Effect of Ultrasonic Treatment on the Physicochemical and Catalytic Properties of Rhodium–Chitosan/HTsVM Catalysts in Dimethyl Ether Conversion to Lower Olefins

Article abstract of DOI:10.1134/S0965544119090032

Abstract: Effect of ultrasonic treatment (UT) on the catalytic and textural properties of composite catalyst systems based on zeolite HTsVM modified with rhodium and chitosan has been studied. The introduction of a rhodium–chitosan composite and UT have a significant effect on the catalytic properties of the zeolite catalyst in the conversion of dimethyl ether (DME) to lower olefins by providing a substantial increase in the catalyst activity; the effect is attributed to a more finely dispersed distribution of rhodium over the catalyst surface. Using transmission electron microscopy (TEM), it has been first shown that UT leads to a decrease in the rhodium particle size and a finely dispersed distribution of Rh over the surface of the zeolite catalyst.

Full text of DOI:10.1134/S0965544119090032

ISSN 0965-5441, Petroleum Chemistry, 2019, Vol. 59, No. 9, pp. 1017–1022. © Pleiades Publishing, Ltd., 2019.
Russian Text © The Author(s), 2019, published in Neftekhimiya, 2019, Vol. 59, No. 5, pp. 569–574.
Effect of Ultrasonic Treatment on the Physicochemical
and Catalytic Properties of Rhodium–Chitosan/HTsVM Catalysts
in Dimethyl Ether Conversion to Lower Olefins
T. I. Batovaa, *, T. K. Obukhovaa, N. V. Kolesnichenkoa, and S. A. Nikolaevb
aTopchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences, Moscow, 119991 Russia
bFaculty of Chemistry, Moscow State University, Moscow, 119991 Russia
*e-mail: batova.ti@ips.ac.ru
Received February 12, 2019; revised April 18, 2019; accepted May 13, 2019
Abstract—Effect of ultrasonic treatment (UT) on the catalytic and textural properties of composite catalyst
systems based on zeolite HTsVM modified with rhodium and chitosan has been studied. The introduction of
a rhodium–chitosan composite and UT have a significant effect on the catalytic properties of the zeolite cat-
alyst in the conversion of dimethyl ether (DME) to lower olefins by providing a substantial increase in the
catalyst activity; the effect is attributed to a more finely dispersed distribution of rhodium over the catalyst
surface. Using transmission electron microscopy (TEM), it has been first shown that UT leads to a decrease
in the rhodium particle size and a finely dispersed distribution of Rh over the surface of the zeolite catalyst.
Keywords: composite zeolite catalysts, chitosan, rhodium, dispersed distribution of active component, ultra-
sonic treatment, dimethyl ether, lower olefins
DOI: 10.1134/S0965544119090032
Synthesis of lower olefins from synthesis gas via led to an increase in the catalyst efficiency in DME
oxygenates (MTO and DTO processes) is the most conversion to lower olefins [11, 12].
promising of the currently available schemes for this
In addition, to provide a finely dispersed distribu-
process. The conversion of dimethyl ether (DME) to
tion of active components, polymers are used as a sub-
lower olefins (DTO process) attracts increasing atten-
strate and a stabilizer; the subsequent heat treatment
tion of researchers, because this process is character-
of the catalyst leads to a finely dispersed distribution of
ized by more favorable thermodynamic characteris-
active catalytic sites [13–15]. The most promising
tics. Effective catalysts for DME conversion to lower
materials for these purposes are biopolymers, such as
olefins are zeolite catalysts modified with various met-
cellulose and cellulose structural analogs—chitin and
als [1, 2]. The composition and structure of the mod-
chitosan—which, unlike cellulose, do not require pre-
ifying active component of the catalyst significantly
modification to be used as supports in catalysis. Chi-
depend on the sample preparation method, in partic-
tosan is a more reactive complexing agent owing to
ular, the catalyst pretreatment conditions and the
deacetylation. It is known that this polymer has a
nature of the active component precursor. Thus, the
unique adsorption capacity with respect to many ions
modification of zeolites with salts of active metals typ-
and even atoms of metals. Chitosan is effectively used
ically leads to the formation of heterogeneities in the
to provide stabilization and a finely dispersed distribu-
chemical and phase compositions and thereby affects
tion of active nanoparticles over the surface of various
the properties of catalysts based on these zeolites. The
supports [16]. Thus, the development of scientific
effect is particularly pronounced in the case of zeolite
approaches to synthesizing efficient catalysts with a
modification with noble metals [3]. The implementa-
high degree of dispersion of the active component for
tion of a finely dispersed distribution of the active
DME conversion to valuable petrochemicals is a prob-
component over the catalyst surface leads to an
lem of high priority.
increase in the catalytic activity. Recently, to provide
this distribution, ultrasonic treatment (UT) and ultra-
This paper describes results of studying the specific
high-frequency microwave-activation have been most features of the formation and distribution of active
commonly used [4–10]. It was previously shown that sites of Rh/HTsVM under conditions of different syn-
an increase in the degree of dispersion of the active thesis methods using chitosan as a medium for dis-
component (Mg) provided by UT during the impreg- persing rhodium; in particular, the effect of ultra-
nation of the support with a salt of the active element sound-assisted synthesis of catalysts for DME conver-
1017

1018
BATOVA et al.
sion to lower olefins on their catalytic properties is the catalyst (0.4–0.6 mm fraction). After that, the cat-
studied.
alyst was activated in an N2 stream at 400°C for 1 h. A
required reactant feed space velocity (2.7 h–1), tem-
perature (320°C), and pressure (~1 atm) were set. To
set the required feed space velocity, the gas flow rate
was controlled by RGG-10 gas flow regulators. For
analysis, the gas stream was supplied via a dosing valve
to a Kristallyuks 4000M chromatograph equipped with
a flame ionization detector. A 27.5 m × 0.32 mm ×
10 μm capillary column was used; the adsorbent was
the CP-PoraPLOT Q-HT nonpolar phase, which was
sufficiently efficient for the isolation of the main groups
of the reaction products (DME, СН3ОН, С1–С6
hydrocarbons). Analysis was conducted using tem-
perature programming (80–200°C at a heating rate of
10°C/min); nitrogen was used as a carrier gas (flow
rate of 30 mL/min). The recorded chromatograms
were processed using the NetChromWin software.
The process parameters were determined in terms of
material balance.
EXPERIMENTAL
The precursor was DME with a purity of 99.8%
(Azot, Novomoskovsk, Russia). The catalyst samples
tested in the synthesis of lower olefins from DME
were based on a high-silica zeolite (TsVM), which is
a domestic counterpart of zeolite ZSM-5, in the
ammonium form with a SiO2/Al2O3 molar ratio of
32.6 (Angarsk catalyst and organic synthesis plant,
Russia). The zeolite was converted to the hydrogen
form (HTsVM) by calcining the powder in air at
500°C for 4 h.
The Rh–chitosan composite was synthesized using
chitosan hydrochloride (Bioprogress, Moscow) with a
molecular weight of 104 Da.
The modification of the HTsVM zeolite with rho-
dium and chitosan was implemented by different
methods.
Sample 1: Rh/HTsVM was synthesized by adding
an aqueous solution of RhCl3 to HTsVM; the mixture
was left to stand for 1 day; after that, the sample was
dried and calcined in air at 500°C for 4 h.
Sample 2: Rh(Ch)/HTsVM was synthesized by
adding a presynthesized composite of chitosan dis-
solved in water and an aqueous solution of RhCl3; the
mixture was left to stand for 1 day; after that, the sam-
ple was dried and calcined in air at 500°C for 4 h.
Sample 3: Rh(Ch-UT)/HTsVM was synthesized in
accordance with a procedure similar to that used for
sample 2, except that the composite of chitosan dis-
solved in water and an aqueous solution of RhCl3 was
subjected to 1-h UT prior to deposition on the zeolite.
Sample 4: Rh(UT)/HTsVM was synthesized in
accordance with a procedure similar to that used for
sample 1, except that the aqueous solution of RhCl3
and HTsVM was subjected to 1-h UT prior to deposi-
tion on the zeolite.
Dimethyl ether conversion (X) was calculated by
formula (1):
m0 − m
m0
(1)
Х =
× 100, %,
where m0 are m is the DME weight at the inlet and out-
let of the reactor, respectively (g).
Olefin selectivity (S) was calculated by formula (2):
molef
mHC
(2)
S =
× 100, wt %,
where molef and mHC is the weight of olefins and all the
synthesized hydrocarbons, respectively (g).
The textural characteristics (specific surface area,
total pore volume, and pore size distribution) of the
samples were studied by low-temperature adsorption–
desorption of molecular nitrogen on a Micromeritics
ASAP-2010 unit. All samples were preevacuated at a
temperature of 350°C to 4 × 10–1 Pa. Nitrogen adsorp-
tion was run at a temperature of 77 K.
The morphology of the nanocomposites and fin-
ished zeolite catalysts was studied by transmission
electron microscopy (TEM) on a LEO912 OMEGA
microscope (Carl Zeiss, Germany) at an accelerating
voltage of 100 kV, a resolution of 0.2 nm, and an illu-
mination area of 1–75 μm. For TEM studies, a solu-
tion of the nanocomposite or a powder of the zeolite
catalyst was placed on a standard TEM copper grid
coated with an amorphous carbon film. The grid was
air dried and then inserted into the microscope to con-
duct the studies.
The calculated amount of Rh and water-soluble chi-
tosan in the composition of the finished Rh/HTsVM
catalyst was 0.1 and 2–7 wt %, respectively. The cata-
lytic tests were conducted using a catalyst fraction of
0.4–0.63 mm; it was prepared by mechanically grinding
the pellets compressed from the prepared Rh/HTsVM
powder.
The UT of the Rh–chitosan composite and the
aqueous solution of RhCl3 and HTsVM was conducted
in an Elmasonic P30H ultrasonic bath at a frequency
of 80 kHz and a power of 130 W (UT). All the samples
contained ~0.1 wt % Rh.
RESULTS AND DISCUSSION
In this study, chitosan in a water-soluble form was
microreactor. Nitrogen was used as the DME diluent. used to synthesize an Rh–chitosan composite, which
The DME concentration in the feed gas mixture was was subsequently deposited on zeolite HTsVM. Ini-
10 vol %. The flow reactor was charged with 0.5 g of tially, the effect of chitosan as a support for rhodium
Catalytic tests on DME conversion to lower olefins
were conducted on a laboratory plant using a flow
PETROLEUM CHEMISTRY
Vol. 59
No. 9
2019

EFFECT OF ULTRASONIC TREATMENT ON THE PHYSICOCHEMICAL
1019
135
125
4
3
2
1
115
105
95
85
75
0.2
0.4
0.6
0.8
1.0
0
Relative pressure (p/p0)
Fig. 1. Low-temperature nitrogen adsorption–desorption isotherms for (1) the Rh/HTsVM, (2) Rh(Ch)/HTsVM,
(3) Rh(Ch-UT)/HTsVM, and (4) Rh(UT)/HTsVM catalysts.
on the textural and catalytic properties of the catalyst of the zeolite catalyst mostly owing to a decrease in the
in DME conversion to olefins was studied.
number of strong Brønsted acid sites.
Figure 1 shows low-temperature nitrogen adsorp-
tion isotherms; Table 1 lists parameters of the pore
structure of the Rh/HTsVM, Rh(Ch)/HTsVM,
Rh(Ch-UT)/HTsVM, and Rh(UT)/HTsVM samples.
It is evident from Fig. 1 that all the catalysts are char-
acterized by a classical isotherm of zeolites with a
micron and submicron crystal size and a small amount
of mesopores. A pronounced horizontal portion of the
adsorption hysteresis, which indicates the presence of
mesopores, is observed in a р/р0 range of 0.5–0.9.
Analysis of the pore structure parameters of the
samples (Table 1) shows that the use of a polymer
matrix in the modification of the zeolite catalyst with
a rhodium salt and subsequent calcining at 500°C has
hardly any effect on the external surface area and vol-
ume of mesopores. Previously [17], in diffuse reflec-
tance infrared Fourier transform spectroscopy studies,
it was shown that the calcining of a zeolite with chi-
tosan leads to the decomposition of chitosan to carbon
dioxide and water; in addition, small amounts of car-
bonate anions were detected on the surface of the
water-soluble zeolite. The sonication of the Rh–chi-
tosan composite prior to deposition on HTsVM does
not affect the volume of meso- and micropores. The
use of chitosan leads to a decrease in the total acidity
Typical micrographs of the ultrasound-treated
Rh–chitosan composite and the Rh–chitosan com-
posite are shown in Fig. 2.
The photographs clearly show dark agglomerates of
chitosan molecules, which contrast with the uniform
gray background of the carbon film of the TEM
holder. The texture of the Rh–chitosan sample is rep-
resented by crosslinked macroagglomerates of chi-
tosan molecules (Fig. 2a). The macroagglomerate size
is 3500 500 nm. The UT leads to the scission of the
bonds between the macroagglomerates and their dis-
persion to 1200 250 nm (Fig. 2b). At a maximum
Table 1. Characteristics of the pore structure of the zeolite
catalysts modified with rhodium and chitosan with and
without sonication of the Rh–chitosan composite
SВЕТ
, Vtot, Vmicro, Vmeso,
No.
Catalyst
m2/g cm3/g cm3/g cm3/g
1
2
3
4
Rh/HTsVM
Rh(Ch)/HTsVM
Rh(Ch-UT)/HTsVM 357 0.184 0.116 0.068
Rh(UT)/HTsVM 362 0.183 0.119 0.064
353 0.182 0.115 0.067
353 0.180 0.114 0.066
PETROLEUM CHEMISTRY Vol. 59 No. 9 2019

1020
BATOVA et al.
20 μm
20 μm
(а)
(b)
1 μm
2 μm
(c)
(d)
20 μm
Fig. 2. Low-resolution micrographs of (a, c) the Rh–chitosan composite and (b, d) the Rh–chitosan composite after ultrasonic
treatment. Microagglomerates are marked with a square.
magnification, it can be evident that both samples attributed to the small size of the crystallites. Unfortu-
contain not only macro- but also microagglomerates nately, a rhodium particle size distribution histogram
of chitosan (Figs. 2c, 2d). The microagglomerate sizes cannot be constructed, because the rhodium content
before and after UT are approximately identical at a in the catalyst is low (0.1 wt %).
level of 200 50 nm.
The impregnation of the zeolite with the Rh–chi-
The electron diffraction pattern of the Rh–chi- tosan composite leads to the formation of agglomer-
tosan composite after UT (inset in Fig. 2d) exhibits a ates in the catalyst than are significantly smaller than
reflection corresponding to an interplanar distance of the agglomerates in the solution. Most probably, the
1.902 Å in the Rh(200) face of rhodium nanocrystal- decrease in the size of similar-type agglomerates is
lites Fm3m [PDF no. 5-0685]. The absence of well- associated with the chemical reaction of the chitosan
defined reflections in the diffraction of electrons from molecules with the acid sites of the zeolite; during
the selected surface of the finished catalyst samples calcining, this reaction leads to the degradation of
with Rh crystallites (Figs. 3a, 3b) is apparently the C–C bonds of the chitosan molecules and the
PETROLEUM CHEMISTRY
Vol. 59
No. 9
2019

EFFECT OF ULTRASONIC TREATMENT ON THE PHYSICOCHEMICAL
1021
20 μm
20 μm
200 nm
100 nm
(а)
(b)
Fig. 3. Low-resolution micrographs of (a) the Rh(Ch)/HTsVM and (b) Rh(Ch-UT)/HTsVM catalyst particles synthesized by
impregnating the support with an Rh–chitosan composite.
defragmentation of the original aggregates. The ten- decrease in the rhodium particle size and a finely dis-
persed distribution of Rh over the catalyst surface.
dency revealed for the Rh(Ch)HTsVM sample is repro-
duced for the Rh(Ch-UT)/HTsVM catalyst (Fig. 3).
The results of studying the effect of the use of chi-
tosan as a polymer matrix and the exposure to ultra-
sound on the catalytic properties of the Rh/HTsVM
catalyst in DME conversion to lower olefins are shown
in Table 2. The use of chitosan in the zeolite modifica-
tion with rhodium, in common with the use of ultra-
sound during deposition, leads to an increase in the
DME conversion, which is most probably attributed
to a more finely dispersed distribution of rhodium
over the catalyst surface. In addition, the eth-
ylene/propylene ratio slightly increases. The total
selectivity for lower olefins remains approximately at
the same level (~76 wt %). The UT does not affect
the distribution of the reaction products and the
selectivity for lower olefin.
The rhodium nanoparticles visible in the micro-
graphs have a spherical shape and a size of (4.5–9.5)
0.4 nm for the Rh(Ch)/HTsVM catalyst (Fig. 3a) and
(1.8–3.3) 0.4 nm for the Rh(Ch-UT)/HTsVM sam-
ple (Fig. 3b). The electron diffraction patterns in
Figs. 3a and 3b correspond to the diffraction pattern of
the Rh–chitosan composite after UT (inset in
Fig. 2d); this finding confirms that the finished zeo-
lite-containing catalyst exhibits the same diffractions
of fine crystallites of metallic Rh as in the Rh–chi-
tosan composite. The widths of the diffraction rings
of the selected sample region differ (insets in
Figs. 3a, 3b); for the Rh(Ch-UT)/HTsVM sample
(inset in Fig. 3b), the diffraction ring is more diffuse;
this fact can indicate a more finely dispersed distribu-
Figure 4 shows the effect of the rhodium deposition
tion of metallic Rh after UT. Thus, UT leads to a method on the stability of the studied catalysts. Com-
Table 2. Effect of the use of chitosan as a polymer matrix and ultrasonic treatment on the catalytic properties of
Rh/HTsVM in DME conversion to lower olefins
Hydrocarbon selectivity, wt %
С2 –С4=
С2= С3=
ХDME,
%
No.
Catalyst
С1–С5
alkanes

ethylene
propylene
butenes
1
2
3
4
Rh/HTsVM
50.1
60.4
60.1
59.4
21.7
23.2
21.1
22.3
37.7
35.6
36.6
37.6
17.0
16.4
17.0
16.9
23.7
24.7
24.9
23.2
76.4
75.2
75.1
76.8
0.6
0.7
0.6
0.6
Rh(Ch)/HTsVM
Rh(Ch-UT)/HTsVM
Rh(UT)/HTsVM
–1
Conditions: Т = 320°С, Р = 1 atm, W = 2.7 h , within 3 h.
PETROLEUM CHEMISTRY Vol. 59 No. 9 2019

1022
90
BATOVA et al.
FUNDING
This work was performed under a state task to Topchiev
Institute of Petrochemical Synthesis of the Russian Acad-
emy of Sciences.
80
70
60
50
40
30
20
2
3
4
1
REFERENCES
1. M. Stoker, Microporous Mesoporous Mater. 29, 3 (1999).
2. I. A. Bakarea, O. Murazaa, M. A. Sanhooba, et al.,
Fuel 211, 18 (2018).
3. G. F. Prozorova, A. S. Pozdnyakov, N. P. Kuznetsova,
et al., Int. J. Nanomed. 9, 1883 (2014).
1.5 2.0 2.5 3.0 3.5 4.0 4.5
Time, h
0.5 1.0
4. L. M. Kustov, in Proceedings of II Russian Congress on
Catalysis “ROSKATALIZ-2014” (Samara, 2014),
Vol. 1, p. 24 [in Russian].
5. A. E. Baranchikov, V. K. Ivanov, and Yu. D. Tret’ya-
Fig. 4. Effect of the use of chitosan as a polymer matrix for
rhodium and ultrasonic treatment on the catalyst stability
in DME conversion to lower olefins: (1) Rh/HTsVM,
(2) Rh(Ch)/HTsVM, (3) Rh(Ch-UT)/HTsVM, and
(4) Rh(UT)/HTsVM.
kov, Usp. Khim. 76, 147 (2007).
6. P. Nimesh, P. K. Vabbina, and P. Nezih, Chem. Mater.
8, 2172 (1996).
7. W. Zhilin, E. Borretto, and J. Medlock, Ultrason.
Sonochem. 29, 104 (2016).
8. Z. Wu, E. Borretto, J. Medlock, W. Bonrath, and
pared with Rh/HTsVM (1), the DME conversion in
the presence of Rh(UT)/HTsVM (4) increases by 20%
within the first hour on stream; however, it decreases
within 4 h; the deactivation rate is high (~8% per
hour). The use of chitosan in the modification of cat-
alyst (2), in common with the UT of the Rh–chitosan
composite (3), provides an increase in the DME con-
version by about 15% compared with the Rh/HTsVM
catalyst (1); in this case, the catalyst stability remains
unchanged; the deactivation rates of the Rh/HTsVM,
Rh(Ch)/HTsVM, and Rh(Ch-UT)/HTsVM samples
are approximately identical (about 2–3% per hour).
G. Cravotto, ChemCatChem 6, 2762 (2014).
9. H. Xu, B.W. Zeiger, and K.S.Suslick, Chem. Soc. Rev.
42, 2555 (2013).
10. E. E. Kolesnikova, T. K. Obukhova, N. V. Kole-
snichenko, et al., Pet. Chem. 58, 573 (2018).
11. A. Tadjarodi, M. Dehghani, and M. Imani, Appl.
Organomet. Chem. 32, 4594 (2018).
12. V. M. Novotortsev, V. V. Kozlov, Yu. M. Korolev, et al.,
Russ. J. Inorg. Chem. 53, 1006 (2008).
13. A. D. Pomogailo, Polymer-Immobilized Metal Complex
Catalysts (Nauka, Moscow, 1988) [in Russian].
Thus, the use of chitosan as a polymer matrix and
the exposure to ultrasound have a significant effect on
the catalytic properties of rhodium-containing zeolite
catalysts for DME conversion to lower olefins by pro-
viding a substantial increase in the catalyst activity. It
has been shown that the UT leads to a decrease in the
rhodium particle size and a finely dispersed distribu-
tion of Rh over the catalyst surface. Most probably, a
finely dispersed distribution of the active component
provides an increase in the component concentration
on the catalyst surface, which leads to an increase in
the activity.
14. G. F. Prozorova, A. S. Pozdnyakov, A. I. Emel’yanov,
et al., Dokl. Chem. 449, 87 (2013).
15. A. D. Pomogailo, A. S. Rozenberg, and I. E. Uflyand,
Metal Nanoparticles in Polymers (Khimiya, Moscow,
2000) [in Russian].
16. A. N. Ozerin, A. N. Zelenetskii, T. A. Akopova, et al.,
Polym. Sci., Ser. A 48, 638 (2006).
17. N. V. Kolesnichenko, E. E. Kolesnikova, T. K. Obu-
khova, and G. N. Bondarenko, Pet. Chem. 58, 1015
(2018).
Translated by M. Timoshinina
PETROLEUM CHEMISTRY
Vol. 59
No. 9
2019