Preparation and Characterization of Core–Shell Composite Zeolite BEA@MFI and Their Catalytic Properties in ESR

Article abstract of DOI:10.1007/s10562-018-2638-3

Abstract: A group of BEA@MFI core–shell zeolite catalysts were successfully constructed with Cu-, Ni-modified Beta core and Co-based Silicalite-1 shell by combining seed induction secondary growth technique and incipient wetness impregnation method. The r

Full text of DOI:10.1007/s10562-018-2638-3

Catalysis Letters
https://doi.org/10.1007/s10562-018-2638-3
Preparation and Characterization of Core–Shell Composite Zeolite
BEA@MFI and Their Catalytic Properties in ESR
1
2
1
1
1
1
1
1
Xing Li · Ziliang Zheng · Shiyao Wang · Chen Sun · Rong Dai · Xu Wu · Xia An · Xianmei Xie
Received: 10 August 2018 / Accepted: 15 December 2018
©
Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract
A group of BEA@MFI core–shell zeolite catalysts were successfully constructed with Cu-, Ni-modified Beta core and
Co-based Silicalite-1 shell by combining seed induction secondary growth technique and incipient wetness impregnation
method. The resulting micro-composite possess a well-defined core–shell structure with high surface area and abundant
mesopore in the outer shell, and a relative large truncated bi-pyramidal shape core of Beta. It was further used to correlate
their performance in ethanol steam reforming. The obtained xNiyCu-Beta@Co-Silicalite-1 catalysts exhibited excellent
catalytic performance compared with the conventional zeolite catalysts. Particularly, the 2.5Ni2.5Cu-Beta@CoSilicalite-1
catalyst exhibited the highest selectivity of hydrogen, the lowest selectivity of by-products, the long-term stability and
coking-resistance capability. This is mainly attributed to the excellent textural properties. And the Ni, Cu nanoparticles
which incorporated into the core–shell micro-composites protected the properties of active phases effectively. Moreover, the
“
purification effect” of the Co-base shell is conducive to obtaining high-purity hydrogen through a multiple reaction system.
Graphical Abstract
Keywords BEA@MFI · Polymetallic catalyst · Ethanol steam reforming · Hydrogen production
1
Introduction
With rising global demands for energy, diminishing avail-
ability of fossil fuels, and growing concerns on pollution
prevention, alternative and environment friendly energy
sources have captured much attention in recent research [1].
Under such a scenario, researchers find that hydrogen is a
clean energy carrier with convenient transportation and high
energy content (120.7 kJ/g). So, hydrogen as a kind of the
most promising energy has been widely used in fuel cells
for pollution-free generation of electricity [2]. Furthermore,
ethanol steam reforming (ESR, Eq. 1) is one of the highly
*
Xianmei Xie
xxmsxty@sina.com
1
College of Chemistry and Chemical Engineering, Shanxi
Key Laboratory of Gas Energy Efficient and Clean
Utilization, Taiyuan University of Technology,
Taiyuan 030024, Shanxi, China
2
Translational Medicine Research Center, Shanxi Medical
University, Taiyuan 030001, Shanxi, China
Vol.:(0
1
123456789)
3

X. Li et al.
attractive route due to the high H yield and the ethanol
contact area of the reactants greatly. As in previous reports
[17], it is well-known that Cu as an active catalyst phase
has the high capability to break the O-H bond, moreover, Ni
can improve the WGS and methane steam reforming (MSR)
reactions to a certain extent, which leads to a complete con-
version of ethanol and by-products on the core. For example,
Damyanova et al. [18] pointed out that the bimetallic CoNi/
Al O catalysts exhibited superior catalytic performance and
2
production mainly derived from biomass [3, 4].
C H OH + 3H O → 6H + 2CO
2
5
2
2
2
(1)
During the past decades, many studies have focused on
investigating different catalytic systems for ESR, including
noble metals (Rh, Ru, Pt and Pd) and non-noble metals (Cu,
Ni, Co) [5–7]. Although noble metal-based catalysts exhibit
outstanding activity and stability for ESR, their utilization
at large scales is limited by their high costs [8, 9]. Non-
noble metal-based catalysts, especially Co-based catalysts,
were widely employed in the reforming process deriving
from their low costs, relatively excellent capability for C–C
2
3
stability compared to the monometallic system. Shawn D.
Lin et al. [19] reported that the synergetic effect between
the Ni and Cu components is considered beneficial for ESR
[
20, 21].
Herein, it is expected that combining the two promotional
2
+
0
and C–H bond rupture. Moreover, both Co and Co have
been found to be active for ethanol dehydrogenation to acet-
strategies to further enhance the performance of products
high-purity hydrogen in ESR reaction. We designed Cu, Ni
as an auxiliary to modify Cobalt-based catalysts to improve
the interaction between active components. The core–shell
structure forms a multiple reaction system to protect active
ingredient and reduce the by-products’ amount in the
reformed gas. We have investigated the effect of core–shell
structure, reaction temperature and contact time on ethanol
0
aldehyde, while Co is highly active for C–C bond cleav-
age and water gas shift (WGS) reaction during ESR [10,
1
1]. However, Co-based catalysts have a tendency towards
rapid catalytic deactivation under reaction conditions and
the relatively low C–C bond-breaking activity is generally
believed to form a large number of intermediate products
and unreacted ethanol in the low temperature range, which
seriously affects the yield of hydrogen [12]. Hence, develop-
ing highly efficient and stable catalysts for ESR also remains
a challenge.
conversion and H selectivity towards the different products.
2
In addition, we also examined the deactivation behaviors of
the catalysts.
There are two approaches typically adopted to overcome
this problem, namely, either by using bimetal (or multi-
metal) catalysts or using active supports to enhance cata-
lytic performance. Among the variety of embedded materi-
als, core–shell structure particles present a class of special
materials attracted particularly increasing interest because
they combine advantages of core and shell with high surface
area and accessible pore channels, which are bring various
opportunities in applications [13], for example, Pirngru-
ber et al. [14] applied core–shell zeolite composite with
enhanced selectivity for the separation of branched paraffin
isomers. In addition, the successful synthesis of several zeo-
lite composites such as MFI/MEL, MFI/MFI, BEA/MFI and
MOR/MFI with core–shell structures has been reported for
different catalytic areas [15]. Recently, zeolites have drawn
growing attention as ideal supports due to their well-defined
channels, tunable acidity and basicity, and high thermal and
chemical stability [16]. Among numerous porous supports,
Silicalite-1 zeolite with MFI zeotype (two-dimensional ten-
member-ring) is an ideal catalyst basic support for ESR.
All-silica Beta zeolite (three-dimensional 12-ring pore sys-
tem) contains abundant microporous structure which slightly
larger than the dimensions of ethanol molecules can extend
the time of reaction and promote the ethanol molecule to
react fully on active sites. From the extensive researches,
the singular structure of Beta as core and Silicalite-1 as shell
should be an effective promotion scheme for ESR reaction,
because of their superior performance in improving the
2
Experimental
2.1 Catalyst Preparation
2
.1.1 Materials
The reactant materials used for the synthesis were tetraethy-
lorthosilicate (TEOS, Aldrich), fumed silica (SiO ·nH O,
2
2
9
9%), tetrapropylammonium hydroxide (TPAOH, 25% in
water, Aldrich), ammonium fluoride (NH F, 96%), poly-
4
diallyldimethylammonium chloride (PDDA, 20%, Aldrich),
anhydrous ethanol (99.7%), sodium hydroxide (99%, Ker-
mel), Ni(NO ) ·6H O (98%, Kermel), Cu(NO ) ·3H O
3
2
2
3 2
2
(
99.5%, Kermel), Co(NO ) ·6H O (98%, Kermel) and dis-
3 2 2
tilled water.
2
.1.2 Synthesis of Beta Crystals Loaded with Ni, Cu Metals
All-silica Beta crystals synthesized with fluoride ion as
mineralizing agent according to a literature procedure [22]
was used as core material. In a typical synthesis, 0.8 g
NH F was dissolved into 9 mL TAPOH solution with vig-
4
orous stirring for 30 min at 30 °C. Then, 2 g of fumed
silica was added into the mixture solution. After stirring
at the room temperature for 3 h, the gel was transferred
into a 100 mL Teflon-lined stainless-steel autoclaves at
1
40 °C for 7 days. After crystallization, the zeolite product
1
3

Preparation and Characterization of Core–Shell Composite Zeolite BEA@MFI and Their Catalytic…
was filtered, washed with distilled water until its pH value
was less than 8, and dried at 100 °C overnight. Finally,
the obtained sample was calcined at 550 °C for 6 h in a
fumace under air atmosphere. Subsequently, the series of
all-silica Beta core catalysts were prepared via incipient
wetness impregnation. The loading amount of Cu and Ni
were varied between 0 and 5 wt% by controlling the con-
centration of metal precursor containing Cu(NO ) ·3H O
2.2 Characterization
Powder X-ray diffraction (XRD) patterns were collected on
a Rigaku D/max-2500 instrument (40 kV, 100 mA) using
Cu/K radiation at a scanning rate of 2°/min, with the scan-
ning range of 5°–80°. Nitrogen adsorption–desorption
isotherms were recorded on a Micromeritics ASAP-2020
instrument. Before adsorption analysis, the samples were
degassed at 300 °C for 24 h. The specific surface areas were
evaluated using the Brunauer–Emmett–Teller (BET) equa-
tion and the amount of micro-pores was calculated by t-plot
analysis. The Barrett–Joyner–Halenda (BJH) method was
applied to analyze the mesoporous structure of the catalyst,
while the microporous structure was calculated by using
the t-plot approach. X-ray photoelectron spectroscopy
3
2
2
and Ni(NO ) ·6H O. After the impregnation, the obtained
3
3
2
wet powders were dried at ambient temperature over night
and calcined at 500 °C for 4 h with a heating rate of 5°C/
min.
2.1.3 Synthesis of Core–Shell Composites
(
XPS) measurements were performed on a V.G. Scientific
The negative surface charge of all-silica Beta core crystals
loaded with Ni, Cu metals were reversed using a 0.5 wt%
aqueous solution of PDDA and then the negatively charged
Silicalite-1 seeds were adsorbed. Silicalite-1 nanocrystals
were synthesized according to Ref [23]. Calcination of the
pretreated all-silica Beta core crystals provided a mate-
rial with seeds firmly fixed to the surface. The secondary
growth of the Silicalite-1 seeds adsorbed on the modified
core crystals was performed at 100 °C with a solution hav-
ing the molar composition 9(TPA) O:25SiO :1500H O:1
ESCALAB250 using Al Kα radiation (1486.6 eV, 150 W).
The recorded spectrum was fitted to a product of Gauss-
ian–Lorentzian function by a least square procedure. After
coating with a thin layer of gold, the morphologies and size
of the samples were observed using a SU-500 (Hitachi,
Tokyo, Japan) scanning electron microscope (SEM). Trans-
mission electron microscopy (TEM) and high-resolution
transmission electron microscopy (HR-TEM) images were
obtained using TF 20, JOEL 2100F instrument with an oper-
ating voltage of 80–200 kV. The sample powders were dis-
persed as a suspension in ethanol solutions and sonicated for
3 min. After being well dispersed, a droplet of suspension
was deposited on copper grids and followed by drying in air.
H temperature programmed reduction (H -TPR) was per-
2
2
2
0
0
0EtOH. After cooling, the composite was treated with a
.1 M ammonia solution in an ultrasonic bath for 10 min
to remove the loosely attached Silicalite-1 crystals, rinsed
repeatedly with distilled water and dried at 80 °C. The
composite was calcined at 550 °C for 6 h to obtain Beta-
Silicalite-1 core–shell materials.
2
2
formed using a FINESORB-3010 apparatus equipped with
a thermal conductivity detector (TCD). The analysis was
carried out by increasing the catalyst temperature from 100
to 850 °C at 5°C/min under a He flow.
The catalysts were obtained by incipient wetness
impregnation (IWI) method. In brief, 1 g of sample was
mixed with an aqueous solution of Co(NO ) ·6H O precur-
3
2
2
sors, the mixture solutions were then stirred at 60 °C until
removed all the water. Furthermore, the synthesized cata-
lysts were air-dried overnight and calcined at 550 °C for
2.3 Catalytic Performance Testing
Catalytic activity tests were conducted in a fixed-bed stain-
less steel reactor at atmospheric pressure. Typically, the cat-
6
h with a heating rate of 2°C/min. The final products were
designated as xNiyCu-Beta@Co-Silicalite-1. In the arti-
cle, the “xNiyCu-Beta@Co-Silicalite-1” catalyst nomen-
clature of xNiyCu-Beta@CoSili, where x and y represent
the nominal Ni and Cu loading (wt%), respectively, xNi-
yCu-Beta and CoSili denotes the Ni-, Cu-modified Beta
core and 10 wt% Co on the Silicalite-1 shell. As reference
materials, the 2.5Cu2.5NiCo-Beta and 2.5Cu2.5NiCo-
Silicalite-1 catalysts were prepared by IWI method under
the similar conditions as the xNiyCu-Beta@CoSili sample.
The contents of the Cu and Ni were both 2.5 wt%, and
Co was 10 wt%. The 2.5Cu2.5Ni-Beta-CoSilicalite-1 was
used by the 2.5Cu2.5Ni-Beta and CoSilicalite-1 mechani-
cal mixing.
alyst (0.3 g, 20–40 mesh) diluted with SiO (dilution ratio
2
was 3) was loaded and sandwiched by two quartz wool beds.
Before the reaction, catalysts were reduced with a H /N
2
2
mixture (70 mL/min, 30 vol% H ) at 550 °C for 2 h, followed
2
by a N purging the reaction system. After which, a mixture
2
of C H OH/H O molar ratio of 1:3 was injected into the
2
5
2
vaporizer (150 °C) by a syringe pump at a rate of 0.07 mL/
min. Nitrogen was used as carrier gas in the reaction. The
reaction products were monitored online using a GC 950 gas
chromatograph equipped with two detectors. Flame ioniza-
tion detector (FID) and thermal conductivity detector (TCD)
with a Porapak-Q column using Ar as a carrier gas to ana-
lyze the organic species, including H , CO , CO and CH .
2
2
4
For the separation and quantification of the liquid products,
1
3

X. Li et al.
another gas chromatography (GC-9890B) with a Porapak Q
capillary column connected to a FID detector was applied.
The catalytic performance was evaluated through the conver-
with diameters of nearly 12 µm (Fig. 1a). Meanwhile, a tab-
let-like structure is observed for Silicalite-1 nanoparticles
with the size of about 90 nm (Fig. 1b). After coating process,
SEM image shows that the core catalyst is covered well by
a dense Silicalite-1 shell and the core–shell particles are
rough (Fig. 1c) with protrusions similar to those of the Beta
core. Compared with single Silicalite-1 zeolite, the crys-
tals formed shell Silicalite-1 on the core surface are much
smaller. This can be explained by the fact that the steric
hindrance provoked by the concurrently growing crystals in
the shell [13]. Beyond that, the section images of the micro-
composite (Fig. 1d) shows that the surfaces of the catalyst
has a small quantity of aggregates and fragments, which may
be due to the fact that the Silicalite-1 shell was destroyed by
the loading process for Co [24].
sion of ethanol (X
), and the selectivity (S ) towards the
EtOH
i
products (i=H , CO , CO, CH or CH CHO):
2
2
4
3
(
moles EtOH − moles EtOHout)
in
XEtOH (%) =
× 100 (2)
moles EtOHin
moles Pi
Si (%) =
× 100
n
∑
(3)
moles Pi
i=1
where S represents the molar amounts of the different
i
components in the products.
The core/shell interface morphologies of 2.5Cu2.5Ni-
Beta@CoSili sample were carefully investigated by TEM
and HR-TEM images as depicted in Fig. 2. Clearly, Fig. 2a,
b shows representative TEM images, where the uniform
growth of the Silicalite-1 shell on the external surface of
Beta zeolite and the shell with an approximately thick-
ness about 50–80 nm can be seen. HR-TEM analysis of the
sample shows the presence lattice fringes and the ordered
channels of Beta (Fig. 2c) with apparent distinction at the
core/shell interface. The close contact and the absence of
3
Results and Discussion
3.1 Characterization of Catalysts
SEM images of the as-synthesized materials are showed in
Fig. 1. Obviously, a truncated bi-pyramidal shape Beta zeo-
lite as the core materials displayed uniform morphologies
Fig. 1 SEM images of a Beta, b Silicalite-1, c low and d high magnifications of the reduced 2.5Cu2.5Ni-Beta@CoSili catalyst
1
3

Preparation and Characterization of Core–Shell Composite Zeolite BEA@MFI and Their Catalytic…
Fig. 2 TEM image of reduced
2
.5Cu2.5Ni-Beta@CoSili at dif-
ferent magnification. The TEM
(
a, b) and HR-TEM (c. d) image
were taken from the area in the
BEA/MFI interface
an intermediate layer between the Beta zeolite and the Sili-
talite-1 frameworks can be better seen in Fig. 2d. There is
not any structural obstacle that could prevent the commu-
nication between the shell and core channel systems. It can
further prove that the core–shell materials were successfully
constructed and guarantee the integrity of each molecular
sieve structure.
Diffraction patterns of the catalysts are presented in
Fig. 3. The xCuyNi-Beta@CoSili and the 2.5Ni2.5Cu-Beta-
CoSilli samples present typical diffraction peaks (7.5°–9.5°
and 23°–24°) of MFI-type material together with the peaks
(
at 7.6°, 21.2°, and 22.5°) characteristic of the BEA-type
framework topology, compared with good crystallinity phase
.5Cu2.5Ni10Co-Silicalite-1 and 2.5Cu2.5Ni10Co-Beta
2
catalysts. It can be indicate that the MFI, BEA structure are
well maintained after the coating process, which is consist-
ent with above mentioned SEM and TEM characterization.
The intensity of the diffraction peaks became weaker than
the 2.5Ni2.5Cu-Beta-CoSilli catalyst, simply because of the
shield effect of the Silicalite-1 shell in xCuyNi-Beta@CoSili
Fig. 3 XRD patterns of the pre-reduction catalysts
as the characteristic peaks of Co O . Two diffraction peaks
3
4
[
25]. For all the samples, the reflection peaks at 2θ of about
1.5°, 36.9°, 44.5°, 59.5° and 77.5°, which can be assigned
are observed at 2θ = 35.7°, 38.9° in 5Cu-Beta@CoSili,
3
which correspond well to CuO. Similarly, the two peaks at
1
3

X. Li et al.
2
θ=44.9° and 63.5° found in 5Ni-Beta@CoSili associated
intercrystalline mesopores formed by stacking of the shell
Silicalite-1 nanoparticles [25], which is in accordance with
the SEM image. The textural property data for all the sam-
ples are listed in Table 1. Apparently, the surface area and
the total pore volume of the xCuyNi-Beta@CoSili catalysts
are slightly higher than 2.5Cu2.5NiCo-Beta catalyst, indicat-
ing that the formation of shell under alkaline condition can
not destroyed the framework of silicon in the sample [28]. It
should be noted here that because Silicalite-1 nanoparticles
were coated on Beta core, the effect of stacking pore and the
second hysteresis loop were decreased, as shown in Fig. 4.
As could be expected, for the core–shell structure materi-
als these values are lower with respect to the single zeolite
catalyst, which is benefit to demonstrate that the core Beta
crystal is well enwrapped by the Silicalite-1 shell [13].
with NiO. However, the reflection peak of CuO and NiO
are less obvious for 2.5Cu2.5Ni-Beta@CoSili catalyst, pos-
sibly ascribe to the CuO and NiO species were low loading
amount and widely spread throughout the support [26].
The N adsorption desorption isotherms of the cata-
2
lysts are exhibited in Fig. 4. 2.5Cu2.5Ni10Co-Beta gave a
typical type I isotherm at a low relative pressure, which is
characteristic of a microporous zeolite materials, suggest-
ing that the microporous structures were largely maintained
after the incorporation of the Cu, Ni and Co. The intrinsic
mesoporosity of Silicalite-1 gave rise to a higher proportion
of mesopores in 2.5Cu2.5Ni10Co-Silicalite-1 and led to a
larger range of hysteresis loop, which stemmed from the
Silicalite-1 shell. The xCuyNi-Beta@CoSili samples dis-
play a combination of type I and type IV isotherms, with
micro-pore filling at low pressures and hysteresis loops in
H -TPR technique was employed to investigate the
2
reduction behavior of the catalysts as shown in Fig. 5. The
2.5Ni2.5Cu-Beta@CoSili TPR curve included four broad
peaks in the range of 238–564 °C. According to the litera-
ture [29], it has been generally accepted that the two main
reduction peak at 288 °C and 386 °C are associated with
the reduction of Co O proceeds in a two-step process,
this isotherm ranging from P/P =(0.4–0.65) and (0.9–1.0),
0
indicating formation of a hierarchical porous system com-
bining microporosity and mesoporosity. These hystere-
sis loops are usually associated with capillary filling and
condensation of N within the homogeneous slit-shaped
2
3
4
which are Co O to CoO, and CoO to Co [30]. The low
3
4
Fig. 4 N adsorption desorption isotherms at 77K of the reduced cat-
2
alysts
Fig. 5 TPR profiles of the pre-reduction catalysts
Table 1 The textural properties of the reduced catalysts
2
2
2
3
3
3
Sample
SBET (m /g) Smicro (m /g) Smeso (m /g) Vtol (cm /g) Vmicro (cm /g) Vmeso (cm /g) Dmicro (nm) Dmesro (nm)
2
2
2
2
5
5
.5Cu2.5NiCo-Beta
.5Cu2.5NiCo-Silicalite-1 322
.5Cu2.5Ni-Beta-CoSili 265
.5Cu2.5Ni-Beta@CoSili 307
233
191
121
122
176
200
220
42
201
143
131
106
84
0.17
0.27
0.23
0.20
0.21
0.21
0.1
0.07
0.21
0.17
0.1
0.11
0.1
0.6
0.49
0.5
0.49
0.55
0.56
2.3
3.09
2.9
2.71
2.55
2.44
0.06
0.06
0.1
0.1
0.11
Cu-Beta@CoSili
Ni-Beta@CoSili
306
304
1
3

Preparation and Characterization of Core–Shell Composite Zeolite BEA@MFI and Their Catalytic…
temperature with a broad peak at 243 °C reduction behavior
may be explained considering the reduction of CuO spe-
cie, yet NiO formation was hardly observed due to its the
low loading amount or the overlapping reduction of NiO
and CoO. Almost similar profile were obtained over 5Ni-
Beta@CoSili and 5Cu-Beta@CoSili catalyst, moreover, for
the 5Ni-Beta@CoSili catalysts, one broad peak at 473 °C
was observed, which was assigned to the reduction of NiO
to Ni. The other broad peak at 272 °C can be seen over
the 5Cu-Beta@CoSili catalyst, which could be ascribed to
the reduction of CuO. Interestingly, from the comparison of
2p1/2, respectively. The two pairs of 2p1/2 and 2p3/2
3
+
2+
peaks indicate the coexistence of both Co and Co
[34, 35]. The strong 2p3/2 and 2p1/2 shake-up satellite
peak indicate that the surface cobalt is largely presented
as CoO [36].
3.2 Catalytic Activity in ESR
The initial catalytic activity tests of all catalysts in the ESR
were performed from 350 °C to 600 °C, aiming to study the
effect of reaction temperature on each catalyst. Since the etha-
nol conversion and product selectivity are the main parameters
to be taken into account for the ESR catalysts, here both of
them were showed in Figs. 7 and 8, respectively. As shown
in Fig. 7, it can be clearly seen that the ethanol conversion
increases with the increase of reaction temperature. Com-
pared to the 2.5Cu2.5NiCo-Beta, 2.5Cu2.5NiCo-Silicalite-1
and 2.5Cu2.5Ni-Beta-CoSili catalysts, all the xCuyNi-Beta@
CoSili catalysts exhibited higher ethanol conversion, which
2
.5Cu2.5Ni-Beta@CoSili and 5Cu-Beta@CoSili, it could
be found that the reduction peak of CuO is slightly shifted
to a lower temperature with the addition of the Ni phase. It
is possible to conclude that the reduction of Cu species is
promoted by the presence of Ni [31]. However, on account
of the effects of particle dispersion degree and strength of
interaction with the support on the reduction kinetics, the
number and location of TPR peaks can vary significantly
[
30, 32, 33]. For example, the H consumption peaks were
2
also observed for 2.5Ni2.5Cu-Beta-CoSili catalyst, all the
peaks shift to lower temperatures than xNiyCu-Beta@CoSili
significantly, indicating the strengthened metal-support
interaction for the core–shell catalyst [6]. In other words,
the core–shell structure 2.5Ni2.5Cu-Beta@CoSili catalyst
exhibits strengthened interaction between active metal Co
and support and the highly dispersed Ni, Cu to achieve large
interface.
To verify the electronic structure and chemical valence
of metals, the XPS spectra for the 2.5Cu2.5Ni-Beta@
CoSili catalyst before reduction was exhibited in Fig. 6.
The high-resolution spectra of Ni 2p (Fig. 6a) and Cu
2
p (Fig. 6b) are hardly detectable for the sample, which
is further demonstrate that the 2.5Cu2.5Ni-Beta core is
well-coated by the Silicalite-1 shell, this is in agree with
the SEM and TEM. The XPS spectra of Co 2p obtained
in Fig. 6c are centered at 780.8 eV and 796.5 eV, cor-
responding to the binding energies of Co 2p3/2 and Co
Fig. 7 Temperature effect on the C H OH conversion over the cata-
2
5
lysts
a
b
c
Co2P3/2
Co2p1/2
9
65
960
955
950
945
940
935
930
885
880
875
870
865
860
855
850
845 810
805
800
795
790
785
780
775
Binding Energy (eV)
Binding Energy (eV)
Binding Energy (eV)
Fig. 6 X-ray spectra of the a Ni2p, b Cu2p and c Co2p region for the 2.5Cu2.5Ni-Beta@CoSili catalyst before reduction
1
3

X. Li et al.
Fig. 8 Products selectivity for the ESR reaction catalyzed with a 2.5Cu2.5Ni-Beta@CoSili, b 2.5Cu2.5Ni-Beta-CoSili, c 2.5Cu2.5NiCo-Sili-
calite-1, d 2.5Cu2.5NiCo-Beta, e 5Cu-Beta@CoSili, f 5Ni-Beta@CoSili catalysts
reached above 80% at 350 °C and almost 100% conversion
around 500 °C. Among them, for the 2.5Cu2.5Ni-Beta@
CoSili catalyst the ethanol conversion has been reached almost
100% at the reaction temperature of 450 °C. It could be related
to the particular core–shell topology structure of the support
1
3

Preparation and Characterization of Core–Shell Composite Zeolite BEA@MFI and Their Catalytic…
and an appropriate amount of metal and the synergetic effect
between the Ni, Cu and Co components [28].
According to Fig. 8b, 2.5Cu2.5Ni-Beta-CoSili catalyst
show the same tendency at any temperatures for product
distributions compared to that of the2.5Cu2.5Ni-Beta@
CoSili. Nonetheless, the by-products selectivity gradually
For the sake of clarity, the catalytic performance data of all
the samples were measured under the same conditions. As is
vividly depicted in Fig. 8, major products are H , CO, CO ,
increases and the selectivity toward H was relatively lower.
2
2
2
CH and CH CHO for all the catalysts in the study. The cata-
On the other hand, the three metals supported on the catalyst
were to constitute a competitive route, resulting in a higher
selectivity for by-products. It can be a good indication that
the core–shell structure can able to efficiently promote the
4
3
lytic performance of the 2.5Cu2.5Ni-Beta@CoSili (Fig. 8a)
was superior to that of other samples, associated with fewer
by-products at the same temperature. At 350 °C, an impor-
tant quantity of CH CHO (19.5%) and CH (10.9%) could be
H selectivity and reduced the production of by-products.
3
4
2
present in products, with a consequent selectivity for H only
In order to further clarify the necessity of the core–shell
structural support in reaction, the catalytic performance of
2.5Cu2.5NiCo-Silicalite-1 (Fig. 8c) and 2.5Cu2.5NiCo-Beta
(Fig. 8d) sample with the same amount of metals and single
zeolite support were measured under the same conditions.
The undesired product distributions were higher than the
values for the 2.5Cu2.5Ni-Beta@CoSili, and a relatively low
2
5
7.8%. It could be related to either the dehydrogenation of
ethanol (Eq. 4) or the decomposition of acetaldehyde (ACD)
Eq. 5) as the main reaction route at this temperature. This
(
demonstrated that the dehydrogenation of ethanol is the first
reaction step in ESR reaction, and acetaldehyde is an inter-
mediate product in the reaction pathway. As the temperature
is raised, CO concentrations decrease. At 400 °C, the con-
tent of CO is reduced to 4%, indicating that the WGS reaction
selectivity to H . The 2.5Cu2.5NiCo-Silicalite-1 exhibited
2
higher CH and CH CHO selectivity in low temperature and
4
3
(
Eq. 6) has higher performance under the reaction conditions
high CH at high temperature. For 2.5Cu2.5NiCo-Beta, the
4
[
31]. Meanwhile, the increases of H and CO were ascribed to
acetaldehyde content is high at low temperature, and as the
temperature is increased to 600 °C, acetaldehyde was still
present. These results reveal that the three metals are sup-
ported on the same support to constitute a competitive route,
and the properties of each metal for the ESR reaction cannot
be fully displayed in this catalyst, resulting in a high selec-
tivity for by-products. In order to further understand the cat-
alytic properties of Cu and Ni phases in improving catalytic
activity, we prepared the monometallic core–shell structure
catalysts (5Ni-Beta@CoSili and 5Cu-Beta@CoSili) for ESR
catalytic test compared to that of the2.5Cu2.5Ni-Beta@
CoSili. The 5Ni-Beta@CoSili (Fig. 8e) catalyst showed
relatively high by-products selectivity. This can well prove
that the metal Ni component was favorable for the acetal-
dehyde decomposition reaction to generate a large amount
2
2
the SR of both ethanol and acetaldehyde (ASR) (Eq. 7). With
the temperature continues to escalate, notable enhancement
in H selectivity was achieved, accompanied by higher CO
2
2
selectivity and lower CO selectivity above 450 °C. It is prob-
ably associated with the acetaldehyde decomposition reaction
occurred obviously in parallel with the acetaldehyde steam
reforming at this temperature. Moreover, the CH selectivity
4
rapidly decreased and finally reached 3.2% at 550 °C, which
was in virtue of the promotion in methane steam reforming
(
MSR) (Eq. 8) at the higher temperature. The production of
CO and CH implied participation by the WGS and methana-
2
4
tion (Eq. 9) reactions, respectively. As the temperature above
5
50 °C, CH and CO concentrations decrease, indicating that
4
methanation was suppressed under the reaction conditions.
Co is known to cleave the C–C bond effectively, which can
of CH and CO by-products, but the WGS and MSR reac-
4
cause catalytic cracking of acetaldehyde into CH and CO. In
tion performance was weak. In addition, the lower activity
of Cu as active phase derives from limited ability in C–C
bond cleavage [37]. Indeed, the 5Cu-Beta@CoSili catalyst
(Fig. 8f) sample was selective mostly towards acetaldehyde.
The result further reveals that a certain amount of metal and
the synergetic effect between the Ni, Cu and Co components
is an effective method to enhance catalytic properties includ-
ing decrease of their by-products selectivity [29]. Combined
with the above analysis, it demonstrated that the core–shell
structure of the 2.5Cu2.5Ni-Beta@CoSili catalyst could not
4
addition, the ASR reaction may occur simultaneously over the
Co-based shell during the formation of acetaldehyde, increas-
ing the production of H , and followed by the WGS reaction
2
to convert CO into CO . A portion of the CH generated from
2
4
ACD can be converted by the Cu, Ni active phase in the MSR
reaction.
CH CH OH → CH CHO + H
2
3
3
2
(4)
(5)
(6)
(7)
(8)
(9)
CH CHO → CH + CO
3
4
only promote catalytic activity and H selectivity, but also
2
CO + H O ↔ CO + H
2
hindered formation of the undesired products in the ESR
2
2
reaction, especially the acetaldehyde molecule.
CH CHO + H O → 2CO + 3H
2
3
2
Based on the above results, the possible mechanism of
action was deduced. First of all, the reactant molecules
CH + H O ↔ CO + 3H
2
4
2
(
ethanol and water) pass through the Co-Silicalite-1 shell,
and adsorbed on the surface of Co nanoparticles form an
CO + 3H → CH + H
2
2
4
1
3

X. Li et al.
ethoxide. Subsequently, the surface adsorbed ethoxide to
a certain extent can dehydrogenate to acetaldehyde. Co is
known to cleave the C–C bond effectively, which can cata-
catalysts were screened out with 21 h stability test at 500 °C.
Typical experimental results are presented in Fig. 9, the data
reports for each catalyst are collected under the same steady-
state conditions. For the 2.5Cu2.5Ni-Beta@CoSili catalyst
(Fig. 9a), it shows enhancement in the catalytic performance
and long term stability. Notably, 100% ethanol conver-
sion and 72% hydrogen selectivity remain almost constant
throughout the 21 h reaction. As for the undesired products,
lytic cracking of acetaldehyde into CH and CO. Then the
4
acetaldehyde steam reforming and acetaldehyde decompo-
sition reaction may occur simultaneously on the Co active
sites. Through a series of catalytic reactions, a variety of
products were obtained such as H , CO , CH , CO and
2
2
4
acetaldehyde intermediate. Subsequently, large number
of intermediate products and unreacted ethanol enter into
the internal reaction system for further catalytic reactions.
Small-molecules (such as CH CHO, CO and CH ) were
small amounts of CO (7%) and CH (6%) were formed and
4
no acetaldehyde was detected during the reaction. In con-
trast, for the 2.5Ni2.5Cu-Beta-CoSili catalyst (Fig. 9b), the
conversion of ethanol continuously declined during reac-
tion together with increasing selectivity of aldehyde after
15 h, indicating that the catalyst’s ability to facilitate C–C
3
4
adsorbed by the high surface area Beta core to enrich around
the Ni, Cu nanoparticles, which can extend the time of reac-
tion and promote the small-molecule to react fully on active
sites. As in previous reports [38], Ni was known to cleave
the C–C bond effectively, which can cause catalytic cracking
bond cleavage was weakened. The initial H selectivity was
2
63% and followed a downwards trend to 48% eventually and
the acetaldehyde selectivity suffered a significant increase,
of acetaldehyde into CH and CO. Moreover, the Cu as an
accompanied by the slightly decreased of CO and CH ,
4
4
active catalyst center to break the O–H bond, the SMR and
WGS reactions may be a strong enhancement, which leads
to a complete conversion of ethanol and by-products. Simul-
taneously, a certain amount of the reactant concentration
may be rapidly increased to some extent in the reaction sys-
tem. Finally, the mixed gas was passed through the Co-Sili-
calite-1 shell again. Owing to the special core–shell structure
can provide a favorable reaction environmental, the reactants
through the system were carried out a series of “purification
which could be ascribed to a loss of activity of acetaldehyde
steam reforming and acetaldehyde decomposition reaction.
Based on the above results, the stability of catalyst for ESR
reaction is highly affected by the character and structure of
support. The special core–shell structure could maintain the
excellent catalytic performance of the 2.5Cu2.5Ni-Beta@
CoSili catalyst for a long period of time, moreover,the syn-
ergetic effect between the Ni, Cu and Co components on the
support reflects to a significant extent on the stability and
ethanol conversion.
effect”. Therefore, the selectivity to H was increased signifi-
2
cantly and the by-products almost disappeared.
3.4 Characterization of Spent Catalysts
3
.3 ESR Stability Test
Active metal sintering and carbon deposition are the primary
reasons for catalyst deactivation in ESR, both of which will
be investigated and compared among the two spent catalysts.
To better understand the effect of core–shell structure in
ESR, 2.5Cu2.5Ni-Beta@CoSili and 2.5Cu2.5Ni-Beta-CoSili
Fig. 9 Ethanol conversion and product distributions versus TOS for ESR at 500 °C over a 2.5Ni2.5Cu-Beta@CoSili and b 2.5Ni2.5Cu-Beta-
CoSili catalysts
1
3

Preparation and Characterization of Core–Shell Composite Zeolite BEA@MFI and Their Catalytic…
The spent catalysts after 21 h of stability test were character-
ized by XRD, TG techniques to investigate the factors result-
ing in different stability performance. First of all, the XRD
patterns of the two spent catalysts are provided in Fig. 10.
The complete diffraction peaks assigned to BEA and MFI
phase are shown in two the samples, indicating the high ther-
mal stability of the Beta and Silicalite-1 zeolite framework
during the ESR. The complete diffraction peaks assigned
to 2.5Ni2.5Cu-Beta@CoSili compare with 2.5Ni2.5Cu-
Beta-CoSili, suggesting the core shell structure enhanced
the thermal stability of zeolite skeleton. In addition, the two
catalysts present characteristic diffraction peaks correspond-
ing to Co metal, indicating excellent anti-sintering perfor-
mance due to the strong metal-support interaction and the
confinement effect of the core shell structure. Moreover, the
reflection peak of graphite (2θ=26.4°) was detected for both
Fig. 11 TG profiles for the spent catalysts
catalysts. In comparison, 2.5Ni2.5Cu-Beta@CoSili catalyst
shows weaker diffraction peak about graphitic carbon, sug-
gesting less graphitic carbon accumulated on the surface of
the catalyst.
and good coking-resistance capability is mainly attributed to
the core–shell structral, well-dispersed and stably confined
Ni, Cu nanoparticles, large surface area, and the novel mul-
tiple catalyst system.
TG investigations of the spent catalysts were carried out
to determine the amount of carbon deposition quantitatively.
As the mass loss profiles shown in Fig. 11, the mass started
decreasing from 355 °C and reached unchanged at 500 °C.
Apparently, the accumulated amount of coke deposition on
spent 2.5Ni2.5Cu-Beta@CoSili was 32%, which, however,
was doubled on the spent 2.5Ni2.5Cu-Beta-CoSili catalyst
In order to understand if the developed catalysts are
promising or not, it is essential to compare their activity,
selectivity and stability with the performance of other sam-
ples employed for ethanol steam reforming, especially in
terms of the amount of carbon and the data found in the
recent literature (Table 2). By comparing similar literature
under similar operative conditions, the catalysts developed in
the present study assured the stability and carbon deposited
values better than that found in other works.
(
61%). The observation illustrates that carbon deposition on
the catalyst surface was effectively suppressed for core–shell
type 2.5Ni2.5Cu-Beta@CoSili catalyst. XRD patterns and
TG measurements also confirm that the composition and
mesoporous shell of the core–shell composite catalyst are
well-retained after running 21 h, which is indicative of good
structure stability. Such an excellent catalytic performance
4 Conclusion
In summary, we have successfully prepared a series of xNi-
yCu-Beta@CoSili catalysts by seed induction method and
evaluated their performances in ethanol steam reforming.
Compared with the results obtained for the 2.5Cu2.5NiCo-
Beta, 2.5Cu2.5NiCo-Silicalite-1 and 2.5Ni2.5Cu-Beta-
CoSili, it was found that the 2.5Ni2.5Cu-Beta@CoSili was
a promising catalyst with superb catalytic activity, high H
2
selectivity, lower selectivity of CO, CH and acetaldehyde as
4
well as a high stability. This could be attributed to the struc-
tural advantages of core–shell support. First, the core–shell
structure could provide a small confinement environment for
Ni, Cu metal clusters, second these highly dispersed metal
ions could show synergistic effects in catalysis processes.
Moreover, the presence of the Cu and Ni phases was effec-
tive in improving the WGS and MSR reactions. On the other
hand, the shell may limit the diffusion of unreacted ethanol
and acetaldehyde to some extent, which greatly enhanced the
concentration in the micro-environment and gave rise to the
Fig. 10 XRD patterns of the spent catalysts
1
3

X. Li et al.
Table 2 Stability performances
of various catalysts for steam
reforming of ethanol
T (℃) Sample
H O /EtOH XEtOH (%) Stability
Carbon
References
2
g g −1h−1
(
)
c
cat
500
500
500
600
500
550
2.5Ni2.5Cu-Beta@CoSili
2.5Cu2.5NiCo-Silicalite-1
PtNi/CeO -Al O
3
3
3
3.7
3
100
90
100
100
98
Stable for 21 h 0.28
Stable for 21 h 0.52
Stable for 24 h 0.66
This work
This work
[39]
[40]
[41]
2
2
3
NiZr/SiO2
Co/Al O
Stable for 8 h
Stable for 6 h
Stable for 3 h
0.62
0.29
0.538
2
3
FeCo/α-Al O
3
100
[42]
2
3
complete catalytic conversion with metal Co, resulting in the
reactants further reacted to produce high-purity hydrogen.
19. Chen LC, Lin SD (2011) Appl Catal B 106:639
2
2
2
2
0. Lorenzut B, Montini T, De Rogatis L, Canton P, Benedetti A,
Fornasiero P (2011) Appl Catal B 101:397
1. De Rogatis L, Montini T, Lorenzut B, Fornasiero P (2008) Energy
Environ Sci 1:501
Acknowledgements We gratefully acknowledge the National Natu-
ral Science Foundation of China (No. 51541210) and Natural science
fundation of Shanxi Province (No. 201701D121042) for our funding.
2. Serrano DP, Van Grieken R, Sanchez P, Sanz P, Rodríguez L
(
2001) Micropor Mesopor Mater 46:35
3. Persson AE, Schoeman BJ, Sterte J, Otterstedt JE (1994) Zeolites
14:557
Compliance with Ethical Standards
2
2
2
4. Wang D, Xu L, Wu P (2014) J Mater Chem A 2:15535
5. Beckers J, Rothenberg G (2010) Green Chem 12:939
6. Shan J, Janvelyan N, Li H, Liu JL, Egle TM, Ye JC, Biener MM,
Biener J, Friend CM, Flytzani-Stephanopoulos M (2017) Appl
Catal B 205:541
Conflict of Interest The authors declare no conflicts of interest.
References
27. Bouizi Y, Rouleau L, Valtchev VP (2006) Chem Mater 18:4959
2
8. Groen JC, Abelló S, Villaescusa LA, Pérez-Ramírez J (2008)
Micropor Mesopor Mater 114:93
1
2
.
.
Li SR, Gong JL (2014) Chem Soc Rev 43:7245
2
3
9. Belambe AR, Oukaci R, Goodwin JJG (1997) J Catal 166:8
0. Prieto G, Martínez A, Concepción P, Moreno-Tost R (2009) J
Catal 266:129
Sharma YC, Kumar A, Prasad R, Upadhyay SN (2017) Renew
Sust Energ Rev 74:89
3
4
.
.
Llorca J, Homs N, Sales J, de la Piscina PR (2002) J Catal 209:306
Zhang J, Zhong Z, Cao XM, Hu P, Sullivan MB (2014) ACS Catal
3
1. Riva R, Miessner H, Vitali R, Del Piero G (2000) Appl Catal A
1
96:111
4
:448
3
3
2. Jongsomjit B, Panpranot J, Goodwin JG Jr (2001) J Catal 204:98
3. Batista MS, Santos RKS, Assaf EM, Assaf JM, Ticianelli EA
5
6
.
.
Coronel L, Múnera JF, Tarditi AM, Moreno MS, Cornaglia LM
(
2014) Appl Catal B 160:254
(
2004) J Power Sour 134:27
Rossetti I, Lasso J, Nichele V, Signoretto M, Finocchio E, Ramis
G, Michele AD (2014) Appl Catal B 150:257
Li D, Li XY, Gong JL (2016) Chem Rev 116:11529
Zhang CX, Li SR, Li MS, Wang SP, Ma XB, Gong JL (2012)
AIChE J 58:516
3
3
4. Pradhan AC, Uyar T (2017) ACS Appl Mater Interfaces 9:35757
5. Zhang JJ, Liang YQ, Mao J, Yang XJ, Cui ZD, Zhu SL, Li ZY
7
8
.
.
(
2016) Sensor Actuat B 235:420
3
3
3
3
4
4
4
6. Huang W, Zuo Z, Han P, Li Z, Zhao T (2009) J Electron Spectrosc
73:88
7. Sun JM, Karim AM, Mei DH, Engelhard M, Bao XH, Wang Y
1
9
.
Palma V, Castaldo P, Ciambelli P, Iaquaniello G (2014) Appl
Catal B 145:73
(
2015) Appl Catal B 162:141
1
1
1
1
1
1
1
0. Mattos LV, Jacobs G, Davis BH, Noronha FB (2012) Chem Rev
8. Li D, Zeng L, Li XY, Assabumrungrat S, Gong JL (2015) Appl
Catal B 176:532
1
12:4094
1. Karim AM, Su Y, Engelhard MH, King DL, Wang Y (2011) ACS
Catal 1:279
9. Sanchez-Sanchez MC, Navarro RM, Espartero I, Ismail AA, Al-
Sayari SA, Fierro JLG (2013) Top Catal 56:1672
0. Calles JA, Carrero A, Vizcaíno AJ (2010) Proceedings of the
WHEC 2010, p. 411
2. Koh AC, Leong WK, Chen L, Ang TP, Lin J, Johnson BF,
Khimyak T (2008) Catal Commun 9:170
3. Bouizi Y, Diaz I, Rouleau L, Valtchev VP (2005) Adv Funct Mater
1. Profeti LPR, Ticianelli EA, Assaf EM (2009) Appl Catal A: Gen
1
5:1955
3
60:17
4. Pirngruber GD, Laroche C, Maricar-Pichon M, Rouleau L, Bouizi
Y, Valtchev V (2013) Micropor Mesopor Mater 169:212
5. Bouizi Y, Rouleau L, Valtchev VP (2006) Micropor Mesopor
Mater 91:70
2. Sekine Y, Nakazawa Y, Oyama K, Shimizu T, Ogo S (2014) Appl
Catal A 472:113
Publisher’s Note Springer Nature remains neutral with regard to
6. Wang N, Sun QM, Bai RS, Li X, Guo GQ, Yu JH (2016) J Am
Chem Soc 138:7484
jurisdictional claims in published maps and institutional affiliations.
1
1
7. Chen LC, Lin SD (2014) Appl Catal B 148:509
8. Andonova S, De Avila CN, Arishtirova K, Bueno JMC, Damy-
anova S (2011) Appl Catal B 105:346
1
3