Vapor-phase dehydration of 1,4-butanediol to 1,3-butadiene over Y2Zr2O7 catalyst

Article abstract of DOI:10.1016/j.mcat.2021.111853

Vapor-phase catalytic dehydration of 1,4-butanediol (1,4-BDO) was investigated over Y₂O₃-ZrO₂ catalysts. In the dehydration, 1,3-butadiene (BD) together with 3-buten-1-ol (3B1OL), tetrahydrofuran, and propylene was produced depending on the reaction conditions. In the dehydration over Y₂O₃-ZrO₂ catalysts with different Y contents at 325°C, Y₂Zr₂O₇ with an equimolar ratio of Y/Zr showed high selectivity to 3B1OL, an intermediate to BD. In the dehydration at 360°C, a BD yield higher than 90% was achieved over the Y₂Zr₂O₇ calcined at 700°C throughout 10 h. In the dehydration of 3B1OL over Y₂Zr₂O₇, however, the catalytic activity affected by the calcination temperature is roughly proportional to the specific surface area of the sample. The highest activity of Y₂Zr₂O₇ calcined at 700 °C for the BD formation from 1,4-BDO is explained by the trade-off relation in the activities for the first-step dehydration of 1,4-BDO to 3B1OL and for the second-step dehydration of 3B1OL to BD. The higher reactivity of 3B1OL than saturated alcohols such as 1-butanol and 2-butanol suggests that the C=C double bond of 3B1OL induces an attractive interaction to anchor the catalyst surface and promotes the dehydration. A probable mechanism for the one-step dehydration of 1,4-BDO to BD was discussed.

Full text of DOI:10.1016/j.mcat.2021.111853

Molecular Catalysis 514 (2021) 111853
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.journals.elsevier.com/molecular-catalysis
Vapor-phase dehydration of 1,4-butanediol to 1,3-butadiene over
Y Zr O catalyst
2
2 7
*
Asami Matsuda, Yoshitaka Matsumura, Yasuhiro Yamada, Satoshi Sato
Graduate School of Engineering, Chiba University, Yayoi, Inage, Chiba, 263-8522, Japan
A R T I C L E I N F O
A B S T R A C T
Vapor-phase catalytic dehydration of 1,4-butanediol (1,4-BDO) was investigated over Y O -ZrO catalysts. In the
Keywords:
2
3
2
Production of 1,3-butadiene
Dehydration of 1,4-butanediol
dehydration, 1,3-butadiene (BD) together with 3-buten-1-ol (3B1OL), tetrahydrofuran, and propylene was pro-
duced depending on the reaction conditions. In the dehydration over Y
2
O
3
-ZrO
2
catalysts with different Y
Y
2
Zr
2
O
7
◦
contents at 325 C, Y
Zr
2 2
O
7
with an equimolar ratio of Y/Zr showed high selectivity to 3B1OL, an intermediate to
◦
Cubic fluorite
◦
BD. In the dehydration at 360 C, a BD yield higher than 90% was achieved over the Y
throughout 10 h. In the dehydration of 3B1OL over Y Zr , however, the catalytic activity affected by the
2 2 7
Zr O calcined at 700 C
2
2 7
O
calcination temperature is roughly proportional to the specific surface area of the sample. The highest activity of
◦
Y
Zr
2 2
O
7
calcined at 700 C for the BD formation from 1,4-BDO is explained by the trade-off relation in the
activities for the first-step dehydration of 1,4-BDO to 3B1OL and for the second-step dehydration of 3B1OL to BD.
The higher reactivity of 3B1OL than saturated alcohols such as 1-butanol and 2-butanol suggests that the C=C
double bond of 3B1OL induces an attractive interaction to anchor the catalyst surface and promotes the dehy-
dration. A probable mechanism for the one-step dehydration of 1,4-BDO to BD was discussed.
1
. Introduction
,3-Butadiene (BD) is a beneficial compound used as a raw material
higher through catalytic conversion processes [7,8].
It has already been reported that BD can be produced from diols [4,
1
9-14]. In the dehydration of 1,4-BDO, solid acid catalysts such as γ-Al
2 3
O
for synthetic rubbers and resins. Currently, 98% of BD is obtained as a
by-product of naphtha cracking in ethylene plants, and thus the supply
of ethylene influences the supply and price of BD. In recent years, with
the revolution of shale gas in the United States, ethylene can be readily
produced by ethane cracking. As a result, the supply for ethylene ob-
tained from naphtha is decreasing, and the shortage of BD supply raised
the market price of BD [1-4]. In addition, aggravation of the global
environment due to carbon dioxide produced from fossil resources is
also a significant problem. In order to solve these problems, there is a
growing demand for a new synthetic process to produce BD from
biomass-derived carbon-neutral raw materials.
and SiO -Al show an efficient cyclization ability to form tetrahy-
2
2 3
O
◦
drofuran (THF) at a temperature of 275 C [4,15]. In the one-step
dehydration of 1,4-BDO to BD, a BD yield of 96.6% is achieved over
◦
Yb
2
O
3
at 360 C [12]. An industrial process to produce BD from 1,4-BDO
has been investigated [16]. It is reported the process consists of two
steps, which include esterification of 1,4-BDO followed by pyrolysis of 1,
4-BDO diester to produce BD. In the process, 0.57 tons of BD, which is a
stoichiometric yield of 95.9%, is produced from a ton of 1,4-BDO.
However, the second step of pyrolysis of 1,4-BDO diester to produce
◦
BD requires a temperature of 500 C or higher.
Scheme 1 shows the production route of BD in the dehydration of
1,4-BDO. 1,4-BDO produces mainly 3-buten-1-ol (3B1OL), followed by
isomerization to form unsaturated alcohols (UOLs) such as 2-buten-1-ol
(2B1OL) and 3-buten-2-ol (3B2OL). The major by-products are THF and
propylene through the decomposition of 3B1OL. Several catalysts such
Biomass-derived 1,4-butanediol (1,4-BDO) can be obtained through
microbial fermentation and catalytic conversion processes. 1,4-BDO can
be obtained from glucose and xylose by microbial fermentation using
ꢀ
1
Escherichia coli [4-6]: 3.83 g L of 1,4-BDO can be produced from xylose
ꢀ 1
with a yield of 0.37g gxylose , which corresponds to the 1,4-BDO
selectivity of 63 mol% [6]. In addition, 1,4-BDO can be produced from
succinic acid and furfural obtained from glucose in good yields of 80% or
2
as rare earth oxides (REO) [15,17-20], ZrO [20-23], and rare earth
zirconates (REZrO) [24,25] have been reported as an efficient catalyst
for the formation of 3B1OL from 1,4-BDO. As for the production of
*
Corresponding author.
E-mail address: satoshi@faculty.chiba-u.jp (S. Sato).
https://doi.org/10.1016/j.mcat.2021.111853
Received 23 July 2021; Received in revised form 25 August 2021; Accepted 25 August 2021
Available online 8 September 2021
2
468-8231/© 2021 Elsevier B.V. All rights reserved.

A. Matsuda et al.
Molecular Catalysis 514 (2021) 111853
Scheme 1. Production of BD from 1,4-BDO
Table 1
◦
Dehydration of 1,4-butanediol over zirconate catalysts at 375 C.
a
2
ꢀ 1
g )
Catalyst
SA (m
Conv. (%)
Selectivity (mol%)
BD
3B1OL
2B1OL
6.0
UOLs
53.2
61.6
54.4
25.3
51.0
28.2
59.9
56.7
61.7
64.6
C3’
0.0
0.2
0.2
1.8
0.2
5.8
0.0
5.5
0.2
4.7
THF
5.5
10.5
8.5
3.0
7.1
3.7
5.8
7.7
8.1
7.7
GBL
0.4
0.3
0.1
0.1
0.0
0.1
0.1
0.4
0.1
0.5
others
4.1
5.6
4.8
3.4
4.3
3.4
7.5
1.3
4.5
1.8
b
Eu
Gd
Tb
2
Zr
Zr
Zr
Zr
Zr
2
O
7
35.1
23.9
34.0
23.8
29.8
36.4
29.6
39.3
27.4
26.3
96.4
36.8
21.7
32.1
66.4
37.3
58.8
26.7
28.3
25.5
20.7
47.2
53.1
44.3
20.6
40.9
23.8
49.4
52.7
52.3
59.8
b
2
2
O
7
99.9
8.6
2
2
O
7
100.0
99.8
10.0
4.7
Dy
Ho
2
2 7
O
2
2
O
7
100.0
100.0
100.0
98.1
10.2
4.4
Y
2
Zr
Zr
Tm Zr
Yb Zr
Lu Zr
2 7
O
Er
2
O
2 7
10.5
4.1
2
2 7
O
2
2
O
7
100.0
98.9
9.3
2
2
O
7
4.8
Conversion and selectivity are averaged between 1-5 h.
Reaction conditions: W/F = 0.31 h; N carrier gas flow rate, 30 cm min
3
ꢀ 1
2
.
BD, 1,3-butadiene; 3B1OL, 3-buten-1-ol; 2B1OL, 2-buten-1-ol; UOLs = 3B1OL + 2B1OL; C3’, propylene; THF, tetrahydrofuran; GBL, γ-butyrolactone. Others include
ethanol, 1-butanol, CO
2
, and some unidentified products.
a
◦
calcined at 900 C; crystallite phase was determined in Ref. [24].
b
2 3 2 3
included rare earth oxide such as Eu O and Gd O .
◦
UOLs, Y
2
Zr
2
O
7
calcined at 900 C shows high activity, such as a 3B1OL
a precipitate prepared from an aqueous solution of Y(NO
3
)
3
⋅H
2
O and
◦
selectivity of 87.7% at a 1,4-BDO conversion of 77.6% at 325 C [25].
ZrO(NO O with an equimolar amount of Y and Zr. The finally
3
)
2
⋅2H
2
Recently, it is reported that the Y
2
Zr
2
O
7
catalyst is also efficient for the
obtained catalyst sample was calcined at a temperature between 600
◦
formation of BD from 1,3-BDO, achieving a high BD yield of 95% at
and 1000 C for 3 h. Hereafter, the calcined yttrium zirconate is named
◦
◦
3
75 C [14]. It is considered that the dehydration of 1,4-BDO and 1,
-BDO to UOLs proceeds selectively on the (111) plane of the cubic
Y
2
Zr
2
O
7
-T, where T is the calcination temperature ( C). Y
2
O
3
-ZrO
2
3
(YZrO) samples with different Y contents were prepared in the same way
as that of Y Zr described above, which is expressed as xYZrO where x
is Y content (mol%). Preparation of other REZrO samples is described
fluorite structure of Yb
2
Zr
2
O
7
[24], Y
2
Zr
2
O
7
[25], and CeO
2
[26].
2
2 7
O
We have reported an important finding that a hydrothermal (HT)
aging of the precursor hydroxide in an ammonia solution provides
elsewhere [25]. Reference acidic catalysts such as Al
2
O
3
(JRC-ALO-7;
2
ꢀ 1
excellent Yb
2
Zr
2
O
7
catalysts with high crystallinity of oxygen-defected
specific surface area, SA =180 m
g
) and SiO
-Al O (JRC-SAL-3;
2 2 3
2
ꢀ 1
fluorite structure as well as relatively high specific surface areas,
which show high activity for the dehydration of 1,3-BDO [24].
In this study, we investigated the selective production BD from 1,4-
BDO in a one-step process (Scheme 1) using REZrO catalysts prepared by
an HT aging method. To clarify the feature of efficient catalyst for the BD
formation, we tested several catalysts for the catalytic activity of UOLs
SA=416 m g ) were supplied by Catalyst Society of Japan.
2
.2. Catalytic reaction
The vaper-phase catalytic reaction was performed in a fixed-bed
down-flow glass tube reactor with an inner diameter of 17 mm under
(
Scheme 1b). As for the second-step dehydration, it has been suggested
◦
an atmospheric pressure of N
2
at a temperature between 300 and 425 C.
that the double bond of UOLs could act as an anchor for the reaction to
proceed [12], so that we also conducted catalytic activity tests of C4
saturated alcohols such as 1-butanol and 2-butanol to discuss the reac-
tion mechanism.
After the catalyst, typically 0.5 g, had been heated in N
2
flow at a flow
rate of 30 cm3 min at the prescribed reaction temperature for 1 h, a
ꢀ 1
ꢀ 1
reactant BDO was fed into the reactor at a liquid feed rate of 1.60 g h
.
◦
The reaction effluent was collected at either 0 or -78 C every hour: the
◦
reaction effluent was collected at 0 C when the conversion was 80% or
2
. Experimental
◦
higher, and the reaction temperature was 350 C or higher. The liquid
products collected in a trap were identified by gas chromatography (GC)
using a mass spectrometer (GCMS-QP5050 A, Shimadzu) with a capil-
lary column (DB-WAX, a length of 30 m, JW Scientific). The products
were quantified by GC (FIDGC-8 A, Shimadzu) with a direct injection
2
.1. Samples
2 2 7
An yttrium zirconate, Y Zr O , was prepared through HT aging using
2

A. Matsuda et al.
Molecular Catalysis 514 (2021) 111853
2
.3. Characterization of catalysts
The X-ray diffraction (XRD) of YZrO catalysts was performed with
New D8 ADVANCE (Bruker) using Cu K
radiation (λ=0.154 nm) to
identify the crystalline phase of the catalyst. The N adsorption mea-
α
2
surement was carried out in a self-constructed volumetric gas adsorption
◦
apparatus at ꢀ 196 C. Three-point data for the N
2 0
adsorption at P/P
range between 0.05 and 0.35 was analyzed by the Brunauer-Emmett-
Teller (BET) equation to calculate the specific surface area (SA) of the
catalyst.
2 2 7 2 3
To construct the crystal structure of Y Zr O and Y O , crystal
structure modeling was performed using VESTA 3 software [27] with
the National Institute for Materials Science (NIMS) AtomWork Database
[
28] including crystallographic data of Y
. Results and discussion
.1. Catalytic dehydration of 1,4-BDO over REZrO catalysts
In our previous work, we have recently reported that the selective
2 2 7 2 3
Zr O [29] and Y O [30].
3
3
dehydration of 1,4-BDO to 3B1OL proceeded over REZrO catalysts with
◦
heavy rare earth at a reaction temperature of 325 C [25]. The REZrO
◦
samples calcined at 900 C have been characterized by using XRD and
X-ray fluorescence analysis. The crystal phase of Y
by Rietveld analysis. Furthermore, the physical properties of
Zr -900 were evaluated by acid-base evaluation, N
2
2 2 7
Zr O was confirmed
Y
2
2
O
7
adsorption-desorption isotherm, and high-resolution field emission
scanning electron microscope images. Eight REZrO samples listed in
◦
Fig. 1. XRD patterns of YZrO catalysts calcined at 900 C at different
Y contents.
Table 1 have a single crystal phase of cubic fluorite such as Y
Tb, Dy, Ho, Er, Tm, Yb, and Lu. However, Eu Zr and Gd Zr
mixture of cubic fluorite and cubic bixbyite Eu or Eu
2
Zr
2
O
7
for
2
2
O
7
2
2 7
O
are a
2
O
3
2 3
O .
In the beginning of this study, the catalytic activity of REZrO was
◦
tested at a high reaction temperature of 375 C to investigate the one-
step formation of BD from 1,4-BDO. Table 1 summarizes the catalytic
activity of ten REZrO samples in the dehydration of 1,4-BDO at a low
space time W/F = 0.31 h, where W and F are catalyst weight and
2 2 7 2 2 7
reactant feed rate, respectively. Y Zr O -900 and Dy Zr O -900 showed
high BD yields of more than 58%, and the major by-products were THF,
3
B1OL, 2B1OL, and propylene. The activity tests of Y
2
Zr
2
O
7
and
◦
Dy Zr
2
2
O
7
, which showed high BD selectivity, were carried out at 360 C
◦
and W/F = 2.5 h for 10 h, where the calcination temperature is 800 C
Fig. S1). Y Zr -900 catalyst showed more stable activity than
Dy Zr -900, so that the Y Zr catalyst was investigated in detail in
the following sections.
(
2
2 7
O
2
2
O
7
2
2 7
O
3
.2. Dehydration of 1,4-BDO over YZrO catalysts with different Y
contents
It is important to determine an active phase of YZrO catalysts. At
first, the optimum ratio of Y/Zr in YZrO catalysts was examined by
varying the Y content from 0 to 100 mol%. Fig. 1 shows the XRD patterns
of the catalysts with different Y contents. It was confirmed that the
◦
Fig. 2. Changes in the 1,4-BDO conversion and the 3B1OL selectivity over
crystal phase of YZrO samples calcined at 900 C varied with Y content.
At Y content of 0 mol%, i.e. ZrO , the crystal phase was monoclinic. At Y
◦
YZrO catalysts at 325 C as well as SA with Y content.
2
◦
3
ꢀ 1
Calcination at 900 C, W/F = 0.31 h; N
2
carrier gas flow rate, 30 cm min ; SA,
contents of 2-4 mol% (XRD patterns not shown), the crystal phase has
been reported to be a mixture of monoclinic and tetragonal zirconia
phases [20]. At Y contents of 6-25 mol%, the crystal phase was a
tetragonal, i.e. Y-stabilized zirconia. At Y contents of 31-43 mol%, the
crystal phase was a cubic fluorite. At a Y content of 50 mol%, it has been
specific surface area.
unit and a capillary column (Inert Cap WAX-HT, inner diameter of 0.53
mm, a length of 30 m, GL-Science). In the quantitative GC analysis, 1-
hexanol was used as an internal standard. Gaseous products such as
reported that 50YZrO sample is Y
structure (PDF 01-074-9311, Fd-3 m, No. 227) [25]. At Y contents of
7-75 mol%, the catalyst sample was a mixture of cubic fluorite Y Zr
with 2θ = 29.9 degree (Fig. 1b) and cubic bixbyite Y with 2θ = 29.2
degree (PDF 01-071-0049, Ia-3, No. 206). At a Y content of 100 mol%,
the crystal phase was a cubic bixbyite Y (100YZrO).
Fig. 2 shows changes in the conversion of 1,4-BDO and the selectivity
2 2 7
Zr O with a defective fluorite-type
3
BD, propylene, and CO
2
were collected with a 0.5 cm gastight syringe
and analyzed in another GC-8A equipped with a thermal conductivity
detector and a packed column (VZ-7, a length of 6 m, GL-Science) using
5
2
2 7
O
2 3
O
2
N carrier gas as an internal standard. The catalytic activity was eval-
uated by averaging the conversion and selectivity data for four hours
from time on stream of 1 to 5 hours, excluding the first period of 1 h.
2 3
O
3

A. Matsuda et al.
Molecular Catalysis 514 (2021) 111853
Fig. 3. Top view of surface structure of (111)
2 2 7 2 3
face of Y Zr O (a-c) and (222) face of Y O (d,
e) modeled by VESTA 3.
2
ꢀ
Large balls represent O , small blue balls are
4
+
3+
either Zr or Y (a), and small red balls are
3
+
Y
(b). The top (white) and the third (gray)
2
ꢀ
anions, and the
layers are composed of O
second layer is composed of the cations. The
crystal structures were investigated using
VESTA 3 [27] and the National Institute for
Materials Science (NIMS) AtomWork Database
[
28-30]. (For interpretation of the references to
color in this figure legend, the reader is referred
to the web version of this article.)
Fig. 4. Changes in the formation rate of 3B1OL in the dehydration of 1,4-BDO
◦
Fig. 5. Change in the catalytic activity of Y Zr O -900 catalyst in the dehy-
over YZrO at 325 C with Y content.
2
2 7
◦
3
ꢀ 1
dration of 1,4-BDO with reaction temperature.
Calcination at 900 C, W/F = 0.31 h; N
2
carrier gas flow rate, 30 cm min .
3
ꢀ 1
.
W/F= 0.31 h; N
2
carrier gas flow rate, 30 cm min
The symbols in the figure indicate the crystal phase of catalyst: T, tetragonal
phase; C, cubic phase.
oxygen defects and cubic bixbyite Y
cussed as follows.
2 3
O [12,25]. This difference is dis-
to 3B1OL with Y content. In comparison, Fig. S2 shows the catalytic
activity of the same samples in the dehydration of 1,3-BDO. The reactant
BDOs have similar reactivity for the dehydration over the YZrO catalysts
From the Rietveld analysis [25], it is confirmed that 50YZrO is a
Zr with an oxygen-defected fluorite structure. The cubic fluorite,
CeO , has a predominantly (111) plane, and the oxygen defects on the
(111) plane serve as adsorption sites for the dehydration of BDOs [26].
The heavy rare earth oxides such as Yb have a cubic bixbyite struc-
ture, and the (222) plane is the most stable [31,32]. In a similar way to
the cubic CeO , the oxygen defects on the surface of Y Zr and Y
Y
2
2 7
O
[
25]: 1,3-BDO is slightly more reactive than 1,4-BDO. In both reactions,
catalysts with a single cubic fluorite phase at Y contents of 31-50 mol%
including Y Zr showed a higher selectivity to UOLs than the other
2
2
2
O
7
2 3
O
phases such as tetragonal Y-stabilized zirconia and mixed phases. It is
considered that the dehydration of 1,4-BDO to produce 3B1OL proceeds
at specific crystal planes [20], and the high 3B1OL selectivity over cubic
fluorite catalysts could be due to single active sites provided on the
fluorite surface. At Y contents of 50 mol% or higher, a decrease in the 1,
2
2
2
O
7
2 3
O
are considered to act as adsorption sites for 1,4-BDO. To discuss the
difference in activity depending on the structure, Fig. 3, which was
constructed by VESTA modeling [27] with crystallographic data
[28-30], shows the top view of the upper three layers of (111) plane of
4
-BDO conversion was observed with a decrease in the SA. The decrease
in the 1,4-BDO conversion in 100YZrO, i.e. Y , could be attributed to
the difference between the crystal phases in cubic fluorite Y Zr with
2
O
3
the cubic oxygen-defected fluorite Y
plane of the cubic bixbyite Y (Fig. 3d,e). The Y
has oxygen in 87.5% of the oxygen-occupied sites of the cubic crystal, i.
2 2 7
Zr O
(Fig. 3a-c) and those of (222)
2
2
O
7
2
O
3
2
2 7
Zr O (111) surface
4

A. Matsuda et al.
Molecular Catalysis 514 (2021) 111853
cluster with four O defects acts as an adsorption site, the ratio of
adsorption sites on the surface corresponds 1/16. In the cubic bixbyite
Y
2
O
3
, the fraction of surface adsorption sites is a half of the cubic
oxygen-defected fluorite Y Zr . The consideration on the defect sites,
therefore, can explain the superiority of the cubic fluorite Y Zr to the
cubic bixbyite Y in the 3B1OL formation rate shown in Fig. 4.
We have further taken into consideration on the active sites as fol-
lows. In a cubic fluorite Y Zr , an average value of the ionic radius of
2
2 7
O
2
2 7
O
2 3
O
2
2 7
O
Y and Zr, which is regarded as a representative distance, is calculated to
be 0.087 nm from the ionic radii of Y (0.0900 nm) and Zr (0.0840 nm)
[
33]. The average value of Y-Zr in Y
2
Zr
2
O
7
is close to that of the ionic
2 3
O that showed high activity pro-
radius of Yb (0.0868 nm [33]) in Yb
duce BD from 1,4-BDO [12], suggesting that the distance between cat-
ions on the surface could be significantly related to the catalytic activity.
Actually, the formation rate of 3B1OL over Yb
that of Y
2
O
3
is 1.3 times as high as
◦
2
O
3
in the dehydration of 1,4-BDO at 375 C [18].
3
.3. Catalytic dehydration of 1,4-BDO over Y
2 2 7
Zr O catalyst under
different conditions
Fig. 5 shows the 1,4-BDO conversion and the selectivity to each
product at different reaction temperatures in the dehydration over
2 2 7
Fig. 6. Changes in the conversion of 1,4-BDO and selectivity over Y Zr O -900
◦
Y
2
Zr
2
O
7
-900. The fed 1,4-BDO was completely consumed at 350 C or
with space time.
◦
3
ꢀ 1
Reaction temperature, 360 C; N
2
carrier gas flow rate, 30 cm min .
e., 12.5% oxygen vacancies of the oxygen-occupied sites [25]. This
means that 1/8 of the surface oxygen occupied sites of Y Zr are va-
2
2 7
O
cancies (12 O defects seen in Fig. 3a). In contrast, vacancies occupy 1/4
of the surface oxygen sites in the stable (222) plane of cubic bixbyite
Y
2
O
3
(24 O defects seen in Fig. 3d).
Fig. 4 shows the formation rate of 3B1OL per unit surface area at
different Y contents. The formation rate varied with Y content, and it
seems dependent on the crystal structure: the formation rates over cubic
catalysts are higher than those over tetragonal catalysts. In the same
cubic structure, the formation rate over fluorite Y
high as that over bixbyite Y , although the ratio of surface oxygen
defects in the cubic bixbyite Y is twice as high as that of fluorite
Zr (Fig. 3d). However, the difference between the catalysts is the
position of the oxygen defects. In the cubic oxygen-defected fluorite
Zr , 1/8 of the surface oxygen are randomly defected with an iso-
2 2 7
Zr O is three times as
2 3
O
2 3
O
Y
2
2 7
O
Y
2
2 7
O
lated position, as shown in Fig. 3a where 12 O defects are observed. It is
reasonable that each defect of the oxygen vacancies acts as an adsorp-
tion site. On the other hand, clusters of vacancies of four oxygen-
occupied sites like a three-rayed star are regularly located on the sur-
Fig. 8. Effect of calcination temperature of Y
2
Zr
2
O
7
catalyst on the dehydration
◦
of 3B1OL at 360 C.
face of cubic bixbyite Y
2 3
O [31,32], as shown in Fig. 3d where 24 O
Conversion and yield are averaged between 1-5 h. Reaction conditions: W/F =
defects constructed from 6 clusters with four O defects are present. If a
3
ꢀ 1
.
0
.31 h; N
2
carrier gas flow rate, 30 cm min
Fig. 7. Changes in the yield of BD and 3B1OL with time on stream in the dehydration of 1,4-BDO over Y
Zr
2 2
O
7
calcined at different temperatures.
Reaction temperature, 360 ◦C; W/F = 2.5 h; N
carrier gas flow rate, 30 cm min
3
ꢀ 1
.
2
5

A. Matsuda et al.
Molecular Catalysis 514 (2021) 111853
selectivity, decreased with time on stream, and the UOLs selectivity
increased (Fig. S4). Therefore, this suggests that the dehydration from
2 2 7
UOLs to BD proceeded on the Y Zr O -900 catalyst, not from THF.
Under the conditions of a high space time of W/F =2.5 h, the UOLs
selectivity became nearly 0, i.e. the dehydration in the second step
proceeded completely.
3
.4. Catalytic dehydration of 1,4-BDO over Y
2 2 7
Zr O calcined at different
temperatures
2 2 7
Fig. 7 shows the effect of calcination temperature of Y Zr O on the
changes in the yield of BD and 3B1OL in the dehydration of 1,4-BDO,
and the numerical data are displayed in Table S1. In the 10-h reac-
2 2 7
tion, Y Zr O -700 maintained the highest BD yield where the 3B1OL
selectivity was 0.0%, indicating that the dehydration of 3B1OL to BD
was complete. On the other hand, the BD yield decreased in the Y
2
Zr
2
O
7
◦
samples calcined at 600, 900, and 1000 C with time on stream. The
yields of BD and 3B1OL were not steady during the 10-h reaction. The
SA of used Y
2
Zr
ꢀ 1
. The SA decreased with increasing the calcination temperature
2 7
O -700 was reduced from the fresh one of 78.6 to 71.8
2
m
g
(
Fig. 8 and Table S1), suggesting that the decrease in the BD yield in the
◦
range of calcination temperatures between 700 and 1000 C is corre-
lated with the decrease in the SA of the samples. The slight decrease in
the BD yield with time on stream even at a high space time would be
caused by the deactivation of active sites for the dehydration of 3B1OL.
The coke deposition on the catalysts was estimated by the decrease in
weight in thermogravimetric analysis, as shown in Fig. S7: the coke
Fig. 9. Effect of calcination temperature on the changes in conversion and
◦
selectivity in the dehydration of 3B1OL over Y
stream.
2 2 7
Zr O at 360 C with time on
3
ꢀ 1
.
Reaction conditions: W/F = 0.31 h; N
2
carrier gas flow rate, 30 cm min
higher. As the reaction temperature increased, the selectivity to BD
increased, and that for 3B1OL decreased. This means that the generated
2 2 7
content was proportional to the SA of the Y Zr O catalyst (Table S1).
◦
3
B1OL is dehydrated to BD at high temperatures. At 375 C, however,
In our previous study, we have confirmed that raising the calcination
temperature improves the crystallinity and the selectivity to 3B1OL
propylene was produced, and the selectivity to propylene increased with
the temperature. This is due to the decomposition of 3B1OL by the retro-
Prince reaction, which is favorable at high temperatures [12,14]. The
2 2 7
[25]. Over the Y Zr O -600 with low crystallinity, however, both the
conversion of 1,4-BDO and the selectivity to 3B1OL were low irre-
◦
total selectivity to UOLs and BD was higher than 85.3% at 400 C or
spective of the high SA. A similar trend was observed in the dehydration
◦
◦
lower, while it was lowered to 79.4% at 425 C. Also, at 425 C, the BD
selectivity decreased steeply with time on stream, and the THF selec-
tivity increased (Fig. S3). Since the catalytic activity was relatively
2 2 7
of 1,3-BDO to BD [14]. Therefore, it is reasonable that the Y Zr O -700
is located at the maximum in the tradeoff curve between the activity in
the dehydration of 1,4-BDO to 3B1OL and the dehydration of 3B1OL to
BD. We will discuss why the dehydration from 3B1OL to BD becomes
more difficult to proceed over the samples calcined at high temperatures
in the following section.
◦
stable at temperatures lower than 425 C, the change in the selectivity
would be caused by deactivation of catalyst probably because of coke
deposition: it is possible that the active site for the dehydration of 3B1OL
to BD was poisoned by coke deposition at high reaction temperatures.
Since propylene and THF are formed as by-products at high reaction
temperatures, the effect of space time was investigated at a low tem-
3
2 2 7
.5. The dehydration of UOLs to BD over Y Zr O
◦
perature of 360 C, at which propylene formation is small.
In the formation of BD from 1,4-BDO, 3B1OL is a key intermediate
12]. To investigate the second-step dehydration (Scheme 1b), the
reactivity test of 3B1OL over Y Zr catalysts was carried out. Fig. 8
Fig. 6 shows the change in the conversion and the selectivity at
[
◦
3
60 C with space time. As the space time increased, the UOLs and THF
2
2 7
O
selectivity decreased, and thus the BD selectivity increased. It is previ-
ously reported the dehydration of THF to BD [34,35]: THF conversion of
shows the effect of the calcination temperature of Y
2
Zr
2
O
7
on the
◦
dehydration of 3B1OL at 360 C. Both the 3B1OL conversion and the BD
8
4
9.1% with BD selectivity of 84.7% is attained over an acid catalyst at
2 2 7
selectivity were the highest in the Y Zr O -600, and they were decreased
◦
00 C [34]. However, in this case, the THF selectivity, as well as the BD
with increasing the calcination temperature. The decrease in the 3B1OL
Table 2
◦
Dehydration of monoalcohols over Y
Zr
2 2
O
7
-900 at 360 C.
Reactant
TOS (h)
Conv. (%)
Selectivity (mol%)
a
BD
64.6
56.1
61.7
61.9
49.4
57.8
-
2B1OL
3B2OL
3B1OL
MEK
-
C3’
3.5
6.9
0.0
0.0
0.0
0.0
-
1-butene
2-butene
others
3.6
3
3
2
1
2
B1OL
B2OL
B1OL
-BuOH
-BuOH
0.5
53.7
25.9
39.6
30.4
51.7
49.1
1.4
17.1
11.2
-
-
-
4
.5
0.5
.5
0.5
.5
0.5
.5
0.5
.5
19.0
12.8
-
-
-
-
5.2
23.4
-
2.5
2.3
3.8
3.1
-
1.7
1.3
0.8
0.6
-
-
-
10.7
10.4
6.0
4
24.1
-
-
-
-
-
-
-
-
-
40.0
-
-
4
33.0
-
-
5.5
-
-
-
-
36.4
59.9
61.7
59.8
-
63.6
40.1
1.1
4
1.4
-
-
-
-
-
17.2
8.9
-
-
29.7
20.7
-
7.5
18.8
4
-
-
-
0.7
3
ꢀ 1
.
Reaction conditions: W/F = 0.31 h; N
2
carrier gas flow rate, 30 cm min
1
-BuOH, 1-butanol; 2-BuOH, 2-butanol; BD, 1,3-butadiene; 2B1OL, 2-buten-1-ol; 3B2OL, 3-buten-2-ol; 3B1OL, 3-buten-1-ol; MEK, butanone; C3’, propylene. Others
include some unidentified products. mixture of cis and trans isomers.
a
6

A. Matsuda et al.
Molecular Catalysis 514 (2021) 111853
Scheme 2. Possible mechanisms for the dehydration of 3B1OL to BD (a) and 1-butanol to 1-butene (b).
Scheme 3. Proposed mechanism for the formation of BD in the dehydration of 1,4-BDO: the first-step dehydration to 3B1OL (a) and the second-step dehydration of
B1OL (b).
3
conversion is comparable to the decrease in SA. In the dehydration of
B1OL over Y Zr [14], we have also reported similar trends to those
2B1OL and 3B2OL (Fig. S5b, c), non-formation of propylene indicates
that propylene forms in the retro-Prince reaction of 3B1OL. In the
dehydration of 2B1OL, the BD yield increased with time on stream. This
is probably because 2B1OL is an intermediate in the formation of BD,
and the dehydration of 2B1OL to BD proceeded because the isomeriza-
tion of 2B1OL to 3B2OL was suppressed with time on stream. In addi-
tion, over typical acidic catalysts such as alumina and silica-alumina,
both the dehydration of 2B1OL to BD and the isomerization of 2B1OL to
3B2OL proceeded (Table S2). However, the catalysts behaved differ-
ently: the dehydration to BD was dominant over silica-alumina, and the
isomerization to 3B2OL was preferable over alumina. This difference
could be possibly explained by the acid property: in general, silica-
alumina has dominantly Brønsted acid and alumina has majorly Lewis
acid. It has been recently found that the isomerization between 2B1OL
2
2
2 7
O
observed in Fig. 8: the conversion of 2B1OL decreased with increasing
calcination temperature together with the decrease in SA of the catalyst,
2 2 7
except the Y Zr O -500. Thus, the decrease in the UOLs conversion with
calcination temperature is caused by the decrease in SA. This suggests
that the active catalysts for the dehydration of UOLs to BD do not need
2 2 7
high crystallinity of fluorite Y Zr O . Namely, the dehydration of UOLs
to BD would be catalyzed by the not-so-well-defined structure of active
sites as well as by the active sites for the dehydration of BDOs, which
need high crystallinity of fluorite Y
the conversion of 3B1OL over Y Zr
tion during the initial period for 5 h was observed in all the Y
2
Zr
2
O
7
. Fig. 9 shows the changes in
with time on stream. Deactiva-
Zr
2
2 7
O
2
2 7
O
samples irrespective of calcination temperature. Therefore, in the
dehydration of 1,4-BDO to BD (Fig. 7), the slight decrease in the BD yield
even at a high space time would be caused by the deactivation of active
sites for the dehydration of 3B1OL.
2 5 2
and 3B2OL is catalyzed by Lewis acid of V O -SiO [36].
The dehydration of C4 linear saturated alcohols such as 1-butanol
and 2-butanol is shown in Table 2, and the time-course of the conver-
sion of 1- and 2-butanol comparing with those of UOLs is displayed in
Fig. S6. The dehydration of 2B1OL and 1-butanol had no deactivation
with time on stream, whereas deactivation with time on stream was
observed in the dehydration of 3B1OL, 3B2OL, and 2-butanol. The
deactivation may be due to the polymerization of alkenes such as pro-
Table 2 and Fig. S5 show the results of the dehydration of UOLs such
◦
as 3B1OL, 2B1OL, and 3B2OL over Y
UOLs, deactivation behaviors of Y Zr
2
Zr
2
O
7
-900 at 360 C. Among the
2
2
O
7
-900 are different (Fig. S5): the
conversion of 3B1OL and 3B2OL decreased with time on stream, while
that of 2B1OL remained constant. In the reaction of 3B1OL (Fig. S5a), a
certain amount of propylene was produced. The propylene yield
remained constant throughout the initial 5 h, suggesting that the reac-
tion was independent and unrelated to the isomerization of 3B1OL to
2 2 7
pylene and butene. The reactivity of UOLs over Y Zr O -900 catalyst is
higher than that of butanols regardless of the position of the OH group.
These results support our previous prediction [12] that the dehydration
of 3B1OL proceeds by anchoring by the double bond of UOLs to the
2
B1OL and the dehydration of 3B1OL to BD. In the dehydration of
7

A. Matsuda et al.
Molecular Catalysis 514 (2021) 111853
surface (Scheme 2a). However, the reactivity of the monoalcohols over a
typical acidic catalyst such as alumina (Table S2) is different from that
4-BDO has been consumed. Both the first-step dehydration of 1,4-BDO to
produce 3B1OL and the second-step dehydration of 3B1OL to BD could
proceed via the base-acid concerted mechanism. The high BD yield
higher than 90% in the dehydration of 1,4-BDO can be supported by the
mechanism of Scheme 3: the two-step dehydration of 1,4-BDO may
separately proceed, but it could simultaneously proceed without
desorbing a 3B1OL intermediate in the BD formation at high tempera-
2 2 7
over Y Zr O -900 (Table 2). The reactivity of the monoalcohols over
alumina was in the order of 2B1OL > 2-butanol > 1-butanol > 3B1OL,
while silica-alumina had the order of 2B1OL > 2-butanol > 3B1OL >
1
3
3
-butanol. In addition, the reactivity order of 2B1OL > 1-butanol >
B1OL over H-ZSM-5 is also reported [35]. In these acid catalysts,
B1OL is not always more reactive than 1-butanol. Furthermore, 3B1OL
◦
tures as 360 C.
is readily decomposed via retro-Prince reaction over acidic catalysts
such as alumina and silica-alumina [37], indicating that a butenyl cation
would be hardly formed over the solid acids. On the other hand, over
4. Conclusions
Y
2
Zr
2
O
7
catalysts, 3B1OL was dehydrated to BD together with the iso-
Vapor-phase catalytic dehydration of 1,4-BDO was investigated over
REZrO catalysts prepared by the HT aging method. Among ten REZrO
mer of 2B1OL and a small amount of propylene, as shown in Fig. S5a.
Thus, the dehydration of 3B1OL would be explained by E2 mechanism
catalysts, Y
catalytic activity, and the Y
Dy Zr . Especially, Y Zr
2
Zr
2
O
7
and Dy
2
Zr
2
O
7
catalysts showed particularly high
catalyst was more stable than the
(
Scheme 2a) rather than E1 mechanism via carbocation over Y
catalysts. Because the much higher reactivity of 3B1OL than 1-butanol
Table 2) is limited by the surface of Y Zr -900, an attractive inter-
action between 3B1OL and Y Zr could accelerate the dehydration.
2
Zr
2
O
7
2
Zr
2
O
7
2
2
O
7
2
2
O
7
-700 shows the complete conversion of
◦
(
2
2
O
7
1,4-BDO with a high BD selectivity of 92.9% at 360 C during the time on
stream of 10 h.
2
2 7
O
Thus, we propose that the double bond of UOLs acts as an anchor to the
surface in the dehydration of 3B1OL (Scheme 2a) in contrast to
In the dehydration of 1,4-BDO over YZrO catalysts with different Y
◦
contents at 325 C, Y
2
Zr
2
O
7
with a Y/Zr ratio of 1, which has a cubic
1
-butanol that has no interaction of alkyl group with the surface
oxygen-defected fluorite structure, showed the highest BD formation
rate with high selectivity for 3B1OL. Although the density of surface
(
Scheme 2b).
In the dehydration of 2-butanol, 2-butene was expected to be the
oxygen vacancies in the cubic bixbyite Y
cubic oxygen-defected fluorite Y Zr , the formation rate of 3B1OL
over the Y Zr was higher than Y . This is explained by the dif-
ference in the aggregate state of the oxygen vacancies.
In the dehydration of 1,4-BDO over Y Zr calcined at different
Zr -700 at 360
Zr at
2 3
O is higher than that in the
main product over acidic catalysts according to the Zaitsev rule (Table
2
2 7
O
S2), whereas 1-butene was the most abundant product over Y
2
Zr
2
O
O
7
-900
and
2
2
O
7
2 3
O
(
Table 2). In the pioneering reports, REO catalysts such as Yb
2
3
Y
2
O
3
have Hoffmann elimination ability in the dehydration of 2-alcohol
38-40]. REO catalysts produced -olefins such as 1-butene and 4-meth-
yl-1-pentene with high selectivity in the dehydration of 2-butanol and
-methyl-2-pentanol, respectively [38,39]. It is also known that ZrO
has Hoffmann elimination ability with acid-base bifunctional catalysis
in the dehydration of 2-butanol [41]. It is interesting that ThO with a
fluorite structure also has a high Hoffmann elimination ability [38].
Therefore, it can be concluded that the Y Zr catalysts also promote
the Hoffmann elimination of 2-alcohol in a similar manner to Y [38]
and ZrO [41]. We have discussed that REO catalyst accelerates the
2
2 7
O
[
α
temperatures, the highest BD yield was observed in Y
2
2 7
O
◦
C. In the first-step dehydration of 1,4-BDO to 3B1OL over Y
2
2 7
O
◦
4
2
325 C, Y
2
Zr
2
O
7
-900 showed the highest yield of 3B1OL [25]. On the
contrary, in the second-step dehydration of 3B1OL to BD over Y
2
Zr
2
O
7
at
◦
2
360 C, the highest BD yield was observed in Y
2
Zr
2
O
7
-600, and the BD
yield decreased with increasing calcination temperature. Therefore, the
highest activity of Y Zr -700 in the one-step dehydration of 1,4-BDO
2
2
O
7
2
2 7
O
2
O
3
to produce BD can be explained by the trade-off relation in the activities
for the first-step dehydration of 1,4-BDO to 3B1OL and for the
second-step dehydration of 3B1OL to BD.
2
Hoffmann elimination in the dehydration of 2-alcohols, but there is no
Hoffmann-ruled product distribution in the dehydration of 1,3-BDO
In addition to 3B1OL, the dehydration reactivity of UOLs such as
2B1OL and 3B2OL that are isomers of 3B1OL as well as that of saturated
alcohols such as 1-butanol and 2-butanol was examined. Regardless of
the position of the OH group, all UOLs had higher reactivity than the
saturated alcohols. The C=C double bond of the UOLs could have an
attractive interaction such as anchoring to the catalyst surface, thereby
promoting the dehydration to BD. This suggests that the two-step
dehydration of 1,4-BDO simultaneously proceeds without desorbing a
3B1OL intermediate in the BD formation in the dehydration of 1,4-BDO
over Yb
groups of 1,3-BDO obviously contribute to the adsorption of 1,3-BDO
25,42,43], it is reasonable that the dehydration of 1,3-BDO does not
obey the elimination rules of Hoffmann and Zaitsev.
2 3 2 2 7
O [42], other REOs [43], and Y Zr O [25]. Because two OH
[
3
.6. Mechanistic consideration of the formation of BD from 1,4-BDO
We have discussed the reactivity of monoalcohols in the previous
◦
section. It is plausible that 3B1OL is much less reactive than 1,4-BDO
over Y Zr because the 3B1OL selectivity is quite high even at high
at high temperatures as 360 C.
2
2 7
O
conversions. After the reactant 1,4-BDO had been consumed at the
complete conversion, 3B1OL could be reacted because the reactivity
order is 1,4-BDO > 3B1OL > 1-butanol, as mentioned above.
Author contributions
AM prepared catalyst samples and carried out catalytic tests for the
Scheme 3 proposes a speculative reaction mechanism for the for-
mation of BD in the dehydration of 1,4-BDO over a defect site of
2
dehydration of 1,4-butanediol, XRD, TG, N adsorption for BET calcu-
lation, and wrote the first draft. YM prepared samples with different Y/
Zr ratios and performed catalytic tests for the dehydration of 1,3-butane-
diol. YY supervised the outline of this work from the viewpoint of
analytical chemistry and edited the manuscript. SS designed the project,
summarized the data, and edited the manuscript.
Y
2
Zr
2 7
O , which is illustrated in Fig. 3c. The selective formation of
3
B1OL in the dehydration of 1,4-BDO over the defect site is explained by
3
+
the adsorption as the form of tridentate coordination of 1,4-BDO to Y
and Zr4 cations and an O anion (Scheme 3a) [25]. A sequential
+
2ꢀ
dehydration proceeds: a hydrogen of 1,4-BDO at position 2 is firstly
abstracted by a basic O2 anion and then an OH group at position 1 is
ꢀ
Declaration of Competing Interest
subsequently or simultaneously abstracted by an acidic Y3 cation. The
+
other OH group at position 4 plays a significant role of anchoring 1,
We declare no conflict of interest. We have no known competing
financial interests or personal relationships that could have appeared to
influence the work reported in this paper.
4
-BDO to the surface. A similar mechanism of the 1,4-BDO dehydra-
tion over Gd catalyst is supported by theoretical calculation [44]. In
2 3
O
the second step, the produced 3B1OL could be promptly adsorbed on the
defect site at the complete conversion of 1,4-BDO (Scheme 3b). Because
1
,4-BDO is more reactive than 3B1OL, 3B1OL can be dehydrated after 1,
8

A. Matsuda et al.
Molecular Catalysis 514 (2021) 111853
Acknowledgments
[17] A. Igarashi, S. Sato, R. Takahashi, T. Sodesawa, M. Kobune, Catal. Commun. 8
(
2007) 807–810.
[
18] S. Sato, R. Takahashi, M. Kobune, H. Inoue, Y. Izawa, H. Ohno, K. Takahashi, Appl.
This work was supported by JSPS KAKENHI KIBAN B Grant Number
JP18H01784 and JP21H01711.
Catal. A: Gen. 356 (2009) 64–71.
[19] Y. He, Q. Li, Y. Wang, Y. Zhao, Chinese J. Catal. 31 (2010) 619–622.
[
20] S. Otsuka, T. Nemoto, R. Yotsumoto, Y. Yamada, F. Sato, R. Takahashi, S. Sato,
Appl. Catal. A: Gen. 575 (2019) 48–57.
Supplementary materials
[
[
21] N. Yamamoto, S. Sato, R. Takahashi, K. Inui, Catal. Commun. 6 (2005) 480–484.
22] N. Yamamoto, S. Sato, R. Takahashi, K. Inui, J. Mol. Catal. A: Chem. 243 (2006)
5
2–59.
23] Q. Zhang, Y. Zhang, H. Li, C. Gao, Y. Zhao, Appl. Catal. A: Gen. 466 (2013)
33–239.
24] T. Nemoto, Y. Yamada, F. Sato, R. Takahashi, S. Sato, Mol. Catal. 473 (2019),
10399.
Supplementary material associated with this article can be found, in
the online version, at doi:10.1016/j.mcat.2021.111853.
[
[
[
[
2
1
References
25] A. Matsuda, Y. Matsumura, K. Nakazono, F. Sato, R. Takahashi, Y. Yamada, S. Sato,
Catalysts 10 (2021) 1392.
26] A. Igarashi, N. Ichikawa, S. Sato, R. Takahashi, T. Sodesawa, Appl. Catal. A: Gen.
[
[
1] C. Angelici, B.M. Weckhuysen, P.C.A. Bruijnincx, ChemSusChem 6 (2013)
1
595–1614.
3
00 (2006) 50–57.
2] E.V. Makshina, M. Dusselier, W. Janssens, J. Degre`ve, P.A. Jacobs, B.F. Sels, Chem.
Soc. Rev. 43 (2014) 7917–7953.
[
[
[
27] K. Momma, F. Izumi, J. Appl. Crystallogr. 44 (2011) 1272–1276.
28] Y. Xu, M. Yamazaki, P. Villars, Jpn. J. Appl. Phys. 50 (2011) 11RH02.
29] D.-J. Kim, S.-H. Hyun, S.-G. Kim, M. Yashima, J. Am. Ceram. Soc. 77 (1994)
[
[
3] J.-H. Park, C.-H. Shin, Appl. Catal. A Gen. 49 (2015) 1–7, 5.
4] D. Sun, Y. Li, C. Yang, Y. Su, Y. Yamada, S. Sato, Fuel Proc. Technol. 197 (2020),
5
97–599.
1
06193.
[
[
30] V.P. Buntushkin, I.V. Romanovich, N.I. Timofeeva, Inorg. Mater. 7 (1971)
[
5] H. Yim, R. Haselbeck, W. Niu, C. Pujol-Baxley, A. Burgard, J. Boldt, J. Khandurina,
J.D Trawick, R.E Osterhout, R. Stephen, J. Estadilla, S. Teisan, H.B. Schreyer,
S. Andrae, T.H. Yang, S.Y. Lee, M.J Burk, S.V. Dien, Nat. Chem. Biol. 7 (2011)
1
456–1457.
31] M. Segawa, S. Sato, M. Kobune, T. Sodesawa, T. Kojima, S. Nishiyama, N. Ishizawa,
J. Mol. Catal. A: Chem. 310 (2009) 166–173.
4
45–452.
[
[
[
32] S. Sato, F. Sato, H. Gotoh, Y. Yamada, ACS Catal 3 (2013) 721–734.
33] R.D. Shanon, Acta Cryst A32 (1976) 751–767.
[
6] Y.S. Tai, M. Xiong, P. Jambunathan, J. Wang, J. Wang, C. Stapleton, K. Zhang, Nat.
Chem. Biol. 12 (2016) 247–253.
34] O.A. Abdelrahman, D.S. Park, K.P. Vinter, C.S. Spanjers, L. Ren, H.J. Cho, D.
G. Vlachos, W. Fan, M. Tsapatsis, P.J. Dauenhauer, ACS Sustainable Chem. Eng. 5
[
[
7] F. Li, T. Lu, B. Chen, Z. Huang, G. Yuan, Appl. Catal. A Gen. 478 (2014) 252–258.
8] B.K. Ly, D.P. Minh, C. Pinel, M. Besson, B. Tapin, F. Epron, C. Especel, Top. Catal.
(
2017) 3732–3736.
5
5 (2012) 466–473.
[
35] S. Li, O.A. Abdelrahman, G. Kumar, M. Tsapatsis, D.G. Vlachos, S. Caratzoulas, P.
J. Dauenhauer, ACS Catal 9 (2019) 10279–10293.
[
9] N.T.T. Nguyen, F. Matei-Rutkovska, M. Huchede, K. Jaillardon, G. Qingyi,
C. Michel, J.M.M. Millet, Catal. Today 323 (2019) 62–68.
[
[
36] T. Ozawa, L. Yu, Y. Yamada, S. Sato, Chem. Lett. 50 (2021) 1635–1638.
37] D. Sun, S. Arai, H. Duan, Y. Yamada, S. Sato, Appl. Catal. A: Gen. 531 (2017)
[
10] III F. Zeng, W.J. Tenn, S.N.V.K. Aki, J. Xu, B. Liu, K.L. Hohn, J. Catal. 344 (2016)
7
7–89.
2
1–28.
[
[
[
11] H. Duan, Y. Yamada, S. Sato, Appl. Catal. A Gen. 491 (2015) 163–169.
12] Y. Wang, D. Sun, Y. Yamada, S. Sato, Appl. Catal. A: Gen. 562 (2018) 11–18.
13] M. Huchede, T.T.N. Nguyen, L. Cardenas, M. Aouine, K. Jaillardon, J.-M.M. Millet,
J. Phys. Chem. C 124 (2020) 2060–2069.
[
[
[
38] A.J. Lundeen, R.V. Hoozer, J. Org. Chem. 32 (1967) 3386–3389.
39] S. Bernal, J.M. Trillo, J. Catal. 66 (1980) 184–190.
40] B.H. Davis, S.N. Russell, P.J. Reucroft, R.B. Shalvoy, J. Chem. Soc. Faraday Trans. I
7
6 (1980) 1917–1922.
[
[
[
14] Y. Matsumura, A. Matsuda, Y. Yamada, S. Sato, Bull. Chem. Soc. Jpn. 94 (2021)
[
[
41] T. Yamaguchi, H. Sasaki, K. Tanabe, Chem. Lett. 2 (1973) 1017–1018.
1
651–1658.
42] S. Sato, R. Takahashi, T. Sodesawa, A. Igarashi, H. Inoue, Appl. Catal. A: Gen. 328
15] S. Sato, R. Takahashi, T. Sodesawa, N. Yamamoto, Catal Commun 5 (2004)
(
2007) 109–116.
3
97–400.
[
[
43] H. Gotoh, Y. Yamada, S. Sato, Appl. Catal. A: Gen. 377 (2010) 92–98.
16] R. Richard, M. Jacquin, M. Dorato, N. Pacheco, C. Rannoux, Patent US10322978
B2(2019).
44] F. Sato, S. Sato, Y. Yamada, M. Nakamura, A. Shiga, Catal. Today 226 (2014)
1
24–133.
9