N-doped carbon supported Pd catalysts for N-formylation of amines with CO2/H2

Article abstract of DOI:10.1007/s11426-017-9215-2

Using mesoporous N-doped carbons (NCs) derived from glucose and melamine as the supports, a series of Pd/NC catalysts were prepared, in which Pd nanoparticles with average size<2.0 nm were uniformly distributed on the supports. It was indicated that the resultant Pd/NC catalysts were effective for N-formylation of amines with CO₂ and H₂ in ethanol without any additives. Especially, the catalyst Pd/NC-800-6.9% containing quaternary N showed the best performance, affording a series of formylamides in good or even excellent yields. Further investigation reveals that the interaction between the Pd nanoparticles and quaternary nitrogen in the NC support was responsible for the good performance of the catalyst.

Full text of DOI:10.1007/s11426-017-9215-2

SCIENCE CHINA
Chemistry
•
ARTICLES• . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . https://doi.org/10.1007/s11426-017-9215-2
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
N-doped carbon supported Pd catalysts for N-formylation of
amines with CO /H2
2
1
,2
1
1,2
1,2
1,2
1,2
Xiaoying Luo , Hongye Zhang , Zhengang Ke , Cailing Wu , Shien Guo , Yunyan Wu ,
1
1,2*
Bo Yu & Zhimin Liu
1
Beijing National Laboratory for Molecular Sciences, Key Laboratory of Colloid, Interface and Thermodynamics, CAS Research/Education
Center for Excellence in Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China;
2
University of Chinese Academy of Sciences, Beijing 100049, China
Received December 8, 2017; accepted January 30, 2017; published online March 23, 2018
Using mesoporous N-doped carbons (NCs) derived from glucose and melamine as the supports, a series of Pd/NC catalysts were
prepared, in which Pd nanoparticles with average size<2.0 nm were uniformly distributed on the supports. It was indicated that
the resultant Pd/NC catalysts were effective for N-formylation of amines with CO and H in ethanol without any additives
.
2
2
Especially, the catalyst Pd/NC-800-6.9% containing quaternary N showed the best performance, affording a series of for-
mylamides in good or even excellent yields. Further investigation reveals that the interaction between the Pd nanoparticles and
quaternary nitrogen in the NC support was responsible for the good performance of the catalyst.
Pd/NC, C–N formation, N-formylation, CO , H2
2
Citation:
2 2
Luo X, Zhang H, Ke Z, Wu C, Guo S, Wu Y, Yu B, Liu Z. N-doped carbon supported Pd catalysts for N-formylation of amines with CO /H . Sci China
Chem, 2018, 61, https://doi.org/10.1007/s11426-017-9215-2
1
Introduction
cluding Pd/Al O [16], Pd/C [22], Cu/ZnO [23], Pd-Au alloy
2 3
nanoparticles on the polyaniline functionalized carbon na-
As a nontoxic, cheap, abundant, and renewable C1 resource,
notubes (PANI-CNT) [20], and Ir/TiO [21], have been re-
2
carbon dioxide (CO ) has been widely applied in the synth-
ported. However, they generally suffered from complicated
preparation process for catalysts, and requirements of bases
[22], and/or long reaction time [20]. Therefore, developing a
more simple way to prepare the heterogeneous catalysts with
high efficiency for N-formylation of amines with CO /H is
2
esis of chemicals [1–6]. Especially, it has been used as a
formylated reagent instead of toxic phosgene or CO to syn-
thesize formylated compounds [4,7–11]. The reductive N-
formylation of amines with CO2 has been widely in-
vestigated using various reducing agents including silanes,
2
2
still desirable.
and H [12–17], and the catalytic N-formylation of amines
Porous carbon materials have been widely applied in cat-
alysis as well as supports [24], and particularly the carbons
doped with heteroatoms (such as N, P, S, B) show promising
potentials due to their superior performances [25–30]. For
example, the N, P, S-codoped carbon catalyst showed high
efficiency for the selective oxidation of aromatic alkanes in
aqueous solution [31]. And the Pd/N-CHT (Pd over N-doped
carbon hollow tubes) showed high catalytic activity in C–C
bond formation via cyanation and Sonogashira reactions, and
2
with CO /H is a green way to produce N-formylamides,
2
2
which have been achieved over homogeneous and hetero-
geneous catalysts [12,16,18–21]. For convenience of re-
cycling catalyst and separating products, heterogeneous
catalysts generally behave obvious advantages compared to
homogeneous catalysts. The heterogeneous catalysts in-
*Corresponding author (email: liuzm@iccas.ac.cn)
©
Science China Press and Springer-Verlag GmbH Germany 2018 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . chem.scichina.com link.springer.com

2
. . . . . . . . . . . . . . . . . . . . . . . . . . . . Luo et al. Sci China Chem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
hydrogenation of various aromatic alkenes, alkynes and nitro
compounds [32]. In our previous work, mesoporous N-doped
carbon (NC) materials were prepared using the eutectic salts
as the soft template via calcining the mixture of glucose/
cellulose and melamine, which showed high surface areas
and high activity for catalyzing the oxidation of ethyl ben-
zene. In addition, they can be used to support the noble
metals (such as Pd, Ru, Rh, Pt) [33].
creased to the final calcining temperature (400, 600, 800 °C),
and kept for 4 h. A series of mesoporous NCs were prepared
via calcining the mixtures of glucose and melamine with the
mass ratio of 2:1 at 400, 600, 800 °C and of 3:1, 20:1 at
800 °C. The resultant NCs were denoted as NC-T-x, where T
represents for the calcining temperature and x for the N
content. In comparison, mesoporous C-800 was prepared
using glucose as the carbon resource at 800 °C.
To explore the application of the NC supported metal
catalysts, a series of Pd/NC nanocatalysts were prepared for
the N-formylation of amines with CO /H in this work. The
In a typical experiment to prepare Pd/NC-T-x with Pd
content of 2.0 wt%, 4.2 mg of Pd(OAc) was dissolved in
2
20 mL of CH Cl , and then was added into 100 mg of NC-T-
2
2
2
2
resultant NC-supported Pd catalysts were characterized by
means of different techniques including transmission elec-
tron microscopy (TEM), X-ray diffraction (XRD), X-ray
photoelectron spectroscopy (XPS), Brunauer-Emmett-Teller
x under magnetically stirring at room temperature. After
12 h, CH Cl was removed completely via distillation under
2
2
reduced pressure, and the resultant solid mixture was re-
duced under the atmosphere of the hydrogen-argon mixture
(
BET) N sorption, element analysis, etc. The activities of the
with the flow rates of 5 mL/min for H and 30 mL/min for
2
2
catalysts for the N-formylation of amines with CO /H were
argon at 250 °C for 3 h. Using different mesoporous NCs or
C-800 as the supports, various NCs or C-800 supported Pd
nanocatalysts with Pd content of 2.0 wt% were obtained.
2
2
investigated under various conditions. As far as we know, the
nitrogen doped carbons supported Pd nanoparticles were first
used in N-formylation of amines with CO /H . Under the
2
2
optimized conditions, the substrate scope was explored, and
a series of N-formylamides were obtained in good or even
excellent yields. Moreover, the reaction mechanism was
explored as well.
2
.3 Characterization
The elemental analysis was conducted on a Flash EA1112
elemental analyzer. XRD analysis was performed on a D/
MAX-RC diffractometer operating at 30 kV and 100 mA
with Cu Kα radiation (λ=1.5418 Å), in range of 3°–80° (2θ)
with a step width of 0.02°. Nitrogen adsorption-desorption
curves were collected by Brunauer-Emmett-Teller (BET)
2
Experimental
2
.1 Materials
(
Quantachrome Autosorb-iQ). The total surface area (Stotal)
Carbon dioxide (99.99%) and hydrogen (99.999%) were
purchased from Beijing Analytical Instrument Company
was calculated using the Brunauer-Emmett-Teller (BET)
method from the nitrogen adsorption data in the relative
pressure (P/P ) of 0.03–0.30. The pore-size-distribution
curves were obtained from the adsorption branches using
density functional theory (DFT) method [34,35]. The value
of the total pore volume was determined at the relative
pressure P/P =0.99. TEM images were obtained with a JEOL
JEM-2100F instrument (Japan) operated at 200 kV. XPS
analysis was performed on an ESCAL Lab 220i-XL spec-
trometer at a pressure about 3×10 mbar using Al Kα as the
excitation source (1486.6 eV) and operated at 15 kV and
(China). Morpholine and different kinds of amine substrates
0
were purchased from J&K Scientific Ltd. (China) and Alfa
Aesar (USA). Metal salts Pd(OAc) was provided by the Alfa
2
Aesar (USA), and glucose, melamine, KCl and ZnCl were
2
purchased from Sinopharm Chemical Reagent Beijing Co.,
Ltd. (China). Ethanol and other solvents were provided by
Beijing Chemical Company (China). The deuterated solvents
0
(
CDCl ) were purchased from J&K Scientific Ltd. (China).
3
−9
All solvents and reagents were used without further pur-
ification.
2
0 mA. The binding energies were referenced to the C 1s line
at 284.8 eV from adventitious carbon.
2
.2 Preparation of the catalysts
Typically, glucose and melamine acted as the carbon and
nitrogen resources, respectively, and the eutectic salt of
ZnCl and KCl (with the molar ratio of ZnCl :KCl=51:49)
2
H2
.4 Procedures for N-formylation of amines with CO2/
2
2
acted as the porogen. First, a certain ratio of glucose, mela-
mine and porogen were mixed well by grinding at room
temperature, and then heated with a gradient temperature
program under argon atmosphere in a tube furnace. The
mixture was first heated at 240 °C for 4 h to make the glu-
cose and melamine dissolve in eutectic salt uniformly and
form cross-linked substances. Then the temperature was in-
Typically, 20 mg of 2.0 wt% Pd/NC-T-x catalyst (0.4 mol%
Pd relative to morpholine), 1.0 mmol of morpholine, and
2 mL of ethanol as the solvent were loaded in sequence into a
25 mL stainless steel autoclave with a Teflon tube. The
sealed autoclave was charged with H at 4 MPa, and then
2
charged with CO up to the total pressure of 7 MPa at room
2
temperature. Subsequently, the reactor was moved into an

3
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Luo et al. Sci China Chem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
oil-bath of 130 °C and stirred for desired time. Then it was
cooled to room temperature, and was released the gas. Then
n-dodecane was added to the reaction solution as an internal
standard for determining the amine conversion and product
yield by gas chromatography with a FID detector and a
nonpolar capillary column (DB-5) (30 m×0.25 mm
46.79, 40.56, 35.34, 33.89, 31.99, 22.35.
N-hexylformamide. H NMR (400 MHz, CDCl ) δ (ppm)
=8.14 (s, 1H), 5.72 (s, 1H), 3.21–3.31 (t, 2H), 1.48–1.53 (m,
2H), 1.24–1.28 (m, 6H), 0.85–0.89 (t, 3H). C NMR
1
3
1
3
(100.6 MHz, CDCl ) δ (ppm)=161.17, 38.21, 31.41, 29.48,
3
26.50, 22.52, 13.98.
N-sec-butylformaide. H NMR (400 MHz, CDCl ) δ (ppm)
1
×
0.25 μm) (GC, Agilent 7890B, USA). High purity nitrogen
3
1
13
was used as a carrier gas. H NMR and C NMR analyses
=8.02, 8.03 (s, 1H), 3.20–3.23 (m, 1H), 1.65 (s, 1H), 1.49–
1.65 (m, 2H), 1.22–1.33 (m, 3H), 0.92–0.95 (m, 3H).
13
were conducted on a Bruker Avance NMR (400 MHz). The
C
1
13
NMR (100.6 MHz, CDCl ) δ (ppm)=160.56, 45.51, 30.78,
H and C NMR data of the resultant compounds are listed
as follows.
3
2
9.58, 22.13, 20.41, 10.26.
N-benzyl-N-methylformamide.
1
1
H
NMR (400 MHz,
N-formylmorpholine. H NMR (400 MHz, CDCl ) δ
ppm)=8.05 (s, 1H), 3.70–3.63 (m, 4H), 3.58–3.55 (t, 2H),
.40–3.37 (t, 2H). C NMR (100.6 MHz, CDCl ) δ (ppm)
3
3
CDCl ) δ (ppm)=8.02, 7.89 (s, 1H), 7.28–6.93 (m, 5H), 4.26
(
3
3
1
3
13
(
m, 1H), 4.13 (s, 1H), 2.58 (s, 3H). C NMR (100.6 MHz,
=
161.52, 67.94, 67.15, 46.50, 41.33.
N-formylpyrrolidine. H NMR (400 MHz, CDCl ) δ (ppm)
CDCl ) δ (ppm)=162.56, 128.91, 128.70, 128.26, 128.11,
3
1
127.64, 127.40, 53.49, 47.79, 34.02, 29.46.
3
1
3
1
=
8.24 (s, 1H), 3.49–3.38 (m, 4H), 1.95–1.86 (m, 4H).
C
N-benzylformamide. H NMR (400 MHz, CDCl
3
) δ (ppm)
NMR (100.6 MHz, CDCl ) δ (ppm)=160.67, 45.80, 42.90,
=8.17 (s, 1H), 7.24–7.32 (m, 5H), 4.41–4.43 (d, 2H), 6.40 (s,
3
1
3
2
4.69, 24.02.
-Formyl-4-methylpiperazine.
CDCl ) δ (ppm)=8.00 (s, 1H), 3.54 (m, 2H), 3.35–3.38 (m,
1H). C NMR (100.6 MHz, CDCl
3
) δ (ppm)=161.22,
1
H
1
NMR (400 MHz,
128.75, 127.75, 127.63, 42.36.
1
N-phenylformamide. H NMR (400 MHz, CDCl
) δ (ppm)
3
3
1
3
13
2
H), 2.38–2.40 (t, 2H), 2.33–2.34 (t, 2H), 2.29 (s, 3H).
C
=8.37 (s, 1H), 7.09–7.56 (m, 5H). C NMR (100.6 MHz,
CDCl ) δ (ppm)=162.77, 159.12, 129.78, 129.13, 125.31,
NMR (100.6 MHz, CDCl ) δ (ppm)=160.87, 55.55, 54.38,
3
3
4
6.28, 45.68, 40.01.
N-formylpiperidine. H NMR (400 MHz, CDCl ) δ (ppm)
124.85, 120.03, 118.84.
N-methyl-N-phenylformamide.
1
1
H
NMR (400 MHz,
3
=
7.89 (s, 1H), 3.38–3.35 (t, 2H), 3.22–3.19 (t, 2H), 1.61–
CDCl ) δ (ppm)=8.36 (s, 1H), 7.05–7.32 (m, 5H), 3.20 (s,
3
13
13
1
.55 (m, 2H), 1.50–1.40 (m, 4H). C NMR (100.6 MHz,
3H). C NMR (100.6 MHz, CDCl
129.60, 126.35, 123.31, 31.97.
) δ (ppm)=162.25,
3
CDCl ) δ (ppm)=160.77, 46.77, 40.55, 26.51, 25.02, 24.62.
3
1
N,N-dipropylformamide. H NMR (400 MHz, CDCl ) δ
3
(
1
ppm)=7.90 (s, 1H), 3.11–3.08 (t, 2H), 3.04–3.00 (t, 2H),
13
.46–1.36 (m, 4H), 0.76–0.72 (m, 6H).
C NMR
3 Results and discussion
(
2
100.6 MHz, CDCl ) δ (ppm)=162.63, 48.99, 43.59, 21.69,
3
3
.1 Characterization of the Pd catalysts
0.40, 11.03.
N,N-dihexylformamide. H NMR (400 MHz, CDCl ) δ
1
The N contents of the resultant NC-T-x were first examined
by elemental analysis. With the same mass ratio of glucose to
melamine at 2:1, the N contents in NC-T-x decreased with
the calcining temperature increase, and the sample prepared
at 800 °C possessed an N content of 6.9%. The sample
prepared with the mass ratio of glucose to melamine at 3:1
and 20:1 at 800 °C possessed N contents at 4.9% and 1.0%,
respectively.
The XRD patterns of NC-800-6.9%, Pd/NC-800-6.9%, C-
800, Pd/C-800 are shown in Figure 1(a). Both NC-800-6.9%
and C-800 showed two broad peaks at about 2θ=23° and 44°
with low intensity, which might be attributed to the partly
graphitic carbon [36,37]. No obvious characteristic diffrac-
tion peaks assigning to Pd particles were detected in the
XRD patterns of Pd/NC-800-6.9% and Pd/C-800, pre-
sumably due to the tiny size of Pd particles with the weak
crystallization, as well as low loading contents [38]. Figure
1(b) shows a TEM image of Pd/NC-800-6.9% with corre-
sponding histogram of particles distribution. It is clear that
the Pd nanoparticles with average size<2.0 nm were uni-
3
(ppm)=7.99 (s, 1H), 3.21–3.25 (t, 2H), 3.13–3.16 (t, 2H),
1
3
1
.47–1.48 (m, 4H), 1.25 (m, 12H), 0.84–0.86 (m, 6H). C
NMR (100.6 MHz, CDCl ) δ (ppm)=161.66, 46.46, 41.14,
3
1
3
0.54, 30.39, 27.66, 26.29, 25.63, 25.16, 21.56, 21.53, 12.99,
2.95.
1
N-cyclopentylformamide. H NMR (400 MHz, CDCl ) δ
3
(ppm)=7.89 (s, 1H), 3.69–3.64 (m, 1H), 1.84–1.76 (m, 2H),
13
1
.56–1.41 (m, 4H), 1.32–1.25 (m, 2H). C NMR (100.6,
MHz CDCl ) δ (ppm)=163.95, 54.02, 32.56, 32.02, 23.12,
2
3
3.00.
N-cyclohexylformamide. H NMR (400 MHz, CDCl ) δ
1
3
(ppm)=8.09 (s, 1H), 5.74–5.55 (s, 1H), 3.89–3.81 (m,
13
0
(
2
.76H), 3.26-3.33 (m, 0.28H), 1.11–1.94 (m, 10H). C NMR
100.6 MHz, CDCl ) δ (ppm)=161.97, 47.09, 34.71, 33.05,
3
5.43, 24.73.
1
N-butylformamide. H NMR (400 MHz, CDCl ) δ (ppm)
3
=
8.13 (s, 1H), 5.74 (s, 1H), 3.24–3.29 (m, 1.6H), 3.16–3.21
(m, 0.4H), 1.47–1.50 (m, 2H), 1.31–1.36 (m, 2H), 0.88–0.92
13
(m, 3H). C NMR (100.6 MHz, CDCl ) δ (ppm)=161.4,
3

4
. . . . . . . . . . . . . . . . . . . . . . . . . . . . Luo et al. Sci China Chem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Table 1 Pore textural properties of catalysts
2
−1 a)
3
−1 a)
Catalyst
NC-800-6.9%
Pd/NC-800-6.9%
NC-800-4.9%
Pd/NC-800-4.9%
NC-600-14.3%
Pd/NC-600-14.3%
NC-400-16.8%
Pd/NC-400-16.8%
C-800
S
total (m g )
1483
1420
1325
1277
994
V
total (cm g )
1.46
1.21
1.35
1.30
0.93
760
0.78
476
0.81
450
0.73
1310
1249
2.14
Pd/C-800
2.09
a) Stotal: total BET specific surface area; Vtotal: total pore volume.
Figure 1 (a) XRD diffraction patterns; (b) TEM image and the histogram
2
(inset) of Pd particles distribution of Pd/NC-800-6.9%; (c) N sorption
isotherm of Pd/NC-800-6.9% and NC-800-6.9%; (d) the distribution of
pore size of Pd/NC-800-6.9% (color online).
formly distributed on the surface of the NC-800-6.9% sup-
port. The other Pd/NC-T-x showed similar morphology with
Pd particle size<2.0 nm.
Figure 1(c, d) shows the N adsorption-desorption iso-
2
therms of Pd/NC-800-6.9% and NC-800-6.9% and the pore
size distribution of Pd/NC-800-6.9%. Obviously, Pd/NC-
8
00-6.9% and NC-800-6.9% exhibited type IV sorption
isotherms, a typical feature of mesoporous materials [35,39],
whose pore diameter were centered on 3.0 nm according to
the DFT model. The immobilization of Pd nanoparticles onto
the nitrogen doped carbon supports resulted in the reduction
in BET surface areas and pore volumes of the catalysts. This
was probably due to the slight blockage of the cavities by the
metal particles [40]. The pore textural properties of all the
catalysts are listed in Table 1. It is obvious that the calcining
temperature remarkably influenced surface areas and pore
volumes of the resultant carbons, and the NCs obtained at
8
00 °C had higher surface areas with larger pore volumes.
The oxidation state of Pd species and the types of the
structural nitrogen in the Pd/NC-T-x were investigated by
means of XPS, and the results are shown in Figure 2. The N
1
s peak in Pd/NC-800-6.9% (Figure 2(a)) can be deconvo-
luted into three ones with BEs at 398.4, 399.4 and 400.7 eV,
which are ascribed to pyrrolic, pyridinic and quaternary N,
respectively, indicating the presence of these three types of N
Figure 2 N 1s XPS spectra of (a) Pd/NC-800-6.9%, (c) Pd/NC-600-
[
41–43]. The presence of Pd(0) on NC-800-6.9% was con-
firmed by the peaks with BEs of 335.7 and 340.9 eV for Pd
d5/2 and Pd 3d3/2, respectively, and Pd (II) was detected
with the BEs at 337.8 and 343.0 eV [44]. The presence of Pd
II) may be caused by the oxidation of Pd nanoparticles
exposed to air. Compared to NC-800-6.9%, Pd/NC-800-
.9% showed lower BE value of quaternary N (400.7 eV for
14.3%, (e) Pd/NC-400-16.8% and (g) NC-800-6.9%; Pd 3d XPS spectra of
(b) Pd/NC-800-6.9%, (d) Pd/NC-600-14.3%, (f) Pd/NC-400-16.8% and (h)
Pd/C-800 (color online).
3
(
trogen. Moreover, the BEs of Pd 3d5/2 and Pd 3d3/2 in Pd/
NC-800-6.9% became larger compared to those in Pd/C-800
(Figure 2(b, h)). This implies the Pd particles might tend to
nucleate and grow at the quaternary nitrogen sites on the NC-
800-6.9% support [39].
6
Pd/NC-800-6.9%, 401.2 eV for NC-800-6.9% (Figure 2(a,
g)), together with similar BEs of pyrrolic and pyridinic ni-

5
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Luo et al. Sci China Chem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Table 2 Reaction conditions optimization of N-formylation of morpho-
line
Compared to that of bulk Pd, the Pd 3d BEs of the resultant
a)
Pd catalysts shifted towards higher values, probably due to
the tiny size of the Pd particles [45,46]. Notably, the shift for
Pd/NC-800-6.9% was more than those of Pd/NC-400-16.8%
and Pd/NC-600-14.3% though they had the similar average
size of the Pd particles (1.62 and 1.56 nm for Pd/NC-400-
1
6.8% and Pd/NC-600-14.3%, respectively). This further
Entry
Catalyst
Amount (mg)
Solvent
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
MeOH
Yield (%)
implies the interaction between the Pd nanoparticle and
quaternary nitrogen.
1
2
Pd/NC-800-6.9%
Pd/C-800
20
20
20
20
20
20
20
20
20
20
10
25
50
50
93
46
45
68
81
72
86
56
64
6
3
Pd/NC-400-16.8%
Pd/NC-600-14.3%
Pd/NC-800-4.9%
Pd/NC-800-1.0%
Pd/NC-800-6.9%
Pd/NC-800-6.9%
Pd/NC-800-6.9%
Pd/NC-800-6.9%
Pd/NC-800-6.9%
Pd/NC-800-6.9%
3
.2 N-formylation of amines with CO /H2
2
4
5
The resultant Pd/NC-T-x catalysts were investigated for N-
formylation of amines with CO and H . The N-formylation
of morpholine was chosen as the model reaction to screen the
catalysts and optimize the reaction conditions. Pd/NC-800-
6
2
2
7
8
3
CH CN
9
1,4-Dioxane
DMSO
EtOH
6
.9% showed the best catalytic activity for this reaction,
affording a product yield of 93% under the experimental
conditions as listed in Table 2, and no byproduct was de-
tectable. In contrast, Pd/C-800 gave a product yield of 46%
10
11
68
96
96
12
EtOH
(Table 2, entry 2), much lower than that obtained over Pd/
1
3 [16]
4 [22]
1 wt% Pd/Al
2
O
3
Octane
MeOH
NC-800-6.9%. Considering that the Pd contents and the
average sizes of Pd nanoparticles in these two samples are
similar, the difference in their catalytic activities may result
from the difference in the catalyst supports. That is, the ni-
trogen-doping in the support of Pd/NC-800-6.9% con-
siderably improved the activity of the Pd catalysts. To further
explore the influence of N-doping in the catalyst supports on
the reaction, Pd/NC-400-16.8% and Pd/NC-600-14.3% were
also examined. It was indicated that they were also effective
for this reaction, but showing lower activity than Pd/NC-
b)
1
1 wt% Pd/C
93
1
.6 wt%Pd-3 wt%Au/
c)
15 [20]
50
1,4-Dioxane
90
PANI-CNT
a) Reaction conditions: 1.0 mmol morpholine, 20 mg catalyst (2 wt%),
2 mL solvent, P(H )=4 MPa, total presure of 7 MPa after charging CO ,
30 °C,24 h; b) The yield of formylpiperidine; c) 48 h.
2
2
1
reaction, and ethanol showed the best performance, while
dimethyl sulfoxide (DMSO) only afforded a little product
yield. To examine the role of methanol in the formylation of
morpholine with CO /H catalyzed by Pd/NC-800-6.9%, we
8
00-6.9% though they had higher nitrogen contents (Table 2,
2
2
entries 3, 4). Especially, Pd/NC-400-16.8% showed similar
activity to Pd/C-800, implying that the N in the Pd/NC-400-
performed two control experiments. One was carried out in
the absence of hydrogen, and a formylmorpholine yield of
4% was obtained. Another experiment was carried out with
deuterated-methanol as the solvent, and deuterium hydro-
gen-containing N-formylmorpholine was detected. These
results indicate that the solvent effect of methanol or ethanol
was related to Pd-promoted transfer hydrogenation [47],
which promoted the formylation of morpholine.
The yield of N-formylmorpholine increased with the cat-
alyst amounts under the similar conditions, and it changed
little as the amount was increased to 25 mg. As for the in-
fluence of reaction time, the N-formylmorpholine yield was
sharply increased from 40% to 93% as the time was extended
from 6 to 24 h. Further prolonging time to 36 h, the N-for-
mylmorpholine yield reached to 98% (Figure 3(b)). The in-
1
6.8% support hardly impact the catalyst activity. The major
difference in the states of N in NC-800-6.9% and NC-400-
6.8% is that NC-800-6.9% had quaternary N while NC-400
1
did not. Therefore, it can be deduced that the presence of
quaternary N in the catalyst supports is favorable to im-
proving the activity of the catalysts. To confirm this deduc-
tion, Pd/NC-800-4.9% and Pd/NC-800-1.0% were examined
for catalyzing the reaction, which afforded an N-for-
mylmorpholine yield of 81% and 72% (Table 2, entries 5, 6).
These results further indicated that the quaternary N in the
NC-800 supports indeed facilitated the activity of the Pd
catalysts.
To optimize the reaction conditions, the influences of
temperature, reaction time, solvents, pressure, and catalyst
amount on the reaction were investigated, and the results are
shown in Table 2 and Figure 3. It was demonstrated that the
N-formylmorpholine yield increased with temperature, and
fluence of pressure was investigated, and the H initial
2
pressure of 4 MPa with total pressure of 7 MPa afforded
appropriate product yield (Figure 3(c)). From the above ex-
periments, the optimized conditions were obtained as fol-
lows: Pd/NC-800-6.9% as the catalyst, ethanol as the
solvent, 4 MPa H and charging CO to total pressure of
1
40 °C was a suitable temperature to achieve high product
yield (Figure 3(a)). The solvents significantly influenced the
2
2

6
. . . . . . . . . . . . . . . . . . . . . . . . . . . . Luo et al. Sci China Chem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
a)
Table 3 Results of N-formylation of amines
Figure 3 The influence of temperature (a), time (b) and pressure (c) on
the yield.
7
MPa, 130 °C, 24 h. Under the optimal conditions, an N-
formylmorpholine yield of 93% was obtained without any
additives, better than those obtained over Pd/C in the pre-
sence of base (Table 2, entry 14), and over 1.6 wt% Pd-3 wt%
Au/PANI-CNT (Table 2, entry 15).
To test the reusability of Pd/NC-800-6.9%, the used cata-
lyst was separated from the reaction solution, and washed
with ethanol for several times, followed by being used in the
next run. After reused for three times, the product yield de-
creased from 93% to 81%. TEM observation indicated that
the Pd nanoparticle size became larger from around 2 to
3
.5 nm, suggesting that under the experimental conditions
the Pd nanoparticle aggregated into larger particles. There-
fore, it is very important to prevent Pd particles from ag-
gregating to achieve stable activity of the catalyst at high
temperature.
Under the optimized conditions, the substrate scope was
explored, and the results are summarized in Table 3. Cyclic
secondary amines could react with CO /H well, producing
2
2
the desired products in 80%–94% yields (Table 2, entries 1–
). Secondary amines, such as di-n-propylamine and di-n-
4
a) Reaction conditions: 1.0 mmol amine, 20 mg catalyst, 2 mL ethanol,
CO /H =3:4 (P/P), 130 °C, 24 h; b) the isolated yield.
2
2
hexylamine achieved excellent product yields (Table 2, en-
tries 5, 6). Primary amines also proceeded well, and mono-
formylated amines were obtained. For example, cyclohex-
ylamine, cyclopentylamine and iso-butylamine afforded ex-
cellent product yields, while n-butylamine and n-hexylamine
displayed lower reactivity, producing the corresponding
formylamines in moderate yields. Inspired by the results of
primary aliphatic amines, we further investigated the N-
formylation of aromatic amines. To our delight, benzylamine
and N-methylbenzylamine could efficiently be converted
into the desired products, and N-methylbenzylamine showed
higher reactivity (Table 2, entries 12, 13). Aniline and N-
methylaniline were also examined, but they showed less
reactivity under the experimental conditions (Table 2, entries
14, 15).
To explore the reaction mechanism of formylation of
morpholine over Pd/NC-800-6.9%, control experiments
were performed. Firstly the reaction of morpholine with CO
was conducted in the presence of Pd/NC-800-6.9% at room
temperature and pressure of 3 MPa for 4 h. Then CO was
released, and H was charged into the reactor up to 7 MPa.
After being stirred at 130 °C for 20 h, N-formylmorpholine
was obtained in a yield of 20%. Based on the above results, a
possible reaction mechanism was proposed as illustrated in
2
2
2
Scheme 1. Morpholine first reacts with CO
formate intermediate, which further reacts with hydrogen
to form an N-
2

7
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Luo et al. Sci China Chem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1
1
Zhao TX, Zhai GW, Liang J, Li P, Hu XB, Wu YT. Chem Commun,
017, 53: 8046–8049
Schmid L, Canonica A, Baiker A. Appl Catal A-Gen, 2003, 255: 23–
2
12
13
14
15
33
Nguyen TVQ, Yoo WJ, Kobayashi S. Angew Chem Int Ed, 2015, 54:
209–9212
9
Jacquet O, Das Neves Gomes C, Ephritikhine M, Cantat T. J Am
Chem Soc, 2012, 134: 2934–2937
Molla RA, Bhanja P, Ghosh K, Islam SS, Bhaumik A, Islam SM.
ChemCatChem, 2017, 9: 1939–1946
1
1
6
7
Cui X, Zhang Y, Deng Y, Shi F. Chem Commun, 2014, 50: 189–191
Yang ZZ, Yu B, Zhang H, Zhao Y, Ji G, Liu Z. RSC Adv, 2015, 5:
19613–19619
18
19
20
21
22
23
24
Federsel C, Boddien A, Jackstell R, Jennerjahn R, Dyson PJ, Scopelliti
R, Laurenczy G, Beller M. Angew Chem Int Ed, 2010, 49: 9777–9780
Zhang L, Han Z, Zhao X, Wang Z, Ding K. Angew Chem Int Ed, 2015,
5
4: 6186–6189
Ju P, Chen J, Chen A, Chen L, Yu Y. ACS Sustain Chem Eng, 2017, 5:
516–2528
Bi QY, Lin JD, Liu YM, Xie SH, He HY, Cao Y. Chem Commun,
014, 50: 9138–9140
Zhang Y, Wang H, Yuan H, Shi F. ACS Sustain Chem Eng, 2017, 5:
758–5765
Liu H, Mei Q, Xu Q, Song J, Liu H, Han B. Green Chem, 2017, 19:
96–201
Yang L, Zhao Y, Chen S, Wu Q, Wang X, Hu Z. Chin J Catal, 2013,
4: 1986–1991
2
2
Scheme 1 The plausible mechanism for Pd-catalysed N-formylation of
morpholine (color online).
5
1
catalyzed by Pd/NC-800-6.9% to form the final product.
3
2
2
2
5
6
7
Kicinski W, Szala M, Bystrzejewski M. Carbon, 2014, 68: 1–32
Li R, Wei Z, Gou X. ACS Catal, 2015, 5: 4133–4142
Ma X, Ning G, Qi C, Xu C, Gao J. ACS Appl Mater Interfaces, 2014,
4
Conclusions
6
: 14415–14422
In conclusion, a Pd nanocatalyst supported on quaternary N-
doped mesoporous carbon material has been developed,
which shows high efficiency for formylation of amines
without any other additives. It was indicated that the nitrogen
doping changed the electronic properties of carbon material,
and have a good correlation with Pd particles, which is
helpful to enhance the activity of the catalyst for N-for-
mylation of amines. This work provides a catalyst with high
2
8
Patel MA, Luo F, Khoshi MR, Rabie E, Zhang Q, Flach CR, Men-
delsohn R, Garfunkel E, Szostak M, He H. ACS Nano, 2016, 10:
2305–2315
29
30
31
32
Wang S, Iyyamperumal E, Roy A, Xue Y, Yu D, Dai L. Angew Chem
Int Ed, 2011, 50: 11756–11760
Xu J, Zhao Y, Shen C, Guan L. ACS Appl Mater Interfaces, 2013, 5:
12594–12601
Yang S, Peng L, Huang P, Wang X, Sun Y, Cao C, Song W. Angew
Chem Int Ed, 2016, 55: 4016–4020
Modak A, Bhaumik A. J Mol Catal A-Chem, 2016, 425: 147–156
performance for N-formylation of amines with CO and H ,
2
2
33 Ma Z, Zhang H, Yang Z, Ji G, Yu B, Liu X, Liu Z. Green Chem, 2016,
which may have a great potential in application.
18: 1976–1982
3
4
Corma A, García H, Navarro MT, Palomares EJ, Rey F. Chem Mater,
000, 12: 3068–3072
2
Acknowledgements This work was supported by the Chinese Academy of
Sciences (QYZDY-SSW-SLH013) and the National Natural Science
Foundation of China (21533011, 21503239).
3
3
5
6
Tanev PT, Chibwe M, Pinnavaia TJ. Nature, 1994, 368: 321–323
Li R, Cao A, Zhang Y, Li G, Jiang F, Li S, Chen D, Wang C, Ge J,
Shu C. ACS Appl Mater Interfaces, 2014, 6: 20574–20578
Wen Z, Ci S, Hou Y, Chen J. Angew Chem Int Ed, 2014, 53: 6496–
6500
37
Conflict of interest The authors declare that they have no conflict of
interest.
3
3
8
9
Yu J, Hai Y, Cheng B. J Phys Chem C, 2011, 115: 4953–4958
Molla RA, Iqubal MA, Ghosh K, Islam SM. Green Chem, 2016, 18:
1
2
Álvarez A, Bansode A, Urakawa A, Bavykina AV, Wezendonk TA,
Makkee M, Gascon J, Kapteijn F. Chem Rev, 2017, 117: 9804–9838
Zhu Q, Ma J, Kang X, Sun X, Liu H, Hu J, Liu Z, Han B. Angew
Chem Int Ed, 2016, 55: 9012–9016
4649–4656
4
4
4
4
4
4
0
1
2
3
4
5
Yang Z, Yu B, Zhang H, Zhao Y, Chen Y, Ma Z, Ji G, Gao X, Han B,
Liu Z. ACS Catal, 2016, 6: 1268–1273
Lai L, Potts JR, Zhan D, Wang L, Poh CK, Tang C, Gong H, Shen Z,
Lin J, Ruoff RS. Energy Environ Sci, 2012, 5: 7936–7942
Li Y, Zhao Y, Cheng H, Hu Y, Shi G, Dai L, Qu L. J Am Chem Soc,
3
4
5
6
7
8
Li Y, Chan SH, Sun Q. Nanoscale, 2015, 7: 8663–8683
Dong K, Razzaq R, Hu Y, Ding K. Top Curr Chem, 2017, 375: 23
Sakakura T, Choi JC, Yasuda H. Chem Rev, 2007, 107: 2365–2387
Cui X, Dai X, Zhang Y, Deng Y, Shi F. Chem Sci, 2014, 5: 649–655
Bobbink FD, Das S, Dyson PJ. Nat Protoc, 2017, 12: 417–428
Daw P, Chakraborty S, Leitus G, Diskin-Posner Y, Ben-David Y,
Milstein D. ACS Catal, 2017, 7: 2500–2504
2012, 134: 15–18
Luo Z, Lim S, Tian Z, Shang J, Lai L, MacDonald B, Fu C, Shen Z,
Yu T, Lin J. J Mater Chem, 2011, 21: 8038–8044
Perini L, Durante C, Favaro M, Agnoli S, Granozzi G, Gennaro A.
Appl Catal B-Environ, 2014, 144: 300–307
9
0
Liu XF, Li XY, Qiao C, Fu HC, He LN. Angew Chem Int Ed, 2017,
He DP, Jiang YL, Lv HF, Pan M, Mu SC. Appl Catal B-Environ, 2013,
56: 7425–7429
132: 379–388
1
Mitsudome T, Urayama T, Fujita S, Maeno Z, Mizugaki T, Jitsukawa
K, Kaneda K. ChemCatChem, 2017, 9: 3632–3636
4
4
6
7
Wertheim GK, Dicenzo SB. Phys Rev B, 1988, 37: 844–847
Creutz C, Chou MH. J Am Chem Soc, 2009, 131: 2794–2795