Investigation of supported Zn(OAc)2 catalyst and its stability in N-phenyl carbamate synthesis

Article abstract of DOI:10.1016/j.apcata.2014.01.008

Zn(OAc)₂ was found to give excellent catalytic performance for methyl N-phenyl carbamate (MPC) synthesis from aniline and dimethyl carbonate (DMC). However, after being used for only once it tended to lose activity because of the formation of ZnO by the reaction of Zn(OAc)₂ and methanol. Zn(OAc)₂/SiO₂ was prepared by incipient impregnation and it gave excellent catalytic performance in MPC synthesis, on which aniline conversion and MPC yield were 98.1% and 93.8%, respectively. And Zn(OAc)₂/SiO₂ was found to be more stable than Zn(OAc)₂ during the reaction. When Zn(OAc)₂/SiO ₂ was used for the fifth time, aniline conversion and MPC yield were found to be 64.3% and 38.1%, respectively. The Zn(OAc)₂/SiO ₂ catalyst was characterized by TG-DTA, ICP, FTIR, XRD and XPS. The characterization results suggested that the deactivation of Zn(OAc) ₂/SiO₂ was also due to the formation of ZnO and there were two reasons for the improved stability of Zn(OAc)₂/SiO₂ catalyst. One was the formation of the Si-O-Zn bonds in Zn(OAc) ₂/SiO₂ catalyst, which increased the steric hindrance of Zn and consequently retarded the reaction between Zn(OAc)₂ and methanol. The other cause was the dehydration between methanol and the hydroxyl group on the SiO₂ surface, which reduced the chance of a reaction between methanol and Zn(OAc)₂.

Full text of DOI:10.1016/j.apcata.2014.01.008

Accepted Manuscript
Title: Investigation of supported Zn(OAc)2 catalyst and its
stability in N-phenyl carbamate synthesis
Author: Fang Li Wenbo Li Jing Li Wei Xue Yanji Wang
Xinqiang Zhao
PII:
S0926-860X(14)00010-6
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2014.01.008
APCATA 14646
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
18-8-2013
30-12-2013
8-1-2014
Please cite this article as: F. Li, W. Li, J. Li, W. Xue, Y. Wang, X. Zhao, Investigation of
supported Zn(OAc)2 catalyst and its stability in N-phenyl carbamate synthesis, Applied
Catalysis A, General (2014), http://dx.doi.org/10.1016/j.apcata.2014.01.008
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

*Graphical Abstract (for review)
Zn(OAc)2
100
80
60
40
20
0
Aniline conversion
MPC yield
MPC selectivity
SiO2
Zn(OAc)2
Aniline
DMC
MPC
Methanol
4
2
3
5
1
Reaction times
Stability of Zn(OAc)2/SiO2 catalyst
Page 1 of 29

*Highlights (for review)
Highlights
·Zn(OAc)2/SiO2 was prepared by incipient impregnation.
·Zn(OAc)2/SiO2 shows the similar catalytic performance as Zn(OAc)2.
·The stability of Zn(OAc)2/SiO2 improves considerably, but Zn(OAc)2 can’t be reused.
·The cause for the improved stability of Zn(OAc)2/SiO2 was investigated.
Page 2 of 29

*Manuscript
Investigation of supported Zn(OAc)2 catalyst and its stability in
N-phenyl carbamate synthesis
Fang Lia*, Wenbo Lia, Jing Lib, Wei Xuea, Yanji Wanga*, Xinqiang Zhaoa
a Hebei Provincial Key Lab of Green Chemical Technology & High Efficient Energy Saving, School of Chemical
Engineering & Technology, Hebei University of Technology, Tianjin, 300130
b School of Civil Engineering, Hebei University of Technology, Tianjin, 300130
Abstract: Zn(OAc)2 was found to give excellent catalytic performance for methyl N-phenyl
carbamate (MPC) synthesis from aniline and dimethyl carbonate (DMC). However,
after being used for only once it tended to lose activity because of the formation of
ZnO by the reaction of Zn(OAc)2 and methanol. Zn(OAc)2/SiO2 was prepared by
incipient impregnation and it gave excellent catalytic performance in MPC synthesis,
on which aniline conversion and MPC yield were 98.1% and 93.8%, respectively. And
Zn(OAc)2/SiO2 was found to be more stable than Zn(OAc)2 during the reaction. When
Zn(OAc)2/SiO2 was used for the fifth time, aniline conversion and MPC yield were
found to be 64.3% and 38.1%, respectively. The Zn(OAc)2/SiO2 catalyst was
characterized by TG-DTA, ICP, FTIR, XRD and XPS. The characterization results
suggested that the deactivation of Zn(OAc)2/SiO2 was also due to the formation of ZnO
and there were two reasons for the improved stability of Zn(OAc)2/SiO2 catalyst. One
was the formation of the Si–O–Zn bonds in Zn(OAc)2/SiO2 catalyst, which increased
the steric hindrance of Zn and consequently retarded the reaction between Zn(OAc)2
and methanol. The other cause was the dehydration between methanol and the
hydroxyl group on the SiO2 surface, which reduced the chance of a reaction between
methanol and Zn(OAc)2.
Keywords: N-phenyl carbamate, aniline, dimethyl carbonate, Zn(OAc)2/SiO2
*Corresponding author. Tel: +86 22 60203540; fax: +86 22 60204294.
E-mail address: lifang@hebut.edu.cn
1
Page 3 of 29

1. Introduction
Methyl N-phenyl carbamate (MPC) is an intermediate for the preparation of methylene diphenyl
diisocyanate (MDI), which is an important monomer for polyurethane synthesis. Initially,
isocyanate was synthesized by the reaction between an amine and phosgene. This method results
in environmental pollution and safety issues, and efforts are being made to develop non-phosgene
routes such as the reductive carbonylation of nitrobenzene [1–4], the oxidative carbonylation of
aniline [4–7] and the aminolysis of dimethyl carbonate (DMC). The reductive carbonylation of
nitrobenzene and the oxidative carbonylation of aniline is carried out under higher temperatures
and pressures, and a noble metal is used as the catalyst. However, the aminolysis of DMC for MDI
synthesis is becoming more attractive. The reaction scheme for the synthesis of MDI from DMC is
described in Scheme 1. In this process the only by-products are methanol and water. In addition,
methanol is used as a raw material for the production of DMC by oxidative carbonylation, which
will reduce the production cost of DMC.
In the first step of Scheme 1, a catalyst is required to achieve a better MPC yield. At present, the
catalysts generally used in this reaction are zinc or lead compounds [8-15] as well as other oxide
catalysts [16–20]. Among the catalysts, Zn(OAc)2 often gives the best activity for aromatic
carbamate synthesis. Gurgiolo [8] prepared MPC over a zinc acetate catalyst and the selectivity of
MPC toward aniline was 99.8% at 140 °C and 0.88 MPa. Additionally, Baba [12] investigated the
methoxycarbonylation of 2,4-toluene diamine and 4,4′-diphenylmethane diamine with DMC to
their corresponding dicarbamates using Zn(OAc)2 as a catalyst at 180 °C for 2 h. The yield of
dimethyltoluene-2,4-dicarbamate
was
96%
while
the
yield
of
dimethyl-4,4′-methylenediphenyldicarbamate was 98%. Although Zn(OAc)2 exhibits high
conversion and selectivity, it possesses certain drawbacks. Apart from the separation and recovery
of the generated homogeneous zinc acetate, it loses activity easily because of its transformation to
ZnO by reaction with methanol and this is a byproduct in the reaction. To overcome these
problems we prepared supported Zn(OAc)2. Active carbon is the best support for these catalysts.
Aniline conversion and MPC selectivity has been found to be 78.0% and 98.0%, respectively, for
the Zn(OAc)2/AC catalyst [13]. The catalytic activity of Zn(OAc)2/AC is lower than that of
Zn(OAc)2. Furthermore, its stability was not investigated. In this paper we report the activity of
2
Page 4 of 29

the Zn(OAc)2/SiO2 catalyst, which has almost the same catalytic activity as Zn(OAc)2 but its
stability was found to be better. The cause of Zn(OAc)2/SiO2 deactivation and the improvement in
stability is also discussed in this paper.
2. Experimental
2.1 Catalyst preparation
The Zn(OAc)2/SiO2 catalyst was prepared by incipient impregnation. Appropriate amounts of
Zn(OAc)2 were dissolved in distilled water. The solution was impregnated into SiO2. The
impregnated samples were left overnight at room temperature and then dried under vacuum at
60 °C. Calcination was carried out at 300 °C for 2 h to obtain the Zn(OAc)2/SiO2 catalyst.
2.2 Catalytic evaluation
MPC was synthesized in a stainless steel autoclave. 1.0 ml aniline, 20 ml DMC and 0.28 g
catalyst were added to the autoclave. The reaction temperature was 170 °C and the reaction time
was 7 h. After the reaction, the reaction mixture was filtered to separate the catalyst and the filtrate
was analyzed by high performance liquid chromatography (HPLC).
2.3 Product analysis
Waters HPLC was used to analyze the reaction liquid. The analytical conditions were as follows:
A Kromasil C-18 (Φ4.6 mm×150 mm, 5 μm) chromatographic column was used at a flow rate of
0.4 ml/min. CH3OH/H2O (70/30, V/V) was used as the mobile phase and the detection wavelength
was 254 nm. A quantitative analysis was carried out using nitrobenzene as the internal standard.
The product mixture was identified qualitatively using a Thermo Fisher TRACE DSQ
GC-MS with a CP-SIL5CB (3.0 m×0.25 mm) capillary column. The injection port temperature
was 250 °C. The column temperature was program-controlled and the initial column temperature
was fixed at 40 °C for 3 min and then increased to 250 °C at a rate of 10 °C /min. The final
temperature was maintained for 3 min. A mass spectrometer was used in Electron Impact (EI)
mode at 200 °C.
2.4 Catalyst characterization
The Zn content of the catalyst was determined using a PerkinElmer Optima ICP 7300V.
X-ray diffraction patterns were recorded on a Rigaku D/MAX2500 with Cu Kα radiation at 40
3
Page 5 of 29

kV and 100 mA. The scanning range was 3-80°.
Thermal analysis was performed using a Rigaku standard model TG-DTA analyzer. The sample
was heated under air and the heating rate was 10 °C /min.
X-ray photoelectron spectroscopy was carried using a PHI-1600 ESCA SYSTEM. Mg Kα
radiation at 300 W and 15 kV was used as the light source and the vacuum system was controlled
at about 1.33×106 Pa. The binding energy values of the XPS signals were calibrated using
adventitious C1s = 284.6 eV as a reference.
FTIR spectra were obtained with a NICOLET NEXUS-470 spectrometer. A DTGS detector was
used and 32 scans were accumulated at a spectral resolution of 4 cm1.
Brunauer Emmett Teller (BET) surface area measurements were done using an ASAP 2020
instrument to analyze the specific surface area, the pore volume and the pore size distribution of
the catalysts.
3. Results and discussion
3.1 Effect of Zn(OAc)2 load on the catalytic performance of Zn(OAc)2/SiO2
The effect of Zn(OAc)2 load on the catalytic performance of Zn(OAc)2/SiO2 is shown in Table 1
(entries 1–5). The aniline conversion and the MPC yield were found to be 37.2% and 5.2%,
respectively, at a Zn(OAc)2 load of 2%. Most of the aniline transformed into N-methylaniline
(NMA) and N,N′-dimethylaniline (DMA). Aniline conversion increased with Zn(OAc)2 load and a
maximum of 98.4% was found for a Zn(OAc)2 load of 20%. Upon increasing the Zn(OAc)2 load
continuously the aniline conversion decreased slightly. The cause might be that the pores of SiO2
are thus blocked by Zn(OAc)2 and ZnO for a small amount of Zn(OAc)2 might be decomposed to
ZnO. And the active surface area of the catalyst decreases with a higher Zn(OAc)2 load. The
catalyst activity decreases accordingly. Additionally, the MPC yield increased with Zn(OAc)2 load
and a maximum value of 93.8% was found for a Zn(OAc)2 load of 30%. Therefore, 30 wt% was
chosen as the Zn(OAc)2 load.
The specific surface area and the pore structure of the Zn(OAc)2/SiO2 catalyst with different
Zn(OAc)2 loadings are summarized in Table 2. The specific surface area, pore volume and pore
4
Page 6 of 29

size decrease with an increase in Zn(OAc)2 loading. This is because Zn(OAc)2and ZnO can block
the pores of SiO2. The X-ray diffraction patterns of the Zn(OAc)2/SiO2 catalysts with different
loadings are shown in Fig. 1. There are no Zn(OAc)2 diffraction peaks detected in the patterns.
This phenomenon may be resulted by the small particle size of Zn(OAc)2 which exceed the lower
detection limit of the XRD apparatus. On the other hand, it may also be the result of the
monolayer dispersion of Zn(OAc)2 on SiO2 surface. Xie et al [21] found that Zn(OAc)2 could
disperse on the surface of SiO2 as a monolayer and the peaks of Zn(OAc)2 crystalline would not
appear in the XRD pattern if the contents of Zn(OAc)2 on SiO2 below the critical dispersion
capacity.
3.2 Effect of calcination temperature on the catalytic performance of Zn(OAc)2/SiO2
To investigate the thermal decomposition of Zn(OAc)2 supported on SiO2, SiO2, Zn(OAc)2 and
Zn(OAc)2/SiO2 (30 wt%, uncalcined) was characterized by TG-DTA and the profiles are shown in
Fig. 2. In the TG-DTA profiles of SiO2 (Fig.2A), there is an obvious weight loss between 50 °C
and 180 °C and the weight loss is about 3.5%, which corresponds to the dehydration of the
catalyst. There are two weight loss processes on the TG-DTA profiles of Zn(OAc)2 (Fig.2B)
accompanied by the absorption of heat. The first stage occurs between 50 °C and 100 °C, which
can be attributed to dehydration of Zn(OAc)2. The second stage begins from about 220 °C to
340 °C and the weight loss is about 73%, which can be attributed to the decomposition of
Zn(OAc)2 for the decomposition temperature of Zn(OAc)2 is about 240 °C [22]. Additionally on
the DTA curve there is an exothermic peak approximately between 320 °C to 450 °C. During the
decomposition of Zn(OAc)2 to ZnO, Timothy et al. [23] found that Zn(OAc)2 was initially
converted to acetic anhydride and ZnO, and then the acetic anhydride was converted into CO2 and
H2O. Accordingly, the exothermic peak is attributed to the degradation of acetic anhydride. There
is an endothermic peak between 50 °C and 200 °C on the DTA profile of Zn(OAc)2/SiO2 (Fig.2C)
accompanied by obvious weight loss on the TG curve, which corresponds to the dehydration of
the catalyst. From 200 °C to 500 °C two weight loss processes are evident on the TG curve. The
first process occurs from 200 °C to 300 °C and the weight loss is only 3%. The first process is
almost the same as that of Zn(OAc)2. The second process occurs from 300 °C to 500 °C and the
5
Page 7 of 29

weight loss is about 10%. Therefore, the thermal stability of most of the supported Zn(OAc)2 is
better than that of Zn(OAc)2. This phenomenon indicates that the dispersion of Zn(OAc)2 is not
good on the SiO2 surface. Some crystals might be in microcrystal form and their decomposition
temperature is similar to that of Zn(OAc)2, while some might have undergone monolayer
dispersion and their decomposition temperature is higher than that of Zn(OAc)2 for the interaction
between Zn(OAc)2 and SiO2. Additionally on the DTA curve there is also an exothermic peak at
approximately 400 °C to 500 °C, which is attributed to the degradation of acetic anhydride.
The effect of calcination temperature on the catalytic performance of Zn(OAc)2/SiO2 (Zn(OAc)2
load was 30 wt%) is shown in Table 1 (entries 4,6–8). The Zn(OAc)2/SiO2 that was calcined at
300 °C exhibited better catalytic performance. At a calcination temperature of 400 °C the catalytic
activity of Zn(OAc)2/SiO2 decreased slightly. However, as the calcination temperature increased
steadily, the aniline conversion and the MPC yield decreased. The mole percent of byproduct,
especially NMA and DMA, increased obviously. The cause is the decomposition of Zn(OAc)2 to
ZnO at higher temperatures and ZnO is known to have less catalytic activity toward MPC
synthesis.
3.3 Catalytic performance and stability comparison between Zn(OAc)2/SiO2 and Zn(OAc)2
Table 3 shows the catalytic performance of Zn(OAc)2/SiO2 and Zn(OAc)2 with the same
amount of Zn. As for Zn(OAc)2, aniline conversion nearly reached 100% at different reaction
time. However, MPC yield increased and a maximum of 96.5% was found for a reaction time of
5 h. Upon prolonging the reaction time continuously MPC yield changed slightly. As for
Zn(OAc)2/SiO2, the aniline conversion and MPC yield increased with reaction time. The
maximal value was obtained with a reaction time of 7h, giving an aniline conversion of 98.1%
and a MPC yield of 93.8%. In addition, the turnover number (TON) over Zn(OAc)2 was almost
the same at different time. Compared with Zn(OAc)2, the TON over Zn(OAc)2/SiO2 increased
with reaction time. When the reaction time was less than 5h, the TON over Zn(OAc)2/SiO2 was
6
Page 8 of 29

lower than that of Zn(OAc)2. Once the reaction time prolonged to 7 h, they were almost the
same.
The stability of Zn(OAc)2/SiO2 was also investigated. After the reaction, the used Zn(OAc)2/SiO2
catalyst was washed with DMC and dried, and was then reused in a following reaction under the
same conditions. The results are shown in Fig. 3. Aniline conversion and MPC selectivity decrease
obviously after second use as the aniline conversion and MPC yield were found to be 73.2% and
57.5%, respectively. However, with further reuse, the aniline conversion and MPC selectivity
decreased but smoothly. After using Zn(OAc)2/SiO2 5 times, the aniline conversion and MPC yield
were found to be 64.3% and 38.1%, respectively.
The stability of Zn(OAc)2 was also investigated for comparison. After Zn(OAc)2 was used once
it showed almost no further activity toward MPC synthesis because aniline conversion and MPC
yield were found to be 9.2% and 1.7%, respectively. Therefore, Zn(OAc)2 is prone to a loss of
activity and cannot be reused.
The deactivated Zn(OAc)2 was characterized by XRD and the result is shown in Fig. 4(b).
Zn(OAc)2 disappears completely and ZnO is formed after single use. ZnO has almost no activity
toward MPC synthesis. We also found that Zn(OAc)2 pretreated with methanol at 170 °C for 3 h
was completely transformed into ZnO. Therefore, the cause of Zn(OAc)2 deactivation is the
formation of ZnO by the reaction between Zn(OAc)2 and methanol. The reaction equation of
Zn(OAc)2 and methanol is as follows [24]:
+
2 CH3OH
Zn(OCOCH3)2
2 CH3COOCH3 + ZnO
+ H2O
Similarly, Zn(OAc)2/SiO2 was pretreated with methanol at 170 °C for 7 h and then the pretreated
Zn(OAc)2/SiO2 was used in the reaction between aniline and DMC. We found that the catalytic
activity of pretreated Zn(OAc)2/SiO2 decreased because the aniline conversion decreased to 67.9%
and the MPC yield decreased to 57.0%, which is close to that of the Zn(OAc)2/SiO2 that was used
a second time. The solution obtained after pretreatment with methanol was analyzed by GC-MS.
The results show that a small amount of methyl acetate is present in the solution, which indicates
that some of the Zn(OAc)2 on the SiO2 surface reacted with methanol to form ZnO.
To understand the cause of Zn(OAc)2/SiO2 deactivation, Zn(OAc)2/SiO2, Zn(OAc)2/SiO2
7
Page 9 of 29

pretreated
with
methanol
(Zn(OAc)2/SiO2-ME)
and
deactivated
Zn(OAc)2/SiO2
(Zn(OAc)2/SiO2-DE) were characterized. The XRD patterns of these catalysts are shown in Fig. 4.
No diffraction peak was present for these catalysts (Fig. 4c, Fig. 4d, Fig. 4e).
FTIR spectra of these catalysts are shown in Fig. 5. The FTIR spectrum of Zn(OAc)2 (Fig. 5a)
has bands at 1557 cm1 and 1448 cm1 that are attributed to the asymmetric and symmetric
stretching vibration of COO in Zn(OAc)2. For the SiO2 spectrum (Fig. 5b), the Si–O–Si
asymmetric and symmetric stretching vibration bands are at 1104 cm1 and 471 cm1. The bands at
804 cm1 and 971 cm1 are attributed to the asymmetric and symmetric stretching vibrations of
O–Si–O [25]. Compared with SiO2 the spectra of Zn(OAc)2/SiO2 (Fig. 5c), Zn(OAc)2/SiO2-ME
(Fig. 5d) and Zn(OAc)2/SiO2-DE (Fig. 5e) show that the band at 971 cm1 broadens toward the
low-frequency side and simultaneously the band at 477 cm1 broadens toward the high-frequency
side, which is caused by the presence of the Si–O–Zn bonds. The Si–O–Zn vibration bands are at
about 940 cm1 and 546 cm1 [26]. The interaction between Zn(OAc)2 and SiO2 results in the
formation of the Si–O–Zn bonds. For Zn(OAc)2/SiO2 (Fig. 5c), the asymmetric stretching
vibration band for COO shifts to a higher frequency and that of the COO- symmetric stretching
vibration broadens toward the low-frequency region. This might come from the interaction
between Zn(OAc)2 and SiO2. The coordination mode of Zn and O in Zn(OAc)2 can be illustrated
as follows where anhydrous zinc acetate adopts a polymeric structure consisting of zinc
coordinated to four oxygen atoms in a tetrahedral environment with each tetrahedron being
connected to neighbors by acetate groups [27].
O
O
O
O
C
H3C
CH3
C
Zn
Once Zn(OAc)2 is loaded onto the surface of SiO2, the coordination mode of the Zn and acetate
groups changes from bidentate coordination to single dentate coordination (illustrated as A and B),
and this comes from the formation of the Si–O–Zn bonds, which release some O atoms in the
acetate groups from the coordination. Accordingly, the COO- stretching vibration band broadens
toward the low-frequency side.
8
Page 10 of 29

Additionally, the band intensity of the COO stretching vibration decreased obviously for
Zn(OAc)2/SiO2-ME (Fig. 5d) and Zn(OAc)2/SiO2-DE (Fig. 5e), which reveals a decrease in the
Zn(OAc)2 content on the surface of the SiO2. There are two causes for this decrease: one is the
formation of ZnO by the reaction between Zn(OAc)2 and methanol and the other is the leakage of
Zn(OAc)2. Therefore, the Zn contents of Zn(OAc)2/SiO2, Zn(OAc)2/SiO2-ME and
Zn(OAc)2/SiO2-DE were analyzed by ICP and the results are 14.2%, 13.4% and 14.6%,
respectively. There is, therefore, no obvious leakage of Zn in Zn(OAc)2/SiO2-DE and
Zn(OAc)2/SiO2-ME. The cause of Zn(OAc)2/SiO2 deactivation is thus the formation of ZnO.
3.4 Reason for the stability of Zn(OAc)2/SiO2
From the aforementioned results, both Zn(OAc)2 and Zn(OAc)2/SiO2 exhibit good catalytic
performance for MPC synthesis. However, Zn(OAc)2 is prone to a loss in activity because of its
reaction with methanol to form ZnO. Once Zn(OAc)2 is supported on SiO2 the stability of
Zn(OAc)2/SiO2 improves considerably.
To determine the reason for the stability of Zn(OAc)2/SiO2 the interaction between Zn(OAc)2
and SiO2 needs to be understood. Zn(OAc)2, SiO2 and Zn(OAc)2/SiO2 were characterized by XPS
and the results are shown in Table 4. Compared with Zn(OAc)2 and SiO2, the binding energy of Si
2p and O 1s in Zn(OAc)2/SiO2 is different. This might be due to the replacement of the Si–O–Si
bonds by Si–O–Zn bonds while Zn(OAc)2 is supported on the SiO2 surface. In addition, the
binding energy of the Zn 2p3 in Zn(OAc)2/SiO2 does not change by comparison with Zn(OAc)2.
Besides The Si–O–Zn bond, the CO–O–Zn bond are also present in the Zn(OAc)2/SiO2 catalyst.
Therefore, it is assumed that the formation of a Si–O–Zn bond has little effect on the binding
9
Page 11 of 29

energy of Zn 2p3.
Many researchers have investigated the reaction between Zn(OAc)2 and methanol to form ZnO
and found that the important intermediate product Zn5(OH)8(Ac)2·2H2O is formed initially. This is
referred to as layered hydroxide zinc acetate (LHZA), which then undergoes hydrolysis and
inorganic polymerization leading to the formation of ZnO [28–30]. When Zn(OAc)2 is supported
on SiO2, the Si–O–Zn bonds are formed by the interaction between SiO2 and Zn(OAc)2.
Compared with free Zn2+ the formation Si–O–Zn bonds imposes steric hindrance on the Zn
present on SiO2, which retards the formation of Zn5(OH)8(Ac)2·2H2O and thus the formation of
ZnO.
Additionally, an interaction exists between methanol and SiO2. SiO2 was pretreated with
methanol at 170 °C for 7 h and characterized by FTIR, and the result is shown in Fig. 5f
Compared with SiO2 (Fig. 5b), three new bands appear in the spectrum of SiO2 pretreated with
methanol. The bands at 2960 cm1 and 2860 cm1 are attributed to the stretching vibration of C–H
in methyl groups while the band at 1465 cm1 is attributed to the bending vibration of C–H in
methyl groups. Therefore, methyl groups are present on the SiO2 surface after SiO2 was pretreated
with methanol. This is due to the dehydration reaction between methanol and the hydroxyl group
on the SiO2 surface and the reaction is as follows [31]:
O
CH3
+ H2O
O
CH3
OH
Si
Si
+
H
From the above analysis, we believe that there are two causes for the stability of the
Zn(OAc)2/SiO2 catalyst. One is the formation of the Si–O–Zn bonds in the Zn(OAc)2/SiO2 catalyst,
which increases the steric hindrance of Zn and consequently retards the reaction between
Zn(OAc)2 and methanol. The other cause is the dehydration of methanol and hydroxyl groups on
the SiO2 surface, which reduces the chance of a reaction between methanol and Zn(OAc)2.
Additionally, Fig. 3 shows that upon Zn(OAc)2/SiO2 catalyst reuse, aniline conversion and MPC
selectivity decreases obviously. However, with further reuse the aniline conversion and MPC
selectivity decreases but smoothly. The cause might be the poor dispersion of Zn(OAc)2 on the
surface of SiO2. Some monolayer dispersion may occur but this is not complete, which might
result in some Zn(OAc)2 not interacting with SiO2 to form the Si–O–Zn bonds and a reaction with
10
Page 12 of 29

methanol may easily result in the formation of ZnO. Therefore, the catalytic performance of
Zn(OAc)2/SiO2 decreases obviously upon reuse. Once this part of Zn(OAc)2 is entirely
transformed to ZnO, the catalytic performance of Zn(OAc)2/SiO2 decreases steadily.
Thus it can be seen that the formation of Si–O–Zn bond plays an important role in the stability of
Zn(OAc)2/SiO2. And to increase the stability of the catalyst, more Si–O–Zn bonds should be
created, which can be achieved by the improved dispersion of Zn(OAc)2 on SiO2 surface. Xie et al
[21] found that spontaneous monolayer dispersion of oxides and salts on supports was a
wide-spread phenomenon. When the content of oxides and salts on supports exceeded the critical
dispersion capacity, there would be a crystalline phase of oxides and salts. The critical dispersion
capacity or the threshold value was almost linearly proportional to the specific surface area of the
support. They also found that Zn(OAc)2 could disperse spontaneously on the surface of SiO2 as a
monolayer at a temperature close to their melting points. So it can be speculated that more
Si–O–Zn bonds can be created using SiO2 with large specific surface area as support and the
stability of Zn(OAc)2/SiO2 catalyst can be improved accordingly.
3.5 A possible reaction mechanism for aniline and DMC over Zn(OAc)2/SiO2 catalyst
The reaction mechanism for the carbamate synthesis by the reaction of amine and DMC over
homogenously Lewis acid catalyst has been investigated widely [33–38]. It is generally believed
that the carbonyl of DMC can interact with nucleophiles for its polarized positive charge and sp2
hybridization of C atom [33, 34]. Over Lewis acid catalyst, there are two different opinions on the
catalytic mechanism. Some people reported that the carbonyl group of DMC could be activated by
the catalyst, which could facilitate the nucleophilic attack accordingly [11,35-37]. And others
thought that the amino group could be activated by the catalyst [38]. Similarly, using Zn(OAc)2 as
catalyst, Baba et al [11] have proposed that the carbonyl of DMC coordinates to Zn2+ to form an
intermediate (Species A). They also reported that methanol which is the by product could help to
-1
transform the bidentate coordination of CH3COO2 to Zn2+ into monodentate coordination and
then facilitated the coordination of DMC to Zn2+. On the contrary, Xiao et al [38] have suggested
that the amino group only can interact with Zn(OAc)2 to form the zinc complex (Species B),
which further react with DMC to produce carbamate.
11
Page 13 of 29

O
C
H
O
C
O
CH3
O
O
O
CH3
Ar
HN
Zn
Zn
H3C
C
O
O
CH3
CH3
NH
Ar
O
C
HO
C
δ-
δ+
Species A
Species B
HO
CH2
As aforementioned in 3.3, there are two modes for the coordination of Zn2+ with acetate group in
Zn(OAc)2/SiO2. With reguard to mode A, the O atom of the carbonyl group in DMC or the N atom
of the amino group in aniline may have the chance to coordinate with Zn2+. However, unlike
Zn(OAc)2, Zn(OAc)2/SiO2 is a heterogeneous catalyst and there exits the resistance for the
reactants to diffuse to the catalyst surface, which can be seen from the results showed in Table 3
that the TON of Zn(OAc)2/SiO2 is lower than that of Zn(OAc)2 in the initial stage of the reaction
and they are almost the same in latter stage. Furthermore, DMC is largely excessive during the
reaction. So it is presumed that more chances for DMC to coordinate with Zn2+ and the reaction
mechanism is illustrated in scheme 2.
As for the mode B, Zn cannot coordinate with O atom or N atom for it has reached the stable
construction of 18 electrons. Thus the mode B does not have the catalytic activity for MPC
synthesis. However, Zn(OAc)2/SiO2 exhibits excellent catalytic performance for MPC synthesis.
Thus the coordination mode of B is few in Zn(OAc)2/SiO2 catalyst.
4. Conclusion
At a Zn(OAc)2 loading of 30 wt% and calcination temperature of 300 °C, Zn(OAc)2/SiO2 shows
excellent catalytic activity toward MPC synthesis with an aniline conversion of 98.1% and a MPC
yield of 93.8%. Moreover, it is more stable than Zn(OAc)2. After 5 cycles of Zn(OAc)2/SiO2 use,
aniline conversion and MPC yield was found to be 64.3% and 38.1%, respectively. The reason for
Zn(OAc)2/SiO2 deactivation is the formation of ZnO by a reaction between methanol and
Zn(OAc)2. For the Zn(OAc)2/SiO2 catalyst, the thermal stability of most supported Zn(OAc)2
improves obviously compared with that of Zn(OAc)2, and this is due to the interaction between
Zn(OAc)2 and SiO2 leading to the formation of a Si–O–Zn bond. In addition, the presence of a
Si–O–Zn bond in the Zn(OAc)2/SiO2 catalyst has an important effect on the improvement of
Zn(OAc)2/SiO2 stability as the steric hindrance of Zn increases and, consequently, the reaction
12
Page 14 of 29

between Zn(OAc)2 and methanol is retarded. Another reason for the improvement in
Zn(OAc)2/SiO2 stability is the dehydration between methanol and the hydroxyl groups on the SiO2
surface, which reduces the possibility of a reaction between methanol and Zn(OAc)2.
Acknowledgements
This work was supported by the National Natural Science Foundation of China (No. 21106031,
21176056) and the Natural Science Foundation of Hebei (No. B2011202139).
References
[1]A. Krogul, J. Skupińska, G. Litwinienko, J. Mol. Catal. A: Chem. 337(2001)9-16
[2]F. Shi, Y. He, D. Li, Y. Ma, Q. Zhang, Y. Deng, J. Mol. Catal. A: Chem. 244(2006)64-67
[3]F. Ragaini, S. Cenini, J. Mol. Catal. A: Chem. 161(2000)31-38
[4]F. Ragaini, Dalton Trans. 32(2009)6251-6266
[5]M. Honda, S. Sonehara, H. Yasuda, Y. Nakagawa, K. Tomishige, Green Chem.13(2011)
3406-3413
[6]B. Chen, S. S. C. Chuang, J. Mol. Catal. A: Chem. 195(2003)37-45.
[7]H. S. Kim, Y. J. Kim, H. Lee, S. D. Lee, C. S. Chin, J. Catal.184(1999)526-534.
[8]A. E. Gurgiolo, U.S. Patent 4,268,683(1981).
[9]Y. Wang, X. Zhao, F. Li, S. Wang, J. Zhang，J. Chem. Technol. Biotechnol. 76(2001)857-861.
[10]W. Xue, J. Cong, F. Li, X. Zhao, Y. Wang, Acta Petrol. Sin. 18(2002)50-55.
[11]T. Baba,A. Kobayashi,Y. Kawanam, K. Inazu, A. Ishikawa, T. Echizenn, K. Murai, S. Asoc, M.
Inomata, Green Chem. 7(2005)159-165.
[12]T. Baba, A. Kobayashi, H. Tanaka, H. Tanaka, S. Aso,M. Inomata, Y. Kawanami, Catal. Lett.
82(2002) 193-197.
[13]X. Zhao, Y. Wang, S. Wang, H. Yang, J. Zhang, Ind. Eng. Chem. Res. 41(2002)5139-5144.
[14]Z. Fu, Y. Ono, J. Mol. Catal. A: Chem. 91(1994)399-405.
[15]T. Baba, M. Fujiwara, A. Oosaku, A. Kobayashi, R. G. Deleon, Y. Ono, Appl. Catal. A:Gen.
13
Page 15 of 29

227(2002)1-6.
[16]N. Katada, H. Fujinaga, Y. Nakamura, K. Okumura, K. Nishigaki, M. Niwa, Catal. Lett.
80(2002)47-51.
[17]W. Kang, T. Kang, F. Ma, Y. Zhao, L.Cui, J. Yao, G. Wang, Chin. J. Catal. 28(2007)5-9.
[18]Q. Li, J. Wang, W. Dong, M. Kang, X. Wang, S. Peng, Chin. J. Catal. 24(2003) 639-642.
[19]F. Li, J. Miao, Y. Wang, X. Zhao, Ind. Eng. Chem. Res. 45(2006)4892-4897.
[20]F. Li, Y. Wang, W. Xue, X. Zhao, J. Chem. Technol. Biotechnol. 84(2009) 48-53.
[21] Y. Xie , Y. Tang, Advan. Catal. 37(1990)1-43
[22]M. Ohyama, H. Kozuka, T.Yoko, Thin Solid Films. 306 (1997) 78-85
[23]T. Biswick, W. Jones, A. Pacula, E. Serwicka, J. Podobinski, Solid State Sci.
11(2009)330-335.
[24]P. Tonto, O. Mekasuwandumrong, S. Phatanasri, V. Pavarajarn, P. Praserthdama, Ceram. Int.
34(2008)57-62
[25]P. A. Tanner, B. Yan, H. Zhang, J. Mater. Sci. 35(2000)4325-4328
[26]A. Roy, S. Polarz, S. Rabe, B. Rellinghaus, H. Zähres, F. E. Kruis, M. Driess, Chem.-A Eur. J.
10(2004)1565-1575
[27]W. Clegg, I. R. Little, B. P. Straughan, Acta. Cryst. C42 (1986) 1701-1703
[28]E. Hosono, S. Fujihara, T. Kimura, H. Imai, J. Colloid Interface Sci. 272(2004)391-393
[29]L. Znaidi, Mater. Sci. Eng. B. 174(2010)18-30
[30]E. Hosono, S. Fujihara, T. Kimura, H. Imai, J. Sol-gel Sci. Technol.29(2004)71-79
[31]T. R. Forester, R. F. Howe, J. Am. Chem. Soc. 109(1987)5076-5082
[32] Y. Xie, Y. Tang, Advan. Catal. 37(1990) 1-43
[33]P. Tundo, L. Rossi, A. Loris, J. Org. Chem. 70(2005)2219-2224
[34]P.Tundo, S. Bressanello, A. Loris, G. Sathicq, Pure Appl. Chem. 77(2005)1719-1725
[35]M. Distaso, E. Quaranta, Appl. Catal. B: Environ. 66 (2006) 72-80
[36]X. Fu, Z. Zhang, C. Li, L. Wang, H. Ji, Y. Yang, T. Zou, G. Gao, Catal. Commun. 10 (2009)
665-668
[37]L. Zhang, Y. Yang, Y. Xue, X. Fu, Y. An, G. Gao, Catal. Today. 158 (2010) 279-285
[38]F. Xiao, D. Zhang, Q. Dong, W. Wei,Y. Sun, React.Kinet.Catal.Lett. 89(2006)131-138
14
Page 16 of 29

Scheme 1 Reactions for the synthesis of MDI
Scheme 2 Reaction mechanism of aniline and DMC catalyzed by Zn(OAc)2/SiO2
Table 1 Effect of loading amount and calcination temperature on catalytic performance a
Table 2 Specific surface area, pore volume and pore size of the Zn(OAc)2/SiO2 catalysts
Table 3 Catalytic activity comparison between Zn(OAc)2/SiO2 and Zn(OAc)2
Table 4 Binding energy of Zn 2p3, Si 2p and O 1s in different catalysts
Fig. 1 XRD patterns of the Zn(OAc)2/SiO2 catalyst with different loadings.
Fig. 2 TG-DTA profiles of SiO2, Zn(OAc)2 and Zn(OAc)2/SiO2.
A-SiO2; B-Zn(OAc)2; C-Zn(OAc)2/SiO2;
.
Fig. 3 Stability of the Zn(OAc)2/SiO2 catalyst.
Fig. 4 XRD patterns of the different catalysts.
a-Zn(OAc)2, b-Zn(OAc)2 after the reaction
c-Fresh Zn(OAc)2/SiO2 catalyst; d-Zn(OAc)2/SiO2-ME; e-Zn(OAc)2/SiO2-DE
Fig. 5 FTIR spectra of the different catalysts.
a-Zn(OAc)2; b-SiO2; c-Zn(OAc)2/SiO2; d-Zn(OAc)2/SiO2-ME; e- Zn(OAc)2/SiO2-DE; f- SiO2 pretreated with
methanol
15
Page 17 of 29

Figure1
Page 18 of 29

Figure2A
Page 19 of 29

Figure2B
Page 20 of 29

Figure2C
Page 21 of 29

Figure3
Page 22 of 29

Figure4
Page 23 of 29

Figure5
Page 24 of 29

scheme 1
Page 25 of 29

scheme 2
Page 26 of 29

Table 1 Effect of loading amount and calcination temperature on catalytic performance a
Amount of Zn in catalyst
Product distribution (mol percent %)
Zn(OAc)2 load
(wt%)
Calcination temperature Aniline conversion MPC yield
Entryb
(°C)
(%)
(%)
(mol)
NMA
56.9
MPC
28.8
DPU
1.2
DMA
13.1
3.1×10-5
1.5×10-4
3.1×10-4
4.6×10-4
5.4×10-4
4.6×10-4
4.6×10-4
4.6×10-4
1
2
3
4
5
6
7
8
2
300
300
300
300
300
400
500
600
37.1
95.4
98.4
98.1
5.2
10
20
30
35
30
30
30
87.7
87.8
93.8
7.2
5.4
4.9
90.6
92.8
93.5
0.2
0.1
0.1
2.0
1.7
1.5
97.0
98.0
90.2
89.5
5.7
6.1
92.6
90.7
0.1
0.1
1.6
3.1
83.5
36.6
64.4
13.3
17.2
49.9
74.5
35.4
0.5
0.7
7.8
14.0
a Reaction conditions: catalyst 0.28 g, DMC 0.2 mol, aniline 0.01 mol, 170 °C, 7 h.
b The catalyst was calcined for 2 h.
1
Page 27 of 29

Table 2 Specific surface area, pore volume and pore size of the Zn(OAc)2/SiO2 catalysts
SBET (m2·g−1)
297.5
Pore volume (cm3·g−1)
Zn(OAc)2 load (wt%)
Pore size (nm)
0
1.8
1.7
1.6
1.4
1.3
1.3
25.0
23.6
23.5
21.9
21.2
18.5
2
289.9
10
20
30
35
254.8
223.6
193.5
189.6
Table 3 Catalytic activity comparison between Zn(OAc)2/SiO2 and Zn(OAc)2
Amount of Zn in catalyst
(mol)
Reaction time Aniline conversion MPC yield
Catalysts
Zn(OAc)2
TON c
(h)
3
(%)
98.7
98.8
99.0
92.5
97.9
98.1
(%)
92.7
96.5
95.4
80.7
89.8
93.8
20.2
21.0
20.7
17.5
19.5
20.4
a
4.6×10-4
4.6×10-4
5
7
3
b
Zn(OAc)2/SiO2
5
7
a Reaction conditions: Catalyst: 0.084 g, DMC: 0.2 mol, aniline: 0.01 mol, temperature: 170 °C.
b Reaction conditions: Catalyst (Zn(OAc)2 loading: 30 wt%, calcination temperature: 300 °C): 0.28 g, DMC: 0.2
mol, aniline: 0.01 mol, temperature: 170 °C.
c TON = mol MPC/mol Zn
2
Page 28 of 29

Table 4 Binding energy of Zn 2p3, Si 2p and O 1s in the different catalysts
Catalysts
Zn(OAc)2
SiO2
Zn 2p3
1022.2eV
-
Si 2p
-
O 1s
531.9eV
534.7 eV
532.4eV
105.7 eV
103.2eV
Zn(OAc)2/SiO2
1022.2eV
3
Page 29 of 29