CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines

Article abstract of DOI:10.1016/j.jcat.2015.11.015

CeO₂ showed higher activity for the direct synthesis of 1,3-dibutylurea (DBU) from CO₂ and n-butylamine than the metal oxides tested. The solvent largely influenced the reaction over CeO₂, and N-methylpyrrolidone (NMP) was preferable among various solvents tested from the viewpoints of activity and selectivity. The catalyst system composed of CeO₂ catalyst and NMP solvent (CeO₂ in NMP) was applicable to the reactions of various amines such as linear primary alkylamines or branched primary alkylamines, although tert-butylamine afforded low conversion. In contrast, secondary amines and aniline provided no yield of the ureas. The combination of 2-cyanopyridine with CeO₂ in NMP (CeO₂ in NMP with 2-cyanopyridine) promoted the transformation of the unreactive amines, showing that tert-butylamine and aniline were converted to the corresponding ureas in 82% and 80% yields, respectively. These yields are much higher than those reported in the previous literatures, indicating that CeO₂ in NMP with 2-cyanopyridine drastically promoted transformation of amines with low reactivity.

Full text of DOI:10.1016/j.jcat.2015.11.015

Journal of Catalysis xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines
⇑
⇑
Masazumi Tamura , Kazuki Ito, Yoshinao Nakagawa, Keiichi Tomishige
Graduate School of Engineering, Tohoku University, Aoba 6-6-07, Aramaki, Aoba-ku, Sendai 980-8579, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
CeO2 showed higher activity for the direct synthesis of 1,3-dibutylurea (DBU) from CO2 and n-butylamine
than the metal oxides tested. The solvent largely influenced the reaction over CeO2, and
N-methylpyrrolidone (NMP) was preferable among various solvents tested from the viewpoints of
activity and selectivity. The catalyst system composed of CeO2 catalyst and NMP solvent (CeO2 in
NMP) was applicable to the reactions of various amines such as linear primary alkylamines or branched
primary alkylamines, although tert-butylamine afforded low conversion. In contrast, secondary amines
and aniline provided no yield of the ureas. The combination of 2-cyanopyridine with CeO2 in NMP
(CeO2 in NMP with 2-cyanopyridine) promoted the transformation of the unreactive amines, showing
that tert-butylamine and aniline were converted to the corresponding ureas in 82% and 80% yields,
respectively. These yields are much higher than those reported in the previous literatures, indicating that
CeO2 in NMP with 2-cyanopyridine drastically promoted transformation of amines with low reactivity.
Ó 2015 Elsevier Inc. All rights reserved.
Received 1 October 2015
Revised 13 November 2015
Accepted 19 November 2015
Available online xxxx
Keywords:
Urea
Carbon dioxide
Cerium oxide
N-Methylpyrrolidone
Heterogeneous catalyst
1. Introduction
because only water is produced. Homogeneous catalysts and cata-
lyst systems were reported such as CsOH, Cs2CO3, [BMIm]OH, Co
CO2 is an abundant, non-toxic, non-flammable and inexpensive
C1 source [1–3]. Many methods for direct conversion of CO2 to
chemicals have been developed [4–16]. Compared with reductive
transformation of CO2 to chemicals such as methanol and formic
acid [4–13], non-reductive transformation to chemicals such as
carbonates, carbamates and ureas [3,14–16] is preferable from
the viewpoint of energy input, while the direct transformation of
CO2 is generally difficult because CO2 is very stable due to the very
strong double bonds [3]. Urea derivatives are an important class of
carbonyl compounds and can be used for many applications as
intermediates for the synthesis of pharmaceuticals or agrochemi-
cals, and also used as additives for fuels [17].
Urea derivatives have been traditionally prepared with phos-
gene or its derivatives [17]. However, the phosgene process has
considerable drawbacks such as toxicity of phosgene and co-
production of large amount of salts by the neutralization. Other
various methods for the synthesis of urea derivatives have been
also studied, such as oxidative carbonylation of amines with CO,
reductive carbonylation of amines with CO, transamidation of
ureas with amines, reaction of carbonates with amines, and reac-
tion of CO2 and amines [17–19]. Among these methods, direct reac-
tion of CO2 with amine to urea derivatives will be promising
(acac)3 and TBA2[WO4] [20–24]. However, heterogeneous catalysts
or non-catalyst systems are more desirable than homogeneous
ones from the industrial, environmental, and economical
viewpoints. Deng and co-workers reported that polymer supported
Au nanoparticles (Au/Poly), which showed the high activity
(TOF = 2900 hꢀ1
) using cyclohexylamine among the reported
heterogeneous catalysts, although the catalyst has drawbacks such
as use of the precious metal of Au [25]. KOH/PEG catalyst and
Mg–Al hydrotalcite catalyst were also reported [26,27], and these
catalysts however have some problems such as low selectivity,
cumbersomeness of catalyst preparation and/or narrow scope of
substrate. Zhao and co-workers reported non-catalyst and
non-solvent system; however, low activity and narrow scope of
substrates were problems to be solved [28].
Recently, CeO2-catalyzed organic reactions have been vigor-
ously studied [29,30] such as transformation of nitriles [31–33],
transformation of CO2 to carbonate derivatives [34–57], reduction
of alkynes and carboxylic acids [58,59], and transamidation and
transesterification of amides or esters [60–69]. In our laboratory,
it was firstly reported that CeO2 showed activity for the activation
of CO2, and applied CeO2 catalyst to transformation of CO2 to
organic carbonates such as dialkyl carbonates and cyclic carbon-
ates, and carbamates such as methyl benzyl carbamate and cyclic
carbamates [15,16,36–49]. It was also reported that CeO2 is effi-
cient for the direct synthesis of cyclic ureas from CO2 and diamines
[37]. In this report, CeO2 catalyst was applied to the synthesis of
⇑
Corresponding authors. Fax: +81 22 795 7214.
E-mail addresses: mtamura@erec.che.tohoku.ac.jp (M. Tamura), tomi@erec.che.
tohoku.ac.jp (K. Tomishige).
http://dx.doi.org/10.1016/j.jcat.2015.11.015
0021-9517/Ó 2015 Elsevier Inc. All rights reserved.
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015

2
M. Tamura et al. / Journal of Catalysis xxx (2015) xxx–xxx
1,3-disubstituted ureas from CO2 and amines, and CeO2 showed
higher activity than the metal oxides tested. NMP is a preferable
solvent for the reaction and enhances the activity of CeO2.
(10 ml) was added to the liquid phase as a solvent, and 1-hexanol
(ꢂ0.2 ml) was also added as an internal standard substance for a
quantitative analysis. The reactor was washed with methanol, and
the liquids used in washing were added to the reaction mixture.
The products were analyzed by a gas chromatograph equipped with
an FID using a CP/Sil 5 CB capillary column (Agilent Technologies
2. Experimental
Inc., length = 50 m, I.D. = 0.25 mm, df = 0.25 lm). The following
2.1. Catalysts and reagents
operating conditions were used: carrier gas: N2, initial flow rate:
0.86 ml/min (pressure control mode), split ratio: 55:1, injection
temperature of 573 K, column temperature program: 313 K
(7 min), 20 K/min to 433 K (4 min), 20 K/min to 473 K (10 min),
and 20 K/min to 573 K (26 min), detection temperature of 573 K.
The qualitative analysis of the products was conducted by a gas
chromatograph equipped with a quadrupole mass spectrometer
(GC–MS) using the same capillary column.
Preparation of pure CeO2 catalyst was carried out by calcining
cerium oxide HS (Daiichi Kigenso, Japan) for 3 h in air at 873 K,
which is the optimized calcination temperature of CeO2 according
to the result of DMC synthesis for CO2 and methanol [41]. The speci-
fic surface area (BET method) of the CeO2 was 84 m2/g. The purity of
the CeO2 is 99.97%. All the chemicals for organic reactions were
commercially available and were used without further purification.
Other metal oxides were commercially available, supplied from the
Catalysis Society of Japan or synthesized by the precipitation
method: ZrO2 (Daiichi Kigenso Kogyo Co., Ltd., Zr(OH)4 was calcined
under air at 873 K for 3 h), MgO (Ube Industries, Ltd., MgO 500A,
Conversion and selectivity to each product in the reaction of
CO2 and amines were calculated on the basis of the amine amount
as follows:
ꢀ
ꢁ
Amount of amine after the reaction
Amount of amine before the reaction
Conv
ersion ð%Þ ¼ 1 ꢀ
ꢃ 100
873 K, 3 h), TiO2 (Nippon Aerosil Co., Ltd., P-25), c-Al2O3 (Nippon
Aerosil), ZnO (FINEX-50, Sakai Chemical Industry Co., Ltd.), and
SiO2 (Fuji Silysia Chemical Ltd., 773 K, 1 h). Y2O3 La2O3, Pr6O11 and
Sm2O3 were prepared by the precipitation method. Y(NO3)3ꢁnH2O
(Wako Pure Chemical Industries Ltd., >99.9%), La(NO3)3ꢁ6H2O
(Wako Pure Chemical Industries Ltd., >99.9%), Pr(NO3)3ꢁnH2O
(Wako Pure Chemical Industries Ltd., >99.5%) and Sm(NO3)3ꢁ6H2O
(Wako Pure Chemical Industries Ltd., >99.5%) were used as precur-
sors. A precursor (25 g) was dissolved in water (100 ml) and NH3aq
(1 M) was dropped with stirring. The pH of the solution was set to
10, resulting in a precipitate. The precipitate was filtered and
washed by water, followed by drying at 383 K overnight (12 h)
and calcining under air at 873 K (673 K for La2O3) for 3 h.
Product amount based on alkyl moiety of the substrate
Selecti
vity ð%Þ ¼
Amount of reacted amine
ꢃ100
The products that were not identified by GC–MS are denoted as
‘‘Others”. The selectivity of ‘‘Others” was calculated as follows:
Selecti
vity of others ð%Þ
Amount of reacted amine ꢀ Amount of identified products
¼
Amount of reacted amine
ꢃ 100
2.2. Catalytic test
The urea formation rate per catalyst gram (V1, mmol hꢀ1 gꢀ1
)
All the reactions were carried out in an autoclave reactor with an
inner volume of 190 ml. The standard procedure of the 1,3-
and the urea formation rate per specific surface area (V2,
mmol hꢀ1 mꢀ2) were calculated as follows:
V1 ðmmol hꢀ1gꢀ1Þ ¼ ½fðamount of urea; mmolÞ ꢀ ðamount of urea at 0 h; mmolÞg=ðreaction time; hÞ
ꢀ fðamount of urea without catalyst; mmolÞ
ꢀ ðamount of urea at 0 h without catalyst; mmolÞg=ðreaction time; hÞꢄ=ðcatalyst; gÞ
V2 ðmmol hꢀ1mꢀ2Þ ¼ ðV1; mmol hꢀ1gꢀ1Þ=ðspecific surface area; m2 gꢀ1
Þ
dibutylurea (DBU) synthesis from CO2 and n-butylamine was as fol-
lows: CeO2 (0.34 g, 2 mmol), n-butylamine 1.46 g (20 mmol) and N-
methylpyrrolidone (NMP) 8 ml (81 mmol) were put into the auto-
clave together with a spinner, and then the reactor was purged with
1 MPa CO2 (Shimakyu Co., Ltd., >99.5%) three times. The autoclave
was pressurized with CO2 to the desired pressure (typically
5.0 MPa) at room temperature, and then the autoclave was heated
to 433 K, where the CO2 pressure was about 12 MPa. In the experi-
ments for determination of the reaction rates, the autoclave with
substrate, solvent and catalyst was first purged with 1 MPa Ar
(Taiyo Nippon Sanso Corporation >99.9999%) three times and pres-
sured with Ar to 0.1 MPa. The autoclave was heated to 433 K and
was pressured with CO2 to 5.0 MPa (CO2 + Ar). The time when the
temperature reached 433 K was set as 0 h. The mixture was con-
stantly stirred during the reaction. After the reaction time, the reac-
tor was cooled in water bath to room temperature. Methanol
The procedure of the reverse reaction of DBU synthesis was as
follows: CeO2 (0.34 g, 2 mmol), DBU 1.72 g (10 mmol), H2O 0.18 g
(10 mmol) and N-methylpyrrolidone (NMP) 8 ml (81 mmol) were
put into the autoclave together with a spinner, and then the reactor
was purged with 1 MPa CO2 (Shimakyu Co., Ltd., >99.5%) three
times. The autoclave was pressurized with CO2 to 5.0 MPa at room
temperature, and then the autoclave was heated to 433 K. The mix-
ture was constantly stirred during the reaction. After the reaction
time, the reactor was cooled in water bath to room temperature.
The analysis of the products was conducted similar to that in the
synthesis of DBU from CO2 and n-butylamine.
The reusability test of CeO2 was conducted as follows: After the
reaction, the used catalyst was collected by decantation. The col-
lected catalyst was washed with ethanol and dried at 383 K for
12 h. After the treatment, the recovered catalyst is applied to
the next reaction. The catalyst was more or less lost during the
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015

M. Tamura et al. / Journal of Catalysis xxx (2015) xxx–xxx
3
recovery procedure of the catalyst, leading to decrease of the
O
catalyst amount (about 30% loss). Therefore, multiple batches (First
run: 4 batches) were conducted at the same time under the same
conditions to collect enough amount of the used catalyst. In order
to conduct the same reaction in the next run, the loss of the used
catalyst was compensated by the used catalyst obtained from
another batch, although the numbers of batches in the next reac-
tion were decreased (First run: 4 batches, Second run: 3 batches,
Third run: 2 batches, and Fourth run: 1 batch).
+
+ H2O
CO2
2
NH2
N
H
N
H
Scheme 1. Chemical equilibrium of DBU synthesis from CO2 and n-butylamine.
chemical equilibrium was about 74%, which was determined by
the experiments of the forward and reverse reactions at 433 K
(Fig. S1 and Table S1).
To compare the reaction rate, the reaction rate per metal oxide
amount (V1, mmol hꢀ1 gꢀ1) was calculated in consideration of both
the blank reaction and the reaction at 0 h. The typical example of
the time-courses below 30% conversion, where the conversion
was clearly lower than that in the equilibrium, is shown in
Fig. S2. Metal oxides except for CeO2 as well as blank showed no
conversion (0%) at 0 h reaction time, while CeO2 showed 5% con-
version with 66% selectivity at 0 h reaction time (Fig. S2). The reac-
tion rate per metal oxide surface (V2, mmol hꢀ1 mꢀ2) was also
calculated. CeO2 showed higher V1 and V2 than those of other metal
oxides. From these results, CeO2 was used in the following
experiments.
2.3. Catalyst characterization
The surface area of CeO2 was measured with BET method (N2
adsorption) using Gemini (Micromeritics). X-ray diffraction (XRD)
patterns were recorded using MiniFlex 600 with Cu K
15 mA) radiation. The amount of eluted metals into the reaction
solution was analyzed by inductively-coupled plasma atomic
emission spectrometry (ICP-AES, Thermo Fisher Scientific iCAP
6500).
a (40 kV,
3. Results and discussion
3.1. Comparison of metal oxides in the 1,3-dibutylurea (DBU) synthesis
from CO2 and n-butylamine
3.2. Optimization of reaction conditions and catalytic performance of
CeO2
At first, we investigated the performance of various metal
oxides for the synthesis of 1,3-dibutylurea (DBU) from CO2 and
n-butylamine as a model reaction (Table 1). Without any catalysts
(blank), n-butylamine was slightly converted with 72% selectivity
to give 6% DBU yield (entry 16). Similar conversion and selectivity
were obtained with ZnO, Sm2O3, Pr6O11, ZrO2, Al2O3, Y2O3, MgO,
TiO2 and SiO2, although ZnO, ZrO2, Sm2O3, MgO, ZrO2 and Pr6O11
provided slightly higher selectivity than blank (entries 3, 5, 7, 9–
15). Among metal oxides tested, CeO2 showed the highest conver-
sion with 94% selectivity. Using metal oxides of CeO2, ZnO, ZrO2
and Sm2O3, the DBU syntheses were conducted with larger amount
of metal oxides (2 mmol) for longer reaction times (4 h) (entries 2,
4, 6 and 8). CeO2 provided 73% conversion with 99% selectivity. On
the other hand, ZnO, ZrO2 and Sm2O3 provided much lower conver-
sion than CeO2 although the selectivity was >90% (ZnO: 97%, ZrO2:
98%, Sm2O3: 94%). In general, the synthesis of DBU from CO2 and
amine is known to be limited by the chemical equilibrium
(Scheme 1) [21,22,70,71]. In the case of the synthesis of DBU from
CO2 and n-butylamine, the conversion of n-butylamine in the
The effect of the reaction temperature was studied in the range
from 383 K to 453 K at 1 h reaction time (Fig. 1). The detailed data
are shown in Table S2. The conversion and selectivity to DBU
increased with increasing the reaction temperature, and leveled
off at P433 K. Considering the chemical equilibrium, the conver-
sion was almost saturated at higher than 433 K. In addition, the
reactions at long reaction time of 24 h using large amount of
CeO2 (1.7 g) were carried out (Fig. S3). The detailed data are shown
in Table S3. The conversion increased in the range from 393 to
403 K, but decreased at higher than 403 K, which is because this
reaction is an exothermic reaction. This tendency is in good agree-
ment with the previous reports [21,22].
The effect of CO2 pressure on the reaction was investigated in
the range from 0.5 to 5 MPa CO2 using CeO2 catalyst (Fig. 2). The
detailed data are shown in Table S4. The selectivity of DBU was
higher than 93% in any cases (Table S4). At any CO2 pressure, the
yield leveled off at longer reaction time than 4 h, and the maxi-
mum yield increased with increasing CO2 pressure from 0.5 MPa
Table 1
Catalyst screening for the synthesis of DBU from CO2 and n-butylamine in NMP.
Entry
Metal oxide
CeO2
Sa (m2 gꢀ1
)
MOx amount (mmol)
t (h)
Conv. (%)
Selec. (%)
V1 (mmol hꢀ1 gꢀ1
)
V2 (mmol hꢀ1 mꢀ2
)
1
2b
3
88
0.1
2
0.1
2
0.1
2
0.1
2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
–
1
4
1
4
1
4
1
4
1
1
1
1
1
1
1
1
24
73
10
42
8
23
9
31
9
7
8
7
6
94
99
86
97
86
98
77
94
70
79
66
70
76
70
68
72
77
–
33
–
12
–
9
0.87
–
0.50
–
0.25
–
0.23
–
0.10
0.01
<0.01
<0.01
<0.01
<0.01
<0.01
–
ZnO
66
46
38
4b
5
ZrO2
6
7
Sm2O3
8b
9
–
La2O3
Pr6O11
28
30
97
70
46
55
535
–
2.8
0.2
<0.1
<0.1
<0.1
<0.1
<0.1
–
10
11
12
13
14
15
16
c
-Al2O3
Y2O3
MgO
TiO2
SiO2
None
5
8
8
Reaction conditions: metal oxide 0.1 mmol (based on metal), n-butylamine 20 mmol, NMP 8 ml, Ar 0.1 MPa, CO2 5 MPa (at 433 K), 433 K.
a
Specific surface area determined by BET method.
CO2 5 MPa (at r.t.).
b
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015

4
M. Tamura et al. / Journal of Catalysis xxx (2015) xxx–xxx
100
90
80
70
60
50
40
30
20
10
0
100
80
60
40
20
383
393
403
413
423
433
443
453
0
0
10
20
30
40
50
Temperature /K
Time / h
Fig. 1. Effect of the reaction temperature in the synthesis of DBU from CO2 and
n-butylamine over CeO2 in NMP after 1 h (s: Conversion, White bar: selectivity
to DBU, Black bar: selectivity to others). Reaction conditions: CeO2 0.34 g (2 mmol),
n-butylamine 20 mmol, NMP 8 ml, CO2 5 MPa (at r.t.), 1 h.
Fig. 3. Time-course of the DBU synthesis from CO2 and n-butylamine over CeO2 in
NMP at 403 K (s: Conversion, d: Selectivity). Reaction conditions: CeO2 1.7 g
(10 mmol), n-butylamine 20 mmol, NMP 8 ml, 403 K, CO2 5 MPa (at r.t.).
100
80
60
40
20
0
100
90
80
70
60
50
40
30
20
10
0
0
4
8
12
16
20
24
Fresh
1
2
3
Time / h
Recycle number
Fig. 2. Effect of CO2 pressure in the DBU synthesis from CO2 and n-butylamine over
CeO2 in NMP. }: 5 MPa at r.t. (12 MPa at 433 K), h: 3 MPa at r.t. (4.5 MPa at 433 K),
M: 1 MPa at r.t. (1.5 MPa at 433 K), s: 0.5 MPa at r.t. (0.7 MPa at 433 K). Reaction
conditions: CeO2 0.34 g, n-butylamine 20 mmol, NMP 8 ml, 433 K.
Fig. 4. Reusability test of CeO2 in the DBU synthesis from CO2 and n-butylamine in
NMP. Reaction conditions: CeO2 0.10 g, n-butylamine 20 mmol, NMP 8 ml, CO2
5 MPa (at r.t.), 433 K, 1 h.
ture of CeO2 was maintained through the recycle tests. In addition,
the filtrate was measured by ICP-AES, resulting in no Ce species
(<0.1%) in the liquid phase. This reaction is catalyzed heteroge-
neously by CeO2.
to 5 MPa, which is explained by the chemical equilibrium of the
reaction.
Fig. 3 shows the time-course of the reaction at 403 K over CeO2
catalyst, and the details are shown in Table S5. The reaction
smoothly proceeded to level off at 24 h and longer reaction time,
providing 87% yield of DBU. The high selectivity (>95%) was main-
tained throughout the reaction.
The reusability of CeO2 for the DBU synthesis from CO2 and
n-butylamine was investigated (Fig. 4). The catalyst was easily
retrieved from the reaction mixture by decantation. CeO2 was
reused at least three times without marked loss of its catalytic
activity and selectivity. Moreover, the specific surface area of
CeO2 was almost constant after the first and second uses (fresh:
84 m2/g, first use: 82 m2/g), and XRD (Fig. S4) patterns were not
changed during the reaction. These results indicate that the struc-
The substrate scope of amines was evaluated using CeO2 in
NMP at 403 and 433 K (Table 2). The reaction conditions were
selected for each substrate near the equilibrium. Linear amines
reacted to afford the corresponding 1,3-dialkylureas in about 80%
and 60–70% yield at 403 and 433 K, respectively, and the selectiv-
ity was P93% (entries 1–8). Allylamine was also converted to the
corresponding 1,3-dialkylurea in about 60% yield at 433 and
403 K, and the selectivity was similar to or slightly lower than that
of the linear amines (entries 9 and 10). b-Branched amines such as
isobutylamine and 2-(N,N-dimethylamino)-ethylamine reacted to
afford the corresponding ureas, and the tendency of the yield and
selectivity is similar to that of the linear amines (entries 11–14).
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015

M. Tamura et al. / Journal of Catalysis xxx (2015) xxx–xxx
5
Table 2
Scope of amines in the synthesis of ureas over CeO2 in NMP.
Entry
Amine
Urea
T (K)
CeO2 amount (g)
t (h)
Conv. (%)
Yield (%)
Selectivity (%)
Urea
Others
O
1
2
403
433
1.7
0.34
24
16
81
66
79
62
98
94
2
2
NH2
N
H
N
H
O
3
4
403
433
1.7
0.34
24
4
87
73
86
72
99
99
1
1
NH2
N
H
N
H
5
6
403
433
1.7
0.34
24
4
84
79
82
77
98
97
2
3
O
NH2
N
N
H
H
O
7
8
403
433
1.7
0.34
24
16
83
75
81
70
96
93
4
7
NH2
N
N
H
H
O
9
10
403
433
1.7
0.34
24
16
71
64
65
58
92
91
8
9
NH2
N
H
N
H
O
11
12
403
433
1.7
0.34
48
16
87
74
86
71
99
96
1
4
NH2
N
N
H
H
O
13
14
403
433
1.7
0.34
48
16
85
83
83
76
96
92
4
8
N
N
N
NH2
N
N
H
O
H
15
16
403
433
1.7
0.34
48
16
70
50
69
49
99
98
1
2
NH2
NH2
N
H
N
H
17
18
403
433
1.7
0.34
48
16
72
57
65
54
90
95
10
5
O
N
H
N
H
O
19
20
403
433
1.7
0.34
48
16
13
16
11
11
86
73
14
27
NH2
N
H
N
H
O
21
22
403
433
1.7
0.34
48
16
14
12
<1
<1
>99
>99
<1
<1
N
H
N
N
23
24
403
433
1.7
0.34
24
4
1
1
<1
<1
>99
>99
<1
<1
NH2
O
N
N
H
H
Reaction conditions: CeO2 1.7 or 0.34 g, amine 20 mmol, NMP 8 ml, CO2 5 MPa (at r.t.).
NH
a
-Branched amines such as 2-aminobutane or cyclohexylamine
N
N
N
R
also reacted to provide the corresponding 1,3-dialkylureas in about
70% and 50% yield at 403 and 433 K, respectively, with the selectiv-
ity similar to those of linear amines (entries 15–18). The yields
were lower than those of linear amines. tert-Butylamine, the bulki-
est primary amine, provided 11% yield of the corresponding urea,
and the selectivity was also low because of the formation of the
corresponding isocyanate (entries 19 and 20). The formed iso-
cyanate will be stable owing to the bulkiness of the tert-butyl
group. On the other hand, diethylamine or aniline hardly reacted
(entries 21–24), which will be due to the bulkiness of secondary
amine and low nucleophilicity of aniline, respectively.
In our previous work [35], the combination catalyst system of
CeO2 and 2-cyanopyridine enhanced the synthesis of dimethyl car-
bonate (DMC) from CO2 and methanol, in which 2-cyanopyridine
works as a dehydration reagent through the hydration reaction
of 2-cyanopyridine to picolinamide (Eq. (1)). Moreover, addition
of 2-cyanopyridine improved the activity of the carboxylation of
CeO2 catalyst.
ð2Þ
N
H
+ R-NH2
On the other hand, tert-butylamine reacted to afford the corre-
sponding urea in high yield of 82% at 403 K (entries 7 and 8). In
general, bulkiness around the amino group in tert-butylamine is
so high that transformation of tert-butylamine is very difficult,
and hence the reported high yield in the literature is at most 33%
[26]. Therefore, the addition of 2-cyanopyridine will considerably
increase the reactivity of the substrate. It should be noted that ani-
line provided 80% and 71% yield of 1,3-diphenylurea at 403 K and
433 K, respectively (entries 9 and 10), although aniline did not give
the 1,3-diphenylurea without 2-cyanopyridine (Table 2, entries 23
and 24). This 1,3-diphenylurea yield is by far higher than the
reported yields (27% [20], 22% [23], 5.3% [72]). These results mean
that the reactivity of aniline was dramatically enhanced, which can
be attributed to removal of the produced H2O from the reaction
media by hydration of 2-cyanopyridine over CeO2 and activation
of amines by both 2-cyanopyridine and CeO2 [73,74]. Therefore,
the combination of 2-cyanopyridine and CeO2 in NMP improved
the yield of the substrates that showed very low reactivity in the
case of CeO2 in NMP without 2-cyanopyridine.
O
N
N
N
CeO2
ð1Þ
NH2
+ H2O
Additive effect of 2-cyanopyridine was examined in the synthe-
3.3. Kinetics in the synthesis of DBU from CO2 and n-butylamine
sis of 1,3-dialkylureas from amines with CO2 at 403 K and 433 K
(Table 3). In the case of n-butylamine, cyclohexylamine and diethy-
lamine (entries 1–6), the selectivity of the corresponding ureas was
low or none, and instead the amidine derivatives, which are
derived from the reaction of 2-cyanopyridine with the amine, were
mainly produced (Eq. (2)):
In order to clarify the reaction mechanism, reaction rates were
estimated under low conversion conditions, where the conversion
is below 30%. In the synthesis of DBU from CO2 and n-butylamine,
the reaction proceeded without CeO2 as shown in Table 1. In
addition, the reaction proceeds during heating to the desired
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015

6
M. Tamura et al. / Journal of Catalysis xxx (2015) xxx–xxx
Table 3
Scope of amines in the synthesis of ureas using the combination catalyst system of CeO2 in NMP and 2-cyanopyridine (CeO2 in NMP with 2-cyanopyridine).
Entry
Amine
Urea
T (K)
CeO2 amount (g)
t (h)
Conv. (%)
Yield (%)
Selectivity (%)
Urea
Others
1
2
403
433
1.7
0.34
24
4
99
98
13
25
13
26
87
74
O
NH2
N
H
N
H
3
4
403
433
1.7
0.34
24
4
99
99
<1
<1
<1
<1
>99
>99
NH2
O
N
H
N
H
5
6
403
433
1.7
0.34
24
4
68
40
<1
<1
<1
<1
>99
>99
O
N
H
N
N
7
8
403
433
1.7
0.34
24
8
95
85
82
65
85
77
15
23
O
NH2
NH2
N
H
N
H
9
10
403
433
1.7
0.34
24
8
95
96
71
80
75
84
25
16
O
N
N
H
H
Reaction conditions: CeO2 1.7 or 0.34 g, amine 20 mmol, 2-cyanopyridine 100 mmol, NMP 8 ml, CO2 5 MPa (at r.t.).
Table 4
Solvent effect in the synthesis of DBU from CO2 and n-butylamine over CeO2.
Entry
Solvent
Conv. (%)
Selectivity (%)
Urea
1-Butyl isocyanate
N-Butylidenebutylamine
Solvent derivatives
Others
1
2
3
4
5
6
7
8
9
N-Methylpyrrolidone (NMP)
Tetrahydrofuran (THF)
Diethylether (DEE)
Benzene
Sulfolane
2-Propanol
Acetonitrile
Dimethyl sulfoxide (DMSO)
Methanol
Dimethyl formamide (DMF)
None
73
25
26
41
59
38
53
82
46
88
34
99
98
97
96
95
89
66
64
43
24
95
0.3
<0.1
<0.1
<0.1
0.2
<0.1
0.5
0.4
0.4
1.5
1.8
1.0
0.5
0.9
0.7
<0.1
0.4
<0.1
0.2
<0.1
<0.1
<0.1
<0.1
<0.1
8.2
32
0.3
0.9
1.5
0.7
4.3
2.1
<0.1
29a
56
<0.1
<0.1
0.5
<0.1
15
3.9
10
11
62
–
Reaction conditions: CeO2 2 mmol (0.34 g), n-butylamine 20 mmol, solvent 8 ml, Ar 0.1 MPa, CO2 5 MPa (at 433 K), 4 h, 433 K.
a
Many peaks including DMSO derivatives were observed.
temperature. Therefore, the reaction rate was calculated in consid-
eration of the blank reaction and heating. The detailed measurement
and calculation method was shown in the experimental section.
Slope = 19
25
20
15
10
5
3.3.1. Kinetic studies on solvent effect
(i) CeO2 in NMP
For transformation reactions of CO2 to chemicals, solvents often
have large influence on selectivity, and the selection of an effective
solvent is pretty important to achieve high selectivity and yield
[36,37]. Table
4 shows the results of the solvent effect.
N-Methylpyrrolidone (NMP), tetrahydrofuran (THF), diethylether
(DEE), benzene, sulfolane and 2-propanol provided P89% selectiv-
ity to DBU, and NMP showed higher conversion than other solvents
tested (entries 1–10). On the other hand, acetonitrile, dimethyl sul-
foxide (DMSO), methanol and dimethyl formamide (DMF) showed
low selectivity (<70%), although the conversion was comparatively
high (entries 7–10). The low selectivity can be explained by the
reaction of n-butylamine with the solvent, and such phenomenon
was often reported in other catalyst systems [27,72,75]. The main
identified by-products in the case of NMP solvent include butyliso-
cyanate, N-butylidenebutylamine and solvent derivatives. In the
case of acetonitrile, N-butylacetamide was produced from
acetonitrile as a main by-product, and the selectivity was 32%.
N-Butylacetamide can be produced by the following two reaction
pathways (Eqs. (3) and (4)): (i) the reaction of acetonitrile with
H2O, which was produced by DBU formation, providing acetamide,
followed by the reaction of the produced acetamide with
n-butylamine (Eq. (3)); (ii) the reaction of acetamide produced
(ii) CeO2 only
(no solvent)
Slope = 6.5
Slope = 5.8
(iii) NMP only
(without CeO2)
Slope = 1.7
(iv) Without CeO2 +NMP
0
0
0.5
1
1.5
Time / h
Fig. 5. Time-courses of DBU synthesis under various conditions (s) CeO2 + NMP,
reaction conditions: CeO2 0.1 mmol (0.017 g), n-butylamine 20 mmol, NMP 8 ml, Ar
0.1 MPa, CO2 5 MPa (at 433 K), 433 K. (h) CeO2 only (no solvent), reaction
conditions: CeO2 0.5 mmol (0.086 g), n-butylamine 100 mmol, Ar 0.1 MPa, CO2
5 MPa (at 433 K), 433 K. (}) NMP only (without CeO2), reaction conditions:
n-butylamine 20 mmol, NMP 8 ml, Ar 0.1 MPa, CO2 5 MPa (at 433 K), 433 K. (M)
without CeO2 and NMP, reaction conditions: n-butylamine 100 mmol, Ar 0.1 MPa,
CO2 5 MPa (at 433 K), 433 K.
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015

M. Tamura et al. / Journal of Catalysis xxx (2015) xxx–xxx
7
Table 5
Comparison of the DBU formation rate in the reaction of CO2 and n-butylamine using CeO2 in various solvents.
Entry
Solvent
CeO2 amount (mmol)
Conv. (%)
Sel. (%)
Dielectric constant
Dissolved CO2
amount (g)a
Formation rateb (mmol hꢀ1
)
Difference of the
rates (mmol hꢀ1
2.5
)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
NMP
0.1
0
0.1
0
0.1
0
0.1
0
0.1
0
0.1
0
0.1
0
0.1
0
24
8
5
1
3
<1
4
2
7
2
56
54
12
6
1
1
92
72
84
20
71
<1
95
19
93
40
75
76
59
26
59
41
62
51
32.2
7.6
3
6
3.7
1.2
0.9
<0.1
0.4
<0.1
0.6
0.1
1.2
0.2
8.2
8.0
1.4
0.3
0.1
0.1
1.0
0.2
THF
0.9
0.4
DEE
4.3
7
Benzene
2-Propanol
DMSO
2.2
8
0.5
18.6
46.5
35.9
32.7
44.0
11
7
1.0
0.2
CH3CN
Methanol
Sulfolane
15
11
5
1.1
<0.1
0.8
0.1
0
10
2
Reaction conditions: CeO2 0.1 mmol (0.017 g) or 0 mmol, n-butylamine 20 mmol, solvent 8 ml, Ar 0.1 MPa, CO2 5 MPa (at 433 K), 433 K, 1 h.
a
Dissolved amount of CO2 was calculated by the following equation: Dissolved amount of CO2 = weight of autoclave + solvent + CO2 (5 MPa) ꢀ weight of autoclave + CO2
(5 MPa) ꢀ solvent.
b
The upper row in each entry is total formation rate and the second row in each entry is formation rate of DBU with solvent only.
O
from acetonitrile with H2O to give acetic acid, followed by the
reaction of the produced acetic acid with n-butylamine (Eq. (4)).
In our previous works, CeO2 can catalyze the hydration of nitriles
to amides [31,33], and in fact, the formation of acetamide was
observed (1.2 mmol) in this reaction, and however, formation of
acetic acid was not observed. This result indicates that the reaction
pathway of Eq. (3) will be a main route for the formation of
N-butylacetamide:
O
NH2
CO2
CH3OH
H3CO
N
H
H3CO
OH
ð6Þ
ð7Þ
O
O
CH3OH
+
CO2
NH2
N
H
OH
H3CO
N
H
O
O
H2O
NH2
CH3CN
ð3Þ
N
H
-NH3
NH2
In the case of dimethylformamide (DMF), N-butylformamide
was produced from DMF as a main by-product, and the selectivity
O
O
O
H2O
H2O
NH2
ð4Þ
CH3CN
N
H
OH
NH2
In the case of methanol solvent, methyl butylcarbamate was
produced from methanol as a main by-product, and the selectivity
was 56%. Methyl butylcarbamate can be produced by the following
three reaction pathways (Eqs. (5)–(7): (i) alcoholysis of DBU with
methanol (Eq. (5)), (ii) reaction of methanol with CO2 to produce
methyl hydrogen carbonate, followed by amidation of the methyl
hydrogen carbonate with n-butylamine (Eq. (6)), and (iii) Carboxy-
lation of n-butylamine with CO2 to give n-butyl carbamic acid, fol-
lowed by esterification of the n-butyl carbamic acid with methanol
(Eq. (7)):
was 62%. N-Butylformamide can be produced by the following reac-
tion pathway, transamidation of DMF with n-butylamine (Eq. (8)):
O
O
NH2
ð8Þ
H
N
H
N
H
-
HN
In the case of dimethylsulfoxide (DMSO), many peaks including
DMSO derivatives were observed in the gas-chromatography chart.
From the viewpoints of conversion and selectivity, we selected
NMP as the optimized solvent for the reaction.
O
O
ð5Þ
+
CH3OH
N
H
N
H
H3CO
N
H
-
NH2
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015

8
M. Tamura et al. / Journal of Catalysis xxx (2015) xxx–xxx
To clarify the effect of NMP, the reactions in four cases were
(a)
9
8
7
6
5
4
3
2
1
0
compared: (i) CeO2 in NMP, (ii) CeO2 only (no solvent), (iii) NMP
only (without CeO2) and (iv) without CeO2 and NMP. The time
dependence is shown in Fig. 5 and the details are shown in
Table S6. In the presence of NMP, 20 mmol n-butylamine was used
for the reaction ((i) and (iii)), and in contrast, in the absent of NMP,
100 mmol n-butylamine was used in order to reduce the experi-
mental error due to the low boiling point of n-butylamine
(350 K), and moreover in the case of (iii) larger amount of CeO2
catalyst (0.5 mmol) was adopted to adjust the molar ratio of
n-butylamine/CeO2 to that of (i) (Table S6). Comparison of (i) with
(iii) shows that the slope of (i) is 3.3 times larger than that of (ii),
and comparison of (ii) with (iv) shows that the slope of (ii) is 3.8
times larger than that of (iv). These results indicate that CeO2
enhances the reaction whether in the presence or in the absence
of NMP. On the other hand, comparison of (i) with (ii) shows that
NMP also enhanced the reaction. Considering the slope of the
time-courses, the formation rate of (i) is larger than the total of
(ii) and (iii), which suggests that CeO2 and NMP cooperatively work
to enhance the formation of DBU. On the other hand, focusing on
the selectivity of DBU (Table S6), the selectivity in the reactions
using CeO2 ((i) and (ii)) is higher than that without CeO2 ((iii)
and (iv)). By-products derived from the reaction of NMP with
n-butylamine were observed, and the amount of the by-products
was larger in the absence of CeO2, which implies that CeO2
suppresses the side reactions of NMP.
From the above results, solvents can catalyze the synthesis of
DBU without CeO2. To precisely evaluate the solvent effect, the
reactions were carried out using various solvents in the absence
and presence of CeO2 (Table 5). The formation rates of DBU
(mmol hꢀ1) in the absence and presence of CeO2 and the difference
of the formation rates were calculated in consideration of the reac-
tions at 0 h (Table 5). The detailed data including the reactions at
0 h are shown in Table S7. In the absence of CeO2 (entries 2, 4, 6,
8, 10, 12, 14, 16 and 18), DMSO showed the highest conversion
with moderate selectivity (entry 12), followed by NMP (entry 2).
Other solvents such as THF, DME, benzene, 2-propanol, CH3CN,
methanol and sulfolane showed quite low yield (62%) (entries 4,
6, 8, 10, 14, 16 and 18). The formation rate of DBU (mmol hꢀ1) in
the case of DMSO is the highest among the solvents. On the other
hand, in the presence of CeO2, DMSO also showed higher conver-
sion than the other solvents used, but the conversion and selectiv-
ity were almost the same as those in the absence of CeO2, meaning
that the catalytic function of CeO2 was not exerted in DMSO
(entries 11 and 12). The possible interpretation of the result is that
DMSO deactivated CeO2 by the adsorption on the surface of CeO2.
Other solvents such as THF, DME, benzene and 2-propanol in the
presence of CeO2 provided a little higher conversion than in the
absence of CeO2, while the selectivity was much improved (entries
3–10). This means that CeO2 can catalyze the reaction in these sol-
vents. On the other hand, acetonitrile or methanol provided low
selectivity at both low and high conversion (Tables 4 and 5), which
indicates that side reactions involved in the solvent occur compar-
atively (Eqs. (3)–(8)). In addition, the total formation rates of DBU,
formation rate of DBU with solvent only and the difference of the
rates were compared with the dielectric constant of the solvents
and dissolved CO2 amount in the solvents using the solvents show-
ing high selectivity (entries 1–12; NMP, THF, DME, benzene,
2-propanol and DMSO). Clear relationship was not found between
the formation rates and the dissolved CO2 amount (Fig. S5). Fig. 6
shows the comparison with the dielectric constant of the solvents.
The correlation between the formation rate of DBU with solvent
only and the dielectric constant shows that the solvent with higher
dielectric constant is effective for the reaction (Fig. 6b). This result
is in accordance with the previous report that polar aprotic sol-
vents are effective due to stabilization of the ionic intermediate
DMSO
NMP
Benzene
DEE 2-Propanol
THF
10
20
30
40
50
0
Dielectric constant
(b)
9
DMSO
8
7
6
5
4
3
2
1
0
Benzene
DEE
NMP
2-Propanol
THF
0
10
20
30
40
50
Dielectric constant
9
8
7
6
5
4
3
2
1
0
(c)
NMP
Benzene
DEE
2-Propanol
THF
DMSO
10
20
30
40
50
0
Dielectric constant
Fig. 6. Correlations between the formation rates and the dielectric constant of
solvents: (a) total formation rate, (b) formation rate with solvent only, and (c)
difference of the rates. (a) Total formation rate is the rate measured using CeO2 and
solvents (Table 5 entries 1, 3, 5, 7, 9 and 11). (b) The formation rate with solvent
only is the rate measured using solvent only (Table 5 entries 2, 4, 6, 8, 10 and 12).
or adsorption of CO2 [72]. Fig. 6c shows the difference of the rates
as a function of the dielectric constant. The volcano-type correla-
tion was obtained, and the difference of the rates gradually
increased with increasing the dielectric constant 630, but
decreased with higher dielectric constant than 30. In addition,
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015

M. Tamura et al. / Journal of Catalysis xxx (2015) xxx–xxx
9
6
5
5
(a)
(b)
4.5
Slope = 0.1
Slope = 0.2
4
4
3
3.5
-0.5
0
1
0.5
-1
0
1
2
Ln(CO2 Pressure / MPa)
Ln (concentration of n-butylamine / M)
Fig. 7. (a) Dependence of CO2 pressure on the reaction rate over CeO2 in NMP, and (b) Dependence of n-butylamine concentration on the reaction rate over CeO2 in NMP.
Detailed results of (a) and (b) are shown in Tables S8 and S9, respectively. Reaction conditions of (a): CeO2 0.017 g, n-butylamine 20 mmol, NMP 8 ml, Ar 0.1 MPa, CO2 0.5–
5 MPa (at 433 K), 433 K. Reaction conditions of (b): CeO2 0.017 g, n-butylamine 10–30 mmol, NMP 8 ml, Ar 0.1 MPa, CO2 5 MPa (at 433 K), 433 K.
Table 6
Reaction orders with respect to CO2 pressure and n-butylamine concentration on various systemsa.
Entry
Catalyst system
CeO2 (g)
NMP (ml)
Reaction order with
Reaction order with respect to
respect to CO2 pressureb
n-butylamine concentrationc
1
2
3
4
CeO2 in NMP
CeO2 (in THF)
NMP only
0.017
0.017
0
8
0.2
0.4
0.5
1.0d
0.1
0.1
0 (THF 8 ml)
8
0
1.0
Without CeO2 and NMP
0
>0.7d
a
b
c
Detailed data are shown in Figs. S7 and S8, and Tables S8–S13.
Reaction conditions: CeO2, NMP (or THF), amine 20 mmol, Ar 0.1 MPa, CO2 0.5–5 MPa or 2–5 MPa (at 433 K), 433 K, 1 h.
Reaction conditions: CeO2, NMP (or THF), n-butylamine 10–30 mmol, Ar 0.1 MPa, CO2 5 MPa (at 433 K), 433 K, 1 h.
The reaction orders were calculated from the data of Ref. [28].
d
mmol hꢀ1 gꢀ1) as a function of Ln(n-butylamine concentration),
affording the reaction order of 0.1 with respect to the
n-butylamine concentration. In order to consider the roles of
CeO2 and NMP individually, the effect of CO2 pressure and
n-butylamine concentration was investigated using CeO2 (in THF)
and NMP only. In the case of CeO2 (in THF), THF was selected as
the solvent because the influence of THF on the reaction rate and
selectivity was quite low (Table 5). The obtained reaction orders
are listed in Table 6 and the details are shown in Figs. S7 and S8,
and Tables S8–S13. The reaction with CeO2 in THF showed the
reaction orders of 0.4 and 0.1 with respect to CO2 pressure and
n-butylamine concentration, respectively (entry 2), and the reac-
tion orders were lower than those without CeO2 and NMP (entries
2 and 4). This means that CeO2 reduced these reaction orders (CO2
pressure and n-butylamine concentration) and indicates that CO2
and n-butylamine are strongly adsorbed on the surface of CeO2.
On the other hand, the reaction in the case of NMP only showed
the reaction orders of 0.5 and 1.0 with respect to CO2 pressure
and n-butylamine concentration, respectively (entry 3). The reac-
tion order with respect to n-butylamine concentration (1.0) is sim-
ilar to that without CeO2 and NMP (entry 4) reported in the
literature (>0.7) [28]; however, the reaction order with respect to
CO2 pressure (0.5) is lower than that without CeO2 and NMP
(1.0) (entries 3 and 4). This result indicates that NMP plays an
important role for activation of CO2. From these results, CeO2
works for activation of CO2 and n-butylamine, and NMP works
for activation of CO2. In addition, the lower reaction order with
respect to CO2 pressure over CeO2 in NMP (0.2, entry 1) than that
over CeO2 (in THF) (0.4, entry 2) and that over NMP only (0.5, entry
3) can be explained by the combination effect of CeO2 and NMP.
In order to clarify whether NMP can catalyze the reaction, the
effect of CeO2 and NMP amount (1 mmol and 5 mmol) was studied
in the DBU synthesis using THF as a solvent (Table S14). The
Table 7
Comparison of reactivities of various amines in the synthesis of 1,3-dialkylureas over
CeO2 in NMP.
Entry
Amine
V1 (mmol hꢀ1 gꢀ1
)
1
2
3
4
77
54
40
43
NH2
NH2
NH2
NH2
5
34
NH2
Reaction conditions: CeO2 0.1 mmol (0.017 g), n-butylamine 20 mmol, NMP 8 ml, Ar
0.1 MPa, CO2 5 MPa (at 433 K), 433 K, 1 h.
Log (formation rate) was plotted as a function of the reciprocal
value of dielectric constants (Fig. S6), which showed the similar
tendency to that in Fig. 6. This result indicates that the solvent with
too high polarity is not preferable, which is due to the deactivation
of the CeO2 by adsorption of the solvent on the surface of CeO2.
Therefore, combination of NMP and CeO2 will be preferable for
the reaction.
3.3.2. Effect of CO2 pressure and n-butylamine concentration in the
reactions using CeO2 in NMP, CeO2 (in THF) and NMP only
The effect of CO2 pressure was examined with CeO2 in NMP in
the range of 0.5–5 MPa. Fig. 7a shows Ln(V1, mmol hꢀ1 gꢀ1) as a
function of Ln(CO2 pressure, MPa). Good linear correlation between
Ln(V1, mmol hꢀ1 gꢀ1) and Ln(CO2 pressure, MPa) was obtained, and
the reaction order with respect to CO2 pressure was calculated to
be 0.2. Next, the effect of n-butylamine concentration was studied
in the range of 10–30 mmol n-butylamine. Fig. 7b shows Ln(V,
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015

10
M. Tamura et al. / Journal of Catalysis xxx (2015) xxx–xxx
amount of NMP has no influence on the reaction in both the pres-
ence and the absence of CeO2, indicating that NMP alone cannot
work as a catalyst for the reaction. Therefore, NMP works as an
effective solvent for the reaction, and the property, that is high
polarity and inertness to CeO2, will be effective for the reaction.
effect of solvents, NMP only promoted the CeO2-catalyzed reaction,
and the combination of CeO2 and NMP showed higher activity than
CeO2 only (non solvent) or NMP only, indicating that the combina-
tion of CeO2 and NMP is preferable for the reaction. Comparison of
the reaction orders with respect to CO2 pressure and n-butylamine
concentration using CeO2 in NMP, CeO2 (in THF) and NMP only
showed that CeO2 (in THF) decreased the reaction orders with
respect to both CO2 pressure (1 ? 0.4) and n-butylamine concen-
tration (>0.7 ? 0.1), and that NMP decreased the reaction order
with respect to CO2 pressure (1 ? 0.5), although the reaction order
with respect to n-butylamine concentration was not changed. CeO2
in NMP furthermore decreased the reaction order with respect to
CO2 (0.2) compared with CeO2 in THF. The behavior can be inter-
preted by the amine activation with CeO2 and CO2 activation with
CeO2 and NMP. The tendency of kinetics in the DBU synthesis
agreed with that of the CeO2-catalyzed cyclic urea synthesis
reported in our previous work, which suggests that the reaction
of the combination system of CeO2 with NMP solvent (CeO2 in
NMP) proceeds through the rate-determining step that amine
adspecies formed by the decomposition of carbamate adspecies
attacked another carbamate adspecies to produce DBU. NMP may
work as an effective solvent for stabilization of the intermediate
in the reaction or activation of the substrate.
3.3.3. Effect of the substituent in amines
The urea formation rates from amines such as n-butylamine,
n-pentylamine, n-hexylamine, 1-amino-2-methylpropane and
2-aminobutane were compared under the same conditions
(Table 7). The details are shown in Table S15. The urea formation
rate from linear alkylamines gradually decreased with increasing
the carbon number (entries 1–3), tendency of which was also
observed in the synthesis of DMC from CO2 and methanol [73].
On the other hand, the urea formation rates from branched
alkylamines were much lower than those of linear alkylamines
with the same carbon number (entries 1, 4 and 5), and the
reactivity of 2-aminobutane is lower than that of 1-amino-2-
methylpropane (entries 4 and 5). To clarify the effect of the steric
hindrance, correlation between the reactivity and steric substituent
constants (Es) which has been proposed by Taft [76,77] was
studied (Fig. S9). Focusing on the reactivity of C4 amines
(n-butylamine, 1-amino-2-methylpropane and 2-aminobutane),
positive linear correlation (Slope = 1) was obtained, which indicates
that steric hindrance largely influences the reactivity. In contrast,
as for the linear amines (n-butylamine, n-pentylamine and
n-hexylamine), the reactivity decreased with increasing the length
of the alkyl chain in spite of the Es being almost the same. There-
fore, the steric hindrance of the substituents mainly influences the
reactivity, but another factor will also influence the reactivity.
From these results, the reaction of amine will be involved in the
rate-determining step.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2015.11.015.
References
The above tendency of the substrate reactivity is similar to that
in the synthesis of cyclic urea [37]. Moreover, the reaction orders
with respect to CO2 pressure and amine concentration in the syn-
thesis of DBU over CeO2 in NMP (reaction order with respect to CO2
pressure: 0.2, reaction order with respect to n-butylamine concen-
tration: 0.1) are almost the same as those in the synthesis of cyclic
urea (reaction order with respect to CO2 pressure: ꢀ0.2, reaction
order with respect to n-butylamine concentration: 0.2). Principally,
the reaction mechanism of the synthesis of DBU over CeO2 in NMP
will be the same as that of the synthesis of cyclic urea over CeO2,
where the amino group generated by the decomposition of carba-
mate adspecies on CeO2 attacks another carbamate adspecies [37].
DBU synthesis will follow the Langmuir–Hinshelwood type mech-
anism through the reaction of the carbamate adspecies and amine
adspecies on CeO2, and the obtained reaction orders can be inter-
preted by the high coverage of the carbamate adspecies. In addi-
tion, NMP may work as an effective solvent for stabilization of
the intermediate in the reaction or activation of the substrate.
[1] S.J. Davis, K. Caldeira, H.D. Matthews, Science 329 (2010) 1330–1333.
[2] S.R. Loarie, P.B. Duffy, H. Hamilton, G.P. Asner, C.B. Field, D.D. Ackerly, Nature
462 (2009) 1052–1055.
[3] T. Sakakura, J.-C. Choi, H. Yasuda, Chem. Rev. 107 (2007) 2365–2387.
[4] C.D.N. Gomes, O. Jacquet, C. Villiers, P. Thuéry, M. Ephritikhine, T. Cantat,
Angew. Chem. Int. Ed. 51 (2012) 187–190.
[5] M. Aresta, Carbon Dioxide as Chemical Feedstock, WILEY-VCH, 2010.
[6] C. Song, Catal. Today 115 (2006) 2–32.
[7] R. Tanaka, M. Yamashita, K. Nozaki, J. Am. Chem. Soc. 131 (2009) 14168–
14169.
[8] P.G. Jessop, T. Ikariya, R. Noyori, Chem. Rev. 99 (1999) 475–494.
[9] P.G. Jessop, Y. Hsiao, T. Ikariya, R. Noyori, J. Am. Chem. Soc. 118 (1996) 344–
355.
[10] F. Arena, K. Barbera, G. Italiano, G. Bonura, L. Spadaro, F. Frusteri, J. Catal. 249
(2007) 185–194.
[11] J. Toyir, P.R. de la Piscina, J.L.G. Fierro, N. Homs, Appl. Catal. B 34 (2001) 255–
266.
[12] O.S. Joo, K.D. Jung, I. Moon, A.Y. Rozovskii, G.I. Lin, S.H. Han, S.J. Uhm, Ind. Eng.
Chem. Res. 38 (1999) 1808–1812.
[13] W. Wang, S. Wang, X. Ma, J. Gong, Chem. Soc. Rev. 40 (2011) 3703–3727.
[14] S.-I. Fujita, M. Arai, J. Jpn. Petrol. Inst. 48 (2005) 67–75.
[15] M. Tamura, M. Honda, Y. Nakagawa, K. Tomishige, J. Chem. Technol.
Biotechnol. 89 (2014) 19–33.
[16] M. Honda, M. Tamura, Y. Nakagawa, K. Tomishige, Catal. Sci. Technol. 4 (2014)
2830–2845.
[17] F. Bigi, R. Maggi, G. Sartori, Green Chem. 2 (2000) 140–148.
[18] O. Kreye, H. Mutice, M.A.R. Meier, Green Chem. 15 (2013) 1431–1455.
[19] D.J. Diaz, A.K. Darko, L. McElwee-White, Eur. J. Org. Chem. (2007) 4453–4465.
[20] F. Shi, Y. Deng, T. SiMa, J. Peng, Y. Gu, B. Qiao, Angew. Chem. Int. Ed. 42 (2003)
3257–3260.
[21] A. Ion, V. Parvulescu, P. Jacobs, D.D. Vos, Green Chem. 9 (2007) 158–161.
[22] T. Jiang, X. Ma, Y. Zhou, S. Liang, J. Zhang, B. Han, Green Chem. 10 (2008) 465–
469.
4. Conclusion
CeO2 showed higher activity in the synthesis of 1,3-dibutylurea
(DBU) from CO2 and n-butylamine in NMP than other metal oxides
tested. The selection of solvent is very important from the view-
points of activity and selectivity, and NMP promoted the synthesis
of DBU over CeO2 catalyst without loss of selectivity. Various ami-
nes such as linear primary alkylamines and branched primary alky-
lamines were converted to the corresponding 1,3-dialkylureas over
CeO2 in NMP, and however secondary amine and aniline were not
converted. Combination of 2-cyanopyridine and CeO2 in NMP
(CeO2 in NMP with 2-cyanopyridine) enhanced the reaction of
low-reactive substrates such as tert-butylamine and aniline, pro-
viding about 80% yield of the corresponding ureas. These yields
are much higher than those reported in the literatures. From the
[23] J. Li, X. Guo, L. Wang, X. Ma, Q. Zhang, F. Shi, Y. Deng, Sci. China Chem. 53
(2010) 1534–1540.
[24] T. Kimura, K. Kamata, N. Mizuno, Angew. Chem. Int. Ed. 51 (2012) 6700–6703.
[25] F. Shi, Q. Zhang, Y. Ma, Y. He, Y. Deng, J. Am. Chem. Soc. 127 (2005) 4182–4183.
[26] D.-L. Kong, L.-N. He, J.-Q. Wang, Synlett 8 (2010) 1276–1280.
[27] R. Zhang, L. Hua, L. Gua, B. Song, J. Chen, Z. Hou, Chin. J. Chem. 31 (2013) 381–
387.
[28] C. Wu, H. Cheng, R. Liu, Q. Wang, Y. Hao, Y. Yu, F. Zhao, Green Chem. 12 (2010)
1811–1816.
[29] L. Vivier, D. Duprez, ChemSusChem 3 (2010) 654–678.
[30] M. Tamura, K. Shimizu, A. Satsuma, Chem. Lett. 41 (2012) 1397–1405.
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015

M. Tamura et al. / Journal of Catalysis xxx (2015) xxx–xxx
11
[31] M. Tamura, K. Sawabe, K. Tomishige, A. Satsuma, K.-I. Shimizu, ACS Catal. 5
(2015) 20–26.
[53] R. Zhang, L. Guo, C. Chen, J. Chen, A. Chen, X. Zhao, X. Liu, Y. Xiu, Z. Hou, Catal.
Sci. Technol. 5 (2015) 2959–2972.
[32] M. Tamura, T. Tonomura, K. Shimizu, A. Satsuma, Green Chem. 14 (2012) 984–
991.
[54] G. Fan, S. Luo, T. Fang, Q. Wu, G. Song, J. Li, J. Mol. Catal. A 404–405 (2015) 92–
97.
[33] M. Tamura, H. Wakasugi, K. Shimizu, A. Satsuma, Chem. Eur. J. 17 (2011)
11428–11431.
[55] R. Juarez, P. Concepcion, A. Corma, V. Fornes, H. Garcia, Angew. Chem. Int. Ed.
49 (2010) 1286–1290.
[34] M. Honda, M. Tamura, K. Nakao, K. Suzuki, Y. Nakagawa, K. Tomishige, ACS
Catal. 4 (2014) 1893–1896.
[35] M. Honda, M. Tamura, Y. Nakagawa, S. Sonehara, K. Suzuki, K. Fujimoto, K.
Tomishige, ChemSusChem 6 (2013) 1341–1344.
[36] M. Tamura, M. Honda, K. Noro, Y. Nakagawa, K. Tomishige, J. Catal. 305 (2013)
191–203.
[56] R. Juarez, A. Corma, H. Garcia, Green Chem. 11 (2009) 949–952.
[57] E. Leino, P. Mäki-Arvela, K. Eränen, M. Tenho, D.Y. Murzin, T. Salmi, J.P.
Mikkola, Chem. Eng. J. 176–177 (2011) 124–133.
[58] Y. Sakata, V. Ponec, Appl. Catal. A 166 (1998) 173–184.
[59] G. Vilé, B. Bridier, J. Wichert, J. Pérez-Ramírez, Angew. Chem. Int. Ed. 51 (2012)
8620–8623.
[37] M. Tamura, K. Noro, M. Honda, Y. Nakagawa, K. Tomishige, Green Chem. 15
(2013) 1567–1577.
[60] M. Tamura, S.M.A. Hakim Siddiki, K. Shimizu, Green Chem. 15 (2013) 1641–
1646.
[38] M. Honda, S. Sonehara, H. Yasuda, Y. Nakagawa, K. Tomishige, Green Chem. 13
(2011) 3406–3413.
[61] M. Tamura, T. Tonomura, K. Shimizu, A. Satsuma, Green Chem. 14 (2012) 717–
724.
[39] M. Honda, S. Kuno, S. Sonehara, K.-I. Fujimoto, K. Suzuki, Y. Nakagawa, K.
Tomishige, ChemCatChem 3 (2011) 365–370.
[62] Y. Wang, F. Wang, C. Zhang, J. Zhang, M. Li, J. Xu, Chem. Commun. 50 (2012)
2438–2441.
[40] M. Honda, A. Suzuki, B. Noorjahan, K. Fujimoto, K. Suzuki, K. Tomishige, Chem.
Commun. (2009) 4596–4598.
[63] Y. Wang, F. Wang, Q. Song, Q. Xin, S. Xu, J. Xu, J. Am. Chem. Soc. 135 (2013)
1506–1515.
[41] Y. Yoshida, Y. Arai, S. Kado, K. Kunimori, K. Tomishige, Catal. Today 115 (2006)
95–101.
[42] K. Tomishige, Y. Furusawa, Y. Ikeda, M. Asadullah, K. Fujimoto, Catal. Lett. 76
(2001) 71–74.
[43] K. Tomishige, H. Yasuda, Y. Yoshida, M. Nurunnabi, B. Li, K. Kunimori, Green
Chem. 6 (2004) 206–214.
[64] M. Tamura, K. Tomishige, Angew. Chem. Int. Ed. 54 (2015) 864–867.
[65] N. Ichikawa, S. Sato, R. Takahashi, T. Sodesawa, H. Fujita, T. Atoguchi, A. Shiga, J.
Catal. 239 (2006) 13–22.
[66] S. Sato, K. Koizumi, F. Nozaki, J. Catal. 178 (1998) 264–274.
[67] S. Sato, R. Takahashi, T. Sodesawa, K. Matsumoto, Y. Kamimura, J. Catal. 184
(1999) 180–188.
[44] K. Tomishige, K. Kunimori, Appl. Catal. A 237 (2002) 103–109.
[45] R. Saada, S. Kellici, T. Heil, D. Morgan, B. Saha, Appl. Catal. B 168–169 (2015)
353–362.
[68] A. Yee, S.J. Morrison, H. Idriss, J. Catal. 186 (1999) 279–295.
[69] Y. Kamimura, S. Sato, R. Takahashi, T. Sodesawa, T. Akashi, Appl. Catal. A 252
(2003) 399–410.
[46] A. Bansode, A. Urakawa, ACS Catal. 4 (2014) 3877–3880.
[47] S.P. Wang, L.F. Zhao, W. Wang, Y.J. Zhao, G.L. Zhang, X.B. Ma, J.L. Gong,
Nanoscale 5 (2013) 5582–5588.
[48] S. Laursen, D. Combita, A.B. Hungria, M. Boronat, A. Corma, Angew. Chem. Int.
Ed. 51 (2012) 4190–4193.
[49] N. Kumar, E. Leino, P. Mäki-Arvela, A. Aho, M. Käldström, M. Tuominen, P.
Laukkanen, K. Eränen, J.P. Mikkola, T. Salmi, D.Yu. Murzin, Microporous
Mesoporous Mater. 152 (2012) 71–77.
[50] M. Aresta, A. Dibenedetto, C. Pastore, A. Angelini, B. Aresta, I. Papai, J. Catal. 269
(2010) 44–52.
[70] E. Otsuka, K. Kanal, T. Sakai, US Pat. 3607937, 1971.
[71] M. Karafian, C.S. Harbor, US Pat. 3668250, 1969.
[72] Y.S. Choi, H. Kim, S.H. Shin, M. Cheong, Y.J. Kim, H.G. Jang, H.S. Kim, J.S. Lee,
Appl. Catal. B 144 (2014) 317–324.
[73] M. Honda, M. Tamura, K. Nakao, K. Suzuki, Y. Nakagawa, K. Tomishige, J. Catal.
318 (2014) 95–107.
[74] M. Tamura, R. Kishi, Y. Nakagawa, K. Tomishige, Nat. Commun. 6 (2015) 8580.
[75] Y.N. Shim, J.K. Lee, J.K. Im, D.K. Mukherjee, D.Q. Nguyen, M. Cheong, H.S. Kim,
Phys. Chem. Chem. Phys. 13 (2011) 6197–6204.
[76] R.W. Taft Jr., J. Am. Chem. Soc. 74 (1952) 3120–3128.
[77] R.W. Taft Jr., J. Am. Chem. Soc. 75 (1953) 4231–4238.
[51] W. Wang, S.P. Wang, X.B. Ma, J.L. Gong, Catal. Today 148 (2009) 323–328.
[52] Z.F. Zhang, Z.T. Liu, Z.W. Liu, J. Lu, Catal. Lett. 129 (2009) 428–436.
Please cite this article in press as: M. Tamura et al., CeO2-catalyzed direct synthesis of dialkylureas from CO2 and amines, J. Catal. (2015), http://dx.doi.org/
10.1016/j.jcat.2015.11.015