-butadiene (BD) [12,13].
Coordination state of niobium in a carrier matrix significantly influences the activity of
niobium-containing sites in a catalyst. Guidotti with co-workers [8] reported that Nb/SiO2
catalysts, prepared by a co-precipitation method, have higher activity and stability in alkenes
epoxidation in the presence of aqueous hydrogen peroxide due to the conservation of
coordination state and uniform distribution of Nb on the carrier surface in contrast to Nb/SiO2
prepared by deposition. NbBEA zeolites, obtained by direct hydrothermal synthesis, catalyze
Meerwein-Ponndorf-Verley (MPV) reduction and etherification processes more effectively than
samples prepared by impregnation [14]. The similar effect was observed for Zr-containing
systems in EtOH conversion to BD [15] i.e. greater specific catalytic activity of ZrBEA in
comparison with Zr/MCM-41 and ZrO2/SiO2 was caused by higher content of isolated Zr(IV)
sites in the ZrBEA catalyst.
We have recently shown [16] that two-step postsynthesis method of tantalum
incorporation in BEA zeolite framework as isolated mononuclear Ta(V) allows preparing highly
selective tantalum-silicate catalysts for the conversion of EtOH and acetaldehyde (AA) mixture
into BD. A clear distinction between postsynthesized and hydrothermally synthesized SnBEA
zeolites was showed in Baeyer-Villiger and MPV reactions [17,18]. Catalytic activity of
dealuminated Sn-containing BEA zeolite is supposed to be associated with the structural
difference of Sn-containing active sites, especially with the presence of frustrated Lewis acid-
base pairs with outersphere coordination of SiO-. Obviously, incorporation of metals in a
4

5
framework of dealuminated BEA zeolite is a more convenient way to form isolated metal species
in the silicate matrix.
This paper is aimed to study the effect of the niobium state present in BEA zeolite matrix
on the activity and selectivity of NbSiBEA catalysts in so-called tandem processes, requiring
different types of active sites (redox and/or acid-base). In particular, the Nb-containing catalysts
were tested in gas- and liquid-phase processes: (1) EtOH conversion to BD, and (2) direct
synthesis of unsymmetrical ethers from aromatic aldehyde and aliphatic alcohol (4-
methoxybenzaldehyde and 2-butanol).
2. Experimental
2.1. Catalysts preparation
A tetraethylammonium BEA was placed in a aqueous solution of 13 mol·L-1 HNO3 and
lasted at 353 K during 4 h to obtain a dealuminated organic-free BEA, as previously described
[19,20]. The obtained SiBEA (Si/Al = 1300) with vacant T-atom sites was recovered by
centrifugation, washed with distilled water and dried in air at 353 K during 24 h.
In order to introduce niobium ions into framework of BEA zeolite, two portion of 2 g of
SiBEA were stirred for 3 h at 353 K in 100 mL of isopropanol (dry) solutions containing 1.5 ×
10-3 and 4.4 × 10-3 mol·L-1 Nb(OC2H5)5 (Alfa Aesar, 99.999 %), respectively. Then, two
suspensions at pH = 5.6 and 2.5, respectively, were stirred in air at 353 K for 1 h to evaporate the
solvent. The resulting samples were washed in distilled water (three times), dried (in air at 353 K
for 24 h) and calcined (in flowing air at 723 K for 3 h). The obtained samples containing 0.7 and
2.0 wt % of Nb were labeled Nb0.7SiBEA and Nb2.0SiBEA, respectively.
2.2. Catalysts characterization
The chemical analysis of zeolites was performed with ICP atom emission spectroscopy at
the CNRS Centre of Chemical Analysis (Vernaison, France).
5

6
Textural properties of studied samples were determined by low-temperature (77 K) N2
sorption on an Autosorb-1 (Quantachrome, USA). The zeolites, before analysis, were treated
under vacuum at 383 K for 18 h. The specific surface area was calculated from the BET method
and the micropore volume was determined using Vt-plot method.
X-ray diffraction (XRD) patterns of the powder samples were recorded using D8
Advance (Bruker AXS GmbH, Germany) diffractometer with monochromatized Cu-Kα radiation
(nickel filter, λ = 0.15418 nm). The d302 spacing was calculated using equation:
nλ=2d·sinθ
where n is an integer (n = 1), λ is the wavelength of incident wave, d is the spacing between the
planes in the atomic lattice, and θ is an angle between the incident ray and the scattering planes.
NMR spectra were registered using Avance 400 spectrometer (Bruker, Germany). 29Si
MAS NMR spectra were recorded at 79.5 MHz using 7 mm (external diameter) zirconia rotors,
29
with CP (1H - Si CP-MAS NMR) and without (29Si MAS NMR). Chemical shifts of silicon
29
were measured relative to tetramethylsilane (TMS). Si MAS NMR spectra were obtained with
rotors spinning speed of 4 kHz, excitation pulse duration of 2.5 μs and recycle delay of 10 s.
Polydimethylsilane (PDMS) was used for setting the Hartmann-Hahn condition. The proton π/2
1
pulse duration, the contact time and recycle delay were 6.8 μs, 5 ms and 5 s, respectively. H
MAS NMR spectra were recorded using 4 mm zirconia rotors with a 90° pulse duration of 3 μs
and a recycle delay of 6 s.
Diffuse reflectance (DR) UV–vis spectra were recorded at ambient atmosphere with
polytetrafluoroethylene as reference using Cary 5000 spectrometer (Varian, USA) equipped with
a double integrator.
Fourier Transform Infrared Spectroscopy (FTIR) spectra were recorded at a spectral
resolution of 1 cm-1 on KBr pellets using a Spectrum One FTIR spectrometer (Perkin Elmer,
USA).
6

7
Transmission FTIR spectra of self-supported wafers were recorded using a Vertex 70
spectrometer (Bruker, Germany) accumulating 128 scans at a resolution of 2 cm-1. The wafers
were activated by calcination at 723 K for 2 h in flowing 2.5% O2 in Ar and then outgassed at
573 K (10-3 Pa) for 1 h before each measurement.
Analysis of acidic properties of the zeolites was performed by adsorption of pyridine and
2,6-di-tert-butylpyridine (DTBP) followed by infrared spectroscopy. Before analysis, the
samples were pressed at ~ 2 ton·cm-2 into thin wafers of ca. 12 mg·cm-2 and placed inside the IR
cell. FTIR spectra with pyridine or DTBP were recorded on a Spectrum One FTIR spectrometer
(Perkin Elmer, USA) accumulating 48 scans at a spectral resolution of 1 cm-1. The cell was
connected to a vacuum-adsorption apparatus allowing a residual pressure below 10-3 Pa. The
spectra were recorded under ambient conditions after pyridine desorption at 423, 523, 573 and
623 K and after DTBP desorption at 423 K. Before pyridine or DTBP adsorption, the samples
were outguessed (10-3 Pa) at 673 K for 1 h. All measured spectra were recalculated to a
“normalized” wafer weight.
2.3. Catalytic activity measurement
2.3.1. Gas-phase process
Catalytic activity tests were carried out at 1 bar in a fixed-bed flow (quartz) reactor with
inner diameter of 4 mm. Samples with grains of 0.25–0.5 mm (0.25 g,) were used. Before
reaction the samples were treated at 673 K under flowing argon for 1 h. EtOH or the mixtures of
EtOH/AA and EtOH/crotonaldehyde were fed into the catalytic reactor (weight hourly space
velocity (WHSV) = 0.8 greagents·gcat-1·h-1, with Ar as a carrier gas and a flow rate of 15 mL/min).
The reagents and obtained products were analyzed chromatographically (KristalLyuks 4000M,
MetaChrom, Russia). To identify CO, CO2 a thermal conductivity detector and a packed column
(10% NiSO4 on coal, 3 m × 3 mm) were used, for organic compounds –a flame ionization
detector and a capillary column (HP-FFAP, 50 m × 0.32 mm).
7

8
Total conversion (TC) of the reagents (EtOH, EtOH/AA and EtOH/crotonaldehyde mixtures),
selectivity to products (Sj) and BD yield (YBD) were calculated by the following formulas:
n0 
n
TC  

i
i
i
100, %
100, %
n0

nj
n0 
Sj 
n


i
i
YBD  TC SBD /100, %
where ni0 is the initial amount of C moles of the reagents, ni and n j are the amount of C moles
of the unreacted reagents i and product j in the stream of the reaction products.
Turnover frequency (TOF) was obtained by dividing the number of moles of reacted
reagent(s) or obtained product(s) with number of moles of Nb per one hour.
2.3.2 Liquid-phase process
Zeolite catalyst (100 mg) after activation (773 K, 2 h) was added to the stock solution of
4-methoxybenzaldehyde (1) (1.25 mmol) in 2-butanol (4 g). The reaction mixture was heated to
353 K with vigorous stirring. At various time intervals, aliquots were taken from the reaction
mixture and analyzed after separation of the catalyst by centrifugation. The progress of the
reaction was analyzed by a gas chromatograph (KristalLyuks 4000M, MetaChrom, Russia)
equipped with an FID detector and a capillary column (HP-FFAP) using mesitylene as an
internal standard for quantification. The reactants and products were identified on the basis of
coincidence of retention times with those of extra pure standards.
Catalytic activity was characterized by the conversion (C) of aldehyde 1 and selectivities
of products (Sk).
nA0ld.1  nAld.1
C 
100, %
nA0ld.1
nk
Sk 
100, %
nA0ld.1  nAld.1
8

9
where n0Ald.1 is the initial amount of C moles of aldehyde 1, nAld.1 and nk are the amount of C
moles of aldehyde 1 and product k in the reaction products.
Turnover number (TON) was calculated as number of moles of reacted aldehyde 1 (after
10 h) divided by number of moles of Nb.
For the etherification reaction of 4-methoxybenzyl alcohol (2) with 2-butanol, the stock
solution, containing alcohol 2 (2.5 mmol) in 2-butanol (4 g) was used. The progress of the
reaction was analyzed as described hereinabove.
3. Results and discussion
3.1. Prove for the incorporation of Nb in BEA zeolite
The samples prepared by postsynthesis method, hereafter referred to as Nb0.7SiBEA and
Nb2.0SiBEA, are white and contain 0.7 and 2.0 Nb wt %.
The textural properties of HAlBEA, SiBEA, Nb0.7SiBEA and Nb2.0SiBEA were studied
by means of low-temperature nitrogen sorption. Fig. 1 exhibits nitrogen sorption isotherms of all
samples. All of them are type I according to IUPAC. The values of BET specific surface area
and micropore volumes of HAlBEA, SiBEA, Nb0.7SiBEA and Nb2.0SiBEA were in the ranges of
460−491 m2 g−1 and 0.19–0.21 cm3 g−1, respectively (Table 1) characteristics of BEA zeolite.
XRD patterns of SiBEA, Nb0.7SiBEA and Nb2.0SiBEA are typical for BEA zeolite
(Fig. 2). Dealumination of BEA zeolite with HNO3 and following incorporation of niobium ions
in the framework of SiBEA zeolite as mononuclear Nb(V) does not affect structure crystallinity,
in line with earlier reports [21,22]. The shift of diffraction peak about 22.6º as a result of removal
of Al atoms from zeolite framework and incorporation of metal ions is used as prove for zeolite
matrix contraction/expansion [19,20,23]. The significant decrease of the d302 spacing from
3.943 Å (HAlBEA; 2θ = 22.55°) to 3.920 Å (SiBEA; 2θ = 22.67°) upon dealumination suggest
contraction of the BEA matrix. A strong increase of d302 spacing from 3.920 Å (SiBEA; 2θ =
22.67°) to 3.957 (Nb0.7SiBEA; 2θ = 22.44º) and 3.946 Å (Nb2.0SiBEA; 2θ = 22.50º) observed
9

10
after incorporation of Nb into SiBEA points out its expansion due to longer Nb–O bond length
(1.89 for tetracoordinated Nb(V) in BEA zeolite [14]) as compared with Si–O (typically 1.60–
1.65 Å in zeolites [24]). The lower increase of d302 value appeared for Nb2.0SiBEA than for
Nb0.7SiBEA and Nb1.0SiBEA, Nb1.5SiBEA, Nb1.6SiBEA [21,22] with only framework
mononuclear Nb(V) suggests that for former zeolite only a part of niobium was incorporated in
the zeolite framework.
The 29Si MAS NMR spectrum of SiBEA (Fig. 3) contains signals at -102.3, -111.0 and
-114.3 ppm. The peaks at -111.0 and -114.3 ppm are caused by framework Si atoms in Si(OSi)4
environment, located at different crystallographic sites [25]. The peak at -102.3 ppm is attributed
to Si atoms in Si(OH)(OSi)3 environment. It is confirmed by a significant increase of intensity of
1
the peak at -102.3 ppm when H – 29Si CP-MAS NMR technique is applied (Fig. 4), since the
signal of 29Si nuclei close to protons is preferentially enhanced.
After incorporation of niobium ions in SiBEA, the intensity of the peak at about
-102 ppm is noticeably decreased for Nb0.7SiBEA and Nb2.0SiBEA (Fig. 3). It confirms the
reaction between niobium ethoxide (Nb(OC2H5)5) precursor and silanol groups of vacant T-
atoms. However, the absence of further decrease of the peak at -102.2 ppm after introduction of
more amount of niobium (Nb2.0SiBEA) suggests that major part of niobium is not incorporated
in framework of Nb2.0SiBEA. It seems that these niobium ions are introduced in extra-
framework position of BEA zeolite as polynuclear Nb(V). In contrast, for Nb0.7SiBEA the peak
at -102.4 ppm prove that almost all niobium is incorporated as isolated mononuclear Nb(V).
The change of the Si(OSi)4 environment takes place upon incorporation of Nb ions in
SiBEA as deduced from the appearance of the well distinguished 29Si MAS NMR peaks at
-114.0 and -114.1 ppm for Nb0.7SiBEA and Nb2.0SiBEA, respectively (Fig. 3). In contrast, for
SiBEA only shoulder is present at around -114.3. It suggests that the incorporation of Nb in
SiBEA modify the silicon environment.
10

11
1
Concerning H MAS NMR spectra (Fig. 5), two main peaks are observed at 1.4 and
5.4 ppm for SiBEA, indicating the presence of protons of isolated (and/or terminal) SiO-H and
hydrogen bonded silanol groups at vacant T-atom sites, respectively [26,27]. A small peak at
3.3 ppm is due to protons of H-bonded silanol groups present in a second kind of
crystallographic sites. This shift is comparable with that observed at 3.20 ppm for H-bonded
silanol groups in silica and silicalite [28,29]. The disappearance of the peak at 5.4 ppm after
incorporation of Nb in SiBEA (Fig. 5) proves the reaction of niobium ethoxide precursor with
both H-bonded silanol groups. The 1H MAS NMR spectra of Nb0.7SiBEA and Nb2.0SiBEA allow
distinguishing the peak with high intensity at around 3.8–4.1 ppm and two other peaks with low
intensity at 2.0 and 1.2 ppm. The peaks with low intensity are probably due to protons of isolated
and terminal silanol, respectively [30], still present in Nb0.7SiBEA and Nb2.0SiBEA after
incorporation of Nb into the vacant T-atom sites of SiBEA.
The DR UV–vis spectrum of Nb0.7SiBEA demonstrates the presence of two absorption
bands at 220 and 234 nm (Fig. 6) corresponding to different mononuclear Nb(V) present in the
framework of BEA zeolite and are assigned to oxygen-tetrahedral Nb(V) ligand-to-metal charge
transfer transitions [21,22,31,32]. The absence of bands at around 330–450 nm assigned to
niobia-like phases [31,33] indicates that these form of Nb species are not present in Nb0.7SiBEA.
In the DR UV–vis spectrum of Nb2.0SiBEA the three bands at 228, 248 and 370 nm
(Fig. 6) appear corresponding to Nb(V) framework mononuclear Nb(V) (band at 228 nm), extra-
framework polynuclear Nb(V) (248 nm) and niobium oxides (370 nm) [6]. Lower intensity of
the band at 228 nm on Nb2.0SiBEA than the bands at 220 and 234 nm on Nb0.7SiBEA spectrum
reveals lower content of mononuclear Nb(V) in Nb2.0SiBEA than in Nb0.7SiBEA. The absence of
(d–d) transitions for both Nb2.0SiBEA and Nb0.7SiBEA in the range of 600–800 nm suggests that
Nb(IV) species are not present.
After treatment of TEABEA zeolite with acid nitric the bands attributed to the OH
stretching modes of AlO-H groups (3782 and 3665 cm-1) and Al-O(H)-Si groups (3608 cm-1)
11

12
[19,34] are not observed in the spectrum of SiBEA (Fig. 7), proving the full removal of
aluminium. The occurring of narrow bands at 3736 and 3705 cm-1 and a broad band at 3515 cm-1
related to isolated SiO-H groups and H-bonded SiO-H groups, respectively, proves the formation
of vacant T-atom sites in SiBEA, as reported previously [34]. The incorporation of niobium in
the zeolite induces a reduction of intensity of all these bands, suggesting the reaction of SiO-H
groups, with the niobium precursor. For Nb0.7SiBEA and Nb2.0SiBEA a band at 3745 cm-1
appears in the spectra, related to Nb(V)O-H.
The niobium incorporation in SiBEA modifies the Si environment as shown by changes
of the stretching vibration of Si–O [22]. The band at 955 cm-1 observed in FTIR spectrum of
SiBEA reveals the presence of Si-O group in vacant T-atom sites, as reported previously [20].
The incorporation of niobium into these sites modifies the Si–O stretching vibration that appears
at higher wavenumber (960 cm-1) on both Nb0.7SiBEA and Nb2.0SiBEA FTIR spectra than on
SiBEA spectrum (955 cm-1) (Fig. 8). The modifications of the band at 1223–1238 cm-1
characteristic of BEA zeolite and associated with the specific arrangement of tetrahedral Si(IV)
sites are due to the increase of the unit cell parameter upon incorporation of niobium into the
framework (XRD results, Fig. 2).
FTIR spectra of pyridine adsorbed on Nb0.7SiBEA and Nb2.0SiBEA are seen in Fig. 8.
The bands at 1445, 1578 and 1596 cm-1 correspond to weakly bonded pyridine [35], as proved
by their disappearance after desorption of pyridine at 523 K. In contrast, the bands appeared at
1450, 1600 and 1611 cm-1 correspond to pyridine coordinatively bonded to Lewis acidic sites
(LAS) (Nb(V) species) [36]). The intensity of absorption bands at 1611 and 1600 cm-1 for
Nb2.0SiBEA is comparable to Nb0.7SiBEA, which contains almost three times less niobium. It
suggests that major amount of LAS are formed by niobium introduced in the framework of
SiBEA as mononuclear Nb(V). The intensity of these bands for SiBEA is much lower than for
Nb0.7SiBEA and Nb2.0SiBEA, as reported earlier [21,37] and proves that they are related to
pyridine adsorbed on niobium.
12

13
The Lewis acidic sites (LAS) concentrations has been calculated from intensity of the
band at 1450 cm-1 corresponding to pyridine coordinately bonded to Lewis acidic sites observed
in FTIR spectra after adsorption of pyridine at room temperature and then desorption at different
temperatures 423, 523, 573 and 623 K. The results for Nb0.7SiBEA and Nb2.0SiBEA are shown
in Table 2. The similar concentration values obtained for Nb0.7SiBEA and Nb2.0SiBEA after
desorption of pyridine at different temperatures indicate that LAS are mainly formed upon
incorporation of niobium in the framework of SiBEA zeolite.
Brønsted acidic sites (BAS) were not identified in both Nb0.7SiBEA and Nb2.0SiBEA
samples (the band of pyridinium cations at 1550 cm-1 on FTIR spectra is absent, Fig. 9 A and B).
DTBP as stronger base than pyridine has been used to establish presence of weak BAS
preferentially on external surface. FTIR spectra of DTBP adsorbed on Nb0.7SiBEA and
Nb2.0SiBEA are presented in Fig. 10. According to previous works [38,39], the bands at 1615
and 1532 cm-1 are assigned to the protonated DTBP adsorbed on BAS. The intensity of these
bands is slightly higher for Nb0.7SiBEA than for Nb2.0SiBEA, indicating the higher amount of
weak BAS in former than in latter zeolite.
Thus, incorporation of niobium into the framework of SiBEA as mononuclear Nb(V) is
proved by XRD, DR UV-vis, NMR and FTIR. DR UV-vis study shows that two types of
framework mononuclear Nb(V) are present in Nb0.7SiBEA, while Nb2.0SiBEA contains both
framework mononuclear Nb(V) and extra-framework polynuclear Nb(V). The presence of Lewis
acidic sites, formed by framework mononuclear Nb(V), and weak Brønsted acidic sites was
identified by FTIR using pyridine and DTBP as probe molecules.
3.2. Catalytic properties of NbSiBEA zeolites in tandem processes
3.2.1. Gas-phase process: conversion of EtOH into BD
Scheme 1 shows consecutive steps of gas-phase process of EtOH conversion to BD, such
as AA formation from EtOH, aldol condensation of AA to acetaldol, dehydration of acetaldol to
13

14
crotonaldehyde, Meerwein–Ponndorf–Verley reduction of crotonaldehyde with EtOH to crotyl
alcohol and AA, and dehydration of crotyl alcohol to BD [40,41]. Scheme 1 also shows the main
side process, which is EtOH dehydration to diethyl ether (DEE) and then to ethylene.
Consequently, bifunctional catalyst capable of catalyzing both hydrogen transfer (redox process),
and aldol condensation and dehydration, is necessary for selective BD synthesis from ethanol
[42–45].
EtOH conversion. The catalytic performance of NbSiBEA zeolites has been shown in
Table 3. The main products of EtOH conversion are BD, AA, ethylene and DEE. Small quantity
of crotonaldehyde was identified. In the presence of Nb0.7SiBEA almost two times higher yield
of BD is achieved than on Nb2.0SiBEA due to greater EtOH conversion and selectivity to BD.
Another difference between two catalysts could be seen from the selectivities toward formation
of the products of EtOH dehydrogenation (AA) and dehydration (ethylene + DEE). Higher
residual amount of AA observed for Nb2.0SiBEA catalyst could indicate the lower quantity of
sites of aldol condensation on its surface. High selectivity to ethylene and DEE for Nb0.7SiBEA
testify to greater quantity of acidic sites, on which dehydration reaction occurs. It should be
pointed out that only trace amount of DEE was observed in the presence of Nb-free SiBEA
zeolite, which may be formed during ethanol etherification on weak Lewis acidic sites [36].
TOF values for NbSiBEA catalysts in the ethanol conversion have been shown in Table
4. Turnover frequency for Nb0.7SiBEA is several times higher than for Nb2.0SiBEA. Thus, it
seems that framework mononuclear Nb(V) efficiently catalyze the processes, which occur on
acidic sites (aldol condensation of AA, EtOH dehydration to ethylene and DEE). In addition,
TOF of the redox reaction of dehydrogenation for Nb0.7SiBEA is almost four times higher than
for Nb2.0SiBEA, which is consistent with the results obtained for alkenes epoxidation in the
presence of aqueous hydrogen peroxide [8].
EtOH/AA mixture conversion. The activity and selectivity of NbSiBEA catalysts were
evaluated in the conversion of ethanol and acetaldehyde mixture. Addition of AA (an
14

15
intermediate, which takes part in the step of aldol condensation) to EtOH will allow estimating
the effect of niobium location in the zeolite matrix on the activity of a catalyst in this step.
In the presence of Nb0.7SiBEA total conversion of EtOH/AA mixture, BD yield (Table 3)
and TOF (Table 4) are higher than for Nb2.0SiBEA. However, the selectivity to BD is
significantly higher for the Nb2.0SiBEA catalyst at 598 K (70.6 %) than for Nb0.7SiBEA (54.8 %)
under the same conditions. The selectivity of BD formation at 623 K for both catalysts is close.
Decrease in selectivity to BD in the case of Nb2.0SiBEA at higher temperature is attributed to the
increase in a rate of thermodynamically more favorable endothermic reaction of EtOH
dehydration to ethylene. Change of selectivity to ethylene for Nb2.0SiBEA catalyst from 11.9 %
to 29.5 % with increasing of the reaction temperature from 598 to 623 K and selectivity
preservation for Nb0.7SiBEA may indicate that dehydration process could occur on LAS, formed
by extra-framework Nb(V) species.
Significantly higher TOF values for the dehydration products (ethylene + DEE) is
observed on the catalyst, containing only framework mononuclear niobium (Nb0.7SiBEA),
compared with the sample, containing extra-framework octahedral Nb(V) (Nb2.0SiBEA) (Table
4). Thus, it is further evidence of higher activity of framework mononuclear Nb(V),
characterized by stronger acidic character than extra-framework octahedral niobium.
It should be noted that higher amount of crotonaldehyde is present in the products of
conversion of EtOH/AA mixture on Nb2.0SiBEA (Table 4). The substantial difference in the
selectivity to crotonaldehyde, probably, indicates that MPV reduction of crotonaldehyde with
EtOH, along with aldol condensation, become key step under the used experimental conditions.
Since Nb2.0SiBEA contains less amount of isolated framework mononuclear Nb(V) than
Nb0.7SiBEA, it can be assumed that MPV reduction of crotonaldehyde proceeds with higher rate
in the presence of such niobium species than Nb2O5 nanoparticles. A similar conclusion was
made for Zr(IV)-based systems [44].
15

16
EtOH/crotonaldehyde mixture conversion. Analogous method of intermediate addition
was applied in order to estimate the effect of niobium location in the zeolite matrix on the
activity of a catalyst in the step of crotonaldehyde MPV reduction to crotyl alcohol. Catalytic
performance of the NbSiBEA catalysts in the conversion of EtOH/crotonaldehyde mixture has
been shown in Table 5.
The total conversion of EtOH/crotonaldehyde mixture is higher for Nb0.7SiBEA than
Nb2.0SiBEA. The residual content of crotonaldehyde in the products for both catalysts is low and
differs slightly, however, amount of BD as well as ethylene (Table 6) and TOF values (Table 4)
are higher for Nb0.7SiBEA than for Nb2.0SiBEA. Higher TOF of BD formation than
crotonaldehyde reaction, especially for Nb0.7SiBEA catalyst, is a result of conversion of AA,
formed in MPV reduction of crotonaldehyde. Thus, the sample containing higher amount of
framework mononuclear Nb(V) is more active in the processes, which occur on the surface
acidic sites. These results confirm the assumption about the different activity of framework and
extra-framework Nb(V) in MPV reduction of crotonaldehyde with EtOH.
3.2.2. Liquid-phase process: synthesis of unsymmetrical ethers
NbSiBEA zeolites with different location of niobium were tested in the liquid-phase
tandem process (Scheme 2), including MPV reduction of aldehyde 1 with 2-butanol to 4-
methoxybenzyl alcohol (2), followed by its etherification with 2-butanol to 4-methoxybenzyl 1-
methylpropyl ether (3). As well as in the case of ethanol conversion into BD, the catalyst of the
process should be bifunctional.
Results of evaluation of catalytic properties of NbSiBEA zeolites in the liquid-phase
tandem process are shown on Fig. 11 and Table 6. Higher conversion of aldehyde 1 is observed
on the catalyst with greater amount of niobium (Nb2.0SiBEA). However, in the presence of the
catalyst, containing only framework Nb(V) species (Nb0.7SiBEA), aldehyde 1 is completely
converted into ether 3, while on Nb2.0SiBEA, containing framework and extra-framework
16

17
niobium species, 67 % selectivity to ether 3 is achieved (Table 7). Nb2.0SiBEA is more active in
MPV reduction than etherification process as evidenced by higher conversion of aldehyde 1 and
incomplete further etherification of alcohol 2.
Different activity of framework and extra-framework Nb(V) in the etherification was
confirmed by the results of the investigation of the reaction of alcohol 2 with 2-butanol in the
presence of Nb0.7SiBEA and Nb2.0SiBEA (Table 8). In the presence of Nb0.7SiBEA 93 %
conversion of alcohol 2 is achieved after 2 h, while for Nb2.0SiBEA only 37 % is observed after
10 h of the reaction. Moreover, calculated value of TON for Nb0.7SiBEA is significantly higher
compared to Nb2.0SiBEA. So, we may conclude that specific activity of framework mononuclear
Nb(V) is superior than extra-framework octahedral Nb(V).
NbSiBEA catalysts are more active in the etherification and tandem processes in
comparison with the zeolites prepared by impregnation of pure silica BEA with ethanolic
solutions of Nb(V) ethoxides [14]. Therefore, prepared by two-step postsynthesis method Nb-
containing SiBEA zeolite catalysts with framework mononuclear Nb(V) can be alternative
materials to hydrothermally synthesized metal-containing zeolites used as catalysts of liquid-
phase tandem processes.
4. Conclusions
Nb0.7SiBEA and Nb2.0SiBEA zeolites contain framework mononuclear Nb(V) and
mixture of framework mononuclear and extra-framework polynuclear Nb(V), respectively, as
evidenced by XRD, NMR, DR UV-vis and FTIR.
The state of niobium in zeolite matrix determines acidic properties and, thus, activity of
NbSiBEA catalysts in tandem processes, as was shown for gas-phase and liquid-phase processes.
Nb0.7SiBEA catalyst is more active than Nb2.0SiBEA in the conversion of ethanol and
ethanol/acetaldehyde mixture into 1,3-butadiene, MPV reduction of crotonaldehyde with ethanol
and etherification of 4-methoxybenzyl alcohol with 2-butanol.
17

18
The higher specific catalytic activity (turnover number/frequency) of Nb0.7SiBEA than
Nb2.0SiBEA has been revealed for the above mentioned processes.
Thus, two-step postsynthesis preparation procedure allows designing bifunctional
NbSiBEA catalysts with redox and acid-base properties required for the occurrence of tandem
processes.
Acknowledgements
Special thanks to Malgorzata Ruggiero (Jerzy Haber Institute of Catalysis and Surface Chemistry
PAS in Krakow, Poland) for BET measurements on HAlBEA, SiBEA, Nb0.7SiBEA and
Nb2.0SiBEA.
18

19
References
[1]
B. Henrique Arpini, A. de Andrade Bartolomeu, C. Kleber Z. Andrade, L. Carlos da
Silva-Filho, V. Lacerda, Curr. Org. Synth. 12 (2015) 570.
[2]
[3]
P. Carniti, A. Gervasini, M. Marzo, Catal. Today 152 (2010) 42.
E.L.S. Ngee, Y. Gao, X. Chen, T.M. Lee, Z. Hu, D. Zhao, Y. Ning, Ind. Eng. Chem. Res.
53 (2014) 14225.
[4]
[5]
[6]
[7]
Z. Xue, Y. Zhang, G. Li, J. Wang, W. Zhao, T. Mu, Catal. Sci. Technol. 6 (2016) 1070.
K. Subramaniyan, P. Arumugam, J. Porous Mater. 23 (2016) 639.
X. Gao, I.E. Wachs, M.S. Wong, J.Y. Ying, J. Catal. 203 (2001) 18.
C. García-Sancho, J.A. Cecilia, A. Moreno-Ruiz, J.M. Mérida-Robles, J. Santamaría-
González, R. Moreno-Tost, P. Maireles-Torres, Appl. Catal. B Environ. 179 (2015) 139.
[8]
C. Tiozzo, C. Palumbo, R. Psaro, C. Bisio, F. Carniato, A. Gervasini, P. Carniti, M.
Guidotti, Inorganica Chim. Acta 431 (2015) 190.
[9]
A. Gallo, C. Tiozzo, R. Psaro, F. Carniato, M. Guidotti, J. Catal. 298 (2013) 77.
[10] A. Feliczak-Guzik, A. Wawrzyńczak, I. Nowak, Microporous Mesoporous Mater. 202
(2015) 80.
[11] O.A. Kholdeeva, Catal. Sci. Technol. 7 (2014) 1869.
[12] V. V. Ordomskiy, V.L. Sushkevich, I.I. Ivanova, One-Step Method for Butadiene
Production, US 8921635 B2, 2014.
[13] W.J. Toussant, J.T. Dunn, D.R. Jackson, Ind. Eng. Chem. 39 (1947) 120.
[14] A. Corma, F.X. Llabrés I Xamena, C. Prestipino, M. Renz, S. Valencia, J. Phys. Chem. C
113 (2009) 11306.
[15] V.L. Sushkevich, D. Palagin, I.I. Ivanova, ACS Catal. 5 (2015) 4833.
19

20
[16] P.I. Kyriienko, O. V. Larina, S.O. Soloviev, S.M. Orlyk, S. Dzwigaj, Catal. Commun. 77
(2016) 123.
[17] J. Dijkmans, M. Dusselier, W. Janssens, M. Trekels, A. Vantomme, E. Breynaert, C.E.A.
Kirschhock, B.F. Sels, ACS Catal. 6 (2015) 31.
[18] J. Dijkmans, J. Demol, K. Houthoofd, S. Huang, Y. Pontikes, B. Sels, J. Catal. 330
(2015) 545.
[19] S. Dzwigaj, M.J. Peltre, P. Massiani, A. Davidson, M. Che, T. Sen, S. Sivasanker, Chem.
Commun. (1998) 87.
[20] S. Dzwigaj, P. Massiani, A. Davidson, M. Che, J. Mol. Catal. A Chem. 155 (2000) 169.
[21] F. Tielens, T. Shishido, S. Dzwigaj, J. Phys. Chem. C 114 (2010) 3140.
[22] S. Dzwigaj, Y. Millot, C. Méthivier, M. Che, Microporous Mesoporous Mater. 130
(2010) 162.
[23] M.A. Camblor, A. Corma, J. Pérez-Pariente, Zeolites 13 (1993) 82.
[24] T. Blasco, M.A. Camblor, A. Corma, P. Esteve, J.M. Guil, A. Martı, S. Valencia, A.
Martinez, J.A. Perdigon-Melon, J. Phys. Chem. B 102 (1998) 75.
[25] C.A. Fyfe, H. Strobl, G.T. Kokotailo, C.T. Pasztor, G.E. Barlow, S. Bradley, Zeolites 8
(1988) 132.
[26] G.L. Woolery, L.B. Alemany, R.M. Dessau, A.W. Chester, Zeolites 6 (1986) 14.
[27] L.W. Beck, J.F. Haw, J. Phys. Chem 99 (1995) 1076.
[28] C. Bronnimann, R. Zeigler, G. Maciel, J. Am. Chem. Soc. 110 (1988) 2023.
[29] A. Maciel, G.E.; Ellis, P.D. in: Bell, A.T.; Pines, NMR Techniques in Catalysis, 1994.
[30] M.A. Aramendia, Y. Avil, V. Borau, C. Jim, J.N.L. Marinas, A. Moreno, J.I.L. Ruiz,
React. Kinet. Catal. Lett. 65 (1998) 207.
20

21
[31] M. Hartmann, A.M. Prakash, L. Kevan, Catal. Today 78 (2003) 467.
[32] A.M. Prakash, L. Kevan, J. Am. Chem. Soc. 120 (1998) 13148.
[33] M. Trejda, A. Tuel, J. Kujawa, B. Kilos, M. Ziolek, Microporous Mesoporous Mater. 110
(2008) 271.
[34] A. Jentys, N.H. Pham, H. Vinek, J. Chem. Soc. Faraday Trans. 92 (1996) 3287.
[35] G. Connell, J.A. Dumesic, J. Catal. 105 (1987) 285.
[36] S. Dzwigaj, N. Popovych, P. Kyriienko, J.-M.J.-M. Krafft, S. Soloviev, Microporous
Mesoporous Mater. 182 (2013) 16.
[37] G. Centi, G. Golinelli, G. Busca, J. Phys. Chem. 94 (1990) 6813.
[38] T. Hibino, M. Niwa, Y. Murakami, Zeolites 13 (1993) 518.
[39] A. Corma, V. Fornes, L. Forni, F. Marquez, J. Martınez-Triguero, D. Moscotti, J. Catal.
179 (1998) 451.
[40] E. V Makshina, M. Dusselier, W. Janssens, J. Degrève, P.A. Jacobs, B.F. Sels, Chem.
Soc. Rev. 43 (2014) 7917.
[41] M. Gao, Z. Liu, M. Zhang, L. Tong, Catal. Letters 144 (2014) 2071.
[42] C. Angelici, M.E.Z. Velthoen, B.M. Weckhuysen, P.C.A. Bruijnincx, Catal. Sci. Technol.
5 (2015) 2869.
[43] O. V. Larina, P.I. Kyriienko, S.O. Soloviev, Catal. Letters 145 (2015) 1162.
[44] V.L. Sushkevich, I.I. Ivanova, E. Taarning, Green Chem. 17 (2015) 2552.
[45] V.L. Sushkevich, I.I. Ivanova, V. V. Ordomsky, E. Taarning, ChemSusChem (2014)
2527.
[46] C.A. Emeis, J. Catal. (1993) 347.
21

22
Fig. 1. Adsorption isotherms of nitrogen at 77 K for HAlBEA, SiBEA, Nb0.7SiBEA and
Nb2.0SiBEA. For convenience, the dataset for SiBEA, Nb0.7SiBEA and Nb2.0SiBEA were shifted
upwards along the Y-axis.
Fig. 2. XRD patterns of as-prepared HAlBEA, SiBEA, Nb0.7SiBEA and Nb2.0SiBEA.
Fig. 3. 29Si MAS NMR spectra of as-prepared SiBEA, Nb0.7SiBEA and Nb2.0SiBEA.
Fig. 4. 1H - 29Si CP MAS NMR spectra of as-prepared SiBEA, Nb0.7SiBEA and Nb2.0SiBEA.
Fig. 5. 1H MAS NMR spectra as-prepared SiBEA, Nb0.7SiBEA and Nb2.0SiBEA.
Fig. 6. DR UV–vis spectra of as-prepared Nb0.7SiBEA and Nb2.0SiBEA.
Fig. 7. FTIR spectra of HABEA, as-prepared SiBEA, Nb0.7SiBEA and Nb2.0SiBEA, outgassed at
573 K (10-3 Pa) for 2 h.
Fig. 8. FTIR spectra of as-prepared SiBEA, Nb0.7SiBEA and Nb2.0SiBEA.
Fig. 9. FTIR spectra of adsorbed pyridine on Nb0.7SiBEA (A) and Nb2.0SiBEA (B) catalysts after
desorption at different temperatures.
Fig. 10. FTIR spectra of adsorbed 2,6-di-tert-butylpyridine on Nb0.7SiBEA and Nb2.0SiBEA
catalysts after desorption at 423 K.
Fig. 11. Synthesis of unsymmetrical ether in the presence of Nb0.7SiBEA and Nb2.0SiBEA:
change of concentrations of 4-methoxybenzaldehyde (1), 4-methoxybenzyl alcohol (2) and 4-
methoxybenzyl 1-methylpropyl ether (3). Reaction conditions: aldehyde 1 (1.25 mmol) in 2-
butanol (4 g), 100 mg of catalyst, 353 K.
Scheme 1. Overall scheme of ethanol conversion to 1,3-butadiene.
Scheme 2. Overall scheme of synthesis of unsymmetrical ether.
22

23
1400
1200
1000
800
600
400
200
0
Nb2.0SiBEA
Nb0.7SiBEA
SiBEA
HAIBEA
0
0.1
0,1
0.2
0,2
0.3
0.4
0.5
0.6
0.7
0,7
0.8
0,8
0.9
0,9
1
0
0,3
0,5
1
Relativ0e,4Pressure0(,P6 /Po)
Relative Pressure (P/Po)
Fig. 1
23

24
22.50
22.44
Nb2.0SiBEA
Nb0.7SiBEA
22.67
22.55
SiBEA
AlBEA
5
10
15
20
25
30
35
40
45
(°)

Fig. 2.
24

25
Fig. 3
25

26
Fig. 4
26

27
Fig. 5
27

28
228
248
220
234
370
Nb2 . 0SiBEA
Nb0 . 7SiBEA
200
300
400
500
600
700
800
Wavelength (nm)
Fig. 6
28

29
3745
3530
Nb2.0SiBEA
3745
3530
3515
Nb0.7SiBEA
3736
3705
SiBEA
3608
3745
3520
3665
HAlBEA
3782
4000
3900
3800
3700
3600
3500
3400
3300
3200
Wavenumber (cm-1)
Fig. 7
29