Conversion of ethanol to 1,3-butadiene over high-performance Mg-ZrO: X/MFI nanosheet catalysts via the two-step method

Article abstract of DOI:10.1039/d0gc00290a

Mg-Zr/MFI nanosheet (NS) catalysts were prepared by a wet impregnation method for ethanol conversion to 1,3-butadiene (1,3-BD) via the two-step method in a dual fixed bed reaction system. Compared with Zr catalysts loaded on MFI(micro) or commercial SiO₂, 16%Zr/MFI(NS) gave the better performance, with 42.3% 1,3-BD selectivity and 60.5% total conversion of ethanol and acetaldehyde. Introducing 1.2 wt% Mg to 16%Zr/MFI(NS) improved the 1,3-BD selectivity to 54.7% at the expense of a 6% drop in the catalytic activity. Reaction conditions imposed remarkable influence on the reaction results. When the reaction was conducted at 350 °C, a WHSV of 1.44 h^-1 and a 2 : 1 ratio of ethanol to acetaldehyde, the 1,3-BD selectivity reached 74.6% with 41.5% total conversion. Such high performance over 1.2%Mg-16%Zr/MFI(NS) was maintained well in a 7 day (168 h) run without deactivation. The catalysts were characterized by XRD, N₂ adsorption, UV-Vis, Raman, and infra-red spectroscopy, NH₃-TPD, TEM and TG. The results showed that the Zr species on MFI(NS) are well distributed with the highest dispersion as compared with the microporous MFI and SiO₂ supported Zr catalysts. The Zr species preferentially occupied the silanol nests of MFI(NS) and eliminated the Br?nsted acid sites at 4 wt% Zr loading, and afforded abundant Lewis acid sites in the form of Zr(OH)(OSi)₃ when the Zr loading was increased to 16 wt%. As a base site, Mg is inactive for MPVO reduction but slightly active for the aldol condensation of acetaldehyde, both of which are much inferior to that of the Lewis acid sites. The 1.2%Mg-16%Zr/MFI(NS) catalyst with hierarchical structures of meso- and micro-pores, abundant weak Lewis acid sites but nearly no Br?nsted acid sites is competent for the two-step ethanol to 1,3-BD conversion process with high activity, selectivity and stability.

Full text of DOI:10.1039/d0gc00290a

Green Chemistry
View Article Online
View Journal
PAPER
Conversion of ethanol to 1,3-butadiene over
high-performance Mg–ZrO /MFI nanosheet
x
Cite this: DOI: 10.1039/d0gc00290a
catalysts via the two-step method†
a,b
a
a
a
a
Xianquan Li, Jifeng Pang, Chan Wang, Lin Li, XiaoLi Pan,
a,c
a
Mingyuan Zheng
*
and Tao Zhang
Mg–Zr/MFI nanosheet (NS) catalysts were prepared by a wet impregnation method for ethanol conversion
to 1,3-butadiene (1,3-BD) via the two-step method in a dual fixed bed reaction system. Compared with Zr
2
catalysts loaded on MFI(micro) or commercial SiO , 16%Zr/MFI(NS) gave the better performance, with
4
2.3% 1,3-BD selectivity and 60.5% total conversion of ethanol and acetaldehyde. Introducing 1.2 wt% Mg
to 16%Zr/MFI(NS) improved the 1,3-BD selectivity to 54.7% at the expense of a 6% drop in the catalytic
activity. Reaction conditions imposed remarkable influence on the reaction results. When the reaction
was conducted at 350 °C, a WHSV of 1.44 h− and a 2 : 1 ratio of ethanol to acetaldehyde, the 1,3-BD
selectivity reached 74.6% with 41.5% total conversion. Such high performance over 1.2%Mg–16%Zr/MFI
1
(
NS) was maintained well in a 7 day (168 h) run without deactivation. The catalysts were characterized by
XRD, N adsorption, UV-Vis, Raman, and infra-red spectroscopy, NH -TPD, TEM and TG. The results
showed that the Zr species on MFI(NS) are well distributed with the highest dispersion as compared with
the microporous MFI and SiO supported Zr catalysts. The Zr species preferentially occupied the silanol
nests of MFI(NS) and eliminated the Brønsted acid sites at 4 wt% Zr loading, and afforded abundant Lewis
acid sites in the form of Zr(OH)(OSi) when the Zr loading was increased to 16 wt%. As a base site, Mg is
2
3
2
3
inactive for MPVO reduction but slightly active for the aldol condensation of acetaldehyde, both of which
are much inferior to that of the Lewis acid sites. The 1.2%Mg–16%Zr/MFI(NS) catalyst with hierarchical
structures of meso- and micro-pores, abundant weak Lewis acid sites but nearly no Brønsted acid sites is
competent for the two-step ethanol to 1,3-BD conversion process with high activity, selectivity and
stability.
Received 23rd January 2020,
Accepted 8th April 2020
DOI: 10.1039/d0gc00290a
rsc.li/greenchem
gas exploitation, more ethylene has been produced via the
ethane dehydrogenation process, which has decreased the
1
. Introduction
1
,3-Butadiene (1,3-BD) is widely used as a monomer for the demand for naphtha cracking. Moreover, in China, the suc-
production of a variety of synthetic polymers such as rubbers, cessful commercialization of the methanol to olefin conver-
1
,2
resins and elastomers. Boosted by the rapid growth of the sion technology has led to 4.5 million tonnes per year ethylene
world economy, 1,3-BD consumption has continuously productivity, accounting for 1/2 of the ethylene production
7
increased in the past decade and reached 12 million tonnes in from the naphtha cracking process in China. Therefore, it is
3
,4
2
018. Nowadays, 1,3-BD is predominantly produced by the necessary to explore alternative routes for 1,3-BD production
extractive distillation of C fractions in the naphtha cracking beyond the naphtha cracking method, particularly in a renew-
process for ethylene production, which largely determines the able way.
4
5
,6
1
,3-BD production. However, with the development of shale
Ethanol is the most productive chemical derived from
biomass. The rapid development in biotechnology and cata-
lytic technology like methanol conversion to ethanol makes
ethanol suitable for use in the synthesis of many important
a
Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023,
P.R. China. E-mail: myzheng@dicp.ac.cn
4
,8,9
chemicals, including 1,3-BD.
typical ways for the catalytic conversion of ethanol to 1,3-BD
ETB). The one-step method, the so-called Lebedev process
Generally, there are two
b
University of Chinese Academy of Sciences, Beijing 100049, China
c
Dalian National Laboratory for Clean Energy, Dalian 116023,
(
People’s Republic of China
which was first developed in the 1920s, solely uses ethanol as
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0gc00290a
the feedstock. The other way is the two-step method, i.e., the
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
Ostromislensky process, in which ethanol is first dehydroge- trated. The Mg–Zr/MFI(NS) catalyst exhibited outstanding
nated to acetaldehyde which then reacts with ethanol to form activity, selectivity and stability in a one week run, and accord-
1
,3-BD. The widely accepted reaction pathway of ETB involves ingly provided a good reference for designing high perform-
four reactions as shown in Scheme 1: (1) ethanol dehydrogena- ance catalysts for the ETB process.
tion to acetaldehyde, (2) aldol condensation of acetaldehyde,
(
3) MPVO reduction of crotonaldehyde with ethanol, and (4)
1
0
dehydration of crotyl alcohol into 1,3-BD. Evidently, this
process involves multiple reaction steps and a complex reac-
tion network, and the reaction intermediates are active and
apt to undergo side reactions. This makes it difficult to realize
high reaction selectivity, which usually varies in the range of
2
. Experimental
2.1 Catalyst preparation
A series of xMg–yZr/MFI(NS) catalysts were synthesized by an
incipient wetness impregnation method. In detail, MFI(NS)
10
1
,2
was prepared following the method reported previously.
4
0–75%.
Meanwhile, the catalyst stability is poor. The
Tetrabutylphosphonium hydroxide (TBPOH) was added dropwise
into tetraethyl orthosilicate under stirring, followed by deionized
water addition. The mixture was stirred for 12 h, and a clear sol
general catalyst regeneration cycle is about 40 h for the one-
3
,11
step process and 144 h for the two-step process.
Therefore,
developing catalysts with high activity, selectivity, and particu-
larly highly stability is of great significance for the renaissance
of ETB application.
with the composition of 1SiO : 0.3TBPOH : 10H O : 4EtOH (mole
2
2
ratio) was formed. The sol was sealed in a Teflon-lined stainless
steel autoclave and heated for 40 h in a pre-heated oven operat-
ing at 388 K. Then, the precipitated white powders were collected
by filtration and dried at 373 K. The dried sample was calcined
at 823 K for 4 h to remove the organics and obtain MFI(NS).
Then, MFI(NS) was impregnated with zirconium oxynitrate
aqueous solution. After drying at 120 °C overnight, the sample
was calcined at 550 °C for 4 h. Then, Mg was introduced to the
as-prepared sample by wetness impregnation of magnesium
nitrate solution, and then dried and calcined again to obtain
xMg–yZr/MFI(NS) catalysts. The Mg loading “x” varied from
According to the reaction and deactivation mechanisms of
ETB, the catalyst with high availability of active sites and high
efficiency of transfer of intermediates and products in the
course of the reaction would have outstanding performance.
In our previous study, we developed a Li–Zn–Hf catalyst sup-
ported on MFI nanosheets (NS), and obtained superior stabi-
lity and 1,3-BD yield compared to that on microporous zeolite
1
0
in the one-step ETB process. In view of the notable advan-
tages of the two-step process over the one-step ETB process in
terms of the reaction stability, it is more attractive to explore
novel high performance catalysts for the former process.
Moreover, in the two-step process, the catalyst in the second
reactor does not require active sites for ethanol dehydrogena-
tion. Therefore, it is more convenient to identify the active site
functions, modulate active sites according to the catalytic be-
havior, and design high performance catalysts rationally.
In this work, we prepared a Zr catalyst supported on MFI
0.48 wt% to 19.2 wt% and the active component Zr loading “y”
varied from 4 wt% to 24 wt%. Mesoporous SiO and micro-
2
porous MFI were also used as supports to load Zr and Mg by the
incipient wetness impregnation method. The Zr/MFI(NS) and
Mg/MFI(NS) catalysts were also prepared by the same method
mentioned above, but without the introduction of the second
metal. The supports of microporous MFI and MFI nanosheets
were synthesized following the method reported in the
(NS), and compared it with conventional microporous MFI and
10,12
literature.
mesoporous silica supported catalysts in the ETB reaction.
Then, Mg was introduced to Zr/MFI(NS) to modulate the acidic
properties of the catalyst and enhance the reaction selectivity.
2.2 Catalyst characterization
According to comprehensive characterization and conditional X-ray diffraction (XRD) patterns of the catalysts were recorded
experiments, the features of MFI(NS) supported catalysts and using a PANalytical X’Pert-Pro powder X-ray diffractometer
the function of active sites for the ETB reaction were illus- with Cu-Kα monochromated radiation (λ = 0.1541 nm) at a
scanning speed of 5° min− in the range of 5–50°.
1
2
N adsorption and desorption experiments of the catalysts
were conducted on a Micromeritics ASAP 2460 system at 77 K.
Before measurements, the samples were pretreated at 300 °C
for 8 h under vacuum conditions. The specific surface areas
were calculated according to the BET equation. The micropore
volume, mesopore volume and external surface areas were
measured by the t-plot method and the adsorption isotherm
was determined by the Barrett–Joyner–Halenda (BJH) method.
Thermogravimetric analysis (TG) was performed on a TA
SDTQ600 analyzer at a temperature-programmed rate of 10 °C
−
1
min from room temperature to 800 °C.
The acidity of the catalysts was measured by infra-red spec-
troscopy of pyridine adsorption using a Bruker Tensor 27
instrument. The samples were pretreated at 350 °C for 1 h in a
Scheme 1 Reaction pathway for the ETB process.
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
vacuum of 1.0 × 10− Pa and then pyridine vapor was intro- products were analyzed with an online GC (gas chromato-
duced at room temperature for 1 min. After that, the pyridine- graph, Panna GC A91) equipped with a FID detector and a
adsorbed system was heated and evacuated at 150 °C, 250 °C 30 m HP-PLOT Q Column (50 mm × 30 mm × 20 mm). The
and 350 °C, respectively, and then the spectra were collected at reactants and products including 1,3-BD, ethylene (ELE),
ambient temperature. The acid quantity was calculated by the ethanol (ETOH), acetaldehyde (ACH), crotonaldehyde (CRH),
2
2
following equations: C (pyridine on B sites) = 1.88 IA(B)R /W, and diethyl ether (DEE) were quantified.
2
and C (pyridine on L sites) = 1.42 IA(L)R /W, where C = concen-
The conversion and product selectivity (based on the reac-
−1
tration (mmol g ), IA (B or L) = integrated absorbance of B or tion in reactor 2) were calculated according to the following
L bands (cm− ), R = radius of the catalyst disk (cm), and W = equations:
weight of the disk (mg).
FTIR of CO adsorption on the catalysts was performed on a
Bruker Vertex 70v spectrometer equipped with a DTGS detec-
tor. The sample was pretreated at 300 °C for 1 h in a vacuum
of 1.0 × 10− Pa and then cooled to −173 °C with liquid nitro-
gen. The FTIR spectra of CO adsorption were collected
at different CO pressures, i.e., 0.5, 1, 2, 5 and 10 Torr,
respectively.
The acidic properties of the samples were also characterized
3
by NH -TPD on a Micromeritics Autochem II apparatus,
equipped with a TCD detector. The catalyst was loaded in a
U-shaped fixed-bed quartz microreactor, and pretreated at
00 °C (ramping rate = 10 °C min ) for 1 h under flowing He
1
Total conversionð%Þ ¼ ðC molethanol tab acetaldehydeꢀ
ðC molunreacted ethanol þ C molunreacted acetaldehydeÞÞ= ð1Þ
C molethanol tabacetaldehyde ꢁ 100%;
5
Ethanol conversionð%Þ ¼ ðC molethanol ꢀ C molunreacted ethanolÞ=
C molethanol ꢁ 100%;
ð2Þ
Acetaldehyde conversionð%Þ ¼
ðC molacetaldehyde ꢀ C molunreacted acetaldehydeÞ=C molacetaldehyde ꢁ 100%;
ð3Þ
−1
3
(
Product selectivityð%Þ ¼ C molproduct=ðC molethanol tab acetaldehydeꢀ
20 mL min− ). When the sample was cooled to 50 °C, 1 ml
NH gas was injected with a syringe many times until the sat-
uration of NH adsorption. When the baseline of the TCD
signal was stable, the sample was heated to 800 °C at a
ramping rate of 10 °C min and the TCD signal was recorded.
Transmission electron microscopy (TEM) images, high-
angle annular dark-field scanning-transmission electron
microscopy (HAADF-STEM) images and energy-dispersive X-ray
1
ðC molunreacted ethanol þ C molunreacted acetaldehydeÞÞ ꢁ 100%:
3
ð4Þ
3
−
1
3
3
. Results and discussion
.1 Effects of Zr loading on the ETB reaction
spectroscopy (EDS) elemental mapping were obtained by using Zirconium is a low cost element and has been widely used in
1
3–15
a JEM-2100F microscope, operated at a voltage of 200 kV.
catalysts to provide acid sites for the ETB reaction.
In this
The morphology of the catalysts was observed by scanning study, Zr was used as the active component for preparing the
electron microscopy (SEM) using
microscope.
a
JEOL JSM-7800F MFI(NS) supported catalysts.
The performances of Zr-based catalysts at different loadings
UV-vis diffuse reflectance spectra of the samples were in ETB were evaluated and are listed in Table 1, entries 1–5.
obtained on a PerkinElmer Lambda 950 spectrometer in the For the blank MFI(NS) support in the ETB reaction, only ethyl-
transmittance mode from 200 nm to 800 nm.
ene and diethyl ether were detected, demonstrating that a
1
6,17
Raman spectra of the catalysts were collected with the number of Brønsted acid sites were present in the zeolite.
+
5
23 nm line of the Ne ion laser on Jobin Yvon U-1000.
After inducing 4 wt% Zr to the MFI(NS) support, 1,3-BD was
obtained with 10.5% selectivity. Meanwhile, crotonaldehyde,
the aldol condensation product, was also produced at a selecti-
2
.3 Catalyst evaluation
The prepared catalysts were evaluated in a dual fixed bed reac- vity of 28.1%. Evidently, Zr provides active catalytic sites for
tion system. The acetaldehyde–ethanol mixture was produced aldol condensation and the MPVO reaction. Further increasing
in the upper reactor to serve as the feedstock for the second the Zr content in the catalyst enhanced the reactant conversion
(150 mL min− ) con- and MPVO reaction, and the highest 1,3-BD selectivity of
1
fixed-bed reactor. In detail, a flow of N
2
taining ethanol was introduced into the first fixed-bed reactor 37.1% was obtained over 16%Zr/MFI(NS). Thus, 16 wt% Zr was
loaded with 1 g of 20%Cu/SiO catalyst, and heated at 250 °C set as the optimal loading on MFI(NS) for further studies in
under atmospheric pressure. After a 24 h on-line induction the following sections.
2
period for catalyst activation, ethanol was converted to acet-
Over the Zr/MFI(NS) catalyst, the selectivity to ethylene and
aldehyde at ca. 96% selectivity and ca. 50% conversion. As diethyl ether was remarkably lower than that over the blank
shown in Fig. S1,† the reaction performance was highly stable MFI(NS) support. It is well known that ethanol dehydration
in a 200 h run (ca. 8 days). Then, the in situ generated feed- reactions are remarkably promoted by the Brønsted acid sites
1
8
stock stream of an ethanol–acetaldehyde (nearly at a ratio of in catalysts. The depressed selectivity to dehydration pro-
: 1) mixture in the N flow entered the second fixed-bed ducts over the Zr/MFI(NS) catalyst suggests that the introduc-
reactor loaded with 1 g of catalyst for the ETB reaction. The tion of Zr onto the MFI(NS) support remarkably decreased the
1
2
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
a
Table 1 Catalytic performance of various Zr catalysts in the ETB process
Conv. (%)
Product selectivity (%)
Entry
Catalysts
TC
7.4
ETOH
ACH
ELE
1,3-BD
DEE
CRH
C-Balance (%)
1
2
3
4
5
6
7
8
9
MFI(NS)
13.7
29
55.9
68.2
77.5
6.5
54.6
70.3
64.3
59.1
0.9
10.3
32.4
52.8
55.9
25.9
55.7
52.2
50.5
44.4
79
0
22
0
101
97
4%Zr/MFI(NS)
8%Zr/MFI(NS) b
16%Zr/MFI(NS)
24%Zr/MFI(NS)
16%Zr/MFI(micro)
19.6
44.2
60.5
64.7
16.2
55.2
61.3
57.4
51.7
12.8
8.8
4.9
7.4
3.6
4.2
4.4
5.1
6.3
10.5
24.5
42.3
31.5
18.2
32.8
37.1
36.4
34.4
37.6
9.4
3.6
3.3
9.4
1.3
3.2
4.3
6.4
28.1
8.3
4.9
4.9
55.4
5.7
2.7
4.9
5.9
78.3
73.2
65.8
97.8
69.1
67.8
71.7
75.7
16%Zr/SiO
2
16%Zr/MFI(NS) (450 °C)
16%Zr/MFI(NS) (650 °C)
16%Zr/MFI(NS) (750 °C)
1
0
a
Reaction conditions: T = 350 °C, WHSV = 1.92 h−1, ethanol : acetaldehyde = 1 : 1. Conv., TC, ETOH, ACH, ELE, 1,3-BD, DEE, CRH and C-balance
are abbreviations for conversion, total conversion, ethanol, acetaldehyde, ethylene, 1,3-butadiene, diethyl ether, crotonaldehyde, and carbon
b
balance, respectively. The data were collected after 4 h of reaction. The MFI(NS) support was calcined at 550 °C in entries 1–6.
quantity of acid sites, particularly the Brønsted acid sites. This was consumed for the MPVO reaction. According to the MPVO
conjecture was further manifested by the Py-IR spectra of cata- mechanism, a six-member ring transition state is involved in
2
2–25
lysts as discussed in the below context. According to Li and co- the reaction (Scheme 2).
In view of the fact that the pore
workers, when Zr species are loaded on de-Al Beta zeolite, the size of MFI(micro) is close to that of the intermediate (0.8 nm
metal ions preferably locate at the silica nest to form Lewis vs. ca. 0.6 nm), it is highly possible that the MPVO reaction is
1
9,20
acid sites.
In turn, the number of Brønsted acid sites notably prohibited by the spatial hindering effect in the micro-
decreased owing to the adjacent silanols in the nest interacting pores. This leads to the low selectivity of 1,3-BD but high
with Zr to form Si–O–Zr bonds. Therefore, in the MFI(NS) selectivity of crotonaldehyde over 16%Zr/MFI(micro) catalysts.
support studied herein, there should also be abundant silanol In addition, the micropore structure also hindered the conden-
nests existing, making MFI(NS) versatile to be modified with sation of acetaldehyde due to the less availability of active sites
various transition metals for designing effective catalysts.
over 16%Zr/MFI(micro).
In addition, it can be noted that in the presence of Zr
2
When commercial mesoporous SiO was used as the
species on MFI(NS), the carbon balance was lowered as com- support to load Zr sites for the ETB reaction in Table 1, entry
pared with that of MFI(NS). In view of the reasonably good 7, the 1,3-BD selectivity and reactant conversion were also
stability of 24%Zr/MFI(NS) in an 8 h run (Fig. S2†), this should inferior to those of 16%Zr/MFI(NS). This suggests that the
be largely attributed to the formation of undetectable large hierarchical structures of MFI(NS) facilitate the active site dis-
molecule products from acetaldehyde condensation over the persion and reactant transportation, and MFI(NS) is a superior
Lewis acid sites on Zr/MFI(NS).
support for the ETB conversion.
The thermal stability of MFI(NS) was checked by calcination
at different temperatures. There are abundant silanols on the
surface of MFI zeolite, and the over-high calcination tempera-
ture may cause loss of the quantity of silanol and affect the
3
.2 Effects of the support structure on the ETB reaction
The pore size and structure of catalysts usually impose remark-
able influence on the catalytic reactions owing to the mass
diffusion and spatial confinement effects. Herein, three
typical supports, i.e., MFI(micro), MFI(NS), and commercial
26
performance of the catalysts. The results listed in Table 1,
2
1
entries 4, 8–10, show that in a wide range of calcination temp-
eratures (450–750 °C), the MFI(NS) supported Zr catalysts
showed similar performances, and MFI(NS) calcined at 550 °C
was the relatively optimal one. Evidently, the nanosheets of
MFI zeolite are rather stable. This is consistent with the good
performance of the regenerated MFI(NS) catalysts as we
2
SiO , which possess microporous, hierarchical and meso-
porous pore structures, respectively, were used to load Zr cata-
lysts for studying the role of the pore structure in the ETB
reaction.
Compared to 16%Zr/MFI(NS) with the hierarchical structure
10
reported previously.
(
Table 1, entry 4), the microporous MFI supported Zr catalyst
showed significantly lower activity for the reactant conversion
16.2% vs. 60.5%) and low selectivity to 1,3-BD (18.2% vs.
2.3%) in Table 1, entry 6, but with an order of magnitude
(
4
higher selectivity to crotonaldehyde (55.4% vs. 4.9%). This
suggests that the MPVO reduction was seriously depressed
over the 16%Zr/MFI(micro) catalyst. The conversion of ethanol
and acetaldehyde over Zr/MFI(micro) was rather low (6%–
22
2
6%), also demonstrating that only a small amount of ethanol Scheme 2 Mechanism of the MPVO reaction, Me = metal.
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
.3 Catalytic performance of Mg–Zr/MFI(NS)
Paper
3
really catalyze the condensation reaction. However, the total
conversions of ethanol and acetaldehyde over them are rather
low (<10%), very close to that over the MFI(NS) support. This
demonstrates that the activity of base sites of Mg for the con-
densation reaction is much lower than that of the acid sites of
Zr on 16%Zr/MFI(NS), which gave 60% total conversion and
47% C product selectivity. This is consistent with the results
4
discussed above that the activity of Mg–Zr/MFI(NS) catalysts is
generally lower than that of 16%Zr/MFI(NS).
The key role of acid–base sites in the Lebedev process,
especially in the Mg–Si systems, was extensively
2
7–29
investigated.
Angelic et al. proposed that the balanced
acid and base sites with a moderate strength catalyzed the
aldol condensation step and depressed the side reactions in
3
0–32
the ETB reaction.
Thus we introduced base sites of Mg to
modulate the acidic properties of Zr/MFI and improve its cata-
lytic performance.
In addition, no 1,3-BD product was obtained over the Mg/
MFI(NS) catalyst, suggesting that Mg sites do not catalyze the
MPVO reaction. This is also consistent with the conclusion
from the performance of 19.2%Mg–16%Zr/MFI(NS) discussed
above. Therefore, the major role of Mg in the ETB reaction
over Mg–Zr/MFI catalysts lies in subtly tuning the acidity of cat-
alysts so that the rates of condensation and MPVO reactions
match well to lead to enhanced 1,3-BD selectivity.
3
.3.1 Effects of Mg loading on the ETB reaction over Mg–
Zr/MFI(NS) catalysts. The 16%Zr/MFI(NS) catalyst was modi-
fied with different amounts of Mg and tested in the ETB reac-
tion. As shown in Table 2, entries 1–5, the activity of Mg–Zr/
MFI(NS) catalysts was slightly lower than that of 16%Zr/MFI
(
NS) after Mg addition. However, the 1,3-BD selectivity was
improved, and the highest value of 54.7% was obtained over
the 1.2%Mg–16%Zr/MFI(NS) catalyst. At higher loadings of
Mg, the total conversions decreased more remarkably, demon-
strating that the presence of Mg in the Zr catalysts depressed
3.4 Reaction condition influence and catalyst stability
the ETB reaction. This is quite contrary to some previous argu- Because the ETB process consists of a series of cascade steps
ments that base sites of Mg greatly contribute to condensation which have distinct activation energies, the reaction tempera-
3
3
of acetaldehyde.
In addition, the crotonaldehyde selectivity was enhanced to and change the reaction selectivity.
2.8% over 19.2%Mg–16%Zr/MFI(NS). Correlating to the
The reaction temperature influence on the reaction was
ture will drastically affect the rate of the rate-determining step
3
lowered conversions of ethanol and acetaldehyde, it can be evaluated. As shown in Table 3, entries 1 and 2, when the reac-
conjectured that the over-abundant base sites of Mg remark- tion was conducted at 250–275 °C, crotonaldehyde was
ably depress the MPVO reduction, and lead to a rather large obtained at a high level (30–40% selectivity), which over-
fraction of the condensation product being intact. In another whelmed the 1,3-BD selectivity (6–24%). Increasing the reac-
word, the base sites of Mg do not contribute to MPVO tion temperature to 300–375 °C (Table 3, entries 3–6) led to
reduction.
.3.2 Catalytic role of Mg in the ETB reaction. To further erature of 400 °C (Table 3, entry 7), the selectivity to ethylene
1,3-BD being changed to the major product. At a higher temp-
3
probe the catalytic role of Mg in the ETB reaction, Mg was was enhanced remarkably. These results suggest that the
loaded on MFI(NS) at different loadings and tested in the reac- MPVO reaction has a higher apparent activation energy than
tion. As shown in Table 2, entries 6–8, the Mg introduction to the aldol condensation over Mg–Zr/MFI catalysts, and
MFI(NS) depressed the dehydration reaction, with ethylene 325–350 °C is a favorable temperature range for the high
1
,13,34
selectivity dropping from 79% to 30%. In contrast to no yield selectivity of 1,3-BD.
of crotonaldehyde over the blank MFI(NS) support, the con-
The reaction space velocity and the ratio of ethanol to acet-
densation product was present over both 1.2%Mg/MFI(NS) and aldehyde in the feedstock are important factors for the two-
1
9.2%Mg/MFI(NS) catalysts. Evidently, the base sites of Mg step ETB process. We modulated the weight of the catalyst and
a
Table 2 Effects of Mg–Zr on the catalytic performance in the ETB process
Conv. (%)
Product selectivity (%)
Entry
Catalysts
TC
ETOH
ACH
ELE
1,3-BD
DEE
CRH
C-Balance (%)
1
2
3
4
5
6
7
8
9
16%Zr/MFI(NS)
60.5
53.6
54.1
36.5
21.2
7.4
9.8
9.7
49.3
68.2
59.4
56.2
32.2
29.3
13.7
6.9
52.8
47.8
52.1
40.9
29.3
0.9
12.7
11.2
40.9
4.9
8.2
4.6
4.7
8.3
79
31.2
30.3
6.4
42.3
46.6
54.7
43.4
50.4
0
0
0
40.4
3.6
6.0
4.8
4.7
11.5
22
36.8
21.6
6.5
4.9
5.3
5.6
8.3
32.8
0
8.7
20.7
6.5
73.2
81.8
83.7
85.8
100
0.48%Mg–16%Zr/MFI(NS)
1.2%Mg–16%Zr/MFI(NS)
4.8%Mg–16%Zr/MFI(NS)
19.2%Mg–16%Zr/MFI(NS)
MFI(NS)
1.2%Mg/MFI(NS)
19.2%Mg/MFI(NS)
100
97.7
97.3
80.3
8.1
57.6
b
1.2%Mg–16%Zr/MFI(NS)
a
Reaction conditions: T = 350 °C, WHSV = 1.92 h−1, ethanol : acetaldehyde = 1 : 1. Conv., TC, ETOH, ACH, ELE, 1,3-BD, DEE, CRH and C-balance
are abbreviations for conversion, total conversion, ethanol, acetaldehyde, ethylene, 1,3-butadiene, diethyl ether, crotonaldehyde, and carbon
b
balance, respectively. The Mg precursor is magnesium acetylacetonate.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
a
Table 3 Reaction results over 1.2%Mg–16%Zr/MFI(NS) under different conditions
Conv. (%)
Product selectivity (%)
Entry
Reaction conditions
TC
ETOH
ACH
ELE
1,3-BD
DEE
CRH
C-Balance (%)
1
1
2
3
4
5
6
7
8
9
250 °C, 1.92 h− , E/A = 1 : 1
21.5
19.5
23
34.7
54.1
63.8
64.5
60.7
41.5
28.1
19.2
16
20
34.7
56.2
75.5
78.8
54.1
30.1
18.1
23.6
23.5
26.2
34.7
56.2
75.7
78.8
67.3
64.4
56
1.2
1.9
4.7
4.9
4.6
12.3
25.7
4.8
7
6.2
4.2
5.5
8.5
5.8
4.8
4.8
5.3
4.5
8.4
4.7
29.5
41.5
19.2
9.3
5.6
4.3
4.5
5.9
5.4
6.1
87.4
94.8
95.1
85.1
73.2
71.8
83.8
86.5
97.9
95.6
−
1
275 °C, 1.92 h , E/A = 1 : 1
300 °C, 1.92 h− , E/A = 1 : 1
325 °C, 1.92 h− , E/A = 1 : 1
24.2
46.2
37.1
54.7
34.4
39.3
62.5
74.6
70.3
1
1
−
1
350 °C, 1.92 h , E/A = 1 : 1
375 °C, 1.92 h− , E/A = 1 : 1
1
−
1
400 °C, 1.92 h , E/A = 1 : 1
350 °C, 0.96 h− , E/A = 1 : 1
1
−
1
350 °C, 1.44 h , E/A = 2 : 1
350 °C, 1.92 h− , E/A = 3 : 1
1
1
0
3.3
a
Reaction conditions: T = 250–350 °C, WHSV = 0.96–1.92 h−1, ethanol : acetaldehyde = 1 : 1–3 : 1. Conv., TC, ETOH, ACH, ELE, 1,3-BD, DEE, CRH
and C-balance are abbreviations for conversion, total conversion, ethanol, acetaldehyde, ethylene, 1,3-butadiene, diethyl ether, crotonaldehyde,
and carbon balance, respectively. E/A represents the ratio of ethanol to acetaldehyde.
the ethanol feeding rate to change the WHSV and feedstock MFI (NS) is rather attractive in view of the high yield of 1,3-BD,
3
,14,16,35,36
composition, and the reaction results over 1.2%Mg–16%Zr/ relatively mild reaction conditions and high stability.
MFI(NS) are listed in Table 3, entries 8–10. Although the effect The TG analysis of the spent catalyst shows that there is
of E/A on 1,3-BD selectivity cannot be identified exactly merely 12 wt% carbonaceous deposition on the catalyst after a
because of the E/A and the reaction WHSV being concurrently seven day run (Fig. S3†). This should be attributed to the MFI
changed in our experiments, the E/A ratio at 2–3 seems to nanosheet structure and abundant active sites of Mg and Zr in
favor 1,3-BD production, which is well consistent with the the catalyst.
1
results over Ta
2
O
5
/SiO
2
discussed previously. When the reac-
−1
tion was conducted at 350 °C, a WHSV of 1.44 h and a 2 : 1
ratio of ethanol to acetaldehyde, the 1,3-BD selectivity reached
3.5 Catalyst characterization
To reveal the high performance and stability of MFI(NS) sup-
ported Mg–Zr catalysts, the catalyst texture and acidic pro-
perties were characterized and correlated to the reaction
results.
7
4.6% with 41.5% ethanol and acetaldehyde conversion.
Under these optimal conditions, the catalyst stability was eval-
uated. As shown in Fig. 1, the catalytic performance was well
maintained in a one-week run. The 1,3-BD selectivity fluctu-
ated between 70–80% and the total conversion of ethanol and
acetaldehyde was maintained at ca. 40%. Compared to the
results in the literature, the performance of 1.2%Mg–16%Zr/
3
.5.1 Zr dispersion on the catalysts. The dispersion of Zr
on different silica supports was investigated using the XRD
pattern shown in Fig. S4.† No peaks owing to crystalline ZrO
2
particles were observed over MFI(NS), MFI(micro) and SiO2
supported catalysts, suggesting the uniform distribution of
2
the ZrO component. Specifically, Zr species on MFI(NS)
were highly dispersed, as evidenced from the TEM and
HAADF-STEM images and EDS mapping results in Fig. S5.†
The Zr dispersion on the supports was further studied by
using UV-vis diffuse reflectance spectra shown in Fig. 2.
The absorbance of bulk ZrO
2
was around 230–240 nm,
ascribed to the octahedral coordination state as reported in
3
7,38
the literature.
For the supported Zr catalysts, a substantial
blue shift in the absorbance was observed particularly on 16%
Zr/MFI(NS). The adsorption bands at 205–215 nm may be
attributed to the ligand-to-metal charge transfer (LMCT)
2
−
4+
from an O
configuration.
ion to an isolated Zr ion in a tetrahedral
3
7–39
According to Li and coworkers, the loading
of Zr on de-Al Beta zeolite formed the tetrahedrally co-
ordinated framework Zr species which acted as Lewis acid
2
0
centers for the reactions. However, considering the differ-
ence in the structure of MFI(NS) and de-Al Beta zeolites, it is
4
+
still not safe to regard that Zr ions stay in MFI(NS) in a tetra-
hedral configuration just based on the UV-vis analysis. Over
Zr/MFI(micro), a small shoulder at ca. 230 nm was observed,
Fig. 1 Stability of 1.2%Mg–16%Zr/MFI(NS) in the ETB reaction (reaction
conditions: T = 350 °C, WHSV = 1.44 h , ethanol : acetaldehyde = 2 : 1).
−1
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
Fig. 2 UV-Vis diffuse reflectance spectra of different supports and Zr-
Fig. 4 Raman spectra of Zr on different supported catalysts.
based catalysts.
−
1
indicating that nanoscopic regions of Zr–O–Zr linkages were and the bands at 550 and 476 cm are ascribed to the three-
3
7
formed on the microporous MFI supported catalyst. The absor- dimensional amorphous ZrO
bance edge was converted to the absorbance band edge energy found in 16%Zr/SiO and 16%Zr/MFI(NS) catalysts. This result
following the equation: αhν = A(hν − Eg) , where α is the indicates that over the MFI(NS) and mesoporous silica sup-
absorbance intensity at a light frequency of ν, A is the absor- ported catalysts, ZrO was highly dispersed without bulk ZrO
bance constant and E
is the absorbance edge energy. As particle formation. This conclusion is consistent with the UV-
shown in Fig. 3, the absorbance edge energies of Zr for bulk vis diffuse reflectance spectra results discussed above.
ZrO , MFI(micro), SiO and MFI(NS) supported Zr catalysts are
3.5.2 Acidic properties of Mg–Zr/MFI(NS) catalysts. The
.8, 5.0, 5.2 and 5.4 eV, respectively. The higher absorbance acidic properties of Mg–Zr catalysts were measured by NH
edge energies of the Zr catalysts can be ascribed to the LMCT TPD, Py-IR and CO adsorption. The NH -TPD profiles are
from O to Zr with lower coordination, due to the formation shown in Fig. 5; it was found that the Zr/MFI(NS) catalysts
2
.
These bands were hardly
2
1
/2
2
2
g
2
2
4
3
-
3
2
−
4+
3
9
of Zr–O–Si and Zr–OH linkages. The energies increase in the have less acid sites compared to the blank MFI(NS) support,
order of 16%Zr/MFI(micro) < 16%Zr/SiO2 < 16%Zr/MFI(NS), and the strength of acidity decreased with the increase of Zr
suggesting that the Zr oxide on MFI(NS) has the highest dis- loading on the support. This suggests that Zr introduction
persion while Zr/MFI(micro) has the lowest. This is well con- onto MFI(NS) weakens the acidity of MFI(NS). This is in agree-
sistent with the trend of catalytic performance, where 16%Zr/ ment with the reaction results listed in Table 1, entries 2–6,
MFI(NS) behaved the best among the three supported catalysts where less yields of dehydration products were obtained over
in the ETB reaction.
Zr/MFI(NS) catalysts compared to the MFI(NS) counterpart.
Py-IR provides quantification information of acid sites. As
Raman spectra provide more information about ZrO dis-
2
−
1
persion. As shown in Fig. 4, the bands at 380 nm of 16%Zr/ shown in Fig. 6 and Table 4, no Lewis acid sites but a large
MFI(micro) should be attributed to the characteristic peak of number of Brønsted acid sites (57.8 µmol g− ) were present on
Si–O bonds in the five membered ring of the MFI-type unit MFI(NS). After introducing 4% or 16% Zr onto MFI(NS), Lewis
1
4
0
−1
structure. The bands at 637 cm stand for monoclinic ZrO2 acid sites were generated at the expense of Brønsted acid sites.
2
Fig. 3 Plots of (αhν) versus hν of the various catalysts.
3
Fig. 5 NH -TPD profiles of MFI(NS) and MFI(NS) supported catalysts.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 6 Py-IR spectra for MFI(NS) and MFI(NS) supported catalysts.
Table 4 Acidic properties of the MFI support and Zr-based catalysts
Lewis acid site
Brønsted acid site
a
a
Catalyst
MFI(NS)
number (µmol g−1) number (µmol g−1) Fig. 7 FTIR spectra of CO absorbed on the Zr-based catalysts with
increased CO pressure.
—
2.0
72.0
57.8
6.0
24.0
11.7
23.9
4
1
1
1
%Zr/MFI(NS)
6%Zr/MFI(NS)
.2%Mg–16%Zr/MFI(NS) 84
.2%Mg–16%Zr/ 80.3
Introducing 1.2 wt% Mg onto 16%Zr/MFI(NS) decreased
MFI(NS)(organic Mg)
the strength of acid sites. As shown in the NH -TPD profiles
3
a
Lewis acid and Brønsted acid sites were calculated according to the
450 and 1540 cm peak areas, respectively, of pyridine adsorption at
50 °C.
(
Fig. 5), the desorption peak of NH
3
decreased from 141 °C to
−1
1
1
116 °C in the presence of Mg. On the other hand, the number
of Lewis acid sites slightly increased at the expense of
Brønsted acid sites (Table 4). Therefore, more abundant Lewis
acid sites with fewer Brønsted sites over the Mg–Zr/MFI(NS)
Over 16%Zr/MFI(NS), the number of Lewis acid sites reached catalyst account for the outstanding catalytic performance in
2.0 µmol g− , which is three times that of the Brønsted acid the ETB reaction.
1
7
sites. Correlating the reaction results with the acid site pro- The location of Mg on 1.2%Mg–16%Zr/MFI(NS) was probed
perties, it can be found that the proceeding of the ETB reaction by conditional experiments. An organic precursor of Mg
majorly depends on the Lewis acid sites, which not only cata- (magnesium acetylacetonate) was used to load Mg onto 16%
lyze the aldol condensation between ethanol and acetaldehyde, Zr/MFI(NS). Since the molecular size of the organic Mg com-
but also realize the critical MPVO reduction of the intermedi- pound (three times the size of benzene) is larger than the
ate. In contrast, the Brønsted acid sites just facilitate the de- micropore size of MFI(NS), the Mg species would largely
hydration to byproducts.
remain on the walls of the mesopores without entering into
The acid sites on Zr/MFI(NS) were further probed with CO the micropores. From the NH -TPD profile, it can be found
3
adsorption. As shown in Fig. 7, three typical FTIR bands were that the organic Mg modified catalyst has stronger acidity than
observed on the spectra of 16%Zr/MFI(NS) but only two peaks the counterpart (131 vs. 116 °C). Quantity analysis according
were present for 4%Zr/MFI(NS). According to the report of to the Py-IR spectra shows that these two catalysts had similar
−
1
Ivanova et al., the FTIR band at 2185 cm was ascribed to the numbers of Lewis acid sites, but there were more Brønsted
CO interacting with the Lewis acid sites (CO-Zr(OH)(OSi) ), acid sites on the 1.2%Mg–16%Zr/MFI(NS)(organic Mg) catalyst as
while the band at 2156 cm was attributed to CO adsorption shown in Table 4. The as-prepared catalyst showed inferior per-
3
−
1
−
1
on silanols, and the band observed at 2136 cm was assigned formance to that of 1.2%Mg–16%Zr/MFI(NS), as shown in
4
1,42
to pseudoliquid CO vibrations.
Obviously, over 16%Zr/MFI Table 2, entry 9. Therefore, it can be conjectured that Mg
(NS), abundant Zr(OH)(OSi)3 exists and provides Lewis acid species over the 1.2%Mg–16%Zr/MFI(NS) catalyst have entered
sites for the ETB reaction. However, nearly no such Lewis sites the mesopores and micropores. The nanosheets of MFI(NS)
were formed on 4%Zr/MFI(NS), in line with its inert activity provide the catalyst with mesopores and micropores, both of
for ETB.
which can host the active sites for the ETB reaction and
Green Chem.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Green Chemistry
Paper
endow the hierarchical-structure catalysts with outstanding
performance.
Acknowledgements
3
.5.3 Structure of the spent catalyst. The structure stability This work was supported by the National Natural Science
of the spent catalyst was probed. From SEM and TEM images Foundation of China (21690081, 21690084, 21721004 and
Fig. S6 and S7†), the morphology of 1.2%Mg–16%Zr/MFI (NS) 21776268) and “Transformational Technologies for Clean
(
was well retained after a 7 day run. However, from the XRD pat- Energy and Demonstration”, Strategic Priority Research
terns (Fig. S8†), it can be noticed that besides peaks belonging Program of the Chinese Academy of Sciences (No. XDA
to the crystalline structure of MFI(NS) zeolite, a new sharp 21060200).
peak appeared, which is attributed to SiO
88-2487). This suggests that MFI(NS) is not essentially stable
under the reaction conditions. After a long time run, the silica-
based support will be gradually changed into crystallized SiO
under the in situ treatment of steam which is generated during
the reaction. It is necessary to improve the structure stability
in future studies.
2
(JCPDS File No. 01-
0
2
Notes and references
1 E. V. Makshina, M. Dusselier, W. Janssens, J. Degreve,
P. A. Jacobsa and B. F. Sels, Chem. Soc. Rev., 2014, 43, 7917–
7953.
2
S. Da Ros, M. D. Jones, D. Mattia, J. C. Pinto, M. Schwaab,
F. B. Noronha, S. A. Kondrat, T. C. Clarke and S. H. Taylor,
ChemCatChem, 2016, 8, 2376–2386.
4
. Conclusions
A series of Mg–Zr/MFI(NS) catalysts were prepared by a wet
impregnation method for the two-step conversion of ethanol
to 1,3-BD. According to the catalyst characterization, MFI(NS)
is a superior support to highly disperse Zr as compared with
3 G. Pomalaza, M. Capron, V. Ordomsky and F. Dumeignil,
Catalysts, 2016, 6, 203–238.
4 J. Sun and Y. Wang, ACS Catal., 2014, 4, 1078–1090.
5 G. M. Cabello González, R. Murciano, A. L. Villanueva
Perales, A. Martínez, F. Vidal-Barrero and M. Campoy, Appl.
Catal., A, 2019, 570, 96–106.
6 W. Janssens, E. V. Makshina, P. Vanelderen, F. De Clippel,
K. Houthoofd, S. Kerkhofs, J. A. Martens, P. A. Jacobs and
B. F. Sels, ChemSusChem, 2015, 8, 994–1008.
2
the microporous MFI and conventional mesoporous SiO . The
introduced Zr species preferentially occupy the silanol nests of
MFI(NS), eliminating the Brønsted sites at low loadings of Zr
and producing abundant weak Lewis acid sites in the form of
3
Zr(OH)(OSi) at high Zr loadings. Over 16%Zr/MFI(NS), 42.3%
1
,3-BD selectivity and 60.5% total conversion of ethanol and
7 Z. Zhao, K. Chong, J. Jiang, K. Wilson, X. Zhang and
F. Wang, Renewable Sustainable Energy Rev., 2018, 97, 580–
591.
8 J. Sun and Y. Wang, ACS Catal., 2014, 4, 1078–1090.
9 D. Cespi, F. Passarini, I. Vassura and F. Cavani, Green
Chem., 2016, 18, 1625–1638.
acetaldehyde were obtained. Introducing 1.2 wt% Mg to 16%
Zr/MFI(NS) enhanced the 1,3-BD selectivity to 54.7% with
slight depression of the catalytic activity. Mg sites have negli-
gible activity in the MPVO reaction and a certain activity in
aldol condensation of acetaldehyde, both of which are much
inferior to that of Lewis acid sites. The major function of Mg 10 C. Wang, M. Zheng, X. Li, X. Li and T. Zhang, Green Chem.,
species in the catalyst should be decreasing the number of 2019, 21, 1006–1010.
Brønsted acid sites so that contributes to 1,3-BD formation. 11 G. Pomalaza, M. C. G. Vofo and F. Dumeignil, Green Chem.,
The reaction conditions have remarkable influence on the 2018, 20, 3203–3209.
reaction results. At optimal reaction conditions of 350 °C, a 12 M. Choi, K. Na, J. Kim, Y. Sakamoto, O. Terasaki and
WHSV of 1.44 h−1 and a 2 : 1 ratio of ethanol to acetaldehyde,
R. Ryoo, Nature, 2009, 461, 246–249.
the 1,3-BD selectivity reached 74.6% with 41.5% of total con- 13 V. L. Sushkevich, I. I. Ivanova and V. V. Ordomsky,
version over 1.2%Mg–16%Zr/MFI(NS), which was maintained ChemSusChem, 2013, 7, 2527–2536.
for 7 days (168 h) without deactivation. Such high performance 14 V. L. Sushkevich, I. I. Ivanov and E. Taarning, Green Chem.,
of 1.2%Mg–16%Zr/MFI(NS) should be attributed to the abun- 2015, 17, 2552–2559.
dant weak Lewis acid sites and their high availability in the 15 J. L. Cheong, Y. Shao, S. J. R. Tan, Y. Zhang and S. S. Lee,
meso- and micro-pores without Brønsted acid sites. This study ACS Sustainable Chem. Eng., 2016, 4, 4887–4894.
demonstrates that the zeolite nanosheet catalysts may have dis- 16 K. Góra-Marek, K. Tarach and M. Choi, J. Phys. Chem. C,
tinguished advantages for the two-step ETB process. The 2014, 118, 12266–12274.
hydrothermal stability of MFI(NS) should be further improved 17 E. Verheyen, C. Jo, M. Kurttepeli, G. Vanbutsele,
in view of the detrimental effect from the water byproduct
E. Gobechiya, T. I. Korányi, S. Bals, G. Van Tendeloo,
during the reaction.
R. Ryoo, C. E. A. Kirschhock and J. A. Martens, J. Catal.,
2
013, 300, 70–80.
8 M. Zhang and Y. Yu, Ind. Eng. Chem. Res., 2013, 52, 9505–
514.
9 B. Tang, W. Dai, G. Wu, N. Guan, L. Li and M. Hunger, ACS
Catal., 2014, 4, 2801–2810.
1
1
9
Conflicts of interest
There are no conflicts to declare.
This journal is © The Royal Society of Chemistry 2020
Green Chem.

View Article Online
Paper
Green Chemistry
2
2
2
2
2
0 B. Tang, W. Dai, X. Sun, G. Wu, N. Guan, M. Hunger and 32 S.-H. Chung, C. Angelici, S. O. M. Hinterding,
L. Li, Green Chem., 2015, 17, 1744–1755.
1 W. Dai, S. Zhang, Z. Yu, T. Yan, G. Wu, N. Guan and L. Li,
ACS Catal., 2017, 7, 3703–3706.
M. Weingarth, M. Baldus, K. Houben, B. M. Weckhuysen
and P. C. A. Bruijnincx, ACS Catal., 2016, 6, 4034–
4045.
2 M. Boronat, A. Corma and M. Renz, J. Phys. Chem. B, 2006, 33 V. V. Ordomsky, V. L. Sushkevich and I. I. Ivanova, J. Mol.
10, 21168–21174. Catal. A: Chem., 2010, 333, 85–93.
3 A. Corma, M. E. Domine, L. Nemeth and S. Valencia, J. Am. 34 C. Angelici, B. M. Weckhuysen and P. C. A. Bruijnincx,
Chem. Soc., 2001, 124, 3194–3195. ChemSusChem, 2013, 6, 1595–1614.
4 F. J. Urbano, M. A. Aramendía, A. Marinas and 35 Y. Xu, Z. Liu, Zheng Han and M. Zhang, RSC Adv., 2017, 7,
1
J. M. Marinas, J. Catal., 2009, 268, 79–88.
5 A. Corma, J. Catal., 2003, 215, 294–304.
6 J. He, X. Yang, D. G. Evans and X. Duan, Mater. Chem.
Phys., 2002, 77, 270–275.
7140–7149.
2
2
36 Q. Zhu, B. Wang and T. Tan, ACS Sustainable Chem. Eng.,
2016, 5, 722–733.
37 X. Gao, J. L. G. Fierro and I. E. Wachs, Langmuir, 1999, 15,
3169–3178.
2
7 J. V. Ochoa, C. Bandinelli, O. Vozniuk, A. Chieregato,
A. Malmusi, C. Recchi and F. Cavani, Green Chem., 2016, 38 Bharat L. Newalkar, Johnson Olanrewaju and Sridhar
1
8, 1653–1663.
Komarneni, J. Phys. Chem. B, 2001, 105, 8356–8360.
39 Y. Tang, E. Zong, H. Wan, Z. Xu, S. Zheng and D. Zhu,
Microporous Mesoporous Mater., 2012, 155, 192–200.
40 Y. Meng, H. C. Genuino, C. H. Kuo, H. Huang, S. Y. Chen,
L. Zhang, A. Rossi and S. L. Suib, J. Am. Chem. Soc., 2013,
135, 8594–8605.
2
2
8 A. Klein and R. Palkovits, Catal. Commun., 2016, 7, 92–95.
9 A. Klein, K. Keisers and R. Palkovits, Appl. Catal., A, 2016,
5
14, 192–202.
3
3
0 C. Angelici, M. E. Z. Velthoen, B. M. Weckhuysen and
P. C. A. Bruijnincx, Catal. Sci. Technol., 2015, 5, 2869–2879.
1 C. Angelici, F. Meirer, A. M. J. van der Eerden, 41 V. L. Sushkevich, D. Palagin and I. I. Ivanova, ACS Catal.,
H. L. Schaink, A. Goryachev, J. P. Hofmann, 2015, 5, 4833–4836.
E. J. M. Hensen, B. M. Weckhuysen and P. C. A. Bruijnincx, 42 V. L. Sushkevich, A. Vimont, A. Travert and I. I. Ivanova,
ACS Catal., 2015, 5, 6005–6015. J. Phys. Chem. C, 2015, 119, 17633–17639.
Green Chem.
This journal is © The Royal Society of Chemistry 2020