Designing a hierarchical nanosheet ZSM-35 zeolite to realize more efficient ethanol synthesis from dimethyl ether and syngas

Article abstract of DOI:10.1016/j.cattod.2019.02.054

In this work, a dual-catalyst bed reactor packed with the combination of hierarchical nanosheet HZSM-35 (Hi-NZ35) zeolite and CuZnAl catalyst was proposed to realize more efficient ethanol synthesis from dimethyl ether (DME) and syngas (CO+H₂). The nanosheet ZSM-35 (NZ35) zeolite was prepared via a direct hydrothermal synthesis route and the CuZnAl catalyst was prepared by co-precipitation method. Moreover, a series of Hi-NZ35_x zeolites were obtained from NZ35 zeolite by further treatment with varied NaOH aqueous solution using hydrothermal process (“x” means the NaOH solution concentration of 0.2-0.6 M). The catalysts properties, such as crystallinity, porosity, acidity, morphology and composition, were characterized by X-ray diffraction (XRD), N₂ adsorption-desorption, NH₃ temperature-programmed desorption (NH₃-TPD), H₂ temperature-programmed reduction (H₂-TPR), scanning electron microscope (SEM) and energy dispersive spectroscopy (EDS). It was found that Hi-NZ35_0.4 zeolite was more effective to obtain hierarchical porosity with mesopore volume up to 0.131 cm³ g^?1. For single DME carbonylation reaction, the NZ35 zeolite exhibited superior catalytic activity (32.2%) and stability compared with conventional ZSM-35 (CZ35) zeolite. Furthermore, improved catalytic activity (42.0%) was observed on Hi-NZ35_0.4 zeolite owing to its abundant mesoporous structure. This result revealed that the hierarchical porosity of zeolite could effectively promote the catalytic performance of zeolite for DME carbonylation reaction. For the ethanol synthesis using the optimized catalysts combination of Hi-NZ35_0.4 zeolite and CuZnAl catalyst, the DME conversion was about 47.2% with higher ethanol productivity of 840.2 mmol kg^?1 h^?1.

Full text of DOI:10.1016/j.cattod.2019.02.054

CatalysisTodayxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Designing a hierarchical nanosheet ZSM-35 zeolite to realize more efficient
ethanol synthesis from dimethyl ether and syngas
Xiaobo Fenga, Peipei Zhanga, Yuan Fanga, Witchakorn Charusirib, Jie Yaoa, Xinhua Gaoa,
Qinhong Weia, Prasert Reubroycharoenc, Tharapong Vitidsantc, Yoshiharu Yoneyamaa,
⁎
⁎
Guohui Yanga, , Noritatsu Tsubakia,
a Department of Applied Chemistry, School of Engineering, University of Toyama, Gofuku 3190, Toyama 930-8555, Japan
b Faculty of Environmental Culture and Ecotourism, Srinakharinwirot University, Bangkok 10110, Thailand
c Department of Chemical Technology, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand
A R T I C L E I N F O
A B S T R A C T
Keywords:
In this work, a dual-catalyst bed reactor packed with the combination of hierarchical nanosheet HZSM-35 (Hi-
NZ35) zeolite and CuZnAl catalyst was proposed to realize more efficient ethanol synthesis from dimethyl ether
(DME) and syngas (CO+H2). The nanosheet ZSM-35 (NZ35) zeolite was prepared via a direct hydrothermal
synthesis route and the CuZnAl catalyst was prepared by co-precipitation method. Moreover, a series of Hi-
NZ35x zeolites were obtained from NZ35 zeolite by further treatment with varied NaOH aqueous solution using
hydrothermal process (“x” means the NaOH solution concentration of 0.2-0.6 M). The catalysts properties, such
as crystallinity, porosity, acidity, morphology and composition, were characterized by X-ray diffraction (XRD),
N2 adsorption-desorption, NH3 temperature-programmed desorption (NH3-TPD), H2 temperature-programmed
reduction (H2-TPR), scanning electron microscope (SEM) and energy dispersive spectroscopy (EDS). It was found
that Hi-NZ350.4 zeolite was more effective to obtain hierarchical porosity with mesopore volume up to
0.131 cm3 g−1. For single DME carbonylation reaction, the NZ35 zeolite exhibited superior catalytic activity
(32.2%) and stability compared with conventional ZSM-35 (CZ35) zeolite. Furthermore, improved catalytic
activity (42.0%) was observed on Hi-NZ350.4 zeolite owing to its abundant mesoporous structure. This result
revealed that the hierarchical porosity of zeolite could effectively promote the catalytic performance of zeolite
for DME carbonylation reaction. For the ethanol synthesis using the optimized catalysts combination of Hi-
NZ350.4 zeolite and CuZnAl catalyst, the DME conversion was about 47.2% with higher ethanol productivity of
Ethanol synthesis
Syngas
DME
Hierarchical nanosheet HZSM-35 zeolite
Dual-catalyst bed
840.2 mmol kg−1 h−1
.
1. Introduction
been developed [8–11]. Among them, the strategy of ethanol synthesis
from dimethyl ether (DME) and syngas, firstly reported by our group,
has received considerable industrial attentions due to the merits of ef-
ficient carbon cycle, facile reaction conditions and low cost catalysts
[12,13]. In this route, methyl acetate (MA) was firstly produced by
DME carbonylation reaction on the upper zeolite bed, and then the
formed MA was converted into methanol and ethanol through hydro-
genation on the lower copper-based catalyst bed. It should be noted that
the produced methanol could be used as the feedstock in subsequent
recycling reaction for DME generation (Scheme 1).
Ethanol is considered as a potential alternative fuel additive or as an
important feedstock to synthesize a variety of industrial chemicals and
polymers [1]. Currently, ethanol is primarily produced by ethylene
hydration and biomass-derived resources fermentation [2]. However,
the established processes have to face great challenges owing to the oil
resource shrinking and costly biological process. Syngas (a mixture of
CO and H2) can be produced by cheap resources such as natural gas,
biomass and coal [3]. Therefore, the catalytic conversion of syngas to
ethanol is considered as one of the promising ethanol synthesis routes
[4]. By far, the research on the direct synthesis of ethanol from syngas
by one-step is limited due to the low selectivity and activity [5–7].
Alternatively, some indirect ethanol synthesis methods via syngas have
As mentioned above, for ethanol synthesis form DME and syngas,
DME carbonylation reaction to MA is considered as a crucial step in this
new ethanol synthesis route. Nowadays, among the different zeolite
candidates those have been evaluated, the ZSM-35 zeolite with a
⁎ Corresponding authors.
E-mail addresses: thomas@eng.u-toyama.ac.jp (G. Yang), tsubaki@eng.u-toyama.ac.jp (N. Tsubaki).
https://doi.org/10.1016/j.cattod.2019.02.054
Received 26 August 2018; Received in revised form 24 January 2019; Accepted 22 February 2019
0920-5861/©2019ElsevierB.V.Allrightsreserved.
Pleasecitethisarticleas:XiaoboFeng,etal.,CatalysisToday,https://doi.org/10.1016/j.cattod.2019.02.054

X. Feng, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Scheme 1. Production of ethanol from syngas and DME.
ferrierite (FER) structure is verified as the most stable catalyst in DME
carbonylation reaction [14,15]. However, the special structure features
(8-membered rings channel: 3.5 × 3.8 Å and 10-membered rings
channel: 4.2 × 5.4 Å) of ZSM-35 zeolite resulted in its poor catalytic
activity. Therefore, the design and modification of ZSM-35 zeolite to
improve its catalytic activity in DME carbonylation should be paid more
attentions. In the previous study, nanocrystal-assembled hierarchical H-
mordenite (HMOR) zeolite had been developed and exhibited good
DME carbonylation catalytic properties because of both the reduced
crystal size and mesoporous structure that could shorten the transport
pathways for reactants and products in the zeolite channels [16,17].
However, there are no studies on improving DME carbonylation per-
formance over hierarchical nanocrystal HZSM-35 (Hi-NZ35) zeolite so
far.
12 h and finally calcined at 550 °C for 5 h. The NaZSM-35 sample was
converted to HZSM-35 through three times of ion-exchange process
with a 0.2 M NH4NO3 aqueous solution at 80 °C for 6 h respectively,
followed by drying at 110 °C overnight and calcination at 550 °C for 5 h.
For comparison, a commercial HZSM-35 zeolite (CZ35) purchased
from Tosoh Catalyst Plant (Tosoh Co., Tokyo, Japan), as a reference
catalyst, was also tested in DME carbonylation reaction.
2.1.2. The preparation of hierarchical nanosheet HZSM-35 zeolite
In this process, the NZ35 zeolite prepared above was used as the
parent sample. Alkaline modification was performed in NaOH (0.2-
0.6 M) aqueous solution according to a reported procedure [20]. In the
treatment process, NZ35 zeolite was firstly treated with NaOH solution
by hydrothermal method at 80 °C for 4 h. Subsequently, the resulting
sample was washed with distilled water until neutral pH and dried at
100 °C for 8 h. The solid product was converted into H-type zeolite
through three times of ion-exchange processes in 0.2 M NH4NO3 solu-
tion at 80 °C for 6 h respectively and then calcined at 550 °C for 4 h. The
final alkali-treated samples were noted as Hi-NZ350.2, Hi-NZ350.4 and
Hi-NZ350.6 corresponding to the varied NaOH solution concentration of
0.2, 0.4 and 0.6 M respectively.
In this work, it was hypothesized that the Hi-NZ35 zeolite could
fulfill the role of improving reactants and products diffusion, thus
leading to better zeolite catalytic performance for DME carbonylation.
Consequently, a nanosheet HZSM-35 (NZ35) zeolite, possessing about
100 nm slice, was synthesized by hydrothermal method, and then the
hierarchical porosity structure was subsequently obtained by post-
treatment with a series of NaOH aqueous solution using hydrothermal
method. Therefore, the NZ35 and a series of Hi-NZ35 zeolites were
developed. The catalytic performance of theses zeolites for DME car-
bonylation to generate MA will be studied in detail. Moreover, for
ethanol synthesis through MA hydrogenation reaction, the employed
CuZnAl catalyst is the industrially well-known catalyst and shows ex-
traordinary catalytic performance [18]. In this report, a more efficient
ethanol synthesis strategy from DME and syngas over a dual-catalyst
bed was presented and realized by combining the hierarchical nano-
crystal HZSM-35 zeolite and CuZnAl catalyst.
2.1.3. The preparation of CuZnAl catalyst
The CuZnAl catalyst was prepared by a co-precipitation method
[18]. 300 mL Na2CO3 aqueous solution (0.5 M), and 300 mL aqueous
solution of Cu(NO3)2·3H2O, Zn(NO3)2·6H2O and Al(NO3)3·9H2O with
the designed molar ratio (Cu/Zn/Al = 1/1/0.1) were added simulta-
neously into 300 mL deionized water under constant stirring. The pre-
cipitation temperature and pH value were maintained at 65 °C and 8.5,
respectively. The obtained deposit was aged overnight, filtrated, wa-
shed with deionized water and dried at 100 °C for 10 h. Subsequently,
the precursor was heated with a rate of 2 °C min−1 to 350 °C and kept
for 1 h. The catalyst reduction process was performed in the flow of 5%
H2 in Ar at 300 °C for 10 h. Finally, 20 mL/min diluted oxygen (1% in
N2) was used for the catalyst surface passivation.
2. Experimental
2.1. Catalysts preparation
2.1.1. Nanosheet HZSM-35 zeolite synthesis
The NZ35 zeolite was synthesized though a hydrothermal method
using trimethylcetylammonium hydroxide (TMCAH) as the organic
structure-directing agent (OSDA) according to the procedure reported
in literature [19]. The molar composition of the precursor mixture was
Na2O:Al2O3:20SiO2:8TMCAH:448.5H2O. In a typical procedure, the
aluminate solution was prepared by adding sodium aluminate (37 wt%
Al2O3, 34 wt% Na2O) into deionized water, named as solution A. An-
other solution B was prepared by adding tetraethoxysilane (TEOS) and
TMCAH into deionized water under stirring. Then, solution B was
added into the solution A with a rigorously stirring to get a clear gel
mixture. Further stirring on the mixture at room temperature was
performed for 1 h, and then the hydrogel was transferred into a Teflon-
lined autoclave. After being crystallized at 160 °C for 144 h, the re-
sultant product was filtrated, washed with water, dried at 110 °C for
2.2. Catalyst characterization
The crystalline phase of catalysts was analyzed by X-ray diffraction
(XRD) using Rigaku Ultima IV X-ray diffractometer with Cu-Kα radia-
tion (λ = 0.154 nm) performed at 40 kV and 20 mA. Crystallinity was
determined according to the sum of the three most intense reflections
(2θ = 9.2, 25.2 and 25.7°). The parent NZ35 zeolite was assigned a
crystallinity of 100% [21]. The surface morphology of the catalysts was
observed by the scanning electron microscopy (SEM, JEOL JSM-
6360LV) equipped with energy dispersive spectroscopy (EDS, JED-
2300), by which to collect the morphology and elemental composition
of the catalysts. The textural properties of the catalysts were de-
termined by N2 physical adsorption (Quantachrome Nova 2200e). Be-
fore measurement, the samples were degassed in a vacuum at 200 °C for
2

X. Feng, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Scheme 2. Illustration of the reactor: (a) single-catalyst bed reactor for DME
Fig. 1. XRD patterns of the CZ35, NZ35 and Hi-NZ35x zeolites.
carbonylation and (b) dual-catalyst bed reactor for ethanol synthesis.
3. Results and discussion
200 min (ramp rate: 10 °C min−1). The reduction behavior of the
CuZnAl catalysts was measured by H2 temperature-programmed re-
duction (H2-TPR) using a BELCAT-B-TT (BEL, Japan) apparatus deco-
rated with a thermal conductivity detector (TCD). The NH3 tempera-
ture-programmed desorption (NH3-TPD) was tested with the same
equipment as for H2-TPR to characterize the acidic properties of zeo-
lites.
3.1. Catalyst characterization
3.1.1. XRD analysis of zeolites
Fig. 1 displays the XRD patterns of CZ35, parent NZ35 and the al-
kali-treated zeolites (Hi-NZ35x, x = 0.2, 0.4 and 0.6). The classic dif-
fraction peaks assigned to typical ZSM-35 zeolite at 9.2, 25.2 and 25.7°
could be observed on all the samples, indicating that the zeolite fra-
mework was well maintained. With the increase of NaOH concentra-
tion, the relative crystallinity increased slightly to 104% on Hi-NZ350.2
and then decreased to 90% on Hi-NZ350.6 zeolite compared with their
parent NZ35 zeolite. These findings implied that the amorphous matter
in the zeolite channels was removed by mild alkaline modification,
whereas the crystal phase was partially dissolved in the strong alkaline
solution. Especially for the Hi-NZ350.6 sample, the intensity of its
characteristic reflection peaks was lower than that of NZ35 parent
zeolite, meaning the loss of some crystallinity during hydrothermal
process [23].
2.3. Catalysts catalytic performance evaluation
All of the catalytic performance evaluation reactions were per-
formed in a fixed-bed stainless steel reactor (9.5 mm OD) (as shown in
Scheme 2). The detailed description on this novel reactor could refer to
our previous study [22]. For the single DME carbonylation, only 0.5 g
zeolite was loaded in the reactor as shown in Scheme 2a. After being
heated at 500 °C under flowing N2 (30 mL min−1) for 3 h, the reactor
was cooled down to reaction temperature, and then the reactants gas
(1.51% Ar/2.06% DME/46.43% CO /50.00% H2, in volume ratio) with
pressure of 1.5 MPa was introduced. For ethanol synthesis, the reaction
was carried out with the same fixed-bed stainless steel reactor loading
with the combination of the upper zeolite catalyst and lower CuZnAl
catalyst (Scheme 2b), that is, 0.5 g zeolite was loaded in the upper layer
while 0.5 g reduced CuZnAl catalyst was fixed at the bottom layer.
The effluent gas phase of DME, CO, CH4, CO2 and Ar (Ar was em-
ployed as the internal standard) from the reactor outlet was con-
tinuously analyzed by an online gas chromatograph (GC) equipped with
a thermal conductivity detector (TCD) and a connected dual columns
(Porapak Q column for DME and Activated Carbon column for Ar, CO,
CH4 and CO2). Liquid products (MA, methanol and ethanol) were col-
lected with 1-butanol as solvent in an ice trap and then analyzed using
GC-8 A with a flame-ionization detector (FID). Here, 1-propanol was
employed as the internal standard for liquid products analysis. All of
the products selectivity in this report was calculated with molecular
base. The DME conversion (CDME), products selectivity (Si) and ethanol
productivity (PEthanol) were calculated by the following equations:
3.1.2. Surface morphology and composition analysis of zeolites
Fig. 2 illustrates SEM images of the tested zeolites. Apparently, the
SEM morphology of NZ35 zeolite presented a discoid shape with the
thickness of around 100 nm, whereas the CZ35 zeolite exhibited stone-
like shape crystal about 10 μm. For all the alkali-treated samples, the
sheets dissolution phenomena were observed. Compared to the uniform
crystal size and smooth surface of NZ35 zeolite, the Hi-NZ350.2 zeolite
showed almost integral crystal phase, as confirmed also by the XRD
analysis. As for the Hi-NZ350.4 zeolite, the crystal morphology dis-
played non-uniform crystals with some small fragments, implying that
the crystals were partially corroded. By further increasing the NaOH
concentration to 0.6 M, more small fragments were observed. Some
cracks and faults were formed on the outer surface of crystalline grains
during severe alkaline treatment, which led to collapse of few large
particles into small fragments [20,24]. These results also agreed well
with the XRD analysis. On the other hand, an obvious decrease trend in
the bulk Si/Al ratio of the alkali-treated zeolites was observed with
increasing NaOH concentration, revealing the preferential dissolution
of silica species in the alkaline treatment conditions [25].
C
DME (%) = (XDME, in /XAr, in –XDME, out /XAr, out)/ (XDME, in /XAr, in)
×100%
Si (%) =Ni/∑Ni ×100% (Ni = molar content of MA, methanol,
ethanol, CO2 and CH4)
PEthanol =NEthanol/ (W·T)
3.1.3. Specific surface area and pore structure of the zeolites
(XDME, in: mole fraction of DME in feed gas; XDME, out: mole fraction
of DME in exit gas; XAr, in: mole fraction of Ar in feed gas; XAr, out: mole
fraction of Ar in exit gas; W and T stand for the catalysts weight and
reaction time, respectively).
Fig. 3a shows the comparison of N2 adsorption-desorption isotherms
of the tested ZSM-35 zeolites. Both the CZ35 and parent NZ35 zeolites
corresponded to typical type Ⅰ isotherm, indicating the existence of rich
micropores. However, as summarized in Table 1, the NZ35 zeolite
3

X. Feng, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 2. SEM images (a, c, e, g, i) and EDS analysis (b, d, f, h, j) of zeolites: CZ35 (a) and (b), NZ35 (c) and (d), Hi-NZ350.2 (e) and (f), Hi-NZ350.4 (g) and (h), Hi-
NZ350.6 (i) and (j).
4

X. Feng, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 3. N2 adsorption-desorption isotherms (a) and pore size distribution (b) of the CZ35, NZ35 and Hi-NZ35x samples.
Table 1
Textural and physicochemical properties of CZ35, NZ35 and Hi-NZ35x zeolites.
a
b
b
c
Sample
Crystallinity (%)
SBET
Sext
Vmicro
Vmeso
Weak acidic sitesd (mmol g−1
)
Strong acidic sitesd (mmol g−1
)
(m2 g−1
)
(m2 g−1
)
(cm3 g−1
)
(cm3 g−1
)
CZ35
80
100
99
217
296
303
320
310
26
0.113
0.163
0.147
0.158
0.145
0.056
0.085
0.102
0.131
0.115
0.390
0.403
0.387
0.398
0.385
0.255
0.302
0.308
0.314
0.315
NZ35
64
Hi-NZ350.2
Hi-NZ350.4
Hi-NZ350.6
96
92
111
100
87
a
b
c
SBET was calculated by multi-point BET equation.
Sext and Vmicro were calculated by the t-plot method.
Vmeso was calculated from BJH method.
d
Estimation by NH3 -TPD curves: weak and strong acidic sites are in the region of 100–300 °C and 300–600 °C, respectively.
Fig. 5. XRD patterns of CuZnAl catalyst before calcination (a), after calcination
(b) and after reduction (c).
Fig. 4. NH3-TPD profiles of the CZ35 (a), NZ35 (b), Hi-NZ350.2 (c), Hi-NZ350.4
(d) and Hi-NZ350.6 (e) zeolites.
modification. However, on Hi-NZ350.6 zeolite, a decrease of SBET with
310 m2 g-1 and mesopore volume with 0.145 cm3 g-1 were achieved,
which were attributed to the dissolution of zeolite into amorphous silica
during the hydrothermal process [23,27]. Thus, the above results sug-
gested that the ZSM-35 with hierarchical porosity structure could be
achieved only by treating with appropriate NaOH concentration.
showed higher BET surface area (SBET) of 296 m2 g−1 compared with
that of CZ35 zeolite (217 m2 g-1), since the small thickness of the na-
nocrystals. It had been proved that the SBET has an important role to the
catalytic stability in DME carbonylation reaction, and the larger SBET
value usually showed better catalyst life [26]. For the other alkali-
treated zeolites, all the isotherms were the combination of type Ⅰ and
type Ⅳ isotherm typical for hierarchical materials. Moreover, a broad
size distribution ranging from 1 to 100 nm was observed with BJH
equation from the desorption branch, suggesting the introduction of
mesopores (as illustrated in Fig. 3b). As expected, the Hi-NZ350.4 zeo-
lite showed the highest SBET of 320 and external surface area (Sext) of
111 m2 g-1. The results from both SBET and the relative intensities in
XRD patterns suggested that the channels of ZSM-35 zeolite can be
purified by cleaning the amorphous phase under mild alkaline
3.1.4. NH3-TPD analysis of the zeolites
The acidic properties of the zeolites were studied by NH3-TPD
analysis. As illustrated in Fig. 4, two NH3 desorption peaks were found
at low and high temperatures, which meant the NH3 desorption from
weak acid sites and strong acid sites, respectively. According to former
reports, the weak acid sites were generally assigned to weak Lewis acid
sites (e.g., extra-framework aluminum), while the strong acid sites were
attributed to the framework Al (Brønsted acid sites) and strong Lewis
5

X. Feng, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Table 2
Physicochemical characteristics of CuZnAl catalyst.
Catalyst
Cu/Zn/Ala
SBET
Sext
Vtol.
Vmeso
Average crystal sizeb
(nm)
(molar ratio)
(m2 g−1
)
(m2 g−1
)
(cm3 g−1
)
(cm3 g−1
)
Cu
ZnO
10.6
CuZnAl
a
25.8/24.8/2.4
61
43
0.384
0.372
9.2
The elemental composition molar ratio was determined by EDS analysis.
b
Crystal sizes were calculated with the Scherrer formula from XRD patterns at 43.28° and 36.38° for the metallic copper and zinc oxide, respectively.
Fig. 6. SEM image (a) and EDS analysis (b) of CuZnAl catalyst.
Fig. 7. N2 adsorption-desorption isotherm (a) and pore size distribution curve (b) of CuZnAl catalyst.
acid sites of ZSM-35 zeolite [24,28]. As shown in Table 1, the weak acid
sites amount of CZ35 zeolite was similar to that of NZ35 zeolite,
whereas the strong acid sites amount was lower. It had been demon-
strated that the catalytic activity was directly determined by the
quantity of the strong acid sites, instead of the weak acid sites, for DME
carbonylation on zeolite catalysts [26]. Therefore, the larger quantity of
strong acid sites on NZ35 zeolite contributes to excellent catalytic ac-
tivity in DME carbonylation reaction. Compared with NZ35 zeolite, the
quantity of the strong acidic sites on the Hi-NZ35x samples slightly
increased, while a reverse trend was observed for the weak acid ones.
This result could be explained by partial extraction of Si atoms, which
was also consistent with former reported results [24,29].
3.1.5. Characterization of CuZnAl catalyst
Fig. 5 shows the XRD patterns of the CuZnAl catalyst. As depicted in
Fig. 5, the peaks of aurichalcite (Cu, Zn)5(CO3)2(OH)6 and malachite
Cu2(CO3)(OH)2 were found over the uncalcined CuZnAl catalyst, while
no diffraction peaks of Al species were found [30]. After calcination at
350 °C for 1 h, the aurichalcite and malachite decomposed into CuO and
Fig. 8. H2-TPR profile of CuZnAl catalyst.
6

X. Feng, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 9. Conversion of DME carbonylation over CZ35, NZ35 and Hi-NZ35X
zeolites; Reaction conditions: T = 220 °C; P = 1.5 MPa; weight (zeolite) = 0.5
g; feed gas: Ar/DME/CO/H2= (1.51/2.06/46.43/50.00); flow rate = 40 mL
Fig. 11. Conversion of DME over combination of (a) Hi-NZ350.4 zeolite with
CuZnAl catalyst and (b) CZ35 zeolite with CuZnAl catalyst for ethanol synth-
esis; Reaction conditions: T = 220 °C; P = 1.5 MPa; weight (Hi-
NZ350.4) = 0.5 g, weight (CZ35) = 0.5 g, weight (CuZnAl) = 0.5 g; feed gas:
min−1
.
Ar/DME/CO/H2= (1.51/2.06/46.43/50.00); flow rate = 40 mL min−1
.
ZnO. The CuO phase will be reduced to Cu metal after reduction pro-
cess. Table 2 lists the crystallite size of Cu and ZnO calculated by using
Scherrer formula. For the SEM image of CuZnAl catalyst, some small
irregular particles were observed. As shown in Fig. 6b, the real Cu/Zn/
Al molar ratio measured by EDS was very close to the initial preparation
molar recipe. Fig. 7a and b present N2 adsorption-desorption isotherm
and pore size distribution of CuZnAl catalyst, respectively. The
isotherm belonged to the type Ⅳ with an apparent hysteresis loop at P/
P0 = 0.8, indicating its typical mesopore structure. As displayed in
Fig. 8, H2-TPR was performed to illustrate the reduction property of
CuZnAl catalyst. The small peak at low temperature is due to the re-
duction of highly dispersed CuO species, whereas the peak appearing at
high temperature is attributed to the reduction in dispersed CuO with
binary support of Al-doped ZnO [13].
Table 3
3.2. The single DME carbonylation on CZ35, NZ35 and Hi-NZ35x zeolites
Product selectivity of DME carbonylation on CZ35, NZ35 and Hi-NZ35x zeolites.
We first evaluated the catalytic performance of zeolite catalysts for
single DME carbonylation to generate MA. Fig. 9 presents the DME
conversion as a function of time on stream over CZ35, NZ35 and Hi-
NZ35x zeolites at 220 °C and 1.5 MPa. For the CZ35 zeolite, DME con-
version increased from 19.9% at 1 h to the maximum value of 22.4% at
2 h, but it decreased sharply to 14.8% at 8 h, showing a rapid deacti-
vation pattern. As for the NZ35 zeolite, the catalytic activity and sta-
bility of DME carbonylation were notably enhanced in comparison with
CZ35 zeolite. The DME conversion of the NZ35 zeolite increased
quickly from 14.4% at 1 h to 32.2% at 9 h and then became stable at
around 26.2%. The Hi-NZ35X zeolites were evaluated for DME carbo-
nylation under the same reaction conditions. For the alkali-treated
nanosheet zeolites, more excellent catalytic performance was observed
on Hi-NZ350.2 and Hi-NZ350.4 zeolites, whereas a low catalytic activity
was obtained on the Hi-NZ350.6 zeolite. As shown in Fig. 9, for Hi-
NZ350.2 zeolite, the DME conversion increased from 20.7% to 38.3%
and then kept constant at around 31.0%. For Hi-NZ350.4 zeolite, the
promoting effect of alkaline treatment was more obvious. The DME
conversion on this Hi-NZ350.4 zeolite firstly increased up to 42.0% and
then maintained at about 37.4%. Meanwhile, the catalytic stability of
both Hi-NZ350.2 and Hi-NZ350.4 zeolites was comparable to that of the
parent NZ35 zeolite. However, for Hi-NZ350.6 zeolite, the DME con-
version decreased gradually to 14.9% at 24 h, indicating its poor cat-
alytic stability. The unsatisfied catalytic performance of Hi-NZ350.6
zeolite should be attributed to its lower SBET, smaller pore volume and
damaged structure, as confirmed by its BET analysis [26]. The MA se-
lectivity in products on all the zeolites catalysts was similar, keeping at
high selectivity above 90% as given in Table 3. Herein, a small amount
of CO2 and CH4 was also detected as the byproducts, which was mainly
attributed to water-gas shift reaction (WGSR) and the oligomerization
reactions, respectively [14,31].
Catalysts
Conv. (%)
Selectivity (%)
DME
MA
Methanol
CH4
CO2
CZ35
14.8
26.1
31.0
37.4
15.0
91.0
93.6
95.2
95.1
92.1
4.7
2.4
3.0
2.6
5.4
3.5
2.6
1.4
1.3
1.7
0.8
1.4
0.4
1.0
0.8
NZ35
Hi-NZ350.2
Hi-NZ350.4
Hi-NZ350.6
Reaction conditions: T = 220 °C; P = 1.5 MPa; weight (zeolite) = 0.5 g; feed
gas: Ar/DME/CO/H2= (1.51/2.06/46.43/50.00); flow rate = 40 mL min−1
;
MA = methyl acetate; for CZ35; reaction time = 8 h; for other zeolite, reaction
time = 24 h.
Fig. 10. Deactivation rates constants on CZ35, NZ35 and Hi-NZ35x. Reaction
conditions: T = 220 °C; P = 1.5 MPa; weight (zeolite) = 0.5 g; feed gas: Ar/
DME/CO/H2 = 1.51/2.06/46.43/50.00; flow rate = 40 mL/min.
The above results indicated that NZ35 zeolite exhibited better cat-
alytic performance than CZ35 zeolite in DME carbonylation reaction.
7

X. Feng, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Table 4
Product selectivity and ethanol productivity of ethanol synthesis over combination of CZ35 or Hi-NZ350.4 zeolite and CuZnAl catalyst.
Catalysts
Productivity
Conv. (%)
Selectivity (%)
(mmol kg−1 h−1
)
DME
MA
Ethanol
Methanol
CH4
CO2
CZ35 and CuZnAl
474.5
840.2
10.0
42.9
0.5
1.2
42.2
43.0
49.3
49.2
1.6
0.8
6.4
5.8
Hi-NZ350.4 and CuZnAl
Reaction conditions: T = 220 °C; P = 1.5 MPa; weight (Hi-NZ350.4) = 0.5 g, weight (CZ35) = 0.5 g, weight (CuZnAl) = 0.5 g; feed gas: Ar/DME/CO/H2= (1.51/
2.06/46.43/50.00); flow rate = 40 mL min−1, MA: methyl acetate; time = 24 h.
Furthermore, the catalytic activity of DME carbonylation was further
improved over the hierarchical nanosheet ZSM-35 zeolites. Among the
tested zeolites, Hi-NZ350.4 exhibited the highest DME conversion of
42.0%. At the same time, it also demonstrated a higher MA selectivity
of 95.1% and better stability. The deactivation rate constants (kd) from
DME carbonylation on the CZ35, NZ35 and Hi-NZ35x were displayed in
4. Conclusions
In conclusion, we designed and prepared a nanosheet HZSM-35
(NZ35) zeolite and a series of hierarchical nanosheet HZSM-35 zeolites
(Hi-NZ35x). For single DME carbonylation reaction, the NZ35 zeolite
showed better catalytic performance than that of the commercial re-
ference HZSM-35 (CZ35) zeolite, because of the larger amount of fra-
mework Brønsted acidic sites and the shorter diffusion pathways of
NZ35 zeolite. Additionally, the specific hierarchical porosity structure
of NZ35 zeolite was obtained by an alkaline treatment method to get
Hi-NZ35x zeolites. The Hi-NZ350.2 and Hi-NZ350.4 zeolites showed
more efficient catalytic performance for DME carbonylation than NZ35
zeolite, since their improved capacity on mass diffusion. For Hi-NZ350.4
zeolite, the conversion of DME reached up to 42.0% with the MA se-
lectivity of 95.1%. By using the combination of Hi-NZ350.4 zeolite and
CuZnAl catalyst in a dual-catalyst bed, more efficient ethanol synthesis
from DME and syngas was realized. The DME conversion was stable at
Fig. 10. The kd values, calculated from the MA formation rate (RMA
)
according to the procedure reported in the literature [32], were
0.16 h−1 for the CZ35 and 0.05 h-1 for the NZ35 zeolite. Clearly, the
NZ35 zeolite had a lower deactivation rate constant than that of the
CZ35 zeolite. Compared with the CZ35 zeolite, the higher DME con-
version of NZ35 zeolite is mainly due to the larger number of strong
acid sites. The excellent stability is mainly attributed to the better mass
transfer and the less diffusion resistance on nanosheet zeolite, thus in-
hibiting its deactivation. In particular, some treatments in mild alkaline
solutions enable the zeolites higher mesopore volume. Therefore, Hi-
NZ350.2 and Hi-NZ350.4 zeolite can further enhance its catalytic activity
mainly because the formed mesopores facilitate the diffusion rate of
reactants and products. Therefore, in the following section, the com-
bination of Hi-NZ350.4 zeolite and CuZnAl catalyst was adopted for
ethanol synthesis from DME and syngas.
42.9% and the ethanol productivity reached 840.2 mmol kg−1 h−1
,
much better than that of the combination of the commercial CZ35
zeolite with CuZnAl catalyst. The more efficient ethanol synthesis route
proposed here, using the combination of hierarchical nanosheet HZSM-
35 zeolite and Cu-based catalyst, is an efficient catalytic process for
syngas conversion to produce ethanol. More importantly, this new al-
ternative ethanol synthetic approach from syngas and DME will reduce
the dependence on petroleum resources and cut the emission of the
greenhouse gas.
3.3. Ethanol synthesis with the combination of zeolite and CuZnAl catalyst
The ethanol synthesis process was carried out on combination of Hi-
NZ350.4 or CZ35 zeolite with CuZnAl catalyst in a dual-catalyst bed
reactor. DME and CO were firstly introduced into the upper zeolite bed
to perform the carbonylation reaction, and the H2 was directly im-
ported, by an independent tube, into the second CuZnAl catalyst bed to
accomplish the hydrogenation of MA from the first catalyst bed. The
DME conversion and product selectivity were illustrated in Fig. 11 and
Table 4, respectively. For the combination of reference CZ35 zeolite
with CuZnAl catalyst, the DME conversion increased from 23.4% at 1 h
to maximum of 25.6% at 2 h, followed by decreasing to 10.0% at 24 h
with ethanol selectivity of 42.2%. By combining Hi-NZ350.4 zeolite and
CuZnAl catalyst, more stable and higher DME conversion of 47.2% was
achieved with ethanol selectivity of 43.0%. The ethanol productivity on
Acknowledgments
X. F. thanks the China Scholarship Council (CSC) for financial
support. This study was financially supported by the grant from JST
ACT-C (Grant Number: JPMJCR12YT) and JST SATREPS, Japan.
References
[1] A.E. Farrell, R.J. Plevin, B.T. Turner, A.D. Jones, M. O’Hare, D.M. Kammen, Science
311 (2006) 506–508.
[2] H. Zabed, J.N. Sahu, A.N. Boyce, G. Faruq, Renew. Sustain. Energy Rev. 66 (2016)
751–774.
this improved zeolite combination reached up to 840.2 mmol kg−1 h−1
.
Here, both of DME conversions over dual-catalyst bed reactor were
slightly higher than those of single zeolite bed under the same reaction
condition. This result should be ascribed to the synergistic and shift
effect of two sequential reactions (the preliminary carbonylation and
the subsequent hydrogenation). On the other hand, methanol as an-
other by-product can be easily in suit recycled into DME through de-
hydration reaction on solid acid catalysts, realizing a green and eco-
nomical route for ethanol synthesis. Herein, for the combination of Hi-
NZ350.4 zeolite and CuZnAl catalyst, the MA selectivity was only 1.2%,
significantly lower than that of DME carbonylation on the single Hi-
NZ350.4 zeolite bed, which suggested that the generated MA from the
first zeolite layer was mainly hydrogenated on the second CuZnAl
catalyst bed. However, the CO2 selectivity was 5.8%, slightly higher
than 1.0% over the single DME carbonylation reaction. It might be
caused by the WGSR that was promoted owing to the presence of Cu-
based catalyst.
[3] K. Göransson, U. Söderllind, J. He, W. Zhang, Renew. Sustain. Energy Rev. 15
(2011) 482–492.
[4] V. Subramani, S.K. Gangwal, Energy Fuels 22 (2008) 814–839.
[5] Y. Man, P. Liu, J. Am. Chem. Soc. 131 (2009) 13054–13061.
[6] H. Yue, X. Ma, J. Gong, Acc. Chem. Res. 47 (2014) 1483–1492.
[7] J.J. Spivey, A. Egbebi, Chem. Soc. Rev. 36 (2007) 1514–1528.
[8] J. Gong, H. Yue, Y. Zhao, S. Zhao, L. Zhao, J. Lv, S. Wang, X. Ma, J. Am. Chem. Soc.
134 (2012) 13922–13925.
[9] Y. Chen, D. Liu, Fuel Process. Technol. 171 (2018) 301–307.
[10] Q. Qian, M. Cui, J. Zhang, J. Xiang, J. Song, G. Yang, B. Han, Green Chem. 20
(2018) 206–213.
[11] X. San, Y. Zhang, W. Shen, N. Tsubaki, Energy Fuels 23 (2009) 2843–2844.
[12] G. Yang, X. San, N. Jiang, Y.K. Tanaka, X. Li, Q. Jin, K. Tao, F. Meng, N. Tsubaki,
Catal. Today 164 (2011) 425–428.
[13] Q. Wei, G. Yang, X. Gao, L. Tan, P. Ai, P. Zhang, P. Lu, Y. Yoneyama, N. Tsubaki,
Chem. Eng. J. 316 (2017) 832–841.
[14] P. Cheung, A. Bhan, G.J. Sunley, E. Iglesia, Angew. Chem. Int. Ed. 45 (2006)
1617–1620.
[15] J. Liu, H. Xue, X. Huang, Y. Li, W. Shen, Catal. Lett. 139 (2010) 33–37.
[16] Y. Yuan, L. Wang, H. Liu, P. Tian, M. Yang, S. Xu, Z. Liu, Chin. J. Catal. 36 (2015)
8

X. Feng, et al.
1910–1919.
CatalysisTodayxxx(xxxx)xxx–xxx
Catal. 312 (2014) 204–215.
[17] M. Liu, W. Jia, J. Li, Y. Wang, S. Ma, H. Chen, Z. Zhu, Catal. Lett. 146 (2016)
249–254.
[25] H. Liu, S. Xie, W. Xin, S. Liu, L. Xu, Catal. Sci. Technol. 6 (2016) 1328–1342.
[26] Y. Liu, N. Zhao, H. Xian, Q. Cheng, Y. Tan, N. Tsubaki, X. Li, ACS Appl. Mater.
Interfaces 7 (2015) 8398–8403.
[18] T. Shishido, Y. Yamamoto, H. Morioka, K. Takaki, K. Takehira, Appl. Catal. A Gen.
263 (2004) 249–253.
[27] W.Yoo.X. Zhang, M. Tsapatsis, A. Stein, Microporous Mesoporous Mater. 149
(2012) 147–157.
[19] R.B. Borade, A. Clearfield, Zeolites 14 (1994) 458–461.
[20] Y.P. Khitev, Y.G. Kolyagin, I.I. Ivanova, O.A. Ponomareva, F. Thibault-Starzyk, J.-
P. Gilson, C. Fernandez, F. Fajula, Microporous Mesoporous Mater. 146 (2011) 207-
207.
[28] M. Milina, S. Mitchell, N.-L. Michels, J. Kenvin, J. Pérez-Ramírez, J. Catal. 308
(2013) 398–407.
[29] W. Yu, L. Deng, P. Yuan, D. Liu, W. Yuan, F. Chen, Chem. Eng. J. 270 (2015)
450–458.
[21] W. Chu, X. Li, X. Zhu, S. Xie, C. Guo, S. Liu, F. Chen, L. Xu, Microporous Mesoporous
Mater. 240 (2017) 189–196.
[30] Q. Zhu, Q. Zhang, L. Wen, Fuel Process. Technol. 156 (2017) 280–289.
[31] X. Li, X. San, Y. Zhang, T. Ichii, M. Meng, Y. Tan, N. Tsubaki, ChemSuschem 3
(2010) 1292-1199.
[22] X. Gao, B. Xu, G. Yang, X. Feng, Y. Yoneyama, U. Taka, N. Tsubaki, Catal. Sci.
Technol. 8 (2018) 2087–2097.
[23] X. Li, X. Liu, S. Liu, S. Xie, X. Zhu, F. Chen, L. Xu, RSC Adv. 3 (2013) 16549–16557.
[24] H. Xin, X. Li, Y. Fang, X. Yi, W. Hu, Y. Chu, F. Zhang, A. Zhang, H. Zhang, X. Li, J.
[32] O. Levenspiel, J. Catal. 25 (1972) 265–272.
9