Formation of a New, Strongly Basic Nitrogen Anion by Metal Oxide Modification

Article abstract of DOI:10.1021/jacs.7b05227

Development of new hybrid materials having unique and unprecedented catalytic properties is a challenge for chemists, and heterogeneous-homogeneous hybrid catalysts have attracted much attention because of the preferable and exceptional properties that are highly expected to result from combination of the components. Base catalysts are widely used in organic synthesis as key materials, and a new class of base catalysts has made a large impact from academic and industrial viewpoints. Here, a principle for creating a new strong base by hybridization of homogeneous and heterogeneous components is presented. It is based on the modification of organic compounds with metal oxides by using the acid-base property of metal oxides. Based on kinetic and DFT studies, combination of CeO₂ and 2-cyanopyridine drastically enhanced the basicity of 2-cyanopyridine by a factor of about 10⁹ (～9 by pK_a (in CH₃CN)), and the pK_a was estimated to be ～21, which locates it in the superbase category. 2-Cyanopyridine and CeO₂ formed a unique adsorption complex via two interaction modes: (i) coordinative interaction between the Ce atom of CeO₂ and the N atom of the pyridine ring in 2-cyanopyridine, and (ii) covalent interaction between the surface O atom of CeO₂ and the C atom of the CN group in 2-cyanopyridine by addition of the lattice oxygen of CeO₂ to the CN group of 2-cyanopyridine. These interactions established a new, strongly basic site of N^- over the CeO₂ surface.

Full text of DOI:10.1021/jacs.7b05227

Subscriber access provided by UNIVERSITY OF THE SUNSHINE COAST
Article
Formation of A New Strong Basic Nitrogen Anion by Metal Oxide Modification
Masazumi Tamura, Ryota Kishi, Akira Nakayama, Yoshinao
Nakagawa, Jun-ya Hasegawa, and Keiichi Tomishige
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b05227 • Publication Date (Web): 28 Jul 2017
Downloaded from http://pubs.acs.org on July 28, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 12
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Formation of A New Strong Basic Nitrogen Anion by Metal Oxide
Modification
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Masazumi Tamura,*,†,‡, Ryota Kishi,† Akira Nakayama,*‡,§ Yoshinao Nakagawa,† Jun-Ya Hasegawa,§
Keiichi Tomishige*,†
†Department of Applied Chemistry, Graduate School of Engineering, Tohoku University, 6-6-07 Aoba, Aramaki, Aoba-
ku, Sendai 980-8579, Japan
‡JST, PRESTO, 4-1-8 Honcho Kawaguchi, Saitama 332-0012, Japan
§Institute for Catalysis, Hokkaido University, Kita 21 Nishi 10, Kita-ku, Sapporo 001-0021, Japan
ABSTRACT: Development of a new hybrid material having unique and unprecedented catalytic property is a challenge for
chemists, and heterogeneous-homogeneous hybrid catalysts have attracted much attention because the preferable and
exceptional properties are highly expected by combination of the components. Base catalysts are widely used in organic
synthesis as key materials, and a new class of base catalysts has a large impact from the academic and industrial viewpoints.
Here, a principle for creating a new strong base by hybridization of homogeneous and heterogeneous components is
presented. It is based on the modification of organic compounds with metal oxides by using the acid-base property of metal
oxides. Based on the kinetic and DFT studies, combination of CeO2 and 2-cyanopyridine drastically enhanced the basicity of
2-cyanopyridine by a factor of about 109 (~9 by pKa (in CH3CN)), and the pKa was estimated to be ~21, which is located in
superbase analogy. 2-Cyanopyridine and CeO2 formed a unique adsorption complex via two interaction modes: (i)
coordinative interaction between Ce atom of CeO2 and N atom of the pyridine ring in 2-cyanopyridine, and (ii) covalent
interaction between the surface O atom of CeO2 and C atom of the CN group in 2-cyanopyridine by addition of the lattice
oxygen of CeO2 to the CN group of 2-cyanopyridine. These interactions established a new strong base site of N- over the
CeO2 surface.
covalent bonds between these components. Among
■
Introduction
heterogeneous components, metal oxides are attractive mate-
rials for the hybrid catalysts due to the unique acid-base and
redox property as well as the high-thermal stability and dura-
bility [9], and they are mainly used as supports for the hetero-
geneous-homogeneous hybrid catalysts. If the acid-base site of
metal oxides can be applied to formation of a new catalytically
active site by direct interaction between metal oxides and
homogeneous components, unique catalytic property will be
expected.
Inorganic-organic hybrid materials are one of the most im-
portant classes in catalysis, and they often show not only a
unique property combining the properties of the individual
components but also an exceptional property derived from
intimate interaction between the components, which leads to
creation of a new catalysis concept [1]. Among the hybrid
materials, heterogeneous-homogeneous hybrid catalysts have
been intensively investigated, and prominent catalyst systems
have been reported, such as organosiliceous hybrid materials
[2], nanoparticles modified with SAMs (Self-assembled
monolayers), linear polymers, and dendrimers [3-5], organic-
inorganic porous materials [6], and supported metals
modified with organic modifiers [7]. The active sites of heter-
ogeneous-homogeneous catalysts are on any one of
heterogeneous components, homogeneous components, and
interface between these components. In the case that the ac-
tive sites are on homogeneous components, heterogeneous
components are typically used for immobilizing the
homogeneous catalytic components, bestowing reusability
and durability, which are characteristic properties of
heterogeneous components, on homogeneous catalytic com-
ponents. In some cases, heterogeneous components work as
co-catalysts for homogeneous catalytic components using
their acid-base property [8]. However, to the best of our
knowledge, there is no reports on formation of a new catalytic
active site by direct interaction between heterogeneous com-
ponents and homogeneous components, such as formation of
Base catalysts are essential tools in organic synthesis be-
cause deprotonation is involved in a variety of organic reac-
tions as a key step [10], and many industrial products have
been manufactured by base catalysts. Among the wide range
of base catalogues, neutral superbases such as DBU
(diazabicycloundecene), amidines, guanidines and phos-
phazenes have been widely applied to various important or-
ganic reactions [11], and recently, various exquisite superbas-
es have been developed by modification or expanding the
above base materials [12]. Therefore, establishment of a new
class of base catalysts is essential from the academic and in-
dustrial viewpoints.
In our preliminary research [13], we found that combina-
tion catalyst of CeO2 and 2-cyanopyridine can be easily formed
by mixing these components, and this catalyst was effective
for hydromethoxylation of acrylonitrile as a model base-
catalyzed reaction, showing more than 2000-fold higher
reaction rate than only CeO2 or only 2-cyanopyridine. Howev-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 12
er, the catalyst structure and reaction mechanism over
was calculated to be about 19, which was much larger than
1
2
3
4
5
6
7
8
CeO2+2-cyanopyridine remains unclear. Herein, based on the
kinetics and DFT studies, we clarified the catalyst structure of
CeO2+2-cyanopyridine, where a strong base site of N anion (N-
) was formed via addition of the lattice oxygen of CeO2 to the
CN group in 2-cyanopyridine. The basicity was much higher
than those over only CeO2 or only 2-cyanopyridine (about 109-
fold), and the estimated pKa is ~21, which is in the superbase
analogy.
those using only CeO2 (estimated pKa of the conjugated acid:
13) or only 2-cyanopyridine (estimated pKa of the conjugated
acid: 8.6). Addition of 2-cyanopyridine to CeO2 drastically
increased the basicity by a factor of more than 106 even in the
Knoevenagel condensation reaction. These results indicate
that the catalyst system has high potential for application to
various base-catalyzed reactions. Therefore, the strong base
site was formed by hybridization of heterogeneous CeO2 and
homogeneous 2-cyanopyridine.
9
Next, the scope of conjugated olefins and alcohols in the hy-
droalkoxylation reaction was investigated using the combina-
tion catalyst of CeO2 and 2-cyanopyridine. Table 1 shows the
■
Result and discussion
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Evaluation of catalytic performance of CeO2+2-
cyanopyridine. We previously reported that CeO2 and 2-
cyanopyridine formed a catalytic complex, and the adsorption
amount of 2-cyanopyridne on CeO2 was estimated to be 0.28
mmol/g [13]. In order to determine the basicity of the active
sites over CeO2+2-cyanopyridine, hydromethoxylation of acry-
lonitrile was carried out as a model reaction using typical or-
ganic base compounds such as pyridine, 4-methoxypyridine,
N,N-dimethylpyridine and 1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU). The logarithmic reaction rate (Log(TOF)) is plotted
versus pKa values [14] of the conjugated acids of the organic
base compounds (Figure 1, and the details are in Table S1).
The linear relationship between the Log(TOF) and pKa was
obtained, meaning that the reaction is a base-catalyzed one.
From the relationship, the pKa of CeO2+2-cyanopyridine based
on the surface 2-cyanopyridine was calculated to be about 21,
which is in the category of superbases [10]. This value was by
far larger than those using only CeO2 (estimated pKa of the
conjugated acid: 12) or only 2-cyanopyridine (estimated pKa
of the conjugated acid: 11), indicating that addition of 2-
cyanopyridine to CeO2 drastically increased the basicity by a
factor of more than 109. The increase of pKa is comparable to
or larger than that in the case of N,N-dimethylaniline and
DMAN (1,8-Bis(dimethylamino)naphthalene), a classical pro-
ton sponge, discovered by Alder [16], where the basicity of
DMAN is 107-fold higher than that of N,N-dimethylaniline. In
addition, the catalyst of CeO2+2-cyanopyridine was applied to
another typical base-catalyzed reaction of Knoevenagel con-
densation between benzaldehyde and ethyl cyanoacetate,
which involves deprotonation of the C-H bond. The correla-
tion between the basicity and pKa was also evaluated by the
same way as above (Figure S1 and Table S2). The pKa of
CeO2+2-cyanopyridine based on the surface 2-cyanopyridine
results
of
various
conjugated
olefins
in
the
hydromethoxylation. Acrylonitrile, methyl vinyl ketone and
methyl acrylate were converted to the corresponding ethers
in high yields (92-98%, Table 1, entries 1-3). In addition,
acrylamide and N,N-dimethyl acrylamide also reacted to af-
ford the target ethers in moderate yields, although a large
amount of CeO2 and longer reaction time were required (Ta-
ble 1, entries 4 and 5).
The scope of alcohols was also studied in the alkoxylation of
acrylonitrile using CeO2 and 2-cyanopyridine (Table 2). Linear
alcohols easily reacted to afford the corresponding ethers in
high yields, although the longer alkyl alcohols showed lower
reactivity (Table 2, entries 1-5). Unsaturated linear alcohols
such as 4-penten-1-ol and 2-hexene-1-ol were also
transformed to the ethers in high yields with maintaining the
olefin moiety (Table 2, entries 6 and 7). In addition, 2-methyl-
1-propanol, cyclohexanemethanol and benzyl alcohol, which
are β-branched alcohols, underwent the reaction to provide
the corresponding products in high yields (Table 2, entries 8-
10). It should be noted that tetrahydrofurfuryl alcohol and
furfuryl alcohol, which are key intermediates from biomass,
were applicable to the reaction, affording the corresponding
ethers in high yields (Table 2, entries 11 and 12). α-Branched
alcohols such as 2-propanol and cyclohexanol could react in
spite of the bulkiness of the substrates, providing the products
in high yields (Table 2, entries 13 and 14). However, no prod-
uct was observed in the case of tert-butyl alcohol, a highly
Table 1. Scope of conjugated olefins in the
hydromethoxylation over CeO2 and 2-cyanopyridine
CeO2 + 2ꢀCyanopyridine
O
CH3OH
+
R
R
323 K, air
4
(R2=0.94)
2-Cyano
pyridine
(mmol)
N
CN
Conjugated CeO2
t
Conv. Sel.
CeO2 +
Entry
N
olefin
(mmol)
(h) (%) (%)
2
0
N
N
N
1
2
1
2
1
4
10
16
98 >98
CN
95
94
72
97
98
80
O
O
3
4
4
6
4
4
48
48
O
N
O
NH2
CeO2
ꢀ2
ꢀ4
O
N
N
CN
N
5
8
4
24
57 >99
O
8 10 12 14 16 18 20 22 24 26 28 30
Reaction conditions: conjugated olefin (10 mmol), methanol
(15 mmol), CeO2 (1-8 mmol), 2-cyanopyridine (1-4 mmol),
323 K, air.
pKa (in CH3CN)
Figure 1. The relationship between the pKa and Log(TOF) of
the organic base compounds in the hydromethoxylation of
acrylonitrile, and estimation of pKa of CeO2+2-cyanopyridine,
CeO2, and 2-cyanopyridine. The detailed data are shown in
Table S1.
ACS Paragon Plus Environment

Page 3 of 12
Journal of the American Chemical Society
Table 2. Scope of alcohols in the hydroalkoxylation of acrylo-
nitrile using CeO2 and 2-cyanopyridine
Table 3. Scope of additives in the hydromethoxylaton of acry-
lonitrile with CeO2
1
2
3
4
5
6
7
8
9
CeO2 + 2ꢀCyanopyridine
Entry
1
Nitrile
t (h) Yield (%) Va (mmol h-1 g-1)
O
R OH
+
CN
R
CN
323 K, air
N
CN
0.5
0.5
20
22
23
25
t
(h)
CeO2
(mmol)
Conv.
(%)
98
97
99
99
90
99
97
Sel.
Entry
Alcohol
N
CN
(%)
>99
>99
2
CH3OH
OH
1
2
1
1
10
N
CN
CN
12
12
8
3
4
1
1
14
8.4
F
3
1
>99
OH
N
4
1
>99
5.3
2.7
3.4
1.5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OH
Cl
5
6
7
2
2
3
24
8
24
>99
>99
>99
OH
N
CN
OH
OH
5
1
NO2
N
OH
NC
CN
CN
8
9
1
1
8
8
99
99
>99
>99
6
7
1
4
11
6.4
OH
N
2.8
8.9
0.42
OH
10
11
12
1
4
4
24
48
48
>99
97
>99
>99
>99
CN
OH
N
O
O
8
24
0.21
N
CN
OH
85
N
9
48
48
0.9
0.6
0.01
0.01
13
14
2.5
4
24
24
90
90
>99
>99
CN
OH
OH
N
10
CN
N
15
3
24
<0.1
-
OH
11
12
13
14
24
48
24
24
5.0
0.12
<0.01
0.03
0.4
Reaction conditions: acrylonitrile (10 mmol), alcohol (20
mmol), CeO2 (1-4 mmol), 2-cyanopyridine (2 mmol), 323 K,
air.
N
CN
0.12
1.2
N
N
bulky alcohol (Table 2, entry 15), which will be due to the
steric hindrance of tert-butyl group. As above, the catalyst
system composed of CeO2 and 2-cyanopyridine was applicable
to the reactions of various alcohols and conjugated olefins.
CN
CN
16
N
CN
CN
15b
16b
1
1
18
11
10
N
N
Kinetics and spectroscopic studies. From our preliminary
study on the effect of 2-cyanopyridine concentration [13],
CeO2 and 2-cyanopyridine followed Michaelis-Menten type
kinetics, resulting in formation of a self-assembled adsorption
complex of CeO2 and 2-cyanopyridine. It was also proposed
that 2-cyanopyridine was adsorbed on the surface of CeO2 at
the two positions of 2-cyanopyridine, the N atoms in pyridine
ring and CN group. In order to obtain further information on
the active sites including the detailed adsorption structure
over CeO2+2-cyanopyridine, kinetic studies such as effect of
calcination temperature, nitrile structure, substrate
concentration, isotope effect of methanol (VCH3OH/VCH3OD) and
effect of the substituents in conjugated olefins were carried
out in the hydromethoxylation of conjugated olefins as a
model reaction. In addition, spectroscopic analyses on the
color of adsorption complexes of CeO2 and additives by UV-Vis
and methanol adsorption by FTIR were also carried out.
6.3
N
CN
N
17
18
0.5
1
20
23
N
CN
7.9
4.6
CN
CN
O
S
19b
20
1
9.4
0.8
5.4
48
0.01
H
N
CN
21
48
1.4
0.02
22b
23
24
3
22
4.1
O
CN
CN
24
4.6
5.9
1.1
0.11
H
N
CN
24
0.14
In order to clarify the influence of the surface state of
CeO2, the effect of calcination temperature was investigated
using CeO2 samples calcined at different temperatures (473-
1273 K, Table S3 and Figure S2). The highest reaction rate per
CeO2 amount (V / mmol h-1 g-1) was obtained at 873 K, and the
V decreased with increasing the calcination temperature over
873 K (Figure S2(a)), which is due to decrease of the surface
area. The reaction rate per the surface area (V’ / mmol h-1 m-2)
25
none
48
0.01
Reaction conditions: acrylonitrile (10 mmol), methanol (20
mmol), CeO2 (1 mmol), additive (1 mmol or b2 mmol), 323 K,
air. aThe reaction rate, V (mmol h-1 g-1) = (Amount of 3-
methoxypropionitrile
(h))/(amount of CeO2 (g)). bThe data were quoted from the
previous report [13].
(mmol))/(Reaction
time
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 12
was plotted as a function of the calcination temperature (Fig-
eroaromatics are not always necessary, however, the position
of the heteroatom with respect to the CN group is important.
1
2
3
4
5
6
7
8
ure S2(b)). The V’ was almost constant above 800 K. Consider-
ing that the main surface of CeO2 calcined at high calcination
temperature is Ce(111) [17,18], which is the most stable facet,
the active site is mainly formed on the Ce(111).
From the above results, the effective additives must fulfill at
least following two factors. (i) Both the CN group and other
heteroatoms such as N or O are necessary. (ii) The heteroatom
is located at the α-position of the CN group. The tendency of
the effective additives is similar to the cases of carbonate syn-
thesis from CO2 and alcohols over CeO2 [16]. In addition, a
bulky substituent around the heteroatom will not be prefera-
ble, which will be reasonable by assuming that the effective
additives are adsorbed on the surface of CeO2 at the two posi-
tions of the CN group and other heteroatom. The detailed ad-
sorption state is discussed later.
The structure of the nitrile additives has large influence on
the formation of the hybrid base site. It is of great importance
to know the detailed structure of effective nitrile additives for
clarification of the active site over CeO2+2-cyanopyridine.
Various additives were applied to the hydromethoxylation of
acrylonitrile with CeO2 (Table 3). p-Substituted 2-
cyanopyridines (Table 3, entries 1-4) were effective, however,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
the reaction rate decreased with
a stronger electron-
withdrawing group. In addition, m-nitro(4-nitro)-2-
cyanopyridine (Table 3, entry 5) was also effective, however,
the reaction rate was lower than those with other 2-
cyanopyridine derivatives. Figure 2 shows the relationship
between the reaction rate and substituent parameters, Ham-
The color of the reaction solution often provides a hint for
the interaction between the components. In the previous re-
port, we demonstrated that the yellow color of the methanol
solution of CeO2 was thickened by addition of 2-cyanopyridine,
which will be related to the electron transfer between CeO2
and 2-cyanopyridine. In order to clarify the difference of color
change between the effective additives and ineffective addi-
tives, the color of the solution of CeO2 and various additives
was investigated using 2-cyanopyridine, 5-methyl-2-
cyanopyridine, 3-cyanopyridine, 4-cyanopyridine, pyridine,
benzonitrile and 2-picolinamide (Figure 3). Obviously, the
color change was observed in the solution of CeO2 with 2-
cyanopyridine and 5- methyl-2-cyanopyridine, which are
effective additives, in contrast, no color change was observed
with other additives, which were ineffective additives (Figure
3(a)). It should be noted that 2-picolinamide, which is pro-
duced by hydration of 2-cyanopyridine, showed no color
change, indicating that the color change in the case of CeO2
and 2-cyanopyridine is not derived from 2-picolinamide. In
addition, the solutions were analyzed by UV-vis (Figure 3(b)).
As expected from the results of the photographs of the solu-
tions, new bands were observed in the solution of CeO2 with
2-cyanopyridine and 5-methyl-2-cyanopyridine, which are
effective additives. In addition, it was also confirmed that such
color change was not observed with other metal oxide+2-
mett parameters (
tween log(Vx/VH) and Hammett parameters. The slope of the
correlation was negative ( =-1.3), indicating that the electron-
σx), providing a good linear correlation be-
ρ
donating group in 2-cyanopyridines is preferable as the
substituent. 6-Substituted 2-cyanopyridines such as 6-methyl-
2-cyanopyridine and 2,6-dicyanopyridine (Table 3, entries 6
and 7) were also applied to the reaction, providing positive
effect. However, the reaction rates were lower than that in the
case of 2-cyanopyridine, which can be interpreted by the ste-
ric hindrance of the substituents at the 6-position. With dimer
of 2-cyanopyridine at the 6-position, 6,6'-Dicyano-2,2'-
bipyridyl, the reaction a little proceeded (Table 3, entry 8) and
the reaction rate was very low (0.21 mmol h-1 g-1), which is
also due to the bulkiness of the structure for adsorption. On
the other hand, 3-cyanopyridine and 4-cyanopyridine (Table
3, entries 9 and 10), which are isomers of 2-cyanopyridine,
were not effective for the reaction. Furthermore, pyridine
(Table 3, entry 11) was not effective. These results indicate
that CN group is essential for the effective additives, and the
position with respect to the N atom in heteroaromatics is im-
portant. This feature for the effective additives was also
confirmed by the results with (2-pyridyl)acetonitrile having
the CN group at the β-positon of the N atom in pyridine ring
(Table 3, entry 12). (3-Pyridyl)acetonitrile and 3-
pyridinepropionitrile (Table 3, entries 13 and 14) showed
almost no positive effect for the reaction.
N
CN
CeO2 +
R
O
CH3OH
+
CN
CN
323 K, air
0.5
Other N-heteroaromatic nitriles such as 2-cyanopyrazine,
pꢀCH3
pyrimidine,
1-isoquinolinecarbonitrile
and
2-
pꢀH
0
ꢀ0.5
ꢀ1
quinolinecarbonitrile were effective, showing high reaction
rate (Table 3, entries 15-18), and the reaction rate with 1-
isoquinolinecarbonitrile was comparable to that of 2-
cyanopyridine. On the other hand, 2-cyanofuran, an O-
heteroaromatic nitrile (Table 3, entry 19), was also effective
for the reaction, however, 2-cyanothiophene, an S-
heteroaromatic nitrile, and 2-cyanopyrrole were not effective
at all (Table 3, entries 20 and 21). These results indicate that
bared O and N atoms in heteroaromatics are necessary for the
effective additives.
ρ
= ꢀ1.3
pꢀOCH3
pꢀF
pꢀCl
mꢀNO2
ꢀ1.5
ꢀ0.5
0
0.5
1
1.5
In addition, linear nitriles were also applied to the reaction
as additives (Table 3, entries 22-24). Methoxyacetonitrile
having O atom at the α-position of the CN group was an
effective additive (Table 3, entry 22), however, n-
σx
Figure 2. Relationship between activity and Hammett param-
eter of p-(5-) or m-(4-)substituted 2-cyanopyridines in the
hydromethoxylation of acrylonitrile using CeO2 and 2-
cyanopyridine derivatives. Reaction conditions: acrylonitrile
(10 mmol), methanol (15 mmol), CeO2 (1 mmol), 5-
substituted 2-cyanopyridine (1 mmol), 323 K, air, 0.5-2 h. The
detailed data are shown in Table S4.
butyronitrile,
a
simple liner alkyl nitrile, and N-
methylaminoacetonitrile were not effective for the reaction
(Table 3, entries 23 and 24). These results indicate that het-
ACS Paragon Plus Environment

Page 5 of 12
Journal of the American Chemical Society
(a)
(a)
4
3
1
A
B
C
D
E
F
G
H
2
3
4
5
CeO2+2ꢀcyanopyridine
Slope=0.3
ꢀ1
2
(b)
6
≈
7
8
9
2
ꢀ2
ꢀ3
ꢀ4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CeO2
Slope=0.0
A,D,E,F,G,H
ꢀ1
0
1
2
3
Ln([Methanol] / M)
C
B
(b)
4
2
400
500
600
700
800
Wavelength / nm
Figure 3. (a) Photographs of CeO2+additive solution. A: none
(only CeO2), B: 2-cyanopyridine, C: 5-methyl-2-cyanopyridine,
D: 3-cyanopyridine, E: 4-cyanopyridine, F: benzonitrile, G:
acetonitrile, H: pyridine, (b) UV-vis spectra of CeO2+additives
in methanol solution. Conditions: methanol (20 mmol), CeO2
(1 mmol), additive (2 mmol), 303 K. Black: none (A), Red: 2-
cyanopyridine (B), green: 5-methyl 2-cyanopyridine (C), blue:
3-cyanopyridine (D), yellow: 4-cyanopyridine (E), light green:
benzonitrile (F), brown: 2-picolinamide (G), pink: pyridine
(H).
CeO2+2ꢀcyanopyridine
Slope=1.0
0
ꢀ2
ꢀ4
ꢀ6
CeO2
Slope=1.0
ꢀ4
ꢀ2
0
2
4
6
Ln([Acrylonitrile] / M)
Figure 4. The effect of methanol concentration (a) and acry-
lonitrile concentration (b) in the hydromethoxylation of acry-
lonitrile using CeO2(ꢀ), and CeO2 + 2-cyanopyridine (). Re-
action conditions of (a): acrylonitrile (40 mmol), methanol (8-
14 mmol), CeO2 (1 mmol), 2-cyanopyridine (0 or 1 mmol), 323
K, air, 0.5 h. Reaction conditions of (b): acrylonitrile (5-30
mmol), methanol (100 mmol), CeO2 (1 mmol), 2-
cyanopyridine (0 or 1 mmol), 323 K, air, 0.5 h. The detailed
data are shown in Tables S5-8.
cyanopyridine (Figure S3). Therefore, the effective additives
caused the color change of the solution of CeO2, and an elec-
tron-donating substituent more thickened the yellow color of
CeO2. These results indicate that the color change is strongly
related to formation of the active site, and it is expected that
effective additives and CeO2 will form a unique adsorption
complex. The detail of the adsorption state of CeO2 and effec-
tive additives is discussed later.
Figure 4(a) shows the dependence of methanol concentra-
tion over CeO2 and CeO2+2-cyanopyridine in the
hydromethoxylation of acrylonitrile by changing the methanol
amount (8-14 mmol). The reaction rates were almost constant
over both the catalysts, and the reaction orders with respect
to the methanol concentration over CeO2 and CeO2+2-
cyanopyridine were calculated to be 0.0 and 0.3, respectively,
implying that methanol was strongly adsorbed on both the
catalysts. Figure 4(b) shows the dependence of acrylonitrile
concentration over CeO2 and CeO2+2-cyanopyridine in the
same reaction by changing acrylonitrile amount (5-30 mmol).
The reaction rates increased with increase of the acrylonitrile
concentration over both the catalysts, and the reaction orders
with respect to acrylonitrile concentration were calculated to
be +1 over both the catalysts, indicating that reaction of acry-
lonitrile was involved in the rate determining step.
to the previous report, methoxy adspecies were observed
over both the catalysts immediately after introduction of
methanol, suggesting that methanol was adsorbed on CeO2
and CeO2+2-cyanopyridine to form methoxy species at both
the catalysts. The intensity of the methoxy adspecies over
CeO2+2-cyanopyridine was smaller than that over CeO2, and
particularly the band at 1034 cm-1, which can be assigned to
double bridged and triple bridged methoxy adspecies [19],
was decreased by addition of 2-cyanopyridine. These results
suggest that methanol can be adsorbed on CeO2+2-
cyanopyridine, while adsorption amount of methanol de-
creased by introduction of 2-cyanopyridine. However, in the
FTIR spectrum of methoxy adspecies over CeO2+2-
cyanopyridine, we cannot detect any new methoxy adspecies
derived from the active sites of CeO2+2-cyanopyridine, which
is due to a small amount of the active sites. In addition, isotope
In order to confirm the adsorption state of methanol on
the surface of CeO2 and CeO2+2-cyanopyridine at the present
reaction conditions, adsorption of methanol on both the cata-
lysts was investigated by transmission FTIR spectroscopy at
323 K, the temperature of the present reaction conditions
(Figure S4). It is well-known that methanol can be easily
dissociated on the acid-base sites of CeO2 to form methoxy
adspecies and proton even at room temperature [19]. Similar
effect
of
methanol
was
investigated
in
the
hydromethoxylation of acrylonitrile using CH3OH and CH3OD,
and the time-courses are shown in Figure 5. The reaction
rates with CH3OH and CH3OD were almost the same. Consider-
ing the result of the FTIR analyses of methanol adspecies, dis-
sociation of methanol to methoxy species and proton on
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 12
CeO2+2-cyanopyridine was very fast, which supports the al-
Figure 6. A good linear relationship was obtained, and the
1
2
3
4
5
6
7
8
most zero reaction order with respect to methanol concentra-
tion (Figure 4(a)). Considering that the activity over CeO2+2-
cyanopyridine is much higher than that over CeO2, the kinetic
studies in Figures 4 and 5 reflect the reaction over the active
sites of CeO2+2-cyanopyridine. Therefore, taking the strong
base property of CeO2+2-cyanopyridine into consideration,
methanol was easily dissociated to form methoxy species and
proton over the catalyst complex of CeO2 and 2-
cyanopyridine.
slope of the plots was positive (+4.6), meaning that a conju-
gated olefin with a more electron-withdrawing substituent
provided higher reaction rate. Therefore, the reaction will
proceed via an anionic intermediate. Taking into considera-
tion that methanol is adsorbed and easily dissociated to form
methoxy species and proton, nucleophilic addition of the
methoxy species on CeO2+2-cyanopyridine to acrylonitrile
will be the rate determining step, as shown in Scheme 1.
From the results of the kinetic and spectroscopic studies,
the reaction route over CeO2+2-cyanopyridine will be
basically similar to that over CeO2. Therefore, the large differ-
ence in the activity between CeO2 and CeO2+2-cyanopyridine
must be derived from the difference of the active sites over
both the catalysts. In order to know the information about the
transition state, effect of reaction temperatures was investi-
gated using both the catalysts (Figure 7), and the apparent
activation energy (HA) was estimated (Table 4). From the
slopes in Figure 7, the Ha over CeO2 and CeO2+2-
cyanopyridine was calculated to be 59 and 41 kJ/mol, respec-
tively. These results indicate that addition of 2-cyanopyridine
to CeO2 decreased the Ha, and hence, the transition state over
CeO2+2-cyanopyridine will be
9
Next, the effect of substituents in the conjugated olefins was
studied over CeO2+2-cyanopryidine, and the relationship be-
tween the reaction rate and Hammett parameters is shown in
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
50
40
30
20
10
0
6
4
0
0.5
1
1.5
Time / h
Figure 5. Isotope effect of methanol in hydromethoxylation of
acrylonitrile with CH3OH and CH3OD using CeO2 and 2-
cyanopyridine. Reaction conditions: acrylonitrile (10 mmol),
CH3OH() or CH3OD(ꢀ) (15 mmol), CeO2 (1 mmol), 2-
cyanopyridine (1 mmol), 323 K, air, 0.5 h. The detailed data
are shown in Table S9.
CeO2 + 2ꢀcyanopyridine
2
0
y = ꢀ4960x + 19 (R2 = 0.99)
CeO2
ꢀ2
ꢀ4
y = ꢀ7030x + 18 (R2 = 0.99)
CeO2 + 2ꢀCyanopyridine
O
CH3OH
+
R
R
323 K, air
1.6
ꢀ6
ꢀCN
1.2
0.8
0.4
0
0.0029
0.003
0.0031 0.0032 0.0033
1 / (Temperature / K)
ρ
= +4.6
Figure
7.
Effect
of reaction
temperatures
in
ꢀCOCH3
ꢀCOOCH3
hydroalkoxylation of acrylonitrile using CeO2 and CeO2+2-
cyanopyridine. Reaction conditions: acrylonitrile (10 mmol),
methanol (15 mmol), CeO2 (1 mmol)+2-cyanopyridine (1
mmol) or CeO2 (3 mmol), 313-333 K, air, 0.5-48 h. The de-
tailed data are shown in Table S11.
ꢀCONH2
0.4
ꢀ0.4
0.2
0.6
0.8
Table 4. Summary of activation energy (Ha) by experiments
and activation free energy (ꢀFa) by DFT-MD calculations.
σx
Figure 6. Relationship between reactivity and Hammett pa-
rameter in the hydromethoxylation of various conjugated
olefins using CeO2 and 2-cyanopyridine. Reaction conditions:
conjugated olefin (10 mmol), methanol (15 mmol), CeO2 (1
mmol), 2-cyanopyridine (4 mmol), 323 K, air, 0.5-12 h. The
detailed data are shown in Table S10.
Experimental
Calculation
Catalyst
Ha (kJ/mol)
ꢀFa (kJ/mol)
CeO2
59
41
45
65
42
45
CeO2+2-cyanopyridine
DBU
different from that over CeO2, which means that the active site
will be different. The adsorption state of 2-cyanopyridine over
CeO2 will be related to the active site over CeO2+2-
cyanopyridine, and therefore clarification of the adsorption
[CH3O-]
+
O
CN
CN
CeO2+2-cyanopyridine
Anion intermediate
Scheme 1. Formation of an anion intermediate by nucleo-
philic addition of methoxy species on CeO2 to acrylonitrile.
ACS Paragon Plus Environment

Page 7 of 12
Journal of the American Chemical Society
state is essential to understand the unique basic property of
the catalyst.
structures are shown in Figure S5. It is seen that no states
1
2
3
4
5
6
7
8
exist between the band gap of CeO2 for St-A and St-B, while
the St-C adsorption leads to the appearance of an additional
state located at 1.2 eV above the valence band, which is mainly
attributed to the non-bonding orbital of the N atom in the CN
group of 2-cyanopyridine. The new band observed in long-
wavelength region in the UV-vis spectra (Figure 3) could be
therefore assigned to the charge-transfer from 2-
cyanopyridine to CeO2, which supports that the formed
complex of CeO2 and 2-cyanopyridine is in the structure of St-
C.
DFT calculation of adsorption state of 2-cyanopyridine on
CeO2. The periodic-boundary DFT calculations were
performed to clarify the adsorption structure of 2-
cyanopyridine on CeO2 and the active site over CeO2+2-
cyanopyridine catalyst. The surface of CeO2(111) facet is
employed for this investigation since the CeO2(111) surface is
the most stable low-index surface among single crystal
terminations of CeO2 [17, 18], which is supported by the
results of the calcination temperature effect (Table S3 and
Figure S2). A periodic p(3×3) hexagonal slab of 27 CeO2 unit
with three O-Ce-O tri-layers is employed to model the
CeO2(111) surface (see below for computational details). For
2-cyanopyridine, three stable adsorption structures (St-A, St-B,
and St-C) were obtained by geometry optimization (Figure 8).
2-Cyanopyridine almost maintains the gas-phase structure at
both the structures of St-A and St-B (Figure 8(a),(b)). At the
structure of St-A, 2-cyanopyridine interacts with the surface
Ce atoms through the two N atoms in the pyridine ring and CN
group with a distance of ~3.0 Å, while only the N atom in the
CN group interacts with the surface Ce atom at St-B. The ad-
sorption energy (EA) of St-A is larger than that of St-B mainly
due to the higher basicity of the N atom in the pyridine ring
than that of the N atom in the CN group. In our previous re-
port, the proposed adsorption structure was similar to St-A
[20]. However, a thorough exploration of adsorption struc-
ture of 2-cyanopyridine on CeO2 leads to the observation of
unique adsorption structure of St-C (Figure 8(c)), and this
structure is found to be the most stable. The CN group is
strongly distorted, and a covalent bond between the C atom in
the CN group and the surface O atom of CeO2 (d(CCN-OS)=1.42
Å) is formed. By virtue of this unique adsorption structure, the
N atom in the CN group carries a negative charge and this site
is expected to work as a strong base site. In fact, the Mulliken
charge analysis is made for St-A and St-C, and we found that
the Mulliken charge of CN group in 2-cyanopyridine at St-C is
more negative than that at St-A by 0.24e−. Analysis of the elec-
tronic structure is also carried out by the projected density of
states (PDOS) and the results for the three adsorption
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Next, adsorption structures of a methanol molecule on CeO2
(111) surface were examined. Optimized adsorption struc-
tures for molecular and dissociative states are shown in Fig-
ure S6. The adsorption energies (EA) are calculated to be
−0.59 and −0.68 eV for molecular and dissociative states, re-
spectively, and the dissociative adsorption is favorable on the
CeO2 surface. In the dissociative adsorption state, methoxy
moiety and a proton are adsorbed on Ce and O atoms, respec-
tively, and a proton is bound to nearby bridging surface O
atom forming a hydrogen bond with the methoxy moiety. This
is in agreement with the experimental observations (Figure
S4), where a methanol molecule is dissociated at the acid-base
sites on CeO2 [19]. The calculated adsorption energies are also
in good agreement with the previous works [21]. When the
proton is diffused further from the methoxy moiety, the ad-
sorption energy becomes −0.29 eV, resulting in less stable
structure (Figure S6). It is seen that the oxidation state of the
methoxy species and Ce atoms does not change on dissocia-
tion, and therefore the methoxy species is adsorbed over Ce as
an anion species.
Several co-adsorption structures of 2-cyanopyridine and
methanol on CeO2(111) were also examined. Co-adsorption
structures where the 2-cyanopyridine is adsorbed with the
structure of St-A or St-B were first investigated. In these cas-
es, regardless of whether the methanol is adsorbed by the
molecular or dissociative state, we could not find more stable
co-adsorption structures than the independent adsorption
state. Here, the energy of the independent adsorption state is
defined as sum of the adsorption energy of 2-cyanopyridine
(St-A or St-B) and methanol in the dissociative state. On the
other hand, although we could not find more stable structures
using the adsorption structure of St-C when the methoxy
(a)
(b)
(c)
d(NCNꢀCe) = 2.90 Å
d(NpyꢀCe) = 3.03 Å
d(NCNꢀCe) = 2.36 Å
d(CCNꢀOS) = 1.42 Å
d(NpyꢀCe) = 2.69 Å
d(NCNꢀCe) = 2.71 Å
(a)
(b)
StꢀA
StꢀC
StꢀB
EA = −1.16 eV
EA = −1.74 eV
EA = −0.34 eV
EA = −0.56 eV
EA = −0.49 eV
Figure 9. Co-adsorption structures of 2-cyanopyridine and
methanol on CeO2(111) surface. The atoms are color coded as
follows: gray, H; dark gray, C; blue, N; red O; dark yellow, Ce.
Figure 8. Adsorption structures of 2-cyanopyridine on the
CeO2(111) surface. The atoms are color coded as follows: gray,
H; dark gray, C; blue, N; red O; dark yellow, Ce.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 12
moiety and proton forms a hydrogen bond (Figure 9(a)),
entry 5). Log(V (reaction rate)) is plotted as a function of the
1
2
3
4
5
6
7
8
proton transfer of methanol moiety to the N atom of CN group
in 2-cyanopyridine makes the energy of this co-adsorption
structure (−1.74 eV, Figure 9(b)) lower than that of the
independent adsorption state (−1.24 eV, 2-cyanoypridine (St-
C): −0.56 eV, methanol (dissociative state): −0.68 eV). This
result indicates that the basicity of the N atom is stronger than
that of the surface O atom. A slightly higher energy at the
structure of Figure 9(a) (−1.16 eV) than that of independent
adsorption (−1.24 eV) is due to the steric interaction between
2-cyanopyridine and methanol fragments. From these results,
2-cyanopyridine is adsorbed on CeO2, and the covalent inter-
action between the surface O atom of CeO2 and C atom of the
CN group in 2-cyanopyridine was formed as the structure of
St-C. This adsorption constructed the strong base site of N- in
CN group of 2-cyanopyridine over CeO2.
energy difference, a basicity index (Figure 10). A good
correlation between the energy difference and Log(V) is
observed, supporting that the nitrogen anion site formed by
adsorption of the nitriles on CeO2 as the St-C acts as a base site
for the reaction.
DFT-based
molecular
dynamics
simulations
of
hydromethoxylation of acrylonitrile in methanol solution.
Reaction mechanism for hydromethoxylation of acrylonitrile
by the CeO2+2-cyanopyridine catalyst is studied by DFT-based
molecular dynamics (MD) simulations in explicit methanol
solvents. From the above kinetic and spectroscopic studies,
nucleophilic addition of methoxy species to acrylonitrile is the
rate-determining step, and therefore we investigate this pro-
cess based on the free energy profile employing the blue
moon ensemble approach (see below for computational de-
tails). Figure 11 shows the free energy profiles as a function of
the distance between O atom of the attacking methanol moie-
ty and C atom of acrylonitrile using CeO2+2-cyanopyridine
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In order to further rationalize the fact that the unique ad-
sorption structure of St-C is responsible for creation of the
new base site at the CN position of 2-cyanopyridine, adsorp-
tion structures of representative nitrile additives given in
Table
cyanopyridine, 4-cyanopyridine, 2-pyridineacetonitrile, 2-
cyanopyrazine, 1-isoquinolinecarbonitrile and 2-
3 were investigated such as 2-cyanopyridine, 3-
70
quinolinecarbonitrile (Table S12 and Figure S7). Table S11
summarizes the adsorption energies of the nitrile additives,
where the covalent bond between the C atom in the CN group
and surface O atom is formed as the structure of St-C. The
adsorption energy and the reactivity shows a good correlation,
where nitriles with high reaction rates have large adsorption
energy of -0.4~-0.6 eV, in contrast, nitriles with low reaction
rates exhibit very week adsorption with the adsorption ener-
gy of ~0 eV. These results indicate that the unique adsorption
similar to St-C is a key structure for the formation of the
strong base site.
CeO + 2ꢀcyanopyridine
CeO₂2+2ꢀPy
60
50
40
30
20
10
0
CeO₂
CeO2
DDBBUU
The effects of the substituent in 2-cyanopyridine deriva-
tives shown in Figure 2 and Table 3 are also examined theo-
retically. The energy difference between the two structures
given in Figure 9 can be used as a simple measure for proton
affinity, namely a base index. In the case of 2-cyanopyridine,
the energy difference is calculated to be −0.58 eV, and more
negative value indicates a higher affinity of a proton (a
stronger base). The calculated values of the energy difference
for the p-(5-) or m-(4-)substituted 2-cyanopyridine are −0.60
eV (p-CH3, Table 3, entry 2), −0.55 eV (p-F, Table 3, entry 3),
−0.56 eV (p-Cl, Table 3, entry 4), and −0.52 eV (m-NO2, Table 3,
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Distance OꢀC / Å
Figure 11. Free energy profiles as a function of the distance
between O atom of the attacking methanol moiety and C atom
of acrylonitrile.
(a)
(b)
2
pꢀCH3
pꢀH
pꢀF
1
pꢀCl
mꢀNO2
0
Figure 12. Snapshot of the MD simulation during the nucleo-
philic attack of methoxy moiety for (a) CeO2+2-cyanopyridine
and (b) only CeO2 catalyst. Blue shaded circle represents the
base site and yellow shaded circle shows the attacking meth-
oxy moiety. The atoms are color coded as follows: gray, H;
dark gray, C; blue, N; red O; dark yellow, Ce.
ꢀ0.65
ꢀ0.6
ꢀ0.55
ꢀ0.5
ꢀ0.45
Energy difference / eV
Figure 10. Relationship between the energy difference and
reaction rate activity in the hydromethoxylation of acryloni-
trile using CeO2 and p- or m-substituted 2-cyanopyridines. The
reaction rates are quoted from Table 3.
ACS Paragon Plus Environment

Page 9 of 12
Journal of the American Chemical Society
catalyst and only CeO2 catalyst. Here, we focus on the activa-
Considering the results of methanol adsorption by FTIR and
1
2
3
4
5
6
7
8
tion barrier of the nucleophilic addition of methoxy species to
acrylonitrile. As shown in Figure 11 and Table 4, the calculat-
ed activation barrier (ꢀFa: 42 kJ/mol for CeO2+2-
cyanopyridine and 65 kJ/mol for only CeO2 catalyst) is in quite
good agreement with the experimental results. Figure 12
shows the snapshot of the MD simulation when the O-C dis-
tance is fixed at 2.0 Å, which is near the transition structure.
For CeO2 catalyst (Figure 12(b)), the surface O atom acts as a
base site to accept a proton, and the methoxy anion attacks
the C atom of CN group in acrylonitrile. On the other hand, the
N atom in the CN group of 2-cyanopyridine acts as a strong
base site for the CeO2+2-cyanopyridine catalysts (Figure
12(a)). It is intriguing to note that the attacking methoxy spe-
cies is hydrogen-bonded with another methanol molecule to
stabilize the transition state structure of the anionic species.
For comparison, we carried out MD simulations for the same
reaction using a DBU as a typical strong base catalyst (Table 4
and Figure S8). In this case, the activation barrier is calculated
to be 45 kJ/mol, which is similar to that by the experimental
results (Table 4). In addition, the value is comparable to that
of CeO2+2-cyanopyridine catalyst. This indicates that the
CeO2+2-cyanopyridine catalyst exhibits similar functionality
to DBU as a strong base.
low reaction order with respect to the methanol concentration
over CeO2+2-cyanopyridine, methanol is adequately supplied
to the active site from methanol adspecies on CeO2 surface,
and activated methoxy species can be easily reformed on the
active sites over CeO2.
■
Conclusion
We demonstrated a new base site composed of CeO2 and 2-
9
cyanopyridine, and the basicity based on the surface 2-
cyanopyridine (pKa=21) is in the category of superbases. The
base catalyst was effective for hydroalkoxylation of
conjugated various olefins with various alcohols, providing
the corresponding ethers in high yields. Based on the kinetic,
spectroscopic, and DFT studies, N- on CN group in 2-
cyanopyridine is responsible for the strong base, which is
formed by adsorption of 2-cyanopyridine on CeO2 via nucleo-
philic addition of the surface O atom on CeO2 to the CN group
in 2-cyanopyridine. The reaction proceeds via nucleophilic
addition of the methoxy species activated by the strong base
sites to the olefin group in the conjugated olefins, and this step
is the rate-determining step. This method is the first example
for fabricating a superbase catalyst by hybridization of heter-
ogeneous and homogeneous components.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Proposed
cyanopyridine. As above, the reaction mechanism of
hydromethoxylation of acrylonitrile over CeO2+2-
reaction
mechanism
over
CeO2+2-
■
EXPERIMENTAL SECTION
Typical procedure for hydromethoxylation of acryloni-
trile. A typical procedure for nucleophilic addition of metha-
nol to acrylonitrile is described in the previous report [13] as
follows: CeO2 (Daiichi Kigenso Kogyo, calcined at 873 K for 3
h) 172 mg (1 mmol) and methanol 0.64 g (20 mmol) were
added to a reaction vessel, and the reaction vessel was capped
with a rubber plug under air. Then, the mixture was vigorous-
ly stirred at 500 r.p.m. at 323 K under air for 1 h. After the
treatment, acrylonitrile 0.53 g (10 mmol) was added into the
mixture with a syringe, and the time was defined as zero reac-
tion time. The mixture was constantly stirred during the reac-
tion. After the reaction, the reaction mixture was filtrated,
diluted with acetone, and transferred to a vial. Details of the
reaction conditions are described in each result. The products
were analyzed by gas chromatography (GC) equipped with a
FID detector and CPSil5 capillary column (length 50 m, i.d.
0.25 mm, film thickness 0.25 ꢀm). Conversion and yield of
products were determined on acrylonitrile basis by GC using
1,4-dioxane as an internal standard.
cyanopyridine is proposed (Scheme 2). At first, the strong
base site of N- is formed by adsorption of 2-cyanopyridine on
CeO2 via addition of the surface O atom on CeO2 to the CN
group in 2-cyanopyridine. Second, methanol is activated by
the strong base site to produce active methoxy species, and
methanol is also adsorbed on CeO2 surface to form methoxy
species and proton. Third, nucleophilic attack of the methoxy
species to acrylonitrile, forming an anion intermediate over
the catalyst. Finally, the target ether is produced by protona-
tion of the anion intermediate. The nucleophilic attack of the
methoxy species to acrylonitrile is the rate-determining step.
N
N
O
O
NC
N
Ce
Ce
Ce
O
O
Ce
Ce
Ce
CH3OH
CH3
H
O
DFT calculations. Geometry optimization was carried out by
the periodic-boundary DFT with the mixed Gaussian and
plane-waves (GPW) approach implemented in the CP2K pro-
gram package [22]. The double-ζ valence plus polarization
(DZVP) basis sets of the MOLOPT type [23] were employed for
H, C, N, O atoms to represent the valence electrons, and the
norm-conserving Goedecker-Teter-Hutter pseudopotentials
[24,25] were used to describe the interactions between the
valence and core electrons. For Ce atoms, we employed the
basis sets and pseudopotential generated by Wang and
coworkers [26]. The energy cutoff of 400 Ry was taken for the
auxiliary plane wave expansion of the density. The
CH3
O
N
N
H
O
O
O
O
Ce
Ce
Ce
Ce
NC
NC
CH3OH
NC
O
CH3
H
NC
CH3
H
O
CH3
O
CH3
N
N
N
N
H
O
H
O
O
O
O
O
O
Ce
Ce
Ce
Ce
Ce
Ce
Ce
Ce
generalized-gradient
approximation
by
the
Perdew−Burke−Ernzerhof (PBE) functional [27] models was
employed as the exchange and correlation potential, and the
DFT+U approach [28,29] was taken in order to correctly
represent the nature of 4f orbitals of Ce atoms, where the
occupancies of 4f orbitals were calculated on the basis of
Scheme
2.
Proposed
reaction
mechanism
of
hydromethoxylation
of
acrylonitrile
over CeO2+2-
cyanopyridine catalyst
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 12
Mulliken population analysis [30,31]. The U value was set to
those of the above CeO2 or CeO2+2-cyanopyridine catalyst
system.
1
7.0 eV following the previous works [17b, 26]. It is worth
noting that the larger value of U = 7.0 eV relative to ones em-
ployed in other theoretical works (4~5 eV) [32,33] is due to
the different implementation of DFT+U method. In fact, Wang
et al. carefully determined this value in the CP2K code to best
reproduce the band gap, gap state location, and the work func-
tion for reduced CeO2(111) surface [26]. The Brillouin zone
integration was performed with a reciprocal space mesh con-
sisting of only the Γ-point in the GPW approach. For geometry
optimization, the forces on all atoms are minimized to less
than 0.02 eV/Å (4.5×10−4 hartree/bohr). The surface of
CeO2(111) is modeled as a periodic p(3×3) hexagonal slab of
27 CeO2 unit with three O-Ce-O tri-layers. The dimensions of a
simulation cell are set to a = b = 11.56, c = 25.0 Å and α = β =
90, γ = 60 degrees, and this slab is separated by ~16 Å of vac-
uum space in the direction perpendicular to the surface. The
lattice parameters of the CeO2(111) slab are determined by
the cell optimization of the bulk CeO2, and the optimized lat-
tice constant of 5.450 Å is in good agreement with the experi-
mental value of 5.411 Å [34,35]. The bottom O-Ce-O tri-layers
are fixed at the bulk positions during the geometry optimiza-
tion and also in molecular dynamics simulations. In calculat-
ing the energies of gas-phase molecules, the simulation cell of
a = b = c = 15.0 Å in the cubic box is used.
2
3
4
5
6
7
8
9
■ ASSOCIATED CONTENT
Supporting Information.
The supporting information is available free of charge on the
ACS Publications web sites at DOI: Details of DFT calculations
in addition to experimental details for materials
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
■ AUTHOR INFORMATION
Corresponding Author
*mtamura@erec.che.tohoku.ac.jp
*nakayama@cat.hokudai.ac.jp
*tomi@erec.che.tohoku.ac.jp
■ ACKNOWLEDGMENT
This work was supported by JST PRESTO Grant Number
JPMJPR15S5 and JPMJPR16S8, Japan. Part of the calculations
was performed on supercomputers at RCCS (Okazaki, Japan),
RIIT (Kyushu University), and ACCMS (Kyoto University).
The adsorption energy of a molecule EA to the CeO2(111)
surface is calculated according to
(0.1)
EA = Emol+surf − Emol − Esurf
■ REFERENCES
where Emol+surf is the total electronic energy of surface-
molecular system while Esurf and Emol are the energies of pris-
tine surface and an isolated molecule, respectively. In this
definition, the more negative value of adsorption energy indi-
cates a stronger binding to the surface.
(1) (a) The Supramolecular Chemistry of Organic-Inorganic Hybrid
Materials; Rurack, K.; Martínez-Máñez, R. Ed.; John Wiley & Sons:
Hoboken, NJ. 2010. (b) Díaz, U.; Brunel, D.; Corma, A. Chem. Soc.
Rev. 2013, 42, 4083-4097. (c)Zhu, K.; Wang, D.; Liu, J. Nano Res.
2009, 2, 1-29. (d) Schätz, A.; Reiser, O.; Stark, W. J. Chem. Eur. J.
2010, 16, 8950-8967.
(2) (a) Margelefsky, E. L.; Zeidan, R. K.; Davis, M. E. Chem. Soc. Rev.
2008, 37, 1118-1126. (b) Tada, M.; Iwasawa, Y. Chem. Commun.
2006, 2833-2844. (c) Hoffmann, F.; Cornelius, M.; Morell, J.; Frö-
ba, M. Angew. Chem. Int. Ed. 2006, 45, 3216-3251. (d) Melero, J. A.;
van Grieken, R.; Morales, G. Chem. Rev. 2006, 106, 3790-3812. (e)
Notestein, J. M.; Katz, A. 2006, 12, 3954-3965.
(3) (a) Christopher Love, J.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R.
G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1103-1169. (b)
Schoenbaum, C. A.; Schwartz, D. K.; Medlin, J. W. Acc. Chem. Res.
2014, 47, 1438-1445. (c) Ulman, A. Chem. Rev. 1996, 96, 1533-
1554.
(4) (a) Pachón, L. D.; Rothenberg, G. Appl. Organomet. Chem. 2008,
22, 288-299. (b) Jia, C.-J.; Schüth, F. Phys. Chem. Chem. Phys. 2011,
13, 2457-2487. (c) Toshima, N.; Yonezawa, T. New J. Chem. 1998,
1179-1201.
(5) (a) Scott, R. W. J.; Wilson, O. M.; Crooks, R. M. J. Phys. Chem. B
2005, 109, 692-704. (b) Wang, D.; Astruc, D. Coord. Chem. Rev.
2013, 257, 2317-2334. (c) Crooks, R. M.; Zhao, M.; Sun, L.; Chechik,
V.; Yeung, L. K. Acc. Chem. Res. 2001, 34, 181-190. (d) Peng, X.; Pan,
Q.; Rempel, G. L. Chem. Soc. Rev. 2008, 37, 1619-1628. (e) Astruc,
D.; Chardac, F. Chem. Rev. 2001, 101, 2991-3023. (f) van Heerbeek,
R.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.; Reek, J. N. H. Chem.
Rev. 2002, 101, 3717-3756.
(6) (a) Kubota, Y.; Nishizaki, Y.; Ikeya, H.; Saeki, M.; Hida, T.; Ka-
wazu, S.; Yoshida, M.; Fujii, H.; Sugi, Y. Micropor. Mesopor. Mater.
2004, 70, 135-149. (b) Hould, N. D.; Foster, A.; Lobo, R. F. Mi-
cropor. Mesopor. Mater. 2011, 142, 104-115. (c) Hould, N. D.;
Senapati, S.; Koller, H.; Lobo R. F. Top. Catal. 2012, 55, 1332-1343.
(7) (a) Burgi, T.; Baiker, A. Acc. Chem. Res. 2004, 37, 909-917.
(b)Mallat, T. Orlmeister, E. Baiker, A. Chem. Rev. 2007, 107, 4863-
4890.
Molecular dynamics simulations. In the molecular dynamics
simulations in explicit methanol solvents, the same simulation
size is employed and methanol molecules are placed between
the slabs of CeO2. The number of methanol molecules is
determined by considering available space for methanol and
the bulk density at room temperature, and 21 and 26 metha-
nol molecules are inserted in the simulation cell for CeO2+2-
cyanopyridine and pure CeO2 system, respectively. Molecular
dynamics simulations are performed with the canonical en-
semble (NVT) condition, where the temperature is controlled
at 330 K by the Nóse-Hoover thermostat. The mass of a hy-
drogen is replaced with that of deuterium allowing for a larger
time step of 1.0 fs. The free energy barrier for the
hydromethoxylation of acrylonitrile is estimated by the blue
moon ensemble method [36]. In this procedure, the reaction
coordinate is defined as a distance between the C atom of ac-
rylonitrile and an O atom of attacking methanol moiety. After
equilibrating the system, a total of 20 ps simulations are
performed for each reaction coordinate. In Figure S9, the
mean force acting on the O-C distance is shown for CeO2+2-
cyanopyridine, CeO2, and DBU catalyst, as a function of fixed
O-C distance. Integration of the mean force leads to the free
energy profile. In this work, we focus on the activation barrier
of the nucleophilic addition of methoxy species to acryloni-
trile, and therefore only the positive area is examined. For
molecular dynamics simulations of hydromethoxylation of
acrylonitrile using DBU as a base catalyst, the simulation cell
size is set to a = b = c = 14.0 Å in the cubic box and 34 metha-
nol, one acrylonitrile, and one DBU molecule are placed in a
simulation box. The other simulation parameters are equal to
ACS Paragon Plus Environment

Page 11 of 12
Journal of the American Chemical Society
(8) (a) Takahashi, T.; Watahiki, T.; Kitazume, S.; Yasuda, H.; Sa-
(16) (a) Honda, M.; Tamura, M.; Nakagawa, Y.; Nakao, K.; Suzuki,
K.; Tomishige, K. J. Catal. 2014, 318, 95-107. (b) Honda, M.; Tamu-
ra, M.; Nakagawa, Y.; Sonehara, S.; Suzuki, K.; Fujimoto, K.-i.;
Tomishige, K. ChemSusChem 2013, 6, 1341-1344.
(17) (a) Paier, J.; Penschke, C.; Sauer, J. Chem. Rev. 2013, 113,
3949-3985. (b) Camellone, M. F.; Ribeiro, F. N.; Szabová, L.;
Tateyama, Y.; Fabris, S. J. Am. Chem. Soc. 2016, 138, 11560-11567.
(18) Vivier, L.; Duprez, D. ChemSusChem 2010, 3, 654-678.
(19) (a) Lamotte, J.; Morávek, V.; Bensitel, M.; Lavalley, J. C. React.
Kint. Catal. Lett. 1988, 36, 113-118. (b) Badri, A.; Binet, C.; La-
valley, J. C. J. Chem. Soc. Faraday Trans. 1997, 93, 1159-1168.
(20) Tamura, M.; Sawabe, K.; Tomishige, K.; Satsuma, A.; Shimizu,
K.-i. ACS Catal. 2015, 5, 20-26.
(21) (a) Beste, A.; Mullins, D. R.; Overbury, S. H.; Harrison, R. J.
Surf. Sci. 2008, 602, 162. (b) Mei, D.; Deskins, N. A.; Dupuis, M.; Ge,
Q. J. Phys. Chem. C 2007, 111, 10514–10522. (c) Kropp, T.; Paier, J.
J. Phys. Chem. C 2014, 118, 23690–23700. (d) Capdevila-Cortada,
M.; García-Melchor, M.; López, N. J. Catal. 2015, 327, 58–64. (e)
Yang, C.; Bebensee, F.; Nefedov, A.; Wöll, C.; Kropp, T.; Komissarov,
L.; Penschke, C.; Moerer, R.; Paier, J.; Sauer, J. J. Catal. 2016, 336,
116–125. (f) Beste, A.; Overbury, S. H. Phys. Chem. Chem. Phys.
2016, 18, 9990–9998.
(22) J. VandeVondele, M. Krack, F. Mohamed, M. Parrinello, T.
Chassaing and J. Hutter, Comput. Phys. Commun. 2005, 167, 103-
128.
(23) VandeVondele, J.; Hutter, J. J. Chem. Phys. 2007, 127, 114105.
(24) Goedecker, S.; Teter, M.; Hutter, J. Phys. Rev. B 1996, 54,
1703-1710.
(25) Hartwigsen, C.; Goedecker, S.; Hutter, J. Phys. Rev. B 1998, 58,
3641-3662.
(26) Wang, Y.-G.; Mei, D.; Li, J.; Rousseau, R. J. Phys. Chem. C 2013,
117, 23082-23089.
(27) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996,
77, 3865-3868.
(28) Herbst, J. F.; Watson, R. E.; Wilkins, J. W. Phys. Rev. B 1978,
17, 3089-3098.
(29) Dudarev, S. L.; Botton, G. A.; Savrasov, S. Y.; Humphreys C. J.;
Sutton, A. P. Phys. Rev. B 1998, 57, 1505-1509.
(30) Esch, F.; Fabris, S.; Zhou, L.; Montini, T.; Africh, C.; Fornasiero,
P.; Comelli, G.; Rosei, R. Science 2005, 309, 752-755.
(31) Paier, J.; Penschke, C.; Sauer, J. Chem. Rev. 2013, 113, 3949-
3985.
(32) Nolan, M.; Parker, S. C.; Watson, G. W. Surf. Sci. 2005, 595,
223-232.
(33) Fabris, S.; Vicario, G.; Balducci, G.; de Gironcoli, S.; Baroni, S. J.
Phys. Chem. B 2005, 109, 22860-22867.
(34) Gerward, L.; Olsen, J. S.; Powder Diffr. 1993, 8, 127-129.
(35) Nakajima, A.; Yoshihara, A.; Ishigame, M. Phys. Rev. B 1994,
50, 13297-13307.
kakura, T. Chem. Commun. 2006, 1664-1666. (b) Motokura, K.;
Tada, M.; Iwasawa, Y. J. Am. Chem. Soc. 2007, 129, 9540-9541. (c)
Shylesh, S.; Wagener, A.; Seifert, A.; Ernst, S.; Thiel, W. R. Chem-
CatChem 2010, 2, 1231-1234.
(9) (a) Solid Acid Catalysis: From Fundamentals to Applications;
Hattori, H.; Ono, Y. Ed.; CRC Press, Stanford: 2015. (b) Solid Base
Catalysis; Ono, Y.; Hattori, H. Springer, Heidelberg: 2011. (c) Hat-
tori, H. Appl. Catal. A 2015, 504, 103-109. (d) Hattori, H. Chem.
Rev. 1995, 95, 537-558. (e) Metal Oxide Catalysis; Jackson, D.;
Hargreaves, J. S. J. Ed.; John Wiley & Sons: New York, 2009.
(10) Superbases for Organic Synthesis: Guanidines, Amidines, Phos-
phazenes and Related Organocatalysts; Ishikawa, T., Ed.; John
Wiley and Sons: New York, 2009.
(11) (a) Palomo, C.; Oiarbide, M.; López, R. Chem. Soc. Rev. 2009,
38, 632-653. (b) Fu, X.; Tan, C.-H. Chem. Commun. 2011, 47, 8210-
8222. (c) Pozharskii, A. F.; Ozeryanskii, V. A.; Filatova, E. A. Chem-
istry of Heterocyclic Compounds, 2012, 48, 200-219. (d) Verkade, J.
G.; Kisanga, P. B. Aldrichimica Acta, 2004, 37, 3-14. (e) Imahori, T.
Kondo, Y. J. Am. Chem. Soc. 2003, 125, 8082-8083.
(12) (a) Takeda, T.; Terada, M. J. Am. Chem. Soc. 2013, 135, 15306-
15309. (b) Takeda, T.; Kondoh, A.; Terada, M. Angew. Chem. Int.
Ed. 2015, 54, 15836-15839. (c) Nacsa, E. D.; Lambert, T. H. J. Am.
Chem. Soc. 2015, 137, 10246-10253. (d) Bandar J. S.; Lambert, T.
H. J. Am. Chem. Soc. 2012, 134, 5552-5555. (e) Kögel, J. F.; Oelkers,
B.; Kovačević, B.; Sundermeyer, J. J. Am. Chem. Soc. 2013, 135,
17768-17774. (f) Coles, M. P.; Aragón-Sáez, P. J.; Oakley, S. H.;
Hitchcock, P. B.; Davidson, M. G.; Maksić, Z. B.; Vianello, R.; Leito, I.;
Kaljurand, I.; Apperley, D. C. J. Am. Chem. Soc. 2009, 131, 16858-
16868. (g) Kolomeitsev, A. A.; Koppel, I. A.; Rodima, T.; Barten, J.;
Lork, E.; Röschenthaler, G. V.; Kaljurand, I.; Kütt, A.; Koppel, I.;
Mäemets, V.; Leito, I. J. Am. Chem. Soc. 2005, 127, 17656-17666.
(h) Kögel, J. F.; Xie, X.; Baal, E.; Gesevičius, D.; Oelkers, B.; Ko-
vačević, B.; Sundermeyer, J. Chem. Eur. J. 2014, 20, 7670-7685. (i)
Raab, V.; Gauchenova, E.; Merkoulov, A.; Harms, K.; Sundermeyer,
J.; Kovacević, B.; Maksić, Z. B. J. Am. Chem. Soc. 2005, 127, 15738-
15743. (j) Estrada, E.; Simón-Manso, Y. Angew. Chem. Int. Ed.
2006, 45, 1719-1721. (k) Ducháccková, L.; Kadlccíková, A.; Ko-
tora, M.; Roithová, J. J. Am. Chem. Soc. 2010, 132, 12660-12667. (l)
Raytchev, P. D.; Martinez, A.; Gornitzka, H.; Dutasta, J. P. J. Am.
Chem. Soc. 2011, 133, 2157-2159.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(13) Tamura, M.; Kishi, R.; Nakagawa, Y.; Tomishige, K. Nat. Com-
mun. 2015, 6, 8580.
(14) Kaljurand, I.; Kütt, A.; Sooväli, L.; Rodima, T.; Mäemets, V.;
Leito, I.; Koppel, I. A. J. Org. Chem. 2005, 70, 1019-1028.
(15) (a) Alder, R. W.; Bowman, P. S.; Steele, W. R. S.; Winterman, D.
R.; Chem. Commun. 1968, 723-724. (b) Alder, R. Chem. Rev. 1989,
89, 1215-1223. (c) Staab, H. A.; Saupe, T. Angew. Chem. Int. Ed.
1988, 27, 865-879. (d) Llamas-Saiz, A. L.; Faces-Foces, C.; Elguero,
J. J. Mol. Struct. 1994, 328, 297-323. (e) Chambron, J.-C.; Meyer, M.
Chem. Soc. Rev. 2009, 38, 1663-1673.
(36) Sprik, M.; Ciccotti, G. J. Chem. Phys. 1998, 109, 7737-7744.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 12
1
2
3
4
5
2-Cyanopyridine
CH3OH
Acrylonitrile
6
7
8
9
Cerium oxide
Strong Basic
Nitrogen Anion
(CeO )
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Keywords: Cerium oxide, 2ꢀcyanopyridine, base catalyst, DFT calculation, hybrid catalyst
12
ACS Paragon Plus Environment