Optimization for catalytic performances of Hβ zeolite in the acylation of 2-methylfuran by surface modification and solvents effect

Article abstract of DOI:10.1007/s11164-016-2715-4

The liquid phase acylation of 2-methylfuran with acetic anhydride over modified Hβ zeolite was first conducted in a continuous flow reactor. The deactivation of Hβ zeolites was attributed to strong adsorption of reactants or products and was verified by GC–MS and ¹³C MAS NMR. Deactivated zeolites can be regenerated to their original state by calcination. The acidic properties was adjusted by surface modification on Hβ, the maximum yield of 89.5?mol% and selectivity of 100?% were obtained over tartaric acid modified by Hβ. The deposition of tetraethoxysilane to silica on Hβ contributed to enhancing the catalytic stability. Combined with the results of NH₃-TPD and Py-FTIR, the amount of Broensted acids played a major role on catalytic activity. A close relationship between the catalytic stability and the ratio of the amount of strong to weak acids at 1:1 was highlighted here. The solvents' effect on the catalytic performances was examined, and 1,2-dichloroethane with moderate polarity exerted a positive effect on catalytic stability.

Full text of DOI:10.1007/s11164-016-2715-4

Res Chem Intermed
DOI 10.1007/s11164-016-2715-4
Optimization for catalytic performances of Hb zeolite
in the acylation of 2-methylfuran by surface
modification and solvents effect
Yuannan Xiong1,2 Wenqi Chen1,2 Aiwu Zeng1,2
•
•
Received: 3 July 2016 / Accepted: 27 August 2016
Ó Springer Science+Business Media Dordrecht 2016
Abstract The liquid phase acylation of 2-methylfuran with acetic anhydride over
modified Hb zeolite was first conducted in a continuous flow reactor. The deacti-
vation of Hb zeolites was attributed to strong adsorption of reactants or products and
was verified by GC–MS and 13C MAS NMR. Deactivated zeolites can be regenerated
to their original state by calcination. The acidic properties was adjusted by surface
modification on Hb, the maximum yield of 89.5 mol% and selectivity of 100 % were
obtained over tartaric acid modified by Hb. The deposition of tetraethoxysilane to
silica on Hb contributed to enhancing the catalytic stability. Combined with the
results of NH3-TPD and Py-FTIR, the amount of Broensted acids played a major role
on catalytic activity. A close relationship between the catalytic stability and the ratio
of the amount of strong to weak acids at 1:1 was highlighted here. The solvents’
effect on the catalytic performances was examined, and 1,2-dichloroethane with
moderate polarity exerted a positive effect on catalytic stability.
Keywords Acylation Á Hb zeolite Á Surface modification Á Catalytic performances Á
Solvents effect
Introduction
Friedel–Crafts acylation of aromatic substances is a well-renowned reaction in the
fine chemical industry [1–4]. The obtained aromatic ketones are widely utilized as
valuable intermediates in the fabrication of fine and special chemicals, such as
&
Aiwu Zeng
awzeng@tju.edu.cn
1
2
Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), State Key
Laboratory of Chemical Engineering, Tianjin, People’s Republic of China
Tianjin University, Institute of Chemical Engineering and Technology, Tianjin 300072,
People’s Republic of China
123

Y. Xiong et al.
drugs, pesticides, elasticizers, dyestuffs, and perfumes [5, 6]. As far as we know,
Broensted acids (e.g., HCl, H2SO4) and Lewis acids (e.g., AlCl3, BF3) are often used
as homogeneous catalysts for the conventional acylation synthesis. Unfortunately,
the inherently encountered defects of such a manufacturing process basically cover
the handling of toxic wastes, equipment corrosion, and products collection [7]. With
increasing pressure on environmental and economic issues, the exploration of
sustainable and available methodologies for the acquisition of high-valued products
with no environmental pollution is a major challenge for our research community.
Toward this purpose, some new and efficient heterogeneous catalysts such as
zeolites [8–10], heteropolyacids [11], clays [12] and metal hydroxide fluorides [13]
are sought to effectively figure out the problems mentioned above. Particularly,
zeolites are extensively used due to their high activity, renewability and eco-friendly
nature, which have been proven to be the most potential catalysts in the catalytic
synthesis field [14]. The convenient applications of acidic zeolites for the Friedel–
Crafts acylation of aromatic and heterocyclic aromatic compounds have been
declared by several research groups [15–17]. For instance, Guignard et al. [18]
recently announced the first commercial applications of zeolites for the acylation of
´
veratrole and anisole. As it was published by Alvaro et al. [19], the high yields of
91 mol% of 2-acetylfuran were effectively synthesized over Hb zeolite under mild
reaction conditions. Nevertheless, zeolites are easily inclined to the fast deactivation
when serving as catalysts for acylation. The deactivation phenomenon that results
from strong adsorption of aromatic substrates or products has been witnessed by
several reports [20, 21]. Therefore, the behavior of Hb zeolite must be optimized
further for dealing with both the problems of catalytic activity and stability, which
are prerequisites for application in industrial production of aromatic ketones. It is
deservedly noted that the Broensted acids on Hb zeolite can account for its high
activity in catalytic synthesis. Reddy et al. [22] reported that the liquid phase
acylation reaction of pyrrole and furan were more positive on Broensted acids of Hb
zeolite. Consequently, a connection between the catalytic performances and acid
sites (both Broensted and Lewis acids) of zeolites should be emphasized. It was
recently found by Zhao et al. [23] that an appropriate balance between the weaker
and stronger acid amounts on the surface of Hb zeolite can enhance the catalytic
stability. There have been several approaches to the optimization of the Friedel–
Crafts acylation reaction. An essential step in preparing the highly active and
stable zeolite is the modification of its surface. From those results reported before
[24–28], we have a solid foundation for believing that the surface modification
using acid treatment and chemical liquid deposition can bring about many benefits,
offering the possibility to tune the intensity and distribution of acid sites.
The acylation of 2-methylfuran with acetic anhydride in liquid phase was used as
a model reaction to study the effects of surface modification, and to examine the
effect of the solvents polarity on the catalytic performances of Hb zeolite. Although
the fast deactivation was still observed by GC–MS and 13C MAS NMR after
acylation reaction, we provided some efficient approach to optimize the catalytic
stability. In our work, Py-FTIR and NH3-TPD analysis were employed to
investigate the relationship between acid sites and catalytic performance of Hb
zeolite. The objective of this study was to enhance the catalytic performance of Hb
123

Optimization for catalytic performances of Hb zeolite in…
zeolite in an acylation reaction through surface modification and solvent effects.
Our contribution to this vast field is that high yield and good catalytic stability were
attained in the acylation of 2-methylfuran under mild conditions. With the catalytic
result of modified Hb zeolite, the modification of the mechanism concerning
Broensted acids and the intensity distribution of acid sites of Hb zeolite was
discussed in detail, which is rarely found in other studies.
Experimental
Materials
2-Methylfuran (2-MF, [98 %) and Tetraethoxysilane (TEOS, [99 %) were
purchased from Energy Chemical Technology Co., Ltd. Acetic acid (99.5 %),
tartaric acid (99.5 %), and nitric acid (65–68 %) were supplied by Tianjin Guangfu
Institute of Fine Chemistry. Hexane (98 %) and acetic anhydride (AA, 98.5 %)
were supplied by Tianjin Jiangtian Chemical Technology Co., Ltd. 2-Acetyl-5-
methylfuran (2-AC-5-MF, [98.0 %) standard and tetraethylammonium hydroxide
(TEAOH) were provided by Aladdin Reagent Co., Ltd. The commercial Hb zeolite
and aluminum oxide powder were procured from Nankai University Catalyst Co.,
Ltd. All chemical reagents were put to use without further purification.
Catalyst preparation and modification
Firstly, 50 g of Hb zeolite powder with acid impregnations in an airtight container,
was stirred with 500 mL of acetic acid (1 mol L-1) at ambient temperature. After
above mixture was stirred for 10 h, the zeolite sample was recovered by filtration
and washed out with distilled water until the solution turned into neutral. Secondly,
the acid-treated Hb was prepared on the basis of the work of Matsukata et al. [29].
The mixture of 10 g zeolite powder, 3 g aluminum oxide and 4 g nitric acid solution
(10 wt%) was agitated into homogeneity, squeezed, and trimmed into the shape of a
short cylinder. Thirdly, the molded sample was dehydrated at 120 °C in dry air for
2 h, and calcination progress was made in a muffle furnace with temperature
programmed to 550 °C for 4 h at a heating rate of 20–30 °C min-1. The obtained
catalyst was called AC-Hb. Moreover, tartaric acid (1 mol L-1) modified Hb
zeolites were prepared by the abovementioned method and designated as TA-Hb.
Approximately 20 g of Hb zeolite (named parent Hb) was suspended in 50 mL
of hexane, then 30 mL of tetraethoxysilane (TEOS) was added into this mixture.
After stirring for 10 h, hexane was removed from its solution by evaporation, and
the mixture was filtered further and dried at 120 °C. The Hb zeolite supported with
TEOS was roasted at 550 °C for 4 h. Thus, the SiO2 chemical liquid deposition
(SiO2-CLD) modified Hb zeolite was attained and labeled as Si-Hb. The carbon
deposits absorbed on the external surface and in the micropores of deactivated
zeolites were recovered by the soxhlet extraction method. After acylation reaction,
the deactivated catalyst was treated in a soxhlet extractor for 8 h using
123

Y. Xiong et al.
dichloromethane as solvent. Then the dichloromethane was completely evaporated
at 60 °C and the extracted organic compounds were obtained.
Catalyst characterization
The X-ray diffraction (XRD) powder patterns were obtained on a Bruker D8-S4
diffractometer operated at 50 mA and 40 kV with a Cu Ka radiation source,
scanning from 4° to 30° range with a scanning rate of 3.21°/min and steps = 2605.
The framework Si/Al molar ratios of zeolites were analyzed by X-ray fluorescence
(XRF) spectroscopy recorded on a Bruker S4 Pioneer spectrometer. The morphol-
ogy of zeolites was characterized by a Hitachi S-4800 Scanning electron microscope
(SEM) with high resolution. The pore diameter distribution and surface area were
achieved on a Micrometrics ASAP 2020 volumetric instrument at -196 °C.
The acid nature of zeolites was detected on a Nicolet 750 infrared spectrometer
through the pyridine adsorption–desorption method. A self-supported wafer was
loaded into an infrared cell coupled with the conventional high vacuum system.
Prior to pyridine adsorption, 25 mg of zeolites were pretreated at 450 °C for 2 h in
the vacuum. When the self-supported wafer was cooled to ambient temperature,
pyridine adsorption proceeded at 60 °C for 1 h. Then, the IR spectra were collected
by the evacuation of adsorbed pyridine at 200 and 350 °C for 0.5 h, respectively.
The information about the surface chemistry of zeolites was gained by FTIR
techniques and recorded on a Bruker TENSOR 37 instrument with a resolution of
4 cm-1
.
The ammonia temperature-programmed desorption (NH3-TPD) for the acids
amount and intensity of zeolites was carried out on a TP-5080 instrument with a
combined thermal conductivity cell detector. First, around 10 mg of zeolite catalyst
was pretreated in situ at 200 °C for 1 h in a flow of helium (50 mL min-1), and then
the NH3 was adsorbed at 60 °C for 0.5 h. The NH3 desorption was taken at the
range of 120–800 °C with a heating rate of 10 °C min-1. The 13C NMR spectra
were acquired on a Varian Infinity plus NMR spectrometer to study the types of
carbon deposits. The scanning frequency and pulse delay were 75.4 MHz and 5.0 s,
respectively. The chemical shifts (d) of spectra with units of ppm stood for the types
of carbon deposits.
Catalytic tests
Continuous acylation reactions of 2-methylfuran (2-MF) by acetic anhydride (AA)
over parent Hb, AC-Hb, TA-Hb, Si-Hb zeolite were conducted on a fixed-bed
reactor (Fig. 1) under atmospheric pressure, consisting of a glass tube (inner
diameter 10 mm) with an electrical heater. Before the acylation reaction, the same
amounts of zeolites were activated at 200 °C in a drying oven for 2 h. The zeolites
were transferred immediately into the reactor to prevent exposure to the atmosphere.
For a standard experiment, acylation temperature was controlled at 60 °C by water
bath heating at the first step. Then, 5.9 g of zeolite catalyst was fixed in the central
section of a fixed-bed reactor. The mixture of 2-MF and AA (molar ratio 1:2.5 or
1:4) was fed to the catalytic column by a metering pump with a given flow rate
123

Optimization for catalytic performances of Hb zeolite in…
Fig. 1 Apparatus diagram of a fixed-bed reactor used for continuous liquid phase acylation of
2-methylfuran
(0.07 mL min-1). The liquid acylated products were periodically collected from the
reactor outlet, and quantitatively analyzed on a gas chromatograph (GC, Agilent
7890A) equipped with the flame ionization detector (FID) and DB-FFAP capillary
column (length 30 m, phase thickness 0.25 lm, diameter 0.25 mm). The target
acylated product 2-acetyl-5-methylfuran (2-AC-5-MF) was obtained from acylation
reaction, and identified by a mass spectrometer (MS, Agilent 5975C) with the mass
selective detector.
The molar yield (mol%) of 2-acetyl-methylfuran (Y2-AC-5-MF) was calculated as
follows:
%mol ð2-AC-5-MF at time ¼ tÞ
Y2-AC-5-MF
¼
 100:
ð1Þ
%mol ð2-MF at time ¼ 0Þ
Results and discussion
Characterization of zeolites
XRD patterns (in Fig. 2) of the parent Hb, AC-Hb, TA-Hb, and Si-Hb are given.
The characteristic diffraction peaks of all zeolite samples were at about 7.8° and
22.5°, indicating that the acid treatment and SiO2-CLD modification had not
changed the basic structure of Hb zeolite, and there are very few visible impurity
peaks. However, the XRD peaks at 7.8° and 22.5° of AC-Hb, TA-Hb and Si-Hb
were weaker than that of the parent Hb and attributed to surface modification.
According to the results of XRF as recorded in Table 1, the framework Si/Al of
parent Hb, AC-Hb, TA- Hb, Si-Hb zeolites used in this work was 27.6, 28.2, 62.9,
and 86.8 respectively, implying that the dealumination occurred obviously on
zeolites in the acid treatment. However, the increase of Si/Al molar ratios in
123

Y. Xiong et al.
Fig. 2 XRD patterns of parent and modified Hb zeolites
SiO2-CLD is mainly ascribed to the deposition of TEOS modifier on the surface of
zeolite. The result in Table 1 presents the specific surface areas (including external
and micropore) and average pore diameter of zeolites. The BET method was used to
determine the specific surface area, and the BJH approach was used to calculate the
pore diameter distribution. In comparison with Hb zeolite, there are certain
increases both in the outer surface area and average pore diameter for acid-modified
Hb zeolites. Particularly, the largest surface area and average pore diameter was
observed on TA-Hb. As a consequence of modification, the acid treatment can
remove the non-framework impurities existing in Hb zeolite micropores, resulting
in increase in specific area. This suggests that the dealuminzation through the acid
treatment can be in favor of the transfer of target products (2-AC-5-MF). However,
SiO2-CLD modification of Hb zeolite led to a small decline in its average pore
Table 1 The framework Si/Al molar ratios, BET surface area, micropore volume, and average pore
diameter of parent Hb and modified Hb zeolites
Sample
Surface areaa (m2/g)
Average pore
diameterb (nm)
Si/Al molar
ratiosc
Total
Micropore
External
Parent
AC-Hb
TA-Hb
Si-Hb
a
365
376
427
351
237
235
222
232
128
142
205
118
0.55
0.61
0.76
0.52
27.6
28.2
62.9
86.8
Surface area calculated by the BET and t-plot method
b
c
Average pore diameter determined by the Horvath–Kawazoe method
Si/Al molar ratios measured by XRF
123

Optimization for catalytic performances of Hb zeolite in…
diameter and external surface area, which can adjust the acidic intensity distribution
of zeolites, similar to the previous report [30].
In order to explore the modification mechanism regarding the acid sites with the
amount and distribution of Hb zeolite, some analytical methods were done in this
study. IR was widely applied to detect the properties of zeolite hydroxyls. Py-FTIR
was also adopted to study the effect of acid treatment and SiO2-CLD modification
on the acid properties. As released by Emeis [31], the concentration of surface acid
sites can be calculated by the integrated peak area of adsorbed pyridine. The
integrated molar extinction coefficients of Broensted acid sites and Lewis acid sites
bands were 1.67 and 2.22 cm/lmol, respectively. The information of Lewis and
Broensted acid sites on Hb zeolite was gained by Py-FTIR through the pyridine
adsorption–desorption method, summarized in Table 2. In a typical NH3-TPD run,
the quantity of desorbed NH3 can be regarded as the acidic quantity of zeolites, and
the acidic intensity was classified by desorption temperature. The results from NH3-
TPD analysis for different zeolites are listed in Table 3.
Effect of acids modification on catalyst properties
From the catalytic point of view, the activity of zeolite can be connected with the
amount of both Broensted and Lewis acid sites in Friedel–Crafts acylation. As
already been pointed out, the total number and nature of acid sites should have a
direct effect on the activity and stability of catalyst [32, 33]. Acid treatments are
utilized to remove the extra-framework Al sites to increase the external surface of
zeolite crystals, which may contribute to improve the adsorption capacity of
reactants. To study the effect of the acid modification, the catalytic performances of
different acid-modified Hb were tested in the acylation of 2-MF. 2-AC-5-MF was
the only product observed in this acylation, so the selectivity for 2-AC-5-MF was
almost 100 %. As shown in Fig. 3, the highest activity and stability are obtained on
TA-Hb catalyst, the yield of 2-AC-5-MF at the beginning is 89.5 mol%, and TA-Hb
can maintain a high yield plateau (C70 mol%) for 45 h in the acylation of 2-MF.
Nevertheless, AC-Hb exhibited the lower initial yield of 82.3 mol%, besides the
yield of 2-AC-5-MF only can remain at about 70 mol% after 12.5 h on stream,
Table 2 Acid site properties and amounts determined by Py-FTIR spectra for parent and modified Hb
zeolites at desorption temperatures of 200 and 350 °C
Samples
200 °C
350 °C
L acid site
B acid site
B/La
L acid site
B acid site
B/La
Parent
AC-Hb
TA-Hb
Si-Hb
a
4.863
4.531
3.668
2.928
0.663
0.547
0.836
0.605
0.136
0.121
0.228
0.207
4.263
3.984
3.152
2.475
0.589
0.506
0.673
0.498
0.138
0.127
0.214
0.201
B/L: the ration of Broensted to Lewis acid site amounts, measured by the ratio of the integrated area of
FTIR peaks to sample weight
123

Y. Xiong et al.
Table 3 NH3-TPD data of parent and modified Hb zeolites
Samples
Weak acidic sites
Strong acidic sites
Acid distribution
L T-peaka
Weak acid
(mmol/g)
H T-peakb
Strong acid
(mmol/g)
Total acid
(mmol/g)
S/Wc
(°C)
(°C)
Parent
TA-Hb
AC-Hb
Si-Hb
206
203
195
202
0.630
0.553
0.658
0.294
0.532
300
315
302
320
298
0.413
0.567
0.371
0.308
0.210
1.043
1.120
1.029
0.742
0.602
0.656
1.025
0.564
1.048
0.395
Deactivated 190
a
L T-peak: low-temperature peak
H T-peak: high-temperature peak
b
c
S/W: the ration of strong acid to weak acid number
Fig. 3 Catalytic performances tests for acylation of 2-methylfuran with acetic anhydride over parent Hb,
TA-Hb and AC-Hb zeolites in a fixed bed reactor (experimental conditions: 60 °C, 2-methylfuran/acetic
anhydride molar ratio = 1:2.5, flow rate: 0.07 mL/min, 5.9 g of zeolites)
which is similar to that of the parent Hb with 13.5 h. This indicates that a serious
catalyst deactivation occurred on AC-Hb and the parent Hb.
Table 2 shows that a decrease in the concentration L acid sites because of the
removal of extra-framework aluminum through the acid treatment. However,
tartaric acids as a special chelating agent can further react with the removed
aluminum, converting into the more stable B acid sites. Meanwhile, the more
stable B acid sites were replenished into the framework structures of Hb zeolite,
thus leading to the increase in B acid sites. What is interesting is that TA-Hb
exhibited higher catalytic stability with the greater amount of Broensted but with
less Lewis acid sites. AC-Hb with the lower activity in acylation can be ascribed to
the lower amounts of Broensted acids (Table 2, line 2) in comparison with parent
123

Optimization for catalytic performances of Hb zeolite in…
Hb. It seems that Broensted acids amounts are of great importance in accounting for
their behaviors in acylation reactions. NH3-TPD tests were carried out to measure
the amounts and types of acid sites of zeolites. The amount of acid sites was
determined on the basis of the peak areas of TPD, and listed in Table 3. The peak at
the range of about 150–200 °C was ascribed to the weaker acids and the peak
approximately at 300–400 °C to the stronger acids announced by Ghiaci et al. [34].
As shown in Fig. 4, by contrast, there was a decrease both in the weaker acid and
stronger acid amounts in TA-Hb and AC-Hb. Nevertheless, TA-Hb shows the better
behavior with the strong to weak acids ratio of nearly 1:1 (Table 3, line 3), and more
appropriate distribution of acid intensity. It implies that the catalytic stability
primarily depended on acid intensity distribution not at large acid amounts.
Effect of SiO2-CLD modification on catalyst properties
For purpose of gaining better performance, some available modification methods
have been utilized to optimize catalysts. Considering some benefits of chemical
liquid deposition (CLD) [35, 36], the deposition of tetraethoxysilane to silica on its
external surface can increase the hydrophobic properties of zeolite, and enhance the
adsorption capacity of organic reactants. It is worthy to study the effect of CLD of
silicates on catalytic performances in our work. The deposition of TEOS modifier to
SiO2 over Hb zeolite was also done in our study. IR technique was put to use to
distinguish the features of zeolite hydroxyls. In line with the previous publication
[37], the framework bridging Al(OH)Si was assigned to the band at
3600–3610 cm-1, Al(OH) of extra-framework was assigned to the band at about
3660 cm-1, while silanol Si(OH) of zeolites was assigned to the band approxi-
mately at 3740 cm-1. As depicted in Fig. 9, the acidic hydroxyls of Hb consist of
the framework bridging Al(OH)Si, extra-framework Al(OH) and silanol group
Si(OH). The silanol group Si(OH) initiates dropping without calcination in SiO2-
Fig. 4 NH3-TPD profiles of the parent Hb and acid modified Hb zeolites
123

Y. Xiong et al.
CLD treatment. Whereas, the amount of Si(OH) at 3740 cm-1 diminished merely
after calcination. Thus, it can be concluded that TEOS reacted with the hydroxyls of
zeolite during the chemical liquid deposition, and thus covered the external surface
of Hb. Besides, the band at about 2965–3070 cm-1 is related to the C-H bond of the
modifier TEOS deposited on surface of Hb zeolite. In contrast to the parent Hb
(Fig. 5), it reveals that the C-H bond can disappear after calcination treatment,
suggesting that TEOS can decompose into silica completely in the SiO2-CLD.
The catalytic performances of Si-Hb and parent Hb were compared in the
acylation of 2-MF with AA. In fact, we can see that SiO2 mainly covered the
external surface of Hb during the chemical liquid deposition, but it is interesting to
find that Si-Hb showed the better behavior. Thus, we might declare that the
acylation of 2-MF with AA primarily took place in the micropores of Hb zeolite. As
shown in Fig. 6, the average lifetime (yield of 70 mol% as benchmark) and yield of
Si-Hb increased by 10 h, and 10 mol%, respectively, compared to the parent Hb.
Moreover, the NH3-TPD profile (Fig. 7) indicates that the strong acids amount of
Si-Hb is far less than that of parent Hb. It should be the coverage of silica on the
outer surface of Hb that results in the reduction of acid amounts and silanol
hydroxyls. The SiO2-CLD modification on Hb led to a decrease of acid sites, but
also a more reasonable acid intensity distribution with the nearly equal amounts of
strong and weak acid (Table 3, line 4). In terms of such results, it may be speculated
that the appropriate acid intensity distribution is the key to maintaining the catalytic
stability.
Fig. 5 IR spectra of parent Hb and modified Hb zeolites in SiO2-CLD before and after calcination
treatment
123

Optimization for catalytic performances of Hb zeolite in…
Fig. 6 Acylation results of 2-methylfuran by acetic anhydride in a fixed bed reactor over the parent Hb,
modified Hb with SiO2-CLD (experimental conditions: 60 °C, 2-methylfuran/acetic anhydride molar
ratio = 1:4, flow rate: 0.07 mL/min, 5.9 g of zeolites)
Fig. 7 NH3-TPD profiles of the parent Hb and Si-Hb zeolites
Catalytic performances
To illustrate the effects of surface modification on the catalytic performances, we
herein compared the catalytic results of modified and parent Hb zeolites.
Concerning acid treatment effects, TA-Hb exhibited the more stable catalytic
performances shown in Table 1, it was found that the total BET surface areas and
average pore diameters of TA-Hb were larger than those of AC-Hb and parent Hb.
These results might be explained by the dealumination effect due to acid
123

Y. Xiong et al.
modification, which also is confirmed by the increase in Si/Al ratio. With regard to
the SiO2-CLD modification, the increase of Si/Al molar ratios in SiO2-CLD is
mainly ascribed to the deposition of TEOS modifier on the surface of zeolite. It
should be noticed that the catalytic performances improved with the increasing
molar ratios of Si/Al to some extent.
The modifications on the acid properties of Hb zeolites were caused the different
behavior involving inthe acylation process. Such a close correlation between the
catalytic performance and the amount of Broensted acids on zeolites was highly
stressed. Unfortunately, a relation was not observed concerning the Lewis acids in
the acylation of 2-methylfuran. In this way, the amount of Broensted acids should be
the restrictive factor, and there are much less Broensted than Lewis acids on each
zeolite catalyst. NH3-TPD results for all zeolites were given in Table 3, and clearly
explain that the higher catalytic stability is well associated with the more
appropriate distribution of acid intensity. Indeed, the strong to weak acids ratio of
TA-Hb, and Si-Hb was almost 1:1, while parent Hb, AC-Hb were not.
Consequently, we have reason to believe that larger numbers of total acid sites of
zeolites would not contribute to the higher catalytic performance, which was
verified by AC-Hb with terrible catalytic stability in acylation of 2-MF. The desired
zeolite catalysts do not have a larger number of acid sites but must with the
appropriate acid intensity distribution, similar to the results published by Zhao et al.
[23].
Informed by the previous report [29], the reusability of zeolites is also a sign of
catalytic stability of zeolites. The reusability of parent Hb, AC-Hb, TA-Hb, and Si-
Hb were investigated in the same acylation of 2-MF with AA, and the results are
given in Table 4. As can be seen, the regenerated zeolites almost exhibited the
identical stability with the same reaction time, and activity with the maximum
yields as their origins. It also shows that zeolites can be recycled by high
temperature calcination. As expected, the parent and modified Hb still remained at
the same time on stream before the deactivated point (yields below 70 mol%), and
catalytic activity (yields of 2-MF). After the eight cyclic regenerations, the average
lifetimes of TA-Hb and Si-Hb were still longer than that of the parent Hb and AC-
Hb.
Origin of deactivation of zeolites
As can be seen from the SEM image (Fig. 11), there are severe carbon deposits
existing on the surface of the deactivated Hb zeolites (the yields of 2-AC-5-MF
below 70 mol%), which might block the acidic sites of the catalyst, leading to the
decrease of the reaction rate. Moreover, the Hb zeolite became dark brown at the
end of the acylation reaction. Figure 7 shows that the yield of 2-AC-5-MF decreased
obviously with increasing time on stream, which means the fast catalyst decay
mainly occur in the acylation of 2-MF. The 13C MAS NMR spectrum of the
deactivated Hb is given in Fig. 8. The spectral feature of deactivated Hb is
associated with the species of carbon deposits, composed of fat carbon, olefin
carbon, aromatic carbon and poly aromatic carbon, etc. [38]. The fat carbon
corresponds to the band at 10–40 ppm, the aromatic carbon is assigned to a band at
123

Optimization for catalytic performances of Hb zeolite in…
Table 4 Results of the regeneration capacity of the parent and modified Hb zeolites
Catalyst
Regeneration Calcination
time
Deactivated Catalyst
point (%)
Catalytic
lifetime
(h)
Maximum
yield of
(mol%)
temperature
(°C)
dosage
(g)
Parent Hb
0
1
2
3
4
5
6
7
8
550
550
550
550
550
550
550
550
550
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
5.92
5.92
5.92
5.92
5.92
5.92
5.92
5.92
5.92
13.5
13.5
13.5
13.5
13.5
13.5
13.4
13.4
13.4
13.5
45.0
45.0
45.0
45.0
45.0
45.0
44.5
44.5
44.5
44.8
12.5
12.5
12.5
12.5
12.4
12.4
12.3
12.3
12.3
12.4
23.5
23.5
23.5
23.5
23.5
23.5
23.2
23.2
23.2
23.4
84.5
84.1
83.8
83.8
83.8
83.4
83.4
83.4
83.4
83.7
89.5
89.5
89.5
89.4
89.3
89.3
89.3
89.2
89.2
89.3
81.4
81.4
81.4
81. 1
81.1
80.9
80.9
80.7
80.5
81.0
87.6
87.6
87.6
87.4
87.3
87.3
87.3
87.2
87.1
87.3
Average/total
TA-Hb
0
1
2
3
4
5
6
7
8
550
550
550
550
550
550
550
550
550
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
5.93
5.93
5.93
5.93
5.93
5.93
5.93
5.93
5.93
Average/total
AC-Hb
0
1
2
3
4
5
6
7
8
550
550
550
550
550
550
550
550
550
550
550
550
550
550
550
550
550
550
550
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
5.95
5.95
5.95
5.95
5.95
5.95
5.95
5.95
5.95
Average/total
Si-Hb
0
1
2
3
4
5
6
7
8
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
70.0
5.91
5.91
5.91
5.91
5.91
5.91
5.91
5.91
5.91
Average/total
123

Y. Xiong et al.
Fig. 8 The 13C MAS NMR patterns of the parent and deactivated Hb zeolites
125–170 ppm, while the band at 170–200 ppm belongs to poly aromatics. In
contrast to parent Hb, the apparent deactivation phenomenon was evidenced by the
strong characteristic signal of carbon deposits in this acylation. Nevertheless, it is
not easy to determine the specific coke content by 13C MAS NMR spectra.
The carbon deposits absorbed on the external surface and in the micropores of
deactivated zeolites were analyzed by GC–MS. It was found acetic acid, 2-AC-5-
MF, and some minor compounds, basically oligomers (Fig. 9), derived from the
consecutive acylation of monoacylated products. In our work reported here, we can
summarize that the prominent deactivation on Hb zeolite exists in the acylation of
2-MF, which matches well with a recent report [39]. The acylated products were
easily adsorbed on the acid sites related to its strong polarity, so the adsorption of
the 2-MF on the channel of zeolites was limited. Therefore, the acid sites on Hb
were covered, which contributes to the reduction of reaction rate. Moreover, the
adsorbed 2-AC-5-MF might react further again, giving by-products of larger
molecular dimensions, i.e., carbon deposits, caused the diffusion problems of
reactants within the channels of the zeolites.
Fig. 9 Second acylation oligomers (E)-1,2-bis(5-methylfuran-2-yl)ethane
123

Optimization for catalytic performances of Hb zeolite in…
Solvent effects
The solvents are generally employed for accelerating reactions rate, and the stability
and selectivity in the organic synthesis of functional chemicals [40]. As it is known,
Dimroth et al. [41]. considered solvents as structured isotropic continuum, mainly
composed of the individual solvent molecule with itself having interactions called
solvent/solvent, and the special solute/solvent interactions like hydrogen-bond,
EPD/EPA and so on. The solvent polarity is represented by experience parameters,
such as ET (30) and normalized values ETN. To delay the deactivation of the catalyst
further, nitromethane and 1,2- dichloroethane were used as solvents in this acylation
reaction. The solvent effects experiments were carried out at 60 °C with the same
flow rate (0.07 mL min-1) in a fixed bed, the equal amount of solvent (20 wt% of
total reactants) was mixed into the mixture of 2-MF and AA (molar ratio 1:2.5), then
was introduced into the catalytic column by a metering pump. After 45 h of
reaction, what is interesting is that 1,2-dichloroethane (ENT = 0.327) was found to
be effective with good activity and stability in the acylation of 2-MF. However, the
other important result is that the yield of 2-AC-5-MF decreases when 1,2-
dichloroethane is substituted by nitromethane (ENT = 0.481). Indeed, the initial yield
of 2-AC-5-MF, which is around 93 mol% in 1,2-dichloroethane, becomes less than
75 mol% in nitromethane shown in Fig. 10 compared to the solvent free.
In this work, SEM was employed to characterize the morphologies of the parent
and deactivated Hb (after reaction of 45 h) with different solvents. As depicted in
Fig. 11, the particle size of the parent and deactivated Hb mainly distribute in the
range of 0.2–0.5 and 0.5–1 lm, respectively. In general, fresh parent Hb mainly
consists of loose aggregates of tiny crystals 30–100 nm in size and uniformly
distributed. The aggregation level of zeolite particles reflects the amounts of carbon
Fig. 10 2-Acetyl-5-methylfuran yield (mol%) versus time on stream in the presence and absence of
solvent in the continuous acylation of 2-methylfuran with acetic anhydride (experimental conditions:
60 °C, 2-methylfuran/acetic anhydride molar ratio = 1:2.5, flow rate: 0.07 mL/min, 5.9 g of parent Hb
zeolite)
123

Y. Xiong et al.
Fig. 11 SEM images of the parent and deactivated Hb zeolites after acylation with different solvent
conditions
deposits. With nitromethane as solvent in acylation of 2-MF, there were serious
carbon deposits remaining on the zeolites, resulting in reaction rate decreases.
However, the lower carbon deposits on catalysts were observed using 1,2-
dichloroethane, indicating that 1,2-dichloroethane of medium polarity could exert a
significant effect to minimize catalyst decay. The solvent with moderate polarity can
favor the generation of acyl cations and desorption of 2-acetyl-5-methylfuran, thus
promoting the acylation process, similar to report from Kim et al. [42]. However,
the stronger polar solvents such as nitromethane (ENT = 0.481) will weaken the
desorption process, and impede the formation of the acyl cation. It is critical to note
that an increasing solvent polarity may lead to the competitive adsorption with
reactants on acid sites of zeolites.
Conclusions
The catalytic performance of all modified Hb zeolites were first tested in continuous
liquid-phase acylation of 2-methylfuran. Among these zeolites, TA-Hb and Si-Hb
showed the more stable catalytic activity, while AC-Hb and parent Hb were easily
deactivated. What’s more, the high yield of 89.5 mol% was achieved on TA-Hb
123

Optimization for catalytic performances of Hb zeolite in…
with 100 % selectivity in our research. In a solvent effect study, 1,2-dichloroethane
as a solvent of intermediate polarity can enhance the catalytic stability in acylation
reaction. The analysis results from GC–MS and 13C MAS NMR clearly elucidate
that Hb zeolite is sensitive to the deactivation, owing to the coke deposits
powerfully adsorbed on its surface. Whereas, deactivated Hb zeolites can be
regenerated to its original status through the high temperature calcination.
Based on the characterization (IR, Py-FTIR, and NH3-TPD) of the acid properties
and catalytic results for the acylation of 2-methylfuran, the proposed modification
mechanism involved in the properties and distribution of the acids intensity on Hb
was discussed. It was clearly found that there was the possibility of adjusting the
distribution of the acid intensity using surface modifications. The acid treatment and
SiO2-CLD modification can be successfully applied for coordinating the distribution
of acid intensity. The best catalytic performances can be achieved as long as the
strong acids were not as intense as possible but with an appropriate acid intensity
distribution, especially, the strong and weak acid amounts were equational (ratio of
1:1). Heterogeneous catalysts in acylation reactions require such a catalytic
flexibility and stability to enlarge the opportunities for the synthesis of other
aromatic ketones. Our work highly supports the potential of surface modification on
Hb zeolite, which could tune acidity of zeolites to meet the requirements.
Acknowledgments This study was supported by the State Key Laboratory of Chemical Engineering
(Tianjin). We gratefully thank the Sinopec Research Institute of Petroleum Processing of China for the
Py-IR measurements.
References
1. M. Bejblova, D. Prochazkova, J. Cejka, ChemSusChem 2, 48 (2009)
2. J.A. Melero, R.V. Grieken, G. Morales, V. Nuno, Catal. Commun. 5, 131 (2004)
3. I. Komoto, J.I. Matsuo, S. Kobayashi, Top. Catal. 19, 43 (2002)
4. P. Moreau, A. Finiels, P. Meric, F. Fajula, Catal. Lett. 85, 199 (2003)
5. H.K. Heinichen, W.F. Holderich, J. Catal. 185, 408 (1999)
6. U. Freese, F. Heinrich, F. Roessner, Catal. Today 49, 237 (1999)
7. G. Sartori, R. Maggi, Chem. Rev. 111, 181 (2011)
8. K.G. Bhattacharyya, A.K. Talukdar, P. Das, S. Sivasanker, Catal. Commun. 2, 105 (2001)
ˇ
ˇ ´ ˇ
9. M.V. Shamzhy, O.V. Shvets, M.V. Opanasenko, L. Kurfirtova, D. Kubicka, J. Cejka, ChemCatChem
7, 1891 (2013)
ˇ
ˇ
´
´
10. C. Jo, R. Ryoo, N. Zilkovac, D. Vitvarovac, J. Cejka, Catal. Sci. Technol. 3, 2119 (2013)
11. J. Kaur, K. Griffin, B. Harrison, I.V. Kozhevnikov, J. Catal. 208, 448 (2002)
12. G.D. Yadav, N.S. Asthana, V.S. Kamble, Appl. Catal. A Gen. 240, 53 (2003)
´ ´
13. F. Frouri, S. Celerier, P. Ayrault, F. Richard, Appl. Catal. B Environ. 168, 515 (2015)
14. M.L. Kantam, K.V.S. Ranganath, M. Sateesh, K.B.S. Kumar, B.M. Choudary, J. Mol. Catal. A Chem.
225, 15 (2005)
15. E.G. Derouane, C.J. Dillon, D. Bethell, S.B.D. Hamid, J. Catal. 187, 209 (1999)
16. E. Fromentin, J.-M. Coustard, M. Guisnet, J. Mol. Catal. A Chem. 159, 377 (2000)
17. P. Botella, A. Corma, G. Sastre, J. Catal. 197, 81 (2001)
18. C. Guignard, V. Pedron, F. Richard, R. Jacquot, M. Spagnol, J.M. Coutard, G. Perot, Appl. Catal.
A Gen. 234, 79 (2002)
´
19. V.F.D. Alvaro, A.F. Brigas, E.G. Derouane, J.P. Lourenco, B.S. Santos, J. Mol. Catal. A Chem. 305,
100 (2009)
20. U. Freese, F. Heinrich, F. Roessner, Catal. Today 49, 237 (1999)
21. D. Rohan, C. Canaff, E. Fromentin, M. Guisnet, J. Catal. 177, 296 (1998)
123

Y. Xiong et al.
22. P.R. Reddy, M. Subrahmanyam, S.J. Kulkarni, Catal. Lett. 54, 95 (1998)
23. D.S. Zhao, J.L. Wang, J. Zhang, Catal. Lett. 126, 188 (2008)
24. W.H. Chen, T. Chang Tsai, S.J. Jong, Q. Zhao, J. Mol. Catal. A Chem. 181, 41 (2002)
25. G.Y. Bai, H.H. Zhang, T.Y. Li, H.X. Dong, J. Han, Res. Chem. Intermed. 41, 5041 (2015)
¨
26. S. Zheng, H.R. Heydenrych, H.P. Rogerb, A. Jentysa, J.A. Lerchera, Top. Catal. 22, 101 (2003)
27. R. Srivastava, N. Iwasa, S.I. Fujita, M. Arai, Catal. Lett. 130, 655 (2009)
28. D.M. Roberge, H. Hausmann, W.F. Holderich, Phys. Chem. Chem. Phys. 4, 3128 (2002)
29. M. Matsukata, M. Ogura, T. Osaki, P.R.H.P. Rao, M. Nomura, E. Kikuchi, Top. Catal. 9, 77 (1999)
30. X.S. Zhao, G.Q. Lu, J. Phys. Chem. B 102, 1556 (1998)
31. C.A. Emeis, J. Catal. 141, 347 (1993)
32. G.Y. Bai, Z. Ma, L.J. Shi, T.Y. Li, J. Han, G.F. Chen, N. Li, P.D. Liu, Res. Chem. Intermed. 38, 2501
(2012)
33. K.K. Cheralathan, I.S. Kumar, M. Palanichamy, V. Murugesan, Appl. Catal. A Gen. 241, 247 (2003)
34. M. Ghiaci, A. Abbaspur, R. Kia, C. Belver, R. Trujillano, V. Rives, M.A. Vicente, Catal. Commun. 8,
49 (2007)
¨
35. R.W. Weber, K.P. MOller, C.T. O’Connor, Microporous Mesoporous Mater. 35–36, 533 (2000)
36. Z.R. Zhu, Z.K. Xie, Q.L. Chen, D.J. Kong, W. Li, W.M. Yang, C. Li, Microporous Mesoporous
Mater. 101, 169 (2007)
37. K.S. Triantafyllidis, L. Nalbandian, P.N. Trikalitis, A.K. Ladavos, T. Mavromoustakos, C.P. Nico-
laides, Microporous Mesoporous Mater. 75, 89 (2004)
38. P. Dejaifve, A. Auroux, P.C. Gravelle, J.C. Vedrine, Z. Gabelica, E.G. Derouane, J. Catal. 70, 123
(1981)
39. Z.H. Chen, W.Q. Chen, T.X. Tong, A.W. Zeng, J. Mol. Catal. A Chem. 396, 231 (2015)
40. M. Casagrande, L. Storaro, M. Lenarda, R. Ganzerla, Appl. Catal. A Gen. 201, 263 (2000)
41. V.K. Dimroth, C. Reichardt, T. Siepmann, F. Bohlmann, Liebigs Ann. Chem. 661, 1 (1963)
42. S.D. Kim, K.H. Lee, J.S. Lee, Y.J. Kim, K.E. Yoon, J. Mol. Catal. A Chem. 152, 33 (2000)
123