Facile synthesis of ordered mesoporous zinc alumina catalysts and their dehydrogenation behavior

Article abstract of DOI:10.1039/c9ra00217k

Ordered mesoporous Zn/Al₂O₃ materials with varying Zn content were simply prepared via an evaporation-induced self-assembly (EISA) method. Dehydrogenation of isobutane to isobutene was carried out on these materials; an isobutane conversion of 45.0% and isobutene yield of 39.0% were obtained over the 10%Zn/Al₂O₃ catalyst at 580?°C with 300 h^?1 GHSV. The obtained materials with Zn content up to 10% possess large specific surface area and big pore volume and zinc species can be highly dispersed on the surface or incorporated into the framework. The acidity of these catalysts was changed by the introduction of Zn, the decrease of strong acid sites is conducive to the promotion of isobutene selectivity and the weak and medium acidic sites played an important role in isobutane conversion. In addition, these catalysts exhibited good catalytic stability, due to the effective inhibition of coke formation by the ordered mesoporous structure.

Full text of DOI:10.1039/c9ra00217k

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Facile synthesis of ordered mesoporous zinc alumina
catalysts and their dehydrogenation behavior†
Cite this: RSC Adv., 2019, 9, 9828
a
a
a
a
a
Ming Cheng,ab Huahua Zhao,
Jian Yang, Jun Zhao, Liang Yan, Huanling Song
*
ac
*
and Lingjun Chou
Ordered mesoporous Zn/Al2O3 materials with varying Zn content were simply prepared via an evaporation-
induced self-assembly (EISA) method. Dehydrogenation of isobutane to isobutene was carried out on these
materials; an isobutane conversion of 45.0% and isobutene yield of 39.0% were obtained over the 10%Zn/
Al2O3 catalyst at 580 ꢀC with 300 hꢁ1 GHSV. The obtained materials with Zn content up to 10% possess large
specific surface area and big pore volume and zinc species can be highly dispersed on the surface or
incorporated into the framework. The acidity of these catalysts was changed by the introduction of Zn,
the decrease of strong acid sites is conducive to the promotion of isobutene selectivity and the weak
and medium acidic sites played an important role in isobutane conversion. In addition, these catalysts
exhibited good catalytic stability, due to the effective inhibition of coke formation by the ordered
mesoporous structure.
Received 10th January 2019
Accepted 23rd March 2019
DOI: 10.1039/c9ra00217k
rsc.li/rsc-advances
the poor selectivity with respect to the formation of many
undesired dry gas (CH4 and C2H6) impeded its more broad
1. Introduction
applications, which probably derived from redundant strong
acidic sites of HZSM-5 zeolite. Liu et al. reported that 27.5%
isobutane conversion and 83.8% isobutene selectivity obtained
in isobutane dehydrogenation over Zn/S-1 catalyst, because
strong acid sites were absent.15 Moreover, Zn/Ti thin lm and
Zn/Ga oxide catalyst has also been referred in isobutane
dehydrogenation.16,17
Isobutene is one of the most important raw materials and
intermediates to produce butyl rubber, ETBE (ethyl tert-butyl
ether), polyisobutene and other chemicals.1,2 To better fulll the
ever-increasing market demand, a small portion of its produc-
tion is by isobutane dehydrogenation.3 Currently, Cr2O3/Al2O3
and Pt/Al2O3 are the two most common and employable cata-
lysts in industrial production.4,5 However, some unavoidable
drawbacks limit their further application, because Pt is expen-
sive and has poor availability, while Cr is not environmentally-
friendly due to its toxicity. Therefore, the exploitation of novel
catalysts with superior catalytic properties, low cost and non-
pollution for isobutane dehydrogenation is highly
recommendable.6–8
At present, all kinds of metal catalysts, such as vanadium-
based, iron-based, molybdenum-based catalysts, have been
widely investigated in this dehydrogenation reaction.9–12 In
particular, owing to the excellent catalytic performance, Zn
based catalysts have been interested and gazed considerably in
the direct dehydrogenation of isobutane. For example, Zn
modied HZSM-5 materials were found effective for activation
of isobutane and high conversion obtained.13,14 Unfortunately,
Recently, ordered and adjustable mesoporous alumina
materials have drawn research much attention. Coupled with
their moderate surface acidity and good thermostability,
ordered mesoporous alumina as heterogeneous catalyst
support has a wonderful applying prospect.18 Morris et al.
successfully synthesized NiO–Al2O3 composites with ordered
mesoporous structure and the metal oxide well dispersed on the
alumina support.19 Yuan et al. reported a simple route to
synthesize g-Al2O3 using P123 template agent, which have high
quality mesoporous structure.20 Schweitzer et al. put forward the
computed minority catalytic pathway consists of undesired C–C
bond cleavage at Zn(II) site had a signicantly higher activation
energy barrier and the high olen selectivity observed for single-
site Zn(II) on SiO2.21 In our previous work, well-ordered meso-
porous Cr2O3/Al2O3 catalysts were synthesized and showed an
advantage in catalytic stability.22 So far, almost no research was
reported on ordered mesoporous zinc alumina composites for
alkane dehydrogenation.
aState Key Laboratory for Oxo Synthesis and Selective Oxidation, Lanzhou Institute of
Chemical Physics (LICP), Chinese Academy of Sciences, Lanzhou 730000, PR China.
E-mail: ljchou@licp.cas.cn; songhl@licp.cas.cn
bUniversity of Chinese Academy of Sciences, Beijing 100049, PR China
cSuzhou Research Institute of LICP, Chinese Academy of Sciences, Suzhou 215123, PR
China
This study rstly prepared ordered mesoporous xZn/Al2O3
catalysts with different Zn content via evaporation-induced self-
assembly (EISA) method and evaluated the catalytic perfor-
mances in isobutane dehydrogenation. The textural properties
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c9ra00217k
9828 | RSC Adv., 2019, 9, 9828–9837
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
of obtained materials were characterized by XRD analysis, N2
Thermogravimetric-differential scanning calorimetry (TG-
adsorption–desorption and TEM. And we also discussed the DSC) was based on a NETZSCH STA 449F3 analyzer under air
form of Zn species, metal-support interaction and surface within 20–800 ꢀC at 10 ꢀC minꢁ1
.
acidity by XPS, H2-TPR and NH3-TPD. Besides, all the charac-
terizations were associated with catalytic reactions.
The Zn content in a series of xZn/Al2O3 catalysts were
measured by inductively coupled plasma optical emission
spectrometer (ICP-OES, 725-ES, Agilent).
2. Experimental
2.1. Catalyst preparation
3. Results and discussion
3.1. XRD analysis
Ordered mesoporous Zn/Al2O3 materials were prepared via
EISA method according to the previous literatures.20,23,24 In
a typical procedure, 1.0 g P123 as surfactant was added to
20 ml ethanol and the solution was stirred for 40 min, followed
by the addition of a mmol aluminum isopropoxide, b mmol
Zn(NO3)2$6H2O (a + b ¼ 10 mmol) and 1.6 ml nitric acid. Aer
6 h of stirring, the mixed solution was transferred to a culture
dish and kept in drying oven at 60 ꢀC for two days. The ob-
tained xerogel was calcined at 600 ꢀC for 5 h. The catalysts were
denoted as xZn/Al2O3, where the nominal molar ratio x ¼ (b/(a
+ b) ꢂ 100%).
The XRD patterns of all the as-synthesized xZn/Al2O3 materials
are presented in Fig. 1. As displayed from Fig. 1A inside,
a distinct reection peak was observed in 3–10%Zn/Al2O3
samples at the characteristic reection (100) of p6mm space
group, which conrming that these materials were composed
of a well ordered mesoporous structure.25 With further
increasing Zn content to 15%, the diffraction peak at 0.8ꢀ
became almost disappeared, indicating the signicant effect
of zinc content in building the ordered mesoporous structure.
Part B of Fig. 1 presents the wide-angle XRD patterns of all the
materials, which revealed the existence of amorphous Al2O3.
Only the 15%Zn/Al2O3 sample exhibited sharp diffraction
peaks associated with hexagonal crystalline ZnO (no. 89-1397
from the ICDD). The average size of these ZnO particles is
23.8 nm. On the contrary, no diffraction peaks for Zn species
came into sight in 3–10%Zn/Al2O3 materials with ordered
mesopore, implying Zn species highly dispersed on support
surface or incorporated into amorphous alumina framework.26
In conclusion, the presence of ordered mesoporous structure
played an important role in promoting the dispersion of Zn
species on the catalyst support.
2.2. Catalytic dehydrogenation
The direct dehydrogenation of isobutane was studied in xed-
bed quartz reactor. Typically, 900 mg of catalyst was sieved at
ꢀ
60–80 mesh. The reaction temperature was set at 560–620 C.
The reactant gas was fed by gas hourly space velocity (GHSV) of
300 hꢁ1. The composition of the gaseous products was analyzed
on-line using gas chromatography tted with ame ionization
detector (FID) and thermal conductivity detector (TCD).
2.3. Characterization
3.2. Nitrogen adsorption–desorption analysis
N2 adsorption–desorption isotherms were measured by
Autosorb-iQ analyzer. The specic surface area was calculated The nitrogen adsorption–desorption analysis of all the as-
by Brunauer–Emmett–Teller (BET) equation and pore size synthesized xZn/Al2O3 materials is displayed in Fig. 2. Each
distribution was calculated by Barret–Joyner–Halenda (BJH) sample but 15%Zn/Al2O3 exhibited typical IV type isotherm as
equation from the N2 sorption isotherm.
well as H1 shaped hysteresis loop, implying the presence of
X-ray diffraction (XRD) were collected on X'Pert Pro diffrac- uniform cylindrical mesoporous channel in these catalysts.27
tion instrument using Cu Ka radiation over the range 0.6ꢀ < 2q < Furthermore, the 3–10%Zn/Al2O3 samples possessed quite
5ꢀ and 5ꢀ < 2q < 80ꢀ.
narrow pore size distribution around 9.5 nm, while that of
Transmission electron microscopy (TEM) images were con- 15%Zn/Al2O3 sample was bigger (around 12.4 nm). It sug-
ducted at TECNAIG2 F20 instrument.
gested that the larger pores could be accumulated by ZnO
X-ray photoelectron spectroscopy (XPS) were recorded on particles. Table 1 listed the detailed data regarding to the
Thermon ESCALAB 250xi spectrometer. The binding energies textural properties of above samples. It has been observed that
were calibrated against at 284.8 eV of C1s.
the specic surface area of as-synthesized 3–10%Zn/Al2O3
H2 temperature-programmed reduction (H2-TPR) was per- materials was similar (around 163 m2 gꢁ1), which because Zn
formed on ChemBET Pulsar Analyzer. Prior to the tests, sample species highly dispersed on support surface or incorporated
was pretreated at 300 ꢀC in He ow for 30 min. Aer cooling to into alumina framework wouldn't lead to pore plugging. At
ambient temperature, then raised temperature to 650 ꢀC in 10% a Zn content of 15%, the specic surface area was only 31 m2
H2–Ar mixed ow.
gꢁ1 owing to the mesoporous structure collapsed. It has been
NH3 temperature-programmed desorption (NH3-TPD) was known that a larger surface area conduced to the better
recorded on ChemBET Pulsar analyzer combining with mass dispersion of active species, thereby can provide more
spectroscopy. Aer being pretreated at 300 C in He ow, the “accessible” active centers for the reactant gas.28 Therefore,
ꢀ
ꢀ
sample adsorbed NH3 to saturation at 120 C. The prole was these as-synthesized 3–10%Zn/Al2O3 catalysts may own pref-
ꢀ
detected from 50 C to 600 ꢀC.
erable catalytic performance.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 9828–9837 | 9829

View Article Online
RSC Advances
Paper
Fig. 1 The XRD patterns of the as-synthesized and spent xZn/Al2O3 catalysts: (a) 3%Zn/Al2O3; (b) 5%Zn/Al2O3; (c) 7%Zn/Al2O3; (d) 10%Zn/Al2O3;
(e) 15%Zn/Al2O3; (f) the spent 10%Zn/Al2O3; (g) the spent 15%Zn/Al2O3.
particles was visible on the surface of 15%Zn/Al2O3 catalyst,
which conformed the result of wide-angle XRD analysis (Fig. 4).
3.3. TEM analysis
The morphology and structure of all the as-synthesized xZn/
Al2O3 materials were performed by TEM (Fig. 3). From the
images, all the sample except 15%Zn/Al2O3 presented one
dimensional parallel channel along [1 1 0], which intuitively
conrmed the presence of well-ordered mesoporous structure
in the 3–10%Zn/Al2O3 catalysts. Notably, no distinct ZnO clus-
ters were invisible in the images, illustrating the high disper-
sion of Zn species on the ordered mesoporous channel. This
result conformed to the low-angle XRD analysis. Only a fraction
of ordered mesopore was formed for 15%Zn/Al2O3 sample. In
the EDX prole of 10%Zn/Al2O3 sample (Fig. 3f), the charac-
teristic peaks of Al, Zn, O element can be observed clearly,
which veried that Zn species had been successfully loaded.
Besides, elemental mapping has further conrmed the ordered
mesoporous structure and shown Zn species on the surface to
be highly dispersed and distributed homogeneously over the
10%Zn/Al2O3 catalyst. However, an obvious aggregation of ZnO
3.4. XPS analysis
The XPS spectra of Zn 2p orbital for all the as-synthesized xZn/
Al2O3 materials are depicted in Fig. 5. As displayed, the binding
energy located at 1021.2–1201.6 eV and 1044.3–1044.7 eV was
assigned to Zn2+, as proved by spin–orbital splitting of 23.1 eV
29,30
between Zn 2p3/2 and Zn 2p1/2
.
Noteworthy, the binding
energy of Zn 2p in the sample 15%Zn/Al2O3 was lower than that
of 10%Zn/Al2O3, which may be derived from the difference of Zn
species in these catalysts. By deconvolution, the peak of 15%Zn/
Al2O3 was divided into two peaks, the big one may be assigned
to bulk ZnO particle on the surface, another small one may
belonged to Zn species incorporated into alumina framework.
The Zn content calculated by XPS and ICP-OES was used to
affirmed and the corresponding data listed in Table 1. We can
see that the bulk Zn contents were closely to the calculated data,
Fig. 2 The nitrogen adsorption–desorption analysis of the as-synthesized xZn/Al2O3 catalysts: (a) 3%Zn/Al2O3; (b) 5%Zn/Al2O3; (c) 7%Zn/Al2O3;
(d) 10%Zn/Al2O3; (e) 15%Zn/Al2O3.
9830 | RSC Adv., 2019, 9, 9828–9837
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
Table 1 The textural data of the as-synthesized xZn/Al2O3 catalystsa
Zn content (%)
Isotherm
type
Samples
SBET (m2 gꢁ1
)
VP (cm3 gꢁ1
)
dP (nm)
XPS
ICP-OES
3%Zn/Al2O3
5%Zn/Al2O3
7%Zn/Al2O3
10%Zn/Al2O3
15%Zn/Al2O3
10%Zn/Al2O3
10%Zn/Al2O3
15%Zn/Al2O3
163.3
162.6
163.4
163.3
31.3
149.7
158.6
25.2
0.41
0.43
0.44
0.43
0.23
0.40
0.42
0.16
9.5
9.6
9.6
9.6
12.4
9.3
IV
IV
IV
IV
III
IV
IV
III
1.6
2.3
3.0
3.8
14.6
—
2.5
4.3
6.4
9.7
12.8
—
a
b
a
9.5
11.8
—
—
—
—
a a,bStand for the corresponding spent and h regenerated catalyst, respectively.
only 15%Zn/Al2O3 had a larger deviation. Moreover, the surface particles on the surface. In contrast, the surface Zn was less
Zn content (14.6%) of this sample from XPS was higher than the than the bulk for 3–10%Zn/Al2O3 samples, which was due to
actual Zn content (12.8%), suggesting aggregation of ZnO most of Zn species incorporated into alumina framework. These
Fig. 3 The TEM images of the as-synthesized xZn/Al2O3 catalysts: (a) 3%Zn/Al2O3; (b) 5%Zn/Al2O3; (c) 7%Zn/Al2O3; (d) 10%Zn/Al2O3; (e) 15%Zn/
Al2O3. (f) EDX characterization for 10%Zn/Al2O3.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 9828–9837 | 9831

View Article Online
RSC Advances
Paper
Fig. 4 TEM image and Zn, Al elemental mappings of the fresh samples: (A) 10%Zn/Al2O3; (B) 15%Zn/Al2O3.
results were accorded with wide-angle XRD analysis and pure bulk ZnO. More specically, this peak belonged to the
conrmed the change of Zn species when increased Zn content reduction of binuclear (Zn–O–Zn)2+ clusters.31 It has been
from 10% to 15%. Besides, the higher binding energy meaning known that the higher reduction temperature, the stronger
a stronger interaction between Zn species and support in 3–10% interaction between metal and support. Therefore, the Zn
Zn/Al2O3 catalysts. This result will be further conrmed by species with the reduction peak around 490 ꢀC have stronger
following H2-TPR characterization.
interaction with support.32 The characterization of H2-TPR was
coincided with the above XRD and XPS analysis.
3.5. H2-TPR proles
3.6. NH3-TPD analysis
H2-TPR characterization is a very useful instrument to investi-
gate the interaction between Zn species and catalyst support. The acidic property of all the as-synthesized xZn/Al2O3 materials
The results for the as-synthesized xZn/Al2O3 materials and pure was determined by NH3-TPD and the corresponding results are
ZnO are shown in Fig. 6. All proles of xZn/Al2O3 showed displayed in Fig. S1.† All the samples exhibited a broad peak at
a broad reduction peak around 490 ꢀC, which could be assigned 50–600 ꢀC, indicating abundant different intensities of acidic
to the reduction of highly dispersed zinc species on support sites in xZn/Al2O3 catalysts. By deconvolution, the broad peak
surface and Zn species incorporated into alumina framework. was divided into four peaks accredited to physically adsorbed
However, 15%Zn/Al2O3 catalyst presented a unique reduction NH3, weak, medium and strong acidic sites respectively.33,34 The
ꢀ
peak at 416 C, which was similar with the reduction peak of detailed calculation result was in Table 2. As displayed, all the
9832 | RSC Adv., 2019, 9, 9828–9837
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
15%Zn/Al2O3 was much smaller, then physically adsorbed NH3
also less than other samples. Besides, we can see that there are
three different acidic sites on pure ordered mesoporous Al2O3.
With the introduction of 3%Zn, the amount of three acidic sites
signicantly enhanced. It indicated that a part of three types
acid sites originated from surface hydroxyl group and coordi-
native unsaturated Al sites on Al2O3 support, and other derived
from Zn species.15,35 Note that the total and weak acidic sites
gradually increased, while the strong acidic sites decreased
slowly with the increase of Zn content until 10%. When the Zn
content reached 15%, the total acidic strength of this catalyst
declined sharply, which may due to bulk ZnO has rarely acidic
sites (Fig. S2†). It was well established that weak and medium
acid sites played a key role to isobutane conversion in dehy-
drogenation reaction.15 However, side reactions (polymeriza-
tion, isomerization and cracking) are mainly catalyzed by strong
acid sites.36 As a conclusion, the various xZn/Al2O3 catalysts with
the different number of acidic sites may present distinct cata-
lytic performance.
Fig. 5 The Zn 2p XPS spectra of the as-synthesized xZn/Al2O3 cata-
lysts: (a) 3%Zn/Al2O3; (b) 5%Zn/Al2O3; (c) 7%Zn/Al2O3; (d) 10%Zn/
Al2O3; (e) 15%Zn/Al2O3.
3.7. Catalytic performance in the isobutane
dehydrogenation
samples except for 15%Zn/Al2O3 exhibited similar amount of
physically adsorbed NH3, which due to their comparable pore
property. However, the specic surface area and pore volume of
The reactivity of isobutane dehydrogenation over a series of
xZn/Al2O3 catalysts is displayed in Fig. 7. As we can notice that
the catalytic activity of xZn/Al2O3 catalysts gradually increased
until 10% Zn content. The 10%Zn/Al2O3 catalyst exhibited
a notably higher initial isobutane conversion (46.6%) and initial
isobutene selectivity (81.8%) by contrast with the 3%Zn/Al2O3
(29.4% conversion and 69.4% selectivity), indicating that Zn
content was a very signicant factor in isobutane dehydroge-
nation reaction. In addition to this, with increasing reaction
time, the reactivity over the 3–10%Zn/Al2O3 catalysts can basi-
cally hold steady, while the 15%Zn/Al2O3 catalyst presented
poor catalytic stability.
Isobutane dehydrogenation is an endothermic reaction in
thermodynamics, which need relatively high temperature to
obtain excellent isobutene yield. The effect of reaction temper-
ature on catalytic activity of 10%Zn/Al2O3 catalyst was investi-
gated in Fig. 8. The conversion of isobutane was 45.0% and the
selectivity of isobutene was 86.7% at 580 ꢀC aer 30 min. With
the increase of the reaction temperature, the conversion of
isobutane obviously increased and selectivity of isobutene
signicantly decreased, which was quite conformed to the
characteristic of alkane dehydrogenation reaction. However, the
Fig. 6 The H2-TPR profiles of the as-synthesized xZn/Al2O3 catalysts:
(a) 3%Zn/Al2O3; (b) 5%Zn/Al2O3; (c) 7%Zn/Al2O3; (d) 10%Zn/Al2O3; (e)
15%Zn/Al2O3; (f) ZnO.
Table 2 The acidic properties of the as-synthesized Al2O3 and xZn/Al2O3 catalysts
The amount of acidic sites (mmol gꢁ1
)
Samples
Physically adsorbed NH3 (mmol gꢁ1
)
Weak
Medium
Strong
Total
Al2O3
0.007
0.008
0.009
0.010
0.009
0.004
0.037
0.048
0.061
0.066
0.069
0.036
0.021
0.051
0.043
0.041
0.048
0.033
0.009
0.026
0.024
0.023
0.016
—
0.067
0.125
0.128
0.130
0.133
0.069
3%Zn/Al2O3
5%Zn/Al2O3
7%Zn/Al2O3
10%Zn/Al2O3
15%Zn/Al2O3
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 9828–9837 | 9833

View Article Online
RSC Advances
Paper
Fig. 9 The reactivity of isobutane dehydrogenation over the 10%Zn/
Al2O3 catalyst. Reaction condition: T ¼ 580 ꢀC.
Fig. 7 The reactivity of isobutane dehydrogenation over the as-
synthesized xZn/Al2O3 catalysts: (a) 3%Zn/Al2O3; (b) 5%Zn/Al2O3; (c)
7%Zn/Al2O3; (d) 10%Zn/Al2O3; (e) 15%Zn/Al2O3. Reaction condition: T
¼ 580 ꢀC, GHSV ¼ 300 hꢁ1
.
conversion and initial isobutene yield over Al2O3 were only 3.1%
and 2.1%, which illustrated that ordered mesoporous Al2O3 was
inactive to the reaction. ZnO exhibited slightly higher initial
dehydrogenation activity (9.2% isobutane conversion and 5.5%
isobutene yield). It seemed that bulk ZnO particle had some
catalytic ability. Surprisingly, with the introduction of Zn, the
xZn/Al2O3 catalysts exhibited very excellent initial catalytic
activity, even then the ordered mesopore collapsed and Zn
species was not highly dispersed over 15%Zn/Al2O3 catalyst.
Considering the possibility of zinc incorporated into Al2O3
framework over 15%Zn/Al2O3 catalyst, we deduced that the
existence of framework zinc may play a crucial role. However,
there is no direct evidence for this. The essential role of Zn
species of xZn/Al2O3 in isobutane dehydrogenation still need to
be identied, which is the focus of the following work in our
laboratory. In addition, it is well-known that the amount of
different acidic sites of catalyst are the key to affect catalytic
activity in alkane dehydrogenation reaction.37 The isobutane
Fig. 8 The reactivity of isobutane dehydrogenation over the 10%Zn/
Al2O3 catalyst. Reaction condition: GHSV ¼ 300 hꢁ1
.
isobutane conversion dropped drastically, isobutene selectivity
just increased slightly with the reaction temperature decreased
to 560 ꢀC. In view of catalytic stability, 580 ꢀC was deemed as the
optimal reaction temperature.
The impact of the GHSV on catalytic activity was also care-
fully studied over 10%Zn/Al2O3 catalyst (Fig. 9). As we can see,
with GHSV increasing from 300 to 600 hꢁ1, the initial isobutane
conversion decreased obviously (from 46.6% to 30.6%), while
initial isobutene selectivity increased (from 81.8% to 91.1%). As
the GHSV continues to rise, it became slope that isobutane
conversion decreased with time on stream.
In order to further investigate what were active sites for
isobutane dehydrogenation, we tested this reaction at 580 ꢀC
over ordered mesoporous Al2O3 and commercial ZnO as
a contrast (Fig. S3†). It can be seen that initial isobutane
Fig. 10 The trend of isobutane conversion rate with the amount of
different kinds of the acidic sites.
9834 | RSC Adv., 2019, 9, 9828–9837
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
conversion rate vs. the number of different acidic sites on the no difference between fresh and spent catalysts. The crystalline
basis of Table 2 were plotted (Fig. 10). The isobutane conversion ZnO phase was also visible over spent 15%Zn/Al2O3 catalyst,
rate versus the amount of strong acidic sites showed the worst and no Zn species over the spent 10%Zn/Al2O3 catalyst as well,
linear correlation (R2 ¼ 0.19), nevertheless, versus the amount of indicating that Zn species were stable without phase trans-
weak and medium acidic sites presented the best linear corre- formation during this dehydrogenation reaction.
lation (R2 ¼ 0.96), indicating that strong acidic sites were not
3.8.2. Nitrogen adsorption–desorption analysis. The
the decisive factor for isobutane conversion, while both the nitrogen adsorption–desorption analysis of the spent and
weak and medium acidic sites took the important part in regenerated catalysts are in Fig. S4.† Compared with the cor-
isobutane conversion. Besides, the isobutane selectivity was responding fresh catalysts, the isotherms of spent catalysts had
obviously increased with the decrease of the amount of strong no obvious changed and still presented uniform pore size. The
acidic sites over 3–10%Zn/Al2O3 catalysts, indicated strong acid textural properties are presented in Table 1. It was clearly seen
sites mainly catalyzed side reactions.
that the BET specic surface areas of the spent 10%Zn/Al2O3
and 15%Zn/Al2O3 catalysts were greatly reduced to 149.7 m2 gꢁ1
and 25.2 m2 gꢁ1, respectively. It was probably due to that coke
deposition on the spent catalyst blocked a portion of the pore,
thus leading to the decrease of textural performance. Aer ve
dehydrogenation–regeneration cycles, the slightly diminished
3.8. The stability of ordered mesoporous Zn/Al2O3 catalyst
3.8.1. XRD analysis. The wide-angle XRD patterns of the
spent catalysts are shown in Fig. 1. It was observed that there is
Fig. 11 TG-DSC curves of the spent catalysts at 580 ꢀC reaction: (a) 3%Zn/Al2O3; (b) 5%Zn/Al2O3; (c) 7%Zn/Al2O3, (d) 10%Zn/Al2O3; (e) 15%Zn/
Al2O3 and the spent catalyst at 620 ꢀC reaction: (f) 10%Zn/Al2O3.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 9828–9837 | 9835

View Article Online
RSC Advances
Paper
specic surface area (163.3 / 158.6 m2 gꢁ1) and pore volume conversion, isobutene selectivity and yield were decrease
(0.43 / 0.42 cm3 gꢁ1) of the regenerated 10%Zn/Al2O3 slightly, indicated the high regenerative ability of the catalyst.
demonstrated the formed coke can be eliminated by regenera-
tion process.
4. Conclusions
3.8.3. TG-DSC analysis. As we all know, coking is usually
the main reason for catalyst deactivation in alkane dehydroge- A series of xZn/Al2O3 materials with various Zn content were
nation reaction. TG-DSC is a very efficient technique to analyze simply prepared via one pot EISA strategy and tested in isobu-
the coke amount and the characterization results are shown in tane dehydrogenation reaction. The obtained materials with Zn
Fig. 11. The TG curves presented downtrend with the increase of content up to 10% possessed well-ordered mesopore with large
temperature, which could be divided into two stages. In the rst specic surface areas, big pore volumes and uniform pore size.
ꢀ
stage, the weight losses in TG curves up to 300 C, ascribed to Zinc species in these catalysts was highly dispersed on support
the loss of physically adsorbed water and impurities. The surface or incorporated into framework, while ZnO crystal
second stage, located in the 350–600 ꢀC, the losses was assigned particles were observed with 23.8 nm size in the case of 15%Zn.
to the removal of coke. The coke amount in the spent 3%Zn/ It was found that these materials presented excellent catalytic
Al2O3, 5%Zn/Al2O3, 7%Zn/Al2O3 and 10%Zn/Al2O3 were 2.97%, activity. Note that the total acidic sites gradually increased,
3.22%, 3.29% and 3.45%, respectively. Which similar coke while the strong acidic sites decreased slowly with the increase
amount was consistent with the same good stability of these of Zn content until 10%. The decrease of strong acid sites is
ordered mesoporous catalysts. However, with increasing Zn conducive to the promotion of isobutene selectivity and the
content to 15%, coke amount was markedly increased (5.19%). weak and medium acidic sites played a role in isobutane
Combined with the poor stability of 15%Zn/Al2O3 catalyst, the conversion. In addition, the catalyst exhibited excellent stability
reason of serious deactivation may caused by more coke, which and high regenerative ability, which demonstrated potential for
also indicated ordered meso-structure could effectiꢀvely inhibi- commercial applications.
ted the formation of coke.38 Moreover, during 620 C reaction
temperature, the coke amount in the spent 10%Zn/Al2O3 cata-
lyst increased to 3.79%. It explained the stability decreased with
Conflicts of interest
the increase of reaction temperature.
There are no conicts to declare.
3.8.4. The dehydrogenation–regeneration cycles. In order
to examine the regenerative ability of the 10%Zn/Al2O3 catalyst,
the ve dehydrogenation–regeneration cycles were investigated
Acknowledgements
(Fig. 12). By calculation, the initial isobutane conversion, iso- The authors sincerely acknowledge the nancial support from
butene selectivity and yield were 45.1%, 83.6% and 37.7% in the the National Natural Science Foundation of China (no.
rst cycle, respectively. With increasing reaction time, the 21573260 and No. 21403255), Youth Innovation Promotion
catalytic activity decreased gradually. Following by 2 h regen- Association CAS (no. 2017460) and Jiangsu Science and Tech-
ꢀ
eration of catalyst at 600 C in air, the activity of catalyst was nology Program (no. BY2016048-01).
obviously restored, which affirmed coke was the main reason of
catalyst deactivation. In the ve cycle, the initial isobutane
References
1 I. Gniot, P. Kirszensztejn and M. Kozłowski, Appl. Catal., A,
2009, 362, 67–74.
2 A. Feller, I. Zuazo, A. Guzman, J. O. Barth and J. A. Lercher, J.
Catal., 2003, 216, 313–323.
3 Y. Tian, X. Liu, M. Rood and Z. Yan, Appl. Catal., A, 2017, 41,
1–9.
4 S. Al-Zahran, N. Elbashir, A. Abasaeed and M. Abdulwahed,
Ind. Eng. Chem. Res., 2001, 40, 781–784.
5 L. Wan, Y. Zhou, Y. Zhang, Y. Duan, X. Liu and M. Xue, Ind.
Eng. Chem. Res., 2011, 58, 4280–4285.
¨
6 J. Zhang, D. Su, A. Zhang, D. Wang, R. Schlogl and C. Hebert,
Angew. Chem., Int. Ed., 2007, 46, 7319–7323.
7 J. Grant, C. Carrero, F. Goeltl, J. Venegas, P. Mueller, S. Burt,
S. Specht, W. McDermott, A. Chieregato and I. Hermans,
Science, 2016, 354, 1570–1573.
8 J. H. B. Sattler, J. Ruiz-Martinez, E. Santillan-Jimenez and
B. M. Weckhuysen, Chem. Rev., 2014, 114, 10613–10653.
9 J. Mu, J. Shi, L. France, Y. Wu, Q. Zeng, B. Liu, Li. Jiang,
J. Long and X. Li, ACS Appl. Mater. Interfaces, 2018, 10,
23112–23121.
Fig. 12 The five dehydrogenation–regeneration cycles of isobutane
dehydrogenation over 10%Zn/Al2O3 catalyst. Reaction condition: T ¼
580 ꢀC, GHSV ¼ 300 hꢁ1
.
9836 | RSC Adv., 2019, 9, 9828–9837
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
10 H. Zhao, H. Song, J. Yang, J. Zhao, L. Yan and L. Chou, Catal. 24 L. Xu, H. Song and L. Chou, Int. J. Hydrogen Energy, 2013, 38,
Sci. Technol., 2017, 15, 3256–3267. 7307–7325.
11 S. Wang, K. Wang and X. Wang, Appl. Surf. Sci., 2016, 386, 25 D. Pan, W. Chen, F. Yu, S. Chen, X. Yan, X. Shi, B. Fan and
442–450. R. Li, Mater. Lett., 2018, 224, 33–36.
12 M. Cheng, H. Zhao, J. Yang, J. Zhao, L. Yan, H. Song and 26 X. Xiang, H. Zhao, J. Yang, J. Zhao, L. Yan, H. Song and
L. Chou, Microporous Mesoporous Mater., 2018, 266, 117–125. L. Chou, Appl. Catal., A, 2016, 520, 140–150.
13 J. Liu, A. Hong, N. He, G. Liu, C. Liang, X. Zhang and H. Guo, 27 M. Kruk and M. Jaroniec, Chem. Mater., 2001, 13, 3169–3183.
Chem. Eng. J., 2013, 218, 1–8. 28 Z. He, Y. Jiao, J. Wang and Y. Chen, Fuel, 2018, 212, 193–201.
14 Y. Sun and T. Brown, Int. J. Chem. Kinet., 2002, 34, 467–480. 29 A. Choi, J. Park, Y. Kang, S. Lee and Y. Kang, Phys. B, 2018,
15 G. Liu, J. Liu, N. He, C. Miao, J. Wang, Q. Xin and H. Guo,
RSC Adv., 2018, 8, 18663–18671.
16 Z. Chen, A. Derking, W. Koot and M. van Dijk, J. Catal., 1996,
161, 730–741.
17 G. Wang, C. Li and H. Shan, Catal. Sci. Technol., 2016, 6,
3128–3136.
18 W. Wang, K. Zhang, Y. Yang, H. Liu, Z. Qiao and H. Luo,
Microporous Mesoporous Mater., 2014, 193, 47–53.
19 S. M. Morris, P. F. Fulvio and M. Jaroniec, J. Am. Chem. Soc.,
2008, 130, 15210–15216.
20 Q. Yuan, A. Yin, C. Luo, L. Sun, Y. Zhang, W. Duan, H. Liu
and C. Yan, J. Am. Chem. Soc., 2008, 130, 3465–3472.
21 N. M. Schweitzer, B. Hu, U. Das, H. Kim and A. S. Hock, ACS
Catal., 2014, 4, 1091–1098.
22 L. Xu, Z. Wang, H. Song and L. Chou, Catal. Commun., 2013,
35, 76–81.
537, 251–257.
30 W. Gu, X. Wu and J. Zhang, Mater. Sci. Semicond. Process.,
2018, 81, 89–93.
31 S. Tamiyakul, W. Ubolcharoen, D. N. Tungasmita and
S. Jongpatiwut, Catal. Today, 2015, 256, 325–335.
32 C. Yu, H. Xu, Q. Ge and W. Li, J. Mol. Catal. A: Chem., 2007,
266, 80–87.
33 H. Ham, M. H. Jeong, H. M. Koo, C. H. Chung and J. W. Bae,
React. Kinet., Mech. Catal., 2015, 116, 173–189.
34 U. Rodemerck, E. V. Kondratenko, T. Otroshchenko and
D. Linke, Chem. Commun., 2016, 52, 12222–12225.
35 S. Zhang, Y. Zhou, Y. Zhang and L. Huang, Catal. Lett., 2010,
135, 76–82.
36 Y. Ren, F. Zhang, W. Hua, Y. Yue and Z. Gao, Catal. Today,
2009, 148, 316–322.
37 M. Dixit, P. Kostetskyy and G. Mpourmpakis, ACS Catal.,
2018, 8, 11570–11578.
38 L. Xu, Z. Miao, H. Song, W. Chen and L. Chou, Catal. Sci.
Technol., 2014, 4, 1759–1770.
23 M. Morris, P. Fulvio and M. Jaroniec, J. Am. Chem. Soc., 2008,
130, 15210–15216.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 9828–9837 | 9837