Catalytic amination of glycerol with dimethylamine over different type ofheteropolyacid/Zr-MCM-41 catalysts

Article abstract of DOI:10.1016/j.mcat.2018.07.017

The effect of different type of heteropolyacid/Zr-MCM-41 catalysts on the catalytic amination of glycerol with dimethylamine to produce Dimethylamino-3-propanal was researched. Under the premise of their respective optimum loading amount, the specific sur

Full text of DOI:10.1016/j.mcat.2018.07.017

MolecularCatalysis457(2018)51–58
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Catalytic amination of glycerol with dimethylamine over different type
ofheteropolyacid/Zr-MCM-41 catalysts
Jianfei Dinga,⁎,1, Mingyu Cuia,1, Tianlin Mab,⁎, Rong Shaoa, Wei Xua, Pengfei Wangc
a
School of Chemistry and Chemical Engineering, Yancheng Institute of Technology, Yancheng 224051, Jiangsu, China
School of Material Science and Chemical Engineering, Chuzhou University, Chuzhou 239000, Anhui, China
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, Shanxi, China
b
c
A R T I C L E I N F O
A B S T R A C T
Keywords:
Catalytic amination
Glycerol
Dimethylamine
Heteropolyacid
Zr-MCM-41
The effect of different type of heteropolyacid/Zr-MCM-41 catalysts on the catalytic amination of glycerol with
dimethylamine to produce Dimethylamino-3-propanal was researched. Under the premise of their respective
optimum loading amount, the specific surface area, pore volume and pore size of the H6P2W18O62/Zr-MCM-
41(Cat-2) were higher than those of the H3PW12O40/Zr-MCM-41(Cat-1). Cat-1 exhibited more total acidity than
that of Cat-2, which resulted in the higher glycerol conversion of Cat-1 at the beginning of the reaction, but the
coke deposited on Cat-1 was more than that of Cat-2. The amount of Brønsted acid sites and ratio of B/(B + L) on
Cat-2 were higher than that of Cat-1. Therefore, the catalytic stability, selectivity and yield of Dimethylamino-3-
propanal of Cat-2 were higher than those of Cat-1, which proved that the Brønsted acid sites are good for the
catalytic amination of glycerol with dimethylamine to produce Dimethylamino-3-propanal. The leaching ratio of
Cat-2 is slightly more than that of Cat-1, because H6P2W18O62 is more hydrophilic than H3PW12O40. However,
the stability of Cat-2 is higher than that of Cat-1, therefore, the main reason for deactivation of catalyst is the
formation of carbon deposit during this reaction. And the reaction pathway of catalytic amination of glycerol
with dimethylamine was proposed. Furthermore, the reaction temperature, GHSV and the molar ratio of di-
methylamine/glycerol were optimized on Cat-2 catalyst. Under the optimum reaction conditions: 300 °C,
H6P2W18O62 loading of 20 wt%, GHSV of 3 h−1, glycerol aqueous concentration of 10 wt%, dimethylamine
aqueous concentration of 30 wt%, molar ratio of dimethylamine/glycerol of 2.5, the glycerol conversion, se-
lectivity and yield of Dimethylamino-3-propanal were 95.6%, 80.8% and 77.2%, respectively.
Introduction
amine chemical product, which can be used as surfactants, pharma-
ceuticals, pesticides, fuel additives, and so on. However, as far as we
Biodiesel is a very significant renewable energy source, which has
received increasing attention for the replacement of fossil resources.
With the rapid development of biodiesel, a huge quantity of byproduct
glycerol is generated. Generally, 10 wt% of glycerol can be generated
during the production of biodiesel. Therefore, it is urgent to transform
glycerol into valuable chemicals. So many relevant researches have
been done, such as dehydration of glycerol to acrolein [1,2], reforming
of glycerol to hydrogen [3], hydrogenation of glycerol to 1,2-propa-
nediol [4], oxidation of glycerol to glyceric acid [5], esterification of
glycerol to monoacylglycerol [6], and catalytic amination of glycerol to
Dimethylamino-2-propanone [7], and so on. In our work, we plan to
produce Dimethylamino-3-propanal by catalytic amination of glycerol
with dimethylamine. Dimethylamino-3-propanal is a very important
know, there are few reports about the catalytic amination of glycerol.
Safariamin et al. [7] researched the direct amination of glycerol over
heteropolyacid-based catalysts. They prepared the Cs2.5H0.5PMo12O40
SiO2 catalysts with different loadings. The highest yield of Dimethyla-
mino-2-propanone (33%) was observed over the Cs2.5H0.5PMo12O40
/
/
SiO2 catalysts with loading of 50 wt%. However, the specific surface
area, pore volume, average pore radius and NH3 uptake of the catalyst
were very low. Therefore, the glycerol conversion and Dimethylamino-
2-propanone selectivity were very low, and there is no Dimethylamino-
3-propanal produced during the process. The catalytic performance
would be further improved.
Heteropolyacid with stable Keggin structure (H3PW12O40) is a
strong Brønsted acid and extensively studied in many acid-catalyzed
⁎
Corresponding authors.
E-mail addresses: ycitjfding@sina.com (J. Ding), matianlin951@163.com (T. Ma).
These authors contributed equally to this work.
1
https://doi.org/10.1016/j.mcat.2018.07.017
Received 22 April 2018; Received in revised form 28 June 2018; Accepted 16 July 2018
2468-8231/©2018ElsevierB.V.Allrightsreserved.

J. Ding et al.
MolecularCatalysis457(2018)51–58
reactions [8]. However, little attention has been given to another het-
eropolyacid with Wells–Dawson structure (H6P2W18O62). According to
the electrostatic theory, Wells–Dawson type H6P2W18O62 possesses
stronger Brønsted acid sites compared to the Keggin type H3PW12O40
[9]. Briand et al. [10] previously gave a detailed introduction about its
catalytic properties and applications. The Wells–Dawson type hetero-
polyacids have super-acidity and excellent stability in the solid phase
and in solution [11]. Various acid catalytic reactions are all effective
with an appropriate Wells–Dawson type heteropolyacid catalyst, such
as esterification [12], synthesis of coumarins [13], synthesis of flavones
[14], oxidation of 1,4-dihydropyridine [15], and oxidation of cyclooc-
tene and cyclohexene with hydrogen peroxide [16].
Considering the low surface area of pure bulk heteropolyacid, sev-
eral acidic or neutral supports including alumina, active carbon, silica,
titanium and zirconia have been employed for heteropolyacid [17],
which can increase abundant acid sites. Mesoporous material MCM-41
is a versatile support due to its large surface area, large pore size, and
high thermal stability [18,19]. However, there is very weak acidity on
the pure MCM-41 support, and it is well-known that pure MCM-41 silica
is Lewis solid without Brønsted acid sites [20–22]. To improve the
acidity of pure silica MCM-41, adding elements into its framework is
essential. For example, García-Sancho et al. [23] studied a family of
zirconium doped mesoporous silica catalysts in the dehydration of
glycerol. The results suggested that Zr doped mesoporous silica cata-
lysts markedly created the Brønsted acid sites on the MCM-41 and the
catalysts showed high catalytic performance.
Wells–Dawson type H6P2W18O62 was prepared according to the litera-
tures with some modifications [25,26]. 15 g of Na2WO4·2H2O was
dissolved in 35 ml of hot water under stirring, afterward 12.5 ml of an
aqueous solution of H3PO4(85%) was added while vigorous stirring.
After 30 min, the resultant solution was sealed in a Teflon-lined auto-
clave and heated at 140 °C for 6 h, and the resultant solution was kept
under static conditions in the Teflon-lined autoclave. Then, the auto-
clave was cooled to room temperature. 15 ml of concentrated HCl
(37%) was added to the above solution. After this, it was extracted with
the same volume of diethyl ether. The yellow powder was obtained by
evaporating ether, and then dried at 120 °C for 8 h.
The H3PW12O40/Zr-MCM-41 (Cat-1) and H6P2W18O62/Zr-MCM-41
(Cat-2) catalysts were synthesized both using wet impregnation
method. 1 g of MCM-41 was dispersed in 20 ml of deionized water
under constant stirring. Then, a calculated amount of H3PW12O40 or
H6P2W18O62 was added with vigorous stirring. The suspension was
stirred for 4 h at room temperature, and then evaporated to dryness.
The resulting solid was dried in an oven at 120 °C for 8 h and then
calcined in air at 300 °C for 3 h. Finally, the Cat-1 and Cat-2 catalysts
with different heteropolyacid loading amounts (2, 4, 8, 12, 16, 20, 24,
28, 30, 32, 36 and 40 wt%) were obtained.
Catalyst characterization
X-ray powder diffraction (XRD) analysis was recorded on
a
Shimadzu XRD-6000 using Cu Kα radiation (λ = 1.5405 Å) with the
second monochromator at 45 kV and 40 mA with a scanning speed of 2°
in 2θ/min. FTIR spectra of the samples were performed on a Nexus 670
IR spectrometer in the range of 4000–400 cm−1 at a resolution of
4 cm−1. Each spectrum was average over 128 scans. The specific sur-
face area, pore size and pore volume of the samples were obtained by
nitrogen adsorption–desorption isotherm using a Micromeritics ASAP
2020 instrument. The specific surface area was calculated by the BET
method. The total pore volume was determined by N2 adsorption at a
relative pressure of 0.99, and the pore diameter was obtained from the
desorption isotherm by the BJH method. Ammonia temperature-pro-
grammed desorption (NH3-TPD) profiles were obtained by Builder PCA-
1200, and the amount of acid sites was calculated by quantifying the
desorbed NH3 from NH3-TPD. Infrared spectroscopy (FT-IR) after pyr-
idine adsorption experiments were performed using a Shimadzu FTIR-
8700 spectrometer. The amount of Brønsted and Lewis acid sites de-
termined from the intensities of the IR bands at 1540 and 1445 cm-1 for
adsorbed pyridine using Emeis' method [27]. The equations are as
follows:
In our work, we research the catalytic amination of glycerol with
dimethylamine over different type of heteropolyacid supported on Zr-
MCM-41 catalysts, and compare the catalytic amination performance of
H3PW12O40/Zr-MCM-41 (Cat-1) and H6P2W18O62/Zr-MCM-41(Cat-2).
Furthermore, the reaction pathway is also proposed by analysing the
results of products distribution and characterization.
Experimental
Catalyst preparation
The MCM-41support was prepared according to our recent report
[24]. Firstly, 1.8 g of cetyltrimethylammonium bromide (CTAB) was
dissolved in 20 ml of distilled water under stirring. After this, 15 ml of
aqueous ammonium was added to the above solution, followed by the
addition of 5.3 g of tetraethylorthosilicate (TEOS) dropwise while stir-
ring. The resultant solution was stirred at room temperature for 2 h, and
then transferred into a Teflon-lined stainless-steel autoclave, heated
under 100 °C for 52 h, and the solution was kept under static conditions
in the autoclave. The solid product was filtered, washed with deionized
water. The white solid obtained was dried at 100 °C for 12 h. Finally,
the MCM-41 material was obtained by calcining at 550 °C for 3 h in air.
The Zr-MCM-41 sample was prepared according to the literature
[23,24]. TEOS was used as Si precursor and zirconium-n-propoxide
(70% in propanol) as Zr source, together with CTAB as synthesis tem-
plate. The preparation procedure of a Zr-MCM-41 sample with a molar
ratio of Si/Zr = 15 was as follows: firstly, two solutions were prepared:
the first solution was prepared by adding 3.0 ml of zirconium-n-prop-
oxide (70% in propanol) and 22.8 ml of TEOS while stirring; the second
solution was made by adding 12.2 g of CTAB into 110 ml water under
stirring, followed by the addition of 110 ml of aqueous NH3 (28 wt%).
Then, the first solution was added into the second solution dropwise.
The resultant solution was vigorously stirred for about 2 h, until a gel
was formed. The resultant gel was transferred into a Teflon-lined
stainless-steel autoclave without stirring and heated at 100 °C for 50 h.
After cooling to room temperature, the solid product was filtered and
washed with deionized water. The white solid obtained was dried in air
at 80 °C for 12 h. Finally, the sample was calcined at 550 °C for 3 h in
air.
C(pyridine on B sites) = 1.88 IA(B) R2/W;
C(pyridine on L sites) = 1.42 IA(L) R2/W;
C = concentration (mmol/g catalyst);
IA(B, L) = integrated absorbance of B or L band (cm−1);
R = radius of catalyst disk (cm);
W = weight of disk (mg).
The contact angles of water in air on the catalysts were measured
using a Krüss DSA100 instrument. Before measurement, the powder
sample was compressed into a disk with a thickness of approximately
1 mm (ca. 2 MPa). A drop of water(1 μL) was injected on the sample
disk. The appearance of the water drop was recorded at ca. 0.1 s with a
digital camera. The contact angle was determined by a photogonio-
metric method. W elemental content analysis was analyzed by means of
inductively coupled plasma (ICP). The carbon content in the spent
catalysts was analyzed for elemental analysis by using a CHNS analyzer
(Vario EL III).
The Keggin type H3PW12O40 was obtained from Sigma-Aldrich.
52

J. Ding et al.
MolecularCatalysis457(2018)51–58
Catalytic reaction
symmetric stretching vibration of Si−OH [28]. The Keggin type
H3PW12O40 shows several strong bands at 1080 cm−1 (PeO), 985 cm−1
(W]O), 889cm−1, 804 cm−1 (WeOeW), and one weak band at
523 cm−1 (WeOeP) [29–34]. For the Cat-1, the characteristic bands of
the Keggin ion are observed clearly, indicating the presence of
H3PW12O40 on the support. However, addition of H3PW12O40 on Zr-
MCM-41 reduces the intensity of these characteristic vibration bands
and some of these bands overlap with that of the Zr-MCM-41 support.
Four characterization bands of Wells–Dawson type anion at 1091, 963,
914, 782 cm−1, which can be assigned to PeO, W]O, W–Ob–W, and
W–Oc–W vibrations respectively [35]. For the Cat-2 catalyst, the cor-
responding characterization bands of Wells–Dawson anion are ob-
viously noticed, suggesting that H6P2W18O62 is incorporated in the
mesoporous molecular sieves. However, a decrease of the intensity of
these characteristic vibration bands is found. In addition, some of these
bands overlap with that of the Zr-MCM-41.
Fig. 2(A) displays the low-angle XRD patterns of the samples. The
patterns consist of a very strong peak along with two weak peaks for all
samples, which accord with a well ordered hexagonal mesoporous
structure [36,37]. However, it is observed that when an amount of
zirconium and heteropolyacid is incorporated into the Si-MCM-41 fra-
mework, intensities of the XRD peaks gradually become weaker. This
suggests that the mesoporous structure of MCM-41 supports remains
almost unchanged but the long-range order is decreased noticeably.
According to the high-angle XRD patterns of the materials (Fig. 2(B)),
for H3PW12O40, many intense diffraction peaks particularly at about 20,
26.7 and 29.6° are clearly found between 2θ = 10 and 90°. For
H6P2W18O62, several intense diffraction peaks especially at about 18.4,
24.6 and 27.2° are observed. Compared to bulk heteropolyacid, the
supported catalysts show only a broad peak in the range of 2θ = 15–40°
corresponding to the formation of an amorphous phase, indicating that
heteropolyacid is finely dispersed on Zr-MCM-41.
The catalytic activity was evaluated at atmospheric pressure in a
vertical fixed-bed reactor (50 cm length, 8 mm i.d.) using 0.5 g of cat-
alyst. The catalyst was ground and sieved to sizes between 20 and 40
mesh and placed in the middle part of the reactor, with quartz particles
packed at both ends. The reaction temperature was monitored by a
thermocouple which was placed near the catalyst bed. Before reaction,
the samples were pretreated at 120 °C under nitrogen flow (40 mL/min)
for about 1 h to remove adsorbed water. The 10 wt% glycerol aqueous
solution and 30 wt% dimethylamine (DMA) aqueous solution were
firstly introduced into a vaporizer by two HPLC pumps at a certain
proportion. Then the mixture was introduced from the top of the re-
actor at a certain flow rate under nitrogen atmosphere. The reaction
carried out under a given temperature. A small part of the reaction
products were used for On-line Analysis, and the left products were
collected in a conical flask cooled in a water-ice mixture.
The products were analyzed by a SP 6890 gas chromatograph (GC)
equipped with a flame ionization detector (FID) and HP-INNOWAX
capillary column(60 m × 0.32 mm × 0.25 μm) using methanol as in-
ternal standard. And the chromatograph is equipped with an on-line
analysis device, which can prevent the product from being oxidized by
air. 0.5 ml samples are injected hourly into the chromatograph. The
column oven was maintained at 40 °C for 2 min. The column was op-
erated from 40 to 100 °C with a ramping rate of 20 °C/min, and kept at
100 °C for 2 min. And then the column was operated from 100 to 180 °C
with a ramping rate of 20 °C/min, and kept at 180 °C for 2 min. Finally,
the column was operated from 180 to 280 °C with a ramping rate of
20 °C/min, and then kept at 280 °C for 2 min. Glycerol conversion and
product selectivity were quantified as follows:
Glycerol conversion (%) = (Moles of glycerol reacted/Moles of glycerol
in the feed) × 100
The results of nitrogen adsorption-desorption of MCM-41, Zr-MCM-
41, Cat-1 and Cat-2 samples are shown in Table 1. The surface area of
MCM-41 is 1013 m2/g. After addition of zirconium, H3PW12O40 and
H6P2W18O62, the surface area of these samples decreases continuously
to 843 (for Zr-MCM-41), 710 (for Cat-1) and 745 m2/g (for Cat-2).
Compared with the MCM-41, the specific surface area, pore volume and
pore size with zirconium and heteropolyacid loading catalysts are sig-
nificantly decreased, indicating that the zirconium and heteropolyacid
are either deposited on the external surface, or being introduced in the
porosity, filling the pores of the mesoporous MCM-41 support. How-
ever, the catalysts still retain uniform mesoporous systems, which are
well in accordance with the XRD results.
The temperature-programmed desorption profiles were collected
using NH3 as a probe molecule, and we deconvoluted the NH3-TPD
spectra of catalysts, as shown in Fig. 3. The amount of acid sites are
summarized in Table 1. It is generally accepted that the acid strength
depends on the ammonia desorption temperature: weak (100–200 °C),
medium (200–400 °C) and strong (> 400 °C) [38]. It is obvious that
MCM-41 has almost no TPD signal, which is in agreement with the
results reported by the research team led by Zheng [30,39–42] and
Sancho et al. [43], indicating the pure MCM-41 did not have obvious
acidity. On the other hand, Zr-MCM-41 exhibits two peaks of NH3-TPD
at 190 °C and 230 °C, which are attributed to the weak and medium acid
site, respectively. When the H3PW12O40 and H6P2W18O62 are supported
on the Zr-MCM-41, both of NH3-TPD temperatures shift right, in-
dicating that the strength of weak and medium acid sites are enhanced.
The amount of weak, medium acid sites and total acidities of the cat-
alysts are in an order of Cat-1 > Cat-2 > Zr-MCM-41. Both of NH3-TPD
temperatures of Cat-1 are higher than those of Cat-2, indicating that the
acid strength of Cat-1 is stronger than that of Cat-2.
Product selectivity (%) = (Moles of carbon in a product formed/Moles
of carbon in glycerol consumed) × 100
Results and discussion
Catalyst characterization
Fig. 1 presents the FT-IR spectra of the Zr-MCM-41, H3PW12O40
,
Cat-1, H6P2W18O62 and Cat-2. On the Zr-MCM-41, the band around
1300–1000 cm−1 is attributed to an asymmetric stretching mode of
Si–O–Si. Moreover, the bands at 791 and 450 cm−1 are ascribed to the
symmetric stretching vibration and bending vibration of the rocking
mode of Si–O–Si, respectively. The band at 960 cm−1 is due to the
Fig. 4 exhibits the FTIR spectra of pyridine adsorption of the sam-
ples. Cat-1, Cat-2 and Zr-MCM-41 catalysts contain Brønsted and Lewis
acid sites, as indicated by the adsorption peaks at about 1538 cm−1 and
Fig. 1. FT-IR spectra of different catalysts. Cat-1: H3PW12O40 loading of 30 wt
%, Cat-2: H6P2W18O62 loading of 20 wt%.
1445 cm−1
, respectively [44]. Furthermore, the peak at about
53

J. Ding et al.
MolecularCatalysis457(2018)51–58
Fig. 2. (A) Low-angle XRD diffraction patterns of MCM-41, Zr-MCM-41, Cat-1 and Cat-2; (B) High-angle XRD diffraction patterns of H3PW12O40, H6P2W18O62, Cat-1
and Cat-2. Cat-1: H3PW12O40 loading of 30 wt%, Cat-2: H6P2W18O62 loading of 20 wt%.
1490 cm−1 is attributed to both Brønsted and Lewis acid sites. The
concentration of Brønsted and Lewis acid sites are reported in Table 1.
There is only Lewis acid site detected on the MCM-41. It is well-
known that pure MCM-41 silica is Lewis solid without Brønsted acid
sites [20–22], and the acidity is very weak, it is almost impossible to
detect the NH3-TPD peak on the MCM-41. The Brønsted acid sites in Zr-
MCM-41 catalyst detected are due to the presence of atoms of Zr re-
placed the Si ones [45]. The generation of these Brønsted acid sites is
associated with the changes in electron density on Si, due to charge
imbalance or difference of electronegativity, resulting from introduc-
tion of Zr atom in the vicinity of the Si−OH, weakens the Si−OH bond
[43,46]. It can be seen from Table 1 that the loading of heteropolyacid
results in a noticeable increase in the Brønsted and Lewis acid sites on
Cat-1 and Cat-2, and the amount of Lewis acid sites on the Cat-1 is
higher than that of the Cat-2, while the amount of Brønsted acid sites on
Cat-2 is higher than that of the Cat-1. According to relevant reference
[47,48], the Brønsted acid sites were increased due to the presence of
small clusters of heteropolyacid, which is a kind of natural protonic
acid, and the Lewis acid sites could be due to the interaction between
heteropolyacid with Zr-MCM-41 support, which creates the ionic group
of [Zr-MCM-41−OH2+]. The possible interaction is shown as follows:
Fig. 3. NH3-TPD spectra of MCM-41, Zr-MCM-41, Cat-1 and Cat-2 catalysts.
Cat-1: H3PW12O40 loading of 30 wt%, Cat-2: H6P2W18O62 loading of 20 wt%.
+
Zr-MCM-41−OH + H3PW12O40/H6P2W18O62 → [Zr-MCM-41−OH2
]
–
[H2PW12O40–/H5P2W18O62
]
Catalytic amination performance
Fig. 5 exhibits the catalytic behavior of Cat-1 with different
H3PW12O40 loadings and Cat-2 with different H6P2W18O62 loadings.
The glycerol conversion gradually increase as adding the H3PW12O40
loading from 2 wt% to 30 wt% on Cat-1. Then, the conversion begin to
decrease with increasing of H3PW12O40 loadings after 30 wt%. As in-
creasing the loading of H3PW12O40 phase, the acidity of Cat-1 gradually
increases, therefore, the glycerol conversion becomes higher and
higher. When the H3PW12O40 loading is more than 30 wt%, excess
Fig. 4. Pyridine-adsorption FT-IR spectra of MCM-41, Zr-MCM-41, Cat-1 and
Cat-2. Cat-1: H3PW12O40 loading of 30 wt%, Cat-2: H6P2W18O62 loading of
20 wt%.
Table 1
Physicochemical properties and acidities of MCM-41, Zr-MCM-41, Cat-1 (H3PW12O40 loading of 30 wt%) and Cat-2 (H6P2W18O62 loading of 20 wt%) catalysts.
Catalyst
Surface area
(m2/g)
Pore volume
(cm3/g)
Pore size
(nm)
Acidity
(mmol NH3/gcat)
Brønsted acidity
(μmol/gcat)
Lewis acidity
(μmol/gcat)
B/(L + B)
Weak
Medium
Strong
MCM-41
Zr-MCM-41
Cat-1
1013
843
710
745
0.82
0.68
0.51
0.59
3.06
2.86
2.54
2.77
—
—
—
—
—
—
—
97.69
94.87
143.75
128.37
—
0.14
0.38
0.18
0.13
0.41
0.20
107.15
238.21
245.75
0.53
0.62
0.66
Cat-2
54

J. Ding et al.
MolecularCatalysis457(2018)51–58
Fig. 5. Effect of amount of H3PW12O40 and H6P2W18O62 loading on the cata-
lytic performance. Reaction conditions: 300 °C, GHSV of 3 h−1, reaction time
2 h, glycerol aqueous concentration of 10 wt%, DMA aqueous concentration of
30 wt%, molar ratio of DMA/glycerol of 2.5.
Fig. 6. The catalytic amination performance of Cat-1 and Cat-2 catalysts.
Reaction conditions: 300 °C, GHSV of 3 h−1, H3PW12O40 loading of 30 wt%
(Cat-1), H6P2W18O62 loading of 20 wt% (Cat-2), glycerol aqueous concentration
of 10 wt%, DMA aqueous concentration of 30 wt%, molar ratio of DMA/gly-
cerol of 2.5.
H3PW12O40 phase would agglomerate or superpose in the mesoporous
channels, which would result in the decrease of the surface area, pore
volume and pore size, and hyperacidity would cause the formation of
coke. Therefore, the catalytic performance begins to become weaker
and weaker. In our work, we chose the H3PW12O40 loading of 30 wt%
on Cat-1. According to the same reason, we chose the H6P2W18O62
loading of 20 wt% on Cat-2.
The catalytic performance of MCM-41, Zr-MCM-41, Cat-1 and Cat-2
catalysts are shown in Table 2. We can see that the glycerol conversion,
D-3 and D-2 selectivity on the Zr-MCM-41 are higher than those of
MCM-41, because there is very weak acidity on the MCM-41. At the
beginning of the reaction, the order of the glycerol conversion is MCM-
41 < Zr-MCM-41 < Cat-2 < Cat-1, it's the same as the order of total
amount of acidity on these catalysts. And the order of the D-3 selectivity
is MCM-41 < Zr-MCM-41 < Cat-1 < Cat-2, it's the same as the order
of amount of Brønsted acid sites on these catalysts (in Table 1).
Therefore, we conclude that more amount of acidity is good for the
glycerol conversion and the Brønsted acid site is good for formation of
D-3.
Table 3
The leaching ratio and coke over Cat-1 (H3PW12O40 loading of 30 wt%) and
Cat-2 (H6P2W18O62 loading of 20 wt%) catalysts.
Catalysts
W (wt%)
Leaching ratio (%)
Coke (wt%)
Cat-1(fresh)
Cat-1 (used)
Cat-2(fresh)
Cat-2(used)
23.0
21.6
15.2
14.1
̶
̶
6.1
3.2
̶
̶
7.3
1.7
glycerol conversion decreased from 98.8% to 67.4% on the Cat-1,
however, the glycerol conversion decreased from 95.2% to 72.5% on
the Cat-2. The stability of Cat-2 is obviously higher than that of Cat-1.
The reason may be that the total amount of acidity on Cat-1 is more
than that of Cat-2, which is good for improving the glycerol conversion.
However, more acidity results in producing more amount of coke on
Cat-1 at the same time. In Table 1, the surface area, pore volume and
pore size of the Cat-2 catalyst are larger than those of Cat-1, which
would reduce the constraints to the mass transfer and usually improve
the performance by facilitating both the access of the reactants to the
active sites and the diffusion of reaction products. However, the active
sites and pore are easy to be covered by the coke on the Cat-1 catalyst.
Therefore, the glycerol conversion on Cat-1 is higher than that of Cat-2
at the beginning of reaction. As the reaction time increasing, the de-
activation of Cat-1 is more serious than that of Cat-2. In Table 3, the
leaching ratio of W over the used Cat-2 (7.3%) is slightly more than that
of Cat-1(6.1%), which may be attributed to that H6P2W18O62 is more
hydrophilic than H3PW12O40. However, the stability of Cat-2 is higher
than that of Cat-1, therefore, the main reason for deactivation of cat-
alyst is the formation of carbon deposit during this reaction.
The hydrophilic/hydrophobic properties of the Cat-1 and Cat-2
catalysts were characterized by the water contact angles measurements.
The results were shown in Fig. 7. The contact angle of H6P2W18O62/Zr-
MCM-41 (23°) is less than that of H3PW12O40/Zr-MCM-41 (44°), in-
dicating that H6P2W18O62/MCM-41 is more hydrophilic than
H3PW12O40/MCM-41.
All kinds of products were shown in Table 4. We can see that the
selectivity of Dimethylamino-3-propanal, acrolein, acetaldehyde and
formaldehyde on the Cat-2 are higher than those of Cat-1, however, the
selectivity of Dimethylamino-2-propanone and hydroxyacetone on the
Cat-1 are higher than those of Cat-2 at the same reaction time. Ac-
cording to the results of pyridine-FTIR in Table 1, the Cat-2 exhibits
more Brønsted and less Lewis acid sites than those of Cat-1, and the
ratio of B/(L + B) on Cat-2 is higher than that of Cat-1. Therefore, we
conclude that the Brønsted acid sites are helpful to produce
The catalytic amination performance comparison between Cat-1
and Cat-2 was shown in Fig. 6, and the coke amounts of the catalyst
used after 10 h and the W elemental content of the fresh catalyst and
catalyst used after 10 h were shown in Table 3. We can see that the
glycerol conversion on Cat-1 is higher than that of Cat-2 before 5 h
reaction time. However, the glycerol conversion on Cat-2 is higher than
that of Cat-1 after 5 h reaction time. During 10 h reaction time, the
Table 2
Glycerol conversion and product distribution results over different catalysts.
Selectivity (%)
Catalyst
MCM-41
Zr-MCM-41
Cat-1
aCgly (%)
bD-3
cD-2
dHA
eAc
fAD
gFD
40.3
(31.5)
65.8
(48.7)
98.8
6.5
(4.7)
39.2
(35.9)
61.5
5.7
(9.1)
34.3
(38.9)
32.6
18.3
(20.5)
7.8
(8.1)
2.1
45.3
(41.6)
5.7
(3.4)
1.7
15.5
(17.3)
8.3
(11.7)
1.3
8.7
(6.8)
2.2
(2.0)
0.8
(67.4)
95.2
(59.8)
80.8
(34.6)
11.4
(2.8)
1.6
(0.3)
3.1
(2.3)
1.8
(0.2)
1.3
Cat-2
(72.5)
(79.4)
(12.9)
(2.8)
(1.0)
(3.6)
(0.3)
Reaction conditions: 300 °C, GHSV of 3 h−1, reaction time 2 h, H3PW12O40
loading of 30 wt% (Cat-1), H6P2W18O62 loading of 20 wt% (Cat-2), glycerol
aqueous concentration of 10 wt%, DMA aqueous concentration of 30 wt%,
molar ratio of DMA/glycerol of 2.5. aConversion of glycerol; bDimethylamino-3-
propanal;
cDimethylamino-2-propanone;
dHydroxyacetone;
eAcrolein;
fAcetaldehyde; gFormaldehyde. In brackets, conversion and selectivity after
10 h on stream.
55

J. Ding et al.
MolecularCatalysis457(2018)51–58
Fig. 7. Contact angles of water on (a) Cat-1
and (b) Cat-2. H3PW12O40 loading of 30 wt%
(Cat-1), H6P2W18O62 loading of 20 wt% (Cat-
2).
Dimethylamino-3-propanal, and the Lewis acid sites easily lead to the
formation of Dimethylamino-2-propanone. And as the reaction time
increasing, the selectivity of Dimethylamino-3-propanal, acrolein and
formaldehyde becomes lower and lower on the same catalyst, however,
the selectivity of Dimethylamino-2-propanone, hydroxyacetone and
acetaldehyde becomes higher and higher. We suggest the reason may be
that the Brønsted acid sites are easier to be covered by the coke than the
Lewis acid sites.
Optimization of reaction conditions
According to the results of Dimethylamino-3-propanal yield in
Fig. 6, we know that the catalytic amination performance of Cat-2 is
superior to that of Cat-1, so we have investigated the effects of reaction
temperature, molar ratio of glycerol/DMA and GHSV of gas mixture on
the catalytic amination performance of Cat-2.
Effect of reaction temperature
The effect of reaction temperature on the catalytic amination per-
formance of Cat-2 is shown in Fig. 9. The high reaction temperature is
helpful to improve the glycerol conversion, because the catalytic ami-
nation of glycerol is a kind of endothermic reaction. Therefore, the
glycerol conversion is gradually improved as increasing the reaction
temperature. After 300 °C, the trend of increasing conversion slows
down as increasing the reaction temperature, and the D-3 selectivity
begins to become lower and lower because of the coke and other by-
products formation under higher temperature. Finally, the yield of D-3
begins to decrease after 300 °C reaction temperature.
Reaction pathway of catalytic amination
Based on the above results, the reaction pathway of catalytic ami-
nation of glycerol with dimethylamine (DMA) is proposed in Fig. 8.
There are two dehydration routes on different types of acid sites. On the
Brønsted acid sites, the protonated glycerol is dehydrated to produce 3-
hydroxypropanal, which can easily convert to acrolein [49,50]. The
produced acrolein furtherly reacts with DMA to produce Dimethyla-
mino-3-propanal. The unstable 3-hydroxypropanal can be decomposed
to formaldehyde and acetaldehyde, and it also can react directly with
DMA to produce Dimethylamino-3-propanal. On the Lewis acid sites,
hydroxyacetone is formed through dehydration and deprotonation of
the protonated intermediate. Further, methylglyoxal is produced by
dehydrogenation of hydroxyacetone. However, the methylglyoxal is
unstable, which can be decarbonylated to acetaldehyde [51], and it also
can react directly with DMA to produce Dimethylamino-2-propanone.
And the acid catalyzed CeC bond cleavage of hydroxyacetone to
acetaldehyde and formaldehyde maybe happened at the same time. The
3-hydroxypropanal, acrolein and hydroxyacetone can be converted into
the coke under the strong acidic condition [52].
Effect of GHSV
The effect of GHSV on the catalytic amination performance of Cat-2
is shown in Fig. 10. From the results, it can be clearly seen that the
GHSV has a significant impact on the catalytic amination performance.
As increasing of GHSV from 1 to 10 h−1, the glycerol conversion gra-
dually decreases, and the D-3 selectivity gradually increases, simulta-
neously. We infer the reason maybe that the higher space velocity
shortens the residence time for glycerol with the catalysts, and the re-
sidence time for by-products of acetaldehyde and formaldehyde with
acid sites on the catalyst is also shortened. Therefore, the probability of
the coke produced from the acetaldehyde, formaldehyde, 3-hydro-
xypropanal and acrolein under the strong acidic condition would be
Table 4
The product distribution results over Cat-1 and Cat-2.
Catalysts
Reaction time (h)
Selectivity (%)
Dimethylamino-
Dimethylamino-
Hydroxyacetone
Acrolein
Acetaldehyde
Formaldehyde
−3-propanal
−2-propanone
Cat-1
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
61.5
61.5
61.3
61.2
61.2
61.1
61.1
60.6
60.2
59.8
80.8
80.7
80.5
80.4
80.3
80.3
80.1
79.9
79.6
79.4
32.6
32.8
32.8
33.0
33.1
33.3
33.5
33.8
34.2
34.6
11.4
11.6
11.6
11.8
11.8
12.1
12.4
12.6
12.8
12.9
2.1
2.2
2.4
2.4
2.6
2.7
2.7
2.8
2.7
2.8
1.6
1.7
1.9
2.0
2.2
2.3
2.4
2.6
2.7
2.8
1.7
1.5
1.4
1.3
1.1
0.7
0.5
0.5
0.4
0.3
3.1
2.9
2.6
2.3
2.1
1.8
1.5
1.3
1.0
1.0
1.3
1.4
1.5
1.6
1.6
1.8
1.9
2.0
2.2
2.3
1.8
2.0
2.4
2.5
2.8
2.9
3.1
3.1
3.5
3.6
0.8
0.6
0.6
0.5
0.4
0.4
0.3
0.3
0.3
0.2
1.3
1.1
1.0
1.0
0.8
0.6
0.5
0.5
0.4
0.3
Cat-2
8
9
10
Reaction conditions: 300 °C, GHSV of 3 h−1, H3PW12O40 loading of 30 wt% (Cat-1), H6P2W18O62 loading of 20 wt% (Cat-2), glycerol aqueous concentration of 10 wt
%, DMA aqueous concentration of 30 wt%, molar ratio of DMA/glycerol of 2.5.
56

J. Ding et al.
MolecularCatalysis457(2018)51–58
Fig. 8. Proposed reaction pathway of catalytic amination of glycerol with dimethylamine.
Fig. 10. Effect of GHSV on the catalytic amination performance of Cat-2.
Reaction conditions: 300 °C, reaction time 2 h, H6P2W18O62 loading of 20 wt%,
glycerol aqueous concentration of 10 wt%, DMA aqueous concentration of
30 wt%, molar ratio of DMA/glycerol of 2.5.
Fig. 9. Effect of reaction temperature on the catalytic amination performance of
Cat-2. Reaction conditions: reaction time 2 h, H6P2W18O62 loading of 20 wt%,
GHSV of 3 h−1, glycerol aqueous concentration of 10 wt%, DMA aqueous
concentration of 30 wt%, molar ratio of DMA/glycerol of 2.5.
and the highest yield of D-3 was appeared at the 2.5 M ratio of DMA/
glycerol.
reduced obviously, and the probability of D-3 formation is increased at
the same time. The highest yield of D-3 was appeared at the GHSV of
3 h−1
.
Conclusions
Effect of molar ratio of DMA/glycerol
In this work, the performance of Cat-1 was compared with that of
Cat-2 during the catalytic amination of glycerol with dimethylamine to
produce Dimethylamino-3-propanal. Firstly, the optimum loading
amount of two kinds catalysts were obtained under the same reaction
conditions. Under the premise of the optimum loading, the specific
surface area, pore volume and pore size of the Cat-2 were higher than
those of the Cat-1. Cat-1 exhibited more total acidity than that of Cat-2,
but the amount of Brønsted acid sites and ratio of B/(B + L) on Cat-2
were higher than that of Cat-1. The amount of coke on Cat-2 was lower
than that of Cat-1. Therefore, the catalytic stability, selectivity and yield
of Dimethylamino-3-propanal of Cat-2 were higher than those of Cat-1,
which proved that the Brønsted acid sites are good for the catalytic
The effect of molar ratio of DMA/glycerol on the catalytic amination
performance of Cat-2 is shown in Fig. 11. As the molar ratio of DMA/
glycerol increasing, the glycerol conversion gradually increased until
the molar ratio is 2.5, then the glycerol conversion gradually decreased.
The reason is that an appropriate amount of DMA is good for reaction
with acrolein, 3-hydroxypropanal or hydroxyacetone, which promotes
the reaction of glycerol conversion. However, the excessive amount of
DMA may cause that the acid sites on the Cat-2 were covered by DMA,
and it is difficult for the access of the glycerol to the acid sites, which
results in decreasing the glycerol conversion. The D-3 selectivity gra-
dually increased with increasing of the molar ratio of DMA/glycerol,
57

J. Ding et al.
MolecularCatalysis457(2018)51–58
Tetrahedron Lett. 45 (2004) 8935–8939.
[14] D.O. Bennardi, G.P. Romanelli, Á.G. Sathicq, J.C. Autino, G.T. Baronetti,
H.J. Thomas, Appl. Catal. A Gen. 404 (2011) 68–73.
[15] L.M. Sanchez, Á.G. Sathicq, G.T. Baronetti, H.J. Thomas, G.P. Romanelli, Catal.
Lett. 144 (2014) 172–180.
[16] M.R. Farsani, F. Jalilian, B. Yadollahi, H.A. Rudbari, Appl. Organometal. Chem. 29
(2015) 7–11.
[17] T.H. Kang, J.H. Choi, Y. Bang, J. Yoo, J.H. Song, W. Joe, J.S. Choi, I.K. Song, J. Mol.
Catal. A: Chem. 396 (2015) 282–289.
[18] X.Y. Chen, M. Jia, G.Z. Liu, X.W. Zhang, L. Wang, Z.T. Mi, Appl. Surf. Sci. 256
(2010) 5856–5861.
[19] J.J. Zou, Y. Xu, X.W. Zhang, L. Wang, Appl. Catal. A Gen. 421-422 (2012) 79–85.
[20] A.E.R.S. Khder, H.M.A. Hassan, M.S. EI-Shall, Appl. Catal. A Gen. 411-412 (2012)
77–86.
[21] F. Cavani, S. Guidetti, L. Martinelli, M. Piccinini, E. Ghedini, M. Signoretto, Appl.
Catal. B 100 (2011) 197–204.
[22] D. Carriazo, C. Domingo, C. Martin, V. Rives, J. Solid State Chem. 181 (2008)
2046–2057.
[23] C. García-Sancho, R. Moreno-Tost, J. Mérida-Robles, J. Santamaría-González,
A. Jiménez-López, P. Maireles-Torres, Appl. Catal. A Gen. 433-434 (2012) 179–187.
[24] T.L. Ma, Z. Yun, W. Xu, L.G. Chen, L. Li, J.F. Ding, R. Shao, Chem. Eng. J. 294
(2016) 343–352.
[25] C.W. Hu, M. Hashimoto, T. Okuhara, M. Misono, J. Catal. 143 (1993) 437–448.
[26] C.R. Graham, R.G. Finke, Inorg. Chem. 47 (2008) 3679–3686.
[27] C.A. Emeis, J. Catal. 141 (1993) 347–354.
Fig. 11. Effect of molar ratio of DMA/glycerol on the catalytic amination
performance of Cat-2. Reaction conditions: 300 °C, reaction time 2 h,
H6P2W18O62 loading of 20 wt%, GHSV of 3 h−1, glycerol aqueous concentration
of 10 wt%, DMA aqueous concentration of 30 wt%.
[28] G. Karthikeyan, A. Pandurangan, J. Mol. Catal. A: Chem. 311 (2009) 36–45.
[29] Q.-H. Xia, K. Hidajat, S. Kawi, J. Catal. 209 (2002) 433–444.
[30] Y. Chen, Y. Cao, G.P. Zheng, B.B. Dong, X.C. Zheng, Adv. Powder Technol. 25
(2014) 1351–1356.
[31] B.R. Jermy, A. Pandurangan, Appl. Catal. A Gen. 295 (2005) 185–192.
[32] X.L. Yang, W.L. Dai, H. Chen, J.H. Xu, Y. Cao, H.X. Li, K.N. Fan, Appl. Catal. A Gen.
283 (2005) 1–8.
amination of glycerol with dimethylamine to produce Dimethylamino-
3-propanal. According the product distribution results, the reaction
pathway of catalytic amination of glycerol with dimethylamine was
proposed. Furthermore, the reaction temperature, GHSV and the molar
ratio of DMA/glycerol were optimized on Cat-2 catalyst.
[33] G.A. Eimer, S.G. Casuscelli, G.E. Ghione, M.E. Crivello, E.R. Herrero, Appl. Catal. A
Gen. 298 (2006) 232–242.
Acknowledgements
[34] G. Karthikeyan, A. Pandurangan, J. Mol. Catal. A Chem. 311 (2009) 36–45.
[35] S.R. Matkovic, G.M. Valle, L.A. Gambaro, L.E. Briand, Catal. Today 133–135 (2008)
192–199.
We are grateful for the financial support of the National Natural
Science Foundation of China (No. 21303154) and the joint research
fund between Collaborative Innovation Center for Ecological Building
Materials and Environmental Protection Equipments and Key
Laboratory for Advanced Technology in Environmental Protection of
Jiangsu Province.
[36] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992)
710–712.
[37] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt,
C.T.W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins, J.L. Schlenker,
J. Am. Chem. Soc. 114 (1992) 10834–10843.
[38] P. Berteau, B. Delmon, Catal. Today 5 (1989) 121–137.
[39] Y. Chen, X.L. Zhang, X. Chen, B.B. Dong, X.C. Zheng, Solid. State Sci. 24 (2013)
21–25.
[40] Y. Chen, X.L. Zhang, M.M. Dong, Y.H. Wu, G.P. Zheng, J. Huang, X.X. Guan,
X.C. Zheng, J. Taiwan Inst. Chem. E. 61 (2016) 147–155.
[41] Y. Chen, X. Chen, B.B. Dong, G.H. Wang, X.C. Zheng, Mater. Lett. 114 (2014) 72–75.
[42] M. Wu, Q.Q. Zhao, J. Li, X.L. Su, H.Y. Wu, X.X. Guan, X.C. Zheng, J. Porous. Mater.
23 (2016) 1329–1338.
[43] C.G. Sancho, R.M. Tost, J.M. Robles, J.S. González, A.J. López, P.M. Torres, Appl.
Catal. A Gen. 433-434 (2012) 179–187.
[44] A.E.R.S. Khder, H.M.A. Hassan, M.S. El-Shall, Appl. Catal. A Gen. 411-412 (2012)
References
[1] T.L. Ma, J.F. Ding, R. Shao, W. Xu, Z. Yun, Chem. Eng. J. 316 (2017) 797–806.
[2] M. Dalil, D. Carnevali, J.L. Dubois, G.S. Patience, Chem. Eng. J. 270 (2015)
557–563.
[3] F.J.G. Ortiz, P. Ollero, A. Serrera, S. Galera, Chem. Eng. J. 218 (2013) 309–318.
[4] C.C. Wang, H. Jiang, C.L. Chen, R.Z. Chen, W.H. Xing, Chem. Eng. J. 264 (2015)
344–350.
77–86.
[45] P. Salas, J.A. Wang, H. Armendariz, C. Angeles-Chavez, L.F. Chen, Mater. Chem.
Phys. 114 (2009) 139–144.
[46] J.A. Anderson, C. Fergusson, I. Rodríguez-Ramos, A. Guerrero-Ruiz, J. Catal. 192
[5] B.N. Zope, D.D. Hibbitts, M. Neurock, R.J. Davis, Science (80-) 330 (2010) 74–78.
[6] S.H. Zhu, Y.L. Zhu, X.Q. Gao, T. Mo, Y.F. Zhu, Y.W. Li, Bioresour. Technol. Rep. 130
(2013) 45–51.
(2000) 344–354.
[7] M. Safariamin, S. Paul, K. Moonen, D. Ulrichts, F. Dumeignil, B. Katryniok, Catal.
[47] Q.H. Xia, K. Hidajat, S. Kawi, J. Catal. 209 (2002) 433–444.
[48] F. Lefebvre, J. Chem. Soc. Chem. Commun. 10 (1992) 756–757.
[49] L.Q. Shen, H.B. Yin, A.L. Wang, Y.H. Feng, Y.T. Shen, Z.N. Wu, T.S. Jiang, Chem.
Eng. J. 180 (2012) 277–283.
Sci. Technol. 6 (2016) 2129–2135.
[8] T. Okuhara, Appl. Catal. A Gen. 256 (2003) 213–224.
[9] C.L. Wang, X.N. Bu, J.W. Ma, C. Liu, K.C. Chou, X.L. Wang, Q. Li, Catal. Today 274
(2016) 82–87.
[50] E. Tsukuda, S. Sato, R. Takahashi, T. Sodesawa, Catal. Commun. 8 (2007)
1349–1353.
[51] A. Corma, G.W. Huber, L. Sauvanaud, P. O’Connor, J. Catal. 257 (2008) 163–171.
[52] D. Yun, Y.S. Yun, T.Y. Kim, H. Park, J.M. Lee, J.W. Han, J. Yi, J. Catal. 341 (2016)
33–43.
[10] L.E. Briand, G.T. Baronetti, H.J. Thomas, Appl. Catal. A Gen. 256 (2003) 37–50.
[11] G.M. Valle, L.E. Briand, Mater. Lett. 57 (2003) 3964–3969.
[12] M. Arabi, M.M. Amini, M. Abedini, A. Nemati, M. Alizadeh, J. Mol. Catal. A: Chem.
200 (2003) 105–110.
[13] G.P. Romanelli, D. Bennardi, D.M. Ruiz, G. Baronetti, H.J. Thomas, J.C. Autino,
58