Dimethyl ether to olefins conversion in a slurry reactor: Effects of the size of particles and the textural and acidic properties of the MFI-type zeolite

Article abstract of DOI:10.1134/S0965544117070052

Effects of the particle size of a suspended zeolite catalyst based on the commercial MFI-type (high modulus zeolite, HMZ) zeolite on its physicochemical characteristics and catalytic properties in the conversion of dimethyl ether (DME) in a three-phase system with a slurry reactor are studied. In order to gain insight into the relationships between the physicochemical characteristics of MFI-type zeolites and their catalytic properties, the textural, acidic, and catalytic properties of HMZ zeolites and synthesized nanocrystallites of the zeolite with the MFI structure are compared. It was found that, in the conversion of DME in the slurry reactor, the catalytic properties of the MFI zeolite are strongly affected by the size of particles of the suspended catalyst: on passage to nanosized suspensions, the activity of the catalyst grows by several times. Reduction in the size of catalyst particles leads to an increase in the yield of hydrocarbons C₅⁺ and a decrease in the yield of lower olefins. At the same dispersity of suspensions, the catalytic properties of suspended catalysts are considerably affected by the distribution of acid sites over strength in the studied zeolite sample and the fraction of amphoteric sites in them.

Full text of DOI:10.1134/S0965544117070052

ISSN 0965-5441, Petroleum Chemistry, 2017, Vol. 57, No. 7, pp. 576–583. © Pleiades Publishing, Ltd., 2017.
Original Russian Text © N.V. Kolesnichenko, S.V. Konnov, V.S. Pavlov, O.V. Yashina, N.N. Ezhova, S.N. Khadzhiev, 2017, published in Nanogeterogennyi Kataliz, 2017, Vol. 2,
No. 1, pp. 29–37.
Dimethyl Ether to Olefins Conversion in a Slurry Reactor:
Effects of the Size of Particles and the Textural and Acidic Properties
of the MFI-Type Zeolite
N. V. Kolesnichenko, S. V. Konnov*, V. S. Pavlov, O. V. Yashina, N. N. Ezhova, and S. N. Khadzhiev
Topchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences, Moscow, Russia
*
e-mail: konnov_sv@ips.ac.ru
Received January 26, 2017
Abstract⎯ Effects of the particle size of a suspended zeolite catalyst based on the commercial MFI-type (high
modulus zeolite, HMZ) zeolite on its physicochemical characteristics and catalytic properties in the conversion
of dimethyl ether (DME) in a three-phase system with a slurry reactor are studied. In order to gain insight
into the relationships between the physicochemical characteristics of MFI-type zeolites and their catalytic
properties, the textural, acidic, and catalytic properties of HMZ zeolites and synthesized nanocrystallites of
the zeolite with the MFI structure are compared. It was found that, in the conversion of DME in the slurry
reactor, the catalytic properties of the MFI zeolite are strongly affected by the size of particles of the sus-
pended catalyst: on passage to nanosized suspensions, the activity of the catalyst grows by several times.
+
5
Reduction in the size of catalyst particles leads to an increase in the yield of hydrocarbons C and a decrease
in the yield of lower olefins. At the same dispersity of suspensions, the catalytic properties of suspended cat-
alysts are considerably affected by the distribution of acid sites over strength in the studied zeolite sample and
the fraction of amphoteric sites in them.
Keywords: catalysis in the dispersed phase, zeolites with the MFI-type structure, physicochemical character-
istics of MFI-type zeolites, nanosized zeolites, dimethyl ether conversion, lower olefins
DOI: 10.1134/S0965544117070052
In recent years, the technologies of deep natural namic characteristics are more favorable than those in
gas processing into lower olefins through methanol the scheme accomplished through methanol.
and/or dimethyl ether (DME) derived from syngas
have been intensively studied and implemented all
over the world [1–11]. These schemes are used for the
synthesis of not only lower olefins but also higher boil-
ing hydrocarbons [3, 5, 12–16].
At the same time, all the studied technologies per-
taining to the synthesis of lower olefins from methanol
and/or dimethyl ether are based on the heteroge-
neous-catalytic gas-phase regime of the reaction.
Because the reaction in question is characterized
by a marked heat effect, problems related to the tech-
nological design arise and a catalyst is usually deacti-
vated rapidly. Therefore, the search is on for the tran-
sition from the conventional gas-phase catalysis to
catalysis in the dispersed phase with the use of a slurry
reactor, in which the process is conducted in a flow
regime in a three-phase system (gas–liquid–solid cat-
The process of syngas conversion to olefins through
methanol makes it possible to produce ethylene and/or
propylene with yields at a level of 75–90 mol %. This
process is being developed by a number of companies, is
being brought to commercialization, and is being
actively implemented in industry now [17].
Along with the methanol-to-olefins (MTO) tech- alyst). The continuous mode of the process (typical of
nologies that have been implemented in industry, a gas-phase catalysis) is combined with the effective
new technology of obtaining of lower olefins from heat removal from the catalyst grain by the liquid
dimethyl ether has attracted considerable attention; medium (typical of catalytic liquid-phase processes).
this technology offers a way to manufacture lower ole- A catalyst is maintained in the state of a fine suspen-
fins with high yield and selectivity (above 88 mol %), sion, and in terms of its parameters and outcome, the
with the fraction of ethylene and propylene in the process approaches homogeneous systems. As the size
hydrocarbon product being above 80 mol % [8–11]. of catalyst particles decreases, the area of the active
This process has not yet been brought to commercial catalyst surface increases, the intradiffusion retarda-
use, but it has great potential, because its thermody- tion declines, and the mass and heat transfer are
576

DIMETHYL ETHER TO OLEFINS CONVERSION IN A SLURRY REACTOR
577
improved; as a result, the rate of reaction grows and in a dry air steam of for 6 h at 550°C. After that, a
the deactivation processes decelerate. Ultrafine sus- threefold ion exchange using an ammonium nitrate
pensions with nanosized catalyst particles exhibit the solution with a concentration of 0.1 mol/L (10 g of
most promise.
solution per gram of zeolite) was conducted at a tem-
perature of 80°C.
This breakthrough technology has already been
successively applied in a number of chemical, petro-
chemical, biochemical, and ecological catalytic pro-
cesses [18–24]. For example, synthetic diesel fuel and
higher alcohols are efficiently produced with the use of
slurry reactors in the Fischer–Tropsch process tech-
nologies [25–28].
The particle size for the aggregates of crystallites of
the synthesized MFI zeolite and the ground HMZ zeo-
lite was measured at a temperature of 25°C by dynamic
light scattering using a Malvern Zetasizer NANO SZ
particle analyzer. A weighed portion of the powder was
mixed with a hydrocarbon oil (DOWTHERM RP) at a
Recently, lower olefins have been synthesized from ratio of 1 : 100 and ultrasonically dispersed for 20 min,
dimethyl ether in a slurry reactor using fine suspen- and the size of zeolite particles was measured.
sions of zeolite-containing catalysts [29], and a tech-
The morphology of zeolites was studied by scan-
nique for preparing ultrafine suspensions of nanosized
ning electron microscopy (SEM). After ultrasonic
catalysts from commercial MFI-type zeolite has been
treatment in water, the sample was thoroughly mixed
developed [30].
and 3 mL of the mixture was sampled. The sample was
diluted with water by a factor of 200, deposited on a
This paper presents the results of our research into
the effect of particle size of the suspended zeolite cat-
carbon strip fastened on a stage, and dried at room
alyst formed from commercial MFI-type zeolite on its
temperature. SEM images were taken on a Hitachi
physicochemical characteristics and catalytic behavior
TM3030 microscope.
in the conversion of DME in a three-phase system.
The textural, acidic, and catalytic characteristics of a
X-ray powder diffraction was applied to study the
nanosized sample of this catalyst are compared with structural characteristics of the framework of original
the analogous characteristics of specially synthesized zeolite samples. XRD data were collected on a
MFI-type zeolite containing nanoparticles formed by Bruker D2 PHASER X-ray diffractometer in the
nanocrystallites with the analogous crystal lattice 2θ geometry. The generator regime was as follows:
structure.
30 kW, 10 mA, and an X-ray tube with a copper
anode (λ СuKα1 = 1.5418 Å). Diffraction patterns
were recorded using rotation of the sample in the
horizontal plane in a 2θ angular range of 5°–50° in
increments of 0.05°, with slit widths at the tube exit
of 0.6 mm and before the detector of 1.15 mm, and an
acquisition time at a point of 3 s.
EXPERIMENTAL
In this study, medium-pore zeolites with the MFI
structure in the Н form were used, namely, commer-
cial high-modulus zeolite (HMZ) with a molar ratio of
SiO /Al O = 37 (ОАО Angarks Plant KiOS) and lab-
oratory-synthesized MFI zeolite with a molar ratio of
SiO /Al O = 57.
Nanoparticles were obtained from the commercial
ultrasonically ground high modulus zeolite (HMZ)
sample. Prior to ultrasonic treatment, the sample was
calcined at a temperature of 500°C for 2 h to obtain its
Н form (НHMZ). The НHMZ powder was thor-
oughly ground in an agate mortar, placed in a Teflon
2
2
3
The pore structure of the sample was studied by
low-temperature molecular nitrogen adsorption–
desorption on an ASAP 2010 porosimeter (Micromer-
itics). Before measurements, all of the samples were
evacuated at a temperature of 350°C to a pressure of
2
2
3
–
1
4
× 10 Pa. The adsorption of N was conducted at a
2
temperature of 77 K.
The acidic properties of the resulting samples
glass, and drenched with distilled water at a ratio of 20 were studied by temperature-programmed desorp-
g of water per gram of catalyst. After mixing in water tion of ammonia (TPD NH ). Experiments were
3
for 5 min, the zeolite aqueous suspension was ultra- conducted on an USGA-101 universal gas sorption
sonically treated in an Elmasonic P30 H ultrasonic analyzer (Unisit). The sample (~0.1 g in the form of
bath (ultrasound frequency of 80 kHz and power of fraction 0.25–0.5 mm) was placed in a quartz reac-
1
30 W). The ultrasonic treatment of the НHMZ tor, heated in a helium stream to a temperature of
(
НHMZ(0)) zeolite in water was performed for 60 and 250°C at a temperature rise rate of 10°C/min, cal-
0 min (НHMZ(60) and НHMZ(90)); the as-treated cined at this temperature for 1 h in a helium stream,
zeolite was dried and calcined at a temperature of and cooled to 60°C. The saturation of zeolites by
00°C for 2 h.
ammonia was conducted in a stream of the dried
Zeolite MFI was synthesized as described in [31] mixture of NH /N (1 : 1) for 15 min. Physically
9
5
3
2
and modified in accordance with [32]. The synthe- adsorbed ammonia was removed at 100°C in a
sized MFI zeolite was separated from the mother solu- helium stream for 1 h. After the sample was cooled to
tion via centrifugation, washed off, dried, and calcined 60°C in a dried helium stream at a feed velocity of
PETROLEUM CHEMISTRY Vol. 57 No. 7 2017

5
78
KOLESNICHENKO et al.
Table 1. Physicochemical characteristics of НHMZ samples with different UST times and synthetic zeolite HMFI
Pore structure**
Zeolite
sample
SiO /Al O , Crystallinity*,
a0***,
μmol/g
2
2
3
SBET
,
Souter surface
,
Vmicro
,
Vmeso + macro
,
mol/mol
%
2
2
3
3
m /g
m /g
cm /g
cm /g
HHMZ(0)
HHMZ(60)
HHMZ(90)
HMFI
37
37
37
57
71
88
88
70
333
343
338
348
88
93
110
102
0.119
0.117
0.108
0.112
0.054
0.058
0.075
0.060
1469
1305
1317
717
*
Calculated from the signal/noise ratio for reflection 101. ** Surface area calculated by the Brunauer–Emmett–Teller method (S
),
BET
area of the outer surface calculated by the t-Plot method (S
= S
– S
), and volume of micropores (V
) and the
outer surface
BET
micropores
micro
volume of meso- and macropores (V
tion of ammonia (TPD NH ).
). *** The total number of acid sites according to the temperature-programmed desorp-
meso + macro
3
3
0 mL/min, the linear temperature rise in the reactor lyuks-4000M chromatograph equipped with a flame-
was performed at a rate of 8°C/min to 800°C.
ionization detector and
a
capillary column
(
27.5 m/0.32 mm) filled with a CP-PoraPLOT-Q liq-
The diffuse reflectance infrared Fourier transform
uid phase (10-μm-thick adsorption layer). Helium
was used as a carrier gas. The chromatograms were
treated using the NetChromWin program.
(
DFIFT) spectra оf the samples in Vaseline oil were
measured in situ in an inert atmosphere (argon). The
spectra were registered at a temperature 450°C in a
Pike Diffus IR high-temperature cell combined with a
Bruker Vertex-70 FTIR spectrometer. IR spectra were
RESULTS AND DISCUSSION
–1
recorded with a resolution of 2 cm in the frequency
–
1
range of 600–4000 cm in the continuous mode for
Figure 1 shows the SEM micrographs of the origi-
5
min (194 scans per spectrum). The spectra were nal НHMZ(0) sample and the same sample after
treated using the Opus-7 software.
ultrasonic treatment. Before SEM measurements, the
НHMZ zeolite samples subjected to ultrasonic treat-
ment for 60 and 90 min were diluted with water by a
factor of 200 to prevent the aggregation of particles
during drying of the sample. As is seen from the
images, the original НHMZ(0) is composed of
agglomerates of crystals (crystal size of ~1–2 μm) with
a wide size distribution of agglomerates (3–100 μm).
Upon ultrasonic treatment (UST), the morphology of
the sample changes appreciably. SEM images
recorded for the НHMZ(60) sample also show coarse
agglomerates of crystals (up to 100 μm); however, they
are surrounded by both minute 3–5 μm agglomerates
and НHMZ zeolite crystallites. An increase in the
time of treatment to 90 min provides an even larger
contribution of minute agglomerates and crystallites.
The synthesis of olefins from DME was accom-
plished using suspended zeolites. Before suspension
preparation, magnesium (1 wt %) was deposited on the
surface of zeolites via impregnation as described in [9].
The magnesium-modified zeolite was ground and dis-
persed in the high-boiling liquid in accordance with the
above-described technique.
The catalyst suspensions in the DME conversion to
olefins were tested in an autoclave with inlet and outlet
fittings and equipped with a mechanical stirrer, a ther-
mocouple (chromel/copel), a reflux condenser, and a
dropping reflux condenser as a baffle. The reaction was
run at a temperature of 300°C and atmospheric pressure
under vigorous mechanical stirring in the continuous
flow regime. Dimethyl ether was bubbled through the
catalyst suspension layer. As a feedstock, dimethyl ether
manufactured by AkzoNobel N.V. (DME weight frac-
tion of 99.99%) diluted with molecular nitrogen was
used. The concentration of DME in the feed gas mix-
ture was 10 vol %. The gas flow velocity was 3.7 nL/h,
During water removal, zeolite crystals and minute
agglomerates aggregate to coarse loose agglomerates;
this finding may be attributed to their aggregative
instability during drying of the sample.
Thus, the ultrasonic treatment of the zeolite in
water causes an increase in the contribution of minute
agglomerates and НHMZ zeolite crystallites; their
content grows in the course of time. The drying of the
sample is accompanied by the development of coarse,
loose agglomerates composed of both zeolite crystal-
lites and smaller agglomerates.
–
1
and the weight load on the catalyst was 0.14 h . The
presented catalytic data refer to a catalyst operation
time of 2 h.
–
1
The VHSV was varied within 400–1200 h . To
attain the desired feed gas velocity, the gas flow was
controlled using Bronkhorst gas flow meters.
The gaseous product of DME conversion was
In accordance with the X-ray powder diffraction,
analyzed by gas-liquid chromatography on a Kristal- during ultrasonic treatment, regardless of its duration,
PETROLEUM CHEMISTRY
Vol. 57
No. 7
2017

DIMETHYL ETHER TO OLEFINS CONVERSION IN A SLURRY REACTOR
(a)
579
500 μm
500 μm
500 μm
100 μμ mm
100 μm
100 μm
30 μm
30 μm
30 μm
(
b)
(
c)
Fig. 1. SEM images for HMZ zeolite samples with different UST times in water: (a) НHMZ (0 min), (b) НHMZ (60 min), and
c) НHMZ (90 min).
(
(
a)
130
20
10
(b)
1
40
30
20
10
00
1
НHMZ 0 min
MFI
1
НHMZ 0 min
НHMZ 60 min
НHMZ 90 min
1
1
1
100
90
80
1
9
0
0
8
70
0
0.2
0.4
0.6
0.8
1.0
0
0.2
0.4
0.6
0.8
1.0
p/p0
p/p0
Fig. 2. Isotherms of low-temperature nitrogen adsorption for (a) the commercial НHMZ sample with different UST times and
b) the synthesized HMFI zeolite.
(
PETROLEUM CHEMISTRY Vol. 57 No. 7 2017

5
80
KOLESNICHENKO et al.
1
.0
accompanied by an increase in the contribution of
meso- and macropores.
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
The isotherm obtained for the HMFI sample fol-
lows the pattern classical for zeolites having micron
and submicron sizes of crystals with a small amount of
mesopores (Fig. 2b). The presence of hysteresis in the
range of p/p = 0.4–0.7 and its absence in the range of
0
НHMZ 0 min
НHMZ 60 min
НHMZ 90 min
high relative pressures provide evidence for the limited
connection of the pore system with the zeolite crystal
outer surface.
MFI
The detailed analysis of the pore structure of the
samples indicates that the outer surface areas of HMFI
and НHMZ(90) samples are similar (Table 1); how-
ever, in the case of НMFI, the volume of micropores
is higher. This is evidence that the agglomerates of
nanocrystals in the НMFI sample represent a densely
intergrown structure rather than freely associated
crystals. A considerable contribution of macropores
for НHMZ(90) suggests a weak aggregation state of
0.1
0
5
0
150 250 350 450 550 650 750
Temperature, °С
Fig. 3. NH TPD spectra of the НHMZ samples with var-
ious times of UST and the synthesized HMFI zeolite.
3
no marked changes in the crystalline structure of the zeolite nanocrystals. The acidic properties of the zeo-
НHMZ zeolite are observed (Table 1).
lites were studied by the temperature-programmed
desorption of ammonia (TPD NH ) (Fig. 3, Table 1).
3
Figure 2 shows the isotherms of nitrogen low-tem-
perature adsorption for НHMZ and HMFI samples.
An analysis of the TPD NH curves reveals that the
3
НHMZ samples, regardless of the UST time, are
characterized by close number and strength of acid
For all of the НHMZ samples (Fig. 2a), regardless
of the UST time, the adsorption isotherms display a sites, albeit ultrasonic treatment entails an insignifi-
well-defined hysteresis in the relative pressure (р/р ) cant reduction in the number of medium- and weak-
0
range of 0.4–1.0, indicating the presence of meso- and strength acid sites. This finding may be attributed to
macropores connecting with the environment through zeolite amorphization under the UST conditions.
micropores. As the time of ultrasonic treatment of the Compared with НHMZ, the synthesized HMFI sam-
НHMZ zeolite increases, the volume of meso- and ple features a smaller number of acid sites, in agree-
macropores tends to increase predominantly owing to ment with a higher silicon/aluminum ratio. For the
the growth of macropores. This observation may be surface of zeolite samples, more detailed information
explained by the appearance of loose agglomerates about the origin, number, and strength of acid sites
formed via the aggregation of smaller particles during was obtained by DFIFT in situ used in combination
with quantum-chemical calculations performed as
described in length in [33]. The IR spectra of all of the
zeolites show four absorption bands corresponding to
the stretching vibrations of ОН groups belonging to
acid sites of different strengths. Bands with vibration
drying of the zeolite after treatment. These results are
in good agreement with the data derived from the
SEM images, because a change of dense agglomerates
(
Fig. 2a) for looser agglomerates (Fig. 2b) should be
–1
–1
frequencies at ν = 3590 cm and ν = 3614 cm indi-
cate that two types of strong Brønsted acid sites occur
on the surface of zeolites—with the end acidic group
Al–OH(H) (type 1) and the bridging bond Al–O(H)–
Table 2. Distribution of different types of acid sites
in НHMZ and HMFI zeolites
Fraction of acid sites
on zeolite surface*, %
–1
Type of acid site
on zeolite surface
Si (type 2). The band in the range ν = 3665 cm cor-
responds to vibrations of two associated hydroxyl
groups with weak acidic properties (type 3). In addi-
НHMZ(90)
НMFI
–1
tion, the band at ν = 3733 cm is assigned to vibra-
tions of the isolated Al–OH bond belonging to the site
with amphoteric properties (type 4).
1
+ 2
3
60.0
26.5
13.5
1317
24.0
27.0
35.0
690
4
Thus, four types of acid sites were identified on
the surface of zeolites НHMZ and HMFI. The dis-
tributions of different types of acid sites on the sur-
face of НHMZ and HMFI zeolites differ appreciably
Total acidity a , kmol/g
0
*
The distribution of acid sites was estimated from the relative inten-
sities of absorption bands in the DFIFT spectrum taking into consid-
eration that the total amount of all ОН groups is 100% (provided that
the samples were prepared under the same conditions).
(
Table 2). As opposed to the synthetic zeolite HMFI,
the total acidity and the content of strong Brønsted
PETROLEUM CHEMISTRY
Vol. 57
No. 7
2017

DIMETHYL ETHER TO OLEFINS CONVERSION IN A SLURRY REACTOR
581
Table 3. Testing of suspended zeolite samples in the synthesis of lower olefins from DME under conditions of three-phase
catalysis
Selectivity for olefins, %
Content of СН ОН
in the product,
wt %
3
Particle size
in suspension, nm
DME conversion,
wt %
Catalyst sample*
∑
С –С
∑С –С
3=
2=
4=
2=
Mg/НHMZ(0)
Mg/НHMZ(60)
Mg/НHMZ(90)
Mg/НMFI
1530–1700
600–813
73–85
28.0
89.0
92.0
81.0
77.0
55.0
67.0
59.8
69.0
47.7
47.3
52.5
1.1
2.5
3.0
61–74
16.6
*
The content of magnesium on the zeolite surface is 1 wt %.
acid sites for the nanosized zeolite HMZ(90) are DME conversions is the intradiffusion retardation of
much higher (type 1 + 2), while the content of weak the reaction in the pores of the primary crystalline
acid sites (type 3) is almost the same. As for the site structure of zeolite. This phenomenon is apparently
with amphoteric properties (type 4), its content for observed at particle sizes above 500–600 nm.
zeolite НHMZ(90) is almost three times lower than
that for the synthetic HMFI.
As the size of Mg/НHMZ catalyst particles
decreases, the rate of all reactions, including second-
Using the zeolite samples, the magnesium catalysts ary transformations, grows; as a consequence, the
of olefin synthesis were prepared [9]. The textural product selectivity changes considerably (Fig. 4).
characteristics of MFI zeolites remained practically Noteworthy is a somewhat higher selectivity for
unchanged after deposition of 1 wt % magnesium on butenes and a lower selectivity for high-boiling hydro-
them, but the total acidity decreased.
carbons in the case of the Mg/НHMZ(0) sample with
the largest sizes of particles. As the size of particles
After drying and calcination, magnesium-contain-
ing zeolite catalysts were ultrasonically dispersed in
the high-boiling liquid Dowtherm for 20 min. The size
of catalyst particles in the suspension was determined
by dynamic light scattering.
The dispersing of НHMZ-based zeolite catalysts in
the high-boiling liquid yielded suspensions with
micron, submicron, and nanosized (after 90-min
ultrasonic treatment) degrees of dispersion (Table 3).
Using the catalyst based on the synthesized zeolite, a
nanodispersed suspension was prepared: during ultra-
sonic dispersing, the agglomerates of the original MFI
zeolite disintegrated into nanocrystallites.
decreases, selectivity for butenes and hydrocarbons С
5
increases (Fig. 4). It appears that this finding may be
explained by an increase in the contribution of the
alkene cycle [34] to the formation of products. It is
interesting that the selectivity for propane and butanes
passes through a maximum. This fact indicates that
the optimum sizes of particles can be found for the
reaction of hydrogen redistribution.
At a close degree of dispersion of suspensions for
the catalysts based on НHMZ(90) and HMFI, the
conversion of DME is noticeably higher for (1%)
The catalyst suspensions were tested in the conver-
sion of DME in the slurry reactor (Table 3). As is seen,
the time of preliminary ultrasonic treatment of the
HMZ zeolite in water has a strong effect on the size of
zeolite catalyst particles in the suspension: the longer
the original HMZ sample was ultrasonically treated, the
smaller the size of particles in the resulting suspensions.
As the size of zeolite particles decreased, the conversion
of DME increased. Note that, on passage from sample
Mg/НHMZ(0) to sample Mg/НHMZ(60), the size of
particles decreased by a factor of 2.3 and the conversion
of DME increased by a factor of 3.2. Further almost 10-
fold reduction in the size of particles led to a small gain
in the conversion of DME (from 89 to 92%). For the
Mg/НMFI sample with a particle size of 61–74 nm, the
conversion of DME was also high and it was much
higher than that for the Mg/НHMZ(0) sample. The
most probable reason for the experimentally measured
4
0
0
НHMZ(0)
НHMZ(60)
НHMZ(90)
3
20
10
0
С С С С С С С4 С5
1
2=
2
3=
3
4=
Hydrocarbons
in the gaseous product
Fig. 4. Composition of hydrocarbon products of DME
conversion under conditions of three-phase catalysis at
different times of UST of НHMZ.
PETROLEUM CHEMISTRY Vol. 57 No. 7 2017

5
82
KOLESNICHENKO et al.
Mg/НHMZ (Table 3), which is due to a much high Mg/НMFI owing to formation of a higher amount of
acidity of the original zeolite НHMZ (Table 2). Prob- propylene. Probably, this is due to the fact that the
ably, the silicon/aluminum ratio in the samples under methylation reaction over the Mg/HMFI zeolite is less
comparison plays a considerable role.
intense. Moreover, the amount of methanol produced
The selectivity for lower olefins is also different for over the (1%) Mg/HMFI catalyst is much higher than
the catalysts based on НHMZ(90) and the synthesized that formed over systems (1%) Mg/НHMZ. This
HMFI zeolite. For example, the selectivity for olefins effect is possibly explained by a high content of type 4
С –С is higher for (1%) Mg/НHMZ(90) owing to
2
=
4=
acid sites in the original HMFI zeolite (Table 3). It is
the contribution due to butenes (Fig. 5) At the same known [35, 36] that, over these amphoteric sites, the
.
time, selectivity for olefins С –С is higher for conversion of DME to methanol occurs to form
2=
3=
methoxy groups in accordance with the following scheme:
−
δ
OH
OCH
Al
CH3 O CH3
+ CH OH
3
Al
Scheme.
+
5
hydrocarbons C . At the same degree of dispersion of
suspensions, the catalytic properties of the suspended
catalysts are considerably affected by the strength dis-
tribution of acid sites. An increase in the share of
amphoteric sites provides a gain in selectivity for
methanol, while selectivity for butenes decreases.
As a result, the conversion of DME shifts to the
production of methanol rather than to the alkylation
of ethylene and propylene with the concomitant
changes in the composition of the products.
Thus, our studies have shown that, in the conver-
sion of DME in the three-phase system in the slurry
reactor, the catalytic properties of MFI zeolites are
strongly influenced by the particle size of the sus-
pended catalyst: on passage to nanosized suspensions,
the activity of the catalyst grows by several times, in
agreement with a change in the textural characteristics
of the zeolites. An increase in the size of catalyst parti-
cles improves selectivity for olefins probably via the
arenium ion mechanism but worsens selectivity for
ACKNOWLEDGMENTS
This work was supported by the Russian Science
Foundation, project no. 15-13-00104.
REFERENCES
. G. Cai, Z. Liu, R. Shi, Ch. He, L. Yang, Ch. Sun, and
Y. Chang, Appl. Catal., A 125, 29 (1995).
. B. V. Vora, P. R. Pujado, L. W. Miller, P. T. Barger,
H. R. Nilsen, S. Kvisle, and T. Fuglerud, Stud. Surf.
Sci. Catal. 136, 537 (2001).
. Makoto Inomata, Akira Higashi, Yoshiteru Makino,
and Yoshinori Mashiko, US Patent No. 6852897
1
2
3
40
30
20
(2005).
Mg/НHMZ(90)
Mg/HMFI
4
5
. M. Hack, U. Koss, P. Konig, M. Rothaemel, and
H.-D. Holtmann, US Patent, No. 7015369.
. U. Olsbye, S. Svelle, M. Bjorgen, P. Beato, T. V. Jans-
sens, F. Joensen, S. Bordiga, and K. P. Lillerud, Angew.
Chem., Int. Ed. Engl. 51, 5810 (2012).
6
. P. Kumar, J. W. Thybaut, G. B. Marin, S. Svelle, and
U. Olsbye, Ind. Eng. Chem. Res. 52, 1491 (2013).
10
7
. A. Takahashi, W. Xia, Q. Wu, T. Furukawa, I. Naka-
mura, H. Shimada, and T. Fujitani, Appl. Catal.,
A 467, 380 (2013).
0
С1 С2= С2 С3= С3 С4= С4 С5
Hydrocarbons in the gaseous product
8
9
. E. N. Biryukova, T. I. Goryainova, R. V. Kulumbegov,
N. V. Kolesnichenko, and S. N. Khadzhiev, Pet. Chem.
51, 50 (2011).
. T. I. Goryainova, E. N. Biryukova, N. V. Kole-
snichenko, and S. N. Khadzhiev, Pet. Chem. 51, 181
(2011).
Fig. 5. Composition of hydrocarbon products of DME
conversion under conditions of three-phase catalysis over
suspended Mg/НHMZ(90) and Mg/HMFI zeolites.
PETROLEUM CHEMISTRY
Vol. 57
No. 7
2017

DIMETHYL ETHER TO OLEFINS CONVERSION IN A SLURRY REACTOR
583
1
0. N. V. Kolesnichenko, E. E. Kolesnikova, L. E. Kitaev, 23. Ya. Liu, T. Hanaoka, T. Miyazawa, K. Murata,
E. N. Biryukova, N. I. Trukhmanova, and S. N. Kha-
K. Okabe, and K. Sakanishi, Fuel Proc. Technol. 90,
dzhiev, Pet. Chem. 52, 179 (2012).
901 (2009).
1
1. S. N. Khadzhiev, N. V. Kolesnichenko, E. N. Khivrich, 24. G. Rispoli, D. Sanfilippo, and A. Amoroso, in Proceed-
E. E. Kolesnikova, and T. I. Batova, Pet. Chem. 53, 259
2013).
ings of the World Heavy Oil Congress (Puerto La Cruz,
2009); Paper 359.
(
1
2. S. M. T. Almutairi, B. Mezari, E. A. Pidko, 25. A. Haghtalab, M. Nabipoor, and S. Farzad, Fuel Proc.
P. C. M. M. Magusin, and E. J. M. Hensen, J. Catal.
07, 194 (2013).
3. S. N. Khadzhiev, N. V. Kolesnichenko, G. I. Lin,
Technol. 104, 73 (2012).
3
26. S. N. Khadzhiev, A. S. Lyadov, M. V. Krylova, and
1
A. Yu. Krylova, Pet. Chem. 51, 25 (2011).
N. A. Markova, Z. M. Bukina, D. A. Ionin, and 27. M. Sadeqzadeh, S. Chambrey, S. Piche, P. Fongar-
G. M. Grafova, RF Patent No. 2442767 (2012).
land, A. Y. Khodakov, F. Luck, D. Curulla-Ferre,
J. Bousquet, and D. Schweich, Catal. Today 215, 52
14. V. L. Baiburskii, V. V. Vints, V. N. Genkin, M. V. Gen-
(2013).
kin, I. I. Lishchiner, O. V. Malova, E. S. Mortikov, and
S. E. Dolinskii, RF Patent No. 2160160 (2000).
28. F. G. Botes, J. Van de Loosdrecht, and J. W. Nie-
mantsverdriet, Catal. Today 215, 112 (2013).
15. O. V. Malova, I. I. Lishchiner, S. E. Dolinskii,
V. A. Plakhotnik, A. N. Kuzmicheva, and K. S. Mor- 29. S. N. Khadzhiev, N. V. Kolesnichenko, N. N. Ezhova,
tikov, RF Patent No. 2160161 (2000).
6. N. V. Kolesnichenko, L. E. Kitaev, Z. M. Bukina,
I. G. Korosteleva, O. V. Yashina, and E. N. Khivrich,
RF Patent No. 2547838 (2015).
1
N. A. Markova, V. V. Yushchenko, O. V. Yashina, and 30. N. V. Kolesnichenko, N. N. Ezhova, and O. V. Yashina,
G. I. Lin, A. Ya. Rozovskii, Kinet. Katal. 48, 846
Pet. Chem. 56 (6), 607 (2016).
2007).
(
3
1. Verified Synthesis of Zeolitic Materials, Ed. By H. Rob-
17. C. Thaller, H. Schmaderer, N. Schodel, Er. Haidegger,
son (Elsevier Science, 2001), p. 288.
H. Schmigalle, Ax. Behrens, and V. Goke, CA Patent
No. 2863285.
3
2. A. G. Popov, V. S. Pavlov, and I. I. Ivanova, J. Catal.
3
35, 155 (2016).
18. B. Buisson, S. Donegan, D. Wray, A. Parracho, 33. A. S. Rodionov, Candidate’s Dissertation in Chemistry
J. Gamble, Ph. Caze, J. Jorda, and C. Guermeur,
Chem. Today 27, 12 (2009).
(Moscow, 2013) [in Russian].
3
4. C. N. Khadzhiev, M. V. Magomedova, and E. G. Pere-
19. A. Lira and R. G. Tailleur, Fuel 97, 49 (2012).
sypkina, Pet. Chem. 54, 243 (2014).
2
0. S. D. Pollington, D. L. Enache, P. Landon, S. Meena- 35. S. Tamm, H. H. Ingelsten, M. Skoglundh, and
kshisundaram, N. Dimitratos, A. Wagland, G. J. Hutch-
A. E. C. Palmqvist, J. Catal. 276, 402 (2010).
ings, and E. H. Stitt, Catal. Today 145, 169 (2009).
3
6. J. G. Chen, P. Basu, T. H. Ballinger, and J. T. Yates,
21. A. Pintar, G. Bercic, M. Besson, and P. Gallezot, Appl.
Langmuir 5, 352 (1989).
Catal. B: Envir 47, 143 (2007).
22. S. N. Khadzhiev, Pet. Chem. 51, 1 (2011).
Translated by T. Soboleva
PETROLEUM CHEMISTRY Vol. 57 No. 7 2017